33059 Social Expenditure and Economic Growth Sharing Growth in Papers a Japanese Way Eiji Tajika and Yuji Yui orking Funded by: The Policy and Human Resources Development Trust Fund W of the Government of Japan WBI Social Expenditure and Economic Growth: Sharing Growth in a Japanese Way Eiji Tajika and Yuji Yui Japan has achieved rapid improvement in social welfare during its high economic growth era, while the expansion of social expenditure as a percentage of GDP remained stable during this period. One of the reasons for this may have been that Japan had a relatively younger population at the time. This paper shows that public policies played a major role in promoting social welfare during this period by adopting appropriate budgetary & tax policies and introducing local delivery of social services. However, this pattern of maintaining high economic growth and low social expenditure came to an end in the early 1970s. The government responded to public demand for more comprehensive welfare policies by increasing benefits such as medical services and pension and as a result, social expenditure, which was once the nation's building block, turned into a stumbling block. World Bank Institute Copyright © 2002 The International Bank for Reconstruction and Development/The World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. The World Bank enjoys copyright under protocol 2 of the Universal Copyright Convention. This material may nonetheless be copied for research, educational, or scholarly purposes only in the member countries of The World Bank. Material in this series is subject to revision. The findings, interpretations, and conclusions expressed in this document are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or the members of its Board of Executive Directors or the countries they represent. Social Expenditure and Economic Growth Sharing Growth in a Japanese Way Eiji Tajika and Yuji Yui 2002. 25 pages. Stock No. 37198 Contents Foreword v The Nature of Japanese Social Expenditure in the Postwar Period 1 Three Aspects of Public Policy and Social Expenditure 7 Conclusion: Lessons from the Japanese Experience 18 References 19 Foreword This paper was prepared for a project on Social Development in East Asia. The project was organized by the World Bank Institute under the auspices of the Program for the Study of the Japanese Development Management Experience (The Brain Trust Program), which is financed by the Policy and Human Resources Development Trust Fund established at the World Bank by the Government of Japan (GOJ). We greatly appreciate the generous support from GOJ. The principal objectives of this Program are to conduct studies on Japanese and East Asian development management experience and to share the lessons of this experience with developing and transition economies. The experiences of other countries are often covered in order to ensure that these lessons are placed in the proper context. This comparative method helps identify factors that influence the effectiveness of specific institutional mechanisms, governance structures, and policy reforms in different contexts. A related objective of the Program is to promote the exchange of ideas on development among Japanese and non-Japanese scholars, technical experts and policy makers. The papers commissioned for this project cover a number of important issues related to Japanese Social Policy. These issues include, among other, Japanese pension system, public health insurance in Japan, evolution of social policy in Japan, role of families, communities and government in improving socioeconomic performance, and Japanese policies towards poverty & public assistance. We hope that these papers will provide important policy findings and lessons for policymakers in developing countries and other development stakeholders. Tsutomu Shibata Advisor & Program Manager of the Brain Trust Program World Bank Institute v 3 Social Expenditure and Economic Growth: Sharing Growth in a Japanese Way Eiji Tajika and Yuji Yui Hitotsubashi University and Seijo University It has now been 30 years that Japan started going through its high economic growth period. During this time the country attained equitable income distribution and a rapid improvement in such social indicators as life expectancy and infant mortality rate. The Japanese economy during this period was characterized not only by its high growth rate, but also by a low ratio of social expenditures to gross domestic product (GDP). The scene changed in the mid-1970s, however. Since then, a low growth rate and high social expenditures have come to characterize Japan. This paper focuses on Japan's high growth period and seeks to show how the country managed to improve social welfare while controlling the expansion of social expenditure. We examine three types of public policies which may have contributed to this outcome. The first is the balanced budget policy, which restrained government activities and helped control inflation. Second, we discuss the role of tax policy. A low tax burden and successive cuts in personal income taxation enabled Japan to distribute the outcome of economic growth to the public and thus promote social welfare. And on another level, the reduction of corporation income tax through generous depreciation allowances and tax-free reserves increased the firms' internal funds, which contributed greatly to promoting investment and consequently securing stable employment. Finally, we look at the role of local governments. Most social expenditure is carried out by local governments in Japan, and the central-local government financial relationship has been effective in distributing services uniformly among local districts. The Nature of Japanese Social Expenditure in the Postwar Period This section surveys Japanese social expenditure in the postwar period with a view to highlighting distinctive patterns and trends. After describing trends in social indicators in the postwar period, we classify the period into three subintervals. The first period is the reconstruction stage of the Japanese economy between the end of World War II and the mid-1950s. The second period is a high economic growth stage between the latter half of the 1950s and the mid-1970s. The third period is the low and stable economic growth stage that started in the late 1970s. We would like to express our gratitude for the many comments we received from the participants of meetings in Tokyo, San Francisco and Cheju island. 