Repot rJo. 7292-MA Malaysia The Housing Sector Getting the Incentives Right April 10, 1989 Infrastructure Division Country Department 11 Asia Regional Office FOR OFFICIAL USE ONLY ,C i beCdisclosed without World Bank authorization. . ' . , )1~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~' :1 .. o * - b~~~~~~~~~~~~~~~~~~- 'I~~____________ '''a 3, ! onyi h efrae of- the r ofica 'uis It otnsmy otews I~ ~b dicoe wihu oldBn uhoiain CURRENCY EOUIVALENTS (as of June 1988) Currency Unit - Ringgit (M$) M$1.0 US$0.40 US$1.0 - M$2.50 FISCAL YEAR January 1 - December 31 WEIGHTS AND MEASURES I meter (m) 3.28 feet 1 square meter (sq m) = 10.8 square feet 1 hectare (ha) - 10,000 sq m or 2.471 acres ABBREVIATIONS AND ACRONYMS EPU - Economic Planning Unit FAR - Floor area ratio JPN - National Housing Department LTV - Loan to value ratio MHLG - Ministry of Housing and Local Government NEP - New Economic Policy PLCHP - Public Low Cost Housing Program SLCHP Special Low Cost Housing Program TLP - Treasury Loan Program UPM - Urban Peninsular Malaysia FOR OFFXC1AL USE. ONLY MALASIA THE HOUSING SECTOR: GETTING THE INCENTIVES RIGHT Table of Contents Page No. EXECUTIVE SUMMARY . .............. i 1. INTRODUCJTION .......... ..1. . A. Background of the Study .1... . . . . . . . . . . . . . . . . B. Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . 1 C. The Approach . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2. THE HOUSING SECTOR IN MALAYSIA .... . . . . . 3 A. Demographic Trends .... . . . . . . . . . . . . . . . . . . . . 3 B. The Housing Stock .... . . . . . . . . . . . . . . . . . . . . . 3 C. Affordability of Housing .... . . . . . . . . . . . . . . . . . 6 D. Government Housing Sector Objectives and Programs . . . . . . . . . 8 E. Financial Intermediation in the Housing Sector . . . . . . . . . . 12 3. THE HOUSING MARKET .... . . . . . . . . . . . . . . . . . . . . . . 19 A. Introdluction .... . . . . . . . . . . . . . . . . . . . . . . . 19 B. Housing Production and Investment Levels . . . . . . . . . . . . . 19 C. Housing Prices .27 D. Summary ..37 4. EVALUATION OF HOUSING POLICIES AND PROGRAMS . . . . . . . . . . . . . 38 A. Issues and Questions .38 B. The Present Value Method .39 C. Examining a Representative SLCHP Investment with the Present Value Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 D. Comparing Other Public Programs to the SLCHP . . . . . . . . . . . 47 E. Housing Regulations and the Private Market . . . . . . . . . . . . 56 F. Summary. . .............. 59 5. STANDARDS AND SELECTED REGULATORY PRACTICES . . . . . . . . . . . . . 64 A. Introduction. . . ......... 64 B. Why Land-use is an Important Issue . . . . . . . . . . . . . . . 64 C. Effects of Standards on Low-Income Housing Supply . . . . . . . 65 D. Land-Use Standards of the Special Low Cost Housing Program . . . 70 E. New Standards for the Special Low Cost Housing Program . . . . . 72 F. Key Parameters to be Considered in Revising Standards . . . . . 75 G. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 This document has a re3tricted distribution and may be used by recipients only in the performance of tneir official duties. Its contents may not otherwise be disclosed without World Bank authorization. TABLES IN THE TEXT 2.1 Population and Housing Stock 2.2 New Housing Units Constructed, 1976-90 2.3 Permanency of Housing, 1980 2.4 Historical Affordability of Housing 2.5 Housing Targets and Performance by Program 2.6 Public Development Budget for Housing Programs 2.7 Housing Loans Outstanding 2.8 Present Value of Selected Government Housing Loan Programs 2.9 Comparison of Treasury Housing Loans and Total Housing Finance 3.1 Housing Completions, 1976-86 3.2 Housing Values, Completed Units, and Value of Residential Output 3.3 Housing Prices and Household Incomes in Malaysia 3.4 Housing Prices and Household Incomes in Selected Cities Worldwide 4.1 Summary of Present Value Model Results 4.2 Comparison of Key Public Sector Programs 4.3 Resale Price of Low Cost Units by State 4.4 Variation in Market Prices, Costs and Subsidies by Location 4.5 Counterexample of Program with Flexible Pricing 4.6 Comparison of Public and Private Sector Housing Units 5.1 Typical Building and Land Development Costs 5.2 Summary Land Use of a Residential Scheme Using Special Low Cost Housing Program Guidelines 5.3 Cost of Land Development Implied by Minimum Standards under a Series of Unit Cost Assumptions 5.4 Developer Profitability and Land-use Standards 5.5 Adjusted Standards to Achieve 55% Salable Land FIGURES IN THE TEXT 2.1 Housing Affordability by Income trcentile 2.2 Housing Loans Relative to Value c- New Housing 3.1 Housing Completions, 1976-86 3.2 Housing Investment and GDP (mid-1970s) 3.3 Housing Investment and GDP (early 1980s) 3.4 Consumer Prices, Rents and Housing Values 3.5 Housing Development Process in Malaysia 4.1 Incentives to the Developer 4.2 Incentives to the Purchaser 4.3 Present Value Model Results 4.4 Net Cost-Benefit, Selected Public Programs 4.5 Variation in Prices, Costs and Subsidies, by Location 4.6 Variation in Cost and Profit/Loss, by Location 4.7 Net Cost-Benefit, Public and Private Investments FIGURES IN TEXT (continued) 5.1 Plot Size and Street Area per Household 5.2 Plot Cost Variations for Variou.s Percentages of Salable Land and Various Land Costs 5.3 Set-back Regulations under the Special Los Cost Housing Program 5.4 Relationship between Profit and Salable Area 1. Vstimating the Long-term Price Elasticity of Housing Supply 2. Iresent Value Model: Summary of Input Data 3. Land Use and Design Standards for Low-cost Housing I THE HOUSING SECTOR: GETTING THE INCENTIVES RIGHT EXECUTIVE SUMMARY A. Introduction Background 1. Late in 1985, the Government of Malaysia was confronted with a deteriorating macroeconomic situation, evidenced by falling exports, declining rates of capital formation, and falling GNP. The construction sector, which for a considerable period had been one of the leading sectors in Malaysian economic growth, had experienced a rapid and severe turnaround. After having grown at an average annual rate of 9.7% between 1980 and 1984, value added in the construction sector fell by 8.4% in 1985 and by 14.0% in 1986. In an attempt to deal with the sharp reversal in this sector, the Government decided to implement a Special Low Cost Housing Program (SLCHP). Under the program, some 80,000 units of housing per year (priced at less than M$25,000) were to be built. It was expected that the combined direct and indirect effects of the program would be to raise the rate of GNP growth by as much as 2% per year, contributing importantly to economic recovery. An additional objective of the program was to sharply change the nature of the product being supplied in the :.;.aysian housing market, offering less-expensive housing to enable proportionally more low-income households to purchase new housing units than had been the case for more than a decade. 2. Not long atter the SLCHP was initiated, the World Bank was requested by the Malaysian Government to help assess the Program. This study, which is a response to that request, was subsequently broadened to include all major interventions in the sector. The expanded coverage was thought necessary in order to provide a context for assessing the SLCHP, which has been formulated and implemented in the context of the overall housing market, with all of the accompanying constraints and incentives. It was judged that any attempt to look at the SLCHP as a single program in isolation from other factors affecting the sector might misjudge the programs's effectiveness and possibly even fail to understand the forces determining the program's evolution, problems and achievements. The study therefore recognizes that there is a single market for housing inputs as well as outputs, that most producers are able to build all types of housing units and that the financing 'or various types of housing often comes from the same sources. This is, therefore, a study of the incentives and disincentives that various forms of government intervention create in a major sector and how efficiency and equity are affected as a result. Obiectives of the Study 3. In light of the above, the objectives of this study are to determine: (a) whether the housing supply system in Malaysia is responsive and efficient; and ii (b) what are the effects of government policies and programs on the sector's development. D. &Rroach of the Study 4. Although the recommendations arising out of this stuJy are focused on expediting and increasing the supply of housing for the lower-income population, the approach taken in the report is to consider the factors and policies influencing the delivery of housing sector-wide. The sector is first analyzed from the perspective of the interaction of supply and demand to produce a particular quantity and price of housing in Malaysia. This, along with various international comparisons, highlights the issue of high prices in the Malaysian housing sector. Next, a financial model of major public policy and program interventions in the sector is used to examine the incentives and disincentives created by government policies and regulations and to identify their net impact on efficiency and equity in the sector. This integrated view of the main government interventions allows an assessment of how changes in one policy or program will affect not only the type of housing which is supplied, through changes in its profitability for the developer, but also the type of housing demanded, through changes in subsidies to the consumer. The main value of such an approach lies in its ability to quantify a number of policy-induced interventions and to see their cumulative effect on the behavior of both suppliers and consumers in the housing market. 5. The financial model used in the report ervloys present va'.ue analysis to examine the economic and financial costs and benefits of two major public housing programs as well as several types of private sector housing development. For each type of program or development, a cash flow model for a representative investment is set up and the present value of each government intervention (e.g., land subsidies, financial subsidles, taxes, regulatory costs) is calculated. The use of present value analysis has the advantage of allowing direct comparison of not only the costs and benefits of quite different interventions in various programs and but also the profitability and economic return of various types of housing -high- versus low-cost, owner-occupied versus rental, one location versus another). It thus permits analysis of how government actions such as land-use regulation, financing policy, infrastructure provision, taxation, price controls and other regulations affect incentives to investors. 