Report No. 49137-MY Malaysia Productivity and Investment Climate Assessment Update August 2009 Poverty Reduction and Economic Management Sector Unit East Asia and Pacific Region Document of the World Bank ACKNOWLEDGEMENTS This report is the result of fruitful collaboration between the Economic Planning Unit (EPU), Prime Minister's Department, Malaysia, and the World Bank. The main data source for this report is the Malaysia Productivity and Investment Climate Survey (PICS) carried out by the Department of Statistics and EPU. Writing the report was a collaborative effort. The World Bank team included Xubei Luo (Task Leader, EASPR), Hiau Looi Kee (DECRG), Philip Schellekens (EASPR), Vatcharin Sirimaneetham (EASPR), Takanobu Terada (University of Maryland), and Charles Chatree Udomsaph (Georgetown University). Albert G. Zeufack managed the design and implementation of the Productivity and Investment Climate Survey and the project before moving to Khazanah. The report builds on the findings of Malaysia Firm Competitiveness, Investment Climate, and Growth report (World Bank, 2005). It was prepared under the overall guidance of Linda Van Gelder (Sector Manager) and Mathew A. Verghis (Lead Economist). Annette Dixon (Country Director) and Vikram Nehru (Sector Director) provided oversight. The World Bank peer reviewers for this report were Deepak Bhattasali (Lead Economist, SASEP) and Mary C. Hallward-Driemeier (Senior Economist, DECRG). Lynn M. Gross skillfully formatted and prepared the report. Susan Middaugh of Have Pen, Will Travel edited it. The report team benefited from inputs and consultations at different stages from stakeholders in Malaysia. We would like to sincerely thank the Economic Planning Unit, Department of Statistics Malaysia and other Ministries/agencies that participated in the PICS surveys either directly or indirectly. The team is grateful to Yukon Huang for fruitful discussions at various stages of the report's preparation, to Ahmad Ahsan, Hamid R. Alavi, Vivian Hon, Yue Li, Omporn Regel, and Hong W. Tan for their useful suggestions and inputs. The team is also indebted to the workers and managers across Malaysia who devoted their time and provided information for the Productivity and Investment Climate Surveys. Table of Contents EXECUTIVE SUMMARY ......................................................................................................................................... I CHAPTER 1. MALAYSIA'S INVESTMENT CLIMATE TODAY ............................................................ 1 GROWTH, INVESTMENT AND PRODUCTIVITY IN MALAYSIA................................................................. 2 PICS 2002 AND PICS 2007 ................................................................................................................................... 5 RECENT DEVELOPMENTS IN MALAYSIA'S INVESTMENT CLIMATE ..................................................... 7 Subjective Indicators ............................................................................................................................................ 8 Perceptions of the Three Top Constraints to Doing Business .............................................................8 Perception of the Severity of Investment Climate Constraints ............................................................9 Objective Indicators ........................................................................................................................................... 13 Infrastructure .........................................................................................................................14 Access to Finance ...................................................................................................................15 Regulatory Framework.............................................................................................................16 Tax Rate and Tax Administration ...............................................................................................18 Supply of Skills ......................................................................................................................19 Innovation/Technological Capabilities ........................................................................................20 Crime and Theft .....................................................................................................................21 MALAYSIA'S INVESTMENT CLIMATE FROM AN INTERNATIONAL PERSPECTIVE .......................... 22 CONCLUSIONS................................................................................................................................................... 30 CHAPTER 2. FIRM PERFORMANCE IN THE MANUFACTURING SECTOR .................................. 32 MANUFACTURING SECTOR PERFORMANCE, PRODUCTIVITY GROWTH AND INVESTMENT CLIMATE ............................................................................................................................................................. 33 Manufacturing Sector Performance ................................................................................................................... 33 Productivity Growth and Investment Climate .................................................................................................... 36 MANUFACTURING SECTOR LABOR PRODUCTIVITY PERFORMANCE ................................................ 37 Measuring Labor Productivity ........................................................................................................................... 37 Performance at the National Level .................................................................................................................... 37 Performance at the Industry Level ..................................................................................................................... 39 Performance at the Regional Level .................................................................................................................... 40 TOTAL FACTOR PRODUCTIVITY AND INVESTMENT CLIMATE ............................................................ 41 Measuring Total Factor Productivity................................................................................................................. 41 Total Factor Productivity in Malaysia ............................................................................................................... 42 Regressing Labor Productivity and TFP on Investment Climate Indicators and Firm Characteristics ............ 43 Correlations Between Firm Characteristics and Performance .......................................................................... 44 Correlations Between Investment Climate and Firm Performance .....................................................47 Correlations Between Changes in Investment Climate and Changes in Productivity .............................49 CONCLUSIONS................................................................................................................................................... 50 CHAPTER 3. FIRM PERFORMANCE IN THE SERVICES SECTOR .................................................. 51 OVERALL SERVICES SECTOR PERFORMANCE.......................................................................................... 52 Performance at the National Level .................................................................................................................... 53 Performance at the Industry Level ..................................................................................................................... 57 Performance at the Firm Level .......................................................................................................................... 60 SKILLS SHORTAGE, INNOVATION READINESS, AND REGULATORY REGIMES IN SERVICES SECTOR ............................................................................................................................................................... 63 Skills Shortage.................................................................................................................................................... 63 Innovation Readiness ......................................................................................................................................... 65 Regulatory Regime ............................................................................................................................................. 68 FIRM PERFORMANCE AND INVESTMENT CLIMATE ................................................................................ 70 Correlates Between Firm Characteristics and Performance ............................................................................. 70 Impact of the Presence of Foreign Firms ........................................................................................................... 73 CONCLUSIONS................................................................................................................................................... 73 CHAPTER 4. REGIONAL PERSPECTIVES OF INVESTMENT CLIMATE ....................................... 75 REGIONAL DEVELOPMENT CORRIDORS .................................................................................................... 76 REGIONS WITH THE BEST AND WORST BUSINESS ENVIRONMENT .................................................... 78 Manufacturing Firms' Perception...................................................................................................................... 78 Services Firms' Perceptions............................................................................................................................... 79 SUBJECTIVE MEASURES OF INVESTMENT CLIMATE ACROSS REGIONS ........................................... 80 Top Investment Climate Constraints .................................................................................................................. 80 Manufacturing Firms ...............................................................................................................80 Services Firms........................................................................................................................81 Severity of Investment Climate Constraints ....................................................................................................... 82 Manufacturing Firms ...............................................................................................................82 Services Firms........................................................................................................................83 OBJECTIVE MEASURES OF INVESTMENT CLIMATE ACROSS REGIONS ............................................. 84 Labor Skills ........................................................................................................................................................ 84 Manufacturing Firms ...............................................................................................................84 Services Firms........................................................................................................................87 Affordability and Quality of Business Support Services..................................................................................... 88 Manufacturing Firms ...............................................................................................................88 Services Firms........................................................................................................................89 Public Infrastructure and Services ..................................................................................................................... 89 Manufacturing Firms ...............................................................................................................89 Services Firms........................................................................................................................92 Government Regulations .................................................................................................................................... 94 Manufacturing Firms ...............................................................................................................94 Services Firms........................................................................................................................97 Innovative Activities and ICT Usage .................................................................................................................. 98 Manufacturing Firms ...............................................................................................................98 Services Firms........................................................................................................................99 Access to Financing ......................................................................................................................................... 100 Manufacturing Firms .............................................................................................................100 Services Firms......................................................................................................................102 CONCLUSIONS................................................................................................................................................. 103 CHAPTER 5. IMPROVING SKILLS ......................................................................................................... 107 MALAYSIA'S SKILLS GAP............................................................................................................................. 108 Macro Evidence ............................................................................................................................................... 108 Upper Secondary Education ....................................................................................................110 Tertiary Education ................................................................................................................111 Micro Evidence ................................................................................................................................................ 112 Employers' Perspectives ........................................................................................................115 Employees' Perspectives ........................................................................................................116 Incidence and Causes of Vacancies ................................................................................................................. 118 SKILLS SHORTAGES AND WAGE PREMIUMS .......................................................................................... 122 Estimated Returns to High School Diplomas and College Degrees................................................................. 122 In-Service Training .......................................................................................................................................... 124 CONCLUSIONS AND POLICY SUGGESTIONS............................................................................................ 126 CHAPTER 6. STRENGTHENING TECHNOLOGICAL CAPABILITIES........................................... 127 MALAYSIA'S TECHNOLOGICAL PERFORMANCE ................................................................................... 127 Global Competitiveness Technology Indicators ............................................................................................... 128 Technological Input Indicators ........................................................................................................................ 129 Research and Development (R&D) Expenditures ........................................................................129 The Shares of Researchers and Technicians in R&D ....................................................................132 Technological Output Indicators ..................................................................................................................... 136 Patents ................................................................................................................................137 Adopters, Adapters or Creators ................................................................................................140 MANUFACTURING SECTOR FIRM LEVEL TECHNOLOGICAL CAPABILITY ANALYSIS ................. 146 Investment, Production, and Linkages Technological Capability Analysis ..................................................... 147 Variations of Technological Capabilities Across Regions ............................................................................... 151 Variations of Technological Capabilities Across Industries ............................................................................ 152 Variations of Technological Capabilities across Firms of Different Sizes and Ownership ............................. 153 CONCLUSIONS AND POLICY SUGGESTIONS............................................................................................ 155 REFERENCES ................................................................................................................................................... 157 Tables Table 1.1: Sources of Growth, Total Economy: 1987-2007 .......................................................................................... 4 Table 1.2: Global Competitiveness Index - Rankings of East Asian Countries (Selected Indicators)......................... 23 Table 3.1: Training and Services Sector Firms ­ Percentage of firms say "yes" to the following questions in 2002 and 2007 ...................................................................................................................................................................... 65 Table 3.2: Technology and Services Firms ­ Percentage of firms say "yes" to the following questions in 2002 and 2007 ............................................................................................................................................................................. 67 Table 3.3: Key Drivers of Innovation (Adaptation and Creation) ............................................................................... 68 Table 3.4: Qualitative Comparison of Business-Support Services Regulation in Malaysia, Taiwan, Korea, Chile, Singapore and Thailand ............................................................................................................................................... 70 Table 4.1: Five development corridors in Malaysia .................................................................................................... 77 Table 4.2: Manufacturing Firms Perception on Regions with the Best Business Environment .................................. 78 Table 4.3: Manufacturing Firms Perception on Regions with the Worst Business Environment ................................ 79 Table 4.4: Services Firms' Perceptions on Regions with the Best Business Environment.......................................... 80 Table 4.5: Services Firms' Perceptions on Regions with the Worst Business Environment ....................................... 80 Table 4.6: Manufacturing firms: The Most Important Business Obstacles in 2007 by Region (open-ended question) ..................................................................................................................................................................................... 81 Table 4.7: Services Firms: The Most Important Business Obstacles in 2007 by Region (open-ended question)........ 82 Table 4.8: Manufacturing firms: Severity of business obstacles by region (close-ended question) ............................ 83 Table 4.9: Services Firms: Severity of Business Obstacles by Region (close-ended question)................................... 84 Table 4.10: Time It Takes Manufacturing Firms to Obtain Basic Infrastructure (days).............................................. 90 Table 4.11: The Quality of Electrical, Water Supply, and Fixed Telephone Services Viewed by Manufacturing Firms ..................................................................................................................................................................................... 91 Table 4.12: Time to Obtain Basic Infrastructure by Services Firms (days) ................................................................. 92 Table 4.13: The Quality of Electrical, Water Supply, and Fixed Telephone Services Viewed by Services Firms ..... 93 Table 4.14: Average Number of Weeks to Obtain Licenses, Permits, and Approvals/Certificates for Manufacturing Firms ............................................................................................................................................................................ 95 Table 4.15: Average Number of Days Spent in Contact with Different Agencies by Manufacturing Firms .............. 96 Table 4.16: Average Number of Weeks to Obtain Licenses for Services Firms ......................................................... 97 Table 4.17: Average Number of Days Spent in Contact with Different Agencies by Services Firms......................... 97 Table 4.18: Innovative Activities Carried Out by Manufacturing Firms ..................................................................... 98 Table 4.19: Innovative Activities Carried Out by Service Firms............................................................................... 100 Table 4.20: Key Investment Climate Challenges by Region ..................................................................................... 104 Table 5.1: Share of Skilled and Unskilled Workers and Their Hourly Wages in the Manufacturing and Business Support Services Sectors (2001 and 2006) ................................................................................................................ 114 Table 5.2: Educational Mismatch in Malaysia .......................................................................................................... 117 Table 6.1: Ranking of Technological Readiness and Innovation Sub-Indexes.......................................................... 128 Table 6.2: Research and Development Expenditure (as percent of GDP) ................................................................. 129 Table 6.3: Researchers and Technicians in R&D (Per Million People)..................................................................... 133 Table 6.4: Patents Issued by the United States to Residents of Selected Countries, 1996-2007 ............................... 139 Table 6.5: Patents Issued by the United States to Residents of Selected Countries (Per Million People), 1996-2007 ................................................................................................................................................................................... 139 Table 6.6: Descriptive Statistics of Manufacturing TCI ............................................................................................ 149 Figures Figure 1.1: Annual growth of GDP and GDP per capita in Malaysia (1987-2007) ....................................................... 3 Figure 1.2: Sectoral composition of GDP in Malaysia (1998-2007) ............................................................................. 4 Figure 1.3: Domestic and foreign direct investment (1987-2006/7) .............................................................................. 4 Figure 1.4: Top Obstacles to Doing Business................................................................................................................ 9 Figure 1.5: Percentages of Manufacturing Firms Perceiving Particular Investment Climate Indicators Major or Severe Obstacles to Doing Business in Malaysia ........................................................................................................ 11 Figure 1.6: Percentages of Services Firms Perceiving Particular Investment Climate Indicators Major or Severe Obstacles to Doing Business in Malaysia .................................................................................................................... 12 Figure 1.7: Change in the Percentage of Firms Rating Indicated Constraint as "Severe" or "Very Severe" over the Period 2002 to 2007 (Manufacturing) ......................................................................................................................... 13 Figure 1.8: Change in the Percentage of Firms Rating Indicated Constraint as "Severe" or "Very Severe" over the Period 2002 to 2007 (Services).................................................................................................................................... 13 Figure 1.9: Objective Indicators of Investment Climate in PICS I and PICS II ­ Infrastructure ................................. 15 Figure 1.10: Objective Indicators of Investment Climate in PICS I and PICS II -- Access to Finance ..................... 16 Figure 1.11: Objective Indicators of Investment Climate in PICS I and PICS II - Regulatory Framework ................ 17 Figure 1.12: Objective Indicators of Investment Climate in PICS I and PICS II ­ Tax Rate and Tax Administration ..................................................................................................................................................................................... 19 Figure 1.13: Objective Indicators of Investment Climate in PICS I and PICS II - Labor Market ............................... 20 Figure 1.14: Objective Indicators of Investment Climate in PICS I and PICS II - Innovation/technological Capabilities .................................................................................................................................................................. 21 Figure 1.15: Objective Indicators of Investment Climate of Services Firms - Crime, Security and Court.................. 22 Figure 1.16: International Comparison of Objective Indicators - Infrastructure ......................................................... 25 Figure 1.17: International Comparison of Objective Indicators ­ Access to Domestic Finance ................................. 26 Figure 1.18: International Comparison of Objective Indicators ­ Labor Market ........................................................ 27 Figure 1.19: International Comparison of Objective Indicators ­ Regulatory framework .......................................... 28 Figure 1.20: International Comparison of Objective Indicators ­ Tax Administration ............................................... 29 Figure 1.21: Corporate Income Tax in East Asian Countries (1997-2007) ................................................................. 29 Figure 1.22: International Comparison of Objective Indicators ­ Crime and Security Cost ....................................... 30 Figure 2.1: Composition of Industrial Production in Malaysia, 1999-2005 ................................................................ 33 Figure 2.2: Industrial Production Index across Industries, 1999-2007 ........................................................................ 34 Figure 2.3: Profit Ratio by Industry (2004-2006) ........................................................................................................ 35 Figure 2.4: Share of Export of the Nine Industries Covered by PICS ......................................................................... 36 Figure 2.5: Labor Productivity in Different Industries ­ International comparison (Median Value-Added per Worker, 2000 U.S. dollar) ......................................................................................................................................................... 39 Figure 2.6: Firm Labor Productivity by Industry (Median Value Added per Worker, 2000 Malaysia Ringgit) ......... 40 Figure 2.7: Firm Labor Productivity by Region (Median Value-Added per Worker, 2000 Malaysia Ringgit) ........... 41 Figure 2.8: Firm Total Factor Productivity by Industry (2004-2006) .......................................................................... 43 Figure 2.9: Firm Total Factor Productivity by Region (2004-2006) ........................................................................... 43 Figure 3.1: Employment Share of Services Sector in 1999 and 2005 ......................................................................... 54 Figure 3.2: Share of Services Sector in GDP in 1999 and 2005 .................................................................................. 55 Figure 3.3: Share of Services Sector in GDP Relative to GDP per Capita .................................................................. 55 Figure 3.4: Labor Productivity of Services Sector in 1999 and 2005 .......................................................................... 56 Figure 3.5: Labor Productivity in All Sectors in 1999 and 2005 ................................................................................. 56 Figure 3.6: Cross-Country Comparison of Labor Productivity in the Services Sector ................................................ 57 Figure 3.7: Services Exports in Selected Countries in 2001 & 2006 ........................................................................... 58 Figure 3.8: Services Imports in Selected Countries in 2001 and 2006 ........................................................................ 58 Figure 3.9: Industry Performance in Services Sector, 2001 & 2006 ........................................................................... 60 Figure 3.10: Locally Trained Skilled Workers vs. Overseas Trained Skilled Workers ............................................... 64 Figure 4.1: GDP per capita in 2000 and Annual Growth Rate .................................................................................... 76 Figure 4.2: Number of Weeks Required to fill Job Vacancies in Manufacturing Firms ............................................. 85 Figure 4.3: Share of Manufacturing Firms Citing Lack of Basic and Technical Skills as the Cause of Job Vacancies ..................................................................................................................................................................................... 86 Figure 4.4: Share of Manufacturing Firms that Carried out Activities to Improve Labor Skills ................................. 87 Figure 4.5: Number of Weeks Required to Fill Job Vacancies in Services Firms ....................................................... 87 Figure 4.6: Share of services firms citing a lack of basic and technical skills as the cause of job vacancies .............. 88 Figure 4.7: The Share of Manufacturing Firms Reporting Selected Business Support Services are Affordable......... 89 Figure 4.8: Estimated Loss from Theft , Robbery or Vandalism and Cost for Providing Security by Manufacturing Firms ............................................................................................................................................................................ 92 Figure 4.9: Estimated Loss from Theft, Robbery or Vandalism and Cost for Providing Security by Services Firms. 94 Figure 4.10: Average Number of Days to Clear Customs Procedure for Exports and Imports by Manufacturing Firms ..................................................................................................................................................................................... 96 Figure 4.11: The Share of Manufacturing Firms that Regularly Use Email and Have a Website for Their Business . 99 Figure 4.12: Share of Working Capital Financed by Retained Earnings and Bank Loans for Manufacturing Firms 100 Figure 4.13: Share of New Investments Financed by Retained Earnings and Bank Loans for Manufacturing Firms ................................................................................................................................................................................... 101 Figure 4.14: Approximate Value of Collateral Required as a Percentage of Loan Value for Manufacturing Firms . 102 Figure 4.15: Share of Working Capital Financed by Retained Earnings and Bank Loans for Services Firms .......... 103 Figure 5.1: Tertiary School Enrollment Rate and Public Spending on Tertiary Education as percent of GDP in East Asian/Upper Middle Income/OECD Countries ......................................................................................................... 110 Figure 5.2: Malaysia's Upper Secondary Education Gross Enrollment Rates .......................................................... 111 Figure 5.3: Malaysia's Tertiary Education Gross Graduation Ratio in International Perspective, 2005 ................... 112 Figure 5.4: Employers' Assessments of Worker Quality: Local Professionals ......................................................... 115 Figure 5.5: Employers' Assessments of Worker Quality: Local Skilled Production Workers .................................. 116 Figure 5.6: Incidence of Vacancies for Professionals and Skilled Production Workers ............................................ 118 Figure 5.7: Three Most Important Considerations in Recruiting ............................................................................... 119 Figure 5.8: Time to Fill Vacancy for Professionals and Skilled Production Workers ............................................... 120 Figure 5.9: Causes for Vacancies as Perceived by Managers .................................................................................... 121 Figure 5.10: Hourly Wage by Skills and the Highest Degree Completed ................................................................. 121 Figure 5.11: Mean Log Hourly Wage by Years of Formal Education....................................................................... 123 Figure 5.12: Three Most Critical Skills/Competencies Needed in Keeping Up with the Emerging Technologies, the Panel Sample ............................................................................................................................................................. 124 Figure 5.13: Training Incidence by Firm Size, 1997-2007 ........................................................................................ 125 Figure 6.1: R&D Expenditures and Level of Development....................................................................................... 130 Figure 6.2: Manufacturing Establishments Reporting R&D Expenditures, 2002 and 2007 ...................................... 132 Figure 6.3: Researchers in R&D and Level of Development .................................................................................... 134 Figure 6.4: Technicians in R&D and Level of Development .................................................................................... 134 Figure 6.5: Manufacturing Establishments Employing Staff Exclusively for R&D, 2002 and 2007 ........................ 136 Figure 6.6: Manufacturing Establishments Filing Patents, 2002 and 2007 ............................................................... 138 Figure 6.7: Patents Issued by the United States to Residents of Foreign Countries and Level of Development, 2006 ................................................................................................................................................................................... 140 Figure 6.8: The Technology Spectrum in Malaysia in 2002 and 2007 ...................................................................... 141 Figure 6.9: The Technology Spectrum by Firm Characteristics in 2002 and 2007 ................................................... 142 Figure 6.10: The Technology Spectrum by Region in 2002 and 2007 ...................................................................... 143 Figure 6.11: The Technology Spectrum by Industry in 2002 and 2007 .................................................................... 144 Figure 6.12: High-Technology Exports can be a Misleading Indicator of Technological Performance.................... 145 Figure 6.13: Kernel Density Plots of Manufacturing TCI in 2002 and 2007............................................................. 150 Figure 6.14: Kernel Density Plots of TCI by Region ................................................................................................ 152 Figure 6.15: Kernel Density Plots of TCI by Industry............................................................................................... 153 Figure 6.16: Kernel Density Plots of TCI by Firm Characteristics............................................................................ 154 Boxes Box 1.1: Malaysia's Tax Regime ................................................................................................................................ 18 Box 3.1: What Counts as Services? ............................................................................................................................. 53 Box 6.1: Global Competitiveness Index .................................................................................................................... 128 Annexes ANNEX 1 .................................................................................................................................................................. 162 ANNEX 1.1 POLICY EFFORTS OF THE GOVERNMENT OF MALAYSIA IN IMPROVING INVESTMENT CLIMATE AND MOVING TOWARD KNOWLEDGE ECONOMY (2001-2008) ........ 162 ANNEX 1.2 SAMPLING METHODOLOGY FOR THE PRODUCTIVITY AND INVESTMENT CLIMATE SURVEY II ­ MALAYSIA 2007.................................................................................................. 170 ANNEX 1.3 SAMPLE COVERAGE AND DISTRIBUTION OF PICS I AND PICS II ............................ 173 ANNEX 1.4 SUMMARY STATISTICS OF SAMPLE FIRMS .................................................................... 188 ANNEX 1.5 SUBJECTIVE AND OBJECTIVE INDICATORS................................................................... 202 ANNEX 2 .................................................................................................................................................................. 206 ANNEX 2.1 FIRM PERFORMANCE INDICATORS OF MANUFACTURING FIRMS ........................ 206 ANNEX 2.2 PRODUCTION FUNCTION ESTIMATION WITH LEVINSOHN-PETRIM METHOD .. 212 ANNEX 2.3 PRODUCTION FUNCTION ESTIMATIONS ......................................................................... 217 ANNEX 2.4 CORRELATES BETWEEN FIRM CHARACTERISTICS AND PERFORMANCE/IC INDICATORS ................................................................................................................................................... 218 ANNEX 2.5 CORRELATES BETWEEN FIRM CHARACTERISTICS AND PERFORMANCE .......... 221 ANNEX 2.6 CORRELATES BETWEEN INVESTMENT CLIMATE AND FIRM PERFORMANCE .. 225 ANNEX 2.7 CORRELATES BETWEEN CHANGES IN INVESTMENT CLIMATE AND CHANGES IN FIRM PERFORMANCE.................................................................................................................................. 227 ANNEX 3 .................................................................................................................................................................. 229 ANNEX 3.1 SALES AND EXPORT PERFORMANCE OF SERVICES FIRMS ...................................... 229 ANNEX 3.2 FIRM CHARACTERISTICS AND FIRM PERFORMANCE ................................................ 232 ANNEX 3.3 ROLE OF PRESENCE OF FOREIGN-OWNED FIRMS IN PRODUCTIVITY ................. 233 ANNEX 4 .................................................................................................................................................................. 235 ANNEX 4.1 PICS SAMPLES AT THE REGIONAL LEVEL...................................................................... 235 ANNEX 5 .................................................................................................................................................................. 237 ANNEX 5.1 DESCRIPTION OF PICS EMPLOYER-EMPLOYEE MATCHING DATA ....................... 237 ANNEX 5.2 METHODOLOGY OF WAGE PREMIUM ESTIMATES ..................................................... 238 ANNEX 5.3 SHARE OF SKILLED AND UNSKILLED WORKERS AND THEIR HOURLY WAGE IN THE MANUFACTURING AND BUSINESS SUPPORT SERVICES......................................................... 246 ANNEX 5.3 SHARE OF SKILLED AND UNSKILLED WORKERS AND THEIR HOURLY WAGE IN THE MANUFACTURING AND BUSINESS SUPPORT SERVICES (CONT.) ........................................ 247 ANNEX 6 .................................................................................................................................................................. 248 ANNEX 6.1 CONSTRUCTION OF THE FIRM-LEVEL TECHNOLOGICAL CAPABILITIES INDEX (TCI) ................................................................................................................................................................... 248 ANNEX 6.2 TCI STATISTICS ACROSS FIRMS IN MALAYSIA ............................................................. 253 ANNEX 6.3 CORRELATES BETWEEN FIRM CHARACTERISTICS AND TCI .................................. 257 ANNEX 6.4 CORRELATES BETWEEN TFP AND TCI ............................................................................. 259 EXECUTIVE SUMMARY In the decades prior to the Asian Financial Crisis, the Malaysian economy experienced rapid growth and a significant structural transformation. It went from an economy that relied on agriculture and commodities to one dominated by manufacturing and services. Since then, however, Malaysia's growth has slowed to a level well below its key competitors in Asia, including the large labor-surplus economies of China and India. The economy seems to be caught in a middle-income trap ­ unable to remain competitive as a high-volume, low-cost producer and unable to move up the value chain and achieve rapid growth by breaking into fast growing markets for knowledge- and innovation-based products and services. The Malaysian authorities have expressed their commitment to regain their earlier growth and reposition their economy as a rapidly growing, knowledge-based, high value-added, and high income economy. A key element of their strategy is to encourage Malaysians to invest more of their savings at home ­ instead of abroad.1 Equally important is the need to improve the quality of that investment. As part of this effort, the Economic Planning Unit of the Prime Minister's Department launched a second Malaysia Productivity and Investment Climate Survey in 2007 (PICS-II) to assess whether and how the investment environment had changed since the first survey in 2002 (PICS-I). This report presents the analytical results of the second survey, which covers nine manufacturing industries and five selected business support services industries. Not surprisingly, the main findings ­ first reported to the authorities in May 2008 -- are similar to the results of the first survey: · Malaysia's investment climate compares favorably with other countries at similar levels of per capita income. The PICS, which reports firms' perceptions of the business environment, suggests that Malaysia is a relatively attractive place for investors for several reasons, including access to financing and efficient customs clearance. This observation is confirmed by other sources such as the Doing Business report, which provides de jure measures of the formal regulatory environment firms' face. For example, in 2009, the Doing Business report ranked Malaysia 20th out of 178 economies for "ease of doing business," up from 25th place the previous year. Similarly, for economic competitiveness, the Global Competitiveness report in 2008-09 placed Malaysia 21st out of 134 economies. · Based on the perceptions of participating firms, the results from PICS-II show virtually no change in the ordinal ranking of the top constraints to doing business from PICS-I. Firms continue to believe that a shortage of skills is the top obstacle: about 40 percent of participating firms reported this as one of their top three constraints. They also continue 1 Malaysia's significant external current account surpluses over the last decade reflect an excess of domestic savings over investment. to consider macroeconomic instability and economic policy uncertainty as major concerns; approximately 20 to 25 percent of firms perceived these factors as severe or very severe obstacles. · Unlike in PICS-I, however, PICS-II suggests deterioration in some aspects of the business environment, most importantly, crime and theft have emerged as big problems. The percentage of firms that perceived crime and theft as a severe or very severe obstacle doubled; it went from less than 10 percent in 2002 to about 20 percent in 2007. Anti- competitive practices have also become a major constraint for business support services firms. The percentage of services firms that perceived it as a severe or very severe obstacle increased from 14 percent to 20 percent. · The remaining indicators show small movements ­ in different directions. For example, aspects of infrastructure and the regulatory framework show improvement, such as the percentage of manufacturing firms that own or share generators and the average number of days that senior managers' spent in dealing with regulatory agencies. On the other hand, innovation and technological capabilities seem to have deteriorated. The breadth and detail of PICS-II provide data that can be productively analyzed for years to come. This report, comprehensive and detailed as it may be, should be considered a first installment in this effort. Even so, some useful insights have emerged from the patterns in the data that warrant further inquiry and understanding. Below are three examples of such patterns (among several in the report) which are worth highlighting: · Firm characteristics are often a good predictor of firm performance. For example, in the manufacturing sector, large, foreign-owned, exporting firms, firms using a large share of computer-controlled machines or engaging in R&D activities tend to have higher labor productivity and total factor productivity (TFP) than others. Not surprisingly, firms facing skilled labor shortages, regulatory burdens, inadequate infrastructure as well as crime and theft often suffer from low labor productivity and TFP. Within the selected business support services sector, the presence of foreign firms, especially those that are not constrained by the 30 percent foreign ownership restrictions, have beneficial effects on the performance of domestic firms. This pattern is consistent with the experience of other countries. · There is considerable evidence that firms find it difficult to locate and recruit the skills they seek. More than 40 percent of firms have reported vacancies of skilled production workers, and the average time to fill a vacancy is about four weeks. Out of desperation, firms often hire people who do not have the appropriate skills for the job. PICS-II shows that many workers lack the appropriate level of education for their jobs or their skills don't match what they were hired to do. Approximately 25 percent of workers with a high-school certificate felt they needed a university education to do their jobs properly. Only 10-15 percent of workers believed their chosen field of education suited their current job; likewise, more than 15 percent believed their educational qualifications were irrelevant to their current occupation. ii · Although the survey shows the expected positive association between technological capabilities and TFP growth, a shortage of skills tends to limit firms' technological capabilities. The supply of researchers and technicians in Malaysia, although increasing, still falls short of the level consistent with the country's income level. Malaysian manufacturing firms continue to function more as "adaptors" rather than "creators" of new technologies. Links between firms and universities are weak, inhibiting the commercialization of new technologies and the transmittal of innovative ideas. Just as the PICS-II data reveal patterns that require further investigation and validation, the data also shed light on several possible policy issues that require further investigation to determine their possible causes before designing an appropriate policy response. A selected few such policy issues that could be of potential interest to senior policy makers are given below: · Why has labor productivity in Malaysias' manufacturing and services sectors not grown faster, despite an international ranking that shows its overall investment climate has been reasonably favorable? Growth in labor productivity in the manufacturing sector declined from 5.5 percent per year during PICS-I to 2.2 percent during PICS-II; overall, the services sector has stagnated since the early 1990s. · Why have crime and theft become big issues? Manufacturing firms estimated their losses from theft, robbery or vandalism have almost doubled from 0.5 percent of total sales in PICS-I to 0.9 percent in PICS-II. · Why has the perception of skills of workers in the manufacturing sector improved while that of workers in the services sector has worsened? From PICS-I to PICS-II, manufacturing firms that perceived a shortage of skills as a severe or very severe constraint declined by 5 percent. On the other hand, business support services firms that shared the same perception increased by 8 percent. · Why have skill shortages remained the top constraint perceived by some 40 percent of firms despite the fact that the number of new college graduates increased by 70 percent from 2000 to 2005? On a related matter, why has the skills premium for college graduates declined -- even though Malaysian firms consistently report job skills are in short supply? The wage premium of college graduates declined more than 25 percent in the manufacturing sector and more than 10 percent in the business support services sector. · Why have Malaysia's overall technological capabilities stagnated even as R&D expenditures and the supply of researchers and technicians increased? From 2000 to 2004, R&D expenditures as a percentage of GDP increased, even if only slightly, from 0.5 percent to 0.6 percent. During the same period, the number of researchers and technicians increased significantly from 279 to 509 per million people and from 40 to 64 per million people, respectively. Yet, roughly one-third of all manufacturing firms failed to upgrade their machinery or product lines or file a patent -- with little change in this share reported between PICS I and PICS II. iii CHAPTER 1. MALAYSIA'S INVESTMENT CLIMATE TODAY 1. The Malaysian economy has successfully transformed itself from one primarily depending on agriculture and commodity production to one dominated by manufacturing and services. The next challenge the country faces is averting the "middle-income trap." Improving the investment climate -- the opportunities and incentives for firms to invest productively, create jobs, and expand -- is the key to sustainable progress in continuously improving living standards and moving toward a more knowledge-based economy. 2. This chapter provides updated information on Malaysia's investment climate.2 It is based on the second round of the Malaysia Productivity and Investment Climate Survey (PICS II) carried out in 2007, which covers nine manufacturing industries and five selected business support services industries. The PICS II results, when compared with those obtained during the first round in 2002, provide an interesting perspective on the evolution of the investment climate. The main findings are: · From an international perspective, Malaysia's investment climate compares favorably with countries having a similar level of income. The PICS, which reports firms' perceptions of the business environment, suggests that Malaysia is a relatively attractive place for investors for several reasons, including access to financing and efficient customs clearance. This observation is confirmed by other sources such as the Doing Business report, which provides de jure measures of the formal regulatory requirements firms face. For example, in 2009, the Doing Business report ranked Malaysia 20th out of 178 economies for "ease of doing business," up from 25th place the previous year. Similarly, for economic competitiveness in 2008-09, the Global Competitiveness report placed Malaysia 21st out of 134 economies. · Based on subjective indicators, the ordinal ranking to top constraints is virtually unchanged: a shortage of skills remains the top obstacle. About 40 percent of firms perceived shortage of skills as one of their top three constraints to doing business. Firms also continue to regard macroeconomic instability and economic policy uncertainty as major concerns. Approximately 20 to 25 percent of firms perceived them as severe or very severe obstacles. Crime and theft have become a big problem. In 2007 about 20 percent of firms perceived them as a severe or very severe obstacle, up from less than 10 percent in 2002. Anti-competitive practices have also become a major constraint on business support services firms. The percentage of services firms that perceived it as a severe or very severe obstacle increased by 50 percent from 12 percent to 20 percent. · Based on reports by firms on specific objective indicators, Malaysia's investment climate has changed little over the past five years. Two aspects on infrastructure and regulatory framework have improved, namely the percentage of manufacturing firms that 2 The impact of the recent global crisis is not captured. 1 own or share generators and the average number of days that senior managers spend on dealing with regulative agencies. What has become worrisome are: crime and theft and trends in innovation and technological capabilities. 3. To achieve its ambitious Vision 2020, it is important for the economy to move in a trajectory toward high growth and high income. The lack of skills and high vacancy rates persist. Improving skills and strengthening technological capabilities are critical to transform Malaysia into a knowledge-based economy; the same is true for enhancing productivity of the existing stock of talent. 4. This chapter is structured as follows: Section 1 reviews the growth, investment, and productivity in Malaysia; Section 2 presents information about the two rounds of PICS in 2002 and 2007; Section 3 examines the changes in investment climate in the past five years for both manufacturing and business support services sectors. It is based on subjective and objective measures; Section 4 benchmarks Malaysia's investment climate from an international perspective and discusses key constraints; Section 5 concludes GROWTH, INVESTMENT AND PRODUCTIVITY IN MALAYSIA 5. Malaysia's economy recovered strongly from the Asian financial crisis in the late 1990s. Foreign reserves increased and the financial sector became less vulnerable to external financial shocks. Malaysia has a significant external current account surplus over the last decade. In 1998-2007, the external balance on goods and services represents in average over 20 percent of GDP. However, the Malaysian economy is still vulnerable to global demands due to its high dependence on international trade, which represents more than 200 percent of GDP. Half of Malaysia's total exports are for electronics and electrical goods. After the Asian financial crisis, the recovery of Malaysian economy benefited from the surge of global demand for information and communication technology products. However, in 2001, when that demand declined, the economy suffered. GDP growth slowed and exports plunged. Economic growth resumed in 2002. From 2003-2007, GDP grew at about 5.7 percent and at 3.9 percent per annum per capita. However, GDP remained significantly below the pre-crisis periods of 8.9 percent and 6.0 percent in 1987-1997, respectively (Figure 1.1). 2 Figure 1.1: Annual growth of GDP and GDP per capita in Malaysia (1987-2007) Source: World Development Indicators 6. In recent years, the combined output of agriculture, forestry and fisheries has expanded even as the relative importance of these rural-based sectors of the economy has continued to decline. This is consistent with other economies in transition. Industry and services are two driving forces of economic growth. In 2007, industry value added represented 30.1 percent of GDP and services value added represented 53.6 percent. In comparison, the importance of agriculture declined to 7.6 percent in 2007. Manufacturing is dominated by export-oriented industries ­ they account for around three-quarters of total manufacturing output. The domestic-oriented sector on the other hand represents only one-quarter. The latter consists mainly of the fabrication of metal products, non-metallic mineral products, food products and transport equipment. 7. Slower growth in output per work was the chief cause of the decline in output growth in 1998-2007 compared with 1987-1997 (Table 1.1). On average, it dropped by 2.6 percentage points from 5.5 percent to 2.9 percent. The slower expansion of employment accounted for about one third of the slower output growth. The slowdown in labor productivity is mainly attributed to the slower accumulation of capital per worker. Both domestic and foreign direct investment declined after the financial crisis (Figure 1.3). Gross domestic investment hovered in the range of 20-25 percent of GDP after 1998; net foreign direct investment ranged from 2-4 percent with the most significant drop to 0.6 percent in 2001. About one-sixth of growth came from the increased use of capital goods in the post crisis period compared to more than one-third before the crisis in 1987-1997. The contribution of improvement in education and total factor productivity remained unchanged at 0.3 percent and 1.6 percent, respectively. More than 50 percent of the growth in labor productivity in the post-crisis period may be attributed to total factor productivity improvement compared to only one-third in the pre-crisis period. However, an average gain of 0.9 percent per year of total factor productivity in 1987-2007 indicates a significant loss during the crisis period. The decline in labor productivity is more significant in the services sector. Understanding the reasons why investment growth stagnated and what factors may help boost the confidence of investors are important for stimulating economic growth and enhancing business productivity. 3 Figure 1.2: Sectoral composition of GDP in Malaysia (1998-2007) Source: The Economic Planning Unit of Malaysia (2009). Table 1.1: Sources of Growth, Total Economy: 1987-2007 (Annual percentage rate of change) 1987-2007 1987-1997 1998-2007 Output 6.8 9.4 5.7 Employment 2.7 3.7 2.0 Output per worker 3.7 5.5 2.9 Contribution of: Capital 2.5 3.4 1.0 Education 0.3 0.3 0.3 Land 0.0 0.0 -0.1 Factor Productivity 0.9 1.7 1.6 Source: cited from the Economic Planning Unit of Malaysia and the World Bank (2008), "Measuring the contribution to GDP and Productivity of the Malaysia Services Sector". Figure 1.3: Domestic and foreign direct investment (1987-2006/7) Source: World Development Indicators 4 8. The Government of Malaysia is aware of concerns about its investment climate and the importance of moving toward a more knowledge-based economy. Active and continuous efforts have been in place in recent years.3 The Second Industrial Master Plan (1996-2005) aimed at promoting productivity of entire clusters of the Malaysian economy. It identified ICT and biotechnology as key growth areas. The Eighth Malaysia Plan (2001-2005) emphasized the need "to enhance the competitiveness of the economy and strengthen economic resilience, concerted efforts will be undertaken to improve total factor productivity as well as facilitate the development of a knowledge-based economy (EPU, 2001: 3)". Under the Third Industrial Master Plan, the Government released the New Economic Plan in 2006, recognizing the "more challenging and competitive global environment." The goal of the Ninth Malaysia Plan (2006- 2010) is to "upscale the manufacturing sector toward higher value added activities and upgrade capacity in the provision of related services. (EPU, 2006: 107)." Greater emphasis is "placed on promoting investment in new areas of growth as well as reinforcing innovation capacity and capacity to augment productivity and competitiveness." To reduce bureaucracy, a special task force, PEMUDAH, was established in 2007 to improve business competitiveness.4 Various measures have enhanced the efficiency of the public services delivery system. PEMUDAH has made progress in reducing impediments to business. They include reducing the time for clearance of exports, facilitating e-payment, developing an online licensing system, improving the employment process for expatriates. Malaysia Industrial Development Authority (MIDA) is also tasked to create a vibrant business environment. PICS 2002 AND PICS 2007 9. The Government of Malaysia and the World Bank jointly conducted Malaysia Productivity and Investment Climate Survey (PICS) to understand the investment climate in Malaysia and how it affects business performance. The goal was to improve the business environment. The Department of Statistics and the Economic Planning Unit carried out the survey in collaboration with the World Bank PICS team. Annex 1.2 provides detailed information on the sampling methodology. The first round (PICS I) was fielded between December 2002 to May 2003; the second round (PICS II) between October 2007 and January 2008. The two rounds share a similar survey framework. · The PICS 2007 covers nine industries in the manufacturing sectors, including: food processing; textiles; garments; wood & furniture; chemical & chemical products; rubber & plastics; machinery & equipment; electrical machinery and electronics (equipment & components); and motor vehicles and parts,5 and 303 firms from five industries of selected business support services sector. The latter includes information technology; telecommunications; advertising & marketing; accounting & related services; and 3 See Annex 1.1 for a summary of policy efforts of the Government of Malaysia. 4 PEMUDAH is the acronym for the taskforce's name in Bahasa Malaysia "Pasukan Petugas Khas Pemudahcara Perniagaan". 5 PICS 2002 covers nine industries in the manufacturing sector, including food processing, textiles, garments, furniture & wood products, chemical & chemical products, rubber & plastics, machinery & equipment, electronics, and auto parts. PICS 2007 added two more industries: office, accounting & computing machinery and electrical appliances. For the purpose of comparison, we regroup the industries in PICS II into nine, the same as PICS I. See Annex 1.3 for details. Five industries in the business support services sector covered in both PICS rounds are the same. 5 business logistics (transportation & related services). In this report, if not otherwise stated, "services sector" is used as a short form to refer to "business support services sector", which covers only these five industries - information technology, telecommunications, advertising and marketing, accounting and related business, and business logistics. The findings are not to be quoted to reflect the services sector as a whole. · The surveyed firms are distributed in nine out of the 13 states: Kuala Lumpur, Selangor, and Melaka (referred to here as the Central Region or Klang Valley); Pulau Pinang and Kedah (referred to here as the North Region); Johor (referred to here as the South Region), Terengganu (referred to here as the East Region), Sabah, and Sarawak. Annex 1.3 provides detailed information on sample coverage and distribution. The survey analysis at the regional level is based on selected firms in the respective regions covered by the survey. The results cannot be generalized to the entire region. For the manufacturing sector, only establishments with more than 10 employees are covered. For the business support services sector, two employment thresholds are used. Only establishments with more than 10 employees are covered for Information Technology, Telecommunications, and Advertising and Marketing, while only establishments with more than 20 employees are covered for Accounting and Related Services and Business Logistics. PICS 2002 surveyed 902 firms in the manufacturing sector and 249 firms in selected business sector; PICS 2007 surveyed 1,115 firms in the manufacturing sector and 303 in selected business support services sector. Four hundred eighty-eight manufacturing firms and 137 services firms participated in both survey rounds.6 Both rounds include interviews with CEOs, chief financial officers, human resource managers, and workers. 