1 2 Eiji Tajika and Yuji Yui Developments with regard to Social Indicators in the Postwar Period Japanese now enjoy much higher levels of income and longer lives compared with many other countries in the world. Japan is also said to be a socially developed and stable country. How did it get to this point? Table 1 looks at some of the main social indicators and shows their trends in the postwar period. If we look at the absolute population trend at first, we see that the population has grown by 50 percent in the 45 years between 1950 and 1995. However, the natural rate of population increase has decreased from 17.2 per million to 2.4 per million in the same period, and population growth is estimated to go down to less than zero about the year 2010. Table 1. Development of Social Indicators after World War II Year 1947 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 Population Total population (thousand) 78,101 83,200 89,276 93,419 98,275 103,720 111,940 117,060 121,049 123,612 125,570 Structure by age group (percent) 65 and over (thousand) -- 4,109 4,747 5,350 6,181 7,331 8,865 10,647 12,468 14,895 18,597 Share(percent) 4.1 4.7 5.3 5.7 6.3 7.1 7.9 9.1 10.3 12.1 14.6 Age under 15 (thousand) -- 29,428 29,798 28,067 25,166 24,823 27,221 27,507 26,033 22,486 19,957 Share(percent) 36.6 35.4 33.4 30.2 25.7 24.0 24.3 23.5 21.5 18.5 16.0 Natural increase rate (per 1,000 pop.) 19.7 17.2 11.6 9.6 11.4 11.8 10.8 7.3 5.6 3.3 2.4 Health Life expectancy at birth Male 50.1 59.6 63.6 65.3 67.7 69.3 71.7 73.4 74.8 75.9 76.4 Female 54.0 63.0 67.8 70.2 72.9 74.7 76.9 78.8 80.5 81.8 82.8 Infant mortality (per 1,000 live births) 76.7 60.1 39.8 30.7 18.5 13.1 10.0 7.5 5.5 4.6 4.2a Total fertility rate 4.54 3.65 2.37 2.00 2.14 2.13 1.91 1.75 1.76 1.53 1.42 Education (enrollment ratio: percent) To high schoolb Male - 48.0 55.5 59.6 71.7 81.6 91.0 93.1 92.8 93.2 94.7 Female - 36.7 47.4 55.9 69.6 82.7 93.0 95.4 94.9 95.6 97.0 To university and junior collegec Male -- -- 15.0 14.9 22.4 29.2 43.0 41.3 40.6 35.2 42.9 Female -- -- 5.0 5.5 11.3 17.7 32.4 33.3 34.5 37.4 47.6 (continues on next page.) Social Expenditure and Economic Growth 3 Year 1947 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 Dwellingd Number of dwellings (thousand) -- 6,896 17,432 20,372 24,198 28,731 32,188 34,705 37,413 40,773 -- Average amount of floor space per dwellings (m2) -- -- -- 72.5 73.9 77.1 80.3 85.9 89.3 91.9 -- Employment Average yearly hours worked (Manufacturing industry) -- -- 2,376 2,484 2,302 2,249 2,014 2,138 2,156 2,119 1,986 Income Per capita GDP (in U.S. $) -- -- 261 477 932 1,964 4,481 9,069 11,124 24,042 40,912 Ratio to U.S. per capita GDP -- -- 0.11 0.17 0.26 0.40 0.61 0.76 0.66 1.08 1.53 --. Not available. a. The number refers to 1994 data. b. The ratio of students who advanced to upper secondary schools or technical colleges over lower secondary school graduates. c. The ratio of students who advanced to universities and junior colleges over graduates of lower secondary school 3 years before. d. The housing survey has been carried out every five years since 1953. Sources:Statistics Bureau, Historical Statistics of Japan, and Japan Statistical Year Book, various editions. OECD, National Accounts: Main Aggregates, 1960-1996. 1998 Economic Planning Agency, Report on National Accounts from 1955 to 1969. 1988 The makeup of the population has also changed drastically in this period. The share of people aged 65 and over has grown to about three times the original value, from 4.7 percent to 14.6 percent, owing to the extension of the average life span and the decrease in the fertility rate. Life expectancy for males has gotten longer by 16.8 years and for females by 19.8 years in this 45-year period. However, these increases should be evaluated based on the fact that such indicators as life expectancy and infant mortality rate were at relatively high levels in 1950; the situation immediately after the war was even worse. As for education, basic education was already available throughout the country in the Meiji era. However, it is notable that enrollment in higher education has increased in the postwar period, owing, in part, to the drastic increase in the number of females entering college. Although such indexes as those on dwellings and working hours do not show substantial improvements, it is striking how rapidly many social indicators improved during the high growth period. In the next section, we investigate the trend of social spending that caused these changes. The Reconstruction Stage of the Japanese Economy: 1945­54 World War II brought Japan a huge loss of human resources and national wealth. A quarter of the national wealth was lost in the war, and one-third of the country's factories and equipment was destroyed. The 4 Eiji Tajika and Yuji Yui level of industrial production in 1946 was equal to one-fifth of the prewar level.1 More than 7 million soldiers were demobilized, 1.5 million people returned to Japan from abroad, and about 4 million workers who worked in war-related industries lost their jobs. The imminent mass starvation of 10 million people was a concern at the end of the war. Understandably, therefore, the priority of economic policies was given to the recovery of the devastated economy, and the application of social expenditure at that time was limited to the prevention of epidemics and support for the extremely poor. The main economic policy--called "Keisha-Seisan" in Japanese, implying a "key-industries-oriented" plan--aimed to restore the production level of the economy and to offer job opportunities to the millions of unemployed people. The government planned to direct scarce resources into key industries such as coal, steel, fertilizers, and a few others. These investments were financed by the Reconstruction Bank, which issued bonds, which the Bank of Japan then bought directly. The money supply was thus increased. The application of this policy was successful in restoring the level of production but it also accelerated inflation. The overall inflation rate based on wholesale prices was nearly 200 percent in 1947. The occupation army invited Mr. Joseph Dodge (President of the Bank of Detroit) as an adviser for economic affairs to restore economic stability promptly. To reduce hyperinflation, Mr. Dodge proposed a comprehensive plan in 1949. Balancing the budget was set as the first priority. Not only general and special accounts of the government but also the budgets of state-owned enterprises had to be balanced. The consolidated budget was balanced under mandatory measures, and the money supply was cut sharply. The Dodge plan was thus comprehensive and drastic. Such severe contraction policies would never have been possible if Japan had not been under the control of a foreign power. The government shrank considerably thanks to this policy. Social Expenditure during the High Growth Period: 1955­73 The Japanese economy restored its prewar production level between 1950 and 1952. Real GDP in 1952 exceeded the 1935 level for the first time since the war. The Annual Economic Report by the Economic Planning Agency declared in 1956 that the restoration process was completed and Japan was stepping into a new growth era. In 1961 the National Pension Scheme and the National Health Insurance System were established, which meant that everyone was to be covered by public pension plans and public health insurance systems. Figure 1 shows Japan's social expenditure as a percentage of the GDP between 1960 and 1981. Social expenditure is defined here as direct public expenditures on education, health services, pension plans, unemployment compensation, and other income-maintenance programs and welfare services.2 The figure also shows the ratio of total government expenditure to the GDP. The figure allows us to observe the following: · Japanese social expenditure as a percentage of nominal GDP had been about 8 to 9 percent in the 1960s, and it started to rise in the mid-1970s. The ratio exceeded 10 percent in 1972, 12 percent in 1974, and 17 percent in the early 1980s. · Education spending accounted for half of total social expenditures in the early 1960s. It was 4 to 5 percent of the GDP throughout the rest of this period. The expenditure for health services was second in size and pensions were third during the early 1960s. Both increased gradually and they 1 Ministry of Finance (1978: 14­15), and Kouzai (1981). 