6. Flowing from che findings of this analysis -- that the land-use standards and infrastructure engineering regulations imposed by the Government constitute the chief constraint to large-scale development of low-cost hdusing, the report concludes with a comprehensive and integrated assessment of the cost implications of the principal design parameters used in the sector in order to determine how best to reduce costs and increase the supply of low-income housing without significantly reducing the benefits to users. The methodology employed for this assessment is to model all of the design standards which determine land-use and to assess their impact on the net salable area of a given site. The model is able to vary individual standards and measure how changes in land- use practices would affect housing development costs. Land-use and iii infrastructure standards are normally established on the basis of an abstract minimum 'need' for each service or facility, and each regulation or standard seems reasonable considering only the narrow purpose or specific objective to which it is addressed. However, the standards often have uninteaded effects on other project characteristics, such as scale, which may result in costly distortions. The analysis carried out in the report shows that all of the standards, taken together, have a significant cumulative influence on costs which necessitates explicit consideration of the trade-offs involvzd. Thus, standards are important determinants of the type and location of housing ultimately produced. C. Basic Findings The Housing Sector 7. The housing sector plays a larger role in the economy than is commonly perceived, with housing investment having been as high as 9% or more of GNP in recent years. Official statistics on hoasing investment appear to have seriously underestimated the actual level of investment in the sector for at least a decade. The main reason for the statistical discrepancy appears to be that official government estimates rely on data on inputs for housing cons.;ruction which themselves appear to be seriously underestimated. The alternative estimates rely on data on housing production and on unit prices of housing. 8. Housing sector investment in the early 1980s was at an unsustainably high level, suggesting that recent declines in the level of activity in the sector represent a return to a more normal and sustainable position rather than a short-term aberration. Comparisons of housing investment levels in Malaysia and in other countries with similar income levels during the early 1980s indicate that investment has been far higher in Malaysia than in other similar countries, particularly over an extended period of time. This high level of housing investment was not characterized by significant increases ir the number of new units prodtvced but was instead attributable to an extraordinarily rapid and sustained rate of increase in the selling price of new housing. 9. Values of newly built houses increased at a compound rate cf about 18.9% per year between 1972 and 1982, dramatically outstriDDing the rate of increase of either rents or consumer prices, wqhich increased at average rates of 6.4% and 7.0% per year, respectively. Such increases also out;aced nominal income increases, which rose by only 10.5% per year over the same period. For a time, such price increases were fueled by expectations of both future price rises (which promised households that future capital gains would make current sacrifices worthwhile) and future income rises (which promised to make the future burden of housing expenditures an increasingly smaller portion of income). When each of these expectations changed in the early 1980s, it becam. apparent that such high housing prices could not be sust&ined. For example, the ratio of the value of the least expensive type of new housing to income rose from 4.6 in 1977 to a peak of 6.9 in 1982, a much greater multiple of household income than is common in other countries. In 1986, the ratio was still 6.0. This suggests that, while many households may have been squeezed out of the iv market for new houses, current prices are likely to fall even further before strong demand resumes. 10. The affordab_l1t& of housing in Malaysia declined, dramatically so for the purchasers of the lowest priced neu housing, from about 1977 through 1982. Affordability improved somewhat after 1982 as housing prices flattened out while incomes continued to rise through 1985. The measures of affordability are computed from reported selling prices of new units which are an im;erfect measure of price movements in the market for existing owner occupied housing and rents but are indicative of the directio-i. Although no direct data on chsnges in the share of households renting was available, the last census shows a large proportion of renters in the major urban areas (over 50%) where prices are the highest and have been -Lsing fastest. Squatter communities are also growing in these same centers. Both indicate an increasing problem of housing affordability especially for the poor in and around major citX1es. Why Housing Prices Are So Hizh 11. The high absolute level of housing prices and their rapid rate of change seem to have resulted from a combination of cost-increasing government regulations, an unresponsive supply system and strong demand. On the supply side, a combination of inappropriate physical standards and excessive regulation in the sector contributed to high costs and insufficient responsiveness in the hovtsing supply system. While the costs of most basic housing inputs have been relatively stable during the 1970s and 1980s, the critical factors seem to be the physical standards for housing construction set by the Government, particularly those related to land-use rlanning and infrastructure provision, and regulatory measures necessitating a Ame-consuming approval process for any proposed housing development. Rapidly rising household incomes and poorly timed housing finance subsidies for civil servants were the main factors stimulating demand. Many housing purchases may have had an important investment motive during the period of rapid price increases. 12. Analysis of the effects of the Vhysical standards for housing construction on heusing prices and output indicates that the design standards for low-cost housing, in terms of ui1it size, choice of materials and efficiency of local contrirtors, seem apprqpriate. However, land development standards, as presently enforced by local authorities, are found to constitute one of the major constraints encountered by developers in responding to the demand for low- cost housing. Using international practice as a yardstick, it appears that about 25% of the land developed for residential purposes is wasted due to excessive road areas, arbitrary setback regulations (i.e., the distance required between a house and the property line) and, to lesser -xtent, redundant community facilities. Due to such wastage, only 25% to 50% of the land area developed may be salable (in contrast to the 65% typically achieved in other countries). The cost of the land which cannot be sold is therefore passed on by the developer to the home buyer, so that housing costs are higher than necessary. 13. To illustrate, current land development standards in Malaysia result in the construction of low-cost residential projects with road areas per household which are up to four times larger than those used in Asia, Europe and v the United States for similar ranges of plot size. Local authorities often require back lanes of up to 6 m, or more than double the 2.6 m typically used in the United States. The set-back requirement of 4.5 m along vehicular roads implies that corner plots must be at least 140 sq m, only 38 sq m of which are buildable. Since regulations restrict the floor area of units on corner lots to that of units on interior lots, the market value of corner plots is hardly more than the value of an interior plot half its size. With the area of corner plots usually representing about 20% of a development's total salable area, the current regulations result in a significant loss of potential revenues for the developer. 14. The effect of these and similar practices is that about 170 sq m of raw land are required to develop the minimum plot size of 68 sq m in Malaysia although at least a third less land or about. 110 sq m would be sufficient if land-use standards were in line with those usid elsewhere. Consequently, where only 60 plots per hectare are possible in i:'laysia with existing land-use standards, 90 plots per hectare are common in other countries for developments with the minimum plot sizes used in Malaysia. The implications of these standards on house costs are profound. Since land and infrastructure costs may represent between 45% and 60% of the total cost of a low-cost unit in Malaysia, depending on location, use of the unnecessarily high standards has made it almost impossible for developers to build units for less than the regulated ceiling price for low-cost housing of M$25,000, which is still not affordable to many of the low-income population. The standards have the;efore made construction of low-cost housing both expensive and inefficient. 15. The standards also make it less profitable for developers to construct low-cost housing than medium-income housing. This is because a developers' profitability is directly linked to the percentage of salable land in a development and the amount of housing floor space built per unit of land. The proportion of salable land in high-density, low-income developments is now less than that of lower density medium income housing, primarily due to the excessive amount of street space required and the stipulation that developers must provide the land for community facilities for large-scale developments, wnl2e land for facilities in smaller schemes with fewer units are financed by tax revenues. Current land-use legislation combined with the nation-wide fixed selling price for low-cost housing also makes developers reluctant to build low- cost housing in and around urban areas where the cost of land is high despite greater demand in such locations. To hold costs under the nationally mandated M$ 25,000 for low-cost housing a developer must reduce the housing floor space to compensate for higher land prices which also reduces the rate of profit. The poor profitability of constructing low-cost housing in urban areas where it is most needed, and most in demand, distorts developers response to market signals. If land-use standards were revised, private developers would likely build more low-cost units because: (a) they would be able to develop land in more expensive areas where the demand for low-cost housing is greater; and (b) they would be able to build a larger area of floor space per unit of land when building low-cost housing and therefore make more profits than when building middle-income housing. 16. Overly complex and time-consuming housing project anproval grocedures at the local level also contribute to the high cost of new housing vi as well as the unresponsiveness of supply. Project land-use and building approvals, for example, normally take more than three years to obtain, and may take as long as five. On top of this, 15 to 20 separate government departments are typically involved in the approval of plans and specifications, adding another two to five years to project completion. The long lead time required to bring a project to fruition due to the lengthy process of obtaining land-use, subdivision and building approvals as well as approval of site p.ans and building specifications prevents developers from responding quickly to changes in demand, thus increasing their risks and eff'ictively raising costs. 