6 Overall, for manufacturing sector and business support services sector, the distribution of firms surveyed in both rounds is similar to that of firms in full samples of PICS 2002 and PICS 2007 across regions, firm sizes, ownerships, and export-orientations. The inclusion of two new industries in manufacturing sector in PICS 2007 results in the difference of distribution between panel firms and PICS 2007 firms across industries. For services sector, the distribution of panel firms and that of firms in full samples of both survey rounds are similar also from the perspective of industries. 6 10. PICS 2002 and PICS 2007 provide subjective assessments by firm managers and objective measures of various aspects of the investment climate, as well as information on corporate finance for 1999-2001 and 2004-2006, respectively. The two rounds of survey data allow an examination of the variation of investment climate across and within industries, regions, firms of different sizes, ownerships, and export-orientations, and the impact of the investment climate on manufacturing firm productivity. Annex 1.4 provides details on these breakdowns. A comparison of the results of PICS 2002 and PICS 2007 based on the two full samples sheds light on the differential effects of business climate on firm performance over time.7 Additional rounds of PICS would provide further useful information to explore the linkages between changes in investment climate and changes in firm performance on a sounder statistics basis and thus offer more solid support to help policy-makers improve investment climate and stimulate growth more effectively. 11. This report mainly draws on the findings from PICS 2007. The study is complemented with comparable PICS/Enterprise Surveys in other countries as well as two other global databases including: Doing Business, which uses a case-study approach with a focus on regulatory regimes; and the Global Competitiveness Report conducted by the World Economic Forum. The latter provides information on competitiveness and productivity of a country based on a set of measures of institutions, policies. RECENT DEVELOPMENTS IN MALAYSIA'S INVESTMENT CLIMATE 12. The PICS provide two types of complementary information of investment climate: subjective and objective. Annex 1.5 provides details. The subjective indicators are key elements for understanding firm managers' perceptions of investment climate issues that influence their investment decision; the objective indicators are an important basis for international comparisons of business environment. At the national level, there are two observations: · Judging by subjective rankings, the four aspects of investment climate ranked among the top constraints of doing business by firms from manufacturing sector as well as from business support services sector in Malaysia remained the same in 2007 as in 2002: skilled labor shortages, tax regulations and high tax rates, lack of business support services, and bureaucratic burdens.8 Judging by subjective rating, firms' perception of skills shortage as one of the biggest problems became more acute in the business support services sector though less so in the manufacturing sector; while firms' perception of tax regulations/taxes rates worsened. In addition, crime and theft became a big problem in the past five years. In 2007, macroeconomic instability and economic policy uncertainty 7 The results on investment climate indicators are similar based on the two full samples as the panel firms that participated in both rounds. We cannot reject the hypotheses that the distribution of the 488 manufacturing firms that participated in both survey rounds is the same as the distribution of the 902 firms surveyed in PICS 2002 at 10 percent significance level across industries, regions, firm sizes, ownerships, and export-orientations. The same distribution hypotheses between the 488 panel firms and the 1115 firms surveyed in PICS 2007 cannot be rejected across regions, firm sizes, ownerships, and export-orientations, either; but can be rejected only across industries. The likely reason of rejection can mainly be related to the inclusion of new industries in the 2007 survey round. For business support services firms, the hypotheses that the distribution of the 249 firms surveyed in PICS 2002, and that of the 303 firms surveyed in PICS 2007, and that of the 137 firms surveyed in both rounds are the same cannot be rejected. 8 See further discussion on tax rate and tax regulations in the following sections. 7 remain among the aspects that the highest percentage of all surveyed firms rated as severe or very severe obstacles. · Judging by objective indicators, some aspects on infrastructure and regulatory framework improved in the past five years. For example, a smaller percentage of manufacturing firms owned or share generators and senior manager's time spent in dealing with regulations decreased. With a sharp increase in supply of college graduates, shortages of skills showed signs of alleviation but the supply of labor with needed skills remained tight. The trends of changes in innovation and technological capabilities and in crimes and thefts seemed however to be worrisome. Subjective Indicators 13. This report uses two sets of statistics ­ ranking and rating ­ in PICS to measure Malaysian firm managers' subjective perceptions of the investment climate. The first set allows ranking investment climate constraints by the percentage of firms that consider a particular issue as one of their top three investment climate constraints to doing business; the question is open-ended. The second set is based on a close-ended question which asks firms to rate on a five-point scale how problematic each of the 18 dimensions of the investment climate is for the operation and growth of their business. 9 Perceptions of the Three Top Constraints to Doing Business 14. The first type of indicators, ranking statistics, shows that the four obstacles ranked most important in 2007 remained unchanged from 2002. These top constraints include skilled labor shortages, tax regulations and high tax rates, lack of business support services, and bureaucratic burden (Figure 1.4).10 The reasons why these four aspects of investment climate remained big constraints are interrelated. As the economy moved from more natural resources to manufacturing and services, the shortage of skilled labor and lack of business support services surfaced. The burdens of tax regulations and bureaucracy relate to a shortage of skilled domestic and foreign professionals who can provide expertise on taxation and other regulatory issues. The shortage of skills may also be a reason for insufficient business support services. 9 In this question, zero stands for not an obstacle, one for minor obstacle, two for moderate obstacle, three for major obstacle, four for severe obstacle, and NA means not applicable. In the Malaysia PICS surveys, the 18 dimensions of investment climate listed in the close-ended rating question are not identical as those in the open-ended ranking question. The results of these two questions are not directly comparable. 10 Given the method of computation of this statistic, the numbers provide information on what firms consider as their most pressing investment constraints in the survey year; but the difference between 2002 and 2007 does not record an absolute increase or decrease in severity of an indicated issue. 8 Figure 1.4: Top Obstacles to Doing Business Source: Malaysia PICS 2002 and 2007. Note: Percent of firms identifying indicated problem as one of top three concerns. · The shortage of skilled labor was perceived as a top obstacle to doing business by the largest percentage of firms in both PICS 2002 and PICS 2007. In manufacturing sector, the percentage of firms that considered it as one of their top three constraints remained virtually unchanged (44.3 percent in 2002 and 43.6 percent in 2007); in business support services sector, it increased from 33.3 percent to 38.6 percent. · Next, tax regulations/tax rates, which were perceived as a top obstacle by the second largest percentage of firms in both survey rounds. In both manufacturing and business support services sectors, the percentage of firms that considered it as one of their top three constraints increased sharply over time, from 22.3 percent in 2002 to 31.7 percent in 2007 for manufacturing, and from 17.1 percent to 30.2 percent for services. · Lack of business support services and bureaucratic burdens remained as big constraints, considered one of their top three constraints by the third and fourth largest percentage of firms in both rounds of the survey. The perception of both worsened over time, though in a smaller magnitude compared with the perception on tax regulations / tax rates. Perception of the Severity of Investment Climate Constraints 15. The second type of indicators, rating statistics, offers an overview of the percentage of firms rating a particular constraint as "major" or "severe" in PICS 2002 and PICS 2007, respectively.11 This complements the data for ranking an obstacle by presenting firms 11 An investment climate issue is considered as problematic if it is perceived as a "major" or "severe" obstacle. 9 perceptions on how a specific aspect of investment climate, which may or may not be among the several biggest constraints, affects the operation and growth of their business. The 18 dimensions12 surveyed in both rounds show the severity of investment climate issues in the manufacturing and services sectors from six main perspectives ­ skills, tax regulations and tax rates, regulatory framework (customs, judiciary, and regulations), infrastructure, access to resources (finance and land), and macro framework, corruption, crime and theft. 16. For manufacturing firms (Figure 1.5): · The shortage of skills was perceived as a "major" or "severe" obstacle by 25 percent of firms in PICS 2002 and 20 percent in PICS 2007. This suggests that skills remain a key obstacle for doing business in Malaysia even though the situation has improved to some extent. In 2007, across regions, firms located in the peninsula are more likely to perceive shortage of skills a major constraint compared with those located in Sabah and Sarawak; across industries, shortage of skills is the most acute in firms producing textiles and E&E and the least constrained in firms producing auto-parts. Shortage of skills is more binding for large firms and exporting firms. There is no significant difference in perception of skills between foreign-owned firms and domestic-owned firms. · Tax rates were perceived by one out of five firms as problematic both in 2002 and 2007. The worsening of perceptions about tax administration ­ 17 percent of firms that perceived it as major or severe constraints in 2007 compared to 13 percent in 2002 ­ may be the key factors that led to the worsening of perceptions about tax issues. · Firms' perceptions on customs and regulations improved slightly, while those on anti- corruption practices worsened somewhat. The percentages of firms perceived customs and trade regulation administration, labor regulations, business licensing and registration, and anti-corruption practices as major or severe obstacles were 16 percent, 15 percent, 11 percent and 14 percent, respectively, in PICS 2007 compared to 15 percent, 13 percent, 10 percent and 16 percent in PICS 2002. · On infrastructure, about 15 percent of firms perceived electricity as a major or severe obstacle in both survey rounds; roughly half as many firms perceived telecommunication in the same way. The percentage of firms that perceived transportation as a major or severe obstacle slightly declined from 12 percent in 2002 to 11 percent in 2007. · There seemed to be a slight improvement in perceptions on access to (domestic and foreign) credit and to land over time, although the percentage of firms considering the cost of financing a major or severe obstacle increased slightly from 18 to 20 percent. · Partly related to the changes in global macroeconomic environment, one out of four firms considered macroeconomic instability a major or severe obstacle in both 2002 and 2007. The percentage of firms considering the uncertainty of economic policy as major or 12 In addition to these 18 dimension, PICS 2007 asked firms to rate how severe immigration was perceived as an obstacle for their operation and growth. 8.6 percent of manufacturing firms and 4.4 percent of services firms considered it a major or severe obstacle. 10 severe obstacles remained pronounced, though it declined by 2 percentage points from 22 percent to 20 percent. The perception of corruption remained unchanged ­ some 14- 16 percent of firms considered it a major or severe obstacle. There seemed to be a significant worsening of perceptions on crime, theft, and disorder; 20 percent of firms considered this a major or severe obstacle in 2007 compared to 11 percent in 2002. Figure 1.5: Percentages of Manufacturing Firms Perceiving Particular Investment Climate Indicators Major or Severe Obstacles to Doing Business in Malaysia Source: Malaysia PICS 2002 and 2007. 17. For services firms (Figure 1.6): · The perception on skills shortages significantly worsened ­ the percentage of firms considering it a major or severe obstacle jumped from 12 percent in 2002 to 20 percent in 2007. In 2007, across regions, firms located in the Central Region and Sarawak are more likely to perceive a shortage of skills a major constraint; across industries, shortage of skills is the most acute in IT and advertising firms and the least constrained in communication and business logistics firms. Shortage of skills is more binding for exporting firms and firms with more than 30 percent of foreign ownership. There is no significant difference in perception of skills between firms of different sizes.13 · The percentages of firms considering tax rates and tax administration major or severe obstacles increased slightly, from 18 percent to 20 percent and from 11 percent to 13 percent respectively. 13 The results with breakdowns at the regional, industry, and firm characteristics levels need to be interpreted with caution due to the limited number of observations. 11 · Changes in firms' perceptions on customs and trade regulation administration, labor regulations, and business licensing and registration were small and mixed; firms' perceptions on anti-competitive practices need to be improved ­ 20 percent of firms considered it a major or severe obstacle in 2007 compared to 14 percent in 2002. · On infrastructure, firms' perceptions on telecommunication and transportation remained unchanged ­ some 8-9 percent of firms perceived them as problematic; while those on electricity worsened ­ 7 percent of firms considered it a major or severe obstacle in 2007 compared to 2 percent in 2002. 14 · There seemed to be a general improvement in firms' perceptions about access to financing and the cost of financing. Only 10 percent of firms considered access to domestic credit and 8 percent access to foreign credit as problematic in 2007 compared to 18 percent and 9 percent in 2002, respectively. Access to land, which was considered a major or severe obstacle by 8 percent of firms in 2007 compared to 5 percent in 2002, remained not a minor concern though there seemed to be a slight worsening in perception. · Firms' perceptions on the four indicators closely related to the macro, political, and administrative sphere seemed to worsen over time. The percentage of firms considering macroeconomic instability problematic remained at some 24 percent; those considering the uncertainty of economic policy a major obstacle increased from 23 percent to 25 percent. More firms considered corruption and crime, theft, and disorder problematic in 2007 compared to 2002. Figure 1.6: Percentages of Services Firms Perceiving Particular Investment Climate Indicators Major or Severe Obstacles to Doing Business in Malaysia Source: Malaysia PICS 2002 and 2007. 14 The worsening of the perceptions may be partly related to the global oil crisis. For example, the expenses on electricity per unit of sales increased by 12 percent over 2004-2006. 12 18. Overall, the uncertainty of economic policy and macroeconomic instability may affect firms' general sentiments and perceptions of other specific aspects of the investment climate. Skills shortages, followed by tax issues, are ranked as the biggest obstacles. Except for a few indicators, changes in perceptions were small and the directions were mixed for both sectors over the past five years. Among the bigger changes, for manufacturing firms, perceptions on skills and education of available workers, economic policy uncertainty, and access to domestic credit improved while perceptions on macroeconomic instability, tax administration, and crime, theft and disorder worsened; for services firms, perceptions on access to domestic credit, cost of financing, and business licensing and registration improved while perceptions about anti-competitive practices, skills and education of available workers, and crime, theft and disorder worsened (Figure 1.7 and Figure 1.8). Figure 1.7: Change in the Percentage of Firms Rating Indicated Constraint as "Severe" or "Very Severe" over the Period 2002 to 2007 (Manufacturing) Source: Malaysia PICS 2002 and 2007. Note: A negative sign indicates improvement in perception; while a positive sign indicates worsening. Figure 1.8: Change in the Percentage of Firms Rating Indicated Constraint as "Severe" or "Very Severe" over the Period 2002 to 2007 (Services) Source: Malaysia PICS 2002 and 2007. Note: A negative sign indicates improvement in perception; while a positive sign indicates worsening. Objective Indicators 19. From an objective perspective the PICS also provides useful information on a wide array of investment climate issues. These include measures of physical infrastructure (for 13 example, yearly number of power outages), access to financing (collateral as percentage of loan value), of regulatory framework (number of licenses/permits applied), of taxation (level of corporate tax rate), of labor market efficiency (number of weeks to fill a vacancy for skilled production worker), of innovation/technological capabilities (percentage of firms with R&D staff), as well as measures of crime (loss from theft, robbery, or vandalism as percentage of sales). Infrastructure 20. Firms in the manufacturing sector reported improvements in some aspects of infrastructure. Electricity and telecommunications services improved, while changes in water and transportation services were small overall (Figure 1.9). · Electricity ­ The percentage of manufacturing firms that own or share generators decreased from 23 percent to 16 percent; few services firms owned or shared generators. The number of days for manufacturing firms to arrange for an electrical hookup declined from 14 days to 9 days. Loss of production due to power outages increased from 2 percent to 3 percent of sales for manufacturing firms while improved from 1.2 percent to 0.6 percent for services firms. The quality of the power supply needs improvement. For manufacturing firms, the yearly number of power outages, namely nine, remained virtually unchanged although the duration of outages declined from 3 to 2 hours. Among services firms, both indicators improved, from 8 to 5 times and from 2 hours to 1 hour, respectively. · Water supply ­ Changes in water supply services were small in magnitude. The average number of insufficiencies is 4 percent for manufacturing firms and 2 percent for services firms. The average number of days to obtain water connection dropped from 11 days to 8 days for manufacturing firms; it remained at 2 for services firms. · Telecommunications ­ Manufacturing firms reported an improvement in telecommunication infrastructure in 2002-2006. The average duration of these interruptions shortened from 6 hours to 1 hour; the number of days to fix a phone connection shortened from 9 days to 8 days; however, the yearly number of fixed phone interruptions remained unchanged at 3. The numbers reported by services firms remained unchanged at 1 hour, 2-4 times per year and 8 days respectively. · Transportation ­ Changes over time were limited in transportation facility. On average, firms in manufacturing as well as services sectors reported 1-2 services disruptions a year; the loss of production was about 0.5 percent of total sales. 14 Figure 1.9: Objective Indicators of Investment Climate in PICS I and PICS II ­ Infrastructure Source: Malaysia PICS 2002 and 2007. Access to Finance 21. Collateral requirement sharply declined for manufacturing firms. In 2002-2007 manufacturing firms reported a sharp decline in the amount of collateral required to qualify for a loan, which indicated that credit was easing. However, the percentage of firms with bank loans and that with overdraft facility remained unchanged or slightly declined (Figure 1.10).15 · Collateral requirement as percent of loan value ­ The amount of collateral that manufacturing firm had to put down dropped in half, from about 80 percent to 40 percent. Services firms had to put down a similar percentage, namely 40 percent in 2007, the number for 2002 is not available. 15 The impact of the recent financial crisis on access to finance is not captured by PICS 2007. 15 · Bank loan ­ The percentage of firms with a term bank loan remained unchanged, about 60 percent for manufacturing and 50 percent for the services sector. · Overdraft facility ­ The percentage of firms with overdraft facility declined, from 72 to 64 percent for manufacturing firms and from 62 percent to 54 percent for services firms. On the other hand, the percentage that used the overdraft facility declined for manufacturing firms but remained unchanged for services firms. Figure 1.10: Objective Indicators of Investment Climate in PICS I and PICS II -- Access to Finance Regulatory Framework 22. The regulatory framework improved to a large extent in 2002-2007. It took firms less time to obtain licenses. As a result, fewer firms hired consultants to file requests. Senior management spent less time dealing with regulatory agencies. Export and import customs clearance, efficient as it is, showed little improvement (Figure 1.11). · Licenses ­ Both manufacturing and services firms reported it took less time to obtain several types of licenses. For example, among manufacturing firms, the average number of days it took to obtain licenses from the federal government shortened from 6 to 5 days and from 6 to 3 days from the state government. The reported number of days to obtain import permit, construction approval and operating license improved significantly for manufacturing firms and for services firms. The total number of licenses manufacturing firms applied for decreased, while those of services firms remained unchanged. · Consultants for regulation ­ The percentage of both manufacturing and services firms that hired consultants to smooth permits, licenses and approvals decreased from 23 percent to 14 percent and from 19 percent to 9 percent, respectively. · Senior managers' time ­ The average number of days that senior managers spent in dealing with regulatory agencies, such as tax offices, labor and social security offices, fire department and department of occupational safety and health, decreased from 16 12 days to 8 days for manufacturing firms and 14 days to 7 days for services firms. Inspectors from regulatory agencies visited firms once or twice a year. In total, the cost of dealing with regulations as a percent of total sales remained about 0.1- 0.2 percent. · Customs ­ It took manufacturing firms slightly longer to get export and import clearances, although it remained low. In 2002, it took 2 days to clear exports and 3 days to clear imports, but it increased to 3 days and 4 days, respectively, in 2007. Figure 1.11: Objective Indicators of Investment Climate in PICS I and PICS II - Regulatory Framework 17 Tax Rate and Tax Administration 23. Contrary to the popular perception that taxation had worsened in 2002 - 2007, tax rates and tax administration did not change for the worse.16 The recent tax reform, conducted after PICS II in 2008, further planned to reduce tax rate and streamline the tax administration (Figure 1.12). Box 1.1: Malaysia's Tax Regime In Malaysia, company income tax had been fixed at a flat rate of 28 percent. In 2007, the government cut the corporate tax by two percentage points from 28 percent to 26 percent. Firms are subject to a flat tax rate of 26 percent as of the Fiscal Year 2008 imposed on income derived from Malaysia, but not on income received in Malaysia from outside. Companies with paid-up capital of RM2.5 million and below are subject to a company tax of 20 percent on chargeable income of up to RM500,000. The effective tax rate is lower due to exemptions. Sales tax is imposed at the import and manufacturing levels except petroleum products. The general rate of sales tax is 10 percent for all goods; 5 percent for fruits, certain food stuffs and building materials; and exempted for basic food-staffs, basic building materials, certain agricultural implements and machinery, computers and etc. Sales tax rates are also exempted for certain tourist and sports goods, books, newspaper, other reading materials and all exports. Manufacturers with sale values of taxable goods not exceeding RM 100,000 per annum are exempted from sale tax. A foreign fund management company is given a concessionary tax treatment. Income arising from services rendered by a foreign fund management company to clients outside Malaysia will be subject to tax at concessionary rate of 10 per cent, while income arising from services to local clients will be subject to the prevailing corporate tax. As a part of efforts to simplify and enhance the efficiency of the tax system, the Malaysian government is moving toward a single-tier tax system. Effective on January, 2008, tax on a company's profits is a final tax and dividends distributed to shareholders will be no longer taxed. Source: "Summary of Tax System 2008", Treasury Malaysia, Ministry of Finance. 17 · Tax rates ­ According to firms' corporate financing information from PICS, changes in total tax payments as a percent of total revenue, corporate taxes as a percent of profit (total revenue minus total cost), and sales tax as percent of total sales were limited in 2002-2007.18 · Tax administration ­ The number of days firm managers spent in contact with the Inland Revenue Board (IRB) to comply with tax regulations declined. It dropped from 5.2 days to 2.0 days and from 5.8 days to 2.1 days, respectively, for manufacturing firms and services firm. Due to a concessionary treatment of a foreign fund management company, 16 Explaining the reasons why perception on taxation worsened while changes in objective indicators were limited would be an interesting topic for further study. PICS does not provide enough explanation although it indicates some useful hints. For example, subjective perception may have worsened temporarily in the transition period because Malaysia is moving toward the Single-tier Tax System while firms need to adjust their routine regarding tax administration; or the worsening of subjective perceptions may be explained partly by relative improvement of other climate investment issues. Firms' perception on taxation burden may include some hidden cost that are not always captured explicitly in tax, for example, accounting and tax services may add additional cost to firms, especially to foreign firms, as foreign accounting firms can provide accounting and tax services only through Malaysian affiliates. 17 Source: http://www2.treasury.gov.my/pdf/percukaian/Summary%20of%20Tax%20System%202008.pdf. "Summary of Tax System 2008" Treasury Malaysia, Ministry of Finance. 18 Tax rates are calculated as averages reported by the firms surveyed. Given some firms were exempted from certain kinds of tax; the reported tax rate can be lower than the level imposed. 18 foreign-owned firms tend to report slightly greater number of days spent in dealing with Tax Inspectorate. Average number of days for domestic and some foreign owned firms in the manufacturing sector are 5.6 days and 1.8 days in the PICS 2002 and PICS 2007 respectively, while 4.3 days and 2.6 days for foreign fund management firms. Average number of days are 5.7 days and 2.1 days for domestic firms in the business support sector in the PICS 2002 and PICS 2007, respectively, while 6.0 days and 2.4 days for foreign firms. Figure 1.12: Objective Indicators of Investment Climate in PICS I and PICS II ­ Tax Rate and Tax Administration Supply of Skills 24. The number of workers who have secondary and tertiary education showed some improvement. The share of better educated employees increased in both manufacturing and services sectors. It now takes shorter time to fill a vacancy for skilled labor. That professionals and skilled production workers (Figure 1.13). · Human capital ­ The composition of employees with different educational backgrounds differs from manufacturing and services firms. In the manufacturing sector, the percentage of employees with university degrees increased from 10 to 15 percent, while the percentage of employees with high school degrees remained unchanged; in the services sector, the two aforementioned percentages increased from 47 percent to 54 percent, and decreased from 49 percent to 41 percent, respectively. · Time to fill vacancies ­ It took firms less time to fill a vacancy for professionals and skilled production workers. The average number of weeks to fill a vacancy for professional and for skilled production workers shortened from 5 weeks to 4 weeks for manufacturing firms; and it significantly shortened from 10 weeks to 5 weeks and from 8 weeks to 4 weeks for services firms. 19 Figure 1.13: Objective Indicators of Investment Climate in PICS I and PICS II - Labor Market Innovation/Technological Capabilities 25. Over time, the trends in innovation and technological capabilities have become a source of concern. Manufacturing and services firms have slowed their investment in new machinery and R&D. Furthermore, government incentives have yet to be more effective (Figure 1.14). · Production equipment ­ Although there seemed to be a slight increase (23 percent to 26 percent) in utilization of computer-controlled machines in manufacturing firms, the decline in the percentage of machinery less than 5 years old ­ from 30 percent to 23 percent in manufacturing firms and from 61 percent to 44 percent in services firms ­ indicates a slowdown in capital depreciation, which might be linked to lower investment. · R&D activities ­ The percentages of manufacturing and services firms that hired staffs exclusively for R&D activities declined from 18 percent to 12 percent and from 9 percent to 5 percent, respectively. The percentage of firms that subcontract R&D to other companies remained unchanged ­ 5-7 percent for manufacturing firms and 3- 4 percent for services firms. Only 15-16 percent of Malaysian firms are working with research institutes, such as Standards and Industrial Research Institute of Malaysia (SIRIM), to upgrade their technology. · Government incentives - The percentage of firms receiving government incentives to conduct technological innovation remained low: only 8 percent for manufacturing sector and 1 percent for services sector. 20 Figure 1.14: Objective Indicators of Investment Climate in PICS I and PICS II - Innovation/technological Capabilities Crime and Theft 26. Manufacturing firms experienced an increase in security costs due to criminal activity. These expenses added to the cost and uncertainty of doing business. · Crime and cost of security ­ Manufacturing firms report larger share of sales is lost due to crime and spending on security in 2007. Manufacturing firms report the estimated loss from theft, robbery or vandalism as a percentage of total sales increased from 0.5 percent to 0.9 percent. As a result, the costs of providing security as a percent of total sales increased from 1.1 percent to 2.4 percent. The cost of crime and the security expense for the manufacturing firms is the highest in Sabah ­ estimated losses from crime and cost of providing security account for 1 percent and 3 percent of the total sales there. No significant increase in loss from crime and security cost is observed among services firms. · Dispute settlement through court ­ In case a dispute over a delay or suspension of payment, return shipment or cancellation happens, the percentage of firms resolve disputes with clients slightly decreased from 26 percent to 22 percent for manufacturing firms and increased from 28 percent to 36 percent for services firms. 21 Figure 1.15: Objective Indicators of Investment Climate of Services Firms - Crime, Security and Court MALAYSIA'S INVESTMENT CLIMATE FROM AN INTERNATIONAL PERSPECTIVE 27. The Doing Business indicators, the Global Competitiveness Index and the Enterprise Surveys/PICS, which complement each other, provide a comprehensive picture of Malaysia's investment climate and put it in a global context. 28. Doing Business focuses on regulatory and bureaucratic burdens of businesses. The 2009 Doing Business report ranks Malaysia in 20th place (out of 178 economies) for its overall "ease of doing business," up from 25th place in the 2008 report. The report concludes that Malaysian business environment is relatively favorable, compared to many other East Asian economies, such as South Korea (23rd place), China (83rd place), Vietnam (92nd place), Indonesia (129th place), Philippines (140th place) and Lao PDR (165th place). However, Malaysia still lags behind the best performers, such as Singapore (1st place) and Thailand (13th place), in the region. The Doing Business report measures a set of regulations affecting 10 stages of a business's life.19 Among those 10, in 2008 and 2009, Malaysia is ranked as the best performer out of 181 economies for ease in getting credit and the fourth best in protecting investors. Compared to the better performers, Malaysia has room to improve when it comes to dealing with construction permits, registering property, starting and closing a business, enforcing contracts, and employing workers. In those categories, Malaysia ranked 48th or lower in 2009. 29. The Global Competitiveness Index, conducted by the World Economic Forum, measures economic competitiveness. It focuses on infrastructure, macroeconomic stability, human capital (health and education), technological readiness, market size, and labor market efficiency. The 2009 Global Competitiveness Index places Malaysia in 21st out of 134 countries, the same as in 2008. In comparison China was in (30th place), Thailand (34th place), Indonesia 19 The ten indicators covered include starting a business, dealing with construction permits, employing workers, registering property, getting credit, protecting investors, paying taxes, trading across borders, enforcing contracts and closing a business. The indicators refer to a specific type of business, generally a local limited liability company operating in the largest business city. The findings provide information on the extent of obstacles to doing business, which, however, is not directly comparable with other reports. In addition, PICS are not available in some countries in recent years. We cannot make direct comparisons between Doing Business ranking and PICS results. 22 (55th place), Vietnam (70th place), and Philippines (71st place), but below Singapore (5th place), South Korea (13th place) in the East Asia region (Table 1.2). In the 2009 Global Competitiveness Report, Malaysia ranked high (the 16th out of 134 economies) for the sophistication of its financial markets. Among its strengths: investor protection and legal rights ranked 4th and 8th, respectively. However, challenges remain. Among them are macroeconomic instability, higher education and technological readiness, which ranked 38th, 35th and 34th, respectively. The government budget imbalance (109th), the low rate of secondary enrollment rate (95th) and its less than prevalent rate of mobile/broadband internet (51st) also lowered Malaysia's overall ranking. Table 1.2: Global Competitiveness Index - Rankings of East Asian Countries (Selected Indicators) Financial Market Technological Macroeconomic Higher Education Overall Ranking S ophistication Readiness S tability and Training Rank Country / Economy 5 Singap ore 1 Hong Kong 7 Singap ore 3 Hong Kong 8 Singap ore 9 Jap an 2 Singap ore 10 Hong Kong 4 South Korea 12 South Korea 11 Hong Kong 16 Malaysia 13 South Korea 11 China 13 Taiwan, China 13 South Korea 37 South Korea 15 Taiwan, China 18 Taiwan, China 23 Jap an 17 Taiwan China 42 Jap an 21 Jap an 21 Singap ore 28 Hong Kong 21 Malaysia 49 Thailand 34 Malaysia 37 M ongolia 35 Malaysia 30 China 57 Indonesia 66 Thailand 38 Malaysia 51 Thailand 34 Thailand 70 Philip p ines 77 China 41 Thailand 60 Philip p ines 55 Indonesia 79 Vietnam 78 Philip p ines 53 Philip p ines 64 China 70 Vietnam 58 Taiwan, China 80 Vietnam 70 Vietnam 71 Indonesia 71 Philip p ines 110 M ongolia 88 Indonesia 72 Indonesia 85 M ongolia 100 M ongolia 109 China 101 M ongolia 98 Jap an 98 Vietnam 109 Cambodia 130 Cambodia 123 Cambodia 105 Cambodia 127 Cambodia Source: World Economic Forum. (2008). "Global Competitiveness Report 2008-2009." Retrieved from: http://www.weforum.org/pdf/GCR08/GCR08.pdf. 30. While Doing Business provides useful book values for the country's legal and regulatory environments,20 PICS provides detailed information about the actual experiences of firms. PICS statistics suggest the de facto environment from a broader perspective. Doing Business captures data on the formal regulatory requirements that firms experience; its rankings describe the de jure environment for business in terms of regulations. Although de jure measures are often correlated with de facto measures, there can be gaps between the two. Changes in official requirements may have little impact on the investment climate on the ground. 31. Based on PICS (2007) data, Malaysia's investment climate is benchmarked with the following East Asian countries ­ China (2003), Indonesia (2003), Republic of Korea (2005), Philippines (2003), Thailand (2006), and Vietnam (2005) ­ and some higher middle income countries ­ Turkey (2005), Brazil (2003), and Mexico (2006). The comparison focuses on six 20 Doing Business survey mainly covers formal requirements, e.g. number of procedures, time and costs of compliance. It uses standardized, hypothetical transactions for comparability. A small number of in-country professionals with experience in these types of transactions provide responses in each area (i.e., 1-5 lawyers/accountants per indicator per country). See "Measuring the Ease of Doing Business in APEC: The Impact of Regulatory Reforms", World Bank, Forthcoming. 23 categories: infrastructure, access to finance, regulation framework, taxation, and crime and theft.21 The PICS results suggest that Malaysia is relatively well positioned in terms of export and import customs clearance and access to finance; on the other hand, skill deficiencies require the attention of policymakers. 32. The quality and accessibility of infrastructure services in Malaysia lies in the middle when compared to other countries (Figure 1.16). Electricity disconnections from public grids are more frequent in Malaysia, but the durations are shorter. Due to the higher frequency of blackout, 16 percent of manufacturing firms in Malaysia owned or shared a generator. This is on a par with China, Brazil and Mexico, but the incidence in Malaysia is four times that of Thailand, half the level in Indonesia and Philippines. The level of production lost due to power outages ­ 2.7 percent for manufacturing and 0.6 percent for services ­ are in line with China, Thailand, Turkey, Mexico and Brazil. Malaysia's loss due to power outages is lower than Indonesia (4.3 percent) and the Philippines (7.1 percent). 21 The selection of comparator countries is subject to data availability in most recent years. The objective of this comparison is to provide a gross picture on Malaysia's investment climate from an international perspective. Due to data availability, the indicators presented here from Enterprise Survey are for manufacturing sectors only. The coverage of industries differs at the country level. The results are to be interpreted with precaution due to the limited number of observations in some comparator countries. 24 Figure 1.16: International Comparison of Objective Indicators - Infrastructure Source: PICS surveys 33. The findings indicate that Malaysia is relatively well positioned when it comes to access to domestic financing, but it still lags behind Thailand in some aspects (Figure 1.17). The collateral requirement in Malaysia, roughly 40 percent of the loan value for both sectors, is the lowest among all comparator countries. In Malaysia, a relatively high share of working capital is financed by domestic financial institutes (26 percent for manufacturing and 19 percent for services), compared to, for example, 2 percent in Mexico and 8 percent in Philippines, but still lags behind Thailand (45 percent). In Malaysia, 64 percent of manufacturing firms and 54 percent of services firms have overdraft facilities. The rate is higher than that in China (29 percent), Indonesia (20 percent), and Philippines (31 percent), but lagged behind Thailand (72 percent) and Brazil (74 percent). 25 Figure 1.17: International Comparison of Objective Indicators ­ Access to Domestic Finance 34. Malaysia suffers from a deficit of university graduates (Figure 1.18). Only 15 percent of employees in manufacturing firms have university degrees or a higher level of education, compared to 38 percent in Korea, 35 percent in Indonesia, and 20 percent in Thailand. Two- thirds of employees in the manufacturing sector in Malaysia have a high-school degree. It takes Malaysian firms longer to fill a vacancy for professional workers (4.2 weeks for manufacturing and 5.3 weeks for services) than firms in countries, such as Indonesia (1.4 weeks), Vietnam (2.5 weeks), and Korea (3.5 weeks). 26 Figure 1.18: International Comparison of Objective Indicators ­ Labor Market 35. Among manufacturing firms, Malaysia's export and import customs clearance process is relatively efficient; its bureaucratic burdens are in the middle range when compared to its comparator economies (Figure 1.19). It takes firms less time to have import and export customs clearance in Malaysia than in other comparator countries, except Thailand. The number of days required to obtain approval for construction (42 days for manufacturing and 36 days for services) is longer in Malaysia, compared to Philippines (8 days) and Thailand (25 days). Malaysia in on a par with Indonesia. This is consistent with findings in the Doing Business report. More effort needs to go into streamlining the construction approval process. The average time senior managers spend in dealing with agencies for business regulation (8 percent for manufacturing and 6 percent for service) is relatively long, compared with other East Asian comparators Thailand (0.4 percent), Vietnam (3 percent), Korea (3 percent) and Indonesia (4 percent). 27 Figure 1.19: International Comparison of Objective Indicators ­ Regulatory framework 36. The objective indicators suggest that Malaysia's tax rate is moderate. We used information from Hong Kong Tax Competitiveness Series to provide a cross-country comparison on tax rates; similar data is not available from PICS/Enterprise survey. From an international perspective, the Malaysian corporate income tax rate (27 percent) is moderate. It is slightly higher than Taiwan (25 percent) and lower than South Korea (27.5 percent), Vietnam (28 percent), Indonesia (30 percent), Thailand (30 percent), China (33 percent) and Philippines (35 percent) (Figure 1.21). However, Malaysia's is higher than in the most competitive economies in East Asia, such as Hong Kong (17.5 percent) and Singapore (20 percent). The findings suggest that Malaysia's corporate income tax rate is not high in relative terms. PICS/Enterprise Survey data suggest that the number of days spent for tax inspectorates in Malaysia (2 days for both manufacturing and services sector) are slightly lower than in South Korea (3 days), Philippines (3 days) and Vietnam (3 days), while slightly higher than that in Mexico (1 day) and Indonesia (1 day) (Figure 1.20). The worsening public perception on taxation may stem from something other than the tax rate per se. 28 Figure 1.20: International Comparison of Objective Indicators ­ Tax Administration Figure 1.21: Corporate Income Tax in East Asian Countries (1997-2007) Made from KPMG, (2007), Hong Kong Tax Competitiveness Series: Corporate Tax Rates. Retrieved from http://www.kpmg.com.hk/en/virtual_library/Tax/competitiveness_series/CorporateTaxRates.pdf. 37. Malaysian firms' suffer relatively large losses from crime and they have to spend a comparatively high cost for security. Manufacturing and services firms both reported almost 1 percent loss of sales related to crime, which is relatively high compared to Korea (0.02 percent), Vietnam (0.2 percent), Thailand (0.4 percent) and Turkey (0.2 percent) (Figure 1.22). The cost of providing security in Malaysia (2 percent) is also high ­ compared to less than 1 percent, for example, in Korea (0.2 percent), Vietnam (0.3 percent), Thailand (0.5 percent) and Turkey (0.5 percent). 29 Figure 1.22: International Comparison of Objective Indicators ­ Crime and Security Cost CONCLUSIONS 38. From an international perspective, Malaysia's overall investment climate compares favorably to countries of similar income levels. The PICS results, which provide de facto statistics on the actual business environment, suggest that Malaysia is relatively well placed in a few areas, such as export and import customs clearance; however, there is room for improvement in other areas, such as skills' inadequacies. The Doing Business indicators, which provide de jure measures of the formal regulatory requirements businesses face, rank Malaysia's investment climate reasonably well within East Asia and among countries with similar income levels. The Global Competitiveness indicators suggest Malaysia's economic competitiveness is reasonably high. 39. The PICS results show changes in the investment climate were limited in the past five years: · Judging by firms' perceptions, the ordinal ranking of top constraints in Malaysia has remained virtually unchanged: a shortage of skills remains the top obstacle. About 40 percent of firms perceived the shortage of skills as one of their top three constraints to doing business, even though the number of workers with upper secondary or tertiary education has increased sharply. Firms continue to consider macroeconomic instability and economic policy uncertainty as major concerns. Crime and theft have become a big problem. · Based on objective indicators, the changes in Malaysia's investment climate in the past five years were limited in many ways. Some aspects of infrastructure and regulatory framework improved. Nevertheless, the trends in innovation and technological capabilities have become worrisome; the same is true of crime and theft. 30 40. Malaysia has much to build on, but decisive action must occur to remove key constraints and help firms enhance their productivity. Improving skills and strengthening technological capabilities and enhancing productivity of the existing stock of talent are critical to transform Malaysia into a knowledge-based economy. 31 CHAPTER 2. FIRM PERFORMANCE IN THE MANUFACTURING SECTOR 41. Malaysia's manufacturing sector is highly concentrated and export-oriented. Electronics and Electrical appliances (E&E), dominated by multinational companies, is the single most important industry in the manufacturing sector. It accounts for over one quarter of total manufacturing value-added and more than half of exports. A high dependency of the E&E industry on imported materials may imply a large exposure to external shocks, especially the difference between price adjustments in final production and intermediate goods. 42. Labor productivity is relatively high in Malaysia compared to other developing countries in East Asia, but its growth rate has slowed in recent years. This chapter reviews firm performance in the manufacturing sector and examines the associations between investment climate and firm productivity based on 1,115 firms surveyed in PICS 2007. Within the economy, firm productivity measured by labor productivity as well as by total factor productivity (TFP) varies considerably. Across regions, firms located in the Klang Valley have the highest overall performance. Across industries, labor productivity is the highest for firms producing chemicals, followed by firms producing auto-parts, E&E, and machinery. It is the lowest for firms producing garments and wood and furniture. Total factor productivity, which accounts for the difference in capital intensity, is the highest for firms producing machinery, closely followed by firms producing chemicals. It is the lowest for firms producing rubber and plastics, textiles, and processing food. The ranking of productivity between industries remained virtually unchanged for 2004-2006. 43. Firm characteristics are generally associated with firm performance; investment climate is also closely linked to firm productivity. Large firms, foreign-owned firms, exporting firms, firms with a larger fraction of computer-controlled machines, and firms engaging in R&D activities tend to have higher labor productivity and TFP. There appears to be a close link between investment climate and firm productivity. Skilled labor shortages, regulations, the quality and adequacy of infrastructure services, along with crime and theft relate to low productivity. An improvement in these features of investment climate is often associated with an improvement in productivity. 44. The chapter is structured as follows: Section 1 briefly describes the association between productivity growth and investment climate. Section 2 measures manufacturing labor productivity, its benchmarks Malaysia with other developing countries in East Asia, and provides detailed information at the industrial and regional levels. Section 3 measures TFP. It examines the correlates between firm characteristics and productivity and between investment climate and firm productivity. Section 4 concludes. 32 MANUFACTURING SECTOR PERFORMANCE, PRODUCTIVITY GROWTH AND INVESTMENT CLIMATE Manufacturing Sector Performance 45. Malaysia's manufacturing sector is highly concentrated and export-oriented. It plays an important role in Malaysia's economy. In 2007, the manufacturing sector accounted for 71 percent of total industrial production. Manufacturing goods accounted for 80 percent of gross export earnings, while the sector accounted for roughly 30 percent of GDP and employed 33 percent of labor force.22 Figure 2.1 shows the composition of industrial production in 1999- 2005. Electronic and electrical (E&E) goods are the single most important category. They account for over a quarter of total value-added of manufacturing goods and over half of total exports; textile, garment and auto parts industries account for only a small percentage of industrial production. Rubber and plastic (petroleum product) industries witnessed a sharp jump in the share in total production ­ the share skyrocketed from 6 percent to 16 percent in 1999- 2005. The share of textile and garments declined significantly, followed by machinery. This report focuses on the nine industries that PICS 2007 covered. They represent almost three- quarters of manufacturing production and over half of total industrial production. 46. Different industries witnessed different real output growth patterns in the past few years. Figure 2.2 shows the Industrial Production Index of surveyed industries. Among the nine industries covered by PICS, the auto parts industry grew most rapidly; the textile and garment industries most slowly. Figure 2.1: Composition of Industrial Production in Malaysia, 1999-2005 Source: Bank Nagara Malaysia (Central Bank of Malaysia) http://www.bnm.gov.my/index.php?ch=109&pg=294&mth=1&yr=2009. 22 Data are for 2007. Share of manufacturing value-added to GDP is obtained from WDI online and percent of labor force employed in the manufacturing sector is obtained from Labor and Human Capital Resources Statistics (http://www.mohr.gov.my/A1.1-A1.10.pdf.) 33 Figure 2.2: Industrial Production Index across Industries, 1999-2007 Source: Source: EPU, Economic Report (2004-2009) 47. The profit ratio varies significantly across industries. PICS II data show that, in 2004- 2006, firms producing auto parts have the highest profit ratio, followed by firms producing chemical and those producing textiles; firms producing rubber and plastics, and E&E are among those with the lowest profit ratios (Figure 2.3).23 E&E firms are dependent on raw imported material ­ E&E firms import approximately 60 percent of raw materials, compared with chemicals (42 percent), auto parts (42 percent), textiles (33 percent), garment (32 percent), rubber and plastics (27 percent), food processing (20 percent), and woods and furniture (13 percent). Given the fierce competition in global market, it is not a surprise that E&E does not have a high profit margin. Looking back, in 2001 during the IT crisis, the changes in import material prices lagged behind the decline in E&E output prices. That resulted in a sharp increase in the share of material costs over total sales revenues. This suggests that E&E is vulnerable to external shocks. More than 50 percent of manufacturing exports are E&E. Exports stand for over 100 percent of GDP. A low profit margin and high vulnerability of E&E may suggest that Malaysia's export performance and total output are sensitive to global market environment. The E&E industry is dominated by multinational companies. The elasticity of entry-exit with respect to investment climate may be particularly high. In general, providing a better business environment would have positive payoffs in terms of growth and productivity. 23 Profit ratio is defined as profit divided by sales. Using information of corporate finance from PICS 2007, profit is calculated as total sales value minus total costs, including direct material costs, energy costs, manpower costs, transport costs, interest charges and financial fees, selling and general administration expenses, and other costs. Profit ratio also varies largely between firms within industries. 34 Figure 2.3: Profit Ratio by Industry (2004-2006) .3 .2 Profit Ratio .1 0 Rubber & plastics Garment Machinery Textiles Auto parts Electronics & electrical Food Woods & furniture Chemicals 48. Malaysia's export of manufacturing goods grew steadily. It increased by 13 percent in 2005-2008 (or 4 percent per annum) in a competitive global environment. The composition of manufacturing exports changed gradually over time.24 Export structure became slightly less concentrated as the share of Electronics and Electricals declined from 65 percent to 61 percent from the first quarter of 2005 to the third quarter of 2007. After the onset of the global crisis, the share further declined to 56 percent the fourth quarter of 2008. In the same period, the share of food products and petroleum products (e.g., plastics) increased from 2 percent to 4 percent and from 4 percent to 5 percent, respectively (Figure 2.4). An increase in export of petroleum products may result from the skyrocketing price of crude oil in 2006-2008. In the short run, export growth largely depends more on the demand side. However, in the long-run, productivity remains a major determinant of economic growth. 24 The share of export of nine industries that the PICS covers declined from 88 percent to 84 percent between the 1st quarter of 2005 and 4th quarter of 2008. 35 Figure 2.4: Share of Export of the Nine Industries Covered by PICS (1st Quarter of 2005­4th Quarter of 2008) Source: Bank Nagara Malaysia (Central Bank of Malaysia) http://www.bnm.gov.my/index.php?ch=109&pg=294&mth=1&yr=2009 Productivity Growth and Investment Climate 49. "Productivity isn't everything, but in the long run it is almost everything."25 A high growth rate is sustainable in the long run only if growth is driven by productivity gains. Due to the diminishing marginal returns to capital, growth that purely depends on factor accumulation will be limited. The increase in labor costs in recent years has contributed to a decline in the competitiveness of several industries. Productivity growth becomes increasingly important as the economy moves from cost- to knowledge-based. 50. Productivity is closely associated with investment climate. Firms choose their location, among other factors, in order to optimize production ­ minimize costs and maximize profits. Productivity and profitability are influenced by the investment climate which conditions costs and risks. Locations with favorable investment climate tend to attract more investments and more firms. Firms in locations with favorable investment climate are more likely to experience higher productivity growth. 51. Investment climate is perceived as an integrated package. It includes a variety of aspects, including macroeconomic stability and the certainty of economic policy, regulation burdens, skilled labor supply, infrastructure services, and so on. Different firms are subject to particular and varying constraints. For example, efficient customs clearance may have a much 25 Paul Krugman (1997), The Age of Diminished Expectations: U.S. Economic Policy in the 1990s. Cambridge, Mass.: MIT Press. 36 larger impact on firms that export and import on a regular basis than on those which occasionally buy or sell a small amount abroad. Addressing a particular aspect of investment climate may hence have different impacts on productivity. For example, improving tax regulation may unleash growth in an industry for which it is the only major binding constraint; but its impact would be much less pronounced on other industries which are hemmed in by other obstacles. 52. An improvement in investment climate over time is often associated with an increase in a firm's productivity. This is a self-reinforcing process as high economic returns encourage capital accumulation and growth. Since development is path-dependent and agglomeration effects play an important role in growth, this can perpetuate existing differences in investment climates if they are not addressed appropriately. The recent World Development Report 2009 highlights the important role of reshaping economic geography by improving institutions, infrastructures, and spatially targeted interventions. Optimizing the impacts of investment climate improvement on productivity growth would require an appropriate pace and sequence of concerted efforts. The following sections examine firm performance using two indicators: labor productivity and total factor productivity. Annex 2.1 provides details on the measurements. MANUFACTURING SECTOR LABOR PRODUCTIVITY PERFORMANCE Measuring Labor Productivity 53. Labor productivity is a common objective indicator of firm performance. It is defined as the value-added produced by each worker. The calculation of labor productivity is straightforward mathematically, equaling the value-added divided by the number of workers. Higher labor productivity mainly results from four factors: more capital or machinery per worker; better skills; more advanced or adapted technology; and a better business environment. The first two factors are often industry-specific. For instance, ceteris paribus, workers are likely to have higher labor productivity in a more capital intensive industry. Thus labor productivity may be a better proxy of productivity for comparisons among firms with similar characteristics than across firms with different capital-to-worker ratios and with a labor force of different quality. Performance at the National Level 54. The level of labor productivity in Malaysia's manufacturing sector is reasonably high given its income level, but its growth rate slowed in recent years. Measured by median value-added per worker in constant 2000 terms, labor productivity of the nine manufacturing industries covered by PICS increased only 2.2 percent per annum from RM62,700 in 2004 to RM65,600 in 2006, compared to 5.5 percent per annum from RM49,300 in 1999 and RM54,900 in 2001.26 26 The numbers reported above are calculated from the full sample of PICS surveys, measured in constant 2000 terms. The level and distribution of labor productivity across regions and industries of the full sample of firms surveyed in each round and those of the panel firms surveyed in both rounds are not statistically different. 37 55. From an international perspective, the level of Malaysia's labor productivity in the manufacturing sector is higher than some other low and middle income comparator countries.27 Malaysia's labor productivity is higher than Thailand, Philippines, and Indonesia, and some other countries with similar development level (GDP per capita US$4000-6000), such as Mexico and Chile (Figure 2.5). Among the comparator developing countries, Malaysia's labor productivity in food processing and electronics/electrical appliances is the highest; its labor productivity in the textiles industry lags slightly behind Chile. Labor productivity in garments industry is in a similar range with Thailand. This suggests that Malaysia commands a reasonable labor productivity premium relative to competitors in the first three industries, but less so in the more labor intensive garment industry. However, the higher wage rate in Malaysia may to some extent offset the advantages; its labor productivity still lags behind the more advanced economies, such as Korea and Japan. As Malaysia's economy develops to a higher stage, it is important to move up the technological ladder and rely more on value-based or knowledge-based competitiveness rather than labor cost advantages. 27 For international comparison, the comparator countries are selected based on data availability of recent PICS. Labor productivity of manufacturing firms of a country is measured by the median value of value-added per worker in each industry. If the most recent PICS provides corporate finance information for a multi-year period, we used a moving average of labor productivity for each firm and calculated the median value of the moving averages. The four industries presented in the report, food processing, textiles, garments, and electronics/electrical appliances, are selected based on data availability in comparator countries. The data are from Bangladesh (2006), Cambodia (2007), Chile (2005), Indonesia (2003), Philippines (2003) and Thailand (2007). If the Enterprise Survey provides corporate finance data for the multi-year period, we calculate value-added per employee of each firm for all surveyed period and average value for each firm. Otherwise, we use the median value of value added per employee based on the one- year data ­ corporate finance data is available only for one year in Chile and Indonesia. Labor productivity is deflated for 2000 USD, using Consumer Price Index (2000 = 100) and Official Exchange Rate of WDI. 38 Figure 2.5: Labor Productivity in Different Industries ­ International comparison (Median Value-Added per Worker, 2000 U.S. dollar) Source: Enterprise Survey (various years) and Malaysia PICS 2007 Performance at the Industry Level 56. In Malaysia, a firm's labor productivity differs considerably by industry, but changes in rankings between industries are limited over time. In the surveyed years 2004- 2006, labor productivity is the highest for firms producing chemicals, remotely followed by firms producing auto-parts, electronics & electrical appliances, machinery, and food; labor productivity is lowest for firms producing garments, followed by those producing wood and furniture (Figure 2.6). 