2 In 1980 the OECD initiated a review of member countries' social expenditures in response to a conflict between economic and social objectives, which was getting more serious during the periods of slow economic growth and high inflation caused by the two oil shocks of 1973­74 and 1979­80. The OECD published social expenditure statistics in 1985, which we can use to examine the characteristics of Japanese social expenditure (OECD 1985). Social Expenditure and Economic Growth 5 were almost the same in 1981. The rate of increase in pensions was extremely high during the 1970s. Expenditure for unemployment compensation had been 0.3­0.4 percent in the 1960s, but it rose to 0.5 percent in the mid-1970s, reflecting the period's recession. · The trend of total government expenditure was similar to that of social expenditure. The ratio to the GDP was around 10 percent in the 1960s, and then it rose throughout the 1970s. The share of social expenditure in total government expenditure, therefore, had been stable at about 47­49 percent in the 1960s and early 1970s--except in the late 1960s when it went up to 50­52 percent. Figure 1. Social Expenditure as a percentage of GDP: OECD Data Note: Left-hand scale is a GDP ratio of social expenditure and right-hand scale is a GDP ratio of total government expenditure. Source: OECD (1985) Figure 1 implies that the course of social expenditure changed drastically in the mid-1970s. Table 2 shows average real growth rates of social expenditure and GDP in two subperiods: the first is from 1960 to 1973, and the second is from 1974 to 1981. The growth rate of social expenditure was a little lower than that of the GDP in the first period. In the second, the GDP growth rate was reduced by almost half, and the growth rate of social expenditure decreased a little but stayed at a relatively high level. The rise of the GDP ratio after the mid-1970s was explained by the declining economic growth rate. Table 2. Real Growth Rate of Social Expenditure and the GDP (percent) Period Social expenditure GDP 1960­73 9.3 9.9 1974­81 7.6 4.5 Note: These numbers are average real growth rates in each period. Source: OECD (1985). 6 Eiji Tajika and Yuji Yui Japanese social indicators improved during the high growth period, but social expenditure stayed at a relatively low level, with the exception of education. Friendly labor-management relations and the equalization of income distribution may have contributed to this. Social Expenditure during the Low Growth Period: 1974 and After This section explores the changing pattern of social policy throughout the low growth period. Table 3 shows the social expenditure in the National Accounts as a share of the GDP from 1970 to 1994. It covers the social expenditure in a broad sense including expenditures for housing and community amenities. From this table, we can see that expenditure for social security and welfare services rose steadily in the 1980s. Although expenditure for pensions is not shown separately, it accounts for most of the growth of spending for social security and welfare services. We can also see that housing had been stable since the mid 1970s and 1980s, but in the early 1990s it increased a little. Table 3. Social Expenditure in the National Accounts: Selected Years, as a Ratio of Nominal GDP (percent) (A) (B) (C) (D) (E) (F) Education Health Social security Housing and Social Social expenditure and welfare community expenditure (includes housing) Year services amenities (A)+(B)+(C) (D)+(E) 1970 3.5 3.0 2.7 1.4 9.2 10.6 1975 4.8 4.0 5.2 2.1 13.9 16.0 1980 4.8 4.5 7.0 2.4 16.3 18.7 1985 4.1 4.7 7.7 2.0 16.6 18.6 1990 3.6 4.6 7.7 2.1 16.0 18.1 1994 3.8 5.3 9.2 2.9 18.3 21.2 Source: Economic Planning Agency (1996). Figure 2 shows the relationship between social expenditure and economic growth. The figure includes three series of social expenditure, namely an Organisation of Economic Co-operation and Development (OECD) series and two series provided by the Economic Planing Agency (EPA): one includes expenditures for housing and community services, and the other excludes it. We can confirm that there is a similar upward trend between social spending provided by the OECD and the other one by the EPA, which is defined in a narrow sense. The figure indicates that social expenditure remained extremely low in the high growth period. As for the slowdown in economic growth during the late 1970s and 1980s, almost all industrialized countries including OECD members suffered from an economic recession and high inflation caused by two oil shocks. Most of the OECD member countries reduced the growth rate of social expenditure drastically in response to recession. The social expenditure ratio increased by only 1.7 percentage points from 23.1 percent in 1975 to 24.8 percent in 1981 in these countries. By contrast, the Japanese GDP ratio increased by 3.3 percentage points from 14.2 percent in 1975 to 17.5 percent in 1981 (figure 1). The expansion of social expenditure after the mid-1970s can be explained partly from the viewpoint of political economy. High economic growth since the mid-1950s improved the standard of living rapidly, but at the same time it caused serious problems such as environmental pollution. In the late 1960s people Social Expenditure and Economic Growth 7 started shifting their major concern from increasing their income to improving their quality of life. The government also shifted its policy objectives in economic plans from attaining maximum growth and full employment to correcting distortion produced by high growth.3 Figure 2. Social Expenditure and Economic Growth: OECD and EPA Series Notes: Social Expenditures are expressed as a ratio of the GDP. Social expenditure of OECD statistics is in calendar-year terms, and other data are in fiscal-year terms. Source: OECD (1985), Economic Planning Agency (1996). Other contributing factors include inflation, demographics, and the level of real benefits. Three Aspects of Public Policy and Social Expenditure The three aspects we explore in this section are the balanced budget principle, tax policy, and the local government's role in delivering social services during the high growth period. The Balanced Budget Principle during the High Growth Period The Public Finance Law (passed in 1947) prohibited the general accounts of the government from issuing public debt in principle.4 The exception was the financing of public investment, but there was no issuance of public debt to finance expenditure until fiscal 1965.5 3 See Economic Planing Agency (1965). 