17. The extensive direct production of housing by government entities in Malaysia has also increased the risks associated witn housing construction for private firms. The public sector's share in the housing market increased from an estimated 25% in 1976-80 to 35% in 1981-85. Not all of this construction has been for the low-income groups; in some cases the public and private sectors are competing for comparable market segments. The public sector has the advantage in this competition, however, since its costs are often lower than those of the private sector due to differential treatment by local officials who facilitate the approval process for public sector developers and apply standards less stringently. The basis for differential treatment is not found in any statute or regulation but is rather a matter of practice. The public firm may also be in a stronger financial position, due to implicit government backing, enabling it to hold unsold inventory and cut prices more aggressively than private competitors. 18. Allocation practices for low-cost housing in Malaysia also directly add costs for developers and may decrease responsiveness by raising uiucertainty as to when completed units may be sold. The specific requirements are that a fixed share of the units, commonly 30% to 40%, be reserved for indigenous Malay purchasers. States can and do change allocation requirements for particular projects if developers have made good faith efforts to meet the quota and fail. But such changes take time and, given current interest rates, a year's delay, which is not uncommon, can easily wipe out any profit margin developers enjoy ex ante. 19. The lack of supoly responsiveness in the Malaysian housing market, due, inter alia, to the lengthy procedures involved in housing development in Malaysia, seems to be supported by comparisons with other countries in a similar position. Despite the s-eady growth in the Malaysian housing stock in the past decade, a comparison wit a number of other countries (which might have been expected to produce less t 4sing) puts Malaysia near the bottom of the range for percentage increase in housing stock, further supporting the conclusion that strong demand was translated relatively more into higher housing prices than into the number of units. Another test of housing supply elasticity is to examine the way in which housing completions vary in response to housing rrice changes. In normally functioning markets, housing completions rise during periods of rapidly increasing housing prices, but in Malaysia housing completions, whether publicly or privately carried out, has shown little responsiveness to rises in the housing price index or to the availability of housirg finance. Lack of market responsiveness therefore seems to be a key constraint in increasing the availability of low-cost housing. vii 20. On the demand side government intervention may have stimulated demand for low-cost units while other policies and practices were increasing costs and discriminating against the supply of these same units. The gXactice of limiting and controlling squatter settlements may have precluded an "escape valve' which has been available in other countries in similar circumstances and has created pent-up demand for low-income housing which cannot be met at current prices. More serious has been the Government-sponsored Treasure Loan Program which makes mortgage loans available to civil servants, without regard to housing costs &nd at interest rates that were typically about 4% per annum, less than half the rate of interest for private mortgage finance. During 1981 and 1982, the years in which the TLP expanded most rapidly, the program accounted for between 30% and 40% (If all rormal sector lending for housing, and for most of the net increase in housing sector lending in those two years. The effect of the subsidy was to reduce the real cost of housing for civil servanrs by approximately half. This, in turn, has removed the incentive for careful shopping and during the early 1980s created considerable upward pressure on housing prices. 21. The high price of housing; in Malaysia is not only due to the induced high costs described above but is the groduct of an unresponsive housing supply system in the face of strcng demand. In Malaysia, there is a strong correlation between annual changes in the prices of newly built housing and changes in the rate of increase of household incomes, with prices rising at a rate that is 50% faster than the rate of increase of household incomes. This is a sign of a highly price-inelastic housing supply, that is, a housing supply that is unable to adjust fully or quickly to demand shifts, with the result that much of the market's supply response is reflected in increasing prices rather than increasing quantities of new housing units. This lack of responsiveness is one of the most critical policy issues in the housing sector. Governmert Response 22. Housing Programs and Subsidies. The Government has for some time recognized the high cost of new housing in Malaysia, and, in keeping with its goal of ensuring that all Malays4-.:a, particularly the low-income group, have access to adequate shelter. 'as sponsored various initiatives to promote the availability of new hous'.. to the lower end of the income distribution. The major interventions with this focus have been the Public Low Cost Housing Program (PLCHP), the SLCHP and directed credit for the purchase of low-cost units, including interest rate ceilings on mortgages. 23. The PLCHP was launched under the Fourth Development Plan (1981-85) and is now winding down from a target of 176,500 units during the Fourth Plan to just over 67,000 units during the Fifth Plan. The units are developed and built directly by the States, although design and construction are normally privately tendered. The maximum sales price is M$25,000, except in Kuala Lumpur where it is higher. The Federal Government develops overall guidelines for the program and provides financing to the States at 4% interest. The States relend these funds to purchasers, typically at a fixed rate of 5.5% for 25 years. The maximum loan size is M$25,000. The PLCHP has not been very successful, as of mid-1987, only 41% and 10% of the Fourth and Fifth Plan targets, resptctively, had been achieved. viii 24. The SLCHP was launched in 1986 with the twin goals of stimulating the economy by a hoped-for 2% a year and increasing the supply of low-income housing. The most important difference between the PLCHP and the SLCHP is that the SLCHP has a wider range of developer options and, in particular, makes greater use of the private sector. This increased reliance on the private sector reflects not only an attempt to reduce the burden carried by the public sector under previous low-cost housing programs but also the recognition that private builders have commonly had more success in building and marketing units than public builders. Two features of the SLCHP are particularly aimed at attracting consumers and producers to participate in the program. For consumers, financial institutions have been instructed by the Government to provide adequate credit to qualifying buyers; for producers, selected infrastructure standards have been reduced and the process for approving development plans has been accelerated. 25. Like the PLCHP, SLCHP progress has been disappointing. Of the 80,000 units to be built during the program's first year (July 1986-June 1987), at the end of April 1988, only 15% were complete, 9% were awaiting infrastructure and 26% were at various stages of construction. The reasons for the shortfall under the SLCHP as well as the PLCHP are very similar. In both programs, developers were being asked to undertake projects which they believed would offer them little or no profit and were therefore understandably reluctant to participate. The cumbersome regulatory environment affecting the programs, and the inappropriate design standards used directly increased costs and made it unadvantageous to produce the low-income r*itv. This was exacerbated during the peak period of the PLCHP when demand in the upper portions of the housing market was high, so that even the majority of public developers strayed far from their declared policy of building low-income housing. When low-income units did get built, sales were frequently slow because the sites selected were often unattractive to prespective buyers. Developers might have chosen poor sites due to government policies and practices such as the fixed maximum selling price of low-cost units which makes the use of more expensive land unprofitable, and public agencies normally use state land, which tends to be in undesirable locations, because the States receive only a nominal fee for the land. 26. The Government has also atte4upted to assist low-income buyers under a directed credit policy which sets m..nimum lending targets and places an interest rate ceiling on low-cost units. In late 1987, when uncontrolled rates were 13-15%, the controlled rates were 9.5% on the finanicing of units costing less than M$60,000 and 11% for units costing M$60,000 to M$100,000. Although these rates are variable, interest rate adjustments are in frequent. Since the commercial lenders who provide the subsidized rates and finance the subsidy element are often required to borrow short and lend long, there is a possibility that future inflation could quickly decapitalize these institutions. 27. Incentives and Disincentives in the SuRRly and Demand of Low-cost Housing. While the general reasons for the Government's inability to stimulate production of low-cost housing are fairly clear, the incentives and disincentives operating in the market are better understood through the use of a financial model which quantifies the effects of government interventions in the sector and how they affect housing costs and incentives to buy and to build. ix For this analysis, a model was designed which permits an evaluation of housing policies and programs from the points of view of the economy, housing suppliers and households, i.e., how combinations of interventions and changes in specific policies and programs influence the attractiveness of individual projects from the three perspectives. The model was used to assess 13 representative low-cost housing investments and the effect on those investments of the following government interventions: land provision, infrastructure and construction subsidies; finance subsidies; the cost of the overly-generous land-use and infrastructure standards commonly used for this type of housing; the lengthy plan approval process; and the cost of relevant taxes. 28. Results obtained from this type of analysis are admittedly indicative, and will vary depending on the assumptions adopted. They are, however, very useful in identifying the relative magnitude of the interven~'cions and in understanding how the various subsidies and costs create incentives which affect the behavior of producers and consumers of housing. For example, low- cost housing may be constructed by public or private sector developers on either public or private land. Depending on the combination of interve-itions affecting the particular unit or project, the economic cost of producing a house may exceed its market value, so that it is a poor use of resources from the perspective' of the overall economy. This is true with some public sector low- cost units on public land in undesirable locations. Developers, for their Dart, may be faced with high regulatory costs which drive up overall production cor.cs so that their financial cost is less than the unit's market value. While such units would normally not be built, in some instances subsidies to developers -an offset the extra financial costs of regulation and make uneconomic housing financially profitable. Similarly, if subsidies to buyers are large enough, uneconomic or unprofitable units will be heavily in demand. The combinations of interventions can therefore create unintended incentives and undesirable outcomes. 