57. For a specific industry, changes in labor productivity over time are small and offer a mixed set of directions. Firms producing food, textile, chemical, electronics, auto parts, and wood and furniture recorded rapid growth in labor productivity in 2006; firms producing machinery and garments recorded slight declines. Firms producing rubber and plastics witnessed limited changes. For some industries, growth in labor productivity can be partly explained by changes in economy of scale and labor intensity; for some industries, this relates to the relative price changes in materials / intermediate goods and final production. For example, the rapid growth in labor productivity by chemical firms can partly be related to the faster increase in sales 39 compared with that in labor. The slow growth of labor productivity of machinery can be partly related to the sharp increase in price of intermediate goods. Figure 2.6: Firm Labor Productivity by Industry (Median Value Added per Worker, 2000 Malaysia Ringgit) Source: Malaysia PICS 2002 and 2007. Performance at the Regional Level 58. Labor productivity also varies considerably by region. Overall, firms in the peninsula tend to have higher labor productivity, compared to firms in Sabah and Sarawak (Figure 2.7). More specifically, firms in Selangor, Kuala Lumpur and Melaka (Klang Valley) have the highest labor productivity, while firms in Sabah the lowest. The increasing trend in labor productivity at the national level is broadly reflected at the regional level, where productivity is higher in four out of the six regions in 2006 compared to 2004. 59. The difference in regional labor productivity reflects a difference in industrial composition. Labor productivity is defined as total value-added divided by the number of workers. Other factors are not accounted for. Capital intensive industries tend to have higher labor productivity. Capital intensive industries, such as chemicals and E&E, more often locate in the Peninsula of Malaysia.28 28 From the survey sample of the PICS 2007, 97 percent of firms producing chemicals and 99 percent of firms producing electronics & electrical appliances locate in the peninsula; while food processing firms and woods & furniture firms together account for 65 percent of all firms in Sabah and for 79 percent in Sarawak. 40 Figure 2.7: Firm Labor Productivity by Region (Median Value-Added per Worker, 2000 Malaysia Ringgit) Source: Malaysia PICS 2002 and 2007. TOTAL FACTOR PRODUCTIVITY AND INVESTMENT CLIMATE Measuring Total Factor Productivity 60. Total Factor Productivity (TFP) is another common measure of firm performance. It is defined as the residual of output or value-added that cannot be explained by factor inputs. It measures the contributions to output beyond those made by skilled and unskilled labor, the intermediate input, and the machinery/capital used. As the contribution of capital and skills is accounted for in the production function estimation, TFP is often considered a more appropriate measure for cross-industry comparisons of firms' productivity, primarily capturing the impact of technology and investment climate. The estimation of TFP, however, differs based on the econometric methods applied.29 This report uses TFP measured as the residuals from a production function estimated for each industry following the Levinsohn and Petrin (2003) techniques. Annex 2.2 provides detailed information on TFP estimations. As a robust check, we also estimate the production functions using Generalized Least Square (GLS) techniques. The production function considered assumes that output is produced by labor (skilled and unskilled), intermediate inputs, and capital. In order to capture the impacts of skills, skilled and unskilled 29 Firms that experience a large positive productivity shock may respond by using more inputs. Potentially, there is correlation between input levels and the unobserved firm-specific productivity shocks in the estimation of the parameters of the production function. Ordinary least squares (OLS) estimates of production functions will thus yield biased parameter estimates of productivity. Different methods have been developed to correct this. Olley and Parkes (1996) use investment to control for correlation between input levels and the unobserved firm specific productivity process. Levinsohn and Petrin's (2003) contribution adds to existing methods of correcting for the potential endogeneity between the choice of inputs and firm productivity by conditioning out serially correlated unobserved shocks to the production technology. 41 labor is included separately in the production function estimation. In Annex 2.3, Tables A2.4 and A2.5 present the results of production function estimations, respectively, by Levinsohn-Petrin and GLS techniques based on the full sample by industry.30 As expected, the marginal contribution to production of skilled labor is higher compared to that of unskilled labor. Total Factor Productivity in Malaysia 61. Based on PICS II, the average level of TFP among Malaysia's manufacturing firms remained virtually unchanged in 2004-2006. It increased slightly at 0.6 percent from 2004 to 2005 and remained unchanged from 2005 to 2006.31 The level of TFP differs across industries and regions in Malaysia. · Across industries: Chemical, garment and machinery industries tend to have high TFP levels; food processing, textile and rubber & plastic industries have low TFP levels; while electronics and electrical appliances, auto-parts, and wood and furniture industries have intermediate TFP levels. Changes of TFP are in general small with mixed direction (Figure 2.8). For example, TFP of machinery producer increased, while those of firms producing foods, garments, and rubber and plastic decreased somewhat.32 The difference in industry rankings for productivity is to a large extent associated with how the contribution of capital is accounted for. For example, garment firms are more labor intensive, while electronics and electrical appliances firms and auto-parts firms are more capital intensive. The relatively low level of labor productivity (value-added in per worker terms) of garment firms is associated with their large number of employees; vice- visa for electronics and electrical appliances firms and auto-parts firms. However, when other productive factors are accounted for, TFP (the residual of output) is relatively high for garment firms. 30 The statistical insignificance and/or ambiguous signs of some coefficients may be related to the insufficient number of observations at the industry level due to missing value of certain corporate finance variables and/or outliers. 31 The limited variation of TFP across years can be related to exclusion of outlying numbers and the small number of observations. Variations of TFP are large with industries. The level and distribution of total factor productivity across regions and industries of the full sample of firms surveyed in each round are not statistically different from those of the panel firms surveyed in both rounds. 32 An increase/decrease in Total Factor Productivity in the industries listed is significant in statistical terms. 42 Figure 2.8: Firm Total Factor Productivity by Industry (2004-2006) Source: Staff calculations based on Malaysia PICS 2002 and 2007. · Across regions: Firms in the Central, North, and South regions have higher TFP levels, while those in East Coast and Sarawak have lower TFP levels (Figure 2.9). More specifically, firms in Selangor, Kuala Lumpur and Melaka (Klang Valley) show the highest TFP, while firms in Sarawak the lowest TFP. Figure 2.9: Firm Total Factor Productivity by Region (2004-2006) Source: Staff calculations based on Malaysia PICS 2002 and 2007. Regressing Labor Productivity and TFP on Investment Climate Indicators and Firm Characteristics 62. This section presents the key results of a set of productivity regressions based on firm characteristics and investment climate. We start by studying the common characteristics of firms with high performance and the contribution of business environment. We use the full 43 sample of 1,115 firms surveyed in PICS 2007 to examine the associations between firm characteristics and firm performance, between investment climate and firm performance after firm characteristics are controlled for. We then attempt to shed light on the potential linkages between changes in investment climate and changes in firm performance by using the 488 panel firms surveyed both in PICS 2002 and PICS 2007, which provides two observation points of investment climate. The precise specification of the regression equations is provided in Annex 2.4. 63. The analytical results allow basic insights into the relationship between productivity, investment climate and firm characteristics. However, a number of analytical caveats need to be observed. First, macro instability and an uncertain economic policy, which are likely to have played a key role in Malaysian firm performance in recent years, cannot be directly captured in this estimate due to limited information. Secondly, the regressions may suffer from a reverse causality problem if specific investment climate indicators at the firm level are determined by productivity. For instance, the size of a firm can impact firm performance through economies of scale, but firm size can in turn be influenced by productivity if firms hire more workers as their profits increase. A third problem is that using firm-level investment climate indicators results in smaller samples; some firms did not answer certain questions. For instance, some firms might not have tried to recruit skilled labor in the survey period, or they may not have recent experience in ordering a new telephone line, but they would face similar constraints as other firms if they had done so. To minimize the problems of endogeneity and sample constraint, this report follows the methodology developed in Dollar, Hallward-Driemeier and Mengiste (2005) and uses regional industry mean levels of the investment climate indicators instead of the firm level data. These can be considered largely exogenous to a specific firm, and are also used to replace missing values.33 Firm characteristic variables are kept at firm level. However, although the findings from the regressions are informative, they potentially could suffer from problems of reverse causality. The findings from the regression analysis need to be interpreted as indicative of a correlation between performance and characteristics but not a causal relationship. Correlations Between Firm Characteristics and Performance 64. A firm's labor productivity and TFP are associated with their specific characteristics. These characteristics mainly include the firm's age, employment size, foreign/domestic ownership, export status, measures of technology and innovations (percentage of computer controlled machines), and R&D status. Regressions of firm characteristics over labor productivity and TFP are conducted separately using the full sample and the panel sample. Although firm characteristics are almost identical between the full sample and the panel sample, firms in the panel sample are older than those in the full sample (Table 2.1).34 In Annex 2.5, regression results from the full sample appear in Table A2.6 and those from the panel sample in 33 See Dollar, Hallward-Driemeier and Mengiste (2005) for detailed discussion. An underlying assumption is that more efficient firms can operate and adopt business schemes under the certain exogenously given business environment. Therefore, using firm-level indicators of investment climate may lead to a biased estimation, because the error term of the estimated specification would be correlated with firm productivity indicators. A simple solution to the problem of erogeneity is to use region-industry averages of firm-level observations. 34 This is an inevitable feature of firms in the panel sample, because the firms included in the panel dataset must have been included in the PICS 2002; while those in the full sample PICS 2007 include newly established firms. 44 Table A2.7. Some firm characteristics, including a dummy for foreign ownership information, one for export orientation, and a percentage of computer-controlled production machinery, as well as a dummy for a firm's R&D activity, are measured with the most recent data at the time of the survey (2002 or 2007); other variables, including firm age, employment size, are measured with corporate finance information reported each year.35 Table 2.1: Descriptive Statistics of Firm Characteristics in 2006, Full Sample and Panel Sample Firm characteristics Full sample Panel sample t-statistics p-value Plant age 18.1 19.8 -3.11 0.00 Size (# of worker) 167.1 161.0 0.33 0.75 Foreign fund management firms 25.9 27.1 -0.50 0.62 (>=30% foreign owned) Exporting firms (>=10% export) 51.2 49.5 0.63 0.53 % of computer controlled 25.4 23.8 0.89 0.38 machines R&D activity (>0 % spent on 16.5 14.7 0.88 0.38 R&D activity) N 996 459 Source: Staff calculations based on Malaysia PICS 2002 and 2007. 65. The relationship between a firm's age and its labor productivity or TFP is mixed and not significant in statistical terms.36 In the full sample, firm age is positively associated with labor productivity and TFP, but the association is not significant in statistical terms. In the panel sample, however, firm age is positively associated with TFP and negatively associated with labor productivity. Firm age is a characteristic that both affects performance and may itself be affected by firm performance. On the one hand firms can cumulate experience and improve performance over time; on the other hand, in a competitive environment, firms with low productivity often have to exit the market and only those with high productivity remain. In this regard, the positive association between age and productivity, as reflected in the full sample results, may be driven by both ways. However, the association between age and TFP remain positive in the panel sample; the association between age and labor productivity changed signs to be negative as the effects of survival selection bias were partially controlled for. Many factors can influence this association. All else being equal, newly-established firms which are more likely to be equipped with new machinery and more adaptive technology may have higher labor productivity. In addition, business life cycle differs across industries. Overall, the relationship between firm age and productivity is weak. 66. A firm's size, measured by the total number of employees, is positively associated with TFP in a significant way. This is true of the full sample as well as the panel sample. It is negatively associated with labor productivity in the panel sample. This is partly related to the 35 The results are to interpret with caution subject to multicolinearity. 36 Square term of firm age was introduced into the regression to test for non-linear relationships. No quadratic relationship was found within normal range of firm age with statistical significance. 45 difference at the industry level. Larger firms are generally more productive and enjoy more benefits of economies of scale. 37 67. Foreign fund management firms, namely those where more than 30 percent of shares are owned by foreign private firms, have higher labor productivity and TFP. The positive associations are significant in statistical terms for both the full sample and panel firms. The PICS results suggest that the median labor productivity of foreign fund management firms is 19 percent higher than that of purely domestic firms and firms managed by less than 30 percent foreign fund in 2006; TFP of foreign fund management firms is more than 30 percent higher. This may be related to the more advanced technology and efficient management scheme used in foreign fund management firms. 68. PICS 2007 results show that exporting firms in Malaysia tend to have higher labor productivity and higher TFP compared with non-exporting firms.38 This is consistent with the findings in many other countries. On average, exporting firms exhibit 18 percent higher labor productivity and 6 percent higher TFP than non-exporting firms based on the results of the full sample, other things being equal. The higher productivity of exporting firms may be related to the higher exposure to fiercely competitively international markets and to more advanced technology and quality control requirement demanded by foreign buyers. 69. A firm's technology level, as measured by the percent of computer-controlled production machinery, is positively associated with labor productivity and TFP in a significant way. In other words, firms with a higher percentage of computer-controlled machinery tend to have higher labor productivity and TFP. The significant relationship is robust for the full sample and the panel sample. Controlling for all other characteristics, the estimates based on the full sample model suggest that a 10 percent increase in computer-controlled production machinery is associated with a one percent higher TFP and two percent higher labor productivity. This suggests that, in general, new machinery has contributed to improving firm productivity. 70. Firms engaging in R&D activities generally have higher productivity. In Malaysia, 16 percent of firms surveyed in the PICS 2007 reported they engaged in R&D activities.39 Based on the full sample results, firms with R&D activities have 22 percent higher labor productivity 37 The PICS results suggest no significant correlation between firm size and capital intensity. 38 The correlation between exporting status and productivity was insignificant in PICS 2002, which could be related to the sharp decline in export following the global economic slowdown in the early 2000s. In 2001, manufacturing sector in Malaysia experienced a sharp contraction of 4.8 percent because of the slump in global demand for manufactured goods, particularly in ICT products. While domestic firms recorded growth of 8 percent, it did not "fully offset the contraction in export-oriented industries. Source: "Asian Development Outlook 2002." (ADB, Manila). 39 Firms of capital intensive industries are more likely to engage in R&D activities compared to those of labor intensive industries. The percentage of firms engaging in R&D activities is the highest in electronics and electrical appliance industry ­ 23 percent of electronics firms engage in R&D activities, followed by chemicals (19 percent), machinery (19 percent), food processing (17 percent), and auto parts (17 percent), while the percentage is lower in labor intensive industries, such as textiles (14 percent), wood products and furniture (12 percent), and garments (7 percent). 46 and 9 percent higher TFP compared to those do not engage R&D activities.40 As technology and innovation are associated with higher firm performance, and more and more so as the economy move towards more knowledge-based, providing incentives for firms to invest in R&D and bridging successful R&D to real productivity enhancement are important. 71. In summary, in Malaysia, large firms, foreign-owned firms, exporting firms, firms with a larger percentage of computer-controlled machinery, and firms engaging in R&D activities tend to have higher labor productivity and TFP. The effects of firm age on performance are weak and mixed. Correlations Between Investment Climate and Firm Performance 72. Firm performance is also associated with the investment climate where the firm is located. For firms with similar characteristics, those that face a more conducive investment climate are more likely to have higher productivity. The results from regression analysis are informative and intuitive; they show a close correlation between performance and investment climate.41 73. This section focuses on examining the correlations between four key aspects of investment climate and labor productivity and TFP among Malaysian firms in 2004-2006. Controlling for firm characteristics, we include four quantifiable objective investment climate indicators ­ percentage of university graduates, percentage of senior managers' time spent in dealing with regulations, percentage of sales loss from crime, and number of days to obtain fixed phone connections ­ to capture the effect on firm productivity of business environment constraints perceived by firms from four aspects ­ skilled labor shortage, regulation, crime and theft, and infrastructure logistics.42 Various elements of investment climate are interrelated, so are their impacts on firm performance. Besides data constraint, the selection of these four variables is partly a balance of the need to include "more indicators" to limited potential bias of omitted variables and to include "fewer indicators" to minimize the potential bias of multi- colinearity. Regression results are presented in Annex 2.6, Table A2.8 on correlates between investment climate and labor productivity and Table A2.9 on correlates between investment climate and TFP. 74. The skills of workers, measured as a percentage of employees with a college degree, play a significant role in explaining variations in labor productivity and TFP. A higher percentage of employees with a college degree is positively associated with firm productivity indicators in a significant way. The results suggest that a 10 percent increase in employees with a 40 Firm performance is associated not only with the current engagement of R&D but more importantly its R&D activities in the past. In addition, given the high risk / high return nature of R&D investment, the rationale and profitability of R&D engagement vary across firms. 41 The statistical significance of the correlations between firm characteristics and investment climate indicators is indicative, and does not imply causality. 42 IC indicators included in this model is selected in the following way: (1) Based on the business managers' perception and availability of objective IC indicators, four major areas of major IC constraints were identified: skills and education of available worker, regulatory framework, crime and infrastructure (see Chapter 1); (2) As a preliminary analysis, we conduct forward selection/backward elimination stepwise regression to identify potentially significant IC indicators within each major IC area; (3)We repeat the operation to identify robust IC indicators that are associated with firm productivity. 47 college degree is associated with a 10 percent increase in labor productivity and a 4 percent increase in TFP, ceteris paribus. If the percentage of employees with a college degree is considered as a proxy of the quality of labor force, then the important role in labor productivity and TFP suggests that improving the skills of the existing labor force and increasing the supply of skilled labor to match market demand are both critical for enhancing firm performance. Investment in human capital is important. 75. Access to infrastructure facilities is significantly correlated with labor productivity and TFP. Although Malaysia's overall quality of infrastructure is good (it ranks in 23rd place out of 134 countries in the Global Competitiveness Report (2008)), its quality of fixed phone infrastructure, which ranks in 71st place, lags behind that of many other comparator countries43 Accessibility of fixed phone lines, measured by the average number of days to obtain fixed phone connection, is negatively correlated with labor productivity and TFP. This indicates that further investment in infrastructure has considerable potential to enhance firm productivity. 76. Crime and theft imposes large social costs. Loss from theft, robbery or vandalism against an establishment as percentage of total sales is negatively associated with firm productivity.44 Increasing crime discourages firms from investing and increases the costs of business, through the direct loss of goods or the cost of taking precautions such as hiring security guards, building fences, or installing alarm systems. In average, Malaysian firms spent 2.3 percent of the sales in providing security with an estimate of 0.8 percent of the total sales loss from crime. More effort is needed to address the increasing trends of loss from crime and cost of providing security. 45 Government can play an important role in changing the incentives for criminals through better law enforcement, stronger deterrence, and more effective crime prevention programs. 77. The indirect cost of complying with business regulations, as measured by the percent of senior manager's time spent in dealing with government requirements, is negatively associated with a firm's productivity.46 This suggests that improvements in streamlining the process of regulations, including tax administration, customs' regulation, and licensing and registration, will help enhance firm performance by reducing unnecessary costs. 78. The correlation between investment climate indicators and firm performance are jointly significant. Good investment climate is closely associated with high firm productivity, others being equal. Investing in skills, providing a good regulatory framework, improving infrastructure services, and increasing access to finance are important measures for alleviating the binding constraints on doing business and enhancing firm performance. 43 The Global Competitiveness Report ranks Malaysia quality of overall infrastructure the 23rd ­ it is better than many regional comparator countries, such as Thailand (29th), China (47th), Indonesia (86th), Philippines (92nd), Cambodia (97th), while Malaysia lags slightly behind Taiwan (19th), South Korea (15th), and Japan (11th). However, the Global Competitiveness Report ranks Malaysian telephone line system 71st ­ it is considerably lower than other infrastructure subindicators, such as quality of roads (17th), quality of port infrastructure (16th), and quality of electricity (31st). 44 The negative association is statistically significant for labor productivity. 45 Loss from theft, robbery or vandalism as percent of sales doubled from 0.5 percent to 0.9 percent, and cost of providing security as percent of sales increased from 1.1 percent to 2.4 percent in 2002-2007. These increases are statistically significant at the 99 percent level. 46 The negative association is statistically significant for TFP. 48 Correlations Between Changes in Investment Climate and Changes in Productivity 79. Using the panel sample, this section examines the relationship between changes in investment climate and changes in firm productivity.47 Each round of PICS provides one observation point of investment climate. Firm characteristics are included to capture firm- specific effects on firm productivity indicators as well as industry and regional dummies. In addition, the initial level of firm productivity, measured by three-year moving average of value- added per worker or TFP in logarithmic form are included as controlling variables. It is often less expensive and more predictable to adopt/adapt existing technology and know-how to move towards a production frontier than to innovate and advance to a new frontier. Similar as Solow's growth model and convergence, firms with higher initial level of productivity may have lower productivity growth due to technological constraints. This objective of the analysis is to shed light on how closely and to what extent an improvement of a specific aspect of investment climate is associated with productivity changes. Annex 2.7 provides detailed information of empirical analysis on the correlates between changes in investment climate and changes in TFP. Results show that changes in most investment climate indicators are correlated with changes in TFP with the expected signs. Firm characteristics and industry and region dummies are controlled for. 80. Firm characteristics jointly play a significant role in changes in productivity. As expected, the initial levels of firm productivity are negatively associated with their growth rates in a significant way. A firm's age is negatively associated with productivity growth while size is positively associated with productivity growth. A negative sign of the export dummy shows that productivity growth rate is lower for exporting firms. A positive sign of the R&D dummy suggests that firms engaging R&D activities grow at a faster pace. Our analysis in the previous section suggests that exporting firms and firms with R&D activities both tend to have higher level of productivity, others being equal. The contrasting signs of their association with growth of productivity provide support that while the effects of export activity are positive they themselves may not be sufficient to enhance productivity in the long run; as the economy becomes more knowledge-based, R&D plays a more and more important role in the continuous improvement of productivity. 81. Changes in investment climate indicators often exhibit a correlation with changes in labor productivity and TFP. However, an improvement in infrastructure is the only one that is significant in statistical terms; improvement in human capital is associated with the largest marginal effects. An improvement in human capital availability, measured by the percentage of workers with a college degree, is positively correlated with more rapid growth in labor productivity. An increase in loss from theft, robbery or vandalism, measured as a percentage of sales, poses adverse effects on labor productivity and TFP. An improvement in 47 The changes of labor productivity and TFP are regressed on the changes of IC indicators, while firm characteristics are controlled for. Changes in IC indicators are calculated by logarithmic changes of an IC indicator between PICS I and PICS II. In other words, where denotes a change in an IC indicator, X, in an industry, j, of region, k; and denote region- industry mean of the IC indicator in PICS I and II, respectively. Therefore, a positive sign of the difference indicates an increased in IC indicators between PICS I and II, and a negative sign a decrease. 49 infrastructure and regulations, measured by the average number of days to fix a phone connection and the percentage of a senior manager's time spent in government compliance, respectively, is correlated with an increase in labor productivity and TFP. CONCLUSIONS 82. Malaysia's labor productivity in manufacturing is reasonably high given its income per capita level compared with other developing countries in East Asia, but its growth rate has slowed in recent years. Labor productivity of the nine manufacturing industries covered by PICS grew at 2.2 percent in 2004-206 compared to 5.5 percent in 1999-2001. 83. Although the patterns of growth varied significantly across industries and firms, the productivity rankings between industries remained virtually unchanged. Labor productivity is the highest for firms producing chemicals, remotely followed by firms producing auto-parts, E&E, and machinery; it was lowest for firms producing garments, wood and furniture. Total factor productivity, which accounts for the difference in capital intensity, is the highest for firms producing machinery, closely followed by firms producing chemicals; it is lowest for firms producing rubber and plastics, textiles, and food processing. 84. Empirical findings suggest a close relationship between firm characteristics and productivity. Large firms, foreign-owned firms, exporters, those with a larger percentage of computer-controlled machinery or engaging in R&D activities tend to have higher labor productivity and TFP. Further, there appears to be a close relationship between investment climate and firm productivity. Firm performance is likely to be higher in areas where the investment climate is better, skilled labor is abundant, regulations and bureaucratic procedures less burdensome, infrastructure services adequate and reliable, and crime and theft are limited. An improvement in one or more of the preceding is often associated with an improvement in productivity. 85. In general, addressing investment climate constraints yields positive payoffs in growth and productivity. However, establishing a clear causality relationship between investment climate and productivity is beyond the scope of this work. Therefore, more research is warranted on the links between the diverse elements of the business climate and productivity. 50 CHAPTER 3. FIRM PERFORMANCE IN THE SERVICES SECTOR 91. The services sector has been identified as a key source of growth in the recent Malaysia Plans. It contributes to more than 50 percent of GDP and employs more than half of the labor force. As Malaysia makes the transition from a middle income to a high income country, the significance of the services sector is likely to increase even further. Both the import and export of services have increased since 2001. Trade deficit in services narrowed and recently became surplus. Tourism is still the most important part of services export. The largest part of services import was communications and computer related services in 2001, but it was overtaken by transport services in 2006. 92. Labor productivity in Malaysia's overall services sector has been stagnant since the early 1990s while other countries have seen significant improvement. From an international perspective, it lags behind countries in the region with a similar income level. The gap between Malaysia and Japan and Korea widened. In 2005, labor productivity of the overall services sector in Malaysia was less than half of that in Korea and less than one-tenth of that in Japan. 93. Business-support services firms, an important sub-sample in the overall services sector covered by PICS, appear to have high profitability, but growth has slowed. This chapter examines at the national and industry levels the performance of the overall services sector using national account data with cross-country comparisons. It assesses at the firm level the performance of the selected business support services sector based on the PICS 2007, which covers a sub-sample of 303 firms.48 At the firm level, average sales of the existing firms in most of the surveyed business support services industries have expanded, even though growth has slowed. The export participation rate and size of export sales are improving, especially for communication services firms. 94. Similar to the manufacturing sector, firm characteristics are closely associated with firm performance in selected business support industries. Firms with high capital intensity tend to perform better in terms of value of sales and labor productivity. Exporting firms and firms offering in-services training also perform better. Older firms tend to be larger than younger firms, but not necessarily more productive. Foreign-owned firms are often larger and more productive compared with domestic-owned firms. Empirical results suggest that foreign-owned firms, especially those that are not constrained by the limit of 30 percent foreign ownership, are more productive.49 Their presence in the business support services sector can boost the productivity of domestic firms. 48 These firms are concentrated in the business-support services industries, and the PICS does not cover banking, retail, tourism, and other services industries. 49 Under the government regulations, foreign ownership of firms in some industries cannot exceed a maximum of 30 percent. See table 3.4 for details. 51 95. A shortage of skills, lack of technological and innovative activities, and regulatory barriers are major constraints on investment in the services sector. Strengthening the services sector as a key source of growth requires reforms on several fronts. While some liberalization has taken place since 2002, the overall services sector in Malaysia is still relatively restrictive when compared to other countries at a similar stage of development. A recent relaxation of ownership restrictions in certain services industries, including some financial sector firms, indicates change is happening in a positive direction.50 96. The chapter is organized as follows. Section 1 presents the performance of the overall services sector at the national, industry, and firm levels. Section 2 investigates some key investment climate aspects in the services sector ­ skills, innovation readiness, and regulatory regimes. Section 3 examines the key drivers of efficiency performance at the firm level empirically. Section 4 concludes. OVERALL SERVICES SECTOR PERFORMANCE 97. This section provides a snapshot of firm performance in the overall services sector in Malaysia. At the national level, we will look at employment, value added, and labor productivity. At the industry level, we will benchmark the services sector in Malaysia to that of some selected countries from an international perspective, as well as other sectors within the Malaysian economy in 2001 and 2007. Industry performance within the services sector is further examined using trade and national account statistics. At the firm level, we will have an in-depth analysis of labor productivity performance of the five selected business support services industries using PICS data. 98. However, several factors limit the robustness of this chapter's findings: · First, measuring the efficiency of services firms is more challenging compared to the manufacturing firms. That's because output, inventory, and material costs are difficult to measure for services firms. Firm level total factor productivity cannot be directly estimated. This chapter uses sales growth, along with value added per worker where data are available, and sales-to-cost ratio as proxies for firm-level efficiency in the services sector. In addition, by carefully isolating the movement of labor productivity due to changes in capital per worker, we examine the effects of various policy variables on firm productivity. · The firm-level analysis in this chapter is based on a survey of 249 services firms in 2002 and 303 firms in 2007 in the following five business support industries: information technology, communications services, accounting and related services, advertising and marketing, and business logistics. Given their special characteristics, the findings based on this sample of business support services firms cannot be generalized to other services industries in Malaysia, such as wholesale, retail, hotels, restaurants, and tourism. 50 In March 2009, the Government has undertaken measures to further liberalize the services sector. Its impact is not captured in this report. 52 Performance at the National Level 99. The services sector is a major part of the Malaysian economy (Box 3.1). It contributes 57 percent of total employment and 55 percent of GDP in 2008, of which, 47.6 per cent was from nongovernment services.51 Complementing the growth and development in the manufacturing sector, the Government is intensifying its efforts to promote and develop the services sector. The Government plans to tap the full potential in the services sector and raise its contribution to 60 per cent of GDP, as identified in the Third Industrial Master Plan. 100. From an international perspective, Malaysia's services sector contributes a relatively small percentage to the country's GDP, the same is true for its share of employment.. 52 Although the share of employment of the services sector has increased in the past,53 it is low relative to more advanced countries, even when compared to countries with big manufacturing sectors, such as Ireland, Japan and Korea (Figure 3.1 and Figure 3.2). For these countries, the share of services employment in 2005 was about 65 percent. When we compare Malaysia to other countries with a comparable GDP per capita level, as shown in Figure 3.3, it is clear that Malaysia is below the sample average (as predicted by the regression line). Box 3.1: What Counts as Services? Services corresponds to International Standard Industrial Classification (ISIC) divisions 50-09 and includes value added in wholesale and retail trade (including hotels and restaurants), transport and government, financial, professional, and personal services such as education, health care, and real estate services. Also included are imputed bank service charges, import duties, and any statistical discrepancies noted by national compilers as well as discrepancies arising for rescaling. Value-added is the net output of a sector after adding up all outputs and subtracting intermediate inputs. It is calculated without making deductions for depreciation of fabricated assets or depletion and degradation of natural resources. The industrial origin of value added is determined by the ISIC, revision 3. The construction industry is not included in the services sector. 101. Overall, labor productivity in Malaysia's services sector has been stagnant in the past while other countries have seen significant improvement. Figure 3.4 suggests that labor productivity, measured by value added per unit of worker in 2000 US$, in the services sector has been stagnant at about US$7,700 in 2005, a level Malaysia achieved in 1993. In other words, 51 Data Source: Ministry of International Trade and Industry. http://www.miti.gov.my/cms/genArticlePdf?id=com.tms.cms.article.Article_ea4b01dd-c0a81573-7cd87cd8- e9437020 52 For international comparisons, data is from World Development Indicators, which gathers information from national and international sources. So it is possible that different countries have slightly different definitions of the variables, and may have gone through various revisions in different periods. The numbers can be slightly different from the statistics of the respective countries. Employment in services is the proportion of total employment recorded as working in the services sector. Employees are people who work for a public or private employer and receive remuneration in wages, salary, commission, tips, piece rates, or pay in kind. Services include wholesale and retail trade and restaurants and hotels; transport, storage, and communications; financing, insurance, real estate, and business services; and community, social, and personal services-corresponding to divisions 6-9 (ISIC Revision 2) or tabulation categories G-P (ISIC Revision 3). To calculate the ratio of value-added to GDP, gross value-added at factor cost is used as the denominator. 53 The share of employment in the services sector was about 50 percent in 1999 and reached 57 percent in 2008. 53 improvement or growth in labor productivity in the services sector of Malaysia was limited in the past decade. For the same time period, labor productivity of the whole economy of Malaysia, which is measured as GDP per employment, has improved from US$9,100 to nearly US$10,000, as shown in Figure 3.5. Thus, labor productivity of services sector is lagging behind in Malaysia. 102. Labor productivity in Malaysia's services sector lags behind countries with a comparable level of GDP per capita. Figure 3.6 shows the labor productivity of Malaysia is lower than the predicted value based on a sample of more than 60 countries for the period of 1980 to 2006, based on GDP per capita. Similar underperformance can be seen when we compare labor productivity of the services sector in Malaysia to that of other countries, as shown in Figure 3.4. In addition, compared to the stagnant performance of the services sector in Malaysia, most of these countries have experienced positive growth in labor productivity and are far more productive. In 2005, the labor productivity of the services sector of Malaysia regressed to less than half of that of Korea and less than 10 percent of that of Japan. Figure 3.1: Employment Share of Services Sector in 1999 and 2005 Employment Share of Services Sector Selected Countries, 1999 & 2005 90 80 70 60 % 50 40 1999 30 2005 20 10 0 FIN HKG IRL JPN KOR MYS SGP THA Source: World Development Indicators (2008) Missing data for Malaysia in 2005 is substituted by data in 2004. 54 Figure 3.2: Share of Services Sector in GDP in 1999 and 2005 Share of Services Sector in GDP Selected Countries, 1999 & 2005 100 90 80 70 60 % 50 40 1999 30 2005 20 10 0 FIN HKG IRL JPN KOR MYS SGP THA Source: World Development Indicators (2008) Figure 3.3: Share of Services Sector in GDP Relative to GDP per Capita Source: The line corresponds to the predicted share of services sector in GDP from a regression on log per capital GDP and a constant. World Development Indicators (2008). 55 Figure 3.4: Labor Productivity of Services Sector in 1999 and 2005 Figure 3.5: Labor Productivity in All Sectors in 1999 and 2005 56 Figure 3.6: Cross-Country Comparison of Labor Productivity in the Services Sector Source: World Development Indicators (2008). The line corresponds to the predicted labor productivity from a regression on log per capita GDP. Total number of observation is 1363, with an R-squared of 0.9532. Performance at the Industry Level 103. We use international financial statistics and national account data to introduce some big picture impressions of the performance of the overall services sector at the industry level for different countries.54 First, we use international finance statistics from the World Development Indicators 2008 for cross-industry, cross-country, and time series comparison of the services sector. Next, we employ Malaysian national accounts data. The advantage of national accounts data is that it provides more detailed coverage of several services sector industries over time. However, data on capital stock for the services sector is not available and therefore TFP growth cannot be estimated. Also, the business support industries covered by the PICS (information technology, communications services, accounting and related services, advertising and marketing, and business logistics) cannot be neatly mapped into the national accounts disaggregation of the services sector. Business support services in the national accounts data come under two categories in Malaysia: transport, storage and communication; and finance, insurance, real estate, and business services. One other services sector category covered in this section is wholesale and retail sales, hotels, and restaurants. 104. Overall, the trade deficit in the services sector has narrowed in the past few years; recently it has become a trade surplus. This resulted from an increase in exports and imports of services. In 1999, the total value of private commercial services exports in Malaysia was about US$12 billion. The volume swelled to US$22 billion in 2006. The average growth rate of services exports in Malaysia is about 15 percent per annum. The fastest growing country in the 54 International time-series data on the performance of different services industries is not available. 57 sample is Ireland, with an impressive 37 percent annual growth rate (Figure 3.7). Figure 3.8 presents the value of commercial services imports of Malaysia, which increased from US$15 billion in 1999 to US$24 billion in 2006. The average annual growth rate of services imports during this period was 12 percent. In 2006, while the growth rate of services import was less than its exports, Malaysia still ran a US$1.8 billion trade deficit in commercial services in 2006, which was 3.8 percent of the total value added for the services sector. In the same period, Hong Kong and Finland ran running trade surpluses in commercial services of US$36 billion and US$0.5 billion, respectively. Figure 3.7: Services Exports in Selected Countries in 2001 & 2006 Value of Service Export Selected Countries, 2001 & 2006 Billlion of US$ 140 120 100 80 60 2001 40 2006 20 0 FIN HKG IRL JPN KOR MYS SGP THA Source: World Development Indicators (2008) Figure 3.8: Services Imports in Selected Countries in 2001 and 2006 Value of Service Import Selected Countries, 2001 & 2006 Billion of US$ 160 140 120 100 80 2001 60 2006 40 20 0 FIN HKG IRL JPN KOR MYS SGP THA Source: World Development Indicators (2008) 58 105. Within the services sector, the private commercial services sector can be broken down into four main industry groups55: (a) Transport Services ­ cover all transport services (sea, air, land, internal waterway, space, and pipeline) performed by residents of one economy for those of another. They involve the carriage of passengers, movement of goods (freight), rental of carriers with crew, and related support and auxiliary services. Excluded are freight insurance, which comes under insurance services; goods procured in ports by nonresident carriers and repairs of transport equipment, which are included under goods; repair of railway facilities, harbors, and airfield facilities, which come under construction services; and rental of carriers without crew, which comes under other services. (b) Travel Services ­ cover goods and services acquired from an economy by travelers in that economy for their own use during visits of less than one year for business or personal purposes. Travel services include goods and services consumed by travelers, such as lodging and meals and transport (within the economy visited). (c) Insurance and Financial Services ­ cover various types of insurance provided to nonresidents by resident insurance enterprises and vice versa, and financial intermediary and auxiliary services (except those of insurance enterprises and pension funds) exchanged between residents and nonresidents. (d) Communications ­ cover international telecommunications and postal and courier services; computer data; news-related services transactions between residents and nonresidents; construction services; royalties and license fees; miscellaneous business, professional, and technical services; personal, cultural, and recreational services; and government services not included elsewhere. 106. Travel services remain the largest services exports, while transport services have overtaken communications as the largest services imports. Figure 3.9 presents the composition of services exports and imports in 2001 and 2006 according to the four main industry groups of the selected countries. The most important industry group within the services sector in Malaysia, in terms of export, is travel services, when the share of travel services in services export was nearly 50 percent in 2001 and 2006. The only other country that had such a high share of travel services export was Thailand. Travel services there accounted for more than 50 percent of its services exports. In terms of services imports, the largest import industry used to be communication services in 2001; in 2006 it was replaced by transport services. 55 Due to data constraint, information on other services industries, such as real estate, hospital, and education, are not presented here. 59 Figure 3.9: Industry Performance in Services Sector, 2001 & 2006 Composition of Commercial Service Export in 2001 Composition of Commercial Service Import in 2001 80 90 70 80 60 70 50 60 40 50 % % 30 40 20 30 10 20 0 10 -10 0 FIN HKG IRL JPN KOR MYS SGP THA FIN HKG IRL JPN KOR MYS SGP THA Transport Travel Insurance and Financial Services Communications, computer, etc. Transport Travel Insurance and Financial Services Communications, computer, etc. Source: World Development Indicators (2008) Source: World Development Indicators (2008) Composition of Commercial Service Export in 2006 Composition of Commercial Service Import in 2006 80 80 70 70 60 60 50 50 40 % 40 % 30 30 20 20 10 0 10 FIN HKG IRL JPN KOR MYS SGP THA 0 Transport Travel Insurance and Financial Services Communications, computer, etc. FIN HKG IRL JPN KOR MYS SGP THA Transport Travel Insurance and Financial Services Communications, computer, etc. Source: World Development Indicators (2008) Missing data for Hongkong in 2006 is substituteed by 2005 Source: World Development Indicators (2008) Data in Communications and computer industry are calculated as the residual of Missing data for Hongkong in 2006 is substituteed by 2005 100 net of shares of all the other three industries for each coutry. Performance at the Firm Level 107. The surveyed firms are diverse in terms of location, age, and ownership structure. More than 70 percent of the firms are from the region of Selangor, Kuala Lumpur, and Melaka. In terms of the age of sample firms, we have new entrant firms and established firms. The average age of firms in the sample is 18.6 years old, while the median firm is about 17 years old. Only 5.2 percent of sample firms are less than 5 years old, and they are mainly in the technical consulting services, advertising agencies, and custom computer programming industries. The majority of the industries sampled did not have new entrants in the past 5 years. That suggests new entries were limited. In terms of ownership structure, domestic-owned firms that have no foreign equity represented 80 percent of the sample; 20 percent of the sample firms had some foreign equity, most of them were concentrated in the advertising industry. In terms of location, age and domestic ownership, the full sample of firms covered in the 2007 survey are rather similar to the 2002 survey. However, a different picture emerges when we focus on those firms 60 that have some foreign equity. Among the firms that have some foreign equity surveyed in the 2007 round, 23 percent of them have no more than 30 percent foreign investment; the ratio was 37 percent in the 2002 round. This suggests that a lower share of foreign-owned firms were affected by the 30 percent foreign equity restriction in the past five years. In addition, seven percent of the sampled firms in 2002 and eight percent in 2007 were majority foreign owned. 108. Compared to the 2002 survey, business support services firms appear to be doing well even though growth has slowed. Firm performance is measured using several indicators, including labor productivity, sales growth and sales-to-cost ratio. Because a large number of firms did not report material and inventory costs, we cannot estimate TFP. In Annex 3.1, Table 3.1a compares the average performance of the business support services in 2004-2006 with that in 1999-2001 (the previous survey). For 2004-2006, the average growth rate of sales per annum was 5.6 percent, whereas for 1999-2001, it was 6.9 percent. To understand whether the slowing down comes from differences in the composition of firms in the two surveys, Table 3.1b reproduces Table 3.1a, but restricts the sample to only those firms covered in both surveys. The result is informative. For the sample of common firms, the annual growth rate of sales decreased from 4.1 percent in 1999-2001 to 2.7 percent in 2004-2006.56 This seems to suggest that there is a genuine slow-down in sales in the business support services sector. On the other hand, the annual growth rate of costs also slowed significantly. For the full sample of firms, cost growth dropped from 12 percent in 1999-2001 to 7.3 percent in 2004-2006; for the sample of common firms in both periods, cost growth dropped from 10.8 percent to 4.7 percent.57 109. The profitability of firms in the business-support services sector remains high. Given that the slow growth in costs more than compensates for the slow growth in sales, it is not too surprising that, on average, the sales-to-cost ratio of firms has remained around 2 in both surveys, as shown in Tables A3.1a and A3.1b in Annex 3.1. This suggests that firms are doing well throughout both survey periods. Thus, even though sales are slowing down, costs of firms are rising less quickly. That means relatively constant profit margins for these firms. 110. While the capital intensity of firms in the survey sample seems to have declined, firms participating in both rounds reported an increase in assets per worker.58 This suggests that the new survey may have included more small firms than the previous survey, which pulled down the overall ratio of average assets per worker. 111. Among business support services firms, the average education level of workers has improved. The average years of schooling per worker in the services sector, as measured at the firm-level for the complete samples, is 13.1 years for the 2004-2006 period, compared to 12.3 years for the 1999-2001 period. For the subset of common firms, the average years of 56 Sales growth is different between firms in the full sample and those in the panel in a significant way. This may be related to the fact that the new survey included smaller firms. Due to limited information, the reasons on why these new smaller firms have stronger performance cannot be fully explored in this report. 57 The limited number of observations can be a factor that contributes to the large changes over time and between two different sets of samples. 58 For the full sample, it drops from RM653 thousand for the earlier period to RM340 thousand in the later period, as shown in Table 3.1a. However, when we restrict the calculation to the set of common firms covered in both periods, the value of asset per worker in fact has improved from RM402 thousand to RM429 thousand. 61 schooling improved from 12.4 year to 13.6 years. Thus, on average, more than 50 percent of workers in services have completed secondary education and have some tertiary education. 112. Among existing firms, average sales in most industries have expanded; however, growth rates have slowed. Table 3.2a presents the average annual total sales of the firms in the sample by industry and survey. In terms of average sales, firms covered in the 2007 survey are smaller than the 2002 survey; however, in terms of sales growth, they are quite compatible, except in the advertising industry and the business logistic industry. Table 3.2b repeats the same exercise but only focuses on the set of firms that are common between the two surveys. Here we see that, with the exception of telecommunication and accounting industries, existing firms in all other industries have improved their sales between the two survey periods. For the accounting industry, average sales have remained similar while the telecommunication industry experienced a significant reduction in sales from the 1999-2001 period to the 2004-2006 period. Nevertheless, the average annual growth rate of sales across all industries has slowed. This is reflected in the aggregate picture shown in Tables A3.1a and A3.1b. 113. The export participation rate has improved for most industries along with the amount of export sales. The lower part of Table A3.2a and A3.2b in Annex 3.1 reports the export activities of firms by industry. By taking the ratio of the number of exporters to the number of firms, we can calculate the export participation rate of the industry. For the 1999- 2001 period, the export participation rate ranged from 5 percent in the accounting services industry to 30 percent in the information technology and communication industries. For the 2004-2006 period, export participation rate improved to 8 percent in the accounting services industry and 34 percent in the information technology industry. While the participation rates went up, given that the new survey covers smaller firms, it is not surprising that the average sizes of export among exporters are smaller, most noticeably in the business logistic industry. For this industry, the average export value dropped from RM300 million to RM39 million. Once we control for sample size and only focus on the group of common firms in the two surveys, as shown in Table A3.2b, the average size of export sales has improved. For the business logistic industry, it increased from RM20 million to RM62 million. Likewise, for the information technology industry, when we consider the full sample of firms, the average size of export sales dropped from RM64 million in the 1999-2001 period to RM28 million in 2004-2006 period, while it improves from RM27 million to RM127 million for the firms that exported in both the survey periods.59 114. Communication services firms tend to be large and grow rapidly. In terms of number of employees, communication services firms are nearly 3 to 4 times larger than firms in other business support services industries. The average annual sales of a typical communication services firm was RM1.6 billion in the 2004-2006 period, and it was RM1.1billion in the 1999- 2001 period. In contrast, firms in the accounting services were the smallest, with average annual sales of about RM10 million in both the survey rounds. Overall, the growth rate for business services was quite high at 13 percent for the 2004-2006 period and 12 percent for the 1999-2001 period. When focusing on the common group of firms to compare the average sales growth between the two surveyed periods, it is clear that information technology firms outperformed the 59 It is not common for business support services firms, to have export. The extremely small sample clearly plays a role in the sharp changes over time in export. 62 other services firms.60 The communication and accounting industries both experienced positive sales growth, while advertising and logistic industries experienced negative sales growth. SKILLS SHORTAGE, INNOVATION READINESS, AND REGULATORY REGIMES IN SERVICES SECTOR Skills Shortage 115. The shortage of skills has persisted. The shortage of skilled labor is perceived as the "biggest obstacle" to doing business by firms across industries, particularly so in the communication services, advertising and marketing services, and information technology services industries. In Klang Valley, Johor and the Eastern Malaysia, the skilled labor shortage is particularly acute and more services firms considered this the key bottleneck over the past five years. The average time to fill vacancies has improved significantly since 2002, but it still takes a long time relative to other countries. Difficulties finding qualified workers were cited as the reason for the long vacancy fill rate. Majority of the firms indicated that the top reasons were because applicants lack the necessary basic and technical skills for the positions. This indicates that there may have been a mismatch between what firms need and what the labor market can provide in terms of the skill set and potential of the candidates.61 116. Experience, education level and technical skills are what firms in the services sector are looking for to fill a vacancy. When firms are asked to identify what are the most important considerations in recruiting, most firms indicated experience, education and technical skills as top criteria in 2007. Similar criteria were also observed in the 2002 survey. Combined with the previous finding that applicants lack basic and technical skills necessary to fill vacancies, the evidence suggests that the current skill shortage can be tied to a lack of qualified and experienced skilled workers. 117. Foreign skilled workers or Malaysian workers trained overseas may provide some relief for the current skills shortage. One way to address the current shortage of experienced and qualified skilled workers and professionals could be through the provision of temporary work permits of foreign skilled and professional workers. As shown in Figure 3.10, more than 20 percent of the firms surveyed in 2007 consider that foreign skilled workers are better qualified than their domestic counterparts. Likewise, close to 30 percent of the firms consider the Malaysian workers trained abroad are better qualified. The percentages were even higher in 2002. This suggests a deficiency in the local education system in producing marketable graduates. 60 Firms in data processing, web-based, and the software-related industry grew at an average annual rate of 8.6 percent in the 2004-2006 period. 61 See further discussions in Chapter 5. 63 Figure 3.10: Locally Trained Skilled Workers vs. Overseas Trained Skilled Workers Percentage of firms that consider the locally trained Malaysian professional to have a lower quality than a foreign trained professionals 40 35 30 25 % 20 2007 15 2002 10 5 0 Foreign Professional Overseas trained Malaysian Professional 118. Because it is hard to find experienced and qualified workers, more than half of firms provided formal in-house training in 2007(Table 3.1). When compared to the previous survey, the proportion of firms provided in-house training was up by about 10 percent. This number could be even higher if the Human Resources Development Fund (HRDF) was more efficiently administered. In PICS 2007, firms reported that it took 33 days to get any reimbursement on the training cost from HRDF, while in PICS 2002 it was about 27 days. In PICS 2007, 84 percent of firms considered that they will train more workers if HRDF was more efficient, while in PICS 2002, it was 76 percent. For those firms that do not provide in-house training in 2006, 90 percent of them indicated that they would have provided the training if HRDF was more efficient. 