8 Eiji Tajika and Yuji Yui Although the general account did not issue public debt, this does not mean that there was no pressure to do so. Rather, the ruling party tried in almost every year to issue public debt to expand public spending. The Ministry of Finance strongly resisted such attempts, because it is very difficult politically to reduce a budget once it has been expanded.6 As we will see later, the government took a large portion of the tax revenue increase brought in by the high economic growth and appropriated it for tax cuts, curbing the expansion of the budget. However, such government resistance, especially by the Ministry of Finance, weakened in the early 1960s. The turning point of public debt policy was the supplementary budget in 1965, when tax revenue fell sharply because of the recession. In the 1966 budget, 15 percent of the public expenditure was financed through public debts. However, as Japanese economy returned to its growth path, the amount of public debt issued decreased in the late 1960s. The budget ran a huge deficit again in 1975, and since then the public debt has continued to climb. Two oil shocks in the 1970s and the soaring value of Japanese currency against the U.S. dollar were reasons, among others, that called for more public spending to boost the economy. Figure 3 shows the general government's (in other words, combining the three sectors--central and local governments, and the social security fund) Saving-Investment Gap in the National Accounts. One can see that there was a surplus until 1974 (with a few exceptions), turning into a huge deficit thereafter. Figure 3. Savings-Investment Gap of General Government: Ratio of GDP Source: Economic Planning Agency (1996). In the background of these developments, social security funds have accumulated a surplus, but this will not be a stable financial source, for the country is aging rapidly and the funds will never be accumulated with the same speed and magnitude as before. Here we investigate two major implications of the balanced budget principle. The first implication is that sticking to principle has made the saving-investment gap of the general government positive and kept the government from being a source of inflation. During the high growth period, private firms had strong 4 This legislation was enacted to enable Japan to develop securely as a disarmed nation, because the government financed most of the war-related expenses by issuing public debt. See Ministry of Finance (1973). 5 The law did not prohibit special accounts and public enterprises such as Japan National Railroad and Nippon Telephone and Telegram to issue debt. These public enterprises began to issue government-guaranteed bonds in 1953. 6 Ministry of Finance (1997). Social Expenditure and Economic Growth 9 incentives to borrow to finance investment. The central bank supplied currency through lending to commercial banks, and there was always excess demand in the money market. Under these financial circumstances, restricting the issuance of public debts contributed to the mitigation of inflation acceleration. The second implication of the balanced budget principle is that it has contributed to restraining the expansion of budget size. As table 4 shows, the Japanese budget was small in relation to the GDP in its early phase of growth, although it expanded significantly in the 1970s and thereafter. We employ two measures here to look into the size of the government: the first is the budget of the General Government in National Accounts, which consists of the central and local governments, and social security funds; and the second is the General Account of National Government, which is the budget of the central government. Each measure saw a turning point in the mid-1970s when the government started to claim more share in the GDP than it did in the period of higher growth. Table 4. Size of Government Expenditure as a Ratio of the GDP, Selected Years (percent) General Account General of National Ratio of FILP to Fiscal year government Government FILP General Accounts 1955 18.0 11.8 3.5 29.2 1960 15.5 10.4 3.7 35.9 1965 17.4 11.0 5.3 47.7 1970 16.8 10.9 5.0 46.4 1975 23.1 13.7 6.9 50.6 1980 26.1 17.7 7.4 41.7 1985 25.3 16.3 6.3 38.7 1990 24.8 15.8 8.2 51.7 1994 28.6 15.4 10.5 68.4 Note: FILP stands for the Fiscal Investment and Loan Program. Expenditure of the General Government includes General Government Final Consumption, Government Fixed Capital Formation, and Social Security Transfer. Source: Economic Planning Agency (1996); Ministry of Finance (1996). Table 4 shows also the size of the Fiscal Investment and Loan Program (FILP). The FILP is a public investment scheme that collects investment resources mostly from postal savings and public pension funds, and invests basically in infrastructure through public enterprises and public financial organizations such as Japan Development Bank and Japan Housing Bank. Interest rates for some investments and lending are set deliberately lower than the market-clearing rate, thus subsidizing certain investments. The FILP grew faster than government expenditure throughout the high growth period. Private investment increased so rapidly that bottlenecks were encountered in the country's infrastructure (roads, harbors, and so on). The government was forced to increase public investments to eliminate these bottlenecks. But under the balanced budget principle, there was a restriction on government expansions of 10 Eiji Tajika and Yuji Yui investment in the general account. The government utilized FILP to satisfy the demand while maintaining the balance of the general account. The national government could use FILP with discretion by taking advantage of the off-budget account system. Table 5 shows that the most of the FILP funds have gone into infrastructure--such as water supply, sewage, transportation, and communication. Loans to individual housing have also been one of the program's most important uses of its funds. A positive side of the program, thus, has been its supplementing and strengthening of on-budget investment in social infrastructure. Table 5. The Fiscal Investment and Loan Program, Classified by Function (percent) Year 1955 1960 1965 1970 1975 1980 1985 1990 1995 (1) Housing 13.8 12.6 13.9 19.3 21.4 26.2 25.4 30.3 35.3 (2) Water supply, sewers, and 7.7 9.1 12.4 11.6 16.7 14.1 15.7 15.3 16.4 other environmental facilities (3) Welfare 2.1 1.7 3.6 2.8 3.4 3.5 2.9 3.1 4.0 (4) Education 4.5 3.4 3.1 2.2 3.0 4.4 3.6 2.0 2.0 (5) Small and medium-sized 8.1 12.5 12.6 15.4 15.6 18.7 18.0 15.7 15.3 enterprises (6) Agriculture, forestry, and 8.9 7.0 7.2 5.0 4.1 4.9 4.3 3.1 3.0 fisheries Subtotal: (1) to (6) 45.1 46.5 52.8 56.4 64.2 71.8 69.8 69.5 76.0 (7) Land conservation and disaster 