29. The model clearly indicate& that the various government interventions in the housing sector, either Introduced in conjunction with the low-cost housing programs or related to other policy objectives, are often internally inconsistent and likely counterproductive. The extensive public sector role in housing development programs seems to have made housing investment less economically efficient, as suggested by the fact that PLCHP units aLe often worth less than they cost to produce -- not due to production inefficiencies but because public developers are frequently constrained in their choice of location to publicly available land (and may also be responding to regional development goals rather than demand). Yet these units may be heavily in demand if subsidies to buyers are large enough. The 'verall impact of the SLCHP is similarly skewed. The SLCHP is superior to the PLCHP in several respects, including its potential economic return, because its more reasonable standards and speedier regulatory procedures reduce costs and the larger private sector role is likely to improve decisions about location. However, the over- zealous application of higher standards and regulatory practices by local authorities, responding to government incentives external to the program, adversely affects the profitability of building the small, lower-cost units. Net incentives under the SLCHP may also work against privatization efforts by making construction of low-cost units less profitable than higher-cost ones and of greater risk due to public sector competition. Thus, two prime policy x objectives -- affordability of housing and privatization -- may be at risk. Somewhat perversely, the land subsidy that significantly lowers the overall financial cost of public sector housing developers normally also means that the unit is poorly located and thus an unacceptable investment from the points of view of both the economy and the house purchaser. 30. The model's quantification of both the subsidies and costs embedded in government policies and public housing programs further indicates that subsidies are deep but are ineffective in reducing costs or stimulating more low-cost housing production because they are offset by high regulatory costs and perverse incentives for suppliers. The main subsidies are publicly supplied land, reduced infrastructure costs and '_elow-market financing. The largest subsidy in many publicly developed units is free or nominally priced land worth about one third of the selling price; however, this is typically offset by an extra regulatory cost of a similar magnitude. Regulatory costs are broadly defined in this analysis, with the value of the land wasted by high land-use and infrastructure standards being the largest component of such costs. The main additional regu' itory costs arise from an unusually extensive, complicated and time-ccnsuming housing plan approval system; housing allocation requirements that delay sales; and risks created by unfair competition from public developers. Of these, only the last is not quantified in this study. 31. The equity implication of this analysis is that, despite large transfers to housing purchasers, the distribution of subsidies is not supporting government housing program goals or enunciated policies. In fact, to the extent that beneficiaries can be identified, the incidence is regressive, with a relatively small number of middle- and even upper-income Malaysians receiving large windfalls. Subsidies are deep and, in the case of the main public programs where the selling price is fixed, related to location rather than any other criteria. Under the present policy of a fixed national maximum selling price for low-income houses, the buyer in a large city may get a unit worth twice the purchase price, while another, in a remote location, may receive almost no subsidy. An additional major subsidy to some home buyers is the TLP which is clearly regressive within the civil service since the amount of subsidy is positively related to income. General directed credit for housing is also not likely to be progressive, although there is no information on the income of borrowers to support this conclusion. 32. In summary, it is clear that the market, as influenced by present policies and programs, is not yielding enough low-cost units to satisfy the potential demand from lower-income groups. Subsidies are deep for various public programs but inefficient because they are offset by high regulatory costs. A major conclusion is that, holding location and design constant, private developers build better and more marketable low-cost units than public developers. However, in the absence of subsidies, even the most efficient builders cannot deliver affordable houses in locations where they are in demand in the current regulatory environment. On the other hand, were regulatory provisions modified, it would be possible to increase affordability substantially while at the same time reducing the reliance on housing subsidies. xi D. Recommendations 33. The Government's attempt to stimulate economic growth through housing construction under the SLCHP does not appear to have been successful to date. In large part this is due to the normal time lag encountered between sectoral investment and its effect on GNP, although in this case the situation is complicated by Malaysia's unresponsive housing supply system. The housing sector would therefore seem to be a poor conduit for macroeconomic stimulation, especially in the Malaysian context. 34. The other goal of the SLCHP, i.e., to enable a larger portion of the low-income population to purchase new housing, nevertheless remains important and still appears viable. But little progress will be made in increasing the availability of low-income housing unless basic policy reforms are introduced to correct the fundamental structural problems affecting the whole housing market, the symptoms of which are unnecessarily high costs and a housing supply system that responds poorly to changes in demand. To do this, the following recommendations are proposed with the objectives of increasing the responsiveness of the housing supply system, lowering housing costs, improving suppliers' incentives to provide low-cost housing, and focusing sector subsidies on the low-income population. While the suggested measures each have a specific overriding objective, they tend to be interrelated and mtutually reinforcing. Increased Resgonsiveness in the Housing Supplv System 35. Malaysia's housing supply system could be made more responsive by reducing the number of steps involved in gaining permission for a proposed housing development and by shortening the time required for review of plans and approval. Even under the SLCHP, which is supposed to allow 'one-stop' processing, the process of gaining approvals of development proposals can delay construction by from 8 to 18 months. An analysis of the differences between short and long approval times under the SLCHP should serve as a model to improve performance nationally. The ultimate objective of this change would be to lower both the risks and costs entailed in the housing development process and to increase suppliers' ability to meet demand by providing the type of housing people want, where they want it. Lowering Housing Development Costs 36. Analysis of the principal housing development design parameters indicates substantial scope for lowering housing development costs, and thus making new housing more affordable to the low-income population, by: (a) adjusting land-use standards to use land more efficiently, (b) revising the system of financing community facilities for large-scale developments, and (c) ensuring that infrastructure engineering design standards reflect an appropriate trade-off between capital and maintenance costs. 37. Revision of Land-use Standards. In determining land-use standards, the guiding principle should be that the profitability of low-cost housing should not be less than that for other types of housing. Rather than xii arbitrarily reducing all standards, it is recommended that global parameters be set as planning targets, the critical parameters being the percentage of salable land in a development and the floor area ratio (FAR), i.e., the ratio of house floor area to plot size. Analysis carried out by the Bank suggests targets of 55% for salable land and a 0.37 FAR (salable land in low-cost developments now ranges from 28% to 47%, with FARs of 0.25 to 0.27). The global measures should be used as screening devices within which developers and local authorities would be allowed to make trade-offs among competing land uses based on their judgement of marketability or demand. If the principle of using global parameters is accepted, then current land-use regulations governing housing set-backs and road area requirements, including those for back lanes, would need to be changed. Since acceptance of appropriate standards by local authorities has already proven to be an issue under the SLCHP, it is further recommended that federal funding for low-cost housing be made contingent on the adoption by local authorities of the improved standards. 38. Financing of Community Facilities. The present system of determining the scope of and financing land for community facilities also raises the cost of large-scale developments aimed at the lower-income population and in so doing distorts developers' decisions about the size and type of projects they will undertake. Under current regulations, the percentage of land allocated to community facilities is different for large- and small-scale developments because certain types of community facilities have to be provided only when a population threshold is reached. Furthermore, if a given area is developed through a number of small schemes, below about 500 plots each, the land to be used for community facilities is purchased by the local authority using general taxation revenue, but if the same area is developed through a few large schemes, above about 2,000 plots per scheme, the cost of land for community facilities is borne by the developer and therefore passed on directly to the future plot owners in the price of the 'Lwelling. These rules discriminate against large-sc3le projects and small plot sizes; they distort the price of land development, as the unit cost of developed land in a large scheme appears higher than in a small scheme, and make the source of financing of land for community facilities dependent on the scale of the scheme. It is therefore recommended that land for larger community facilities be financed out of a development fee. A study to establish a detailed plan of how such a fee might be structured for financing community facilities is also needed. 