64 Table 3.1: Training and Services Sector Firms ­ Percentage of firms say "yes" to the following questions in 2002 and 2007 Employer-Provided Training in the Business-Support Services Sector Malaysia Did your establishment run formal training program? (%) 2007 55 2002 51 Did you hire fresh graduate from public vocational training institutions? (%) 2007 26 2002 28 Have you sent your worker for training in SDI in the past three years? (%) 2007 13 2002 22 How long, on average, did it take HRDF to process your claims for reimbursement in 2001? (days) 2007 33 2002 27 Would you train more workers if HRDF was more efficient? (%) 2007 84 2002 76 Source: Productivity and Investment Climate Surveys, 2002 and 2007. Innovation Readiness 119. The PICS results suggest that technology and innovative activities are declining among those services firms participating in the survey. 62 While more than half of the sampled firms reported that they have upgraded their machinery and equipment in the last two years, the proportion is less when compared to the previous survey five years ago. This kind of decline is common across all the questions listed in Table 3.2 for the column of full sample, which judge the services firms according to their technology and innovative activities. Only 28 percent of the surveyed firms indicated that they entered new markets in 2007, a drop of 8 percentage point relative to 2002. Likewise, the percentage of firms developed a new service line was 18 percent in 2007, and it was 23 percent in 2002; the percentage of firms that upgraded an existing service line was 42 percent, while it was 50 percent in 2002. Finally, 22 percent of the firms signaled that they introduced a new technology which has substantially changed their business practice in 2007, the proportion was 31 percent in 2002. While the decline could be due to the differences in the composition of firms between the two surveys, it is nonetheless worrisome. Technology and innovative activities of the common firms, those that were covered in both rounds of the surveys are however relatively constant. The last column of Table 3.2 restricts the sample to the subset of firms that were covered in both the 2002 survey and the 2007 62 See more discussions in Chapter 6. 65 survey. Indeed in this subset of common firms we see that technology and innovative activities of firms have been steady. While the percentage of firms upgraded their machinery or equipment marginally decline from 64 percent in 2002 to 61 percent in 2007, the proportions of firms introduced new service line or upgraded existing service line have slightly improved to 19 percent and 47 percent respectively in 2007, while it was 17 percent and 44 percent in 2002. 120. While R&D activities are declining among services firms, more of them are getting ISO certification. That indicates the services they offer are becoming more reliable and consistent. Table 3.3 presents the percentage of firms in the business support services sector that have in-house R&D staffs and outsource R&D activities, in 2002 and 2007 for both the full sample as well as the subset of common firms in both the surveys. Results indicating that R&D activities are on a decline, for both the full sample and the group of common firms. This suggests that firms are not doing as much R&D as they were before. In 2007, for the full sample of firms, only 5 percent of the firms have full time R&D staffs in-house and 3 percent of the firms outsourced their R&D activities. The corresponding numbers in the 2002 survey were 9 percent and 4 percent respectively. For the set of common firms, the numbers were 3 percent and 5 percent respectively in 2007, changed from 7 percent and 4 percent respectively in the 2002 survey. However, the percentage of firms that received ISO certifications have increased significantly, from 16 percent to 25 percent for the full sample of firms, and 18 percent to 27 percent for the group of common firms. This suggests that the reliability and consistency of services provided by these firms are improving over time. 121. Firm to firm research linkages improved along with collaboration with government and multilateral agencies. Most firms indicated they learn the most from other firms when it comes to research output. As shown in Table 3.3, in 2007, 49 percent of the surveyed firms and 48 percent of the common firms participated in firm-firm collaboration. The percentage of firms participated in firm-firm collaboration in 2002 was 41 percent of all firms and 42 percent of the common firms respectively. Thus, there is a clear upward swing in firm-firm linkages. Other research linkages that are improving include firm-government and firm-multilateral agencies linkages. For the full sample of firms, 16 percent of them were collaborating with the government agencies in 2002 and the share increases to 21 percent in 2007. Likewise, 15 percent of the full sample firms were collaborating with multilateral agencies in 2002, and the proportion increases to 19 percent in 2007. Similar upward swing is observed in the restricted sample of common firms between the two surveys. The overall profile of research activities have also improved. As more firms collaborate on research, more of them are obtaining ISO certifications. Relative to 2002, we observe that more firms in the business services sector are aware of research and quality controls. This is supported by the improved collaboration between firms as well as more firms are obtaining ISO certifications. 66 Table 3.2: Technology and Services Firms ­ Percentage of firms say "yes" to the following questions in 2002 and 2007 Malaysia Technological activities Full sample Common firms Upgraded machinery and equipment in the last 2 years 2007 54 61 2002 67 64 Entered new markets due to process or service improvements in quality or cost 2007 28 30 2002 36 31 Filed any patents/utility models or copyright protected materials 2007 6 6 2002 6 5 Developed a major new service line 2007 18 19 2002 23 17 Upgraded an existing service line 2007 42 47 2002 50 44 Introduced new technology that has substantially changed the way that the main service is provided 2007 22 27 2002 31 26 Source: Productivity and Investment Climate Surveys, 2002 and 2007. 67 Table 3.3: Key Drivers of Innovation (Adaptation and Creation) Common Innovative activities and linkages (%) Full sample firms Received any ISO certification 2007 25 27 2002 16 18 Have full time R&D staff in house 2007 5 3 2002 9 7 Outsourced R&D activities 2007 3 5 2002 4 4 Collaborate with other firms in research activities 2007 49 48 2002 41 42 Collaborate with local universities in research activities 2007 14 12 2002 14 15 Collaborate with local research institutes in research activities 2007 20 18 2002 17 17 Collaborate with government agencies in research activities 2007 21 20 2002 16 15 Collaborate with multilateral agencies in research activities 2007 19 16 2002 15 15 Source: Productivity and Investment Climate Surveys, 2002 and 2007. Regulatory Regime 122. This section examines business regulations and barriers affecting business support industries. Since the last survey, Malaysia has implemented reforms to reduce regulatory barriers to the services industries. However, to date, the services industries in Malaysia are still considered highly protected relative to other countries with a similar level of development, with many industries facing foreign ownership and key posts restrictions. Below are details on the restrictions facing firms operating in the business support services. 123. Accounting and Taxation Services. No substantial changes in terms of regulatory barriers have been made in this industry since the last survey in 2002. Foreign accounting firms may operate in Malaysia through the local affiliates. Before applying for a license at the 68 Ministry of Finance, all accountants must register with the Malaysian Institute of Accountants (MIA), which has a citizenship or permanent residency requirement. Registration also requires either a local university degree or a membership of at least one of the 11 recognized overseas professional bodies recognized by Commonwealth countries. 124. Architectural Services. No substantial changes in terms of regulatory barriers have been made in this industry since the last survey in 2002. With the approval of the Board of Architects for operating in Malaysia for a specific project, a foreign architectural firm may act as a joint- venture participant of a local firm. However, Malaysian architectural firms may not have foreign architectural firms as registered partners. Foreign architects may not be licensed in Malaysia, which is necessary for any submissions of architectural plans. Nevertheless, foreign architects are allowed to be managers, shareholders, or employees of Malaysian firms. 125. Engineering Services. No substantial changes in terms of regulatory barriers have been made in this industry since the last survey in 2002. Similar to architectural services, foreign engineers must be sponsored by Malaysian firms to operate in Malaysia for any specific project, which are subject to approval of the Board of Engineers. The license is project-specific. To qualify for the license, a foreign engineer must be registered as a professional engineer in his or her home country, have at least 10 years of experience, and physically presence in Malaysia for at least 180 days in a year. To obtain temporary licensing for a foreign engineer, the Malaysian firm must show to the Board that they cannot find a local engineer for the job. Foreign engineers are not allowed to operate independently of Malaysian partners, or serve as directors or shareholders of an engineering consulting firm. A foreign engineering firm may establish a commercial presence in Malaysia if all directors and shareholders are Malaysian. Foreign engineering firms may collaborate with a Malaysian firm, but the Malaysian firm is expected to design and is required to submit the plans. 126. Telecommunications. Changes have been made in January 2005 in the revised services offer in the WTO. Prior to 2005, under the WTO Basic Telecommunications Agreement, Malaysia guarantees market access and national treatment for most basic telecommunications services only through acquisition of up to 30 percent of the shares of existing licensed public telecommunication operators and limits market access commitments to facilities-based providers. This makes Malaysia's regime one of the most restrictive among economies with similar levels of development. Value-added services suppliers also face the 30 percent foreign equity restriction, which limits competition and may benefit incumbent dominant providers. However, since February 1, 2000, the Communications and Multimedia Act has come into effect, which is Malaysia's first legislation to address anticompetitive activities. The results of the law have yet to be seen. Currently the government has allowed foreign equity in the sector to a maximum of 61 percent for the first five years, and 49 percent afterwards. However, companies with MSC- status are allowed to have full foreign ownership. Since January 2005, foreign equity is allowed up to 49 percent in the "application service providers" category. 127. Wholesale and Retail Trade. More restrictions have been put in place since December 2004. Foreign involvement in the wholesale and retail trade requires having a local incorporation. Foreign equity of 30 percent is allowed with at least 30 percent to be reserved for Bumiputeras. There are also minimum capital requirements and restrictions on the number of expatriates for managerial or technical posts (one key post per company with a maximum of 69 10 posts total). However, recent developments show that increasingly more licenses are issued to foreign-owned hypermarkets, which works to promote liberalization and competition in this sector. 128. Advertising and Marketing. Advertising commercials are restricted to a maximum of 20 percent foreign film content. The government recently relaxed enforcement of regulations governing the appearance of foreign actors in commercials. Commercials cannot "promote a foreign lifestyle." Table 3.4: Qualitative Comparison of Business-Support Services Regulation in Malaysia, Taiwan, Korea, Chile, Singapore and Thailand Foreign Ownership Restrictions Foreign Professionals Restrictions Malaysia <=30% in general except in insurance, Not permitted to practice as lawyers, banking and financial services industries. engineers, architects, accountants, among other professions. Taiwan No restriction in other industries in general; Restrictions have lifted on foreign Some technical limits on portfolio investment may lawyers and security brokers. apply; Technical limits in telecommunication services . Korea No specific restriction in general, except the media Generally permissable if that the necessary sector which is partially closed to foreigners: local qualification are met or obtained. <=49% in cable TV, network operator; <=33% in satellite broadcast; foreign investment not permitted for broadcast TV operation. Chile No specific restriction in general. No specific restrictions in general. Singapore No restriction in basic telecommunication services Highly restricted to foreign lawyers. <=49% of foreign ownership of companies Local professional certification required for broadcasting to the domestic market engineers and architects. <=5% per shareholder, local or foreign, in newspaper. Thailand <=49% in telecommunication series; Not permitted to practice as lawyers, <49% in land transport; engineers, architects, accountants, <=25% in banks; among other professions. Prohibited in legal services (except the US) Source: USTR (2006). Note: On March 10, 2009, the Government took steps to liberalize the services sector. Measures have been taken to liberalize 27 services subsectors, with no equity condition imposed. These sub-sectors are in the areas of health and social services, tourism services, transport services, business services and computer and related services. FIRM PERFORMANCE AND INVESTMENT CLIMATE Correlates Between Firm Characteristics and Performance 129. This section investigates the extent to which a business support services firm's efficiency is associated with the level of education of its workers, their training, export orientation, ownership structure, and new entrants. Several performance indicators are 70 examined, specifically, sales, growth rate of sales, labor productivity (as measured by sales per worker), and firm productivity. The first three indicators are observable and are positively correlated. Firm productivity, on the other hand, is not observable. Given that information on the quantity and prices of output at the firm level is unavailable in the survey, neither construction of TFP measures nor the estimation of firm productivity is feasible. Nevertheless, it is still possible to estimate the effects of the aforementioned factors on firm productivity in a regression analysis framework, which follows. 130. The literature contains many hypotheses about the driving force behind a firm's performance. They can be summarized as follows: · Competition enhances efficiency ­ entry and exports promote productivity gains (Bernard and Jensen [1999]) · Foreign investment enhances technology transfer ­ restrictions on foreign ownership deter productivity gains (Blomstrom and Sjoholm [1999], Javorick and Spatareanu [2003]) · Foreign firms or new entrants steal market shares from existing domestic firms ­ restrictions on foreign ownership or entry may help protect domestic firms (Aitken and Harrison [1999]) · Physical and human capital investments enhance labor productivity (Mankiw, Romer and Weil [1992]) 131. Regression analysis helped to test the preceding hypotheses on the services industry firms sampled in the 2002 and 2007 surveys.63 Of these surveyed firms, 137 are common across the two surveys and 128 of these common firms have positive sales figures. The observable performance indicators of firms are first examined, and then the estimation procedure on unobserved productivity will be discussed. The pair-wise correlation coefficients between log of sales, and growth of sales, and log of labor productivity are 0.15 and 0.68, respectively, and the correlation between the growth of sales and the log of labor productivity is 0.19. Annex 3.2 reports the partial correlations of the different observable performance indicators of the firms with the log assets value per worker, share of exports in sales, foreign ownership (positive foreign equity) and firm age, controlling for year, region and industry fixed effects. The first three columns of the table are estimated by panel regressions using the between firm variation, while the last column uses the data for 1999-2001 and 2004-2006 in a Ordinary Least Square (OLS) regression, controlling for year, region and industry fixed effects. 132. It is clear that firms with more capital (as measured by the asset value per worker), tend to perform better when it comes to sales revenue and labor productivity. This could be related to economies of scale which reward larger firms and the contribution of capital. Similarly, exporting firms are better performers than the non-exporting firms. Older firms tend to be larger, but not necessarily more productive in terms of growth and labor productivity than the younger firms. While these three factors are robust in expected signs in all columns, they are 63 Outliers have been trimmed for the analysis. 71 statistically significant in at least three out of the four columns reported in Annex 3.2. On the other hand, the effect of foreign ownership on firm size and sales is positive and significant, which suggests that firms with foreign equity tend to be larger, and may be more productive. Given that some of the industries in the business-support services sector are subjected to the 30 percent foreign ownership restriction, a better specification of this variable is therefore necessary. 64 As previously mentioned, firm productivity is unobservable, and given the lack of information on price and quantity of output, construction of TFP is infeasible. Nevertheless, the following specification allows for the estimation of the effect of various determinants on firm productivity without explicitly estimating firm productivity. Specifically, labor productivity (log value of sales per worker) is regressed on the log value of assets per worker. Under the assumption of constant returns to scale production function, the residuals of such a regression capture the unobserved productivity, provided that good controls for firm prices are included. Year, region and industry fixed effects are used, in a between estimation to control for firm prices. In other words, once the movement of labor productivity that is due to prices and capital intensity is isolated, the remaining part of the movement of labor productivity is attributed to the movement of firm productivity.65 133. We test the hypothesis that unconstrained firms are more productive than their constrained counterparts and domestic firms. We expect this variable to be positive and significant.66 Table A3.4 in Annex 3.3 reports the regression results on the effects of human capital (log of average years of education per worker), export share in sales, foreign ownership, training, and new firms (aged 5 years or less) on firm productivity, where column (1) shows the between panel estimation, controlling for only year fixed effects and column (2) further controls for industry and region fixed effects as a robustness check. In addition, given that many industries in Malaysia are subject to the 30 percent foreign ownership restriction, and the PICS I report has found that the constrained firms are less productive, we only include a foreign ownership dummy which equals one if the firm's foreign equity is more than 30 percent; given that many variables of interest are firm-specific and time invariant, the hypotheses can only be tested using between estimation, which only takes into account the variation of the variables between firms. It is clear that these results are very similar, even after we control for the human capital stock of the firms (log of average years of education per worker), export exposure (share of exports in sales), firm age and in-house training of the firms. 134. The results suggest that firm productivity is highly correlated to other firm performance indicators (Table A3.4). Firm with unconstrained foreign equity are in average 40 percent more productive than other firms in the business support services sector. This effect is large and significant. Exporting firms are more productive, so are those firms that provide in- house worker training, although these effects cannot be precisely estimated. Average year of 64 According to PICS II, 13 percent of services firms' list ownership regulations as one of three biggest constraints ­ it increased from 9 percent in the PICS I; foreign owned firms are more likely to recognize ownership regulation as the biggest obstacle. About 20 percent of foreign owned firms consider it one of the biggest obstacles. However, the numbers should be interpreted with caution due to the limited number of observations. 65 In order to explain the effects of various determinants on firm productivity when firm productivity is unobservable, those variables can simply be included in the regression to obtain unbiased estimates. 66 In other words, the assumption is that the constrained firms (that is, firms that have foreign equity less than or equal to 30 percent) are less productive, and that the unconstrained firms with foreign equity greater than 30 percent, are more productive. 72 schooling of workers are however not positively correlated with the unobserved productivity. Younger firms are again shown to be more productive than the older firms, suggesting that younger firms may be able to tap into the latest technology which makes them leap frog the older firms. Impact of the Presence of Foreign Firms 135. Domestic firms tend to benefit from the presence of foreign firms (Table A3.5 in Annex 3.3).67 Restricting the sample to domestic firms that have no foreign investment in their equity, regression results suggest that domestic firms benefit significantly from the presence of foreign firms in the same industry. For the business support services sector, a 10 percent increase in industry sales that is originated from firms with foreign equity increases the productivity of domestic firms in the same industry by 7.5 percent. The effect is larger when we only consider those firms with more than 30 percent foreign equity ­ a 10 percent increase in industry sales that is originated from these unconstrained firms increases the productivity of domestic firms in the same industry by 9.8 percent. 136. Industry-wide, the presence of foreign-owned firms has a big impact on the productivity of domestic firms. This can be related to the fact that foreign firms bring in advance technology and know-how, from which domestic firms may learn through the networking of personnel and contacts. Such an effect is also confirmed by a firm survey conducted in Lithuania, UK, and US.68 CONCLUSIONS 137. A number of salient features about the Malaysian services sector emerge from a data analysis of national accounts, the investment climate and cross-country comparisons. First, the performance of the different sectors within the services sector varies. Communication services leads in size and IT in growth. Second, the firm-level indicators of performance (e.g., sales growth, growth in value-added per worker) suggest that the business support services sector in Malaysia is doing well, despite signs of slowing. Third, a shortage of skilled labor and high taxes top the list of obstacles to doing business. In the past five years firms appeared to be more concerned about macroeconomic instability, economic policy uncertainty and corruption. 138. In the business support services sector, firm performance is closely aligned with the characteristics of individual businesses. These characteristics include asset per worker, foreign ownership, export orientation, training of workers, age of firms, and the education level of workers employed in the firms. The presence of foreign-owned firms, especially those not constrained by the 30 percent foreign equity restrictions, tends to increase the productivity of domestic firms in the industry through knowledge spill-over. 139. For the services sector to grow, reform is imperative. First, costly regulatory burdens need to go down. Second, reforming the education system, enhancing skills training, and 67 Firms' type of ownership varies across industries in the services sector. Based on PICS 2007, about one-third of advertising and marketing firms have over 30 percent foreign ownership, compared with one-quarter of IT firms, one-fifth on telecommunication firms, one-seventh of business logistics firms, and one-thirtieth of accounting firms. 68 See Smarzynska (2002), Haskel et al. (2001), and Keller and Yeaple (2003). 73 liberalizing restrictions on the inflow of professionals would be helpful in closing the gap on the shortage of skills. Third, more analysis is needed to identify specific constraints facing different services sub-sectors. The next survey will need to include more firms from a larger group of the services sector to accurately assess the investment climate, supply side constraints, and deficiencies in the business environment that are likely to hold back this sector as a source of growth over the medium-term in Malaysia. 140. Specific objectives that follow from the findings include: improving and monitoring the investment climate in the services sector; increasing on-the-job and off-the-job training, especially among SMEs. Activities to achieve these goals for the services sector include carrying out periodic firm-level surveys; improving and adapting the survey questionnaire especially questions on firm operations such as material cost; carrying out detailed diagnostics of the regulatory regime; encouraging firms, especially SMEs, to establish and expand in-house training; and improving the efficiency of the HDRF. Furthermore, interagency coordination in the services sector needs improvement. 74 CHAPTER 4. REGIONAL PERSPECTIVES OF INVESTMENT CLIMATE 142. Malaysia's economic performance varies by state and region. Five development corridors have been established to reduce regional disparity, move up the value chain, raise the capacity for knowledge and innovation, improve the standard and sustainability of quality of life, and strengthen institutional and implementation capacity as thrusts of the Ninth Malaysia Plan. Improving investment climate is important for the development corridors to reach their full potential. 143. This chapter investigates the subjective perceptions firms have at the regional level as well as objective measures of investment climate. 69 It focuses on labor skills, affordability of business support services, public infrastructure, government regulatory framework, access to finance, and firm's effort for innovative activities. Klang Valley is perceived by manufacturing firms as well as selected business supporting services firms as the region with the best investment climate, while Sabah the least business-friendly. Moving a production line from a less business-friendly location to a more business-friendly location is believed to significantly save firms' operating costs. In the past years, the perceived gaps between regions with the best and the least satisfactory investment climate appear to remain wide. However, the picture that the objective measures provide is mixed. Indeed, firms in Sabah are more likely to experience a less satisfactory business environment, such as limited access to affordable business support services, slower import and export customs procedures, and larger costs associated with crime and theft. The investment climate in Klang Valley appears to be middling among other regions in many aspects. 144. Inadequate skills are a major obstacle for firms in doing business in almost all regions. For manufacturing firms, the skills issues seemed to have been slightly improved in the past five years in Klang Valley, in the South, and in Sarawak; for business support services firms, this constraint became more acute, especially in Sabah and the South. 145. This chapter is structured as follows. Section 1 discusses the regional development corridors. Section 2 presents firms' perceptions of regions with the best and the least satisfactory investment climate. Section 3 describes the subjective ranking of top investment climate constraints and rating of severity of obstacles in each region. Section 4 reviews the objective measures of key investment climate aspects. Section 5 concludes. 69 The analysis at the regional level is based on the selected firms located in the respective regions covered by the surveys. The findings cannot be generalized. The number of the observations is limited in certain cases, especially for business support services firms. For this reason, this chapter places more emphasis on manufacturing firms. 75 REGIONAL DEVELOPMENT CORRIDORS 146. Malaysia's economic performance varies by state and region. In 2000, GDP per capita is the highest in Wilayah Persekutuan Kuala Lumpur, followed by Terengganu and Penang.70 The level of GDP per capita in Wilayah Persekutuan Kuala Lumpur is almost five times that in Kelantan. In 2000-2005, the two richest states, Wilayah Persekutuan Kuala Lumpur and Terengganu, grew at a lower rate compared to the national average; while Sarawak and three of the four states in the North, including Penang, Perak, and Kedah, registered a high growth rate.71 A large share of FDI locates in Klang Valley, while only a small share in the East and Sabah. Figure 4.1: GDP per capita in 2000 and Annual Growth Rate 35000 6 30000 5 25000 4 20000 3 15000 2 10000 5000 1 0 0 Johor Terengganu Selangor Melaka Sembilan Kedah Kelantan Penang Pahang Sabah Persekutuan Perak Perlis Sarawak Negeri Wilayah Kuala Klang Valley North South East Coast Sabah Sarawak Per Capita GDP in 2000 - Left Scale Annual Per Capita GDP Growth (2001-05) (%) - Right Scale Sources: Eighth Malaysia Plan and Ninth Malaysia Plan, Economic Planning Unit Note: GDP per capita is measured in RM 1987. 147. Five development corridors have been established. The key objectives are to reduce regional disparity, move up the value chain, raise the capacity for knowledge and innovation, improve the standard and sustainability of quality of life, and strengthen institutional and implementation capacity. As part of the Ninth Malaysia Plan (2006-2010), the Malaysian government has introduced five development corridors that span throughout the country. Three of these are located in the Northern, Eastern and Southern parts of Peninsular Malaysia, while the other two, Sabah and Sarawak, are in the Borneo Island. Table 4.1 provides details of the corridors. The principal aim of each development corridor is to attract new private investments, both local and foreign, into the focus sectors or industries of the region. To accomplish this goal, there are plans to enhance key factors that influence investment decisions such as physical infrastructure, human capital, and government regulations. In many cases, specific incentive packages are also given to investors. It is envisaged that the five development corridors will not 70 Source: http://www.epu.jpm.my/new%20folder/development%20plan/RM8/c5_cont.pdf 71 Source: http://www2.seri.com.my/9MP_presentations/RMK9_PPinang29042006%2020Presentation.pdf 76 only support Malaysia to become a developed nation by 2020 (Vision 2020) but also reduce income disparity between Klang Valley and other parts of the country. Table 4.1: Five development corridors in Malaysia Iskandar Northern East Coast Sabah Sarawak Corridor Malaysia Corridor Economic Development of Renewable Economic Region Region Corridor Energy Region name in PICS South North East Sabah Sarawak Coverage area District of Penang, Kedah, Pahang, Entire Sabah Tanjung Manis- Johor Bahru Perlis and Kelantan, state Similajau and and partial Northern Perak Terengganu and hinterland district of district of Pontian Mersing, Johor Area size (km2) 2,216 17,816 66,736 73,997 70,708 Development period 2006-2025 2007-2025 2007-2020 2008-2025 2008-2030 Focus sector - Education - Agriculture - Agriculture - Agriculture - Aluminium - Financial - Human capital - Education - Environment - Glass - Healthcare - Infrastructure - Manufacturing - Human capital - Marine engineering - ICT/creative - Manufacturing - Oil, gas & - Infrastructure - Metal-based industries - Tourism petrochemical - Manufacturing - Petroleum-based - Logistics - Tourism - Tourism - Timber-based - Tourism - Aquaculture - Livestock - Palm oil - Tourism Expected investment 382 178 112 113 334 (RM billion) Expected 1.4 3.1 1.9 2.1 3.0 employment (million) Source: Adapted from Box 3-2 in Mid-term Review of the Ninth Malaysia Plan, Economic Planning Unit (2008) Note: Expected values of investment and employment are by the end of respective development period. 148. Although they share a similar development goal, the five development corridors focus on different sectors. In Iskandar Malaysia, the focus sectors are more services-oriented, including education, financial, healthcare, ICT and creative industries, logistics, and tourism; while the Sarawak Corridor focuses more on resource-based production. However, manufacturing is a common focus sector of all these development corridors. Manufacturing sector will remain critical in the Malaysian economy. Vision 2020 envisages that by 2020 the share of manufacturing activities in GDP will remain relatively unchanged from the current level. 149. These five economic regions aim to generate substantial investments and jobs. Total expected investment amounts to RM 1,119 billion or 1.5 times Malaysia's GDP in 2008. 72 Total job creation is targeted at 11.5 million workers, which is about the same size as the current workforce. Achieving these targets is especially challenging given the current global economic crisis that may dampen FDI to emerging economies worldwide. 72 With comparable development period (up to 2025), the Iskandar Malaysia plans to attract investment value that is about 3.4 times higher than the Sabah Development Corridor although its area size covered is 33 times smaller. 77 REGIONS WITH THE BEST AND WORST BUSINESS ENVIRONMENT 150. Overall, manufacturing and services firms view the Klang Valley as the region with the best investment climate; they regard Sabah as least satisfactory. Manufacturing Firms' Perception 151. Manufacturing firms in all regions viewed Klang Valley as the region with the best business environment.73 Table 4.2 reports manufacturing firms' perception about the best region to do business in Malaysia. Over 80 percent of firms located in Klang Valley believe their region has the most favorable investment environment. In all other regions, the share of firms that viewed Klang Valley as the best region is higher than those that cited their own region. The South and the North appear to be the next best places, although the North has become less attractive as a manufacturing base in recent years. In PICS 2002, up to 16-38 percent of firms located in Klang Valley, the South, and Sabah cited the North as the best region. In PICS 2007, these figures went down to only 0-7 percent. Table 4.2: Manufacturing Firms Perception on Regions with the Best Business Environment Region where the firm is located now Klang Valley North South East Sabah Sarawak Klang Valley 82.8 45.1 47.3 60.7 69.0 62.5 Region with North 5.2 44.7 7.3 3.6 0.0 0.0 best business South 9.9 5.8 39.8 7.1 0.0 0.0 environment East 1.5 2.7 2.5 25.0 0.0 3.1 in firms view Sabah 0.6 1.3 1.9 3.6 24.1 3.1 Sarawak 0.0 0.4 1.3 0.0 6.9 31.3 Source: Malaysia PICS 2007 152. Relocating to a more business-friendly region could markedly enhance a firm's cost advantage. For example, firms in the East estimated that moving to Klang Valley would lower their production costs by nearly 20 percent. Across regions, the mean value is about 13 percent in 2007.74 This is comparable to the value in PICS 2002 (12 percent), thus suggesting that the gaps between the best and worst performing regions remain relatively wide. The perceived worsening business environment in the North is also reflected in this estimate. In 2002, firms in the North believed moving to Klang Valley would save the production cost by 9 percent. By 2007, this increased to 14 percent. Meanwhile, the superior business environment in Klang Valley does not 73 Different aspects of the investment climate may be more or less binding constraints depending on the characteristics of each enterprise. The percentage of firms perceiving a particular region as having the best investment climate was much higher among firms actually located in that region compared to firms located elsewhere. This is expected because firms choose to establish themselves where they can maximize profit (therefore, where they perceive that the investment climate is most suitable for their growth, ceteris paribus). For the same reason, very few firms perceived the region where they are located as having a poor business climate. The high percentage of firms perceived Klang Valley as the region with the best investment climate could be partly associated with the concentration of firms in the region. 74 These estimates are derived from the samples that exclude firms that believe they are already located in the best region, i.e. bolded cells in Table 4.2. 78 seem to benefit firms in different industries equally. A large share of firms that wish to move into and out of Klang Valley is in the food processing and rubber/plastics industries.75 153. Sabah has the least satisfactory business environment; the East is next. Table 4.3 reports firms' perception on regions with the worst business environment. Around 20 percent of firms currently based in Sabah perceived their own region as the worst location. At least a quarter of firms in all regions other than Sabah also indicated Sabah the least business- friendly. Similar is true for the East. Moving a production line to an area with poor business environment can be very costly. On average, firms estimated that production cost would rise by approximately 17 and 25 percent if they were based in the East and Sabah, respectively. Compared with the estimates in PICS 2002 (11 and 27 percent), there appears no major improvement in Sabah and worsening environment in the East. Klang Valley is the preferred choice for producers, as opposed to Sabah and the East, irrespective of angles we looked at. Table 4.3: Manufacturing Firms Perception on Regions with the Worst Business Environment Region where the firm is located now Klang Valley North South East Sabah Sarawak Klang Valley 3.3 2.5 4.9 9.1 0.0 8.7 Region with North 7.3 4.0 9.2 9.1 15.0 13.0 worst business South 4.4 7.0 7.2 18.2 15.0 4.4 environment in East 37.4 39.8 26.5 9.1 25.0 47.8 firms view Sabah 23.8 24.9 29.1 31.8 20.0 26.1 Sarawak 23.8 21.9 23.2 22.7 25.0 0.0 Source: Malaysia PICS 2007 Services Firms' Perceptions 154. In general, business support companies view the Klang Valley as the most attractive region. Nearly 90 percent of services firms in Klang Valley believe their current base is already the most competitive (Table 4.4). At least 60 percent of those in the North and the South also share this view.76 On average, relocating a services company to Klang Valley could save around 9 percent of operating cost, which is not much different from the estimate in PICS 2002 and slightly lower than the potential saving for manufacturing firms. Again, this suggests that regional investment climate disparity is still relatively wide over the past several years. 75 Firms that wish to move into Klang Valley are firms that are currently located outside Klang Valley but believe Klang Valley has the best business environment. Firms that wish to move away from Klang Valley are firms that are currently based in Klang Valley but view other region as the best location to do business. 76 The two states on the Borneo Island (Sabah and Sarawak) prefer Sarawak over Klang Valley but this may be related to the limited number of observations in the sample and the specific characteristics of the firms surveyed. 79 Table 4.4: Services Firms' Perceptions on Regions with the Best Business Environment Region where the firm is located now Klang Valley North South Sabah Sarawak Klang Valley 89.7 61.8 61.9 33.3 40.0 Region with North 5.4 38.2 0.0 0.0 0.0 best business South 3.9 0.0 28.6 0.0 0.0 environment East 1.0 0.0 4.8 0.0 0.0 in firms view Sabah 0.0 0.0 4.8 25.0 6.7 Sarawak 0.0 0.0 0.0 41.7 53.3 Source: Malaysia PICS 2007 155. Services providers view the business environment in the East and Sabah as the least favorable. Except in Sabah, at least a third of firms in all other regions view the East as the least business-friendly region. The PICS samples for services firms unfortunately do not cover the East, thus not allowing us to see if the firms there also share this view.77 Moving a services firm to the East would raise the operating cost by up to 13 percent on average, which is only 2 percentage points lower than the estimate in 2002. The perception of poorer business environment in the East has been persistent; this was also the case in PICS 2002. Table 4.5: Services Firms' Perceptions on Regions with the Worst Business Environment Region where the firm is located now Klang Valley North South Sabah Sarawak Klang Valley 1.1 2.9 0.0 10.0 0.0 Region with North 5.0 2.9 10.5 10.0 0.0 worst business South 6.1 8.8 5.3 0.0 6.3 environment in East 47.8 47.1 36.8 20.0 37.5 firms view Sabah 20.0 11.8 31.6 20.0 50.0 Sarawak 20.0 23.5 15.8 40.0 6.3 Source: Malaysia PICS 2007 SUBJECTIVE MEASURES OF INVESTMENT CLIMATE ACROSS REGIONS 156. A shortage of skilled labor, tax rates and tax regulations are considered the top business obstacles across regions. Top Investment Climate Constraints Manufacturing Firms 157. Manufacturing firms in all regions except the East ranked the shortage of skilled workers as their top obstacle (Table 4.6). Over 20 percent of firms in Klang Valley, North, South, and Sabah cited skills shortage their top concerns. Between 2002 and 2007, firms' perception about skills shortage has eased in most regions, while tax issues gained more significance. This seems to suggest that a gradual cut in the corporate tax rate since 2006 has not 77 More importantly, the absence of survey responses by services firms in the East makes it difficult to clearly understand what specific business obstacle(s) that is discouraging the operation of business support services there. 80 been seen as substantial enough by manufacturers78 and there might be distortions in the taxation system which lowered firms' overall sentiment, which was expressed as concerns of tax rate and tax regulations. Table 4.6: Manufacturing firms: The Most Important Business Obstacles in 2007 by Region (open-ended question) Share of firms citing Change Region Most important business obstacle these obstacles in PICS from 2002 2007 1. Skilled labor shortage 23.1 - 5.0 Klang 2. Tax regulations and/or high taxes 12.9 + 3.1 Valley 3. Insufficient demand for products 8.0 - 0.2 1. Skilled labor shortage 20.2 - 0.7 North 2. Tax regulations and/or high taxes 13.0 + 2.9 3. Ownership regulations 8.3 + 6.6 1. Skilled labor shortage 22.1 - 7.8 South 2. Bureaucratic burden 8.5 + 5.2 3. Tax regulations and/or high taxes 8.2 - 2.2 1. Lack of business support services 17.9 - 1.2 East 2. Ownership regulations 14.3 + 14.3 3. Skilled labor shortage 14.3 - 4.8 1. Skilled labor shortage 23.3 + 6.7 Sabah 2. Tax regulations and/or high taxes 16.7 + 16.7 3. Competition from imports 13.3 + 9.2 1. Skilled labor shortage 13.0 - 10.5 Sarawak 2. Tax regulations and/or high taxes 10.9 + 7.9 3. Ownership regulations 8.7 + 2.8 Source: Malaysia PICS 2002 and PICS 2007 Services Firms 158. Similar as manufacturing firms, a shortage of skilled labor, tax regulations and high tax rates are key constraints for the business services industry. The lack of skills is perceived to be more severe in Klang Valley, the South, and Sabah (Table 4.7). The issues of crime and theft and bureaucratic burden in the North are new. Both manufacturing and services firms in the North are increasingly more concerned about security. 78 The government has cut corporate income tax rate by one percentage point per year since 2006 (from 28 percent in 2006 to 25 percent in 2009). 81 Table 4.7: Services Firms: The Most Important Business Obstacles in 2007 by Region (open-ended question) Share of firms citing Change Region Most important business obstacle these obstacles in PICS from 2002 2007 1. Skilled labor shortage 21.4 + 7.2 Klang 2. Tax regulations and/or high taxes 12.6 + 7.3 Valley 3. Lack of business support services 9.7 + 5.6 1. Tax regulations and/or high taxes 11.4 + 5.9 North 2. Crime and theft 11.4 + 11.4 3. Bureaucratic burden 11.4 + 11.4 1. Skilled labor shortage 18.2 + 9.9 2. Tax regulations and/or high taxes 13.6 + 13.6 South 3. Lack of business support services 9.1 + 0.8 4. Crime and theft 9.1 + 0.8 1. Skilled labor shortage 42.9 + 34.5 Sabah 2. Tax regulations and/or high taxes 14.3 + 14.3 3. Political instability 14.3 + 6.0 Sarawak 1. Tax regulations and/or high taxes 44.4 + 38.2 Source: Malaysia PICS 2002 and PICS 2007 Note: For Sarawak, the number of observations is limited. There are many business obstacles that are equally ranked the second. These include, among others, bureaucratic burden, crime and theft, labor regulations, lack of business support services, ownership regulations, and regulations on starting/expanding businesses. Severity of Investment Climate Constraints Manufacturing Firms 159. Throughout the regions, a large percentage of firms rated a shortage of skills and the tax rate/tax regulations as severe or very severe constraints (Table 4.8). In addition, manufacturers in Klang Valley and Sabah (as well as those in the North and Sarawak) considered economic environment and uncertain macroeconomic conditions as important concerns, while those in the East were more concerned about the limited availability of affordable business support services. Region-specific problems include poor electricity services and inflexible labor regulations in Sabah79, and high perception of crime and theft rates in the North. 79 Around one quarter of Sabah's population is immigrants (both regular and irregular) who come mainly from the Philippines and Indonesia. This could explain why firms in Sabah feel that labour regulations are more stringent there. In particular, around 25 percent of firms in Sabah indicated that dealing with procedures of expatriates and of foreign immigrant workers is a major business obstacle. Other aspects of labour regulations surveyed in PICS such as procedures of local workers, limits on temporary hiring, and inflexible salary scale for skilled workers are cited by much fewer firms. 82 Table 4.8: Manufacturing firms: Severity of business obstacles by region (close-ended question) Share of firms viewing these Klang obstacles as severe Valley North South East Sabah Sarawak Telecommunications 8.4 9.3 7.1 0.0 6.5 8.7 Electricity 16.1 10.5 18.8 0.0 29.0 10.9 Transportation 12.0 8.2 9.7 10.3 9.7 15.2 Access to land 9.4 6.2 1.8 6.9 6.5 10.9 Tax rates 21.7 24.9 15.3 20.7 32.3 13.0 Tax admin 18.4 18.7 10.9 10.3 25.8 10.9 Customs and trade regulations 13.8 16.0 12.4 3.5 16.7 8.7 Labor regulations 12.8 11.7 13.6 3.5 22.6 17.4 Skills and education of workers 20.0 19.3 21.2 20.7 12.9 8.7 Business licensing and registration 10.5 11.3 9.2 0.0 16.1 8.7 Access to domestic credit 11.2 14.8 10.9 6.9 16.7 8.7 Access to foreign credit 6.9 6.6 4.7 6.9 3.2 0.0 Cost of financing (interest rates) 20.4 21.3 13.6 6.9 22.6 23.9 Econ policy uncertainty 22.5 18.3 14.4 6.9 12.9 10.9 Macroeconomic instability 30.2 30.4 19.4 3.5 35.5 28.3 Corruption 14.6 16.4 10.9 3.5 0.0 10.9 Crime, theft and disorder 17.1 21.8 16.8 10.3 6.5 13.0 Anti-competitive practices 16.3 15.6 8.5 6.9 6.7 10.9 Immigration 7.4 9.7 9.1 3.5 16.1 4.4 Source: Malaysia PICS 2007 Note: Numbers in bold represent at least 20 percent of all surveyed firms in respective region Services Firms 160. Services firms share a common concern over economic policy uncertainty and macroeconomic instability (Table 4.9). Firms in all regions are concerned about certain aspects of macroeconomic conditions, i.e., access to credits in the South, high borrowing costs in the North and Sabah, and macroeconomic instability in Klang Valley and Sarawak. Corruption, Crime and theft, and anti-competitive practices also came out as big constraints in some regions. 83 Table 4.9: Services Firms: Severity of Business Obstacles by Region (close-ended question) Share of firms viewing these Klang obstacles as severe Valley North South Sabah Sarawak Telecommunications 8.3 11.4 4.8 14.3 16.7 Electricity 7.8 5.9 4.8 7.1 11.8 Transportation 10.4 3.1 9.5 7.1 5.9 Access to land 9.8 3.3 5.6 0.0 5.9 Tax rates 19.4 28.1 19.0 28.6 5.9 Tax admin 11.6 21.2 9.5 21.4 11.8 Customs and trade regulations 14.1 14.8 4.8 7.7 16.7 Labor regulations 10.8 9.1 9.5 7.7 16.7 Skills and education of workers 22.3 14.7 9.5 14.3 22.2 Business licensing and registration 9.9 8.6 4.8 0.0 16.7 Access to domestic credit 8.9 16.7 5.6 0.0 20.0 Access to foreign credit 6.7 11.8 7.7 0.0 33.3 Cost of financing (interest rates) 14.1 21.2 5.0 8.3 12.5 Econ policy uncertainty 22.5 33.3 15.8 54.5 37.5 Macroeconomic instability 23.7 29.0 10.5 30.8 38.5 Corruption 22.9 27.6 17.6 50.0 28.6 Crime, theft and disorder 17.7 31.3 29.4 0.0 35.7 Anti-competitive practices 20.7 16.1 11.8 9.1 41.7 Source: Malaysia PICS 2007 Note: Numbers in bold represent at least 20 percent of all surveyed firms in respective region OBJECTIVE MEASURES OF INVESTMENT CLIMATE ACROSS REGIONS 161. Objective measures create a mixed picture. Indeed, firms in Sabah are more likely to experience a less satisfactory business environment, such as limited access to affordable business support services, slower import and export customs procedures, and larger costs associated with crime and theft. The investment climate in Klang Valley appears to be middling among other regions in many aspects. Labor Skills Manufacturing Firms 162. Filling job vacancies often takes a long time in the North, followed by the South and Klang Valley. Figure 4.2 depicts the number of weeks that manufacturing firms on average took to fill the most recent job vacancy for professional staff, skilled production worker, and unskilled production worker. Here professionals include trained and certified specialists such as engineers, scientists, software programmers, lawyers, and other university graduates. Skilled production workers generally refer to skilled technicians who involved directly in the production process. Firms in the North spent the longest time to find suitable employees of all types, followed closely by the South and Klang Valley. For example, finding professional staff and skilled production workers took on average over six weeks in the North relative to only about two weeks in Sabah. Although the overall results suggest that highly trained workers are in short supply in 84 Peninsular Malaysia, this can be related to the industrial composition in each region.80 Compared with 2002, time to fill vacancies has become much longer in the East (and the North to a lesser extent). Figure 4.2: Number of Weeks Required to fill Job Vacancies in Manufacturing Firms Sabah Sarawak East Klang Valley South North 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 Profes sional Skilled production workers Uns killed production workers Source: Malaysia PICS 2007 163. A higher share of firms in the Klang Valley and the East perceived a shortage of skills and a mismatch of skills are the chief reasons why job vacancies are difficult to fill. Job vacancies arise because, among others, the quantity of available workforce is inadequate locally (labor shortage), most workers exhibit poor basic skills (skill shortage), and/or workers sufficiently possess certain skills but these are not skills looked for by firms (skill mismatch). The PICS data suggest that labor shortage is not very severe in Malaysia. When asked to rate the importance of selected causes of job vacancies, only 2-13 percent of manufacturing firms across regions cited insufficient number of university graduates as important. Also, no applicants for unskilled worker position are viewed as important by 17-23 percent of firms across regions. Skill shortage and skill mismatch are more influential. Figure 4.3 displays the share of producers that rated the lack of required basic and technical skills as important causes of vacancies. Overall, workforce in the East appears to be least satisfactory. At least half of firms in these two regions believe job seekers lack basic and technical skills firms looked for. Shortage of basic skills is also notable in Sarawak. In addition to skill issues, some vacancies arise because applicants demand high wage, a problem which is more common in the North, the East, and Sarawak. High turnover of new recruits also leads to job openings in Sarawak. Overall, skill shortage, skill 80 A region may experience long time-to-fill vacancies when qualified workers working in all (or most) industries are in short supply or, alternatively, when that region is dominated by only a few industries in which suitable staffs are hard to find. It takes firms shorter time to fill vacancies in Sabah and Sarawak. One possible reason can be because they have much higher share of food processing firms (where skills needed are relatively more abundant) than states in Peninsular Malaysia. On average, food firms spent 4.2 weeks to find a qualified professional staff, which is lower than the all-industry average of 5.4 weeks. Rubber and plastics firms, for example, took 5.7 weeks. 85 mismatch, and high wage demanded largely explain why firms in the North spent longer time to find suitable employees.81 Figure 4.3: Share of Manufacturing Firms Citing Lack of Basic and Technical Skills as the Cause of Job Vacancies Sarawak Sabah South North Klang Valley East 0.0 10.0 20.0 30.0 40.0 50.0 60.0 Applicants lack technical skills Applicants lack basic skills Source: Malaysia PICS 2007 164. What production workers lack most is proficiency in English, IT and communication skills. Figure 4.3 above noted that basic skills are particularly poor in the East and Sarawak. At least half of firms in these two regions rated the following three basic attributes of local skilled production workers as poor or very poor: English language proficiency, IT, and communication skills. In the East, leadership and problem solving skills are also perceived as lacking, while entrepreneurs in Sarawak pointed out creativity of workers. Among the three regions that experience severe skill shortage and/or skill mismatch (the East, Klang Valley, and Sarawak); firms in Sarawak placed less effort to enhance skills of their workers. Figure 4.4 below shows that only one-fifth of Sarawak firms run formal training programs for their staff, compared to 55 percent of establishments in Klang Valley. According to PICS, a large share of firms in Sarawak is unaware of skills development institute, while others believe services that the institute provides are not fully relevant to firms' needs. 81 In terms of industrial composition, the North has many more electronics firms than other states. This industry spent seven weeks on average to secure a professional worker, much longer than the all-industry average of 5.4 weeks. 86 Figure 4.4: Share of Manufacturing Firms that Carried out Activities to Improve Labor Skills Sarawak East South Sabah North Klang Valley 0.0 10.0 20.0 30.0 40.0 50.0 60.0 Run form al training program m e Sent workers for training in Skills Developm ent Ins titute Source: Malaysia PICS 2007 Services Firms 165. Finding professional and skilled workers is more time-consuming for business services in Sabah. It generally takes over ten and six weeks to recruit suitable professional staff and skilled services workers in Sabah, which is around 5.5 and 1.5 weeks longer than the national averages, respectively (Figure 4.5). Skilled laborers are also relatively in short supply in Klang Valley82. Nonetheless, Sabah, where unskilled migrant workers constitute a large part of the population, seems to have abundant unskilled laborers. Relative to 2002, time required to fill these job vacancies decreased across regions in 2007. This is particularly true in the South, which made it currently least time-consuming in securing new staffs of all categories. Figure 4.5: Number of Weeks Required to Fill Job Vacancies in Services Firms South Saraw ak North Klang Valley Sabah 0.0 2.0 4.0 6.0 8.0 10.0 12.0 Prof essional Skilled w orkers Unskilled w orkers Source: Malaysia PICS 2007 82 Unlike the case of manufacturing firms, it is unlikely that industrial composition will be influential here. Except Klang Valley, services establishments in other regions are largely dominated by accounting and business logistics activities. 87 166. Basic skills inadequacy is common in all regions. In Sarawak, and technical inadequacy is a problem. At least half of firms in all regions cited lack of basic skills as an important reason for job vacancies (Figure 4.6). A view on technical skills is more diverse, with Sarawak as the worst-performing area. Over 60 percent of firms there believed poor technical skills led to unfilled vacancies. In contrast, services enterprises in the North viewed the lack of technical skills as less significant an obstacle but face with higher wages demanded by job applicants, an issue which was also voiced by manufacturing firms there. Finally, labor shortage is also not severe for services firms. Only a small number of firms cited insufficient number of university graduates or no applicants for unskilled worker position as critical obstacles. Figure 4.6: Share of services firms citing a lack of basic and technical skills as the cause of job vacancies North Sabah Klang Valley South Saraw ak 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Applicants lack basic skills Applicants lack technical skills Source: Malaysia PICS 2007 Affordability and Quality of Business Support Services Manufacturing Firms 167. The affordability and quality of business support services varies by region. Among the six aspects of business support services that PICS covers (engineering and design, management and marketing, accounting, legal, insurance, and IT services), the general services such as management and marketing, accounting, and insurance services are more affordable. At least 70 percent of manufacturing firms in each region reporting this in PICS 2007. However, the more technical support services such as engineering and design, legal, and IT services, which contribute directly to production and technological capabilities of manufacturers, are less affordable (Figure 4.7). Across regions, these business services are the least affordable in Sabah and the most affordable in Klang Valley. 88 Figure 4.7: The Share of Manufacturing Firms Reporting Selected Business Support Services are Affordable Sabah East Sarawak South North Klang Valley 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 Legal services IT services Engineering and design Source: Malaysia PICS 2007 168. The perceived quality of services also varies. For example, around 25 and 15 percent of manufacturers in the East viewed legal and insurance services as having poor/very poor quality respectively, compared to the average values of only 3-4 percent in the remaining five regions. The perceived quality of engineering and design, legal, and accounting services in the East worsened rather noticeably between 2002 and 2007, while nearly all services in most regions improved. This is in line with the response from the open-ended question where the lack of business support services is cited as the most binding business obstacle in the East. Services Firms 169. Business support services firms, who participated in the survey, reported they were generally satisfied with other services providers they used. This was in contrast to manufacturing firms. The share of services firms that rated the quality of the business support services as poor or very poor is relatively low across regions. More worrying cases are engineering and design services in the North and insurance services in the South, where up to 18 percent of firms cited these as having poor/very poor quality. This is somewhat consistent with the open-ended question, which shows that the lack of business support services was ranked as one of most critical investment constraints in the South. Over time, it appears that the quality of these services improved across regions. Public Infrastructure and Services Manufacturing Firms 170. Manufacturing firms in the North, the East, and Klang Valley reported it takes longer to obtain basic infrastructure services. Table 4.10 lists the average number of days it took for manufacturing firms to obtain three basic infrastructure services--electricity, water supply, and fixed telephone line--and their corresponding standard deviations in each region. 89 The time to obtain these infrastructure services is generally longer in the North, the East, and Klang Valley. As suggested by the standard deviation, uncertainty is also high. In some extreme cases, connecting electricity in North could take up to nearly 2.5 months.83 Such delay could be due to inadequate physical infrastructure or slow bureaucratic procedure or both. Table 4.10: Time It Takes Manufacturing Firms to Obtain Basic Infrastructure (days) Electrical connection Water connection Fixed telephone line PICS Change Standard PICS Change Standard PICS Change Standard 2007 from 2002 deviation 2007 from 2002 deviation 2007 from 2002 deviation Klang Valley 10.4 -4.2 14.8 10.3 -0.2 13.5 8.4 -1.0 9.6 North 10.2 -5.5 15.0 8.5 -3.2 9.0 7.8 -1.1 8.9 South 8.0 -4.1 13.4 7.2 -4.1 11.4 7.9 -2.0 15.56 East 12.5 +3.5 15.3 8.2 +2.3 4.7 9.2 +4.5 4.2 Sabah 5.5 -6.7 6.4 5.2 -6.8 4.6 5.0 -6.4 3.3 Sarawak 7.2 -7.3 7.0 8.2 +1.8 10.2 5.6 +0.3 4.8 Source: Malaysia PICS 2002 and PICS 2007 171. The quality of infrastructure services varies. Table 4.11 records the average frequency and duration of power outages, insufficient water supply, and interrupted telephone services by region. Despite being the region that takes the shortest time to provide basic infrastructure services in Table 4.10, the quality of these services in Sabah is poorer. There, power surge occurs on average more than twice a month and inadequate water supply about three times a month. This worsened significantly relative to the situation in 2002.84 The service interruption is not only frequent but also lengthy in Sabah. Water supply was lacking for over six hours, while power and telephone failures could each last around three hours. Unreliable electricity services incur a reasonably high cost to producers. Firms in Sabah estimated that the loss due to power interruption is around 4.6 percent of their total production value, up from 3.5 percent in 2002. The magnitude of loss is similar in the South where electricity services are also less reliable. In response to this, 36 and 20 percent of producers in Sabah and the South own a power generator respectively, relative to an average of 13 percent in the remaining four regions. Finally, Table 4.11 also shows that the three infrastructure services are more reliable in the East, both in terms of lower frequency and shorter duration of interruptions. The PICS data nonetheless suggests that the transport system is less reliable in the East. Transport disruption took place about five times a year, each lasting on average for three hours. This is much longer than the average of one hour in other regions. 83 Outlying observations are trimmed out in the analaysis. 84 The large differences in the frequency of power outages and insufficient water supply between Sabah and other regions are not due to outlying responses, defined as firms' responses that are unusually high. Firms that reported most frequent power failure and inadequate water supply are in fact located in Klang Valley. 