7.7 6.4 3.1 1.6 1.2 1.7 2.3 1.2 1.3 reconstruction (8) Roads 3.7 4.4 7.9 8.6 8.0 5.7 8.8 9.8 7.7 (9) Transportation and 12.2 14.6 13.9 13.2 12.7 9.6 8.5 8.3 4.6 communication (10) Improvement of developing 8.5 7.0 7.0 4.0 3.3 2.6 2.5 2.5 2.6 regions Subtotal: (7) to (10) 32.1 32.4 31.9 27.4 25.2 19.6 21.9 21.8 16.2 (11) Industries and technology 15.8 13.4 7.8 5.7 3.0 3.0 2.9 2.9 3.1 (12) Trade and economic 7.0 7.8 7.5 10.6 7.7 5.6 5.3 5.8 4.7 cooperation Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Note: In addition to the fields above, there is a fund operation service. Source: Ministry of Finance (Various years). The FILP has produced negative effects on the economy as well, however, in the process of raising funds and allocating spending. The isolation of its fiscal management (regarding both revenue and expenditure) from market activities has engendered inefficiencies in the economy. Social Expenditure and Economic Growth 11 In financing the FILP, postal savings had the largest share until 1980. Postal savings absorbed the sizable domestic savings and were in an advantageous position in the competition with private financial institutions. This dominance of postal savings has had an oppressive effect on the financial market's flourishing, efficient private institutions. In spending, the FILP caused overinvestment in housing, highways, bridges, and so forth--especially following the high growth period. Public investment opportunities with high expected rates have decreased since the mid-1970s. The FILP has continued to direct many public investment projects with low economic rates of return, thanks to its revenue-raising advantage. Anecdotes of overinvestment abound : three toll bridges over the Japanese inland sea, called Seto Naikai, and local highways that can never recoup the costs thrown into them. Tax Policy during the High Growth Period The focus of the argument here is finding a link between tax policy and social welfare. The issue of concern is the extent of the trickle-down effect from national income growth to household welfare. PERSONAL INCOME TAX. The most important channel through which the growth in national income was distributed to Japanese households was by cutting taxes. As table 6 shows, Japanese tax burden as measured by its proportion to the GDP was lower during the high growth period than in the later years. And it has been much lower than that of member countries of the OECD.7 However, it was higher in this growth period than in the prewar period,8 and people felt the burden much more in the high growth period because before the war the level of income was so low that most of one's income had to be allocated to such necessities of life as food. Moreover, high economic growth was accompanied by inflation,9 and some adjustment was needed for tax increases caused by nominal increases in income. The first priority in personal tax policy, therefore, was in cutting taxes during the high growth period. During the 1960s, the high growth of the economy had brought much larger tax revenues (called a "natural increase") to the fiscal authority than the analysts' estimates in the initial budget. This "natural increase" was utilized to adjust the nominal increase caused by inflation. The government's Tax Council published a report in 1964 saying that at least 20 percent of the "natural increase" had to be allocated to cutting taxes.10 Figure 4 shows "natural increases" in the personal income tax and the national tax total, with the amount of the tax cut in each year. As the figure illustrates, throughout the high growth period, the reduction of taxes on personal income continued, except for 1960 when a huge typhoon hit the country and public works subsequently consumed the resources set aside for the tax cut. In the late 1950s, the personal income tax cut was greater than the "natural increase." The tax cuts were implemented by both raising the minimum taxable income and cutting tax rates. However, the manipulation of minimum taxable income was used as a policy tool more often than for revising the tax rate structure. A substantial revision the of tax rate structure was undertaken only twice between 1955 and 1973. This is because raising the minimum taxable income was thought to be more effective in alleviating the tax burden of the low-income class. Although effectiveness of such a tax reduction policy depends on the level of income earned, a simulation study11 shows that average tax rates on personal income, irrespective of income levels, decreased from 1955 to 1975, rose in the early 1980s, 7 However, along with the increase in social security, the tax burden expanded enormously in the 1970s, especially toward the end of decade. 8 If we use the ratio of total amount of national and local tax over national income as the index of tax burden, this index was about 20­ 22 percent in the mid-1950s, much higher than the prewar level of 13 percent (Ministry of Finance 1990). 9 Average growth rate of nominal GDP was 16.3 percent, and average rate of inflation was 6.0 percent during the period from 1960 to 1970. 10 Ministry of Finance (1990). 11 Tajika and Hayashi (1995). 12 Eiji Tajika and Yuji Yui and thereafter decreased again. Japanese tax policy during the high growth period was effective in alleviating income tax on households rather significantly over a wide income range. Table 6. Tax Revenues as a Percentage of the GDP: Selected OECD Countries and Selected Years 1965 1970 1975 1980 1985 1990 1994 Japan Tax revenue 14.3 15.3 14.9 18.0 19.3 22.2 18.1 Social security 4.0 4.4 6.0 7.4 8.3 9.1 9.7 contribution Total 18.3 19.7 20.9 25.4 27.6 31.3 27.8 United Tax revenue 21.1 23.0 21.3 21.0 19.5 19.8 20.6 States Social security 3.2 4.4 5.4 5.9 6.5 6.9 7.0 contribution Total 24.3 27.4 26.7 26.9 26.0 26.7 27.6 United Tax revenue 25.7 31.8 29.4 29.4 31.2 30.2 27.9 Kingdom Social security 4.7 5.1 6.1 5.9 6.7 6.2 6.2 contribution Total 30.4 36.9 35.5 35.3 37.9 36.4 34.1 France Tax revenue 22.7 22.3 21.9 23.9 25.2 24.4 24.9 Social security 11.8 12.8 15.0 17.8 19.3 19.3 19.2 contribution Total 34.5 35.1 36.9 41.7 44.5 43.7 44.1 Germany Tax revenue 23.1 22.9 23.8 25.1 24.2 22.9 23.9 Social security 8.5 10.0 12.2 13.1 13.9 13.8 15.4 contribution Total 31.6 32.9 36.0 38.2 38.1 36.7 39.3 OECD Tax revenue 21.5 23.8 25.1 26.0 27.2 28.0 28.2 average Social security 4.8 5.7 7.4 8.1 8.7 9.0 10.2 contribution Total 26.3 29.5 32.5 34.1 35.9 37.0 38.4 Note: The first row for each country shows tax revenue (excluding social security contribution), the second one shows the social security contribution, and the third one shows total tax revenue as a percentage of the GDP. Source: OECD (1996). CORPORATION INCOME TAX. The objective of corporation income tax policy had been to stimulate capital accumulation since the war, and the policy in the high growth period was continuously geared toward reducing the tax burden. The movement of the statutory corporate-income tax rate is shown in figure 5. The corporation tax rate had been lowered throughout the late 1950s and 1960s. In the late 1960s, the firms and the government were deeply concerned about the consequences of opening up the economy, and this facilitated the cut in the tax rate further. Social Expenditure and Economic Growth 13 Figure 4. Income Tax Cut during the High Economic Growth Period Source: Ministry of Finance. Kuni no Yosan (National Government Budget), various years. Figure 5. Corporation Income Tax Rate Source: Ministry of Finance Statistics Monthly, various editions, and Authors' estimates. 