39. Infrastructure Design Standards. Local authorities are currently allowed to require more than official national minimum standards for roads constructed in conjunction with housing developments; as a result, both road standards and development costs are unnecessarily high. Their reluctance to allow lower, but acceptable, standards is motivated by two factors. First, they have a strong vested interest in ensuring that a maximum length of streets in new developments are at least 30 feet wide since they receive a Federal Government grant for road maintenan-e, the amount of which is based, among other things, on the length of streets in their areas that are wider than 30 feet. Since the grant is fungible and does not have to be spent on maintenance, it is an attractive local source of income. Second, they tend to prefer higher road standards overall in order to reduce subsequent maintenance costs, which they themselves must finance. It is therefore recommended that an objective formula for calculating the least-cost mix of capital varsus recurrent costs for all xiii infrastructure included in low-cost developments be established and applied. It is further recommended that the formula for Federal Government transfers, such as the road maintenance grant, be revised to support more efficient designs. ImRroving DeveloRers' Incentives 40. Flexible Pricing of Low-cost Housing. Variations in market conditions throughout Malaysia are not matched by variations in the price of low-e-ost units (-9 ich is controlled at a maximum of M$25,000). In better locations, where the market value of these units is highest, the financial loss to the developer charging the much lower controlled price is greatest. The developer thus has an incentive to minimize production of low-cost units to only the number required to obtain permission for some other profitable activities, such as building more expensive housing. To improve the correspondence between signals to developers (their profitability) and the economic desirability of building each unit, it is recommended that the price of low-cost units be based on the unit's location using existing information on local prices of a standard unit. Since the current standardized pricing policy is also the primary determinant of buyer subsidies, with the largest subsidy going to the buyer in the larger centers where housing price levels are highest, this proposal would also reduce "untargeted" subsidies. 41. Market Pricing of Land. While well-intended, the present practice of providing state land at nominal prices (currently M$0.20 per square foot) may actually distort housing production in two ways. First, since the States know that they will not be fully compensated for any land provided for low-cost housing, they have little incentive to select more attractive and better located parcels. Second, developers (public or private) who receive free land may not be as conscious of using it efficiently even though other associated costs, such as infrastructure, may be higher as a result of inefficient land-use plans. All transfers of land for low-cost housing from State Governments should therefore be on the basis of market prices, with any subsidies provided separately under an explicit plan to assist developers or buyers. 42. Redefinition of Public Sector Role in Low-cost Housing. A major area for reform of policies concerns the need to clarify the role of the public sector in the housing market. The Government's role in regulating the sector shcould be re-examined, with the objective of ensuring that the regulations and standards imposed protect the buyer while not unnecessarily constraining sector growth and efficiency. Also, public sector housing investments have not always been economically efficient. If it is decided to continue direct public sector involvement in housing development, it is recommended that incentives for the production of low-cost housing be reviewed and adjustments made so that they are neutral in respect to public-private developers, as well as to project scale, location, and house value. 43. Improved Procedures for Allocation Ouotas. Requirements that 30% to 40% of low-cost units be allocated to Malays and that the State Governments approve all buyers tend to discourage private sector involvement in low-cost housing development since sale of completed units is delayed, thus raising developers' holding costs. Replacing the quotas with direct subsidies to xiv Bumiputras would lessen this adverse impact. Alternatively, the cost of the quotas could be reduced if the authorities formalized what constitutes "good efforts" on the part of developers to meet the requirements, then released the obligation once those conditions had been met. Targeting of Subsidies 44. DeveloRment of a Subsidies Policy. A primary purpose of subsidies is usually to achieve an equity objective, be it assisting a deserving housahold to purchase better accommodation or reducing the price of housing by lowering costs for a developer. At present, the numerous explicit and implicit subsidies in the housing sector in Malaysia do neither very well. Large subsidies to developers, in the form of reduced land costs and cheaper infrastt;cture, are off-set by high regulatory costs. The substantial subsidies conferred on some buyers of low-cost housing do not seem related to need or anj other program goal. The subsidized TLP goes to a group not near the bottormz of the income distribution. Consequently, it is recommended that an explicit subsidy policy be formulated and agreed, specifying the purpose, level and target recipients for housing subsidies. 45. Phasing out of Directed Credit. Lending quotas for commercial banks and finance companies may have originally been necessary to introduce those institutions to mortgage lending but, given the size of the housing loan portfolios they presently hold and the general liquidity of the banking system, the time may be appropriate to consider phasing out these quotas. Administered interest rates subsidizing the purchase of smaller houses represent a significant, but by no means the largest, distortion in the housing market. Subsidies resulting from this policy are roughly estimated at M$123 million a year, or about 25% of all housing credit subsidies provided. It is not clear that the present system of compulsory lending at rates only slightly below the market is better for home buyers in general than better access to credit, with banks allowed to charge rates matching their risk assessment of each type of loan. It is thus recommended that the present directed credit regulations for housing be reviewed with the objective of establishing a timetable for phasing them out. 46. Reform of the TLP. Although the TLP was primarily designed as part of the compensation package for public employees its may have had an important effect on the housing sector. TLP loans bear a much higher subsidy than those provided under directed credit arrangements and represent about 60% of all housing credit subsidies provided annually by the Government. A typical TLP loan is subsidized at up to 40% of its face value if evaluated ex ante at current market rates for similar loans. While current TLP contracts specify that interest rates are variable, the Treasury has not been able to exercise this option in the past, and many civil servants believe the rate to be fixed in practice (or at least extremely sticky). The aggregate subsidy under the program is roughly estimated at about M$400 million a year or over twice the annual development budget for housing. Further, this subsidy is not explicitly budgeted. While the subsidies in the other programs are not negligible, further reform of the Treasury program could have a high payoff. The Government may also wish to examine the need to tax below-market loans given by private firms to prevent any distorting effect on the housing market. xv Action PlAn 47. The above recommendations might be implemented under the following three-part action program addressing the key requirements for the development of a more dynamic low-cost housing supply system in Malaysia. The main areas to be addressed are listed, along with a description of required changes in policies and practices as well as the main objective of these changes. (a) Revise land-use and infrastructure standards and selected regulatory practices. The obiective would be to increase the supply of low- cost units by making it at least as profitable for developers to build them as other types of housing and by making it financially attractive for local authorities to accept more appropriate standards. The main mechanisms for achieving these objectives would be (i) the revision of kay land-use planiing parameters to achieve at least a 55% salable land area and 0.37 FAR on low-cost sites and (ii) revision of the formula by which federal road grants are calculated. In cooperation with the Town and Country Planning Department of the Ministry of Housing and Local Government (MHLG), the Bank could assist the Government to develop prototype site plans which would beco'te the basis for revised land subdivision legislation and revised infrastructure financing mechanisms which could be implemented on a pilot basis to demonstrate the feasibility and attractiveness of such plans to developers and home buyers. Changes in the formula for financing infrastructure would be studied and discussed with the Federal Treasury. (b) Rationalize and reduce subsidies. The obiective would be to make the present pattern of subsidies in housing as rational as possible while reducing the subsidy level to the minimum consistent with the Government's social and political objectives. The following steps are proposed: (i) measure existing subsidies, (ii) develop an explicit policy on subsidies, and (iii) draft an action plan to change the size and pattern of subsidies. The measures to be changed involve a number of ministries including MHLG and the Treasury. Step (ii) could be undertaken with MHLG and subsequent steps possibly under the guidance of the interministerial policy committee which presently exists for housing. The main actions required include the introduction of flexible pricing for low-cost housing units. based on local market prices; reimbursing State Governments for the market price of land they supply for low-cost housing developments; adjust incentives for the production of low- cost housing so that they are neutral with respect to public and private sector developers; and formalize procedures for meeting ethnically-based allocation quotas. (c) Reduce regulatory costs. The objective would be to make the housing supply system more responsive, primarily by streamlining the approval process required for housing development, thus lowering costs and reducing risks to suppliers. This might be done by analyzing implementation experience under the SLCHP and determining xvi why the time required for approvals varied so much among the individual SLCHP developments. The expected result of this exercise would be a clearer understanding of the costs of unnecessary regulation and inefficient processing of applications as well as recommendations on how to shorten and streamline the proce;a. 1. INTiTRODUCg[ON A. Background of the Stvft 1.1 Late in 1985, the Government of Malaysia was confronted with a deteriorating macroeconomic situation, evidenced Dy falling exports, declining rates of capital formation, and falling GNP. The construction sector, whlich for a considerable period had been one of the leading sectors in Malaysian economic growth, had experienced a rapid and severe turnaround. After having grown at an average annual rate of 9.7% between 1980 and 1984, value added in the construction sector fell by 8.4% in 1985 and by 1.4.0% in 1986. In an attempt to deal with a sharp reversal in the sector, the Government decided to implement a Special Low Cost Housing Program (SLCHP), under which some 80,000 housing units per year (priced at less than N$25,000) would be built. It was expected that the combined direct ard indirect effects of the program would be to raise the rate of growth of GNP by a much as 2% per year, contributing importantly to economic recovery. An additional objective of the program was to sharply change the nature of the product being supplied in the Malaysian housing market, offering less-expensive housing .o enable a larger proportion of low-income households to purchase new housing units than had been the case for more than a decade. 