90 Table 4.11: The Quality of Electrical, Water Supply, and Fixed Telephone Services Viewed by Manufacturing Firms Power outages or surges from the public Insufficient water Interruption fixed grid supply telephone service PICS Change PICS Change PICS Change 2007 from 2002 2007 from 2002 2007 from 2002 Klang Valley 11.1 - 1.6 6.2 - 1.8 4.1 - 1.4 North 9.6 - 1.7 4.4 - 4.0 6.0 + 0.3 Time per year South 12.2 + 2.2 5.4 + 0.2 4.6 + 0.1 experiencing... East 5.1 - 9.1 0.9 - 4.2 1.7 + 1.1 Sabah 26.7 + 5.4 36.4 + 26.8 7.6 + 5.5 Sarawak 7.5 - 2.5 0.3 - 2.2 2.7 - 4.8 Klang Valley 2.5 - 1.5 2.0 - 7.0 2.5 - 6.9 North 1.5 - 2.2 1.1 - 6.4 2.8 - 9.6 Average South 3.1 + 0.1 1.8 - 3.3 1.9 - 5.0 duration of... (hours) East 0.9 - 3.9 0.2 - 1.5 1.2 - 5.3 Sabah 2.9 - 0.2 6.3 + 3.0 2.8 + 0.8 Sarawak 1.3 - 0.1 0.0 - 2.6 2.0 - 3.8 Source: Malaysia PICS 2002 and PICS 2007 172. Producers in Sabah also suffer more from crime and theft. In addition to less reliable basic infrastructure services, the incidence of theft, robbery or vandalism is also high in Sabah. As depicted in Figure 4.8, the estimated loss from theft, robbery or vandalism is about 3.6 percent of total sales relative to the average of 1.4 percent in other regions. Crime in Sabah has intensified significantly in recent years. This is despite the fact that Sabah firms already spent up to 5.4 percent of their total sales in providing security to establishments. In 2002, the estimated loss and cost of providing security each stood at only 0.6 percent of total sales. In 2007, the combined cost/loss associated with crime is as high as nine percent of total sales.85 85 The results can be partly related to the limited number of observations and need to be interpreted with caution. 91 Figure 4.8: Estimated Loss from Theft , Robbery or Vandalism and Cost for Providing Security by Manufacturing Firms East Sarawak Klang Valley South North Sabah 0.0 1.0 2.0 3.0 4.0 5.0 6.0 Shipment cost due to breakage and theft / total shipment cost Estimated loss from theft, robbery or vandalism / total sales Estimated cost of providing security / total sales Source: Malaysia PICS 2007 Services Firms 173. It takes services firms a long time to obtain electrical and water connections in Sarawak and telephone lines in Klang Valley. Services firms in Sarawak on average waited 45 and 22 days for electrical and water connections (Table 4.12).86 Time used to obtain these basic infrastructure services can also be very uncertain, as indicated by high standard deviations. Connecting electrical and water services in Sarawak could take up to four and two months, respectively. In contrast, the South performs consistently better in this area, while there is a notable improvement in the North. Table 4.12: Time to Obtain Basic Infrastructure by Services Firms (days) Electrical connection Water connection Fixed telephone line PICS Change Standard PICS Change Standard PICS Change Standard 2007 from 2002 deviation 2007 from 2002 deviation 2007 from 2002 deviation Klang Valley 11.0 + 2.0 18.2 9.5 - 1.7 13.2 10.9 + 0.7 14.3 North 7.1 - 6.9 7.9 7.2 - 3.0 10.0 7.8 - 5.2 7.6 South 4.3 - 0.2 7.8 6.5 + 3.2 15.3 4.3 + 1.2 7.5 Sabah 9.9 + 2.7 9.6 8.8 + 3.4 7.5 8.3 + 3.1 8.3 Sarawak 45.4 + 30.9 76.0 21.6 + 7.6 23.9 6.0 + 2.2 3.9 Source: Malaysia PICS 2002 and PICS 2007 174. Nonetheless, once installed, interruptions in infrastructure services in Sarawak are limited, but interruptions in Sabah are frequent and lengthy. As in the case of manufacturing firms, the quality of basic infrastructure services in Sabah in services firms' view is clearly lagging behind other regions. Services interruptions in Sabah are more often and lasted longer than other regions (Table 4.13). For example, power outages and insufficient water supply took 86 The results need to be interpreted with caution due to the limited number of observations. 92 place 1.5 and 2.5 times a month on average, respectively. The quality also worsened compared to 2002. In 2007, despite the result shows that power supply services is the poorest in Sabah, services enterprises in the North reported the highest loss incurred from power surges (3.7 percent of operating cost) and owned more power generators (11.4 percent of northern firms). One possible reason might be the difference in composition of the services sector across regions. Business logistics firms are more concentrated in the North (71.4 percent of northern firms,). Business logistics sub-industry's average loss from power interruptions was estimated at nearly five percent of operating cost, much higher than a range of 0.1-1.1 percent in the other four services sub-industries. Table 4.13: The Quality of Electrical, Water Supply, and Fixed Telephone Services Viewed by Services Firms Power outages or surges from the public Insufficient water Interruption fixed grid supply telephone service PICS Change PICS Change PICS Change 2007 from 2002 2007 from 2002 2007 from 2002 Klang Valley 6.2 - 4.9 3.5 - 1.2 2.4 - 4.9 North 12.0 + 4.4 3.1 + 1.8 6.8 + 3.2 Time per year South 12.0 + 3.3 10.0 + 6.4 6.4 + 4.0 experiencing... Sabah 18.9 + 4.1 30.9 + 27.1 13.7 + 12.4 Sarawak 3.8 - 5.6 0.0 0.0 1.4 - 3.4 Klang Valley 0.6 - 2.7 0.9 - 1.9 0.7 - 1.1 Average North 0.8 - 0.8 0.3 - 0.2 0.4 - 0.6 duration of... South 2.2 - 6.1 3.3 - 6.3 2.2 + 1.5 (hours) Sabah 2.2 + 0.4 6.2 + 1.9 2.9 + 2.7 Sarawak 0.2 - 1.7 0.0 0.0 3.1 + 1.1 Source: Malaysia PICS 2002 and PICS 2007 175. Crime is more detrimental in the North and the South. As opposed to the views of manufacturing firms' in Sabah, services providers in the North and the South spent a sizeable amount of financial resources to protect their assets, yet the associated losses remained high. The combined spending/losses amounted to 8.4 percent of total sales in the North and 11.3 percent in the South. The situation worsened quite remarkably in the South. In 2002 this figure was only 2.3 percent of total sales. This result confirms the message in the open-ended question in Table 4.7, which showed that crime and theft is one of the top business obstacles in both regions. 93 Figure 4.9: Estimated Loss from Theft, Robbery or Vandalism and Cost for Providing Security by Services Firms Klang Valley Saraw ak Sabah South North 0.0 2.0 4.0 6.0 8.0 10.0 Estimated cost of providing security / total sales Estimated loss f rom thef t, robbery or vandalism / total sales Source: Malaysia PICS 2007 Government Regulations Manufacturing Firms 176. Government agencies in the North took longer to issue official documents. Table 4.14 records the number of weeks that firms spent on average to obtain licenses, permits, and approvals/certificates from different levels of state authority. It clearly shows that, except for approvals/certificates issued by the federal government, obtaining these documents is most time- consuming in the North. There it takes between 6-9 weeks to obtain licenses from different agencies, while this is only around 3.3-4.3 weeks on average in other regions. For example, acquiring operating license typically takes 34 days in the North, which is up to 20 days longer than the mean level of other regions. Obtaining import permit and construction is also six and nine days slower than other regions on average, respectively. To deal with this red tape, over one-fifth of firms in the North hired agents to process the requests for these government documents. The share of firms that used agents is also high in the South and the East. The cost incurred from hiring these agents varies around 0-0.5 percent of total sales in different regions. Although this seems very small, it is comparable to what firms spend on research and development (0-0.6 percent of total sales). Such resources could have been utilized in more productive activities. 94 Table 4.14: Average Number of Weeks to Obtain Licenses, Permits, and Approvals/Certificates for Manufacturing Firms Federal government State government Local authority PICS Change PICS Change PICS Change 2007 from 2002 2007 from 2002 2007 from 2002 Klang Valley 5.3 - 1.6 4.7 - 1.5 4.6 - 0.6 North 9.0 + 0.6 7.7 - 0.9 6.0 - 1.4 South 5.7 - 0.8 5.0 - 1.4 3.9 - 3.6 Licenses East 3.5 + 1.5 4.5 + 1.1 3.3 + 1.7 Sabah 2.4 - 0.9 1.7 - 2.8 2.3 + 0.2 Sarawak 4.5 - 2.7 1.9 - 4.7 2.6 + 0.3 Klang Valley 4.8 - 1.7 4.9 - 4.9 2.3 - 4.5 North 5.1 - 2.5 7.4 - 4.8 4.0 - 2.1 South 3.5 - 3.3 3.7 - 4.9 3.0 - 2.0 Permits East 4.6 + 2.1 4.7 0.0 3.6 + 1.6 Sabah 1.9 - 1.1 1.7 - 5.5 4.0 + 2.6 Sarawak 3.5 - 6.0 3.0 - 4.6 2.3 - 4.3 Klang Valley 6.7 - 5.1 3.9 - 1.7 3.1 - 7.4 North 6.4 - 1.8 17.9 + 12.2 7.6 + 0.9 Approvals South 6.0 - 3.0 6.8 + 3.0 3.9 - 4.4 and certificates East 4.6 - 3.4 4.3 + 3.3 4.5 0.0 Sabah 4.4 + 0.4 12.5 + 11.2 1.0 - 7.2 Sarawak 12.8 + 2.0 8.2 + 2.1 4.8 + 0.5 Source: Malaysia PICS 2002 and PICS 2007 177. Firms in the East experience the most frequent visits and inspections from government officials. Table 4.15 shows the average number of days that firms spent in contact with selected government agencies in 2006. Official visits and inspections have generally become less frequent across regions over the past few years, especially for Tax Inspectorate. Despite this, firms in the East spent up to 16 days in dealing with these four agencies alone. There are also other agencies that manufacturers tend to deal with, such as environmental and industrial standards agencies. The last column notes that nearly 20 percent of senior managers' time in the East was spent to meet various government requirements and regulations on taxes, customs, labor, licensing, registration, etc. This increased sharply from 2002, and suggests that although official visits and inspections are less common, complying with government regulations is more burdensome. 95 Table 4.15: Average Number of Days Spent in Contact with Different Agencies by Manufacturing Firms Fire and Rescue Percent of senior Department / management time Labor and Department of Tax Inspectorate Local authority per year spent in Social Security Occupational dealing with Safety and regulations Health PICS Change PICS Change PICS Change PICS Change PICS Change 2007 from 2002 2007 from 2002 2007 from 2002 2007 from 2002 2007 from 2002 Klang Valley 1.8 - 19.8 2.8 - 3.8 3.4 - 1.1 2.5 - 6.3 9.5 + 0.6 North 3.3 - 18.8 3.2 - 3.1 3.0 - 1.5 2.2 - 4.0 12.5 + 3.0 South 1.5 - 16.8 3.1 - 2.3 2.7 - 0.6 1.9 - 2.8 9.6 + 1.1 East 3.7 - 10.9 4.5 - 1.8 4.0 + 1.9 3.8 - 0.3 18.7 + 13.0 Sabah 1.9 - 21.3 2.5 - 0.7 1.9 - 0.5 2.3 + 0.6 6.2 + 2.9 Sarawak 0.7 - 21.0 1.7 - 1.7 1.3 - 1.9 1.3 - 1.4 4.4 + 0.7 Source: Malaysia PICS 2002 and PICS 2007 178. Customs clearance is the slowest in Sabah. Figure 4.10 displays the number of days that firms in each region took on average to clear customs process for imports and exports. These procedures are clearly more cumbersome in Sabah than in other regions. On average, firms took at least ten days to clear goods, which increased sharply from 3.7 days for imports and 1.6 days for exports in 2002. In other regions, the changes between 2002 and 2007 are small. Figure 4.10: Average Number of Days to Clear Customs Procedure for Exports and Imports by Manufacturing Firms East South North Saraw ak Klang Valley Sabah 0.0 2.0 4.0 6.0 8.0 10.0 12.0 Exports Imports Source: Malaysia PICS 2007 96 Services Firms 179. Services companies in Sarawak waited the longest to receive licenses they had applied for. Obtaining licenses from the state government took up to three weeks on average in Sarawak, as opposed to only one week in the South (Table 4.16). The result on time taken to obtain a construction permit is similar. Lower regulatory burden in the South benefits from a significant improvement between 2002 and 2007. In 2002, the state government in the South in fact has the least favorable performance. Table 4.16: Average Number of Weeks to Obtain Licenses for Services Firms Licenses Approval for Federal construction government State government Local authority PICS Change PICS Change PICS Change PICS Change 2007 from 2002 2007 from 2002 2007 from 2002 2007 from 2002 Klang Valley 2.5 + 0.5 1.3 + 0.1 1.3 + 0.1 6.7 - 1.1 North 4.0 + 2.3 2.0 + 0.3 1.4 + 0.2 4.7 - 3.8 South 5.6 + 2.5 1.0 - 6.0 1.4 + 0.2 9.8 + 2.1 Sabah 2.3 + 0.3 1.6 + 0.4 1.1 - 0.3 3.8 + 1.8 Sarawak 1.2 - 1.2 2.8 + 1.6 1.0 0.0 13.9 + 6.8 Source: Malaysia PICS 2002 and PICS 2007 180. In addition to licensing delays, firms in Sarawak spend more time contacting government agencies. Services companies there spent over four days each in dealing with Tax Inspectorate and Fire and Rescue Department per year, which is 1.5-2 days longer than firms in the other regions on average (Table 4.17). Over the past several years, senior management of firms in Sarawak had to spend more time in dealing with government regulations, i.e. up from 6.3 percent of total time in 2002 to 10.3 percent in 2007. As in the case of manufacturing firms, bureaucratic process still consumes the sizeable share of managers' time in the North. Table 4.17: Average Number of Days Spent in Contact with Different Agencies by Services Firms Fire and Rescue Percent of senior Department / management time Labor and Tax Inspectorate Department of Local authority per year spent in Social Security Occupational dealing with Safety and Health regulations PICS Change PICS Change PICS Change PICS Change PICS Change 2007 from 2002 2007 from 2002 2007 from 2002 2007 from 2002 2007 from 2002 Klang Valley 3.2 - 4.6 4.1 - 5.9 1.0 - 2.7 1.9 - 13.4 7.5 - 1.1 North 2.4 - 0.7 4.5 - 3.5 1.5 + 0.3 2.5 - 0.9 12.3 + 0.6 South 2.1 - 1.5 4.7 - 3.1 3.4 + 1.8 3.0 + 0.5 8.5 + 1.1 Sabah 3.9 - 3.3 4.6 - 3.2 2.7 - 7.3 22.8 + 1.4 8.1 + 2.4 Sarawak 4.2 - 37.8 4.4 - 0.4 4.2 - 1.8 6.8 - 33.9 10.3 + 4.0 Source: Malaysia PICS 2002 and PICS 2007 97 Innovative Activities and ICT Usage Manufacturing Firms 181. Firms in Sarawak carried out fewer innovations. Table 4.18 shows the shares of firms that undertook various innovative activities in the past two years. Here innovation refers not only to technological innovations but also process, product, and commercial innovations. Innovation is the least common in Sarawak. This is true across various types and sophistication levels of innovation. For example, for innovative activities those require less effort by firms such as upgrading machinery and equipment (technological innovation) and upgrading an existing product line (product innovation), only 30 percent of firms in Sarawak recently accomplished them. This is low compared to 40-70 percent of firms in other regions. The results are similar for more sophisticated innovative activities, such as introducing new technology and products, and filing patents. Activities that could enhance firm's innovation capability such as subcontracting R&D activities and entering a new joint venture with foreign partner are also negligible in Sarawak.87 Relative to 2002, enterprises in Sarawak have also become less innovative.88 Table 4.18: Innovative Activities Carried Out by Manufacturing Firms Filed Introduced new patent/utility Upgraded Upgraded an technology that Developed a new model or machinery and existing product markedly changed major product copyright equipment line how to produce key line protected product materials PICS Change PICS Change PICS Change PICS Change PICS Change 2007 from 2002 2007 from 2002 2007 from 2002 2007 from 2002 2007 from 2002 Klang Valley 59.7 - 6.1 49.2 - 5.4 26.3 - 10.3 27.3 - 8.1 17.6 + 5.5 North 62.7 + 1.9 51.8 - 1.3 31.5 + 1.7 31.1 - 0.4 14.4 - 2.3 South 58.7 - 0.5 40.3 - 4.2 23.2 + 0.2 22.9 - 1.3 14.1 + 8.0 East 69.0 + 7.1 48.3 + 5.4 31.0 + 7.2 27.6 + 18.1 17.2 + 7.7 Sabah 54.8 + 21.5 54.8 + 13.2 25.8 + 5.0 25.8 + 17.5 12.9 + 4.6 Sarawak 30.4 - 7.8 28.3 - 10.0 10.9 - 6.8 10.9 - 0.9 4.4 - 4.5 Source: Malaysia PICS 2002 and PICS 2007 182. Sarawak firms are less apt to use ICT. In addition to being less innovative, fewer producers in Sarawak use information and communications technology (ICT) in their operations. Figure 4.11 depicts the shares of firms that regularly use email and website to interact with their suppliers and customers. Email usage is common in three quarters of firms in Klang Valley, compared with less than one-third of firms in Sarawak. Between 2002 and 2007, the extent of ICT usage in Sarawak rarely changed. In contrast, strong improvement took place in Sabah, where the shares of firms that used website and email regularly rose by 30 and 9 percent, respectively. 87 Subcontracting R&D activities and entering a new joint venture with foreign partner allow firms to better explore global knowledge and more knowledgeable human resources available elsewhere. This potentially leads to higher and more efficient R&D spending. 88 Among other factors, industrial composition of producers in Sarawak helps to explain this. More than three quarters of firms there are in the food processing, wood products, and furniture which engage in fewer innovative activities. 98 Figure 4.11: The Share of Manufacturing Firms that Regularly Use Email and Have a Website for Their Business Sarawak East South North Sabah Klang Valley 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Email Website Source: Malaysia PICS 2007 Services Firms 183. Compared to the South, services firms in Sabah appear more innovative than others. As in the case of producers, less sophisticated innovative activities are rather widely done among services firms. At least half of firms in all regions upgraded their equipment in the past two years, although Klang Valley has seen a large drop in this ratio (Table 4.19). Overall, more Sabah companies upgraded existing and developed new service line, adopted new major technology, and entered a joint venture with foreign firm. Between the two rounds of PICS, they also exhibited greater effort for innovation. Meanwhile, it does not appear that there is a strong link between government incentives for promoting innovation and firms' innovation efforts. The share of firms that received government incentives to conduct technological innovations and R&D is the lowest in Sabah (11.1 percent, about half of the national mean).89 This suggests that strengthening the effectiveness of government incentives can further boost firm level innovations. 89 Sabah has large services firms more than others (14 percent of all firms, same as Klang Valley). This might help to explain its greater innovation effort. The least innovative region (South) has similar industrial composition and foreign ownership share but the lowest share of large firms in Malaysia. 99 Table 4.19: Innovative Activities Carried Out by Service Firms Introduced new Entered a new Upgraded an technology that joint venture with Upgraded Developed a new existing service markedly changed foreign partner equipment major service line line how to deliver key service PICS Change PICS Change PICS Change PICS Change PICS Change 2007 from 2002 2007 from 2002 2007 from 2002 2007 from 2002 2007 from 2002 Klang Valley 52.6 - 21.0 43.8 - 13.9 22.0 - 15.9 19.1 - 7.4 3.9 - 9.9 North 54.3 + 16.2 37.1 + 8.6 28.6 + 14.3 17.1 + 7.6 2.9 - 1.9 South 59.1 - 5.2 27.3 - 1.3 9.1 + 2.0 9.1 + 2.0 0.0 0.0 Sabah 64.3 + 18.1 50.0 + 19.2 28.6 + 5.5 21.4 - 1.7 7.1 + 7.1 Sarawak 50.0 + 5.6 38.9 + 16.7 27.8 + 22.2 22.2 + 11.1 0.0 - 5.6 Source: Malaysia PICS 2002 and PICS 2007 Access to Financing Manufacturing Firms 184. Producers in the East rely more on retained earnings than bank loans to finance their operations. To attain firm-level data on access to finance, PICS asked firms to indicate how they finance working capital and new investments. Possible sources of funds given are internal funds/retained earnings, commercial bank loans (local and foreign), leasing arrangement, investment funds, trade credit, sale of stock, family/friends, and informal sources. Figure 4.12 compares the share of working capital financed by internal funds/retained earning plus family/friends and that by local commercial bank loans. Figure 4.13 depicts similar sources of funds for new investments. It can be seen that the contribution of local bank loans to both working capital and new investments is very small (below 20 percent) in the East. The contribution of local bank loans to working capital is also low in Sarawak. Figure 4.12: Share of Working Capital Financed by Retained Earnings and Bank Loans for Manufacturing Firms 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 East Saraw ak Klang Valley South North Sabah Internal f unds, retained earning, f amily and f riends Local commercial bank loans Source: Malaysia PICS 2007 100 Figure 4.13: Share of New Investments Financed by Retained Earnings and Bank Loans for Manufacturing Firms 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 East Klang Valley North Sabah Saraw ak South Internal funds, retained earning, family and friends Local commercial bank loans Source: Malaysia PICS 2007 185. Higher collateral requirements partly explain the limited use of bank loans in the East and Sarawak. Figure 4.14 highlights much lower shares of collateral required in loan value in 2007 relative to the levels in 2002. This partly reflects the more stable financial market conditions compared to 2002 when the Malaysian economy was recovering from the 1997 economic crisis. Overall, a more limited role of bank loans in financing firms in the East and Sarawak could be explained by higher collateral requirements. Collateral required is highest in Sarawak at around 65 percent of loan value. In the East, although the current collateral requirement is not much higher than the rest of the country, the requirement dropped only marginally between 2002 and 2007. In other regions, the decrease is much more significant. For instance, collateral requirement in Klang Valley fell from 91 percent in 2002 to only 40 percent in 2007. Note that the limited use of bank loans in the East and Sarawak is unlikely due to an inadequate access to finance. Nearly 70 percent of firms in the East and Sarawak currently have a term loan from commercial banks or financial institutions, which is higher than all other regions except Sabah. High borrowing rates are also unlikely an explanation. The average interest rates for short and long-term domestic liabilities that firms in the East and Sarawak paid were among the lowest in the country. For example, the average short-term rates in 2007 in these two regions are around 7.0-7.8 percent, compared to 11.1-12.9 percent in Sabah, the South, and the North.90 90 In addition to limited access to finance and high borrowing costs, an industrial composition of firms in the East and Sarawak does not seem to explain this. 101 Figure 4.14: Approximate Value of Collateral Required as a Percentage of Loan Value for Manufacturing Firms South Klang Valley Sabah North East Saraw ak 0.0 20.0 40.0 60.0 80.0 100.0 2007 2002 Source: Malaysia PICS 2007 Services Firms 186. Services companies in the South appear to have limited access to financing. Only a third of firms in the South have term loans, compared to nearly half for the whole country. There are also fewer South firms with bank overdraft facilities. These numbers are consistent with the close-ended question in Table 4.9, which noted that almost 30 percent of services firms in the South cited access to both domestic and foreign credits as a major business obstacle. As a result, the South relies much more on retained earnings and family and friends to finance their operations, as opposed to formal commercial banks loans (Figure 4.15). A likely explanation is the concentration of small services firms in the South (and Sarawak). Both regions have low reliance on bank loans.91 91 Although the South pays higher lending rates and is subject to high collateral requirement, these factors do not seem to explain this. Both borrowing cost and collateral requirement in Klang Valley are lower than the national means, still their reliance on term loans is limited. 102 Figure 4.15: Share of Working Capital Financed by Retained Earnings and Bank Loans for Services Firms Sabah North Klang Valley South Saraw ak 0.0 20.0 40.0 60.0 80.0 100.0 Internal f unds, retained earnings, f amily and f riends Local commercial bank loans Source: Malaysia PICS 2007 CONCLUSIONS 187. In Malaysia the investment climate varies from state to state and region by region. Moving a production line to a more business-friendly location could significantly enhance a firm's competitive advantage. Manufacturing firms as well as services firms perceived the Klang Valley region as having the best business environment, Sabah the least satisfactory. Over the past few years the gaps between the best and worst performing regions loomed large based on these estimates. However, the picture that emerged from the objective measures is not so black-and- white. It is true that firms in Sabah are more likely to experience a less satisfactory business environment when it comes to limited access to affordable business support services, slower customs procedures for imports and exports and greater expenses due to crime and theft. Nevertheless, these objective measures also reveal that the investment climate in the Klang Valley needs further improvement based on the quality of its electrical and water supply, fixed telephone services and the number of weeks it takes to obtain licenses, permits and approvals. 188. In all regions, a shortage of skilled labor and high taxes constrain investment. However, the concern about taxes could be exaggerated. The government cut the corporate tax rate in 2006. Because the most recent PICS occurred in 2007, the firms surveyed may not have experienced the effects of that reduction. The lack of skilled labor, cited in PICS 2002 as the most severe business obstacle, persists. Even in regions with a relatively good business environment like Klang Valley, the North, and South, skilled workers remain in short supply. Other region-specific issues that tend to limit investment appear in Table 4.20. Sabah, followed by the East, is perceived as having the worst investment climate because firms there suffer from a wide range of business obstacles. Services firms in Sarawak also face major obstacles. 103 Table 4.20: Key Investment Climate Challenges by Region Labor skills Government rules Infrastructure and Access to Business R&D and and regulations public services finance support Innovation services Klang Skill shortage Valley & mismatch (M&S) North Skill shortage Slow process to obtain Electricity & mismatch licenses, permits, and connection is time- (M&S) approvals (M) consuming (M) High loss and cost associated with theft, robbery or vandalism (S) East Skill shortage Frequent visits and Unreliable transport High Poor quality & mismatch inspections by system (M) collateral of business (M&S) government officials requirements support (M) (M) services (M) South Skill shortage High loss and cost Weak & mismatch associated with innovation (M&S) theft, robbery, or effort (S) vandalism (S) Sabah Skill shortage Rigid labor regulations Poor quality of High Less & mismatch (M) electricity, water collateral affordable (M&S) supply, and fixed requirements business Slow customs telephone services (M) support procedures (M) (M&S) services (M) High loss and cost associated with theft, robbery or vandalism (M) Sarawak Skill shortage Slow process to obtain Electricity and High Weak & mismatch licenses (S) water connections collateral innovation (M&S) are time-consuming requirements effort and low Frequent visits by (S) (M) ICT usage (M) government agencies (S) Source: Findings based on Malaysia PICS 2002 and PICS 2007 Note: "M" indicates development challenges identified by manufacturing firms. "S" is those by services firms. 189. Improving the business climate is critical for the Development Corridors to reach their full potential. · For the Sabah Development Corridor, increasing the value of existing industries is a key strategy for the region. Among others, it entails transforming and expanding agricultural and bio-technological sectors in the region. For example, a palm oil industrial cluster will be established to boost the value of the palm oil industry. The development plan also promotes downstream activities in the agricultural sector and new manufacturing, namely furniture, sustainable materials, and oil and gas. To fulfill these plans, Sabah's investment climate needs to get stronger. In particular, the region needs to: (i) improve the affordability of key business support services; these tend to strengthen production and 104 technological capabilities of producers such as engineering and design, legal, and IT services; (ii) enhance the reliability of electrical, water supply, and fixed telephone services; (iii) enact and enforce stronger measures to reduce the frequency of crime and theft; and (iv) speed up customs to reduce the cost of conducting international trade. · Enhancing the quality of key business support services and the reliability of transport system are important for the East Coast Development Corridors (ECER), In the East, frequent visits and inspections by government officials represents lost time, time that senior managers could have spent on more productive activities. In ECER, the chief manufacturing activities, namely the petrochemical, wood-based, and food processing industries, are resource-based. The proposed development plan will extend the current value chains and strengthen emerging industries that offer significant potential; the latter include the following products and services: Halal, handicrafts, boat building, garments and textiles and automotive and transport equipment. These plans encompass the establishment of new free zones, specialized industrial parks, stronger inter-industry linkages, R&D, and product development activities. · Promoting innovation in the Sarawak Development Corridor is critical to strengthening the development of renewable energy. Among others, the Sarawak Corridor of Renewable Energy focuses on resource-based manufacturing activities, such as aluminum, glass, metal, petroleum, timber, livestock, and palm oil. After attracting these industries into the region, the second phase of the development plan (2016-2020) will emphasize industrial cluster development and the R&D capabilities of individual firms. R&D is also a key element of the final phase (2021-2030). Innovative activities occur least frequently among producers in Sarawak. Fewer firms there upgraded machinery and product lines, introduced new technology and products, filed patents, and outsourced R&D activities compared to other regions. ICT such as e-mail and website is also less prevalent there. Therefore, government incentives that increase innovation ICT usage are desirable. · Streamlining government regulations is important for the North Development Corridor. Although PICS 2007 found the North to be a relatively business-friendly region, its attractiveness as a production base has declined in recent years. As an example, firms experience delays in connecting to basic public infrastructure such as electricity. Obtaining licenses permits, and approvals/certificates are also more time-consuming than in all other regions. Overcoming these obstacles could help to boost new investments into the Northern Corridor Economic Region (NCER). The latter aims to deepen industrial linkages and to propel industries up the value chain. The role of the North as a hub for Malaysia's electrical and electronics sector will get stronger with the establishment of a micro-electronics center of excellence. To raise the capability of local E&E producers, the Northern corridor will support more advanced activities such as silicon, automation and materials design. NCER also plans to promote new industries such as biotechnology, downstream agriculture, sustainable materials and oil and gas. 190. An improvement in the investment climate goes hand in hand with the development corridor initiatives. Overall, improving the business environment in the five development corridors is important to reach targeted investments. Furthermore, this would help reduce the 105 socio-economic inequalities between Klang Valley and other regions in Malaysia. Different economic regions need to mitigate different investment constraints. This is especially critical for the Sabah Development Corridor and the East Coast Economic Region where several business obstacles hinder operations. In addition to improving the business climate, other development strategies planned by the corridors, such as industrial and science parks, industrial clusters, and investment incentive will also contribute to their success. 106 CHAPTER 5. IMPROVING SKILLS 191. Inadequate skills have long been a major constraint on Malaysia's investment climate. The widespread perception of skills inadequacy consists of two interrelated types of disequilibria in the market for skills. The first is skills shortage, and the second is skills mismatch or skills deficiency. This situation has been fuelled by a steady increase in the demand for skilled workers in Malaysia over the last decade. This chapter examines the level, distribution, and quality of the skills and competencies of workers in Malaysia, as well as the match between the qualifications supplied and those demanded by businesses. Improving skills will play an important role as the economy moved from more cost- to value- and knowledge- base. 192. In the past few years, the supply of skills has improved, but challenges remain. At the macro level, the gap in human capital stock between Malaysia and economies at similar levels of development has narrowed, but the deficit remains large. At the micro level, from the perspectives of employers as well as employees, although the quality of local professionals and skilled production workers has improved, the supply of skills still falls short of the rapidly increasing demand. The incidence and intensity of hard-to-fill vacancies remain high in the manufacturing sector of Malaysia. Over time, firms in manufacturing sector reported an improvement in perception while firms in business support services sector a worsening of concerns. 193. Although the quantity of high-school and college graduates has increased, it has been insufficient in addressing the skills constraints. The stagnant real wage level and declining wage premium suggest that although skills inadequacies have been perceived as a major obstacle and increasingly so by services firms, the willingness to pay for the scarce factor of education seems to be limited and even declining over time under some circumstances. However, the wage premiums for those who had studied abroad, who received employer- provided trainings in the areas of marketing, information technology and management, and who received off-the-job training are high and increasingly so in particular in the services sector. This indicates that, to some extent, general education level may not be the best indicator of skills: an average worker with high-school or college education may not have the skills that firms required. To improve the quality of the labor force, firms need to provide in-service training to improve workers' general skills and firm-specific skills. To attract and retain the workers with needed skills, of which supply is limited, firms need to pay higher wage premium. The coexistence of high job vacancy rates of skilled workers/professionals and unemployment of certain college graduates suggests possible imperfections in the labor market and education system. Addressing the distortions which have been persisting in the past years in preventing labor market from sending signals to effectively match supply and demand and improving the general education system to provide workers with skills and potential required by firms are of critical importance. 107 194. This chapter is structured as follows. Section 1 reviews Malaysia's gaps in skills for work with macro evidence and micro evidence. Section 2 examines wage premium in manufacturing and services sectors and discusses the role of in-service training. Section 3 concludes with policy suggestions. MALAYSIA'S SKILLS GAP 195. Inadequate skills have long been a major constraint for Malaysian businesses despite a rapid increase in the number of high-school and university graduates in recent years. Recognizing that "the quality of the nation's human capital will be the most crucial element" to attain the national goal of moving towards a knowledge-based economy, the Ninth Malaysia Plan placed high priority on human capital development. The Bio Nexus program offers tax breaks and research grants to biotechnology companies to accelerate the progress of "Bio Valley" projects.92 196. The Ninth Malaysia Plan focused on increasing accessibility to higher education. In order to achieve the target of 40 percent tertiary education participation rate of the age group 17- 23 in 2010, measures implemented includes: providing financial assistance to students; benchmarking quality of higher education institutions against international standards; increasing coverage and utilization of ICT; introducing a quality assurance system by increasing teachers with degree qualification in primary and secondary schools; and improving in infrastructure facilities, such as clean water facility and electricity supply to bridge a gap in school quality between urban and rural schools (EPU, 2006). The Plan also took steps to address the mismatch of skills. It emphasized education in science, mathematics and foreign languages and provided programs for increasing the supply of quality science and technology human resources to enhance scientific and technological capabilities in support of intensified innovative R&D activities. Toward this end, the Government of Malaysia will increase budget allocation to tertiary from RM 13,404 million in the Eighth Malaysia Plan to RM 16,069 million (EPU, 2006). Macro Evidence 197. At the macro level, the supply of labor with higher education increased sharply in the recent past. The number of fresh college graduates increased about 70 percent from 30,732 in 2000 to 51,771 in 2005.93 The gap in human capital stock between Malaysia and economies at similar levels of development has narrowed over the past decade, but the deficit remains large. We assess the gap in the stock of human capital using two macro indicators. The first is the gap between Malaysia's actual stock of human capital as measured by the gross upper secondary school enrollment rate and its predicted value as given by the country's level of income. The second is the gap between Malaysia's actual stock of human capital as measured by the gross tertiary graduation ratio and its predicted value as given by the country's level of income. The indicators of skills gap in Malaysia is measured with respect to international benchmarks, including countries that have successfully completed the transition from low-skills manufacturing to high-tech, high-skills dominated manufacturing. The gap in the stock of upper 92 The Bio Valley project was launched in 2003 in a 500 acre site of Cyberjaya to facilitate industries related to biotechnology. See Cyranoski, D. (2005, August). "The Valley of Ghosts." Nature, 436, 620-621. 93 Data source: World Bank (2007), Malaysia and the Knowledge Economy: Building a World-Class Higher Education System". 108 secondary education for Malaysia has averaged 30 percent over the period of 1998 and 2005, and the gap in the stock of tertiary education has averaged 10 percent over the period of 1998 and 2005. 198. The government has taken bold measures to expand education to address the constraints on skilled labor. Those steps have narrowed the gaps. Government spending on education as percent of total governmental expenditure increased from 18 percent in 1991 to 25 percent in 2004. In 2000-2004, the Government of Malaysia allocated, in average, 24 percent of the government expenditure on education, of which one-third to tertiary schools94. Malaysia's public spending on tertiary education (2.2 percent of GDP) is high compared to comparator countries in East Asia (less than 1 percent). 199. However, the gross tertiary school enrollment in Malaysia is still relatively low. Malaysia's gross tertiary enrollment rate is 30 percent, which is lagging behind South Korea (89 percent), Japan (55 percent), Thailand (46 percent) and Mongolia (39 percent) (Figure 5.1). One reason of the low tertiary enrollment in Malaysia compared to many other East Asian countries is the only a small percentage of secondary school graduates continue to pursue full university education (World Bank, 2007). Although Malaysia has an overall secondary graduation rate of 84 percent, only 20 percent of secondary graduates pursued the university degree in 2004. In contrast, for example, 64 percent of secondary graduates pursued the college degree in South Korea; 59 percent in the Philippines and 31 percent in Thailand. Although high completion rate of secondary school provides a great potential to promote tertiary education, more appropriate efforts are needed. Students' limited basic skills in engineering and technology may be one important factor that discourages the pursuance of university in the fields of sciences; while the textbook-oriented teaching approach may also weaken the interests of students.95 94 Source: Public spending on education data are from WDI online, and the percentage distribution of public expenditure on tertiary school are from UNESCO World Education Indicators online. 95 See Mahadevan (2005). 109 Figure 5.1: Tertiary School Enrollment Rate and Public Spending on Tertiary Education as percent of GDP in East Asian/Upper Middle Income/OECD Countries Note: Based a sample of 47 countries, the red best fit line has a slope of 17.2 that is significant at the 1 percent level and an intercept of 37.8. The tertiary education gross enrollment rate is regressed on the public spending on tertiary education as percent of GDP and a constant. Source: Tertiary school enrollment rate from WDI. Public spending on tertiary education as percent of GDP is calculated by authors based on the total public spending on education as percent of GDP from WDI and percent distribution of public current expenditure on education by level (tertiary) from UNESCO's World Education Indicators online. Data are for 2004-2006, the most recent data available. Upper Secondary Education 200. The gap in enrollment rates at the upper secondary level that existed between Malaysia and economies at a similar level of development has narrowed over the past decade, but the deficit has persisted as well. Figure 5.2 shows the relationship between the development level, measured as real per capita GDP in logarithmic form (on the horizontal axis) and the upper secondary education gross enrollment rates (on the vertical axis) in 2005 for 177 countries for which data are available from UNESCO.96 The gap in the stock of human capital is measured by the distance between the actual gross upper secondary enrollment rate of a given country and the norm for its income level. Malaysia has averaged 30 percent deficit in upper secondary education over the period 1998 and 2005. In 1998, the actual gross upper secondary enrollment rate was 44 percent, whereas the predicted values for a country at Malaysia' level of development is 76 percent. In 2005, the actual gross upper secondary enrollment rate was 52 percent and the predicted value for a country at Malaysia's level of development is 81 percent. Overtime, this gap has narrowed from 32 percent in 1998 to 28 percent in 2005. 96 Upper secondary education gross enrollment rate is the total number of students enrolled in upper secondary, regardless of age, expressed as a percentage of the population in the theoretical age group for upper secondary education. 110 Figure 5.2: Malaysia's Upper Secondary Education Gross Enrollment Rates in International Perspective, 2005 Upper Secondary Education Gross Enrollment Rate 2005 (%) 120% United Kingdom Japan 100% Russia Korea, Rep. United States Chile 80% Philippines Hong Kong, China 60% Thailand China MALAYSIA Indonesia 1998-2005 40% India Sub-Saharan Africa 20% 0% 5.0 6.0 7.0 8.0 9.0 10.0 11.0 Natural Logarithm of Real GDP Per Capita, 2005 Note: Based on a sample of 177 countries, the red best fit line has a slope of 23.4 that is significant at the 1 percent level and an intercept of -138.2. The upper secondary education gross enrollment rate is regressed on the log of real per capita GDP and a constant, and the resulting red line correspond to the predicted upper secondary education gross enrollment rates given the income level of the country. The historic path of upper secondary education gross enrollment rates and income in Malaysia over the period 1998 to 2005 is traced along the blue curve. Source: UNESCO Institute for Statistics Tertiary Education 201. Similarly, the gap in graduation rate at the tertiary level that existed between Malaysia and economies at a similar level of development has narrowed over the past decade, and yet, this gap has persisted. Figure 5.3 shows the relationship between the log of real per capita GDP (on the horizontal axis) and the gross tertiary graduation ratio (on the vertical axis) in 2005 for 93 countries for which data are available from UNESCO.97 As measured by the distance between the gross tertiary graduation ratio of a given country and the norm for its income level, the gap in the stock of human capital has averaged 10 percent over the period 1999 and 2005. In 1999, the actual gross tertiary graduation rate was 11 percent, whereas the predicted values for a country at Malaysia's level of development is 22 percent. In 2005, the actual gross tertiary graduation rate was 15 percent and the predicted value for a country at Malaysia's level of development is 24 percent. Overtime, this gap has narrowed from 12 percent in 1999 to 10 percent in 2005. 97 Gross tertiary graduation ratio is the total number of graduates from first degree, regardless of age, expressed as a percentage of the population at the theoretical graduation age for such programs. 111 Figure 5.3: Malaysia's Tertiary Education Gross Graduation Ratio in International Perspective, 2005 70% Gross Tertiary Graduation Ratio (First Degree) 2005 (%) 60% 50% Russia Ireland 40% United Kingdom Japan Korea, Rep. United States 30% Chile 20% Philippines Hong Kong, China China MALAYSIA 10% Indonesia 1999-2005 0% 5.0 6.0 7.0 8.0 9.0 10.0 11.0 Natural Logarithm of Real GDP Per Capita, 2005 Note: Based a sample of 93 countries, the red best fit line has a slope of 8.5 that is significant at the 1 percent level and an intercept of -56.2. The gross tertiary graduation ratio is regressed on the log of real per capita GDP and a constant, and the resulting red line correspond to the predicted gross tertiary graduation ratio given the income level of the country. The historic path of gross tertiary gross graduation rates and income in Malaysia over the period 1998 to 2005 is traced along the blue curve. Source: UNESCO Institute for Statistics. Micro Evidence 202. At the micro level, PICS data show that since 2002 the quality of local professionals and skilled production workers has improved. Nevertheless, the prevalence of the skills shortages and the mismatch of skills in Malaysia has remained a constraint on business. 98 In 2002, employers believed the skills of their locally-hired professionals and skilled production workers were poor in several areas, especially IT, creativity/innovation, and communication skills, while their assessment of worker skills is higher in 2007. While firm managers' perception on worker skills seemed to improve in manufacturing, the problem is, however, still cited by a significant percentage of establishments as a major constraint to doing business: 24.7 percent (the most cited obstacle) in 2002 and 19.3 percent (the third most cited obstacle) in 2007. Conversely, the perception on skills has worsened largely in the business support services from 12.0 percent in 2002 to 20.1 percent in 2007, moving from the seventh to the third ranked problem identified. This pattern of responses across the manufacturing and business support services consistently holds across regions, firm size categories, exporters and non-exports, ownership groups, and industries. 98 We use the worker survey of PICS data for this analysis on micro evidence. See Annex 5.1 for details. 112 203. Workers have a major concern with their lack of educational attainment particularly as it relates to their own lack of preparedness for jobs. Many workers in Malaysia believe they lack of skills needed and/or do not have the relevant educational background for their work. Malaysia experienced a rather tight labor market over the past years. 99 Because of the shortage of upper secondary school and university graduates, firms have to resort to hiring high school dropouts to do the job of a high school graduate, and similarly, high school graduates do the job of a college graduate. Skills shortage and mismatch have led to a sub-optimal hiring policy to recruit less qualified worker out of desperate, which has resulted in productivity losses at the establishment level (World Bank, 2005a). 204. Manufacturing firms have levels of skilled labor that vary by industry. The shares are relatively high in firms producing machinery, chemicals, auto parts and electronics and electrical. These firms also tend to offer higher hourly wages to skilled laborers. The share of skilled labor declined most significantly in firms producing garments and wood products. Average wages are highest in chemical firms.100 The share of skilled labor is much higher in the business support services sector compared with the manufacturing sector. Over 90 percent are skilled labor in IT, telecommunication, accounting, and advertising firms; and about 65 percent are skilled labor in business logistics firms. Changes are small over time in the business support services sector. Across industries, the share declined slightly in IT, telecommunication and advertising industries. The average wage is the highest in advertising firms, followed by telecommunication and IT firms. 99 Unemployment rate is about 3-4 percent in Malaysia in 2000-2005. 100 This is consistent with the findings in Chapter 2 where chemical firms have the highest labor productivity. 113 Table 5.1: Share of Skilled and Unskilled Workers and Their Hourly Wages in the Manufacturing and Business Support Services Sectors (2001 and 2006) Manufacturing Sector Share of skilled and unskilled labor (%) Rubber & Electronics & Woods & 2001 Food Textiles Garments Chemicals Machinery Auto plastics electrical furniture Skilled 56.8 64.0 62.0 61.6 51.3 73.0 61.2 57.6 51.6 labor Unskilled 43.2 36.0 38.0 38.4 48.7 27.0 38.8 42.4 48.4 labor 2006 Skilled 48.2 51.2 46.6 61.3 46.0 64.3 53.0 56.8 38.5 labor Unskilled 51.8 48.8 53.4 38.7 54.0 35.7 47.0 43.2 61.5 labor Hourly wage (in RM) Rubber & Electronics & Woods & 2001 Food Textiles Garments Chemicals Machinery Auto plastics electrical furniture Skilled 9.0 7.6 6.3 12.7 9.3 10.1 9.7 11.7 7.5 labor Unskilled 4.5 5.1 4.2 5.9 4.7 6.9 5.6 6.4 4.4 labor 2006 Skilled 9.5 6.6 6.9 13.2 9.3 10.9 10.2 10.4 8.3 labor Unskilled 5.1 4.9 4.4 7.2 5.3 6.6 6.5 6.1 4.5 labor Business support Services Sector Share of skilled and unskilled labor (%) Information Business 2001 technology Telecommunication Accounting advertising logistics Skilled labor 99.4 96.0 92.4 94.7 65.1 Unskilled labor 0.6 4.0 7.6 5.3 34.9 2006 Skilled labor 97.9 92.8 92.3 90.1 65.3 Unskilled labor 2.1 7.2 7.7 9.9 34.8 Hourly wage (in RM) 2001 Skilled labor 23.9 22.6 14.4 22.7 15.3 12.9 12.5 8.2 13.4 7.5 Unskilled labor 2006 Skilled labor 19.2 19.9 16.4 25.7 14.3 11.2 14.2 10.8 12.7 8.3 Unskilled labor Note: Skilled labor includes management, professional, skilled production worker and unskilled labor includes the rest of categories. Wage is measured in nominal terms. Source: Malaysia PICS 2002 and 2007. 114 Employers' Perspectives 205. Firms have long perceived inadequate skills as a major constraint to doing business. As discussed in the first chapter, skills inadequacy is ranked by the highest percentage of firm managers as one of the biggest obstacles in both PICS 2002 and 2007. 206. Survey results indicate a positive trend of change in the quality of local professionals, especially in areas that are critical to technological innovation. The PICS included questions that ask managers to assess the quality of their employees on a four point scale, as "very poor", "poor", "fairly good", or "very good". In 2002, employers believe the skills of their locally-hired professionals were no better than poor in several areas, especially IT and creativity/innovation skills, but by 2007, the assessment of worker skills are higher, as depicted in Figure 5.4. In particular, a high share of managers at manufacturing establishments in 2002 believed that IT and Creativity/Innovation skills of local professionals were of low quality, but a lower percentage felt this way by 2007 (13 and 10 percent in 2002, respectively, to 7 and 8 percent in 2007).101 Similarly, among managers in the business support services firms, assessment of the IT and Creativity/Innovation skills of local professionals as being low in quality fell from 8 and 11 percent in 2002, respectively, to 5 and 8 percent in 2007. Figure 5.4: Employers' Assessments of Worker Quality: Local Professionals Manufacturing Business support Services English Proficiency 2007 English language proficiency 2007 2002 2002 Communication Professional Communication skills Social Social Skills Teamworking Teamworking skills Leadership Leadership skills Time Management Time Management skills Adaptability Adaptability skills Creativity/Innovation Creativity/innovation skills Numerical Numerical skills Problem Solving Problem solving skills IT IT skills Technical/Professional Technical/professional skills 0% 5% 10% 15 0% 5% 10% 15% % of Managers Ranking Local Professional Workers as Having "Poor" or "Very Poor" Skil % of Managers Ranking Local Professional Workers as Having "Poor" or "Very Poor" Skill Source: Malaysia PICS 2002 and 2007 101 Manufacturing firms are more open to foreign workers. While workers of foreign nationals accounts only for 3- 4 percent of the total permanent skilled workers in business support services, it accounts for 11 percent in manufacturing. This could be associated with the difference in perception on skills of manufacturing and services firms. 115 207. The perception of improvement in worker quality of local skilled production workers is larger compared to local professionals, especially in areas critical to technological innovation. A majority of managers at manufacturing establishments in 2002 believed that the Communication, Technical/Professional, and IT skills of local skilled production workers were of low quality, but a much lower percentage felt this way by 2007 (67, 63, and 48 percent in 2002, respectively, to 35, 38, and 28 percent in 2007). Similarly, among managers in the business support services, assessment of the Communication, Technical/Professional, and IT skills of local professionals as being low in quality fell from 18, 22, and 19 percent in 2002, respectively, to 11, 16, and 13 percent in 2007. The results for all skill areas are presented in Figure 5.5. Figure 5.5: Employers' Assessments of Worker Quality: Local Skilled Production Workers Manufacturing Business support Services English Proficiency 2007 English language proficiency 2007 2002 2002 Communication Professional Communication skills Social Social Skills Teamworking Teamworking skills Leadership Leadership skills Time Management Time Management skills Adaptability Adaptability skills Creativity/Innovation Creativity/innovation skills Numerical Numerical skills Problem Solving Problem solving skills IT IT skills Technical/Professional Technical/professional skills 0% 10% 20% 30% 40% 50% 60% 70% 0% 5% 10% 15% 20% 25% % of Managers Ranking Local S killed Production Workers as Having "Poor" or "Very Poor" Skil % of Managers Ranking Local Skilled Production Workers as Having "Poor" or "Very Poor" Skills Source: Malaysia PICS 2002 and 2007. Employees' Perspectives 208. Workers tend to perceive the skills problem in two ways: some recognize that they do not have the level of education/skills for their jobs (skills shortage); others realize that the skills they have would be a better fit for other jobs (skills mismatch). 209. As a result, firms resort to sub-optimal hiring. Consequently, a high percentage of employees reported their jobs required more education/skills than they actually have (Table 5.2). In 2002, 18.1 percent of manufacturing workers that attended upper secondary school but did not graduate (i.e., no diploma) estimated that their current job required a high-school diploma, and 21.5 percent of high school graduates (i.e., with diploma) believed that they need a university 116 degree to do their jobs properly.102 In 2007, these percentages of manufacturing workers have decreased slightly, but remain 16.7 and 18.3 percent, respectively. At establishments in the business support services, in 2002, 26.0 percent of workers attended upper secondary school but did not graduate estimated that their current job requires a high school diploma, and this percentage remained virtually unchanged at 25.7 percent in 2007. Lastly, of high school graduates working in the business support services, 32.6 percent in 2002 believed that they need a university degree to do their jobs properly. While the percentage of high school graduates who felt this way has decreased in 2007, the level remained high at 24.1 percent. Table 5.2: Educational Mismatch in Malaysia Manufacturing What level of education is most appropriate for your work? (%) 2002 Degree Diploma Upper Sec. Lower sec. Primary Informal/None Total Degree 77.7 14.8 3.9 2.6 0.7 0.3 100 Worker's Educational Diploma 21.5 56.7 16.7 4.1 0.3 0.7 100 Upper secondary 3.5 18.1 54.8 16.8 4.2 2.7 100 Attainment Lower secondary 1.5 6.8 31.2 44.1 10.6 5.7 100 Primary 0.4 2.4 16.3 32.3 36.3 12.3 100 2007 Degree Diploma Upper Sec. Lower sec. Primary Informal/None Total Degree 74.2 14.7 7.9 1.8 0.2 1.2 100 Diploma 18.3 59.1 17.2 3.5 0.7 1.1 100 Upper secondary 3.6 16.7 57.4 15.3 3.7 3.4 100 Lower secondary 1.8 7.4 30.0 46.2 8.1 6.7 100 Primary 0.7 3.0 21.7 31.4 31.5 11.6 100 Business Support Services What level of education is most appropriate for your work? 2002 Degree Diploma Upper Sec. Lower sec. Primary Informal/None Total Degree 88.7 8.9 1.9 0.4 0.0 0.1 100 Worker's Educational Diploma 32.6 54.3 11.1 1.4 0.2 0.4 100 Upper secondary 7.8 26.0 57.1 7.1 1.4 0.6 100 Attainment Lower secondary 3.9 11.7 17.1 56.8 8.2 2.3 100 Primary 4.0 2.0 6.0 26.0 56.0 6.0 100 2007 Degree Diploma Upper Sec. Lower sec. Primary Informal/None Total Degree 85.8 10.4 2.5 1.2 0.1 0.0 100 Diploma 24.1 63.6 8.6 3.3 0.3 0.1 100 Upper secondary 8.4 25.7 57.6 5.7 1.5 1.1 100 Lower secondary 1.1 15.2 30.9 38.7 10.0 4.1 100 Primary 0.0 6.7 10.7 29.3 44.0 9.3 100 Source: Malaysia PICS 2002 and 2007. 210. A large percentage of workers in Malaysia believe they do not have the appropriate educational background to do their jobs. Employees were asked to rate the level of relevance of their field of education to the work they were doing. Only 9 and 7 percent of manufacturing workers in 2002 and 2007, respectively, believe that the ideal field of education best suited for their job is the one they possess. Furthermore, 14 and 17 percent of manufacturing workers in 2002 and 2007, respectively, are employed in areas in which the ideal field of education is 102 In a similar study of 2,460 Dutch graduates, Allen and Velden (2001) find that 14 percent of higher vocational education graduates and 8 percent of university graduates were working in jobs for which they considered a higher level of education was more appropriate. 117 completely different from their own. Of workers in business support services, the percentages and changes over time are dramatically different. In 2002, 63 percent of business support services workers believed that the ideal field of education best suited for their job is the one they possess, but this number fell dramatically to 15 percent in 2007. 103 Furthermore, 7 and 15 percent of business support services workers in 2002 and 2007, respectively, are employed in areas in which the ideal field of education is completely different from their own. Skills mismatch has lead to productivity losses at the establishment level (World Bank, 2005a). Incidence and Causes of Vacancies 211. To measure the shortage of skills, we used the incidence and intensity of hard-to-fill vacancies; two indicators cited most often in the literature (see Green et al., 1998). As shown in Figure 5.6, the incidence of vacancies for professionals in Malaysia is high compared with some neighboring countries, such as Indonesia and Thailand (Figure 5.6). In 2002 is 26 and 42 percent of manufacturing and business support services establishments, respectively, reported vacancies of professionals. These percentages have remained unchanged / increased slightly in 2007 to 27 and 47 percent, respectively. Similarly, 52 and 36 percent of manufacturing and business support establishments in 2002 have had vacancies for skilled production workers. The incidence of vacancies among manufacturing establishment has fallen to 44 percent, while the percentage of business support services has increased to 45 percent in 2007. In 2007, the incidence of manufacturing vacancies for professional and skilled production workers in Malaysia is much higher than the levels observed in Indonesia in 2003 (17 and 25 percent, respectively) and Thailand in 2007 (20 and 30 percent, respectively). Figure 5.6: Incidence of Vacancies for Professionals and Skilled Production Workers Source: Productivity and Investment Climate Surveys, Select Countries. 212. Many employers valued experience and technical skills over educational level, loyalty or interpersonal skills when recruiting employees. The results of Figure 5.7 show the three most important considerations that an employer uses in recruiting. Almost 80 percent of 103 This is also true among business support services workers in the panel sample ­ the share of workers who believed that the ideal field of education best suited for their job dropped from 60 percent to 17 percent. The reasons for this drastically changes are of interest for further study. 118 employers in both manufacturing and business support services sectors consider experience or technical skills one of three most important factors for hiring potential employees. However, business support services firms tend to place higher weights on education level than manufacturing firms. While almost 70 percent of business support services firms consider education level of potential employees as one of three most important factors, only 53 percent of manufacturing firms do. Figure 5.7: Three Most Important Considerations in Recruiting Source: Malaysia PICS 2002 and 2007. 213. In 2007, it took Malaysian firms about four weeks to fill a vacancy for a professional or a skilled production worker. For manufacturing sector, this is longer than in some comparator countries.104 For example, the time needed to fill a vacancy for a professional in manufacturing is less than 3 weeks in Bangladesh (2002) and less than 2 weeks in India (2005) and Indonesia (2003). Similarly, filling a vacancy for a skilled production worker in manufacturing takes on average less than 2 weeks in Bangladesh (2002), Brazil (2003), and India (2005), and less than one week in Indonesia (2003). 104 The information on the number of weeks to fill job vacancies in comparable business support services sector is not available in other countries. 119 Figure 5.8: Time to Fill Vacancy for Professionals and Skilled Production Workers Note: Establishments with the number of weeks to fill a vacancy greater (less) than the mean plus (minus) 3 times the standard deviation of the respective sector, country, and year are defined as outliers and are dropped from the calculation of the means presented in the above chart. 214. Workers lack basic and technical skills that are appropriate to their jobs. That is the main reason why there is such a high vacancy rate for professionals and skilled production workers, according to firm mangers in Malaysia.105 As shown in Figure 5.9, a majority of establishments in both sectors of the two PICS rounds cited the fact that applicants do not have the required basic nor technical skills as one of the main causes for these vacancies. Consequently, a majority of the employers also believe that workers demand too high of a wage relative to their limited skills. Figure 5.10 shows the hourly wage by skills and educations, measure RM in nominal terms. In relative terms, the wage level of the unskilled production labor and the least educated (worker with informal education or illiterate worker) increased more rapidly than the other groups. 106 105 The PICS included questions asking firms to identify the most important factors leading to their difficulties in filling these vacancies. 