14 Eiji Tajika and Yuji Yui However, cutting the corporate income tax was not the only measure used by the government. In terms of strengthening the firms' investment capacity, increasing their cash flows would be as important as increasing their income. In this respect, tax-free reserves and accelerated depreciation were the two most noteworthy policies in postwar Japan that made corporate cash flows thicker and their long-term investment easier. According to the Tax Council's estimation, firms could increase their cash flow by more than 10 percent through such special tax concessions in the early 1960s.12 Increases in cash flow created by generous tax policy contributed a great deal to promoting investments, because there were plenty of investment opportunities, and the firms had until then suffered from a lack of investable funds. It also strengthened the firms' power to resist the deterioration in economic conditions. We see that the series of policies that lessened corporate tax liabilities eventually stabilized the economy and increased employment. That is, cutting corporate taxes was another route by which the outcome of growth was carried to individual households. However, an important series of events occurred in the middle of the 1970s when the corporate tax rate was raised. The increase in public expenditure--which had been made necessary by offering better social security and boosting the economy by public investment--started to change the course of tax policy in Japan. Since the increase, the effective Japanese corporate income tax rate--incorporating national (central) government as well as local taxes--has been about 50 percent, which was probably the highest in the world throughout the 1980s and the first half of the 1990s. The Local Government's Role in Delivering Social Services In the relationship between Japan's national (central) government and local governments, gross local government expenditure has been a little larger than that of the national government. However, the national government provides a transfer of funds to the local government, which amounts to nearly 40 percent of the local government's total expenditure. Therefore local governments have provided 60­70 percent of total government expenditure after netting out intergovernmental transfer. This kind of relationship was built after the war and was maintained throughout the high economic growth period. Major social services such as education and public health are provided through the activities of local governments in Japan. For example, table 7 shows the funding structure of public education. Although the funding is split between the national government and local governments nearly equally, most of the national funding is used as grants, and the direct funding source of the central government for public education activities is only 10 percent of total funding. Furthermore, the share of local funding is increasing in recent local public finance. The same reasoning is applied to social welfare and public health activities. The national government finances 60 to 80 percent of those expenditures (except for pension plans and medical services); however, most of the national funding is used as grants. Pensions and medical services are provided as social insurance, and each local government (cities, towns, and villages) manages the National Pension System and National Health Insurance as an insurer.13 Thus, we can say that the provision of most of the services is the responsibility of local governments. As shown in figure 1, the expenditure for education had the largest share in social expenditure in the high growth period. In the following argument we will examine the delivery system of education in local government activities. One can say that the Japanese national-local government financial relationship has been effective in providing education uniformly among local districts. The emphasis on distributional goals in the 12 This estimation was derived from a sample survey by the Tax Council, and magnitude differed by firm. See Tajika and Yui (1996). 13 For an outline of the Japanese public pension system and public medical insurance, see Tajika, The Public Pension System in Japan: The Consequences of Rapid Expansion, and Fukawa, Public Health Insurance in Japan, in the World Bank Institute (WBI) Working Paper Series. Social Expenditure and Economic Growth 15 provision of social services, however, is likely to cause local governments to depend excessively on the national government policy, which leads to the inefficient allocation of services from a local autonomy viewpoint. Table 7. Financial Sources of Public Expenditure for Education (percent) Sources of funds 1950 1955 1960 1965 1970 1975 1980 1985 1990 Central government 51.6 46.7 47.8 50.0 47.1 46.5 47.3 43.4 42.1 Direct expenditure 10.2 11.2 11.6 14.6 13.4 12.2 13.7 13.5 13.5 Subsidies to local 41.4 35.5 36.2 35.4 33.7 34.3 33.6 29.9 28.6 governments Subsidies 3.7 22.8 22.5 21.4 19.6 20.6 20.2 17.9 15.5 Fund through the fiscal 37.7 12.7 13.7 14.0 14.1 13.7 13.4 12.0 13.1 equalization system Local governments 48.4 53.3 52.2 50.0 52.9 53.5 52.7 56.6 57.9 Prefecture 27.6 30.9 30.8 30.0 30.1 30.2 28.1 31.2 31.8 Municipal 20.8 22.3 21.4 20.0 22.8 23.3 24.6 25.4 26.1 Public expenditure for 160 372 613 1,385 2,883 8,119 14,006 16,568 20,258 education (billion yen) Note: Central government subsidy includes direct transfer of funds to local government as well as transfer of funds through the Fiscal Equalization system. Source: Ministry of Education. Various years. We will discuss first the provision of compulsory education. The general outline of compulsory education in postwar Japan consists of six years of elementary school and three years of junior high school. The basic law on school education stipulates the establishment and management of both elementary and junior high schools by local governments. Ninety-nine percent of elementary school students and 95 percent of junior high school students enroll in schools managed by local governments. Officially, local governments have discretion regarding the content of compulsory education. However, a school board in each locality has to choose textbooks out of those approved by the Ministry of Education. Moreover, the national government makes five-year plans that determine such important regulations as the number of students allowed per classroom and of students per teacher. The national government then provides funds for local governments to guarantee the uniform provision of compulsory