1.2 Not long after the SLCHP was initiated, the World Bank was requested by the Malaysian Government to help assess the Program. This study, which is a response to that request, was subsequently broadened to include all major interventions in the sector. The expanded coverage was thought necessary in view of the single market for both inputs and outputs in the housing sector which make it difficult to accurately assess one program in isolation from the larger sectoral environment. The study has therefore evolved into a consideration of the incentives and disincentives that various forms of government intervention create in the sector. B. Obiectives 1.3 The objectives of this study are to determine: (a) whether the housing supply system in Malaysia is responsive and efficient; and (b) how government policies and programs affect the sector. As a result of this focus, it has not been considered necessary to treat all aspects of housing policy either exhaustively or equally. - 2 - C. The Approach 1.4 The report considers the sector from the larger perspective of the major influences on its development and the factors responsible for its current structure, particularly the lack of adequate low-cost housing. The sector is first analyzed from the perspective of the housing market, which, along with various international comparisons, highlights the issue of high housing prices. Next, with the help of a financial model and present value analysis, the report quantifies the incentives and disincentives created by the major public policy and program interventions in the housing sector and identifies the net effects of these actions on efficiency and equity in the sector. This integrated view of the main government interventions allows a detailed consideration of how changes in one policy or program affect either the type of housing supplied, by changing the profitability to the developer, or the type of housing demanded, by changing the subsidies available to the consumer. The main value of this approach lies in its ability to quantify a number of policy-induced interventions and to examine their cumulative effect on the behavior of both suppliers and consumers in the housing market. Finally, some of the principal design parameters and regulations for the sector are analyzed, again in a comprehensive and integrated framework which permits measurement of the cumulative effects of these standards on housing costs. 1.5 The study has five chapters. This introduction is followed by a general description of the sector - the housing stock, its growth and quality, financial intermediation in the sector, and the Government's major objectives and programs for the sector. The housing market is examined in Chapter 3 and the major facto:s responsible for high ho~using prices in Malaysia are identified. A rep-resentative sample of the government initiatives are analyzed in Chapter 4, using the financial model mentioned above to provide a quantification of their costs. Chapter 5 considers the most *ostly of these interventions, i.e., standards and regulations for the housing sector as a whole, and identifies the design parameters with the largest impact on costs and developers' profitability. 2. THE HOUSING SECTOR IN MALAYSIA A. Demographic Trends 2.1 More than one third of Ialaysia's nearly 17 million citizens live in urban centers of over 10,000 people. Between 1970 and 1980, the urban population grew at an annual average of almost 5%, considerably higher than the less than 3% growth of the total population. Furthermore, as a result of declining household size, the number of urban households grew by almost 6% per annum during the same period. 2.2 In a typical recent year, about 70,000 new urban households were formed in Malaysia, about half due to natural population growth and the other half reflecting migration to urban areas. Since migration is expected to continue at a similar rate, the level of urbanization should surpass 40% by the end of the century. The current dominance of the capital region and the Klang Valley area is expected to continue. B. The Housing Stock Growth 2.3 Growth in the stock of dwellings in Malaysia between 1970 and 1980 reflects both the increasing urbanizatio-± of the country and the strong economy during that period. While the total number of dwellings increased by only 3.8% p.a., the number of urban housing units rose at an annual average rate of 7.1A (Table 2.1). Therefore, despite the rapid rise in the number of urban households, the availability of urban housing actually improved. This is also indicated by Malaysia's housing supply ratio (i.e., the ratio of dwellings to households) which showed a sibstantial increase in urban areas froux 78% in 1970 to 87% in 1980 and is estimiated at about 90% at present. This level is relatively high compared to that of other countries at similar income levels and represents a considerable achievement. The value of the supply ratio is limited, of course, by not capturing the qualitative and distributional dimensions of the housing stock. Table 2.1: PoRulation and Housing Stock Rate of 1970 1975 1980 1985 Change 1970-80 Total Population (thousands) 10439.4 11985.1 13745.2 15677.0 2.8% Urban Population (thousands) 2798.6 3645.5 4492.4 5679.2 4.8% Urban Share of Population (%) 27% 30% 33% 36% - Urban Households (thousands) 480.7 668.3 855.8 1081.9 5.9% Total Dwellings (thousands) 1601.1 1966.9 2332.6 n.a. 3.8% Urban Dwellings (thousands) 375.0 558.2 741.3 n.a. 7.1% Urban Dwellings/Households (%) 78% 84% 87% n.a. - Sources: Department of Statistics Malaysia, Population and Housing Census, 1970 and 1980; 1985 data obtained from MHLG; 1975 data estimated by interpolating. -4- Public vs. Plivate Role 2.4 As shown in Table 2.2, the private sector built about 75% of all new housing units during the period of the Third Malaysia Plan (1976-80) but only about 65% during the Fourth Plan period (1981-85). This decline in private sector participation resulted partly from generous federal funding for public programs and the attractiveness of the property market to numerous state and local governments. The number of public enterprises active in the housing market grew rapidly from the late 1970s on, when they were often producing new housing for middle- and upper-income households. However, due to a slowdown in sales and the accompanying financial problems of private and public housing developers in the mid-1980s, the public sector's share of new units to be constructed during the Fifth Plan period (1986-90) is targeted to decline to about 21%. Table 2.2: New Housing Units Constructed. 1976-90 1976-80 1981-85 1986-90 Implementing Sector Third Plan Fourth Plan Fifth Plan (go0 No. (0) No. (M) No. (%) Private Sector 362,680 75 347,876 65 552,500 79 Public Sector 120,791 25 189,051 35 149,000 21 Total 483,471 100 536,927 100 701,500 100 2.5 Private sector housing construction is carried out by developers, cooperatives and individuals as discussed below. While most housing is supplied for the formal market, an increasing share is constructed informally, that is, outside of normal government regulations and procedures: ,a) Private Develoner Houslng is formal sector private housing built by registered developers, often in large-scale developments. The formal sector is well developed in Malaysia, and many registered developers are members of a well- organized Housing Developers Association. Few direct incentives are offered by Government to encourage production by private developers. (b) CooDerative Housing is housing provided by cooperative associations which are private entities. Housing cooperatives receive some government assistance, primarily through tax concessions. (c) Individual Housing is formal housing, legally built, but not in a large-scale development (and hence not subject to some planning regulations for large developments such as those requiring the provision of infrastructure and community facilities). In general, these are constructed by small-scale builders catering to the market for individual units in existing neighborhoods, as opposed to new developments. While individually small, -5- collectively these builders construct a substantial share of the formal sector total, almost as many as the larger developers did during the Fourth Plan. (d) Informal Sector Housing is built illegally, occasionally on private but mostly on public land, in contravention of land-use regulations and often without payment for use of the land. Precise statistics on the number of urban squatters and informal housing production are not available, but census data indicate that there are approximately six undocumented units (presumably in the informal sf,tor) built for every ten formally recorded units (see para. 3.6). During the last several years, when incomes and formal sector housing production have been falling, it is likely that informal sector housing has been increasing in relative terms. Housing Quality 1.6 The quality of shelter in Malaysia has also been improving as the quantity grew. According to the 1980 Population and Housing Census, the proportion of all housing units with piped water, electricity, adequate toilets and separate bathrooms and kitchens rose from 57% in 1970 to 75% in 1980 in Peninsular Malaysia. This considerable aggregate improvement was largely due to major changes in the quality of rural housing since urban dwellers experienced only a modest rise in the availability of services, from an already high level of 84% to 88%. In urban areas, about 90% of the improvement came from the addition of new full'-serviced units rather than extension of services to existing dwellings. 