106 The wage of unskilled production worker grew by 28 percent and that of the least educated by 30 percent. 120 Figure 5.9: Causes for Vacancies as Perceived by Managers Manufacturing 2002 Manufacturing 2007 Business-Supporting Services 2002 Business-Supporting Services 2007 % of Establishments Indicating that Factor is One of 75 the Top 3 Most Important Causes of Vacancies 50 25 0 Applicants Universities Applicants do Applicants do No applicants High turnover demand very not producing a not have not have for unskilled of new recruits high wage sufficient required basic required workers number of skills technical position graduates skills Source: Malaysia PICS 2002 and 2007 Figure 5.10: Hourly Wage by Skills and the Highest Degree Completed Source: Malaysia PICS 2002 and 2007 121 SKILLS SHORTAGES AND WAGE PREMIUMS Estimated Returns to High School Diplomas and College Degrees 215. Wage premium captures the firm managers' willingness to pay for skills that are scarce. Following the methodologies of wage premium estimation developed in Jaeger and Page (1996), Bauer, Dross, and Haisken-DeNew (2003), and Gibson (2004) to estimate wage effects of academic credentials, this section examine the extent of skills shortages in Malaysia.107 Using the PICS worker survey for manufacturing firms and services firms, respectively, we estimate the marginal returns to high school diploma and college degree controlling for workers' characteristics and establishment fixed effects.108 The sample includes both local worker whose nationality is Malaysia and foreign workers who are not native Malaysian. The results show that, in Malaysia, the returns to high school diploma are higher for workers in manufacturing sector than those in business support services sector and the difference widened over time from 2002 to 2007 with an increase in returns for workers in manufacturing sector and a decline in returns for workers in services sector. The effect of receiving a high school diploma is 11 percent in 2002 and 14 percent in 2007 for a worker in the manufacturing sector, compared to 9 percent and 5 percent, respectively, for a worker in the services sector. As expected, the returns to college degree are higher than those to high school diploma. In 2002, the marginal returns to college degree are similar for workings in manufacturing sector and services sector. However, over time, for workers in both sector, returns to college degree declined, especially in manufacturing sector. The marginal effect of receiving a college degree above a high school diploma for a worker in the manufacturing sector has fallen from 23 percent in 2002 to 17 percent in 2007; similarly, the marginal effect of receiving a college degree above a high school diploma for a worker in business support services has fallen slightly from 22 percent in 2002 to 20 percent in 2007. The wage premium in Malaysia for high school diploma is relatively high and that for college degree is relatively low, compared with the returns to education in Thai manufacturing firms109 - 5 percent to high school diploma, 6 percent to vocational certificate (Por Wor Chor), and 30 percent to college degree. 216. Furthermore, hourly wages tend to increase as the number of years of schooling goes up. (Figure 5.11).110 For both sectors in 2002 and 2007, the curves show clear increases in the slope for upper secondary and tertiary education.111 However, the wage premium for additional years of formal education fell from 2002 to 2007 in both sectors in Malaysia. This finding is consistent with other data which shows a decline in the wage premium for advanced credentials. Compared with Thailand, the premiums to workers with more than 16 years of schooling are lower in Malaysia. 107 See Annex 5.1 for details on estimation methodology. 108 See Annex 5.2 for details on definitions of workers' characteristics. 109 The Thailand PICS 2002 covers the same industries in manufacturing sector as the Malaysia PICS. 110 The curves connect the average predicted values of log hourly wages (for each year of formal education). 111 For business support services sector, the clear increasing trend starts from the 13th year of schooling and beyond. 122 Figure 5.11: Mean Log Hourly Wage by Years of Formal Education Total returns to completed years of formal education. (Relative to comparison group with 9 or less completed years of formal education.) 90 Malaysia Manufacturing 2007 Malaysia Manufacturing 2002 Malaysia Business-Supporting Services 2007 Malaysia Business-Supporting Services 2002 80 Percentage Increase in Hourly Wages 70 60 50 40 30 20 10 0 10 11 12 13 14 15 16 over 16 Completed Years of Formal Education Note: The percentage increase in wages associated with a dummy variable coefficient is calculated as e - 1. Coefficients come from the regressions identical to specifications in Table 3.8, except the base comparison group for the set of schooling dummy variables is workers with 9 or less completed years of formal education. Source: Malaysia Productivity and Investment Climate Survey 2002 and 2007; Staff's calculations. 217. However, recent trends suggest that, over time, firms in the services sector are less willing to pay for the skills of high school graduates; college graduates faced a similar problem in both sectors. The decreasing trend of wage premium for higher education may relate to the mismatch of the education profile of workers and their job requirements. In particular, from the firm managers' perspective, the important skills needed in keeping up with the emerging technology changed overtime, which further adds to the obstacles of shortages and mismatches of workers' skills. For example, in PICS 2002, IT skills, technical/professional skills and creativity/innovation for manufacturing/professional communication skills for services were the three most critical skills/competences expected of potential employees to meet the future needs. In 2007, while IT skills and technical/professional skills remained important, the percentage of firms that perceived those two among the top three most critical skills both declined. However, the percentage of firms that perceived English proficiency and numerical as important skills more than doubled for manufacturing as well as for services. Compared with manufacturing, the changes in important skills needed are larger in services sectors, for example, the demand for team-working skills, leadership skills, time management skills, and problem solving skills has grown rapidly, in addition to that for English and numeric skills aforementioned. This offers support to the worsening perception on skills of services firms. 123 Figure 5.12: Three Most Critical Skills/Competencies Needed in Keeping Up with the Emerging Technologies, the Panel Sample Manufacturing Business support Services Source: Malaysia PICS 2002 and 2007. In-Service Training 218. In-service training plays an important and popular role among Malaysian employers from many industries.112 Of manufacturing establishments, 48 percent in 2002 and 58 percent in 2007 provide some formal training for their employees. Of establishments in the business support services, 60 percent in 2002 and 55 percent in 2007 provide in-service training. This result is consistent with previous findings in the literature. Tan and Batra (1995) have shown that, in the early 1990s, the training incidence in Malaysia was at levels comparable with the United States. In 1994, 83 percent of Malaysian firms surveyed did informal training and 35 percent did formal training from any sources ­ the training incidence of Malaysia was the highest compared with other comparator countries, such as, Colombia, Indonesia, Mexico and Taiwan.113 219. Within Malaysia, the incidence of in-service training varies according to the size and type of firm. Figure 5.13 presents the incidence of formal training in Malaysian manufacturing by firm size from the Inter-Firm Linkages and Technology Development (ILTD) Survey conducted in 1997 and the PICS 2002 and 2007, as well as the business support services from the PICS 2002 and 2007. Among large-sized establishments (150 or more employees), in-service 112 In-service training is defined here as formal in-house or outside training, that is, a learning event taking place in a classroom setting inside or outside the premises of the firm. 113 The firm level training data was collected for 2,200 in Malaysia in 1994. The survey covers a wide range of firms with different characteristics in terms of age, region, firm size, foreign capital, export orientation, industry. For details, see Tan and Batra (1995). 124 training incidence is twice or three times higher than among small-sized establishments (less than 50 employees). Training incidence among, small-, medium-, and large-sized manufacturing establishments have averaged 30, 62, and 84 percent, respectively, over the past decade. Similarly, training incidence among small-, medium-, and large-sized establishments in the business support services have averaged 44, 69, and 86 percent, respectively. While this positive relationship between firm-size and training incidence is a standard result in the literature as bigger firms often finds training more effective in raising productivities and are capable to afford it at the first place, the magnitude is very important in Malaysia and may signal impediments to training for small firms. 220. For manufacturing sector, the proportion of small manufacturing establishments providing formal training to their employees changed largely in the past years. It declined from 34 percent in 1997 to 25 percent in 2002, but recovered to 31 percent in 2007. Training incidence among medium-sized manufacturing establishments increased from 56 percent in 1997 to 57 percent in 2002 to 72 percent in 2007. Similarly, training incidence among large-sized manufacturing establishments increased from 78 in 1997 to 82 percent in 2002 to 91 percent in 2007. For business support services sector, training incidence among establishments remains at their high levels over the past five years. Figure 5.13: Training Incidence by Firm Size, 1997-2007 100 1997 2002 2007 80 Percentages of Firms 60 40 20 0 Small Medium Large Small Medium Large Manufacturing Business-Support Services Source: ILTD Survey 1997; Malaysia PICS 2002 and 2007. 221. Generally, in-service training has a positive impact on productivity. Workers with firm-specific training in technical skills often get higher salaries after other characteristics (such as formal education and experience) are controlled for (Table A5.1). Workers in manufacturing who have received formal training from their current employers in the areas of management/quality technologies earned 8 percent higher wages in 2002 and 2007 than those who did not receive such training. Similarly, workers in the business support services who have received employer-provided training in the areas of management/quality technologies and information technology earned 9 percent higher wages than those who did not receive such training. 125 CONCLUSIONS AND POLICY SUGGESTIONS 222. The quantity of labor supply with high-school and tertiary education has increased over the past five years in Malaysia. However, as the economy moves towards one that is more knowledge based, these improvements still fall short of the demand. The widespread perception that Malaysia tends to have inadequate skills has persisted. First, the gap that existed between Malaysia and similar economies when it comes to human capital and overall skills at the national level continues. Second, employers and employees perceive a serious shortage and mismatch of skills, perceptions supported by empirical findings. Third, real wages and average wage premiums paid to graduates with upper secondary or tertiary education have either been stagnant or showed signs of decline. That said, some firms are willing to pay high wage premiums to attract and retain workers with the skills they need, especially to those who have studied abroad, who received employer-provided training in marketing, information technology and management or who received similar training outside of work. Due to the shortage of skills, firms often have to resort to a sub-optimal hiring policy. These practices have resulted in costly loss in productivity for individual businesses. 223. To enhance its competitiveness in a knowledge-based economy, Malaysia must improve its human capital. Expansion of upper secondary and tertiary education is an important ingredient for countries to develop the technological capabilities necessary to absorb knowledge spillovers from FDI and trade. The same applies to company training. The Government and the private sector need to foster skills demanded by the market. Public-private sector collaboration will be critical to identify specific skills that firms need and to define the skills content of secondary and tertiary education. 224. However, an increase in the number of high-school and college graduates is insufficient to correct this skills constraint. High job vacancy rates among skilled workers/professionals and unemployment among certain college graduates suggests imperfections in the labor market and education system. It is important to balance expanding the higher education system and improving quality (World Bank, 2007).These flaws, which exist side by side, have persisted over the past few years. They prevent the labor market from matching supply and demand and they hamper the education system from providing future workers with the skills and potential that firms require. 126 CHAPTER 6. STRENGTHENING TECHNOLOGICAL CAPABILITIES 226. Technological progress plays an increasingly important role in the modern economy. As technological capabilities go up, it translates into better performance. Malaysia faces a daunting challenge to improve its technological capabilities and gear its economy on a path to higher growth and transition to a higher level of development. 227. The inadequacy of skills limits improvements in its technological capabilities. PICS found that Malaysian firms continue to be characterized as "adaptors" rather than "creators" when it comes to technology. Despite significant improvement in technological input in the recent past, as measured by the supply of researchers and technicians and R&D expenditures, and in technological output, as measured by the number of patents issued, Malaysia's technological capacities still lag behind those of better performers, such as Korea and Singapore. 228. Malaysian firms need to have stronger technological capabilities, especially linkages capabilities. In order to efficiently translate technological advances into higher productivity and performance, stronger linkages are needed between local suppliers to transfer technology and greater cooperative efforts are needed between firms and other institutions on technology research and development. As technological capacities are closely associated with firms' productivity, focusing on fostering the environment that promotes and enables the development of firm-level technological capabilities, namely through investment in human capital, linkages, and competition, would be important for Malaysia. 229. The chapter is structured as follows. Section 1 presents Malaysia's technological performance. It starts by benchmarking with other countries, and follows by detail information from both the input and output aspects. Section 2 examines the technological capacity of manufacturing firms using the Technological Capability Index (TCI) at the national level as well as across regions, industries, and different types of firms. It also explores the relationship between firm-level technological capabilities and TFP through regression analysis. Section 3 concludes with a discussion of policy recommendations. MALAYSIA'S TECHNOLOGICAL PERFORMANCE 230. This section provides a snapshot of technological performance by Malaysian manufacturers. Based on the Global Competitiveness Indicators, Malaysia's technological performance is reasonably good compared with countries at similar development level, but improvements in several areas are needed to catch-up with the more advanced economies, such as Korea and Singapore. Available data indicate that the country falls short in terms of the level of R&D expenditures, the supply of researchers, and the number of patents filed that are consistent with its level of development. A large ratio of high-tech export, which mainly reflects its ability to attract FDI into capital-intensive activities involved in the assembly of "high tech" 127 products, as opposed to being engaged in creation and design processes, does not indicate strong technological capacity in Malaysia. Global Competitiveness Technology Indicators 231. Malaysia's technological performance is reasonably good compared to regional competitors and countries at the same level of development, but it still lags behind more advanced economies, such as Korea and Singapore. Table 6.1 shows that benchmarking with 131 countries in the sample, Malaysia ranks 30 and 21, respectively, in terms of technological readiness and innovation using Global Competitiveness Index (Box 6.1). Table 6.1: Ranking of Technological Readiness and Innovation Sub-Indexes Technological Readiness Innovation (rank out of 131 countries) Malaysia 30 21 Chile 42 45 China 73 38 India 62 28 Indonesia 75 41 Japan 20 4 Korea, Rep. 7 8 Singapore 12 11 Thailand 45 36 United Kingdom 16 14 United States 9 1 Source: World Economic Forum (2007). Box 6.1: Global Competitiveness Index Global Competitiveness Index (GCI), developed by the World Economic Forum (WEF), takes into account the differences in the role of specific pillars of growth across countries in their stage of development. The index is derived from hard data and an Executive Opinion Survey, including both qualitative and quantitative data. It is composed of sub-indices related to twelve pillars of growth including, among other factors, institutions, macroeconomic stability, human capital, technological readiness, and innovation. Technological readiness and innovation are two pillars closely relevant to technology. · The technological readiness pillar includes: technological readiness, firm-level technology absorption, quality of competition in the ISP sector, laws relating to ICT, cellular telephones, internet users, local availability of specialized research and training services. · The innovation pillar includes: quality of scientific research institutions, company spending on R&D, university/industry research collaboration, government procurement of advanced technology products, availability of scientists and engineers, utility patents, IPR protection, capacity for innovation (GDP - exports + imports). It is relatively more important for countries that are close to the knowledge frontier. Technological non-core economies are those where innovation is not crucial for growth. They are relatively (compared to technological core economies) distant from the technological frontier, and can benefit more from technology adoption/technology transfer, copy and imitation. The index for the core technology economies consist of two sub-indices: innovation and ICT. As countries are not necessarily at the same stage of development, the relative importance of different pillars in explaining competitiveness and economic performance vary. Based on this, the weights of the pillars in the index change with a country's level of development (a detailed discussion on the maximum likelihood-based weighing methodology used to compute the GCI is included in Chapter 1.3 of the 2005 Global Competitiveness Report). For instance, the innovation pillar is given a lower weight for countries below a certain income threshold. This is a practical approach that highlights the most important pillars for each country and helps facilitate country-specific policy recommendations. 128 232. Technological readiness and innovation are two important pillars closely related to technology. Technological readiness takes into account the level of technology available to firms; and innovation looks at a country's ability to innovate. The former is more important for countries that are below the production frontier, while the latter is more important for countries that are close to the production frontier. Malaysia ranks higher in innovation, than in technology readiness, but India, Indonesia, and China follow more closely behind. Technological Input Indicators 233. Technological input reflects the country's ability to use, develop, and absorb technology. This includes indicators such as the number of graduates from university and training institutes, R&D expenditures, and research staff,. While the previous chapter discusses skills and education related issues, this sub-section focused on R&D expenditure and staff. Research and Development (R&D) Expenditures 234. At the national level, R&D expenditures in Malaysia grew from 1996 to 2005. And yet, they remain far below their regional competitors and what is expected for Malaysia's level of income. 114 Table 6.2 shows that the level of R&D expenditures as a percentage of GDP in Malaysia is far below Singapore and, to a lesser extent, China. Figure 6.1 shows that over the period of 1996 to 2004 the level of R&D expenditures in Malaysia has been consistently below the expected value for its level of income.115 These results warrant in-depth sector and industry- specific analyses, as well as an assessment of past government strategies to stimulate R&D. Table 6.2: Research and Development Expenditure (as percent of GDP) 1996 1997 1998 1999 2000 2001 2002 2003 2004 Malaysia 0.2 .. 0.4 .. 0.5 .. 0.7 .. 0.6 Chile 0.5 0.5 0.5 0.5 0.5 0.5 0.7 0.7 0.7 China 0.6 0.6 0.7 0.8 0.9 1.0 1.1 1.1 1.2 India 0.5 0.7 0.7 0.8 0.8 0.7 0.7 0.7 0.6 Japan 2.8 2.9 3.0 3.0 3.1 3.1 3.2 3.2 3.2 Korea, Rep. 2.4 2.5 2.3 2.3 2.4 2.6 2.5 2.6 2.8 Singapore 1.4 1.5 1.8 1.9 1.9 2.1 2.1 2.1 2.2 Thailand 0.1 0.1 .. 0.3 0.3 0.3 0.2 0.3 0.3 United Kingdom 1.9 1.8 1.8 1.9 1.9 1.8 1.8 1.8 1.7 United States 2.5 2.6 2.6 2.7 2.7 2.8 2.7 2.7 2.7 Note: Expenditures for research and development are current and capital expenditures (both public and private) on creative work undertaken systematically to increase knowledge, including knowledge of humanity, culture, and society, and the use of knowledge for new applications. R&D covers basic research, applied research, and experimental development. Source: United Nations Educational, Scientific, and Cultural Organization (UNESCO) Institute for Statistics. 114 Ideally, cross-country comparisons of technological indicators should take into account that countries are at different stages on the technological path, and therefore, indicators should be normalized in some manner before comparisons are made. Unfortunately, lack of data prohibits such weighting from being undertaken. 115 For a more rigorous analysis and interpretation, Figure 6.1 should normally only take into account countries with comparable economic and industrial structures since the relevance and level of R&D depends on these structures. This is not possible due to lack of data. The figure should therefore be interpreted with caution. However, notwithstanding these concerns, it gives an approximate sense of the country's standing. This remark also applies to Figure 6.3 which shows the number of researchers. 129 Figure 6.1: R&D Expenditures and Level of Development Note: Israel (10.0, 4.6 percent) and Congo, Dem. Rep. (5.6, 0.4 percent) have been removed from the figure to better visualize the gaps. Based a sample of 101 countries, the red best fit line has a slope of 0.462 that is significant at the 1 percent level and an intercept of -3.2. The historic path of R&D expenditures and income in Malaysia over the period 1996 to 2004 is traced along the blue curve. Source: World Development Indicators Database 2008. Staff calculations. 235. In Malaysia's 2007 PICS sample, 16.8 percent of manufacturing establishments reported positive expenditures for R&D activities in the previous year, which reflects a decline from 18.8 percent in 2002.116 As greater R&D expenditures are often associated with an improving readiness of moving up the technological ladder, the decline over time indicates worrisome trends.117 236. Usually large, exporting, and foreign-owned establishments spent more on R&D (Figure 6.2). In 2002 and 2007, 38.2 and 29.4 percent of large-sized manufacturing establishments, respectively, reported R&D expenditures, whereas only 12.6 and 11.9 percent of SMEs in manufacturing did so.118 Similarly 26.4 and 22.1 percent of exporters versus 10.8 and 116 The difficulty in interpreting this percentage as well as the change over time is due to the lack of knowledge about the nature, quality, and sophistication of such activities, which may well be very different across firm characteristics and survey years. 117 In the future, regression analysis based on annual manufacturing census data would be useful to better assess the contribution of R&D to TFP growth in Malaysia in order to better guide policy recommendations. Several studies have investigated the contribution of TFP in other countries. Comin (2004), for example, argues that R&D is actually a small part of TFP growth. 118 Large-sized establishments employ 150 or more permanent workers, medium-sized establishments employ 50- 150 permanent workers, and small-sized establishment employ less than 50 workers. 130 10.3 percent of non-exporters report R&D expenditures in 2002 and 2007, respectively.119 In addition, 25.4 and 23.7 percent of foreign-owned versus 15.9 and 13.7 percent of domestically- owned establishments report R&D expenditures in 2002 and 2007, respectively.120 Percentages decline from 2002 to 2007 in all groups. 237. The North and the Klang Valley regions spent the highest percentage on R&D; Sabah and the South spent the least. In the North region, 23.8 and 21.9 percent of establishments reported R&D expenditures in 2002 and 2007, respectively, and in the Klang Valley region, 22.9 and 20.7 percent did so. In Sarawak, the percentages increased from 5.9 to 10.9 percent. Conversely, in Sabah and the South region, the incidence of R&D expenditures decreased from 16.7 and 14.4 percent in 2002 to 9.7 and 10.4 percent, respectively. 238. Generally, the Electronics and Electrical Appliances industries spent more on R&D; the Garments and Wood & Furniture industries spent the least. This is in general consistent with the observations in other countries that industries with higher-technical content often spend more on R&D. In Electronics, 29.3 and 28.2 percent of establishments reported R&D expenditures in 2002 and 2007, respectively. Conversely, in Garments and Wood & Furniture, the percentages declined from 14.7 to 6.7 percent and from 13.6 to 12.4 percent in 2002-2007, respectively. Except for Food Processing, where incidence rose from 15.5 percent in 2002 to 16.5 percent in 2007, the percentages fell in all industries, particularly in Motor Vehicles (26.3 to 11.6 percent), Textiles (26.7 to 15.4 percent), Garments (14.7 to 6.7 percent), Machinery & Equipment (24.1 to 17.2 percent), and Chemicals (24.2 to 18.5 percent). 119 An establishment is defined as being foreign-owned if 10 percent of equity is held by foreign nationals, in accordance with the International Monetary Fund criterion that distinguishes between portfolio and direct foreign investment flows. 120 Exporters are those establishments with direct and indirect exports amounting at least 10 percent of total sales. 131 Figure 6.2: Manufacturing Establishments Reporting R&D Expenditures, 2002 and 2007 By Firm Characteristics 40% 2002 2007 30% 20% 10% 0% All SME Large Non- Exporter Domes tic Foreign Firms exporter By Region 40% 2002 2007 30% 20% 10% 0% Klang Valley North South Eas t Coas t Sabah Sarawak By Industry 40% 2002 2007 30% 20% 10% 0% Food Proces s ing Textiles Garments Chemicals Rubber & Plas tics 40% 2002 2007 30% 20% 10% 0% Machinery & Equipment Electronics Motor Vehicles Wood & Furniture Source: Malaysia PICS 2002 and 2007. The Shares of Researchers and Technicians in R&D 239. In Malaysia the supply of R&D researchers and technicians increased over the past ten years; however, this percentage still lags significantly behind the comparator countries. The shares of researchers and technicians in the population are indicators of technological input that reflect the capability of a country to build a knowledge base, and hence, approach or challenge its technological frontier. Researchers in R&D are professionals engaged in the conception or creation of new knowledge, products, processes, methods, or systems and in the management of the projects concerned, and as an indicator of skills in a country, is measured by the total number per million people. Technicians in R&D and equivalent staff are people whose main tasks require technical knowledge and experience in engineering, physical and life sciences 132 (technicians), or social sciences and humanities (equivalent staff), and this indicator is also measured as the total number per million people. Technicians participate in R&D by performing scientific and technical tasks involving the application of concepts and operational methods, normally under the supervision of researchers. 240. The share of Malaysian researchers and technicians in R&D remains a fraction of that in industrialized nations. This gap may negatively affect the technological capabilities and innovative capacity of Malaysia. As seen in Table 6.3, the number of researchers and technicians per million people has grown tremendously from 91 and 31 in 1996 to 509 and 64 in 2004, respectively, yet Malaysia considerably lags behind Singapore, and to a lesser extent, China. As illustrated in Figures 6.3 and 6.4, the number of researchers and technicians in Malaysia is below not only the level of regional competitors but also the expected value given its per capital GDP. The deficits have been persistent as the supply of researchers and technicians has not kept up with the pace of economic development in Malaysia. Table 6.3: Researchers and Technicians in R&D (Per Million People) Researchers in R&D (Per Million People) 1996 1997 1998 1999 2000 2001 2002 2003 2004 Malaysia 91 .. 156 .. 279 .. 299 .. 509 Chile 390 395 402 404 411 413 440 772 833 China 445 474 387 420 546 579 627 663 708 India 157 .. 119 115 .. .. .. .. .. Indonesia .. .. .. .. 215 202 .. .. .. Japan 4,907 4,958 5,163 5,198 5,098 5,310 5,070 5,287 5,294 Korea, Rep. 2,190 2,242 2,005 2,156 2,317 2,898 3,003 3,187 3,279 Singapore 2,538 2,621 2,986 3,211 4,140 4,086 4,353 4,745 4,999 Thailand 102 74 .. 171 .. 286 .. 287 .. United Kingdom 2,501 2,508 2,706 .. .. .. .. .. .. United States .. 4,211 .. 4,483 4,537 4,600 4,605 .. .. Technicians in R&D (Per Million People) 1996 1997 1998 1999 2000 2001 2002 2003 2004 Malaysia 31 .. 44 .. 40 .. 58 .. 64 Chile 266 267 276 303 308 303 .. 296 302 India 114 .. 102 90 .. .. .. .. .. Japan 667 662 687 667 621 540 526 528 .. Korea, Rep. 630 576 530 563 453 453 496 567 .. Singapore .. 370 353 370 339 350 381 .. .. Thailand 39 75 .. 87 .. 115 .. 208 .. Notes: Researchers in R&D are professionals engaged in the conception or creation of new knowledge, products, processes, methods, or systems and in the management of the projects concerned. Postgraduate PhD students (ISCED97 level 6) engaged in R&D are included. Technicians in R&D and equivalent staff are people whose main tasks require technical knowledge and experience in engineering, physical and life sciences (technicians), or social sciences and humanities (equivalent staff). They participate in R&D by performing scientific and technical tasks involving the application of concepts and operational methods, normally under the supervision of researchers. Source: UNESCO Institute for Statistics. 133 Figure 6.3: Researchers in R&D and Level of Development Note: Finland (10.3, 7,832), Iceland (10.4, 6,807), and 5 African countries with log GDP per capita less than 7 have been removed from the figure to better visualize the gaps. The historic path of researcher in R&D (per million people) and income in Malaysia over the period 1996 to 2004 is traced along the blue curve. Source: World Development Indicators Database 2008. Staff calculations. Figure 6.4: Technicians in R&D and Level of Development Note: Luxembourg (11.1, 3,719), Switzerland (10.5, 2,366), Iceland (10.4, 2,052), Netherlands (10.4, 1,765) and 5 African countries with log GDP per capita less than 7 have been removed from the figure to better visualize the gaps. The historic path of technicians in R&D (per million people) and income in Malaysia over the period 1996 to 2004 is traced along the blue curve. Source: World Development Indicators Database 2008. Staff calculations. 134 241. PICS data indicates that less than one-sixth of manufacturing establishments in Malaysia employ researchers and technicians in R&D. This number has declined over the period 2002 to 2007. In the Malaysia 2007 PICS sample, only 11.9 percent of manufacturing establishments employed staff exclusively for R&D, down from 15.0 percent in 2002. In contrast, the percentages are much higher and growing in Thailand-- from 20.9 percent in PICS 2002 to and 22.3 percent in PICS 2007.121 Figure 6.5 presents the percentage of manufacturing establishments in Malaysia employing staff exclusively for R&D across firm size, exports, ownership, region, and industry groups. 242. R&D expenditures vary across firm with different characteristics. Consistent with the findings on the incidence of R&D expenditures among the PICS sample, more large-sized, exporting, and foreign-owned establishments employ researchers and technicians in R&D than SMEs, non-exporters, and domestically-owned establishments, respectively, and the declining trend is evident in all groups, as well. As depicted in Figure 6.5, 33.2 and 23.9 percent of large- sized manufacturing establishments in 2002 and 2007, respectively, employed staff exclusively for R&D, whereas only 9.2 and 7.6 percent of SMEs in manufacturing did so in 2002 and 2007. Similarly 19.3 and 17.2 percent of exporters versus 10.3 and 5.4 percent of non-exporters employed researchers and technicians in R&D in 2002 and 2007, respectively. In addition, 20.7 and 16.5 percent of foreign-owned versus 12.6 and 10.0 percent of domestically-owned establishments employed researchers and technicians in R&D in 2002 and 2007, respectively. Percentages decline from 2002 to 2007 in all groups. 243. The highest percentage of manufacturing establishments employing researchers and technicians in R&D are in the Klang Valley and the North regions; Sabah and Sarawak has the lowest. In the Klang Valley region, 19.0 and 14.6 percent reported employing staff exclusively for R&D in 2002 and 2007, respectively, and similarly, in the North region, 18.2 and 14.4 percent did so. Notably, the percentage increased from 0 to 10.3 in the East Coast over the period 2002 to 2007. Conversely, in Sarawak, the incidence decreased from 8.8 percent in 2002 to 2.2 percent in 2007. The percentages have remained low in Sabah over this period. 244. Electronic firms tend to employ the highest percentage of researchers and technicians in R&D, motor vehicles the least. In Electronics, 29.3 and 25.0 percent of establishments employed staff exclusively for R&D in 2002 and 2007, respectively. In Garments, incidence rose sharply from 4.9 percent in 2002 to 12.2 percent in 2007. Conversely, the incidence has dropped dramatically in Motor Vehicles, from 21.1 percent in 2002 to 2.9 percent in 2007. The percentages fell in all other industries, particularly in Chemicals (30.3 to 17.1 percent), Wood & Furniture (21.0 to 9.4 percent), and Machinery & Equipment (16.1 to 7.5 percent).122 121 Source: Thailand 2002 and 2007 PICS. 122 The reasons for the sharp changes over time may be related to the small number of firms which employed staff exclusively for R&D, and is of interest of further research. 135 Figure 6.5: Manufacturing Establishments Employing Staff Exclusively for R&D, 2002 and 2007 By Firm Characteristics 40% 2002 2007 30% 20% 10% 0% All SME Large Non- Exporter Domestic Foreign Firms exporter By Region 40% 2002 2007 30% 20% 10% 0% Klang Valley North South Eas t Coast Sabah Sarawak By Industry 40% 2002 2007 30% 20% 10% 0% Food Processing Textiles Garments Chemicals Rubber & Plastics 40% 2002 2007 30% 20% 10% 0% Machinery & Equipment Electronics Motor Vehicles Wood & Furniture Source: Malaysia PICS 2002 and 2007. Technological Output Indicators 245. Technological outputs are the result of a successful combination of firm-level capabilities and the institutional framework supporting them. They can therefore be used to assess the efficiency of the technological inputs and, more generally, to give a sense of technological performance. Firm and national-level indicators taken together can provide a 136 sense of technological output performance in Malaysia. At the firm level, the PICS data provides information on patents, specifically the number of utility models, patents, and copyright protected materials filed during the last two years. At the national level, the most commonly used technological output indicators are patents issued by the United States to residents in foreign countries and high-technology exports. Patents 246. PICS data show that the number of firms filing patents rose from 2002 to 2007. This increase is consistent across most firm characteristics, regions, and industry groups (Figure 6.6). In 2007, 15.2 percent of manufacturing establishments in Malaysia surveyed declared that they did file a patent during the last two years, up from 10.6 percent in 2002. Among large-sized establishments, 17.0 and 20.0 percent declared filing patents in 2002 and 2007, respectively, compared to 8.5 and 15.1 percent of SMEs in 2002 and 2007. The percentage is also higher for exporters (13.6 percent in 2002 and 19.3 percent in 2007) than for non-exporters (7.4 percent in 2002 and 10.1 percent in 2007). Among foreign-owned establishments, 12.4 and 17.1 percent declared filing patents in 2002 and 2007, respectively, compared to 9.9 and 14.3 percent of domestically-owned ones in 2002 and 2007. Regionally, the percentage of manufacturing establishment filing patents increased from 2002 to 2007 in the Klang Valley, South, and East Coast regions and Sabah, whereas in the North and Sarawak the percentage fell. The incidence of establishment filing patents increased in all manufacturing industries, except Chemicals, Wood & Furniture, and Textiles. 137 Figure 6.6: Manufacturing Establishments Filing Patents, 2002 and 2007 By Firm Characteristics 30% 2002 2007 20% 10% 0% All SME Large Non- Exporter Domestic Foreign Firms exporter By Region 30% 2002 2007 20% 10% 0% Klang Valley North South East Coast Sabah Sarawak By Industry 30% 2002 2007 20% 10% 0% Food Processing Textiles Garments Chemicals Rubber & Plastics 30% 2002 2007 20% 10% 0% Machinery & Equipment Electronics Motor Vehicles Wood & Furniture Source: Malaysia PICS 2002 and 2007. 247. Malaysia's innovation can be more accurately assessed by using the number of patents that Malaysian residents filed in the United States. Cross-country comparison using patent indicators from the respective PICS data remains difficult since national requirements for filing patents are not necessarily the same and the commercial value of these patents is also unknown. A comparison of the number of patents filed in the United States can provide clearer assessment of the relative performance of countries since the requirements are the same for all applicants everywhere. Tables 6.4 and 6.5 shows that the number of patents issued by the United States to residents in Malaysia has been persistently very low over the 1990 to 2006 period and 138 trails far behind Korea and Singapore. The number of U.S. patents issued to Malaysia is well below expected levels given its per capita GDP, as depicted in Figure 6.7. Table 6.4: Patents Issued by the United States to Residents of Selected Countries, 1996- 2007 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Malaysia 21 26 38 27 51 51 57 65 86 95 124 154 Chile 4 4 16 12 15 15 13 16 17 15 12 25 China 51 59 87 86 143 239 347 442 551 583 868 1,139 India 37 43 80 109 123 159 254 339 366 405 470 560 Indonesia 2 12 9 4 15 9 14 13 12 36 11 16 Japan 22,979 24,314 30,490 30,425 34,563 34,875 34,954 37,860 37,734 34,079 36,482 36,656 Korea, Rep. 1,428 1,828 3,052 3,477 3,699 3,783 3,775 4,198 4,590 4,811 5,835 6,882 Singapore 87 111 122 134 220 299 392 443 498 420 424 457 Thailand 8 16 14 23 36 46 49 53 33 28 38 29 United Kingdom 2,668 2,787 3,548 3,686 4,241 4,425 4,076 4,117 4,047 3,744 3,978 4,100 United States 64,562 69,294 85,783 89,119 100,548 101,619 93,347 99,898 97,913 85,238 96,174 94,618 Notes: Data include utility, design, plant, and reissue patents. FY 2007 numbers are preliminary. Source: United States Patent and Trademark Office (USPTO), 2007 Performance and Accountability Report. Table 6.5: Patents Issued by the United States to Residents of Selected Countries (Per Million People), 1996-2007 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Malaysia 0.994 1.200 1.711 1.187 2.191 2.145 2.350 2.629 3.414 3.703 4.748 5.800 Chile 0.274 0.270 1.065 0.788 0.973 0.962 0.824 1.003 1.054 0.921 0.730 1.507 China 0.042 0.048 0.070 0.069 0.113 0.188 0.271 0.343 0.425 0.447 0.662 0.863 India 0.039 0.045 0.081 0.109 0.121 0.154 0.242 0.318 0.339 0.370 0.423 0.499 Indonesia 0.010 0.061 0.045 0.020 0.073 0.043 0.066 0.061 0.055 0.163 0.049 0.071 Japan 182.720 192.829 241.199 240.229 272.428 274.285 274.267 296.434 295.348 266.715 285.560 286.889 Korea, Rep. 31.367 39.779 65.936 74.587 78.689 79.888 79.282 87.734 95.462 99.619 120.513 141.808 Singapore 23.701 29.241 31.065 33.849 54.619 72.257 93.870 107.660 119.519 98.457 96.333 99.595 Thailand 0.138 0.272 0.235 0.383 0.593 0.752 0.794 0.853 0.527 0.444 0.599 0.454 United Kingdom 45.868 47.791 60.663 62.813 72.013 74.862 68.703 69.113 67.585 62.165 65.648 67.175 United States 239.656 254.143 310.972 319.377 356.279 356.275 323.980 343.533 333.448 287.474 321.225 313.698 Notes: Data include utility, design, plant, and reissue patents. FY 2007 numbers are preliminary. Source: United States Patent and Trademark Office (USPTO), 2007 Performance and Accountability Report. World Development Indicators Database 2008. Staff calculations. 139 Figure 6.7: Patents Issued by the United States to Residents of Foreign Countries and Level of Development, 2006 Note: Taiwan, China (10.2, 322.3) and Japan (10.3, 285.6) have been removed from the figure to better visualize the gaps. The historic path of the number of patents per million people and income in Malaysia over the period 1990 to 2004 is traced along the blue curve. Source: United States Patent and Trademark Office (USPTO), 2007 Performance and Accountability Report. World Development Indicators Database 2008. Staff calculations. Adopters, Adapters or Creators 248. Malaysia continues to be more an "adapter" than a "creator" based on the framework for technological transition developed in the first productivity and competitiveness report (World Bank, 2005). Manufacturing establishments can be classified into adopter, adapter, or creator. Specifically, · An establishment is classified as an adopter if it has either upgraded machinery and equipment; or introduced new technology that has substantially changed the way the main product is produced in the last two years, but has not undertaken any of the three activities listed below. · An establishment is classified as an adapter if it has either upgraded of an existing product line or entered new markets due process or product improvements in quality or cost, but has not filed patents in the last two years · An establishment is classified as a creator if in the last two years it has filed any patents/utility models or copyright protected materials. 140 As seen in Figure 6.8, only 10.6 percent of the firms sampled in 2002 were creators, 48.4 percent were adapters, 9.5 percent were adopters, and 31.6 percent did not undertake any technological innovation-related initiatives in the last two years. In 2007, the percentage of creators improved substantially to 15.1 percent, but remained far below that of adapters and with no activity. Figure 6.8: The Technology Spectrum in Malaysia in 2002 and 2007 50 % of PICS Sample in 2002 2007 40 Respective Year 30 20 10 0 No Activity Adopters Adapters Creators Source: Malaysia PICS 2002 and 2007. 249. There is room for improving the diffusion of technology to several specific groups. Many firms do not engage in any activities related to technological innovation. Figure 6.9 shows, as expected, large-sized, exporting, and foreign-owned establishments tend to be more technologically active than SMEs, non-exporters, and domestically-owned establishments, respectively. Public policies can play an important role to bridge the gap between large and small-sized establishments, as well as the broader rationale for promoting exports and FDI in general, by diffusing technologies through explicit linkages between firms or demonstration effects. Studying the extent of linkages between large multinational corporations and small, local firms is a subject that is being taken seriously in Malaysia at present and it should continue to be the focus of investment and technology policies. 141 Figure 6.9: The Technology Spectrum by Firm Characteristics in 2002 and 2007 Source: Malaysia PICS 2002 and 2007. 250. Variations exist in regional and industrial technological activity. These activities need to expand, especially in certain critical areas of the economy. Figure 6.10 shows the variation in technological spectrum across regions. In 2007, 67.4 and 41.9 percent of manufacturing establishments in Sarawak and Sabah, respectively, undertook no technological innovation-related initiative, respectively. Figure 6.11 shows the variation in technological spectrum across manufacturing industries. In 2007, large segments of several industries did not engage in any technological innovation-related activities at all: Garments (53.3 percent), Wood & Furniture (40.8 percent), Textiles (40 percent), Machinery & Equipment (38.7 percent), and Food Processing (38.3 percent). 142 Figure 6.10: The Technology Spectrum by Region in 2002 and 2007 2002 2007 Source: Malaysia PICS 2002 and 2007. 143 Figure 6.11: The Technology Spectrum by Industry in 2002 and 2007 2002 2007 Source: Malaysia PICS 2002 and 2007. 251. At the first glance, the "adapters" status might seem to be in contrast with Malaysia's large share of total manufacturing exports in high-technology products.123 Figure 6.12 shows that Malaysia is well above its high-technology exports potential. This seemingly controversy can be explained by the following two factors. First, the same high- technology products exported can be the outcome of different processes for different countries; a process involving sophisticated design and fabrication as opposed to a simple, relatively low- skilled based assembly process that is less productivity-enhancing. Second, the nature or composition of high-technology exports may vary from country to country. Firms within the same industry can vary largely in their technical content. For example, firms in high-tech industry can be divided into three types: (i) firms assembling simple products; (ii) firms designing products while learning how to innovate; and (iii) firms designing products and 123 High-technology exports are provided by the United Nations COMTRADE database, and are defined as "products with high R&D intensity, such as aerospace, computers, pharmaceuticals, scientific instruments, and electrical machinery". The database has its limitations since it also does not take into account assembly activities. Manufactures comprise commodities in SITC sections 5 (chemicals), 6 (basic manufactures), 7 (machinery and transport equipment), and 8 (miscellaneous manufactured goods), excluding division 68 (non-ferrous metals) and are provided by World Bank staff estimates from the COMTRADE database maintained by the United Nations Statistics Division. Both series are available from the World Development Indicator 2008. The share of high-technology exports in total manufactured product exports is expected to increase with industrialization. A large share of high- technology exports in total manufactured products exports often gives an indication of a high degree of sophistication and technological capabilities of the manufacturing sector. 144 conducting R&D for product innovation.124 Subject to the relative importance of each of these three types of firms, a high ratio of exports coming from firms in high-tech industries may not necessarily indicate the a high share of exports are of high-technological intensity.125 Therefore, while the increase in high-technology exports is a good sign, it does not offer enough support to rapid productivity-enhancing technological progress in the Malaysia. 252. International competition offers a double-edged sword: it raises a threat to export markets and a stimulus for upgrading technology. As Malaysia becomes increasingly challenged by the entry of new competitors in the global marketplace, the country needs to strengthen its national technological capabilities, learn by doing, build its own brand images, and upgrade to more sophisticated products. Forward-looking strategies that focus on skills development, importing of commercial firm-level R&D, and access to foreign technology should be implemented. Figure 6.12: High-Technology Exports can be a Misleading Indicator of Technological Performance Source: World Development Indicators Database, 2008. 124 See Hobday (1995) for a discussion of this classification and Hobday (2000) for a review of case studies of countries that have successfully managed their transition from low-tech to high-tech status. 125 Unfortunately, information of the distribution of the different types of firms is not currently available. Information on import content of exports would help in better characterizing high-technology exports, but that would require detailed data on all the imported parts used by different industries, which are not available. For example, it is difficult to find data on all the parts imported to make a specific type of automobile. Also, it is difficult to know whether the final products are exported or sold locally. Concerted efforts of setting up a mechanism to systematically collect, update, and analyze these type of indicators in the future would be useful. 145 253. In short, Malaysia needs to improve its technological performance to catch up with its leading regional competitors. The second part of this chapter goes beyond the experience of successful regional competitors and focuses on identifying the potential key drivers of technological upgrading and learning using the PICS data. Adopting a microeconomic perspective and regression analysis will help shed light on the complex relationship between technological capabilities and firm performance. Empirical findings will help identify the most pressing issues in technology development facing Malaysian firms and guide policy recommendations. MANUFACTURING SECTOR FIRM LEVEL TECHNOLOGICAL CAPABILITY ANALYSIS 254. A firm's technological capacity is important for growth in productivity. Industrial technological development is not a process that can be promoted instantly or easily by only investing in new equipment or imported technology, but rather requires conscious and continuous investments by firms in their own technological capability. That is, simply purchasing new machinery or entering into a partnership with an MNC is insufficient for establishments in Malaysia to catch up with global leaders. A growing literature stresses the difficult firm-specific processes involved in building technological capabilities and argues that enterprises have to undertake conscious investments to put technology to productive use.126 Transfer necessarily requires learning because many aspects of technologies are tacit; technological knowledge is difficult to embody in hardware or written instructions. The process of getting a new technology into production requires the development of new skills and information. Mastery of new technologies, ultimately, can only be acquired through concerted effort, skill upgrading, and investments in training, research and development activities, and extensive experience. This section uses the Technological Capabilities Index (TCI), developed by Lall (1992), to examine technological capabilities at the firm level in manufacturing sector. 255. Firm-level technological capability (FTC) encompasses the skills, knowledge, and experience that enterprises must accumulate to operate efficiently, and more important, to innovate their products and processes.127 It is at the firm-level where empirical analysis can help identify sources of low technological capabilities, and derive appropriate policy responses. At a macro level, national technological capability (NTC) is defined as the aggregation of firm- level technological capabilities, taking into account linkages between firms and other economic agents, combined with all the factors and institutions that support the "technology" enabling environment.128 The nature of these factors differs across countries and can play a significant role in explaining differences in NTC. 256. This section draws on PICS data to examine technological factors that affect productivity and competitiveness at the firm-level. 129 The empirical evidence from PICS suggests that poor technological performance in Malaysia is associated with low levels and 126 See Pack and Westphal (1986), Katz (1987), Lall (1992), Bell and Pavitt (1993), Pietrobelli (1997), Wignaraja (1998), Romijn (1999), Wignaraja (2002), and Metcalfe (2003). 127 Lall (2002). 128 Lall (2000) defines NTC as "the complex interaction of skills, experience, and effort that enables a country's enterprises to efficiently buy, use, adapt, improve, and create technologies...It comprises the non-market system of interfirm networking and linkages, way of doing business, and the web of supporting institutions". 129 Information is available only for manufacturing firms, but not available for business support services firms. 146 stagnant growth of activities related to generating linkages within the economy; such an environment leads to less innovation. Our study fills the gaps in existing research by testing econometrically how this affects capability acquisition and examining the reasons the key determinants of technological capacities and performance is important in Malaysia. Investment, Production, and Linkages Technological Capability Analysis 257. The Technological Capabilities Index (TCI) captures a variety of objective information into coherent measures of a firm's capacity to establish, operate, and transfer technology. Following Wignaraja (1998, 2002), TCI draws on the taxonomy developed by Lall (1992), which identifies and categorizes firm-level technological capabilities into investment, production, and linkages activities. TCI provides a composite measure of technological capabilities composed of information about firm-level technological behavior that is provided by the rich data collected in the PICS. A high TCI score is associated with a close engagement with technological related activities. 130 258. Manufacturing establishments in Malaysia lag far behind in creating linkages to boost their technological capabilities. Moreover, average scores on technological capabilities associated with developing linkages fell from 2002 to 2007 (Table 6.6). 259. The technological capabilities indicator can be broken down into three categories ­ investment, production, and linkages. It permits useful comparisons of the technological capabilities across firms and enables econometric analysis of the influences on the acquisition of firm-level technological capabilities. (i) Investment technological capabilities are skills and information required before the investment is undertaken and those needed for carrying out the investment itself. These activities include project preparation, technology identification and transfer, and workforce training. The understanding gained by the operating firm of the basic technologies involved affect the efficiency with which it later operates its facilities. A crucial factor in the development of technological capabilities is the stock of skilled employees and additions to this stock by training, both in-house and externally. (ii) Production technological capabilities range from basic skills such as adoption, operation, and maintenance to more advanced knowledge such as adaptation, improvement, and equipment "stretching" to ultimately the highest and most demanding technical proficiencies of research, design, and innovation. This category covers both process and product technologies as well as the monitoring and control functions included under industrial engineering. Process technological capability includes quality control, maintenance, plant layout, inventory control, and improvements in equipment and processes. Product technological capability includes mastering product design and specifications, improving existing products, developing new products, and licensing product technology. The skills classified under this production category determine not only how efficiently and effectively given technologies are operated and improved, but 130 See Annex 6.1 for details on the construction of TCI. 147 also how successfully in-house efforts are utilized to absorb technologies bought or imitated from other firms. (iii) Linkage technological capabilities are the skills needed to exchange information, knowledge, and technology with component or raw material suppliers, buyers of output, subcontractors, consultants, service firms, and technology institutions. These linkages affect not only the productive efficiency of the firm, but also the diffusion of technology throughout the economy and the deepening of the industrial structure, both of which are essential to industrial development. Such extra-market linkages play a significant role in promoting productivity increase and building technological capability at the national level. 260. Decomposing TCI into the three main types of activities shows that linkage technological capability is the weakest part. Average Linkages TCI score has fallen from 0.295 in 2002 to 0.284 in 2007. As seen in Table 6.6 and Figure 6.13, over 70 percent of the samples do less than 4 of the 9 technical activities related to linkages (i.e., TCI scores less than 0.4). Conversely, only one percent of all manufacturing establishments have attained high levels of linkages technological capabilities (i.e., TCI scores over 0.8). As linkages are crucial in expanding the knowledge economy throughout Malaysia, policy should endeavor to facilitate the development of an economic and political environment that fosters cooperation, interaction, and interdependence between economic agents. 261. There is much room for improvement in the technological capabilities of Malaysia's manufacturing firms. As seen in Table 6.6, more than half the sample do less than 12 of the 30 activities that composed TCI (score of 0.4) where the median TCI score is only 0.346. While the average TCI score increased from 0.357 in 2002 to 0.372 in 2007, Figure 6.13 shows that the shift in the distribution from 2002 to 2007 has been marginal. Nearly a third of the PICS establishments continue to do less than 6 of the 30 technological capabilities activities (i.e., TCI scores between 0 and 0.2). Moreover, less than 5 percent of the manufacturing establishments have attained high levels of technological capabilities (i.e., TCI scores over 0.8). 148 Table 6.6: Descriptive Statistics of Manufacturing TCI 2002 2007 Median Mean Std. Dev. Median Mean Std. Dev. Overall TCI 0.346 0.357 0.218 0.346 0.372 0.226 Investment TCI 0.250 0.313 0.247 0.250 0.332 0.251 Production TCI 0.286 0.299 0.231 0.286 0.307 0.246 Linkages TCI 0.222 0.295 0.181 0.300 0.284 0.170 Frequency distribution (%) 0otoni SukiIBokri JoMrBoh", G ~~" KangwTebrou I : : ;,""" ~""" CD i ~ :::::::,'.-~ , _. PlCSIl .J o 1 IIong; M Sengkong Kg. .... 0 PlCS I AND Pies II CD Banting Sebe.CO'IQT.-.ir I Bob>g Betiunlai ,ON ) ."" · 1IoN8er..dom " Tepoh I". . . ., I Bukil~a KfOAH ® NATIONAL CAPITAL lIoN C""'" (Sebohogion) ,,~ 1'- ..... ,.'" .-.- STATE BOUNDARIES tl lIoN roga NORTHERN REGION: Ayerllam o I PoritRojo "",GOO.., ~B.,I,;iIBetvnIul>g - - REGION BOUNDARIES 14 8er""""'9 Bayon lepo> 1 Plenlong \./ - INTERNATIONAl. BOUNDARIES II , , ...,." 8'*,. Angkat n Beropil fZt .....,- .. Ponlion Kec~ Saleng KfLANTAN I 10 I ® Jin;ong Cyl-jayo .len""om 41» Kojong &. SQ. chua 6i) Bukinengoh O ~ 0 · Geo food ptocess.n9 Te>dileo " " REGION Kg. s.....go;P....d>oIo ® ® Woe. ","",", , " " ® Kg.CempoI.;o Kg.MeIoyu Kebun Ikongg 0 Moc:hor.gBuboi< Nibong lebal a...."" pIoSIics Rubbe.ond Mg T"" ..... , K...oIoSg.BuIuh K""hoi eo Si~~ SOmpong. 4 Bukir T""90h ,-, Compomng mach;,- a.3 a.m""" plastic. G . 0 5 " l.....bahJaya Sungc>iNo . ", Rubbe. and ,. MmjKl Tanoh ® Sungoo Sakop Auloparb G MeIako ® SoJngoi Duo "" wuuo.lUlloJlUllituou ...,. @ MeN ® Songoi Nibong Totol " ,. ® 0 G G 1'0>;" Pet>ombong. PeL.nSgBuluh Puc:hongBl.12 iM 1. 1 5...nggaOPuyu Sungo.seIuong ing a...,,,,,, . ..os, " " , " " PlCS II ®. '"'"""'" 4?' ::::::-- i ::::;;r.::: S1 Sl SQ. Monggi. SQ. Way Subong " " """'AOm Rubber ond pIosI;a Mod.i""'Y ond equ;pr-m '''' " " " ~ ,""",,''''- ~ """"'"~ C!) """"'''''' ~- AuIO porI> Woocl and lunoIure " " " " sa Tome.> Mesto 1II;o""" __ ~by~"""'~ IhI ~,,,"WMd __ ElecJ,-icow!ionce Compu~ng mod.inM " , 0 0 " VI AI Totliongf(\;t)g TeIoIobioo PHIl.IPPlNES - THAIlAND f."tai(.~ INDONESIA "'" ..c" -... ~ SOUTHERN REGION SOUTH I SEA CHINA IIRDJ6'U ~e- rN food processing Go<_. T ~1es ..... " n " 16 50 2"N " "'""""" Rohber and pb.t;a Machinery oM equipment " " " " .."""'"" s.:::~:'K } .J , , 50.""- " " " 2Q " EIeaOllic. \.~~ j ___ '''''_.i . A~_ 5 I Woodond ..... ilure " sbMile. Electricopplionce 31 0 " H,... ,-, ~mod., ..... 0 INDONESIA · '" 338 "'" FEBRUARY 2009 175 l000E 10S0E 110"E 11S0E PHIliPPINES SURVEYED TOWNS: THAILAND SABAH: CD Donggoogon SOUTH CHINA CD Kasigui SEA · Kola Kinabalu ~ Menggotal 5 Putatan 6 Inanam Rubber and plastics Machinery and equipment INDONESIA SARAWAK: Auto ports · Bandaroya Kuching Selatan wood and funi lure · Bandaroya Kuching Utara Total 24 CD Kola Padawan ®Sg. Moong Bazaar o· o· ® Balu Kawa Bazaar INDONESIA ® Bt. Kitang Bazaar @ Kota Sentosa l000E IOSOE J100E 11S0E ® lambir Bazaar SABAH ® Miri Townland REGION 5°N MALAYSIA .!, BRUNEI....._ SABAH INVESTMENT CLIMATE SURVEYS I \ CD SURVEYED TOWNS: PICSI I '/ '- :) .; "-.., . ;-' 0 --. "- o.,.."\'" "_ 0 _ 0 ""' · , CD PICSII " .j ' I o PICS I AND PICS II REGION BOUNDARIES ,I f' INTERNATIONAL BOUNDARIES ') SARAWAK , / . ,,: SARAWAK (\ PlCS I PlCS Il ,r- Food processing 24 30 REGION I .J" Garments Chemicals 2 0 3 2 \ '-> Rubber and plastics Electronics 3 5 1 -" wood and funiture 4 5 Total 34 46 % , - o- . r - ) .~ . - _ _ ....... / I '- .. ' / / ... - -.,/' . ...... ( . o 1 I 0 50 1 I 50 100 Kilometers 1 I 100 MiJes ~ .....'. _ ..,..J, r ' .-v~ · INDONESIA Thjs map we. produced by the Map Desjgn Unjt 01 The Wcxld Bonk. - ~ AI The boundaries, coJon,.denomjnolio'" ond any other jnfom,otion 'CJ () shown on thj. mop do not jmply, on the port of The world Bonk W ;3 Group, ony judgmenl 011 the legol sIolus 01 any !enllory, Of" ooy ~ 8 11 00 E 11 50 E endorsement Of" occeplonce of svch boundories. 00 ~ 00 176 tBRD 36819 "'" ,'t-,\ ,~ Iv\ALAYSIA SURVEYED TOWNS: _'_' ~'. - i PfRUS ..... "". ', " ) .... " - -.".J. . . ....... I THAILAND BUSINESS SUPPORTING SERVICES SURVEYS SlANEYEO T(MfNS; JOIiOR: ! ..... BaM./Kotogia l""'-/~ -~ ' ~,'""""' _KoociI ) ( CD PlCSI KUAl..I.LU/llll'UR: (:"'i O ~_ PICS. .~ .- ./ CD e PlCS I MI) PICS II ' "N KIDAH ,.~ . . . .,. ....... /3 MUAKA: lValD'or. Sungto; NiboRg ' \I.Tok>nt.! ........ ......., '\ L ":s I PlCS2 rifNGGANU ......, ....... ,w, sn"GOR , ~- ,- , 1Ivti....>~ " " " I r EAST COAST REGION "-- ... ~ ,........ · · --.,;- Sco>dot&.-us.:.l.* T ~& ~ _a.... ;,.w Sor; Sooio/Sg. Way~_"IIJaya/I(;j.s....g..;,,- ...... ~I<{,J,b>J~It!j"""""" PICS 1 ":$2 ~'N I ~~ ,""""-- , , " " ...-., ...... - p""'-'onoI_o ~"8 & --W;"II ,~ ..... ~...,. , n " " " " '00 -..... \ PAHANG SOUTH SEA CHINA ,oio< _ _ "--- ...... PICS I " " PlC52 ......... - "'=<~ ~" 25 5(11(;","-> A<'~ & ~ pooIo.