education throughout the country. There are two channels of financial assistance from the national government to local governments to subsidize the management of compulsory education. The first is a payment of specific-purpose grants. For example, the national government subsidizes 50 percent of staff salaries in compulsory education. It also subsidizes 50 percent of school facilities and equipment. The second channel works through the local allocation tax system, whose purpose is to equalize financial capacity among local governments. The national government prescribes the standard level of local service and then calculates the necessary expenditure for each local government to provide standard educational services (defined as part of the area's "basic financial needs"). The national government guarantees the funds by allocating tax grants to the local government whose tax revenue falls short of its basic financial needs. The local allocation tax grant is an important source of funds for local governments, constituting about 20 percent of their total revenue.14 14 The uses of the local allocation tax grant are not specified. This distinguishes it from a specific-purpose grant. 16 Eiji Tajika and Yuji Yui Thus, services considered to be part of the "basic financial needs" are to be provided uniformly by local governments. Expenditures for compulsory education are 22.1 percent of the basic financial needs at the prefecture level and 13.0 percent at the municipal level according to recent data.15 Table 8 shows coefficients of variation per capita revenues and expenditures of local governments in selected years. The table shows that per capita tax revenue has the largest disparity among localities. This is because it reflects regional economic levels directly. Although local governments in Japan can have discretion on tax rates within certain permitted levels, the rates and tax bases are actually uniform among localities. The level of coefficients decreased from the 1960s to the 1980s, and started to increase in the rest of the "low and stable economic growth" period, reflecting the change in regional income distribution. The regional disparity of tax revenues is standardized through the financial equalization system. The coefficient of variation of total revenue is much lower than that of local tax revenue. Table 8. Local Revenue and Expenditure, Coefficients of Variation Revenue Expenditure Welfare Local Total Total Public and Welfare Labor Public Year Tax Revenue Expenditure Investment Education Labor service service health 1956 33.7 16.3 17.3 38.7 8.6 31.6 -- -- 28.7 1960 42.1 17.0 17.0 43.9 8.8 28.5 -- -- 31.9 1965 41.9 14.1 14.5 24.9 11.2 32.9 31.4 54.8 22.2 1970 38.8 11.8 12.3 17.5 8.9 32.5 30.1 60.4 16.2 1975 31.3 15.6 16.0 16.2 10.1 27.5 25.8 63.4 15.7 1980 27.8 16.8 17.1 19.5 9.2 24.9 23.9 56.8 14.3 1985 28.8 16.9 17.2 17.5 9.4 26.0 25.8 52.6 13.6 1990 32.7 16.2 16.7 16.2 10.1 23.4 24.0 44.8 12.6 --. Not available. Note: This table shows coefficients of variation per head in local expenditure and revenue; "local" refers to the total of prefectures and municipalities. Source: Calculated from Ministry of Home Affairs, Local Public Finance Statistics, various years. On the spending side, one can see that the coefficient of variation of educational expenditure is the smallest among various categories. The high coefficients of public health and welfare reveal that there is wide disparity in expenditure levels among local districts, but the overall level has been decreasing throughout the postwar period. The low coefficient of variation of education service implies that per capita local spending for education is more evenly distributed than other categories of expenditures. We can demonstrate this in another fashion too. Figure 6 depicts per capita local tax revenue and educational expenditure of prefectures in 1960. Revenue and expenditure are both normalized by setting the national average to 1. There were 46 prefectures in 1960 and they are set in the order of their per capita local tax revenue in the figure. Tokyo has the highest per capita tax revenue, collecting 2.5 times as much as the national average. The lowest is the Kagoshima Prefecture, which collects 50 percent of the national average. Since the volume of per capita local tax revenue reflects regional economic levels, there are significant differences in regional income distribution. 15 Ministry of Home Affairs (1996). Social Expenditure and Economic Growth 17 From another perspective, though, as shown by the dotted line representing per capita education expenditure, most local governments remain within 15 percent of the national average. This shows that compulsory education service is provided evenly. We can see a similar tendency for the years up to the 1990s. Figure 7 shows changing patterns of per capita local tax revenue and per capita education expenditure every five years between 1956 and 1990. First we group local governments into six sets based on their level of per capita local tax revenue, which is a proxy of regional income level in each year. We count Tokyo as one group (G0 corresponds to Tokyo). The highest and lowest groups (in terms of per capita local tax revenue) are designated by G1 and G5, respectively. Each group contains nine prefectures. Then we calculate average normalized per capita expenditure for each group. The figure shows the deviation of the normalized educational expenditure of each group from the average. Figure 7. Expenditure for Education Note: This shows the deviation of the per capita expenditure of each group from the average. Source: Authors' estimates. The difference of education expenditures among groups are found to have stayed within a small range--except for Tokyo--throughout the research period. Furthermore, expenditures for education in the lower income groups are shown to be relatively larger. This means that there is a strong distributional policy adjustment to provide educational service in a uniform way. Thus we can say that Japan's central-local government financial system has been effective in providing educational services uniformly between local districts, but has also led to excessive spending in the name of redistribution. The following argument provides an illustration. Figure 8 shows the allocation of per capita local expenditure for public investment.16 Similar to the methodology in figure7, prefectures are classified into six groups according to their ranking of per capita local tax revenue, and to simplify calculations, both per capita expenditure and revenue are normalized. We can observe the following from this figure: 16 Local expenditure for public investment includes expenses for the construction and repair of roads, harbors, bridges, public housing, and so on. 18 Eiji Tajika and Yuji Yui Figure 8. Expenditure for Public Investment Note: See Note in Figure 7. Source: Authors' estimates. First, the disparity of per capita expenditure for public investment is much larger than that of the education expenditure found in figure 