2.7 Despite this indication of increasing housing qualWty, census data on the permanency of housing (Table 2.3) indicate that the quality of housing in Malay-la is poor. The latter statistics, however, are misleading since, as the census itself cautions, the reported statistics on the permanency of housing units classify dwellings only according to the materials used in constructing the walls and roof. Thus, permanent dwellings are constructed of materials such as cement or brick, semi- permanent dwellings include a mixture of permanent materials and less durable materials such as.. corrugated iron and woven bamboo, and temporary dwellings are comprised entirely of traditional materials. A more meaningful measure of the quality of housing in Malaysia is the census report on the external structural condition of housing, which lists 93% of all housing units as "sound." Although this obviously reflects a judgement, the overall finding is that housing quality in Malaysia is high. Table 2.3: Permanency of Housing. 1980 Temporary Permanent Semi-permanent All Malaysia Urban 48% 44% 9% Rural 11% 64% 25% Source: 1980 Census, p. 52. - 6 - C. Affordability of Housing 2.8 Although avaiiable data on housing transactions are not classified by income groups, it is possible to make some tentative judgments about housing affordability, based on th6 changing relationship between housing prices and income. As the Malaysian economy responded to external stimuli and grew strongly from the mid-1970s, housing prices rose rapidly. From 1976 until the trend flattened out in 1982, the reported selling price of a single-story terrace house increased by an average of 18.6% pa. During the same period, household income rose by 10.8% p.a., indicating a general decline in the ability of households to purchase the most typical new unit on the market. According to the same measure, housing prices stayed almost constant from 1982 to 1984, then fell by about 20% from the peak. Incomes continued to rise after housing prizes had peaked, and the eventual fall in incomes appears to have been smaller than zhe decrease in housing prices. 2.9 Table 2.4 and Figure 2.1 illustrate the trend in affordability between 1976 and 1986, using data on average house prices appearing in the Ministry of Finance's ProDerty Market Reports. Assuming commonly available financing terms, the average house bought or sola.d in 1976 was affordable only to households at the 88th percentile of the income distribution who devoted one quarter of their income to mortgage payments. Tnis level remained approximately the same until 1980 when housing prices rose rapidly and only the 90th percentile and above could afford the average house under these assumptions, approximately the level where it has remained since. A house costing 30% below the average was affordable to the 85th percentile, rising to the 88th in 1981. A lower-priced formal housing unit, assumed to cost 50% of an average-priced unit, woUld have been affordable at the 68th percentile in 1976, but would have been affordabie only to the 85th percentile during the housing price peak of 1981-83, and to the 81st percentile in 1986. By any standard, this is very expensive housing for the relative income group. While income distribution during this period may have changed, it is unlikely that any such change would have been large enough to alter this overall conclusion. 2.10 The above examples assume a mortgage for 90% of the value of the unit, with the household spending 25% of total income on mortgage payments. In reality, many households exceed the 25% norm, especially those purchasing new housing for the first time. If it is assumed that 35% of income is spent on housing, the average unit would have been affordable at the 87th percentile in 1986 and the low-cost unit at the 60th percentile. Where family resources are available to make a larger downpayment, the loan amount would be lower, making the monthly payments more feasible in terms of cash flow. 2.11 The price of home ownership is not necessarily the price of housing. Many families rent, which costs much less. Nationwide, some 48% of all households are renters. This is based on the 1980 census which reports that 57% of urban housing units are owner-occupied, adjusted by the supply ratio which indicates the existence of 10% more households than units (para. 2.3). The proportion of renters differs, however, depending on the area. The seven more rural states had owner-occupancy levels of nearly 80% in 1980, while the Federal Territory, which is largely urban, had less than 50% owner-occupancy or (adjusting for the supply ratio) 55% tenants. - 7 - Table 2.4: Historical Affordability of Housint -30X below -501 below Amiual Average Affordable Average Affordeble Averge Atfordable Household House at House at House at Yeor Income Price rPercentile) Price (percentile) Price (percentile) 1976 S5,941 $28,800 88 820,160 85 S14,400 68 1977 86,627 $30,500 87 $21,350 85 S15,250 64 1978 87,538 $35,300 88 S24,710 85 S17,650 66 1979 n8,421 843,400 88 S3.,380 86 S21,700 72 1980 S9,888 S59,600 90 841,720 87 S29,800 81 1981 S10,966 s73,500 91 S51,450 88 836,750 85 1982 811,619 880,200 91 $56,140 88 S40,100 85 1983 S12,282 880,700 91 856,490 87 S40,350 85 1984 813,179 S83,700 90 s53,590 87 S41,850 84 1985 S13,106 S79,600 90 855,720 87 339,800 82 1986 811,587 $70,000 90 849,000 87 835,000 81 Sources: Ministry of Finance, Ministry of Housing and Local Government. Notes: Affordability is determined by assuming 251 of household income devoted to making level payments on a 25 year, 10X mortgage for 90X of unit value (LTVJ Housing price is for single story terrace house in a sample of cities. Household income is private consuantion per household based on national income accounts. Figure 2.1: Husant Affordability by Income Percentile 92 - so L Average 7~~ - ~30% (Caverage OD - 76- 1 72- 70- 61 1976 1979 19a2 IBM - 8 - 2.12 In addition, an unknown number of squatters cannot afford formal sector housing and have therefore turned to informal arrangements. While no reliable data are available on this segment of the population, their numbers seem to be growing, as indicated by expanding communities of squatters in many urban areas and the informal reports of local authorities that as much as a fifth of all housing is being constructed by squatters. 2.13 In summary, housing conditions in Malaysia have been directly influenced by the strong overall economic performance of the past decade. The boom, caused in part by the rising value of the country's exports, stimulated demand for housing both by inducing urbanization and by offering a refuge for investors. But since even the least expensive formal sector units are not affordable by the lower-income groups, the latter are increasingly turning to the informal sector to meet their housing needs. The problem, then, is apparently not that there is an absolute shortage of housing, but that the mix or distribution of new housing prevents wide-scale ownership in the urban areas. D. Government Housing Sector Obiectives and Programs Obiectives 2.14 The Government of Malaysia has affirmed that it has a basic responsibility to ensure 'that all Malaysians, particularly the low-income group, have access to adequate shelter and related facilities."1l Government objectives for the sector on the macroeconomic level are to stimulate overall economic activity by encouraging housing investment and, on the microeconomic level, to improve the efficiency and equity of the current housing delivery system. Toward these objectives the Government has adopted policies and programs which are intended to make housing more available to lower-income groups, and particularly to the Bumiputra, to make housing finance readily available to certain segments of the population, and to set land-use, planning and infrastructure standards to ensure housing adequacy. 2.15 Regarding the Government's physical objectives for the sector under its five- year development plans, Table 2.5 summarizes past performance under the Fourth Plan (1981-85) and targets for the Fifth Plan (1986-90), and Table 2.6 presents the Federal Government's development budget for public housing programs under both Plans. While the categories in these tables do not correspond exactly, the tables nevertheless illustrate the same general trend: a reduction in the public secter's role in the direct production of housing units and increasing reliance on the private sector. 1. Government of Malaysia, Fifth Malaysia Plan 1986-1990, p. 521. Table 2.5: Housing Targets and Prformance by Program +-- Fourth Plan --+ +-- Fifth Plan --4 1981-1985 1986-1990 Pat of Pet of Target Actual Target Target Actual Target c/ --------------------------------------------------------------__---__-----_--__----------------__--_-------- Public Sector (Including Public-Private Joint Progrms) Special Lov Coot Housing Program 240000 800S 32 Public Lov Cost Housing Program 176500 72308 411 67193 6593 102 KL City Hall Program Regional Dvivelopmnt Authorities 110010 36112 33X 57500 13682 2*X Staff Institutional Housing 58500 23258 401 27000 2105 8X Housing by Commercial Agencies 53560 58373 109S 18700 10610 572 -------------------------------------------------------------------__--------__----------------------------- Subtotal. Public Sector 398570 190051 482 410393 40995 102 -------------------------------------------------------------------__--------__----------------------------- Private Sector ___________________________ Private Developer Houslg 349470 101799 292 540000 23054 41 Cooperative Housing 25260 5414 212 12500 938 81 Individual Housing a/ 150000 94660 632 n.&. n.a. Informal Housing n. a. unknown n.a. unknown ----------------------------------------------------------------__-----------__----------------------------- Subtotal: Private Sector bl 524730 201873 382 552500 23992 42 --------------------------------------------------------------------__-------__----------------------------- Total: bl 923300 391924 421 962893 64987 72 Notes: aI Targets for Indlvidual HousLng not yet available for Fifth Plan. bI Subtotal and total do not include informal sector. el As of January 1988. Table 2.6: Public Development Budget for Housing Programs a/ (M$ million) Fourth Plan Estimated Fifth Plan Allocation Expenditure Allocation Program 1981-85 1981-85 1986-90 Public low-cost housing 1,712.22 1,659.06 691.79 Sites and services 1.61 1.21 78.41 Government quarters 89.29 44.47 56.62 Squatter control 3.36 3.36 17.00 SEDCs and UDA b/ 45.38 45.38 142.71 Total 1,851.86 1,753.48 986.53 a/ Institutional quarters and housing in land development schemes are not included in table as they are provided directly to the respective agencies. b/ SEDC - State Economic Development Corporations; UDA - Urban Development Authority. Source: Fifth Malaysian Plan 1986-90, p. 530. - 10 - Public Sector Housing Programs 2.16 The Government currently oversees five major housing programs as follows. 2.17 The Public Low Cost Housing Program (PLCHP). The PLCHP was the centerpiece housing program during the Fourth Plan. It is now winding down from a target of 176,500 units during the Fourth Plan (19% of planned formal output) to just over 67,000 units during the Fifth Plan. This still represents 7% of total planned output and perhaps a higher share of scarce government administrative skills and resources. Most PLCHP units have been produced for sale, but some units are initially leased to tenants, who have an option to buy after 10 years.Y The units are developed and built directly by the states, although design and construction are normally privately tendered. Units are usually of a moderately high physical standard, on the order of 70 sq.m. The maximum sales price is MH25,000, except in Kuala Lumpur where it is higher. 