-O--":_ o ,sAUWAK REGION / .... ! ! -.~_ ..I _ .......... _.i , , ,- , e...u-.~ ,. " '0 2S .loOMilM '" I N DO NE SIA """""'" IN DO NE SIA , ~ "'" """""" 177 l(l()OE loseE ll00E 115"E PHILIPPINES THAIlAND For detail. see IBRD 36183 Area 01 Map SOUTH CHINA SURVEYED TOWNS: SEA SABAH: O lnanam/Kola Kinobclu/Menggatol 7 7 CD Ken ingou 6 7 13 14 INDONESIA " SARAWAK i.."'; i SARAWAK: . , Bondoroyo Kuching Salotan/ Bondoroya Kuch ing Uklro J REGION ,.' , l " i ......, .,... ._ ..1.--- . ..",._. I ND ON ESIA 0" 0" I ND ON E SIA lOO"E 105"f .' IIO"E 115"E SABAH MALAYSIA CD REGION S· N BUSINESS SUPPORTING SARAH l BRUNEI,-. A \ SERVICES SURVEYS I \ SURVEYED TOWNS: \ I . 1' - - /·-.'·r'\..· ......... · _ ·-" . , - ·. ,l CD "/ CD ~es I ~es II ". / . I r · Pies I AND Pies II r ~.' ~ REGION BOUNDARIES \ INTERNATIONAL BOUNDARIES r ~ SARAWAK . f . ..,...· < SARAWAK f\ PICS I P1eS II / Information technology REGION Y " Accounting & related ( professional service 11 11 '-. Bu!>ineu logistics 6 ~ Total 18 18 '- / ", r ' - ' - -' . / ' - ) ' - -- _ . /j ........ ( o 50 100 Kilometen I I I "\. ·. . / .../' . r r 1 \ ...... ,., .- . ....... . "\. . _ ./ o 50 I OO~iles INDONESIA Thi. map was prodvcecJby.the Map Design Unit of The wond Bonk ~ ~ ~ .- The bovndorioM, roIoo.._ denominatioros"onrJ O"Y other ;nformot;o., 0 I shown "'" "'is mop do noI imply, 0<1 the port 01 The World Bonk W t-.> Group, coy i~ 00 the legal >Iotu. ofony _ritary, or any 0- § 110 0 E 1 15° E ~. or occeplonce of wen bovndotill$. ~ 178 Distribution of Manufacturing and Services Firms by Industry 4. PICS I covers 10 industries in the manufacturing sector, including food processing, textiles, garments, wood products, chemical & chemical products, rubber & plastics, machinery & equipment, electronics, auto parts, and furniture, across six regions in Malaysia. PICS II added two more industries, i.e., office, accounting & computing machines and electrical appliances. In this report, following the previous investment climate assessment, wood product industry and furniture industry are grouped into one. PICS I and II cover five industries in business support services sector, including information technology; communication services; accounting & related professional services; advertising & marketing; and business logistics Table A1.7 and Table A1.8 show the distribution of firms by industry and by region in PICS I & II for manufacturing and services sectors, respectively. Table A1.7: Distribution of Manufacturing Firms in PICS I & II PICS I Industry description Central North South East Coast Sabah Sarawak Total Food processing 62 31 72 8 10 24 207 Textiles 4 8 14 4 0 0 30 Garments 13 13 73 0 1 2 102 Chemicals 9 8 14 1 1 0 33 Rubber and plastics 97 56 88 2 3 3 249 Machinery and equipment 55 20 11 0 1 0 87 Electronics 19 36 19 0 0 1 75 Auto parts 24 4 5 2 3 0 38 Wood & Furnitutre 32 5 31 4 5 4 81 PICS II Industry description Central North South East Coast Sabah Sarawak Total Food processing 76 42 74 9 12 30 243 Textiles 8 9 18 5 0 0 40 Garments 17 15 53 0 2 3 90 Chemicals 34 19 21 5 1 2 82 Rubber and plastics 111 83 79 0 3 5 281 Machinery and equipment 51 20 21 0 1 0 93 Electronics 22 41 20 0 0 1 84 Auto parts 19 7 4 2 3 0 35 Wood and funiture 43 17 48 8 9 5 130 Electric appliance 15 8 4 0 0 0 27 Computing machines 1 5 4 0 0 0 10 179 Table A1.8: Distribution of Services Firms in PICS I & II PICS I Central North South Sabah Sarawak Total Information technology 29 0 0 0 0 29 Telecommunication 9 0 0 0 0 9 Advertising & marketing 69 7 4 7 11 98 Accounting & related business 17 0 1 0 0 18 Business logistics 60 14 9 6 7 96 PICS II Central North South Sabah Sarawak Total Information technology 37 1 0 0 1 39 Telecommunication 10 0 0 0 0 10 Advertising & marketing 82 9 8 7 11 117 Accounting & related business 26 0 0 0 0 26 Business logistics 57 25 16 7 6 111 5. PICS 2002 surveyed 902 firms in the manufacturing sector and 249 firms in selected business sector; PICS 2007 surveyed 1115 firms in manufacturing sector and 303 in selected business support services sector. 488 manufacturing firms and 137 services firms participated in both survey rounds, called panel firms. For the purpose of comparison, the two new industries surveyed in PICS II "office, accounting & computing machines" and "electrical appliances" are grouped with "machinery" and "electronics" into "machinery and equipment" and "electronics and appliances", respectively. 6. Table A1.9 and Table A1.10 show the slight difference in distribution of surveyed firm by industry, region, export orientation, ownership and firm size in PICS I, PICS II, panel and the pooled sample (PICS I and II) in manufacturing and services sectors, respectively. Chi-test results suggest that the differences in distribution between PICS I and PICS II and between panel firms and full samples are not significant by region, firm-size, ownership, and export-orientation for both manufacturing and services firms, but significant by industry only for manufacturing firms likely related to the new industry grouping. For the purpose of direct comparison, we follow the approach of the previous PICS report (World Bank, 2005) focusing on un-weighted analysis. 180 Table A1.9: Distribution of Manufacturing Firms by Industry, Region, Export Orientation, Ownership and Firm Size in the PICS I, PICS II, Panel Sample, and the Pooled Sample Manufacturing PICS I PICS II Panel Pooled PICS I PICS II Food processing 23.0 21.8 28.7 28.5 22.3 Textiles 3.3 3.6 3.1 3.3 3.5 Garments 11.3 8.1 10.3 9.8 9.5 Chemicals 3.7 7.4 4.1 5.1 5.7 Rubber and plastics 27.6 25.2 26.0 25.8 26.3 Machinery and equipment 9.7 9.2 8.4 9.0 9.4 Electronics and appliances 8.3 10.0 7.6 7.4 9.2 Auto parts 4.2 3.1 3.9 3.5 3.6 Wood and furniture 9.0 11.7 8.0 7.6 10.5 Total 100.0 100.0 100.0 100.0 100.0 Central 34.9 35.6 34.0 34.0 35.3 North 20.1 23.9 18.7 18.7 22.2 South 36.3 31.0 35.5 35.5 33.4 East Coast 2.3 2.6 3.1 3.1 2.5 Sabah 2.7 2.8 2.9 2.9 2.7 Sarawak 3.8 4.1 5.9 5.9 4.0 Total 100.0 100.0 100.0 100.0 100.0 Non-exporter 46.3 44.9 47.5 45.8 45.5 Exporter 53.7 55.1 52.5 54.2 54.5 Total 100.0 100.0 100.0 100.0 100.0 Purely domestically-owned 67.1 69.8 66.3 68.7 68.6 Less than 30% foreign-owned 5.5 4.5 6.0 3.8 5.0 More than 30% foreign-owned 27.4 25.7 27.7 27.6 26.5 Total 100.0 100.0 100.0 100.0 100.0 Small (employment<50) 48.3 45.1 45.6 43.1 46.5 Medium (50<=employment<150) 27.2 28.3 27.4 29.2 27.8 Large (employment >=150) 24.5 26.6 27.0 27.7 25.7 Total 100.0 100.0 100.0 100.0 100.0 181 Table A1.10: Distribution of Services Firms by Industry, Region, Export Orientation, Ownership and Firm Size in the PICS I, PICS II, Panel Sample, and the Pooled Sample Business support Service PICS PICS I II Panel Pooled PICS PICS I II Information technology 11.6 12.9 9.5 9.5 12.3 Telecommunications 3.6 3.3 2.2 4.4 3.4 Accounting & related business 39.2 38.6 47.4 47.4 38.9 Advertising & marketing 7.2 8.6 8.0 5.8 8.0 Business logistics 38.4 36.6 32.8 32.8 37.4 Total 100.0 100.0 100.0 100.0 100.0 Central 73.6 70.0 67.9 69.3 71.6 North 8.4 11.6 9.5 10.2 10.1 South 5.6 7.9 7.3 5.8 6.9 East Coast 0.0 0.0 0.0 0.0 0.0 Sabah 5.2 4.6 6.6 5.8 4.9 Sarawak 7.2 5.9 8.8 8.8 6.5 Total 100.0 100.0 100.0 100.0 100.0 Non-exporter 84.2 82.9 86.6 87.3 83.5 Exporter 15.8 17.1 13.4 12.7 16.5 Total 100.0 100.0 100.0 100.0 100.0 Purely domestically-owned 79.7 82.6 80.5 83.5 79.2 Less than 30% foreign-owned 7.5 4.0 5.3 3.8 4.2 More than 30% foreign-owned 12.9 13.4 14.3 12.8 16.6 Total 100.0 100.0 100.0 100.0 100.0 Small (employment<50) 51.8 57.4 50.0 47.0 56.1 Medium (50<=employment<150) 29.3 27.5 35.3 35.6 26.0 Large (employment >=150) 18.9 15.1 14.7 17.4 17.9 Total 100.0 100.0 100.0 100.0 100.0 Panel Samples 7. An additional objective of the PICS II is to build a panel by re-interviewing establishments included in PICS I. For surveys beyond the first iteration attrition becomes a major concern. This problem compounds the non-response bias present in most enterprise surveys and may seriously compromise sample sizes per industry/size stratum. Consequently, substitutions are made to maintain the original sample design and reach the target sample size per stratum. Substitutions by stratum are made in PICS II to reconstruct the original sample design of the survey at the first iteration, and in the case of Telecommunications and Advertising & Marketing, to expand the original coverage. Establishments included in PICS I were retained in the sample except for those that stopped operation. In this regard, 556 manufacturing and 182 103 business support services establishments from PICS I were included in the PICS II sample selection. (See Appendix Tables A1.11 and A1.12.) Table A1.11: Sample Size and Response Rate for Manufacturing Industries Successful Sample Selection Enumeration Response Rate Manufacturing Industries (Including Panel Panel Total Total Substitution) PICS I New PICS I New 1. Food Processing (15) 142 114 256 139 104 243 97.9 91.2 94.9 2. Textiles (17) 17 24 41 16 24 40 94.1 100.0 97.6 3. Garments (18) 59 47 106 48 42 90 81.4 89.4 84.9 4. Wood & Wood Products (20) 3 32 35 2 26 28 66.7 81.3 80.0 5. Chemical & Chemical Products (24) 34 59 93 25 57 82 73.5 96.6 88.2 6. Rubber & Plastics (25) 154 136 290 126 155 281 81.8 114.0 96.9 7. Machinery & Equipment (29) 53 60 113 44 49 93 83.0 81.7 82.3 8. Office, Accounting, & Computing Machinery (30) 0 11 11 0 10 10 .. 90.9 90.9 9. Electrical Machinery and Apparatus (31) 3 30 33 3 24 27 100.0 80.0 81.8 10. Electronics (Equipment & Components) (32) 36 45 81 33 51 84 91.7 113.3 103.7 11. Motor Vehicles and Parts (34) 15 23 38 17 18 35 113.3 78.3 92.1 12. Furniture (3) 40 63 103 35 67 102 87.5 106.3 99.0 Total 556 644 1,200 488 627 1,115 87.8 97.4 92.9 Notes: MSIC Divisions (2-digit codes) denoted in parentheses. Table A1.12: Sample Size and Response Rate for Business support Services Industries Business support Services Successful Industries Sample Selection Enumeration Response Rate (Including Panel Panel Total Total Substitution) PICS I New PICS I New 1. Information Technology 8 29 37 13 26 39 162.5 89.7 105.4 2. Telecommunications 1 4 5 6 4 10 600.0 100.0 200.0 3. Accounting & Related Services 56 72 128 65 52 117 116.1 72.2 91.4 4. Advertising & Marketing 5 20 25 8 18 26 160.0 90.0 104.0 5. Business Logistics (Transportation & Related 33 72 105 45 66 111 136.4 91.7 105.7 Services) Total 103 197 300 137 166 303 133.0 84.3 101.0 8. The panel firms appear on the both rounds of PICS. Table A1.13 and A1.14 show distribution of surveyed firms by industry, region, export orientation, ownership and firm size in PICS I, PICS II, and panel (PICS I and II). Some firms are regrouped into another region or industry, because their location or main product has changed since the PICS I. The distributions of firms by industry, region, export orientation, ownership and firm size differ slightly between samples. 183 Table A1.13: Distribution of Manufacturing Firms by Industry, Region, Export Orientation, Ownership and Firm Size in the PICS I, PICS II, and Panel Sample Manufacturing PICS I PICS II Panel 1. Food processing (15) 23.0 21.8 28.5 2. Textiles (17) 3.3 3.6 3.3 3. Garments (18) 11.3 8.1 9.8 4. Chemicals (24) 3.7 7.4 5.1 5. Rubber and plastics (25) 27.6 25.2 25.8 6. Machinery and equipment (29 & 30) 9.7 9.2 9.0 7. Electronics and appliances (31 & 32) 8.3 10.0 7.4 8. Auto parts (34) 4.2 3.1 3.5 9. Wood and furniture (20 & 36) 9.0 11.7 7.6 Total 100.0 100.0 100.0 1. Klang Valley 34.9 35.6 34.0 2. North 20.1 23.9 18.7 3. South 36.3 31.0 35.5 4. East Coast 2.3 2.6 3.1 5. Sabah 2.7 2.8 2.9 6. Sarawak 3.8 4.1 5.9 Total 100.0 100.0 100.0 1. Non-exporter 46.3 44.9 45.8 2. Exporter 53.7 55.1 54.2 Total 100.0 100.0 100.0 1. Purely domestically-owned 67.1 69.8 68.7 2. Less than 30% foreign-owned 5.5 4.5 3.8 3. More than 30% foreign-owned 27.4 25.7 27.6 Total 100.0 100.0 100.0 1. Small (employment<50) 48.3 45.1 43.1 2. Medium (50<=employment<150) 27.2 28.3 29.2 3. Large (employment >=150) 24.5 26.6 27.7 Total 100.0 100.0 100.0 184 Table A1.14: Distribution of Services Firms by Industry, Region, Export Orientation, Ownership and Firm Size in the PICS I, PICS II, and Panel Sample Business support Service PICS I PICS II Panel 1. Information technology 11.6 12.9 9.5 2. Telecommunications 3.6 3.3 4.4 3. Accounting & related business 39.2 38.6 47.4 4. Advertising & marketing 7.2 8.6 5.8 5. Business logistics (Transportation & Related Services) 38.4 36.6 32.8 Total 100.0 100.0 100 1. Klang Valley 73.6 70.0 69.3 2. North 8.4 11.6 10.2 3. South 5.6 7.9 5.8 4. Sabah 5.2 4.6 5.8 5. Sarawak 7.2 5.9 8.8 Total 100.0 100.0 100 1. Non-exporter 84.2 82.9 87.3 2. Exporter 15.8 17.1 12.7 Total 100.0 100.0 100 1. Purely domestically-owned 79.7 82.6 83.5 2. Less than 30% foreign-owned 7.5 4.0 3.8 3. More than 30% foreign-owned 12.9 13.4 12.8 Total 100.0 100.0 100.0 1. Small (employment<50) 51.8 57.4 47.0 2. Medium (50<=employment<150) 29.3 27.5 35.6 3. Large (employment >=150) 18.9 15.1 17.4 Total 100.0 100.0 100.0 9. Distributions of firms by region, ownership, export orientation, and firm size are consistent between PICS I and II, while it is slightly different by industry type in statistical term. In order to test if the distributions differ significantly across samples in a statistical term, we conduct chi-square test of independence. We perform the hypothesis test: H0: Distributions by each breakdown are not associated between PICS I and II. Ha: Distributions by each breakdown are associated between PICS I and II. 10. The Table A1.15 and A1.16 show the results of chi-square tests for distributions of manufacturing firms and services firms, respectively, surveyed by each breakdown between PICS I and II. The results indicate that firm distributions across industry are significantly different between two PICS in a statistical term, when we merge electrical appliances and computing machines into electronics and machinery industries, respectively. Distributions of firms across industry also differ significantly between PICS II and Panel PICS I, PICS II and 185 Panel PICS II, and PICS II. The distributions of firms do not, however, differ significantly by region, export orientation, ownership and firm size. 11. For services sector, the distribution of panel firms and that of firms in full samples of both survey rounds are similar also from the perspective of industries, as shown in Table A1.15. Table A1.15: Results of Chi-square Test of Independence for Distributions of Manufacturing Firms Industry Region PICS I PICS II PICS I PICS II 25.48 PICS II 7.71 PICS II 0.00*** 0.17 6.2 21.84 7.04 9.11 Panel 0.63 0.01*** Panel 0.22 0.11 Exporter Ownership PICS I PICS II PICS I PICS II 0.08 PICS II 2.17 PICS II 0.78 0.34 0.44 0.86 0.17 2.71 Panel 0.51 0.36 Panel 0.92 0.26 Firm size PICS I PICS II 2.15 PICS II 0.34 1.26 0.14 Panel 0.53 0.93 Note: Upper: chi-sq statistics. Lower: p-value. *** indicates rejecting the null hypothesis. 186 Table A1.16: Results of Chi-square Test of Independence for Distributions of Services Firms Industry Region PICS I PICS II PICS I PICS II 0.67 PICS II 3.1 PICS II 0.96 0.54 3.19 3.5 Panel 1.51 2.26 Panel 0.53 0.48 0.83 0.69 Exporter Ownership PICS I PICS II PICS I PICS II 0.16 PICS II 3 PICS II 0.69 0.22 0.38 0.92 Panel 0.76 0.42 Panel 0.54 0.34 0.68 0.81 Firm size PICS I PICS II 1.38 PICS II 0.5 1.78 2.91 Panel 0.41 0.23 Note: Upper: chi-sq statistics. Lower: p-value. *** indicates rejecting the null hypothesis. 187 Annex 1.4 Summary Statistics of Sample Firms 1. The PICS I and PICS II cover 902 and 1,115 manufacturing firms and 249 and 303 services firms, respectively. These firms are diverse in terms of geographical location, types of products/services, export orientation, ownership structure and employment size. We break down the locations of firms, types of business, sizes of firms, export orientation and ownership structure for the purpose of presenting the summary statistics and conducting regression analysis, based on the followings: · Industry ­ 9 industry codes that are constructed based on 4 digit ISIC for manufacturing firms (food processing, textile, garment, chemical, rubber & plastics, machinery, electronics & appliances, auto parts, and wood products & furniture) are used for summary statistics. In services, 5 service codes that are constructed based on the service number reported in the PICS (information technology, communication services, accounting & related services, advertising & marketing, and business logistics) are used. · Region ­ 6 region codes are used for summary statistics indicator each firms as its location of the manufacturing establishment (Klang Valley, North, South, East Coast, Sabah and Sarawak) and 5 region codes are used as its location of the services firms (Klang Valley, North, South, Sabah and Sarawak). · Firm size ­ A firm is categorized into 3 groups depending on their size or the number of employment. A firm is small-size if its total number of employment in 2006 is less than 50; medium-size if its number of current employment in 2006 is more than or equal to 50 and less than or equal to 200; and large-scale if the number of employment in 2006 is more than 200. · Exporter ­ A firm is considered an exporting firm if the firm exports more than 10 percent of its output directly or indirectly in 2006 as indicated in Part II ­ question 8.9. · Domestic firms ­ A firm is categorized into 3 groups depending on percent of foreign ownership as indicated in Part I ­ question 1.5. A firm is considered the purely domestically-owned firm if 0 percent was owned by a foreign private sector firm in 2006; less than 30 percent foreign-owned if more than 0 percent and less than and equal to 30 percent is owned; and more than 30 percent foreign- owned if more than 30 percent is owned. Industrial Coverage 2. A wide variety of industries is covered in the PICS. Table A1.17 and A1.18 detail distributions of manufacturing and services firms surveyed in PICS II by ISIC codes. PICS II added two industries: office, accounting and computing machinery and electrical appliances. Each of these industries is regrouped into machinery and electronics, respectively, given the limited number of coverage. Because of the broad definition of ISIC, services firms of the same ISIC code do not necessarily belong to the same service code. 188 State Distribution 3. Manufacturing firms are surveyed in nine states and services firms in seven states. Table A1.19 and A 1.20 detail distribution of manufacturing and services firms surveyed in PICS II. More than one-third of manufacturing firms and more than two-thirds of services firms are from Selangor, Kuala Lumpur and Melaka, while 7 percent of manufacturing and 11 percent of services firms are from Sabah and Sarawak. Ownership Structure 4. Twenty-Six percent of manufacturing firms and 16 percent of services firms are foreign fund management firms of which more than 30 percent are owned by the foreign private sector (Table A1.21 and A1.22). A share of foreign fund management firms is highest in the electronics and electrical appliance industry, lowest in the wood and furniture industry ­ 63 percent of electronics and electrical appliance producers are more than 30 percent foreign owned; 11 percent of wood product and furniture producers are. Direct foreign investment in services sector is less penetrating in the services sector. Foreign ownership is relatively common in the advertising industry, other transport agencies and software publishing. Export Orientation 5. Sales from export accounts for more than 10 percent of total sales in the majority of manufacturing firms (52 percent), while in less than one-fifth of services firms (17 percent) (Table A1.23 and Table A1.24). Electronic and electrical appliance producers are most likely to export their products ­ approximately four-fifths of electronic firms are exporting firms. The majority of firm in machinery, chemical, rubber and plastics, and, wood product and furniture industries are classified as exporters. Export orientation is relatively low in the auto parts (26 percent) and food processing (39 percent) industries. Most of businesses supporting industries, except the business logistic transport sector, are less likely to export sales more than 10 percent of total sales. Firm Size 6. Twenty-seven percent of manufacturing firms surveyed are classified as large and 16 percent of services firms are as well (Table A1.25 and Table A1.26). Electronics and electrical appliance firms are most likely to be large (58 percent), while food processing firms are least likely to be large (17 percent). In the services sector, telecommunication firms are most likely to be classified as large in terms of employment. 189 Table A1.17: Distribution of Manufacturing Firms by ISIC Codes Industry MISIC De scription PIC S I PIC S II 150 1 Production, processing and preservation of meat, fish, fruit, 151 vegetables, oils and fats 54 84 Food Proce ssing 152 Dairy products 6 5 Grain mill products, starches and starch products, and prepared animal 153 feeds 19 24 154 Other food products 127 130 171 Spinning, weaving and finishing of textiles 29 31 Te xtile s 172 Other textiles 7 173 Knitted and crocheted fabrics and articles 1 2 Garme nts 181 Wearing apparel, except fur apparel 102 90 201 Sawmilling and planing of wood 2 18 W ood products 202 Products of wood, cork, straw and plaiting materials 2 10 and furniture 361 Funiture 76 102 369 n.e.c. 1 241 Basic chemicals 14 44 C he micals 242 Other chemical products 19 38 251 Rubber products 61 98 Rubbe r and 252 Plastic products 186 183 plastics 254 1 255 1 291 General-purpose machinery 26 38 Machine ry and 292 Special-purpose machinery 53 52 e quipme nts 293 Domestic appliances n.e.c. 8 3 300 Office, accounting and computing machinery 10 311 Electric motors, generators and transformers 7 312 Electricity distribution and control apparatus 6 313 Insulated wire and cable 10 314 Accumulators, primary cells and primary batteries 2 Ele ctronics and e le ctrical 319 Other electrical equipment n.e.c. 2 appliance s 321 Electronic valves and tubes and other electronic components 40 58 Eelevision and radio transmitters and apparatus for line telephony 322 and line telegraphy 8 7 T elevision and radio receivers, sound or video recording or 323 reproducing apparatus, and associated goods 27 19 340 2 341 Motor vehicles 6 4 Auto parts Bodies (coachwork) for motor vehicles; manufacture of trailers and 342 semi-trailers 12 13 343 Parts and accessories for motor vehicles and their engines 18 18 Total 902 1,115 190 Table A1.18: Distribution of Services Firms by 4 Digit ISIC Codes, PICS I PICS I SID Industry description IT Telecommunication Accounting Advertising Business logistics Total 4510 Construction site preparation 1 1 6010 Railways 1 1 6021 Other scheduled passenger land transport 4 4 6023 Freight transport by road 34 34 6110 Sea & coastal water transport 12 12 6112 1 1 6120 Inland water transport 0 6210 Scheduled air transport 1 1 6220 Nonscheduled air transport 3 3 6301 Cargo handling 6 6 6302 Storage & warehousing 1 1 6303 Other supporting transport activities 4 4 6309 Activities of other transport agencies 21 21 6412 Courier services 4 4 6420 T elecommunications 10 1 11 7112 Renting of water transport equipment 1 1 Renting of agricultural machinery and 7121 equipment 2 2 7123 Renting of office machinery & equipment 1 1 7210 Hardware consultancy 5 5 7220 Software publishing, consultancy & supply 15 2 1 18 7230 Data processing 2 2 Data activities & on-line distribution of 7240 electronic content 1 1 Maintenance & repair o f office, accounting 7250 & computing machine 1 1 Accounting, book-keeping & auditing 7412 activities; tax consultancy 4 17 21 7413 Market research & public opinion polling 0 Business & management consultancy 7414 activities 0 Architectural & engineering activities & 7421 related technical consultancy 72 72 7422 T echnical testing & analysis 1 1 7430 Advertising 1 15 16 8322 4 4 Total 28 12 97 16 96 249 191 Table A1.18: Distribution of Services Firms by 4 Digit ISIC Codes (cont'd), PICS II PICS II SID Industry de scription IT Te le communicationAccounting Adve rtising Busine ss logistics Total 4510 Construction site preparation 0 6010 Railways 1 1 6021 Other scheduled passenger land transport 0 6023 Freight transport by road 46 46 6110 Sea & coastal water transport 13 13 6112 0 6120 Inland water transport 1 1 6210 Scheduled air transport 1 1 6220 Nonscheduled air transport 2 2 6301 Cargo handling 8 8 6302 Storage & warehousing 2 2 6303 Other supporting transport activities 2 2 6309 Activities of other transport agencies 33 33 6412 Courier services 2 2 6420 T elecommunications 7 7 7112 Renting of water transport equipment 0 Renting of agricultural machinery and 7121 equipment 0 7123 Renting of office machinery & equipment 0 7210 Hardware consultancy 8 1 9 7220 Software publishing, consultancy & supply 23 2 25 7230 Data processing 0 Data activities & on-line distribution of 7240 electronic content 0 Maintenance & repair o f office, accounting 7250 & computing machine 1 1 Accounting, book-keeping & auditing 7412 activities; tax consultancy 7 19 26 7413 Market research & public opinion polling 1 1 Business & management consultancy 7414 activities 4 2 6 Architectural & engineering activities & 7421 related technical consultancy 91 91 7422 T echnical testing & analysis 0 7430 Advertising 3 23 26 8322 0 Total 39 10 117 26 111 303 192 Table A1.19: Distribution of Manufacturing Firms by State, PICS II Industry ISIC De scription Klang Valle y North Panang Kuala (Pulau Se labor Lumpur Me laka Pinang) Kedah Production, processing and preservation of meat, fish, fruit, 151 vegetables, oils and fats 14 13 2 14 1 152 Dairy products 2 0 0 1 0 Food Proce ssing Grain mill products, starches and starch products, and prepared 153 animal feeds 2 1 1 7 1 154 Other food products 18 14 9 15 3 171 Spinning, weaving and finishing of textiles 5 2 0 5 2 Textile s 172 Other textiles 1 0 0 0 2 173 Knitted and crocheted fabrics and articles 0 0 0 0 0 Garme nts 181 Wearing apparel, except fur apparel 4 12 1 13 2 201 Sawmilling and planing of wood 2 1 1 2 1 Wood products 202 Products of wood, cork, straw and plaiting materials 3 0 1 2 2 and furniture 361 Funiture 24 2 9 5 5 241 Basic chemicals 15 2 2 5 2 Che micals 242 Other chemical products 14 1 0 10 2 Rubbe r and 251 Rubber products 27 10 7 17 12 plastics 252 Plastic products 50 11 6 33 21 291 General-purpose machinery 21 1 1 5 0 Machine ry and 292 Special-purpose machinery 16 11 0 12 2 e quipme nts 293 Domestic appliances n.e.c. 1 0 0 0 1 300 Office, accounting and computing machinery 1 0 0 3 2 311 Electric motors, generators and transformers 4 0 0 2 0 312 Electricity distribution and control apparatus 2 1 0 2 0 313 Insulated wire and cable 4 2 0 0 2 314 Accumulators, primary cells and primary batteries 1 0 0 1 0 Ele ctronics and e lectrical 319 Other electrical equipment n.e.c. 1 0 0 1 0 appliances 321 Electronic valves and tubes and other electronic components 14 2 1 24 3 Eelevision and radio transmitters and apparatus for line 322 telephony and line telegraphy 1 0 0 0 2 T elevision and radio receivers, sound or video recording or 323 reproducing apparatus, and associated goods 4 0 0 4 8 341 Motor vehicles 2 0 0 0 1 Bodies (coachwork) for motor vehicles; manufacture of trailers Auto parts 342 and semi-trailers 2 2 0 2 0 343 Parts and accessories for motor vehicles and their engines 11 2 0 4 0 Total 266 90 41 189 77 193 Table A1.19: Distribution of Manufacturing Firms by State (cont'd), PICS II Industry ISIC De scription South East Coast Sabah Sarawak Johor Te re nggan Sabah Sarawak Production, processing and preservation of meat, fish, fruit, 151 vegetables, oils and fats 26 4 4 6 152 Dairy products 0 0 1 1 Food Proce ssing Grain mill products, starches and starch products, and prepared 153 animal feeds 6 0 1 5 154 Other food products 42 5 6 18 171 Spinning, weaving and finishing of textiles 12 5 0 0 Te xtile s 172 Other textiles 4 0 0 0 173 Knitted and crocheted fabrics and articles 2 0 0 0 Garme nts 181 Wearing apparel, except fur apparel 53 0 2 3 201 Sawmilling and planing of wood 5 3 1 2 Wood products 202 Products of wood, cork, straw and plaiting materials 2 0 0 0 and furniture 361 Funiture 41 5 8 3 241 Basic chemicals 11 5 0 2 C he micals 242 Other chemical products 10 0 1 0 Rubbe r and 251 Rubber products 23 0 2 0 plastics 252 Plastic products 56 0 1 5 291 General-purpose machinery 9 0 1 0 Machine ry and 292 Special-purpose machinery 11 0 0 0 e quipme nts 293 Domestic appliances n.e.c. 1 0 0 0 300 Office, accounting and computing machinery 4 0 0 0 311 Electric motors, generators and transformers 1 0 0 0 312 Electricity distribution and control apparatus 1 0 0 0 313 Insulated wire and cable 2 0 0 0 314 Accumulators, primary cells and primary batteries 0 0 0 0 Ele ctronics and e le ctrical 319 Other electrical equipment n.e.c. 0 0 0 0 appliance s 321 Electronic valves and tubes and other electronic components 13 0 0 1 Eelevision and radio transmitters and apparatus for line 322 telephony and line telegraphy 4 0 0 0 T elevision and radio receivers, sound or video recording or 323 reproducing apparatus, and associated goods 3 0 0 0 341 Motor vehicles 0 0 1 0 Bodies (coachwork) for motor vehicles; manufacture of trailers Auto parts 342 and semi-trailers 4 1 2 0 343 Parts and accessories for motor vehicles and their engines 0 1 0 0 Total 346 29 31 46 194 Table A1.20: Distribution of Services Firms by State, PICS II Klang Valle y North South Sabah Sarawak Panang Kuala (Pulau SID Industry de scription Se lagor Lumpur Me laka Pinang) Johor Sabah Sarawak 6010 Railways 0 1 0 0 0 0 0 6023 Freight transport by road 9 8 0 15 10 3 1 6110 Sea & coastal water transport 5 3 0 1 1 2 1 6120 Inland water transport 0 0 0 0 1 0 0 6210 Scheduled air transport 0 0 0 1 0 0 0 6220 Nonscheduled air transport 0 1 0 0 0 1 0 6301 Cargo handling 2 0 0 3 1 1 1 6302 Storage & warehousing 2 0 0 0 0 0 0 6303 Other supporting transport activities 1 0 0 1 0 0 0 6309 Activities of other transport agencies 24 0 0 4 2 0 3 6412 Courier services 0 1 0 0 1 0 0 6420 T elecommunications 2 5 0 0 0 0 0 7210 Hardware consultancy 7 1 0 0 0 0 1 7220 Software publishing, consultancy & supply 8 16 0 1 0 0 0 Maintenance & repair o f office, accounting 7250 & computing machine 1 0 0 0 0 0 0 Accounting, book-keeping & auditing 7412 activities; tax consultancy 2 19 0 1 2 0 2 7413 Market research & public opinion polling 0 1 0 0 0 0 0 Business & management consultancy 7414 activities 2 4 0 0 0 0 0 Architectural & engineering activities & 7421 related technical consultancy 25 35 1 8 6 7 9 7430 Advertising 10 16 0 0 0 0 0 Total 100 111 1 35 24 14 18 195 Table A1.21: Distribution of Manufacturing Firms by Ownership (PICS II) Less than More than Pure ly 30% 30% Industry MISIC De scription domestical fore ign foreign ly owne d owned owne d Production, processing and preservation of meat, fish, fruit, 151 vegetables, oils and fats 68 2 11 152 Dairy products 2 1 2 Food Processing Grain mill products, starches and starch products, and prepared animal 153 feeds 18 2 4 154 Other food products 111 1 16 171 Spinning, weaving and finishing of textiles 24 2 5 Textiles 172 Other textiles 7 0 0 173 Knitted and crocheted fabrics and articles 2 0 0 Garme nts 181 Wearing apparel, except fur apparel 74 3 12 201 Sawmilling and planing of wood 15 0 3 Wood products 202 Products of wood, cork, straw and plaiting materials 7 1 2 and furniture 361 Funiture 87 5 9 241 Basic chemicals 25 4 15 Chemicals 242 Other chemical products 21 2 15 Rubbe r and 251 Rubber products 54 4 38 plastics 252 Plastic products 126 10 45 291 General-purpose machinery 22 4 12 Machine ry and 292 Special-purpose machinery 42 0 10 e quipments 293 Domestic appliances n.e.c. 1 1 1 300 Office, accounting and computing machinery 2 0 6 311 Electric motors, generators and transformers 2 0 5 312 Electricity distribution and control apparatus 3 0 3 313 Insulated wire and cable 6 1 3 314 Accumulators, primary cells and primary batteries 0 0 2 Ele ctronics and e lectrical 319 Other electrical equipment n.e.c. 0 1 1 appliance s 321 Electronic valves and tubes and other electronic components 17 1 36 Eelevision and radio transmitters and apparatus for line telephony 322 and line telegraphy 1 1 4 T elevision and radio receivers, sound or video recording or 323 reproducing apparatus, and associated goods 5 0 12 341 Motor vehicles 2 1 1 Bodies (coachwork) for motor vehicles; manufacture of trailers and Auto parts 342 semi-trailers 11 0 2 343 Parts and accessories for motor vehicles and their engines 9 2 6 Total 764 49 281 196 Table A1.22: Distribution of Services Firms by Ownership, PICS II Pure ly Le ss than More than dome stically 30% fore ign 30% fore ign SID Industry de scription owne d owne d owne d 6010 Railways 1 0 0 6023 Freight transport by road 40 1 5 6110 Sea & coastal water transport 9 4 0 6120 Inland water transport 0 0 1 6210 Scheduled air transport 1 0 0 6220 Nonscheduled air transport 2 0 0 6301 Cargo handling 6 0 2 6302 Storage & warehousing 1 0 1 6303 Other supporting transport activities 2 0 0 6309 Activities of other transport agencies 22 0 11 6412 Courier services 2 0 0 6420 T elecommunications 5 1 1 7210 Hardware consultancy 6 0 3 7220 Software publishing, consultancy & supply 17 1 7 Maintenance & repair o f office, accounting & 7250 computing machine 1 0 0 Accounting, book-keeping & auditing activities; 7412 tax consultancy 24 0 2 7413 Market research & public opinion polling 0 0 1 7414 Business & management consultancy activities 5 1 0 Architectural & engineering activities & related 7421 technical consultancy 88 1 2 7430 Advertising 14 1 11 Total 246 10 47 197 Table A1.23: Distribution of Manufacturing Firms by Export Orientation (PICS II) Exporte r (>10% Industry MISIC De scription None xporte r sales) Production, processing and preservation of meat, fish, fruit, 151 vegetables, oils and fats 44 40 152 Dairy products 1 4 Food Proce ssing Grain mill products, starches and starch products, and prepared animal 153 feeds 19 4 154 Other food products 83 46 171 Spinning, weaving and finishing of textiles 16 14 Te xtile s 172 Other textiles 3 4 173 Knitted and crocheted fabrics and articles 1 1 Garme nts 181 Wearing apparel, except fur apparel 53 37 201 Sawmilling and planing of wood 11 7 Wood products 202 Products of wood, cork, straw and plaiting materials 7 3 and furniture 361 Funiture 45 56 241 Basic chemicals 18 26 C he micals 242 Other chemical products 17 20 Rubbe r and 251 Rubber products 23 73 plastics 252 Plastic products 98 84 291 General-purpose machinery 16 22 Machine ry and 292 Special-purpose machinery 25 27 e quipme nts 293 Domestic appliances n.e.c. 0 3 300 Office, accounting and computing machinery 1 7 311 Electric motors, generators and transformers 2 5 312 Electricity distribution and control apparatus 2 4 313 Insulated wire and cable 4 6 314 Accumulators, primary cells and primary batteries 0 2 Ele ctronics and e le ctrical 319 Other electrical equipment n.e.c. 0 2 appliance s 321 Electronic valves and tubes and other electronic components 10 45 Eelevision and radio transmitters and apparatus for line telephony 322 and line telegraphy 2 5 T elevision and radio receivers, sound or video recording or 323 reproducing apparatus, and associated goods 3 15 341 Motor vehicles 4 0 Bodies (coachwork) for motor vehicles; manufacture of trailers and Auto parts 342 semi-trailers 10 2 343 Parts and accessories for motor vehicles and their engines 11 7 Total 529 571 198 Table A1.24: Distribution of Services Firms by Export Orientation (PICS II) Exporter (>10% SID Industry de scription Nonexporter sales) 6010 Railways 1 0 6023 Freight transport by road 38 7 6110 Sea & coastal water transport 8 5 6120 Inland water transport 1 0 6210 Scheduled air transport 1 0 6220 Nonscheduled air transport 1 1 6301 Cargo handling 7 1 6302 Storage & warehousing 1 1 6303 Other supporting transport activities 1 1 6309 Activities of other transport agencies 19 12 6412 Courier services 1 1 6420 T elecommunications 6 1 7210 Hardware consultancy 6 3 7220 Software publishing, consultancy & supply 18 7 Maintenance & repair o f office, accounting & 7250 computing machine 1 0 Accounting, book-keeping & auditing activities; 7412 tax consultancy 25 1 7413 Market research & public opinion polling 0 1 7414 Business & management consultancy activities 5 0 Architectural & engineering activities & related 7421 technical consultancy 85 5 7430 Advertising 22 4 Total 247 51 199 Table A1.25: Distribution of Manufacturing Firms by Size (PICS II) Small Medium (>=50 Large Industry MISIC Description (<50) & <150) (>=150) Production, processing and preservation of meat, fish, fruit, 151 vegetables, oils and fats 37 31 16 152 Dairy products 1 1 3 Food Processing Grain mill products, starches and starch products, and prepared 153 animal feeds 11 13 0 154 Other food products 84 23 22 171 Spinning, weaving and finishing of textiles 17 6 7 Textiles 172 Other textiles 5 1 1 173 Knitted and crocheted fabrics and articles 1 1 0 Garments 181 Wearing apparel, except fur apparel 53 15 22 201 Sawmilling and planing of wood 9 6 3 Wood products and 202 Products of wood, cork, straw and plaiting materials 5 4 1 furniture 361 Funiture 50 27 25 241 Basic chemicals 13 15 15 Chemicals 242 Other chemical products 15 14 9 251 Rubber products 29 42 25 Rubber and plastics 252 Plastic products 81 58 43 291 General-purpose machinery 19 8 11 Machinery and 292 Special-purpose machinery 33 13 6 equipments 293 Domestic appliances n.e.c. 1 0 2 300 Office, accounting and computing machinery 0 0 9 311 Electric motors, generators and transformers 2 3 2 312 Electricity distribution and control apparatus 2 3 1 313 Insulated wire and cable 4 2 4 314 Accumulators, primary cells and primary batteries 0 0 2 Electronics and electrical 319 Other electrical equipment n.e.c. 0 2 0 appliances 321 Electronic valves and tubes and other electronic components 11 10 36 Eelevision and radio transmitters and apparatus for line telephony 322 and line telegraphy 0 3 4 Television and radio receivers, sound or video recording or 323 reproducing apparatus, and associated goods 1 3 14 341 Motor vehicles 0 2 2 Bodies (coachwork) for motor vehicles; manufacture of trailers and Auto parts 342 semi-trailers 9 3 0 343 Parts and accessories for motor vehicles and their engines 5 4 9 Total 498 313 294 200 Table A1.26: Distribution of Services Firms by Size (PICS II) Me dium (>=50 Large SID Industry de scription Small (<50) & <150) (>=150) 6010 Railways 0 0 1 6023 Freight transport by road 28 11 7 6110 Sea & coastal water transport 7 3 3 6120 Inland water transport 0 1 0 6210 Scheduled air transport 1 0 0 6220 Nonscheduled air transport 0 1 1 6301 Cargo handling 2 3 3 6302 Storage & warehousing 0 1 1 6303 Other supporting transport activities 0 0 2 6309 Activities of other transport agencies 20 10 3 6412 Courier services 2 0 0 6420 T elecommunications 1 0 6 7210 Hardware consultancy 5 2 2 7220 Software publishing, consultancy & supply 11 8 6 Maintenance & repair o f office, accounting & 7250 computing machine 0 0 1 Accounting, book-keeping & auditing activities; 7412 tax consultancy 13 9 4 7413 Market research & public opinion polling 1 0 0 7414 Business & management consultancy activities 4 1 1 Architectural & engineering activities & related 7421 technical consultancy 67 21 3 7430 Advertising 17 6 3 Total 179 77 47 201 Annex 1.5 Subjective and Objective Indicators 1. PICS covers a wide variety of questions business manger's perception and objective indicators of Malaysia investment climate. Followings are sources of information presented in chapter 1. Codes for questions are consistent between manufacturing PICS and services PICS, if not exceptionally mentioned. Perceptions of The Business Climate By The Firms · Severity of investment climate constraints (closed question) ­ We calculate the number (percentage) of firms that respond 3 or 4 (major or very severe) to each of 1-18 investment climate constraint in questions of Part I ­ question 5.1. We divide this number of firms by a common denominator to obtain the percentage of firms consider them a major or severe obstacle. The common denominator is the maximum number of non-missing responses among items 1-15 for manufacturing and items 1-19 for services. · Top three business constraints (open question) ­ We obtain the list of the three biggest obstacles perceived by Malaysian firms surveyed in Part I ­ question 5.2. We divide the number of firms that name an IC obstacle out of 22 items for 1st, 2nd, and 3rd biggest obstacles to doing business by the total number of Malaysian firms surveyed. Objective Investment Climate Indicators In presenting objective indicators, we systematically applied the outlier criteria of three standard deviations to eliminate extreme observations from open-ended questions. If the IC indicators reported by a firm is greater than the mean in the corresponding industry by more than three standard deviations or smaller than the mean by more than three standard deviations, the observation is considered as an outlier and dropped from the calculation of average score. Infrastructure · Yearly number of power outages and average hours of the duration: The average number of power outages or surges from the public grid per month and its average duration (hours) in 2006 are calculated from Part I ­ question 6.14(b). We calculate the yearly number of power outage by multiplying the average monthly number by 12 (months) after dropping missing values. · Days to obtain electricity connection: Days required obtaining fixed telephone line in 2005-2006 is obtained from Part I ­ question 6.6(2). · Percentage of firms with own or shared generator in Part I ­ question 6.16: A firm is considered to have a generator, if the firm owns, rents or has shared access to a generator in 2006. We calculate the percent of firms with generator by dividing firms with generator by the total number of firms after dropping missing values. 202 · Production lost due to power outage in Part I ­ question 6.15: Percentage of production value lost due to power interruptions from the public, e.g., losses dues to lost production time from the outage, time needed to reset machines and production rejected due to processes being interrupted, grid in 2006. · Yearly number of fixed phone interruptions and average hours of the duration: The average number of fixed phone interruptions per month and its average duration (hours) in 2006 are obtained from Part I ­ question 6.14(c). We calculate the yearly number of fixed phone interruptions by multiplying the average monthly number by 12 (months) after dropping missing values. · Days to obtain fixed phone line: Days required obtaining fixed telephone line in 2005-2006 is obtained from Part I ­ question 6.6(1). · Yearly number of insufficient water supply: The average number of insufficient water supply per month is obtained from Part I ­ question 6.14(b). We calculate the yearly number of by multiplying the average monthly number by 12 (months) after dropping missing values. · Days to obtain water connection: Days required obtaining water connection in 2005-2006 is obtained from Part I ­ question 6.6(3). · Yearly number of transport disruption: The average number of transport disruption per month is obtained from Part I ­ question 6.14(d). We calculate the yearly number of by multiplying the average monthly number by 12 (months) after dropping missing values. · Transportation cost as percentage of export earnings: Transport cost as percent of the value of export earnings is obtained from 7.3.y.c. of manufacturing PICS. · Production lost while in transit in part I ­ question 6.13 of both manufacturing services PICS: Estimated lost shipment due to breakage, theft or spoilage as a percent of total sales in 2006. Access to Finance · Percentage of firms with overdraft facility in Part IIA ­ question 9.18: This indicates firms with a bank overdraft facility. We calculate a percent of firm with an overdraft facility by dividing it by the total number of firms surveyed after dropping missing values in 2006. · Percentage of firms with bank loan in Part IIA ­ question 9.15: This indicates a percent of firms with a term loan from a bank or financial institution in 2006. We calculate a percent of firm with a bank loan by dividing it by the total number of firms surveyed after dropping missing values. 203 · Collateral as percent of the loan value in Part IIA ­ question 9.29: Average percentage of the value of collateral required for the loan was calculated after dropping missing values. Regulatory Framework · Total number of licenses applied in the past two years and average weeks to obtain a license from the federal government agency/state government agency in Part I ­ question 6.2: The total number of licenses was calculated by the sum of the numbers of licenses applied to the federal government, e.g., MITI, MIDA, JAKIM, LPKP; the state government, e.g., land office, district office; and local authority. · Days to obtain approval for construction/import permit/operating licenses in Part I ­ question 6.6: They imply average days required to obtaining approval for construction/import permit/operating license in 2005-2006. · Days to clear export/import custom in Part I ­ question 7.3.y.b and 7.4.y.a. of manufacturing PICS and 7.6.y.a of services PICS138. This implies the average number of days required to clearing Malaysian customs for exporting any part of output/importing equipment and other inputs. · Percentage of firms using consultant for licenses/permits/approvals in Part I ­ question 6.4: This indicates a percent of firms that used agents, consultants, or one or more employees to help the firms to process licenses/permits/approvals. · Total number of days spent with regulation and cost spent in dealing with regulation of the business as percent of sales in Part I ­ question 6.7: This is the sum of the number of days and total cost spent in contact with agency dealing with regulations of the business, e.g., tax inspectorates (IRB), labor and social security, fire & rescue department/department of occupational health and local authority, etc. Total cost spent in dealing with these agencies is divided by total sales. · Number of inspections per year in Part I ­ question 6.8. This indicates that the number of times inspectors from agencies listed above visit the firm's establishment in 2006. Tax Rate and Tax Administration · Total tax expenses as percentage of total sales/corporate tax as percentage of profit/sales tax as percentage of total sales ­ The total tax expenses are the sum of corporate tax, import duties on materials and components, import duties on capital goods, sales tax, excise tax, local authority tax and other direct tax paid by the establishment in 2006. The total tax expense is divided by the firm's revenue; 138 Average number of days to clear Malaysian export custom is available only in the manufacturing PICS. 204 The corporate tax expenses is divided by the firm's profit; and the total sales tax payment is divided by total sales to make them comparable between firms. Tax payment data is obtained from Part IIA ­ question 9.4. · Total number of days spent with tax regulation (IRB) in Part I ­ question 6.7: This is the sum of the number of days spent in contact with agency dealing with regulations of tax inspectorates (IRB) Labor Market · Percentage of employees with secondary school degree/college degree in Part IIB ­ question X.17. We obtain the number of employees with university degree for a firm and calculate the percentage of employees with university degree by dividing it by the total number of employees indicated in Part IIB ­ question 10.13a. · Average number of weeks to fill vacancy for professionals and skilled production workers in Part I ­ question 4.10. This implies that weeks required to filling the most recent vacancy for a professional, e.g., engineers, scientists and university graduate, and skilled production workers, e.g., skilled technicians. The average weeks are calculated for all respondents. Innovation/Technological Capabilities · Percentage of computer-controlled machinery in Part I ­ question 3.7: It indicates the percentage of a firm's production machines that is computer- controlled139. · Vintage of capital is the machinery & equipment of the firm that is less than five years old in Part I. - question 3.6: It indicates the percentage of a firm's plant machinery and equipment that is of age younger than 5 years old. · Percentage of firms that received government incentives in Part I ­ question 3.21 of manufacturing PICS and 3.27 of services PICS: It indicates that the percentage of firms that received any government incentives to conduct technological innovation and R&D. Crime and Theft · Losses from theft, robbery or vandalism and estimated cost of providing security in Part I ­ question 6.11a: The losses from theft, robbery or vandalism as a percent of total sales vandalism in 2006. Estimated costs as a percent of its total sales of providing security for the establishment indicated. 139 Available in the manufacturing PICS only. 205 ANNEX 2 Annex 2.1 Firm Performance Indicators of Manufacturing Firms Labor Productivity (Value-added per worker) and Total Factor Productivity 1. In identifying how different dimensions of the business climate affect firm performance, we used two different methods to measure firm performance derived from the PICS survey in 2002 and 2007: Value added per employee and Total Factor Productivity. The variables used in calculating firm performance indicators, we use following production variables from the 2007 PICS survey. · Output (y) for years 2004-2006 are given by total sales in the table of Part IIA ­ question 9.1: It is deflated by corresponding Production Price Index from the Department of statics. · Intermediate costs (m) for years 2004-2006 are defined as the sum of direct material cost, electricity expenditures, and fuel and other energy expenditures in the table of Part IIA ­ question 9.13. These costs are obtained from question 9.13 and deflated by Production Price Index from the Department of Statistics. · Skilled labor (s) for years 2004-2006 are the sum of the number of management, professionals, skilled production workers and nonproduction workers in the table of Part IIB ­ question 10.8 for manufacturing. · Unskilled labor (u) for years 2004-2006 are given as the number of unskilled workers in the table of Part IIB -question 10.8. · Capital stock (k) is defined as the book value of machinery and equipment in the Part IIA ­ question 9.14 deflated by Production Price Index from the Department of Statistics. 2. Two measures of firm performance indicators ­ Labor Productivity (VAL) and total factor productivity (TFP) are included which capture different dimensions of firm performance. · VAL measures a productivity of labor. It is the ratio of value-added defined as output minus intermediate cost to total employment. Mathematically, it is defined as: y it - mit VALit = sit + u it where i and t stand for firm and time respectively; y it is output measured as operating revenue for firm i at a given time t ; mit is material input plus energy cost (electricity and fuel); and sit u it is the sum of skilled and unskilled labors or the total number of labor. 206 · TFP measures multi-factor production that represents the efficiency of the firms in transforming a set of inputs into output. TFP captures how a firm converts factors of production, i.e., capital and labor, into output. TFP reflect the level of technology that a firm holds, marginal quality of products and government policies, etc. In estimating production function, we use Levinsohn and Petrin's method to use intermediates as a proxy. The Levinsohn Petrin's metho yield more unbiased estimators to correct firm's decision of input adjustment in response to productivity shock. Outlier Criteria 3. In calculating the summary statistics of these firm performance indicators, we identify outlying observations, i.e. establishment, with respect to output-labor ratio, capital-labor ratio, materials-labor ratio, and labor share in revenue, and material share in revenue. If any of these ratios or shares is larger than its mean value in the corresponding 2-digit industry by more than 3 standard deviations or if the ratio/share is smaller than its mean value by more than 3 standard deviations, the observations are considered to be outliers. For the material (input) revenue shares, two additional rules are applied: (i) all firms with labor or materials revenue shares larger than 1 are classified as outliers and (ii) all firms with materials shares smaller than 0.1 are classified as outliers. Applying all outlier rules, the sample size reduced from 1,115 to 1,009. Full Sample vs. Panel Sample 4. Measures of firm performance, i.e., labor productivity and TFP, obtained from the panel sample and from the full sample are not statistically significantly different. Table A2.1 and A2.2 present mean labor productivity and TFP obtained from the panel sample and the full sample by industry and by region in 2004-2006. We test the null hypothesis and alternate hypothesis: H0: mean values from the full sample and from the panel sample are equal. Ha: mean values are not equal. Results of T-test to see if the mean values are statistically different are also shown. In all breakdowns, t statistics fail to reject the null hypothesis, meaning that these values are not significantly different in a statistical term. 207 Table A2.1: Firm Level Mean Labor Productivity in Malaysia (2004-2006) ­ Common Samples vs. Full Samples 2006 Common sample Full sample Student's T -T est mean mean p-value Malaysia 166,932 183,032 0.57 Central 149,612 217,373 0.22 North 163,132 138,016 0.56 South 206,093 188,235 0.73 East Coast 73,561 285,680 0.27 Sabah 197,419 129,672 0.59 Sarawak 84,036 80,839 0.92 mean mean p-value Food processing 285,424 256,467 0.66 T extiles 77,011 85,940 0.76 Garments 37,153 36,078 0.83 Chemicals 396,734 492,897 0.55 Rubber & plastics 115,467 188,776 0.31 Machinery 133,421 124,900 0.78 Electronics & appliances 93,197 108,241 0.59 Auto parts 161,810 155,493 0.92 Wood & furniture 55,332 70,131 0.35 2005 Common sample Full sample Student's T -T est mean mean p-value Malaysia 165,747 169,333 0.89 Central 167,514 213,011 0.36 North 134,864 128,354 0.86 South 196,062 158,121 0.43 East Coast 142,276 296,932 0.40 Sabah 213,353 130,683 0.47 Sarawak 64,233 62,110 0.93 mean mean p-value Food processing 249,797 243,610 0.94 T extiles 69,201 79,870 0.67 Garments 39,713 37,591 0.65 Chemicals 407,371 409,545 0.99 Rubber & plastics 128,230 162,464 0.42 Machinery 122,647 119,494 0.90 Electronics & appliances 123,764 128,232 0.89 Auto parts 180,768 155,362 0.68 Wood & furniture 105,440 83,395 0.52 208 Table A2.1: Firm Level Mean Labor Productivity in Malaysia (2004-2006) ­ Common Samples vs. Full Samples (cont'd) 2004 Common sample Full sample Student's T -T est mean mean p-value Malaysia 166,932 183,032 0.57 Central 149,612 217,373 0.22 North 163,132 138,016 0.56 South 206,093 188,235 0.73 East Coast 73,561 285,680 0.27 Sabah 197,419 129,672 0.59 Sarawak 84,036 80,839 0.92 mean mean p-value Food processing 285,424 256,467 0.66 T extiles 77,011 85,940 0.76 Garments 37,153 36,078 0.83 Chemicals 396,734 492,897 0.55 Rubber & plastics 115,467 188,776 0.31 Machinery 133,421 124,900 0.78 Electronics & appliances 93,197 108,241 0.59 Auto parts 161,810 155,493 0.92 Wood & furniture 55,332 70,131 0.35 209 Table A2. 2: Firm Level Mean Total Factor Productivity in Malaysia (2004-2006) ­ Common Samples vs. Full Samples 2006 Common sample Full sample Student's T -T est mean mean p-value Malaysia 5.22 5.21 0.89 Central 5.54 5.52 0.97 North 5.74 5.10 0.33 South 5.13 5.30 0.66 East Coast 3.43 3.91 . Sabah 3.37 4.43 0.30 Sarawak 3.16 3.62 0.39 mean mean p-value Food processing 2.93 2.96 0.52 T extiles 2.67 2.64 0.90 Garments 10.59 10.32 0.26 Chemicals 12.06 12.24 0.51 Rubber & plastics 2.10 2.13 0.52 Machinery 12.80 12.61 0.17 Electronics & appliances 6.38 6.34 0.46 Auto parts 4.93 4.89 0.72 Wood & furniture 4.21 4.19 0.79 2005 Common sample Full sample Student's T -T est mean mean p-value Malaysia 5.20 5.26 0.81 Central 5.49 5.46 0.94 North 5.55 5.17 0.52 South 5.27 5.50 0.64 East Coast 3.73 3.84 0.84 Sabah 3.22 4.42 0.24 Sarawak 3.19 3.61 0.43 mean mean p-value Food processing 2.96 2.98 0.74 T extiles 2.74 2.60 0.52 Garments 10.75 10.47 0.27 Chemicals 12.03 12.18 0.60 Rubber & plastics 2.12 2.16 0.40 Machinery 12.59 12.34 0.41 Electronics & appliances 6.40 6.33 0.57 Auto parts 4.89 4.85 0.80 Wood & furniture 4.25 4.18 0.41 210 Table A2.2: Firm Level Mean Total Factor Productivity in Malaysia (2004-2006) ­ Common Samples vs. Full Samples (cont'd) 2004 Common sample Full sample Student's T -T est mean mean p-value Malaysia 5.10 5.23 0.63 Central 5.55 5.49 0.89 North 5.68 5.21 0.44 South 4.90 5.31 0.37 East Coast 3.67 3.89 0.70 Sabah 3.38 4.55 0.23 Sarawak 3.15 3.62 0.40 mean mean p-value Food processing 2.95 2.96 0.81 T extiles 2.61 2.56 0.83 Garments 10.41 10.34 0.78 Chemicals 12.15 12.17 0.94 Rubber & plastics 2.18 2.19 0.94 Machinery 12.72 12.24 0.13 Electronics & appliances 6.43 6.28 0.20 Auto parts 5.02 4.90 0.38 Wood & furniture 4.21 4.23 0.75 211 Annex 2.2 Production Function Estimation with Levinsohn-Petrim Method 1. Total factor productivity is measured as the residual from the estimation of a log-linear four factor Cobb-Douglas production function with Hicks-neutral technical change in logarithmic form. A simultaneity problem, however, arises when there is contemporaneous correlation between the factors of production and the errors, often thought as Hicks neutral productivity shocks. The firm, for example, may observe productivity shocks early enough to allow for a change in factor input decisions. For example, firms that are more productive will hire more labor and use more materials in order to produce more. As pointed out by Griliches and Mareisse (1998), profit-maximizing firms immediately adjust their inputs each time a productivity shock is observed, resulting in input levels correlated with in the regression. This simultaneity violates the OLS conditions for unbiased and consistent estimation. 2. In the context of the Cobb-Douglas production function, the error term is therefore assumed to be additively separable in two distinct components. Formally, the production function of firm i in MSIC 2-digit manufacturing industry j (=15, 17, 18, 24, 25, 29, 32, 34, 36) at time t for the regression analysis is assumed to have the following form: qitj = kitj + mitj + sitj + uitj + itj [2] where output q is produced combining the capital stock k, raw materials m, skilled labor s, and unskilled labor u. is residual firm productivity and can be decomposed as follows: itj = itj + itj [3] where is a component of firm productivity that is known to the firm manager and possibly affects input choices but is unknown to the econometrician and is a random shock to output/productivity that is realized after input choices are made (therefore it is not correlated with input choices.140 3. Olley and Pakes (1996) and Levinsohn and Petrin (2003) have developed two similar semi-parametric estimation procedures to overcome the simultaneity problem when estimating production functions. Olley and Pakes include the investment decision of the firm in the estimation equation to proxy for unobserved productivity shocks. Derived from a structural model of the optimizing firm, the proxy controls for the part of the error correlated with inputs, , by "annihilating" any variation that is possibly related to the productivity term. The method suggested by Olley and Pakes, however, generates consistent and unbiased estimates if and only if there is a strictly monotonous relationship between the proxy and output. Consequently, firms that make only intermittent investments will have their zero-investment observations truncated from the estimation routine because the monotonicity condition does not hold for these observations. For the PICS samples, this is a large portion of the data. 4. Given the considerable attrition in the PICS sample when using the Olley and Pakes approach, the paper adopts the method developed by Levinsohn and Petrin (2003) to estimate 140 More specifically, the following equalities are verified for the conditional expected values: E[it/sit]=0, E[it/ uit]=0, E[it /mit]=0, and E[it /kit]=0. 212 production functions. Levinsohn and Petrin offer an estimation technique that is very close in spirit to the Olley and Pakes approach but uses intermediate inputs in lieu of investment as a proxy for unobserved productivity shocks. Nearly all firms in the PICS data almost always report positive material costs. Therefore, the Levinsohn-Petrin intermediate input proxy estimator is the optimal choice for the PICS samples. 5. Follow Levinsohn and Petrin (2003) to obtain production function parameters that correct for the endogeneity of input choices, the basic specification is given by equations 2 and 3. The main idea is that firms choose their variable inputs (i.e., skilled labor, unskilled labor and materials) with a knowledge of their productivity , whereas capital is a quasi-fixed input that is costly to adjust. The demand for variable inputs by a firm can be derived from profit maximization and depends on capital k and on productivity . In particular, the demand for m = m(it , kit ) .141, 142 Under fairly general technical conditions (described materials is given by it in detail in Levinsohn and Petrin, 2003), it is possible to invert the materials demand function and express the unobserved (to the econometrician) firm productivity as a function of two variables (materials and capital) that are observable: it = ( mit , kit ) . 6. The crucial idea behind this estimation method is to use this proxy for productivity to control for the endogeneity of input choices with respect to productivity. The estimation proceeds in two stages. Stage 1: 7. Replacing the proxy equation for productivity into equation 3 and this equation itself into equation 2, we obtain: qit = mit + kit + sit + uit + (mit , kit ) + it [4] or qit = sit + uit + h(mit , kit ) + it [5] where h(mit , kit ) = mit + uit + (mit , kit ) . 8. The function h(.) groups all the terms on materials and capital. Its functional form is unknown but we can approximate it either (i) using a polynomial in materials and capital or (ii) using locally weighted least squares as follows: 141 For simplicity of notation, we drop the superscript j in the rest of the section, but the estimation is performed for each industry j separately. 142 Ideally, the function m(.) should vary across years (i.e., mt (it , kit ) to account, for example, for changes in the price of materials, other inputs, or output. This would be a way (though an imperfect one) to account for these prices for which data are not available. However, in practical terms, given the small number of observations available for each individual year and industry in the Malaysia PICS datasets, we choose to consider a function m (.) that does not vary across years. 213 (i) Adding the terms of a 4th degree polynomial in m and k to substitute for h(.) in equation 5, it can be estimated by OLS to provide consistent parameter estimates for skilled and unskilled labor. (ii) Taking the expectation of both sides of equation 5, conditional on m and k gives: E [ qit / mit , kit ] = E [ sit / mit , kit ] + E [uit / mit , kit ] + E [ h(mit , kit ) / mit , kit ] + E [ it / mit , kit ] [6] Now E [ h( mit , kit ) / mit , kit ] = h(mit , kit ) and since it is uncorrelated with any of the inputs E [ it / mit , kit ] = 0 . So equation 6 can be rewritten as: E [ qit / mit , kit ] = E [ sit / mit , kit ] + E [uit / mit , kit ] + h( mit , kit ) [7] Taking the difference between Equation (D4) and Equation (D5'), one obtains: ( qit - E [ qit / mit , kit ]) = ( sit - E [ sit / mit , kit ]) + (uit - E [uit / mit , kit ]) + it [8] 9. This equation with "transformed regressors" can be estimated by OLS (with no constant term) if one obtains estimates for the conditional expected values. These conditional expected values are approximated by quadratic locally weighted least squares regressions of yit, si,t, and uit on (mit, kit).143 It should be noted that one also obtains an estimate for the unknown function h(mit, kit) that is used in Stage 2. Stage 2: 10. An additional assumption is required to obtain the coefficients on materials and on capital. We assume that productivity is serially correlated: it follows a Markov process it = E [it / it -1 ] + it . Also, the main identification assumption in Stage 2 is that capital is slow to adjust: it may adjust to the expected part of productivity conditional on lagged productivity E [it / it -1 ] , but it does not adjust to the unexpected shock it . So, one can derive the following moment condition that identifies the coefficient on capital144: E [ it + it / kit ] = 0 [9] 11. A separate moment condition is needed to identify the coefficient on materials. A condition for materials exactly parallel to that for capital cannot be used since materials are 143 The quadratic locally weighted least squares procedure is described in Levinsohn and Petrin (2002). Basically, weighted least squares estimation is used to construct predictions of yit, sit, and uit given (mit, kit) including as regressors the terms from a second order polynomial in (mit, kit). For any data point (mit*, kit*), for which one needs an estimate of the three conditional expected values, the regression gives greater weights to the observations closest to the point (mit*, kit*). A consistent estimator for the conditional expected value is the intercept from this locally weighted regression. 