7, although it has been reduced recently. Second, in terms of the relationship between levels of public investment and regional income, roughly speaking, the level of relatively high income groups (G1 and G2) are shown to be going up, while the level of relatively low income groups (G4 and G5) are shown to be coming down. However, the watershed of this polarization occurred in the early 1970s. The public investment levels of low-income groups (G4 and G5) were relatively higher than those of high-income groups. This fact indicates that public investments, in substance, have functioned as policy measures to satisfy redistributional objectives since 1975. One does not intrinsically associate per capita public investment spending with per capita income. Therefore, the existence of a redistributional tendency seems to show that the purpose of local public investment is not providing public goods but rather creating jobs and income in regions with relatively low income. Local public investments seem to be utilized as a means of income distribution, and this aspect becomes obvious after the mid-1970s. Conclusion: Lessons from the Japanese Experience Japan achieved high economic growth and rapid improvement of social welfare through the 1960s and early 1970s. Moreover, during this period, social expenditures remained constant relative to rates observed in OECD countries. One reason for this may be that the population structure of Japan was young at that time. The majority of the population belonged to the group contributing to the fund. Moreover, the growth rate of social expenditure was around 9 percent in the high economic growth period. Social expenditure in absolute volume increased almost proportionately with the GDP. However, as we stressed in this paper, public policy played a major role in promoting social welfare in the high growth period. The government orchestrated budgetary policy, tax policy, and the local delivery of social services to contribute toward social development and to distribute the fruits of economic growth to the public. Specifically we examined the role played by such policies as balanced budget policy, personal and corporate income tax policies, and local delivery of education service. Balancing the government budget was strictly disciplined in Japanese public finance. This discipline was necessary for keeping the Social Expenditure and Economic Growth 19 government slim and controlling inflation. Loosening fiscal policy easily resulted in the overheating of the economy and caused inflation. The Fiscal Investment and Loan Program (FILP) was utilized as a device to promote public investment in social infrastructure while balancing the budget. By cutting taxes, the income tax policy distributed the outcome of high economic growth. As for the delivery of social expenditure, basic expenditures such as education were delivered equally across localities, helped by Japan's central-local government financial system. However, things changed significantly at the turn of the 1970s. The harmony of high economic growth and low social expenditure disappeared. The Japanese economy since then has been characterized by the combination of low economic growth and high social expenditures. The economic growth rate was reduced by half after the two oil shocks, and the postwar high growth era came to an end. Instead of reducing social expenditure, there were demands to increase it further. Although the high economic growth period created an affluent society on the one hand, it also brought such serious by-products as environmental problems on the other. People's concern shifted from promoting economic growth to improving the quality of life. Politically, the ruling Liberal Democratic Party was losing its power and populists' platforms were gaining public approval. The diversification of social expenditure and its expansion led people to overlook the cost-benefit ratio of social welfare. Those who insisted on an increase in social expenditure supposed implicitly that high economic growth would be sustained in the future. In only a few years the Pension Plan for the self- employed virtually went bankrupt. Medical care also needed cuts in benefits and an increase in contributions. It was evident that public social welfare policies of the high growth era needed reexamining. Thus we may derive two lessons from postwar Japanese experience. Countries need to adopt a broad perspective in formulating public policies for social development; and they should be aware that the building blocks of social development in one era might be transformed into stumbling blocks by socioeconomic changes in another. References The word processed describes informally reproduced works that may not be commonly available through libraries. Economic Planning Agency. 1965. Chuuki Keizai Keikaku (Medium-Term Economic Plan). Tokyo. In Japanese. ______. 1996. Annual Report of National Accounts. Tokyo. Kouzai, Yutaka. 1981. Kodo Seicho no Jidai (The Era of High Economic Growth). Tokyo: Nihon Hyouron Sha. In Japanese. Ministry of Education. Various years. School Basic Survey. Tokyo. Ministry of Finance. 1973. History of Fiscal and Monetary Policies in Japan, from the End of the War to the Peace Treaty: Government Debt, Vol. 11. Tokyo. In Japanese. ______. 1978. History of Fiscal and Monetary Policies in Japan, from the End of the War and to Peace Treaty: Statistics, Vol. 19. In Japanese. _____. 1990. History of Fiscal and Monetary Policies in Japan, 1952­1973: Tax Policy, 1952­1973, Vol. 6. Tokyo. In Japanese. 20 Eiji Tajika and Yuji Yui _____. 1997. History of Fiscal and Monetary Policies in Japan, 1952­1973: National Debt, Vol. 7. Tokyo. In Japanese. _____. Various years. Zaisei Kinyuu Tokei Geppou (Statistics Monthly). Various editions. Tokyo. In Japanese. _____. Various years. Kuni no Yosan (National Government Budget). Tokyo. Ministry of Home Affairs. 1996. Local Public Finance System and its Management. In Japanese. OECD. 1985. Social Expenditure 1960­1990. Paris. _____. 1996. Revenue Statistics. Paris. Tajika, Eiji, and Fumiko Hayashi. 1995. "Tax Burden and Pensions from the Viewpoint of Lifecycle of Salaried Worker." Processed. In Japanese. Tajika, Eiji, and Yuji Yui. 1996. "Public Policies and Capital Accumulation: Japan at the Dawn of Economic Growth." Hitotsubashi Journal of Economics 37(2): 135­53. Mission of World Bank Institute This mission of WBI is to help World Bank clients and staff acquire new development knowledge and skills through a variety of courses, seminars, and other learning events. It designs programs on topics related to economic and social development for governments, nongovernmental organi- zations, and other stakeholders. The Institute produces and disseminates publications and electronic information products that support these objectives. For information on WBI publications write to: Publications WBI The World Bank 1818 H Street, NW Washington, DC 20433 Tel: 202-473-6349 Fax: 202-522-1492 Visit us on the World Wide Web at: http://www.worldbank.org/wbi SN 37198