144 E[it+it/kit]=E[it /kit]+E[it /kit]. The main identification assumption says that E[it/kit]=0. The definition of it implies that E[it /kit]=0. So E[it+it /kit]=0. 214 correlated with productivity (both the expected and unexpected parts): [ it E + it / mit ] 0 .145 But one can derive a moment condition for lagged materials. Materials in year t-1 are chosen by the firm without knowledge of the productivity shocks realized only in year t, therefore E [ it + it / mit -1 ] = 0 [10] The residual it + it in the moment conditions is obtained replacing productivity it by its Markov process and switching sides for some of the terms in equation 2: it + it = q it - ^ s it - u it - m it - k it - E[ it / it -1 ] ^ ^ [11] 12. It should be noted that in equation 10 we replaced the coefficients on skilled and unskilled labor by their estimated values from Stage 1 and we included an estimate for the expected value of productivity conditional on lagged productivity. This conditional expected g ( it-1 ) value is a function of it -1 . We call it but its functional form is unknown. We approximate this g(.) function by locally weighted least squares of an estimate of it on an estimate of it -1 .146 13. Sample analogs for the two moment conditions in equations 9 and 10 are obtained for all firms and a general method of moments (GMM) criterion function is constructed. The estimates for the coefficients on materials and capital are those making the sample analogs of the moment conditions as close to 0 as possible:147 T T Min , ([( it + it ) * kit ]) + ([( it + it ) * mit -1 ]) 2 2 [11] i t =2 i t =2 An iterative procedure needs to be used to minimize this function starting from some candidate initial parameters and (e.g, those obtained by OLS). Outlier Criteria Used in Estimating Production Functions 14. We apply slightly different outlier criteria to estimate production function from that used in calculating summary statistics. We define an observation (i.e., establishment) to be an outlier for variable X, if its value is larger than the mean of X in the corresponding 2-digit MSIC industry by more than 3 inter-quartile deviations of X (in the industry) or if its value is smaller 145 Although E[it /mit] is 0, E[it /kit] is different from 0. 146 Specifically, the estimate for the expected productivity conditional on lagged productivity is given by an LWLS regression of an estimate for it given by (it + it ) = yit - sit - uit - * mit - * kit on an estimate for ^ ^ ^ it-1 given by it -1 = h (mit -1 , kit -1 ) - * mit -1 - * kit -1 . Note that both of these estimates depend on the ^ ^ parameters of interest in Stage 2: and . 147 The sample analogs for the moment conditions are summed across firms i (the index i in the first summation symbol). For each firm the moment conditions are summed from the second year of available data on since the procedure uses lagged inputs (the index t in the second summation symbol). No moment condition can be computed in the first year of available data for a firm. 215 than the industry mean by more than 3 standard deviations. The variables for which this outlier rule is applied are: output-labor ratio, capital-labor ratio, materials-labor ratio, labor share in revenue, materials share in revenue. The definition of each of the variables entering these ratios and of the shares is provided immediately below. Outliers are identified for the ratios and shares in each of the 6 sample years 1999, 2000, 2001, 2004, 2005, and 2006 when data permit. For the input revenue shares, two additional rules are applied: (i) all firms with labor or materials revenue shares larger than 1 are classified as outliers; and (ii) all firms with materials shares smaller than 0.1 are classified as outliers. These outliers are dropped from the calculation of industry means and standard deviations required to apply the three standard deviation rule above. 216 Annex 2.3 Production Function Estimations Table A2.3: Levinsohn-Petrin Production Function Estimation (full sample) Electronics Auto Woods & Food Rubber & Textiles Garments Chemical Machinery & electrical parts furniture processing plastics appliances Skilled labor -0.02 0.07 0.32 0.16 0.12 0.30 0.14 0.20 0.16 (in log) (0.094) (0.062) (0.115)*** (0.034)*** (0.024)*** (0.077)*** (0.270)*** (0.046)*** (0.0507)*** Unskilled -0.02 0.03 0.04 0.17 0.08 0.01 0.09 0.00 0.08 labor (in log) (0.076) (0.051) (0.077) (0.041)*** (0.022)*** (0.043) (0.022)*** (0.037) (0.047)* Intermediates 0.86 0.23 0.56 0.63 0.78 0.12 0.78 0.01 0.49 (in log) (0.264)*** (0.242) (0.249)*** (0.169)*** (0.277)*** (0.228) (0.172)*** (0.333) (0.267)* Capital (in 0.04 0.09 0.11 0.07 0.03 0.11 0.08 0.16 0.12 log) (0.147) (0.113) (0.136) (.071) (0.057) (0.129) (0.070) (0.086) (0.106) Observations 108 236 113 375 685 167 266 298 Wald test of constant returns to scale chi2 0.48 6.33 0.02 0.07 0 7.85 0.74 4.48 0.5 p-value 0.4896 0.0119 0.877 0.795 0.9463 0.0051 0.39 0.344 0.479 Source: Staff calculations based on Malaysia PICS 2002 and 2007. Note: Standard errors in parentheses. *, **, and *** indicate significant at 10 percent 5 percent and 1 percent, respectively. Table A2.4: Generalized Least Square Production Function Estimation (full sample) Auto Electronics Woods & Food Rubber Textiles Garments parts Chemical Machinery & electrical furniture processing & plastics appliances Skilled labor -0.016 -0.022 0.098 0.098 0.09 0.237 0.084 0.236*** 0.09 (in log) (0.059) (0.065) (0.06) (0.06) (0.028)*** (0.075)*** (0.03)*** (0.071) (0.057) Unskilled 0.009 0.01 -0.04 -0.04 0.048 -0.011 0.038 -0.029 0.058 labor (in log) (0.093) (0.047) (0.061) (0.061) (0.017) *** (0.038) (0.024) (0.056) (0.047) Intermediate 0.802 0.833 (0.83 0.828 0.766 0.662 0.769 0.760*** 0.753 s (in log) (0.085)** (0.045)** (0.046)** (0.043)** (0.076)*** (0.022) *** (0.051)*** (0.091) (0.048)*** * * * * Capital (in 0.129 0.099 0.06 0.064 0.094 0.11 0.082 0.039 0.105 log) (0.039)** (0.026)** (0.028)*** (0.031)** (0.031)** (0.015) *** (0.038)*** (0.043) (0.048)** * * Observations 110 265 124 124 829 187 985 295 319 Source: Staff calculations based on Malaysia PICS 2002 and 2007. Note: Standard errors in parentheses. *, **, and *** indicate significant at 10 percent 5 percent and 1 percent, respectively. 217 Annex 2.4 Correlates Between Firm Characteristics and Performance/IC Indicators 1. Firm characteristics are significantly correlated with each other. Plant age is negatively correlated with exporter dummy, foreign ownership and percentage of computer controlled production machines. This suggests that older firms tend to be less export-oriented, foreign owned, and endowed with computer controlled machines. Size of a firm is positively correlated with export dummy, foreign ownership, percentage of computer controlled machines, and R&D activity ­ larger firms are more likely to be exporting, foreign owned, installed with computer- controlled machines and engaging in R&D activities. Export dummy, foreign ownership dummy, percentage of computer controlled production machines and R&D activity are positively correlated with each other. Table A2.5: Correlation between Firm Characteristics Foreign Exporter Owned Computer R&D Dummy Dummy Controlled Dummy Plant Age Firm Size (>10%) (>30%) Machine (>0%) Plant Age 1 Firm Size 0.0167 1 Exporter Dummy -0.0720* 0.3937* 1 Foreign Owned Dummy -0.0355* 0.3290* 0.3131* 1 Computer Controlled Machine -0.0699* 0.2857* 0.1659* 0.2417* 1 R&D Dummy -0.029 0.2561* 0.1634* 0.1003* 0.0931* 1 Note: * indicates significance at 10 percent confidence level. These correlations are calculated for the sample without outlying firms. 2. To identify how firm characteristics interact with firm performance, we conduct fixed effect regressions with robust standard error. The model tested in 2.3.1 is formally specified as the following: where is a measure of firm performance indicator (value-added per worker or TFP) of a firm, i, at time, t, and, explanatory variables include plant age ( ); firm size measured by the number of labor employed in the establishment ( ); a dummy variable for a foreign fund management firm ( = 1 if foreign ownership is more than and equal to 30 percent); a dummy for an exporting firm ( = 1 if export of sales is more than and equal to 10 percent); and, two other variables of technology innovation of a firm: percent of computer controlled production machines ( ), and a dummy for R&D activity ( = 1 if expense on R&D activities is more than 0 percent). We include matrices of industries dummies (Food processing as a reference industry), region dummies (Klang Valley as a reference region) and time dummies (2004 as a reference year) to capture industry-, region-, and time- specific effects. Parameters represent coefficients estimating effects of firm characteristics and represent vectors of parameters for dummy matrices. We estimate the 218 effects using Ordinary Least Square (OLS) with robust standard errors to address heteroskedasticity. Comments on Regression Results 3. The results from regression analysis are very informative and intuitive. But it should be reiterated that the statistical significance of these firm characteristics and IC indicators does not necessarily show a causality of these indicators. For example, firms with foreign ownership tend to show higher levels of labor productivity and total factor productivity. This may reflect a fact that firms with foreign ownership tend to be innovative as a result of technology transfer from foreign stakeholders. But it is also true that the best performing domestic firms are very attractive for foreign investors, resulting in higher share of foreign ownership. Thus, the findings from regression analysis on relationship between IC indicators and firm performance do not necessarily indicate a causal relationship. 4. Analysis of IC indicators is subject to a potential econometric problem of multicolinearity ­ some of explaining variables are strongly correlated among IC indicators. We observe significant correlation among some of these indicators. In order to identify the problem of multicolinearity, we complement analysis of the regressions of firm performance and IC indicators with partial regressions of IC indicators of each IC category. The partial regressions tend to show stronger and more significant impacts of these IC indicators given multicolinearity. Correlates between IC Indicators and Performance 5. In 2.3.2, we explored the interaction between IC indicators and firm performance. To address a non-respondent in PICS and a reverse causality issue, we used region-industry mean of IC indicators instead of firm level data. The mean indicators are obtained by: where Xik and nj denotes an IC indicator reported by a firm, i, and the number of firms in an industry, j, of region, k, that provided a valid response to an item of the questionnaire. 6. Using the industry mean value, we identified IC indicators that are associated with firm performance indicators. To select potentially significant IC indicators included, we systematically conduct stepwise regressions within each category of investment climate and across categories in the following way: · Based on the business managers' perception and availability of objective IC indicators, four major areas of major IC constraints were identified: skills and education of available worker, regulatory framework, crime and theft and infrastructure; · As a preliminary analysis, we conduct forward selection/backward elimination stepwise regression to identify potentially significant IC indicators within each major IC area; and 219 · We repeat the operation to identify robust IC indicators that are associated with firm productivity. 7. After these operations, four IC indicators remain significant and are used in the full model in Chapter 2. The formal specification of the full model can be written as follows: where denotes average share of employees with university degree to the total employees in industry j, of region, k, average number of days to obtain fixed phone connection, average production loss (as percentage of sales) due to theft, robbery or vandalism, and average senior manager's time dealing with agencies to comply with regulations. The regression model also includes the firm characteristics mentioned above, except two measures of technology. Again, we tested the model with the fixed-effect OLS with robust standard error to address heteroskedasticity, based on data obtained from the PICS II. Correlates between IC Indicators and Performance Using the Panel Data 8. In section 2.3.3, we explored how changes in IC indicators in 2003-2007 are associated with measures of firm performance. Changes in IC indicators are calculated by: Changes in IC indicators are obtained by subtracting a region-industry mean of an IC indicator in PICS I from that in PICS II. The fixed-effect OLS with robust standard error is conducted for the panel sample of 488 firms without outlying firms. 220 Annex 2.5 Correlates between Firm Characteristics and Performance Table A2.6: Correlates between Firm Characteristics and Performance (PICS 2007, Full Sample) Firm Characteristics and Labor Productivity ­ Full Sample Dependent Varialbe = (Value added per worker in log) #1 #2 #3 Firm Characteristics Age (in log) 0.027 0.004 0.01 (0.033) (0.033) (0.033) Firms size (in log) -0.003 -0.014 -0.015 (0.018) (0.018) (0.018) Foreign management firm (=1 if more than 30% ownred by foreign firm) 0.267 0.291 0.291 (0.055)*** (0.054)*** (0.054)*** Exporter (=1 if more than 10% sales exported) 0.202 0.175 0.175 (0.044)*** (0.044)*** (0.044)*** % of computer controlled machines 0.003 0.002 0.002 (0.001)*** (0.001)*** (0.001)*** R&D dummy (=1 if any R&D expenditure) 0.224 0.218 0.218 (0.055)*** (0.055)*** (0.055)*** Dummies Inudstry dummies yes+++ yes+++ yes+++ Region dummies no yes+++ yes+++ Year dummies no no yes Constant 10.907 11.264 11.243 (0.105)*** (0.114)*** (0.116)*** Observations 2953 2953 2953 Adjusted R-squared 0.18 0.2 0.21 Source: Staff calculations based on Malaysia PICS 2002 and 2007. Note: Food processing industry, Klang Valley, and 1999 are respectively used as reference industry, region, and year for fixed-effect dummies. Robust standard errors in parentheses. * significant at 10 percent; ** significant at 5 percent; *** significant at 1 percent. + jointly significant at 10 percent; ++ significant at 5 percent; +++ significant at 1 percent. 221 Firm Characteristics and Total Factor Productivity ­ Full Sample Dependent Varialbe = (TFP) #1 #2 #3 Firm Characteristics Age (in log) 0.035 0.025 0.026 (0.022) (0.022) (0.022) Firms size (in log) 0.071 0.07 0.07 (0.023)*** (0.023)*** (0.023)*** Foreign management firm (=1 if more than 30% 0.124 0.135 0.135 ownred by foreign firm) (0.042)*** (0.041)*** (0.041)*** Exporter (=1 if more than 10% sales exported) 0.045 0.059 0.059 (0.028) (0.030)** (0.030)** % of computer controlled machines 0.001 0.001 0.001 (0.001)* (0.001)* (0.001)* R&D dummy (=1 if any R&D expenditure) 0.108 0.093 0.093 (0.048)** (0.047)** (0.047)** Dummies Industry dummies yes+++ yes+++ yes+++ Region dummies no yes yes Year Dummies no no yes Constant 2.525 2.609 2.612 (0.092)*** (0.104)*** (0.106)*** Observations 1886 1886 1886 Adjusted R-squared 0.97 0.97 0.97 Source: Staff calculations based on Malaysia PICS 2002 and 2007. Note: Food processing industry, Klang Valley, and 1999 are respectively used as reference industry, region, and year for fixed-effect dummies. Robust standard errors in parentheses. * significant at 10 percent; ** significant at 5 percent; *** significant at 1 percent. + jointly significant at 10 percent; ++ significant at 5 percent; +++ significant at 1 percent. 222 Table A2.7: Correlates between Firm Characteristics and Performance (Panel Sample) Firm Characteristics and Labor Productivity ­ Panel Sample Dependent Varialbe = (Value added per worker in log) #1 #2 #3 Firm Characteristics Age (in log) -0.026 -0.064 -0.136 (0.038) (0.039) (0.042)*** Firms size (in log) -0.013 -0.037 -0.054 (0.024) (0.024) (0.024)** Foreign management firm (=1 if more than 30% ownred by foreign firm) 0.24 0.272 0.27 (0.070)*** (0.071)*** (0.070)*** Exporter (=1 if more than 10% sales exported) 0.127 0.096 0.07 (0.057)** (0.059) (0.057) % of computer controlled machines 0.003 0.003 0.003 (0.001)*** (0.001)*** (0.001)*** R&D dummy (=1 if any R&D expenditure) 0.018 0.005 0.276 (0.048) (0.048) (0.062)*** Dummies Inudstry dummies yes+++ yes+++ yes+++ Region dummies no yes+++ yes+++ Year dummies no no yes Constant 11.022 11.471 11.333 (0.138)*** (0.161)*** (0.163)*** Observations 2463 2463 2463 Adjusted R-squared 0.1 0.12 0.14 Source: Staff calculations based on Malaysia PICS 2002 and 2007. Note: Food processing industry, Klang Valley, and 1999 are respectively used as reference industry, region, and year for fixed-effect dummies. Robust standard errors in parentheses. * significant at 10 percent; ** significant at 5 percent; *** significant at 1 percent. + jointly significant at 10 percent; ++ significant at 5 percent; +++ significant at 1 percent. 223 Firm Characteristics and Total Factor Productivity ­ Panel Sample Dependent Varialbe = (TFP) #1 #2 #3 Firm Characteristics Age (in log) 0.027 0.035 0.033 (0.048) (0.048) (0.05) Firms size (in log) 0.744 0.733 0.733 (0.024)*** (0.024)*** (0.024)*** Foreign management firm (=1 if more than 30% ownred by foreign firm) 0.181 0.193 0.193 (0.057)*** (0.058)*** (0.058)*** Exporter (=1 if more than 10% sales exported) 0.216 0.146 0.146 (0.059)*** (0.060)** (0.060)** % of computer controlled machines 0.004 0.003 0.003 (0.001)*** (0.001)*** (0.001)*** R&D dummy (=1 if any R&D expenditure) 0.028 0.058 0.062 (0.055) (0.055) (0.067) Dummies Industry dummies yes+++ yes+++ yes+++ Region dummies no yes+++ yes+++ Year Dummies no no yes Constant 8.855 9.09 9.105 (0.171)*** (0.171)*** (0.176)*** Observations 1882 1882 1882 Adjusted R-squared 0.95 0.95 0.95 Source: Staff calculations based on Malaysia PICS 2002 and 2007. Note: Food processing industry, Klang Valley, and 1999 are respectively used as reference industry, region, and year for fixed-effect dummies. Robust standard errors in parentheses. * significant at 10 percent; ** significant at 5 percent; *** significant at 1 percent. + jointly significant at 10 percent; ++ significant at 5 percent; +++ significant at 1 percent. 224 Annex 2.6 Correlates between Investment Climate and Firm Performance Table A2.8: Correlates between IC Indicators and Labor Productivity (PICS 2007) Dependent Varialbe = (Value added per worker in log) #1 #2 #3 IC Indicators % of university graduate 0.011 0.01 0.01 (0.001)*** (0.001)*** (0.001)*** Average number of days to obtain fixed phone connection -0.004 -0.008 -0.008 (0.003) (0.004)** (0.004)** Loss from crime (as % of sales) -0.065 -0.053 -0.054 (0.015)*** (0.019)*** (0.019)*** Senior managers' time in dealing with regulations (%) -0.004 -0.005 -0.005 (0.005) (0.006) (0.006) Firm Characteristics Age (in log) 0.038 0.02 0.026 (0.032) (0.033) (0.033) Firms size (in log) -0.009 -0.021 -0.021 (0.018) (0.018) (0.018) Foreign management firm (=1 if more than 30% ownred by foreign firm) 0.246 0.263 0.263 (0.053)*** (0.052)*** (0.052)*** Exporter (=1 if more than 10% sales exported) 0.185 0.165 0.165 (0.044)*** (0.044)*** (0.044)*** % of computer controlled machines 0.002 0.002 0.002 (0.001)*** (0.001)*** (0.001)*** R&D dummy (=1 if any R&D expenditure) 0.171 0.18 0.18 (0.056)*** (0.056)*** (0.056)*** Dummies Industry dummies yes+++ yes+++ yes+++ Region dummies no yes+++ yes+++ Year dummies no no yes Constant 10.86 11.202 11.182 (0.115)*** (0.121)*** (0.123)*** Observations 2944 2944 2944 Adjusted R-squared 0.21 0.22 0.23 Source: Staff calculations based on Malaysia PICS 2002 and 2007. Note: Food processing industry, Klang Valley, and 1999 are respectively used as reference industry, region, and year for fixed-effect dummies. Robust standard errors in parentheses. * significant at 10 percent; ** significant at 5 percent; *** significant at 1 percent. + jointly significant at 10 percent; ++ significant at 5 percent; +++ significant at 1 percent. 225 Table A2.9: Correlates between IC Indicators and Total Factor Productivity (PICS 2007) Dependent Varialbe = (TFP) #1 #2 #3 IC Indicators % of university graduate 0.004 0.004 0.004 (0.001)*** (0.001)*** (0.001)*** Average number of days to obtain fixed phone connection -0.011 -0.012 -0.012 (0.006)* (0.006)* (0.006)* Loss from crime (as % of sales) -0.023 -0.023 -0.023 (0.014)* (0.017) (0.017) Senior managers' time in dealing with regulations (%) -0.006 -0.012 -0.012 (0.003)* (0.005)** (0.005)** Firm Characteristics Age (in log) 0.035 0.025 0.026 (0.022) (0.022) (0.022) Firms size (in log) 0.071 0.07 0.07 (0.023)*** (0.023)*** (0.023)*** Foreign management firm (=1 if more than 30% 0.124 0.135 0.135 ownred by foreign firm) (0.042)*** (0.041)*** (0.041)*** Exporter (=1 if more than 10% sales exported) 0.045 0.059 0.059 (0.028) (0.030)** (0.030)** % of computer controlled machines 0.001 0.001 0.001 (0.001)* (0.001)* (0.001)* R&D dummy (=1 if any R&D expenditure) 0.108 0.093 0.093 (0.048)** (0.047)** (0.047)** Dummies Industry dummies yes+++ yes+++ yes+++ Region dummies no yes yes Year dummies no no yes Constant 2.622 2.738 2.741 (0.112)*** (0.139)*** (0.140)*** Observations 1880 1880 1880 Adjusted R-squared 0.97 0.97 0.97 Source: Staff calculations based on Malaysia PICS 2002 and 2007. Note: Food processing industry, Klang Valley, and 1999 are respectively used as reference industry, region, and year for fixed-effect dummies. Robust standard errors in parentheses. * significant at 10 percent; ** significant at 5 percent; *** significant at 1 percent. + jointly significant at 10 percent; ++ significant at 5 percent; +++ significant at 1 percent. 226 Annex 2.7 Correlates between Changes in Investment Climate and Changes in Firm Performance Table A2.10: Correlates between Changes in IC Indicators and Changes in Labor Productivity in PICS I and PICS II (Panel Sample) Dependent variable (ln(moving average of VAL in PICS II) - ln(moving average of VAL in PICS I) #1 #2 Firm Characteristics Moving average of VAL in PICS I (in log) -0.882 -0.894 (0.079)*** (0.081)*** Age (in log) -0.266 -0.317 (0.137)* (0.144)** Firms size (in log) 0.084 0.1 (0.053) (0.052)* Foreign management firm (=1 if more than 30% ownred by foreign firm) 0.216 0.207 (0.145) (0.147) Exporter (=1 if more than 10% sales exported) -0.245 -0.241 (0.133)* (0.138)* % of computer controlled machines 0.003 0.004 (0.002) (0.002) R&D dummy (=1 if any R&D expenditure) 0.375 0.344 (0.168)** (0.171)** Changes in IC Indicators (PICS II - PICS I) % of university graduate 0.446 (0.336) Average number of days to obtain fixed phone connection -0.716 (0.274)*** Loss from crime (as % of sales) -0.093 (0.134) Senior managers' time in dealing with regulations (%) -0.282 (0.431) Dummies Industry dummies yes yes Region dummies yes+++ yes Constant 10.39 10.27 (1.130)*** (1.180)*** Observations 184 178 Adjusted R-squared 0.63 0.64 Source: Staff calculations based on Malaysia PICS 2002 and 2007. Note: Food processing and Klang Valley are respectively used as reference industry and region for dummies. Robust standard errors in parentheses. * significant at 10 percent; ** significant at 5 percent; *** significant at 1 percent. + jointly significant at 10 percent; ++ significant at 5 percent; +++ significant at 1 percent. 227 Table A2.11: Correlates between Changes in IC Indicators and Changes in Total Factor Productivity in PICS I and PICS II (Panel Sample) Dependent variable (ln(moving average of TFP in PICS II) - ln(moving average of TFP in PICS I) #1 #2 Firm Characteristics Moving average of TFP in PICS I (in log) -0.939 -0.945 (0.047)*** (0.051)*** Age (in log) -0.023 -0.028 (0.019) (0.019) Firms size (in log) 0.025 0.028 (0.008)*** (0.008)*** Foreign management firm (=1 if more than 30% ownred by foreign firm) 0.009 0.007 (0.018) (0.019) Exporter (=1 if more than 10% sales exported) -0.046 -0.047 (0.017)*** (0.017)*** % of computer controlled machines 0.0002 0.0001 (0.0003) (0.0003) R&D dummy (=1 if any R&D expenditure) 0.054 0.051 (0.019)*** (0.01951)** Changes in IC Indicators (PICS II - PICS I) % of university graduate 0.028 (0.045) Average number of days to obtain fixed phone connection -0.055 (0.028)** Loss from crime (as % of sales) -0.011 (0.022) Senior managers' time in dealing with regulations (%) -0.034 (0.030) Dummies Industry dummies yes+++ yes+++ Region dummies yes yes Constant 1.29 1.28 (0.096)*** (0.112)*** Observations 185 179 Adjusted R-squared 0.770 0.77 Source: Staff calculations based on Malaysia PICS 2002 and 2007. Note: Food processing and Klang Valley are respectively used as reference industry and region for dummies. Robust standard errors in parentheses. * significant at 10 percent; ** significant at 5 percent; *** significant at 1 percent. + jointly significant at 10 percent; ++ significant at 5 percent; +++ significant at 1 percent. 228 ANNEX 3 Annex 3.1 Sales and Export Performance of Services Firms Table A3. 1a: Firm Performance in Services 1999-2006 ­ Full Sample 1999-2001 2004-2006 Sales Growth 6.9% 5.6% Costs Growth 12.0% 7.3% Sales-to-costs Ratio 2.2 2.1 Assets Per Worker (RM) 652,794.90 340,165.90 Average Years of Schooling Per Worker 12.3 13.1 Number of Firms 249 303 Source: Productivity and Investment Climate Surveys, 2002 and 2007. Table A3.1b: Firm Performance in Services 1999-2006 ­ Common Firms 1999-2001 2004-2006 Sales Growth 4.1% 2.7% Costs Growth 10.8% 4.7% Sales-to-costs Ratio 2.1 2.2 Assets Per Worker (RM) 402,083.40 429,280.00 Average Years of Schooling Per Worker 12.4 13.6 Number of Firms 137 137 Source: Productivity and Investment Climate Surveys, 2002 and 2007, for the set of common firms in both the surveys. 229 Table A3.2a: Firm Performance in Services 1999-2006 ­ Common Firms Sales of Firms 1999-2001 2004-2006 Average Average Number of (Thousands Annual Number of (Thousands Annual Firms of RM) Growth Firms of RM) Growth Information Technology 27 45,059 26.7% 35 44,555 13.6% Communication Services 10 1,112,347 22.6% 7 1,611,201 39.8% Accounting and Related Services 92 10,716 1.7% 122 10,476 4.9% Advertising and Marketing 16 48,871 16.3% 27 37,803 6.0% Business Logistics 88 169,314 6.3% 108 41,420 1.4% Exports of Firms 1999-2001 2004-2006 Average Average Number of (Thousands Annual Number of (Thousands Annual Exporters of RM) Growth Exporters of RM) Growth Information Technology 8 97,354 15.8% 12 47,497 8.2% Communication Services 3 437,633 32.8% 2 884,763 2.2% Accounting and Related Services 5 53,048 -8.4% 10 44,885 0.2% Advertising and Marketing 3 72,362 4.2% 6 68,199 17.2% Business Logistics 24 538,535 13.5% 30 39,380 3.9% Source: Productivity and Investment Climate Surveys 2002 and 2007. 230 Table A3.2b: Sales and Export Performance of Firms in the Business-Support Services Sectors ­ Common Firms Sales of Firms 1999-2001 2004-2006 Average Average Number (Thousands of Annual Number of (Thousands Annual of Firms RM) Growth Firms of RM) Growth Information Technology 11 39,696 22.8% 11 75,441 8.6% Communication Services 3 2,596,492 32.7% 3 1,932,151 4.1% Accounting and Related Services 61 13,508 3.8% 61 13,066 6.5% Advertising and Marketing 10 58,874 18.6% 10 69,763 -1.2% Business Logistics 40 32,918 2.7% 40 58,428 -3.7% Exports of Firms 1999-2001 2004-2006 Number Average Average of (Thousands of Annual Number of (Thousands Annual Exporters RM) Growth Exporters of RM) Growth Information Technology 2 48,180 14.2% 2 203,580 23.5% Communication Services 1 550 - 1 1,764,016 12.8% Accounting and Related Services 0 0 - 0 - Advertising and Marketing 1 102,789 6.6% 1 12,928 0.5% Business Logistics 6 47,624 8.3% 6 118,512 -8.7% Source: Productivity and Investment Climate Surveys 2002 and 2007. 231 Annex 3.2 Firm Characteristics and Firm Performance Table A3.3: Correlates between Firm Characteristics and Firm Performance (3) Log of sales (1) Log of sales (2) Growth of sales per worker (4) Log of sales Explanatory Variables (between estimate) (between estimate) (between estimate) per worker (OLS) Log of assets value per worker 0.53*** 0.00 0.55*** 0.52*** (0.06) (0.01) (0.04) (0.02) Share of export in sales 0.81** 0.15* 0.22 0.43*** (0.37) (0.09) (0.27) (0.14) Foreign ownership dummy 0.59*** -0.02 0.17 0.09 (positive foreign equity) (0.21) (0.05) (0.15) (0.08) Firm age 0.03*** -0.00** -0.01** -0.01*** (0.01) (0.00) (0.01) (0.00) Constant 8.09*** 0.18 5.62*** 5.61*** (1.54) (0.32) (1.11) (0.30) Industry fixed effects yes yes yes yes Region fixed effects yes yes yes yes Year fixed effects yes yes yes yes Number of observations 1476 985 1476 1476 Number of firms 393 390 393 393 R-squared 0.4048 0.0847 0.4037 0.3552 Note: Standard errors in parentheses. *, **, and *** indicate significance at 90 percent, 95 percent, and 99 percent confidence levels respectively. 232 Annex 3.3 Role of Presence of Foreign-Owned Firms in Productivity Table A3.4: Determinants of Firm Performance Dependent Variable: Log of Sales per Worker Explanatory Variables (1) Between Estimate (2) Between Estimate Log of assets value 0.57*** 0.56*** per worker (0.04) (0.04) Log of average years of -0.13 -0.08 education per worker (0.18) (0.20) Share of sales for exports 0.24 0.17 (0.25) (0.26) Training 0.14 0.16 (0.15) (0.16) Foreign equity more than 30% 0.41** 0.42** (0.17) (0.17) Age -0.01* -0.01** (0.01) (0.01) Constant 5.32*** 5.56*** (1.15) (1.19) Industry fixed effects no yes Region fixed effects no yes Year fixed effects yes yes Number of observations 1408 1408 Number of firms 377 377 Note: Standard errors in parentheses. *, **, and *** indicate significance at 90 percent, 95 percent, and 99 percent confidence levels respectively. 233 Table A3. 5: Domestic Firms Tend to Benefit from the Presence of Foreign Firms Dependent Variable: Log of Sales per Worker Estimation: Fixed Effect Panel Regressions Explanatory Variables (1) (2) Log of assets value 0.31*** 0.31*** per worker (0.03) (0.03) Log of average years of 0.32** 0.32** education per worker (0.14) (0.14) Share of sales for exports 0.69*** 0.68*** (0.24) (0.24) Foreign_Presence 0.75* 0.98** (0.45) (0.45) Entry Competition 0.01 -0.00 (0.26) (0.26) Training 0.16 0.15 (0.11) (0.11) age -0.00 -0.00 (0.00) (0.00) Constant 7.24*** 7.23*** (0.49) (0.49) Firm fixed effects yes yes Year fixed effects yes yes Number of observation 1143 1143 Notes: Standard errors in parentheses. *, **, and *** indicate significance at 90 percent, 95 percent, and 99 percent confidence levels respectively. Column (1) uses all foreign firms in the calculation of foreign presence variable; Column (2) only use those firms with foreign equity > 30 percent in the calculation of foreign presence variable. In both columns, sample only consists of the 310 strictly domestic firms. 234 ANNEX 4 Annex 4.1 PICS Samples at the Regional Level Manufacturing Firms 1. The distribution of manufacturing firms across industries varies widely across regions. About 90 percent of all surveyed producers are in Klang Valley, the South and the North regions. Half operates in the food and rubber and plastic industries, which are dominant in most regions. Some industries are more concentrated in certain areas, e.g., machinery and equipment in Klang Valley, electronics equipment in the North, garments in the South, and furniture in Sabah. The most concentrated region appears to be Sarawak, where over 70 percent of all firms are food processing businesses. There is a high degree of consistency between the two rounds of PICS, both in terms of regional and industrial distributions. Refer to Table A1.7 for details. 2. Employment size and foreign ownership are reasonably diverse across regions (Table A4.1). The proportions of small firms (employing fewer than 50 workers) are much higher in the East, Sabah, and Sarawak. These range between 58-78 percent of all firms compared to 40 percent in the other three regions. In addition to being larger, more companies in Klang Valley, the North, and the South are foreign-owned. Sampling consistency between the two PICS remain relatively high. Table A4.1: Distribution of Manufacturing Firms by State, Employment and Ownership in PICS 2002 and 2007 Employment size Foreign- Small Medium Large owned Klang Valley 51.8 31.3 16.9 16.8 North 40.2 31.8 27.9 27.5 South 42.6 40.5 16.9 30.4 PICS 2002 East 85.7 4.8 9.5 9.5 (902 firms) Sabah 58.3 33.3 8.3 8.7 Sarawak 82.4 11.8 5.9 8.8 Share in whole sample 48.3 33.4 18.3 23.2 Klang Valley 40.9 36.8 22.3 21.2 North 39.5 35.4 25.1 28.0 South 47.1 32.2 20.8 25.4 PICS 2007 East 62.1 20.7 17.2 10.3 (1,115 firms) Sabah 58.1 35.5 6.5 3.2 Sarawak 78.3 17.4 4.4 6.5 Share in whole sample 45.1 33.8 21.2 22.7 Source: Malaysia PICS 2002 and PICS 2007 Note: Small enterprises employ 1-49 workers, 50-199 for medium, and over 200 for large Services Firms 3. The distribution of services firms across industries also varies widely by regions. A large share of the services companies in PICS are accounting firms and business logistics companies 235 concentrated in Klang Valley. Up to 70 percent of all services firms in PICS 2007 are located in Klang Valley. There are accounting and business logistics firms in all regions but businesses such as telecommunications and advertising and marking were only surveyed in Klang Valley. Note that there is no coverage of services firms located in the East, so there will always be five regions for the analysis of services firms. Refer to Table A1.8 for details. 4. In terms of employment size, table A4.2 shows that, except in Klang Valley, most services firms are small. A higher share of companies in Klang Valley is foreign-owned. There appears to be high consistency between the two rounds of surveys. Table A4.2: Distribution of Services Firms by State, Employment Size and Ownership in PICS 2002 and 2007 Employment size Foreign- Small Medium Large owned Klang Valley 48.1 35.0 16.9 9.6 North 42.9 42.9 14.3 0.0 South 57.1 14.3 28.6 0.0 PICS 2002 Sabah 69.2 30.8 0.0 0.0 (250 firms) Sarawak 72.2 22.2 5.6 0.0 Share in whole sample 51.0 33.3 15.7 7.1 Klang Valley 50.0 35.7 14.3 10.1 North 75.8 15.2 9.1 5.7 South 78.3 17.4 4.4 0.0 PICS 2007 Sabah 64.3 21.4 14.3 0.0 (303 firms) Sarawak 77.8 16.7 5.6 0.0 Share in whole sample 57.4 30.2 12.4 7.7 Source: Malaysia PICS 2002 and PICS 2007 Note: Small firms employ 1-49 workers, 50-199 for medium, and over 200 for large 236 ANNEX 5 Annex 5.1 Description of PICS Employer-Employee Matching Data 1. The matched employer-employee data used in this section are taken from the 2002 Productivity and Investment Climate Survey (PICS) of establishments in Malaysia conducted by the Economic Planning Unit of the Prime Minister's Department (EPU) with technical assistance from the World Bank. 2. The PICS is comprised of two components: a workplace questionnaire administered to management and a worker questionnaire administered to up to 10 randomly sampled full-time employees in each establishment. Structured interviews with management consisted of approximately 300 questions aimed to collect information on a wide array of firm and plant characteristics in the following 9 areas: governance and ownership structure; investment, technology, and innovation; labor market, education, and skills; investment climate constraints and business relations; infrastructure, access to land, and government regulations; international trade; products and inputs; corporate finance; and labor and human resources. The survey of employees was conducted in conjunction with the main survey of establishments and involved the administration of a 75 item worker questionnaire in each establishment. 3. Permission to interview employees was granted at all establishments in the sample. In 2002, response rate of the worker survey was very high with the targeted number of 10 full-time workers being interviewed in 88.8 percent of the establishments in manufacturing and in 79.0 percent of the establishments in business support services. In 2007, the response rate of the worker survey was similar high with the targeted number of 10 full-time workers being interviewed in 93.8 percent of the establishments in manufacturing and in 93.0 percent of the establishments in business support services. The final result is a very large and representative matched employer-employee dataset, with several direct indicators of worker heterogeneity captured in an array of human capital and demographic variables in addition to the detailed information collected on workplace and firm characteristics. The final 2002 sample comprises 8,590 individuals employed in the 902 manufacturing establishments surveyed and 2,232 individuals employed in the 250 business support services sector establishments surveyed. The final 2007 sample comprises 10,774 individuals employed in the 1,115 manufacturing establishments surveyed and 2,915 individuals employed in the 303 business support services sector establishments surveyed. 237 Annex 5.2 Methodology of Wage Premium Estimates 1. The empirical strategy sets out to estimate the wage effects of a high school diploma and college degree in Malaysia. An alternative to using the categorical measure of completed schooling years is to enter a string of dummy variables, one for each year of completed schooling (Hungerford and Solon, 1987; Jaeger and Page, 1996). By relaxing the assumption that education enters the wage equation linearly, the following specifications impose fewer restrictions on the shape of the wage-schooling profile and maximize the amount of wage variation that can be explained by complete years of schooling: ln Wi = + k Di (credential = k ) + g Di ( Si = g ) + X i + X i2 + i , [1] kK gG where the dependent variable ln Wij , measured in nominal terms, is the natural logarithm of hourly wage of worker i in establishment j (subscript suppressed); K = {high school diploma, college degree} and Di (credential = k ) is a dummy variable equal to 1 if the individual i possesses educational credential k, and 0 otherwise; G = {9 or less, 10, 11, , 16, 17, 18 or more years} and Di (Si = g ) is a dummy variable equal to 1 if individual i has completed g years of schooling, and 0 otherwise; X i are years of potential experience defined in the usual way (that is, age - Si - 6); and i is the error term.148 Since information on both years of completed schooling as well as diploma and degree received is available in the match employer-employee dataset, direct estimates of academic credentials can be estimated from the above equation. Equation [1] is similar to those used by Jaeger and Page (1996), Bauer, Dross, and Haisken- DeNew (2003), and Gibson (2004) to estimate wage effects of academic credentials in the US, Japan, and New Zealand, respectively. 2. The above definition of the high school diploma dummy variable follows Jaeger and Page (1996) and is equal to 1 for all workers who reported receiving a high school diploma as well as those who reported receiving a college degree. That is, workers who reported receiving a college degree are assumed to have also received a high school diploma (since the matched employer-employee datasets provides information only on the highest educational credential obtained). The reported coefficients on the college degree dummy variable therefore can be interpreted as the marginal effect over a high school degree. Equations with the high school diploma dummy variable alternatively defined as equal to 1 only for those workers whose highest educational credential is the high school diploma are also estimated, and the reported coefficients on the college degree dummy variable from these regressions can be interpreted as the total returns to a college degree. 3. Equation [1] is then augmented with additional covariates to obtain the fully-specified regression model as follows: 148 Two variables are used to construct wages: 2002 and 2007 salary in ringgits divided by four times the average number of hours worked per week. Workers with log hourly wages that are more (less) than the 75th (25th) percentile plus (minus) 3 times the interquartile range are considered extreme outliers and excluded from the analysis. 238 ln Wi = + k Di (credential = k ) + g Di ( Si = g ) + X i + X i2 + iw + a j + i , [2] kK gG where iw denotes a set of individual characteristics of worker i and a j is a fixed effect for establishment j. A complete listing of individual-level variables used in the paper and their definitions as well as descriptive statistics is provided in the Udomsaph (2006). 4. Columns (1) through (4) of Table A5.1 report the results obtained from estimating equations [2] for the sample of workers in the match employer-employee dataset for Malaysian manufacturing and business support services in 2002 and 2007. Coefficients in Table A5.1 are estimated using ordinary least squares with standard errors adjusted for clustering on establishment, and fixed effects estimation is used to remove establishment characteristics. The marginal effect of receiving a college degree over a high school diploma is presented along with the other main covariates, while the total returns of receiving a college degree are reported at the bottom of Table A5.1. Column (5) present result for Thai manufacturing in 2005 for comparative purposes. 239 Table A5.1: Estimated Returns to High School Diplomas and College Degrees, Selected Worker Characteristics [Dependent Variable: individual log hourly wage] Malaysia Thailand Business-Supporting Manufacturing Manufacutring Services 2002 2007 2002 2007 2005 ** ** * High School Diploma 0.105 0.140 0.087 0.046 0.052* /Por Wor Chor (0.024) (0.021) (0.036) (0.037) (0.024) Marginal Effect Over High School Diploma/Por Wor Chor Vocational Certificate - - - - 0.062 (0.033) ** ** ** ** ** College Degree 0.225 0.165 0.222 0.196 0.300 (0.031) (0.025) (0.036) (0.033) (0.040) Years of Completed Schooling 9 or less -0.143** -0.131** -0.209** -0.219** -0.103** (0.022) (0.019) (0.055) (0.046) (0.025) * ** 10 0.003 -0.058 -0.140 -0.082 -0.029 (0.031) (0.027) (0.040) (0.070) (0.052) 11 -0.017 -0.038* -0.067 -0.07 -0.052 (0.021) (0.018) (0.045) (0.038) (0.068) 12 ref. ref. ref. ref. ref. 13 0.061* 0.006 0.059 -0.055 0.036 (0.027) (0.024) (0.044) (0.048) (0.081) 14 0.053 0.022 0.04 0.066 0.070* (0.033) (0.027) (0.047) (0.048) (0.033) 15 0.103** 0.075* 0.120* 0.055 0.157* (0.038) (0.030) (0.049) (0.051) (0.064) 16 0.125** 0.073 * 0.084 0.046 0.085 * (0.043) (0.036) (0.055) (0.052) (0.040) more than 16 0.130** 0.110** 0.177** 0.124* 0.336** (0.041) (0.035) (0.059) (0.054) (0.046) Potential experience 0.029** 0.024** 0.058** 0.035** 0.028** (0.002) (0.002) (0.004) (0.003) (0.001) ** Potential experience squared/100 -0.048 -0.038** -0.099** -0.047** -0.041** (0.003) (0.003) (0.010) (0.006) (0.003) Management 0.555** 0.429** 0.363** 0.437** 0.765** (0.022) (0.019) (0.050) (0.035) (0.018) Professionals 0.450** 0.420 ** 0.257 ** 0.369 ** 0.412 ** (0.026) (0.023) (0.047) (0.034) (0.015) Skilled production workers 0.224** 0.177** 0.156** 0.195** 0.194** (0.015) (0.014) (0.046) (0.031) (0.009) Nonproduction workers 0.226** 0.162** 0.007 0.074* 0.178** (0.018) (0.016) (0.045) (0.032) (0.011) Studied abroad 0.080** -0.060** 0.051 0.155** 0.321** (0.025) (0.023) (0.033) (0.027) (0.037) 240 Table A5.1: Estimated Returns to High School Diplomas and College Degrees in Malaysia, Selected Worker Characteristics (cont.) Malaysia Thailand Business-Supporting Manufacturing Manufacutring Services 2002 2007 2002 2007 2005 ** ** Basic (Computer Skills) 0.102 0.112 0.055 0.017 0.094** (0.015) (0.014) (0.049) (0.043) (0.011) Moderate (Computer Skills) 0.203** 0.216** 0.103* 0.065 0.173** (0.018) (0.015) (0.049) (0.040) (0.013) Complex (Computer Skills) 0.166** 0.235** 0.131* 0.139** 0.256** (0.030) (0.024) (0.057) (0.046) (0.023) Employer-provided training in the areas of: People skills 0.056** 0.042** 0.064** 0.008 0.070** (0.013) (0.011) (0.023) (0.016) (0.009) Production technologies 0.040* 0.031 0.112 0.043 0.024 (0.018) (0.017) (0.068) (0.045) (0.015) Marketing (0.028) 0.074* (0.022) 0.161** 0.120** (0.040) (0.034) (0.073) (0.048) (0.030) Information technology 0.059* (0.007) (0.052) 0.088** 0.000 (0.026) (0.029) (0.043) (0.030) (0.023) Management/quality technologies 0.076** 0.077** 0.012 0.082** 0.045** (0.018) (0.018) (0.033) (0.029) (0.014) Intellectual property 0.032 (0.053) 0.006 (0.066) (0.010) (0.086) (0.075) (0.098) (0.096) (0.040) Safety procedures 0.010 (0.025) (0.030) (0.016) -0.046** (0.021) (0.021) (0.060) (0.050) (0.014) Language skills (0.023) (0.012) (0.045) 0.024 0.062 (0.034) (0.036) (0.045) (0.040) (0.033) Other (On-the-job Training) 0.020 -0.116* 0.013 0.067 0.082** (0.044) (0.055) (0.057) (0.048) (0.020) On-the-job training received from previous employer 0.077** 0.028* 0.048 0.026 0.048** (0.016) (0.013) (0.027) (0.022) (0.010) Off-the-job training/certification 0.044** 0.035* 0.007 0.078** (0.010) (0.014) (0.014) (0.025) (0.020) (0.021) Lack English proficiency in doing job -0.003 -0.034** -0.036 -0.015 -0.026** (0.012) (0.010) (0.026) (0.017) (0.008) Unemployed in last two years -0.039 -0.025 -0.067 -0.050 -0.039 (0.033) (0.016) (0.057) (0.030) (0.033) Distance from job 0.193** 0.075** -0.127** 0.018 0.287** (0.045) (0.026) (0.049) (0.040) (0.028) Unmarried female -0.138** -0.145** (0.042) 0.003 -0.093** (0.017) (0.015) (0.031) (0.023) (0.011) Married male 0.138** 0.104** 0.080* 0.076** 0.061** (0.016) (0.014) (0.032) (0.024) (0.011) Married female -0.130** -0.106** (0.026) 0.017 -0.102** (0.017) (0.014) (0.033) (0.024) (0.011) 241 Table A5.1: Estimated Returns to High School Diplomas and College Degrees in Malaysia, Selected Worker Characteristics (cont.) Malaysia Thailand Business support Manufacturing Manufacturing Services 2002 2007 2002 2007 2005 Tenure 0.022** 0.024** 0.013** 0.020** 0.022** (0.002) (0.002) (0.005) (0.002) (0.002) - Tenure squared -0.021** 0.021** 0.006 -0.031** -0.0002** (0.007) (0.006) (0.016) (0.013) Malaysian nationality: Ethnicity Yes Yes Yes Yes Yes Foreign worker: Country of origin Yes Yes Yes Yes Yes Constant 0.844** 1.152** 1.747** 1.598** (0.030) (0.026) (0.251) (0.061) Establishment fixed effects Yes Yes Yes Yes Yes Observations 7,812 10,585 2,068 2,826 13,476 Number of Firms 893 1,101 245 296 13,86 Adjusted R2 0.487 0.381 0.389 0.386 0.576 _____________________________________________________________________________ Note: Robust standard errors are denoted in parentheses. * denotes significance at 5 percent level; ** denotes significance at 1 percent level. Source: Malaysia PICS 2002 and 2007; Thailand PICS 2005. Staff calculations. 242 Worker Variables Definition Hourly wage 2001 and 2007 salary in ringgits divided by 4 times the average number of hours worked per week. High school diploma Equal to 1 if the highest level of formal education attained is high school, and equal to 0 otherwise. College degree Equal to 1 if the highest level of formal education attained is university, and equal to 0 otherwise. Potential experience Age minus the number of years of formal education minus 6. Management Equal to 1 if employee is making management decisions (excludes supervisors.) Professionals Equal to 1 if employee is a trained and certified specialist outside of management such as engineers, accountants, lawyers, chemists, scientists, software programmers. Generally, Professionals hold a University-level degree. Skilled production workers Equal to 1 if employee is a technician involved directly in the production process or at a supervisory level and is considered by management to be skilled. Unskilled production workers Equal to 1 if employee is involved in production process and is considered by management to be unskilled. Nonproduction workers Equal to 1 if employee is a support, administrative, or sales worker and not included in management or among professionals. Studied abroad Equal to 1 if studied aboard, and equal to 0 otherwise. None (Computer Skills) Equal to 1 if rated self as having no computer skills, and equal to 0 otherwise. Basic (Computer Skills) Equal to 1 if rated self as having basic computer skills such as printing invoices, and equal to 0 otherwise. Moderate (Computer Skills) Equal to 1 if rated self as having moderate computer skills such as word processing and e-mail, and equal to 0 otherwise. Complex (Computer Skills) Equal to 1 if rated self as having complex computer skills such as programming, and equal to 0 otherwise. 243 Worker Variables Definition People skills Equal to 1 if "4 (Very Important)" is the response to the question: "In your job, how important is dealing with people?" (possible answers range from 1-4), and equal to 0 otherwise. Production technologies Equal to 1 if received formal training in production technologies from current employer, and equal to 0 otherwise. Marketing Equal to 1 if received formal training in marketing from current employer, and equal to 0 otherwise. Information technology Equal to 1 if received formal training in information technologies from current employer, and equal to 0 otherwise. Management/quality technologies Equal to 1 if received formal training in management/quality technologies from current employer, and equal to 0 otherwise. Intellectual property Equal to 1 if received formal training in intellectual property from current employer, and equal to 0 otherwise. Safety procedures Equal to 1 if received formal training in safety procedures from current employer, and equal to 0 otherwise. Language skills Equal to 1 if received formal training in language skills from current employer, and equal to 0 otherwise. Other (On-the-job Training) Equal to 1 if received formal training from current employer in areas other than those listed above, and equal to 0 otherwise. On-the-job training received from Equal to 1 if received formal training from previous employer, and previous employer equal to 0 otherwise. Off-the-job training/certification Equal to 1 if completed a professional certification or skills training program such as attending a vocational school or if currently enrolled in an after work learning program, and equal to 0 otherwise. Lack English proficiency in doing job Equal to 1 if "English language proficiency" is the response to the question: "What skill do you lack the most in doing your job?" (there are 12 possible answers), and equal to 0 otherwise. Unemployed in last 2 years Equal to 1 if unemployed within the last 2 years, and equal to 0 otherwise Distance from job Number of kilometers from job. Unmarried male Equal to 1 if single, separated, divorced, or widowed man, and equal to 0 otherwise. 244 Worker Variables Definition Unmarried female Equal to 1 if single, separated, divorced, or widowed woman, and equal to 0 otherwise. Married male Equal to 1 if married man, and equal to 0 otherwise. Married female Equal to 1 if married woman, and equal to 0 otherwise. Tenure Number of years worked in current establishment. Source Author's definitions using Malaysia PICS 2002 and 2007 data. :: 245 Annex 5.3 Share of Skilled and Unskilled Workers and Their Hourly Wage in the Manufacturing and Business support Services Panel Sample, 2001 and 2006 Manufacturing Share of skilled and unskilled labor ( percent) Electronics Rubber & & Woods & 2001 Food Textiles Garments Chemicals plastics Machinery electrical Auto furniture Skilled labor 54.5 62.0 65.2 63.8 49.5 69.5 64.3 51.0 54.0 Unskilled labor 45.5 38.0 34.8 36.2 50.5 30.5 35.7 49.0 46.0 2006 Skilled labor 46.4 61.6 48.2 62.5 46.2 64.3 51.4 51.7 35.0 Unskilled labor 53.6 38.4 51.8 37.5 53.8 35.7 48.6 48.3 65.0 Hourly wage (in RM) Electronics Rubber & & Woods & 2001 Food Textiles Garments Chemicals plastics Machinery electrical Auto furniture Skilled labor 8.8 7.7 6.6 13.0 9.3 10.7 10.5 10.8 8.1 Unskilled labor 4.5 4.2 4.3 5.9 4.5 6.9 6.1 6.7 4.4 2006 Skilled labor 9.4 7.1 6.6 13.1 8.9 11.1 9.9 11.1 8.4 Unskilled labor 4.9 5.4 4.4 6.2 5.2 7.9 7.0 6.2 4.7 246 Annex 5.3 Share of Skilled and Unskilled Workers and Their Hourly Wage in the Manufacturing and Business support Services (cont.) Business support services Share of skilled and unskilled labor (%) Information Business 2001 technology Telecommunication Accounting advertising logistics Skilled labor 99.5 98.6 92.8 94.3 66.2 Unskilled labor 0.5 1.4 7.2 5.7 33.8 2006 Skilled labor 98.0 96.0 93.9 90.6 63.0 Unskilled labor 2.0 4.0 6.1 9.4 37.2 Hourly wage (in RM) 2001 Skilled labor 23.4 26.1 14.9 22.8 14.6 Unskilled labor 13.1 8.2 16.1 7.5 2006 Skilled labor 23.9 21.2 15.9 38.4 15.2 Unskilled labor 14.1 17.6 9.6 13.5 9.0 Note: Skilled labor includes management, professional, skilled production worker and unskilled labor includes the rest of categories. Wages are measured in nominal terms. Source: Malaysia PICS 2002 and 2007. 247 ANNEX 6 Annex 6.1 Construction of the Firm-Level Technological Capabilities Index (TCI) 1. A functional index--Technological Capabilities Index (TCI)--quantifies and summarizes differences in firm-level technological capabilities and serves as useful tool for the assessment of technological development. TCI permits useful comparison of the technological capabilities across firms and enables econometric analysis of the influences on the acquisition of firm-level technological capabilities. TCI captures and integrates a variety of objective and subjective information into coherent measures of a firm's capacity to establish, operate, and transfer technology. TCI provides a composite measure of technological capabilities composed of information about firm-level technological behavior that is provided by the rich data collected in the PICS. 2. Following Wignaraja (1998, 2002), TCI draws on the taxonomy developed by Lall (1992), which identifies and categorizes firm-level technological capabilities into investment, production, and linkages activities. Table A6.1 recreates the illustrative matrix of the major technical functions involved in the development of firm-level technological capabilities from Lall (1992). Firm level technological capacities are divided broadly into three categories-- investment, production, and linkages. The columns categorize the major firm-level technological capabilities by function, and the rows identify the degree of complexity or difficulty as measured by the nature of activity from which the technological capability arises.149 The advantage of this framework is that it provides a clear continuum of technical functions that are necessary for the development of firm-level technological capabilities from the time new technology enters a given firm to when it exits to other firms and institutions. 3. TCI is constructed by mapping questions available in the PICS onto the taxonomy of firm-level technological capabilities developed by Lall (1992) described above, with the scoring system shown in Table A6.2. TCI is composed of 30 separate technical activities. Investment technological capabilities are represented by 7 separate technical activities covering investment planning, technology transfer, and workforce training. Production technological capabilities is represented by 14 separate technical activities, which range from common process engineering tasks (such as upgrading machinery and equipment, introducing new technology, and ISO 9,000 quality management status) to product engineering tasks (such as improving existing products, introducing new products, and researching and developing new designs). Linkages technological capabilities are represented by 9 separate technical activities involving technological interactions with buyers of output, suppliers, and research institutions. A single point is given for each technical activity the firm has performed; for higher levels of IT-related investments and computer-controlled machinery, an additional point is scored. Therefore, each firm is ranked out 149 Since the simplicity or complexity of a particular function is difficult to judge a priori, this categorization is simply indicative rather than exact. The technological capabilities matrix is not intended to represent a necessary sequence of learning since different firms do adopt different technologies using different sequences. See Lall (1992). 248 of a total technological capability score of 33 and the result is normalized to give a value between 0 and 1.150 4. While all activities listed in the matrix of technological capabilities do not have to be performed for every industrial venture, there is a core set of basic competencies that have to be internalized by the firm to ensure successful commercial operation. If a firm is unable by itself to decide on its investment plans or selection of equipment processes, or to reach minimum levels of operating efficiency, quality control, equipment maintenance, and cost improvement, or to adapt its product designs to changing market conditions, or to establish effective linkages with reliable suppliers, it is unlikely to be able to compete effectively in open markets. Moreover, the basic core of skills, knowledge, and experience must grow over time as the firm undertakes more complex tasks. The hallmark of a technologically "mature" firm is the ability to identify prerequisite skills for efficient specialization in technological activities, to extend and deepen this knowledge with experience and effort, and to draw selectively on other economic agents to complement the existing capabilities of the firm.151 150 Weighting complex activities more than simple ones was considered in the construction of TCI. However, the very nature of technological learning through the accumulation of experience in problem solving aided through formal research effort or aided by external inputs dictates that mastery would proceed from simpler to more difficult activities. For example, an establishment undertaking an advanced activity such as in-house process innovation would have also completed the basic task of assimilating process technology. Therefore, that particular establishment would have gained 2 points and further weighting would skew TCI in favor of establishments with high levels of technological capabilities. 151 See Lall (1992). 249 Table A6.1: Illustrative Matrix of Technological Capabilities by Lall (2002) INVESTMENT PRODUCTION LINKAGES PRE PROJECT PROCESS PRODUCT INDUSTRIAL WITHIN INVESTMENT EXECUTION ENGINEERING ENGINEERING ENGINEERING ECONOMY SIMPLE · Pre-feasibility · Civil construction · Debugging, · Assimilation of · Work flow · Local procurement of ROUTINE and feasibility · Ancillary services balancing product design · Scheduling goods and services (Experienced studies · Equipment · Equality control · Minor adaptation to · Time-motion · Information exchange based) · Site selection erection preventive market needs studies Inventory with suppliers · Scheduling of · Commissioning maintenance control investment · Assimilation of process technology BASIC ADAPTIVE · Search for · Equipment · Equipment · Product quality · Monitoring · Technology transfer DUPLICATIVE technology procurement stretching improvement productivity of local suppliers (Search based) source · Detailed · Process adaptation · Licensing and · Improved · Coordinated design · Negotiation of engineering and cost saving assimilating new coordination · Science & technology contracts · Training and · Licensing new imported product links · Bargaining recruitment of technology technology suitable terms skilled personnel · Information systems INTERMEDIATE DEGREE OF COMPLEXITY INNOVATIVE · Basic process · In-house process · In-house product · Turnkey capability RISKY design innovation innovation · Cooperative R&D (Research based) · Equipment design · Basic research · Basic research · Licensing own and supply technology to others ADVANCED 250 Table A6.2: Technological Capabilities Index (TCI) for the Malaysia PICS Sample LINKAGES WITHIN INVESTMENT (7) PRODUCTION (14) ECONOMY (9) Expect to make a substantial What percentage of your production What percentage of your increase in investment in order to machines is computer controlled? (Q.3.7) domestic inputs comes from the increase capacity or improve same region that your quality? (Q.3.3) [None=0] establishment is located? (Q.8.7) [>0 and 45 Percent=1] [Yes=1] [No=0] [>45 Percent=2] [None=0] [>0 and 80 Percent=1] What percentage of your next Employ staff exclusively for design/doing [>80 Percent=2] investment will be IT related? innovation/R&D? (Q.3.12) (Q3.4.y) Do your suppliers make your raw [Yes=1] [No=0] materials to your unique [None=0] specification? (Q8.14) [>0 and 10 Percent=1] Subcontract R&D project to other [>10 Percent=2] companies or organizations? (Q.3.13) [Yes=1] [No=0] Training of workforce to [Yes=1] [No=0] Use e-mail in interactions with implement technology transferred clients and suppliers? (Q3.25.a) from parent establishment? Paid royalties? (Q.3.14) (Q.3.19.2) [Yes=1] [No=0] [Yes=1] [No=0] [Yes=1] [No=0] Do you sell through your Planning to introduce new designs or website? (Q.25.b.y) If a supplier to a MNC and products in the next 2 years? (Q.3.15) learned new technology from that [Yes=1] [No=0] MNC, was it explicitly via MNC [Yes=1] [No=0] licensing, training, quality If a supplier to a MNC, did you certification programs? Upgraded machinery & equipment? learn any new technology from (Q.20.y) (Q.3.16.1) that MNC? (Q.20) [Yes=1] [No=0] [Yes=1] [No=0] [Yes=1] [No=0] Run formal in-house training Entered new markets due to process or Technology innovation developed programs for employees? product improvements in cost or quality? in collaboration with other firms? (Q.10.14) (Q.3.16.2) (Q.3.18.1) [Yes=1] [No=0] [Yes=1] [No=0] [Yes=1] [No=0] Send employees to formal Filed any patent/ utility models or Technology innovation developed training programs run by other copyright protected materials? (Q.3.16.3) in collaboration with universities? organizations? (Q.3.18.2) (Q.10.17) [Yes=1] [No=0] [Yes=1] [No=0] [Yes=1] [No=0] Developed a new product line? (Q.3.16.4) [Yes=1] [No=0] (TableA 6.2 continued on next page) 251 Table A6.2: Technological Capabilities Index (TCI) for the Malaysia PICS Sample (cont.) LINKAGES WITHIN INVESTMENT PRODUCTION ECONOMY Did you spend on design or R&D Upgraded an existing product line? Technology innovation developed in previous year? (Q.9.9) (Q.3.16.5) in collaboration with research institutions? [Yes=1] [No=0] [Yes=1] [No=0] (Q.3.18.3) Introduced new technology that has [Yes=1] [No=0] substantially changed the way the main product is produced? (Q.3.16.6) Technology innovation developed in collaboration other [Yes=1] [No=0] institutions? (Q.3.18.5) Adaptation or R&D of technology [Yes=1] [No=0] transferred from parent establishment to suit local conditions? (Q.3.19.1) [Yes=1] [No=0] Have you received any government incentives to conduct technological innovation and R&D? (Q.3.21) [Yes=1] [No=0] Has your firm received any ISO (e.g. 9000, 9002, or 14,000) certification? (Q,26) [Yes=1] [No=0] Note: Question number for variables in Malaysia PICS 2007 dataset denoted in parentheses. 252 Annex 6.2 TCI Statistics Across Firms in Malaysia Table A6.3: Descriptive Statistics of Manufacturing TCI by Region Mean Scores Klang Valley North South East Coast Sabah Sarawak Overall TCI 2002 0.390 0.400 0.323 0.322 0.245 0.244 2007 0.408 0.385 0.339 0.406 0.339 0.234 Investment TCI 2002 0.337 0.376 0.279 0.256 0.208 0.176 2007 0.365 0.341 0.302 0.366 0.347 0.193 Production TCI 2002 0.336 0.341 0.271 0.238 0.167 0.153 2007 0.335 0.338 0.277 0.293 0.242 0.151 Linkages TCI 2002 0.311 0.304 0.272 0.339 0.264 0.310 2007 0.313 0.270 0.259 0.362 0.281 0.243 Frequency distribution of Overall TCI (%) Observations 0