2.18 The Federal Government develops overall guidelines for the program and provides financing to the states at 4% interest. The states relend these funds to purchasers, typically at a fixed rate of 5.5% for 25 years with a 2-year grace period. The maximum loan size per individual is M$25,000. The loan-to-value (LTV) ratio can be up to 100%. Each state sets ethnic quotas by location. Other eligibility criteria are normally that beneficiaries currently reside in the state, and that family income be less than M$750 per month. (At current terms, this implies a typical debt-service-to-income ratio of 20%.) There is no statutory minimum income, but in practice most states require at least M$350- 400 per month (corresponding to a debt service ratio of about 40%). 2.19 To date, the PLCHP has not met its targets. During the Fourth Plan, actual production was only 41% of the target, and as of mid-1987, only 10% of the Fifth Plan target had been met due to higher housing prices and falling real incomes which seriously dampened demand for all types of new housing. Although PLCHP housing is heavily subsidized, it is still too expensive for a substantial share of the low-income population. Chapters 4 and 5 below examine aspects of the PLCHP program in some detail and suggest ways in which, by lowering costs, demand could be stimulated for such units. 2.20 The Special Low Cost Housing Program (SLCHP). The SLCHP was undertaken in 1986 in response to the cyclical downturn in Malaysia's economy and in the construction industry in particular. After having grown at an average annual rate of 9.7% between 1980 and 1984, value added in the construction sector fell by 8.4% in 1985 and by 14% in 1986. The SLCHP had two objectives: to increase the supply of low- and moderate-income housing, and to stiimuiate the economy with net additions to supply. The latter objective was the immediate impetus for the program, the combined direct and indirect effects of which were axpected to raise the GNP growth rate by as much as 2% a year. Presuming that most or all of this 2. Sale means that the structure Is sold but the land is normaUy on a long-term lease for 33, 66 or 99 years. 3. Data on the proportion of sales versus long-term leases with options to purchase are not available. Deions conceming the mix are left to the states. - 11 - housing would be incremental, that is, unlikely to be built in the absence of the program, the initial calculations of the potential stimulatory effect on the economy were not unreasonable. A similar program was instituted in Chile in response to deteriorating economic conditions in the early 1980s, and detailed calculations of the potential economic impact of the program on the macroeconomy were nearly identical to those estimated in Malaysia. 2.21 An innovative feature of the SLCHP is its reliance on private sector developers to produce most of the units. About 60% of the planned 80,000 units a year are to be built on state land provided for low-cost 4 development to private sector developers. Forty percent (32,000) were to be built o.i private sector land; to improve the financial viability of using private land for such development, developers using this variant were to build 40% low-cost units, and 60% low-medium and medium-cost units (i.e., up to M$100,000). Since only low- cost units count towards the 80,000 goal, this implies 48,000 low-medium and medium-cost units in addition to the 80,000 low-cost. Higher profits on the more expensive units are used to cross-subsidize the low-cost units. 2.22 The main supply-side incentives in the program are reduced infrastructure standards and speedier approval for land conversions and other regulatory matters. Even before the SLCHP, private developers were generally required to build 30% of new units to sell as low-cost housing (i.e., below M$25,000). Bank Negara also requires financial institutions to allocate certain percentages of their portfolios for low-cost housing. While such incentives and regulations underline the Government's commitment to housing equity, there has never been extensive construction by force account or other measures to achieve compliance. 2.23 The actual progress of the SLCHP has been somewhat disappointing relative to expectations. Although 80,000 units were to be built during the first year of the program (July 1986-June 1987), as of the end of April 1988, only 15% (12,159 units) had been issued certificates of fitness, while another 9% (6,841 units) were completed and awaiting infrastructure, and 26% (20,685 units) more were at various stages of construction. 2.24 The most serious problems in implementing the SLCHP include lack of demand due to inappropriate pricing, poor choice of locations and designs by developers, and administrative constraints. Pricing has been a problem since the standard M$25,000 per house is too low to cover developers' costs in expensive areas like Kuala Lumpur, and too high to attract buyers in the more remote states where house prices are low. Problems with sites and designs have arisen because attempts by the private sector to create what is a fundamentally different proeuct line of inexpensive houses have required experimentation in designs and site configurations that have not always been immediate market successes. The desire to reduce costs by choosing locations with cheap land prices has also frequently led to producing houses that are far from existing areas of work and senrices, and which, as a result, have not sold rapidly. The 4. In this report, low-cost housing refers to that which meets guidelines of the PLCHP or the SLCHP. The physical design is typicalt, a single-story terrace house of 40-46 sq.m., containing two bedrooms, a living area, a kitchen and a bathroon: and toilet. The sales price of such units is usually Mt25,000. - 12 - administrative constraints hampering program implementation reflect the problems posed by the Government's normal housing regulations and rrocedures which result in lengthy delays and high costs for developers. Implemenltation has been further hampered by the reluctance of local planning authorities to approve plans consistent with acceptable new, lower infrastructure standards included in the program. 2.25 Regional Development Authorities (RDAs). The RDAs focus mainly on rural development, for example, by developing new agricultural land. Programs involve two federal agencies, the Federal Land Development Authority and the Federal Land Consolidation and Rehabilitation Authority (within the Ministry of Land and Regional Development), as well as executing agencies of the state governments. Houses are constructed for participants as part of these programs. 2.26 Staff Institutional Housing (SIH). This is housing built for police, customs officials, the military, and other public safety personnel whose duties require that they reside in particular locations and/or who may be frequently relocated. Responsibility for these programs rests with the line agencies involved, although public works departments may provide design standards. Actual construction is normally contracted out. 2.27 Housing by Commercial Agencies (HCA). This is housing built by State Economic Development Corporations (SEDCs) as part of their Integrated Area Development and other industrial and commercial development schemes. SEDCs are quasi-public corporations which operate somewhat like private commercial developers. 2.28 The RDA, SIH and HCA programs are not discussed further in this report since, unlike the PLCHP and SLCHP, they have no explicit housing policy or macroeconomic goals. These programs, however, are not trivial. Housing construction under the three programs comprised half of the targeted Fourth Plan public sector output, and 60% of actual output. Their share increases to about 60% of planned public output in the Fifth Plan, and represents 80% of actual production to date, although in the context of declining direct public construction overall. If more comprehensive studies of the sector are undertaken in the future, RDA, SIH and HCA merit inclusion. E. Financial Intermediation in the Housing Sector Growth of Mortgage Lending 2.29 Growth of financial intermediation in the Malaysian housing sector has been extremely rapid, involving a variety of institutions. Table 2.7 indicates the principal types of institutions involved in housing sector lending and the total loan amounts outstanding for each category since 1980. The two building societies, the Malaysia Building Society Berhad (MBSB) and the Borneo B&ilding Society Berhad (BBSB) established in 1950 and 1958, respectively, were the first institutions to begin major lending for housing. Growth in lending by the two housing credit institutions has been steady, at a compound rate in excess of 12% per year in nominal terms and about 7% in real terms. Despite this growth, however, lending by commercial banks, finance companies, and the Government has come to overshadow the lending activities of the housing credit institutions. - 13 - Lending by commercial banks and the Government was nonexistent before the mid- 1970s, but by 1986 comprised about 39% and 38%, respectively, of all outstanding loans for housing. Lending by housing cooperatives and other institutions (e.g., rural credit cooperative societies) has never played a major role relative to that of other housing finance intermediaries. 2.30 The rapid growth in lending for housing, at a compound annual rate for all housing lenders of 22% since 1960, represents not only a major source uf resource mobilization within the Malaysian economy, but also a major component of the portfolios of Malaysian financial institutions. On an annual basis, net new lending for housing was equal to 2.6% of GNP in 1986, while outstanding mortgage debt was equal to about 27% of GNP. In 1987, housing loans represented 14% and 16% of the outstanding portfolios of commercial banks and finance companies, respectively, compared to less than 10% for each institution during the early 1970s. Table 2.7: Outstanding Housing Loans (M$ Million) Treasury Commercial Finance Building Loan Banks CompanLes SocietLes Division Total 1980 $2,233 $620 $986 $1,103 $4,941 45Z 132 20x 222 1OOX 1981 $2,811 $833 $1,214 $2,093 $6,952 401 122 171 30x 1002 1982 $3,498 $1,085 $1,450 $3,359 $9,391 372 122 152 36x 1001 1983 $4,158 $1,283 $1,597 $4,230 $11,267 371 llX 14x 38s 100l 1984 $5,130 $1,543 $1,741 $5,313 $13,726 371 1ix 13X 392 100l 1985 $6,306 $1,829 $1,906 $6,423 $16,464 381 lix 122 39s 1001 Ann. Growth Rate: 23.11 24.2x 14.11 42.2X 27.2% 2.31 Growth in lending for housing has paralleled the rapid growth in the level of overall housing investment. Figure 2.2 illustrates the way in which annual new lending for housing has varied in relation to the estimated value of new housing built between 1976 and 1986. On average, each year's incremental loans have been about 36% of the estimated value of newly built housiig, with a low of about 25% during 1979 and 1980, when housing prices were incteasing dramatically, and a high of 50% in 1986, when prices had fallen significantly. - 14 - Figure 2.2: Housing Loans Relative to Value of Few Housing 0.48- 0.46- ,- 4.2-.' 42 I 3S 02a -I I476 a 0 Lmra/HaisingVahum Housing Finance Policies 2.32 ImRlications for Financial Intermediaries. Malaysia's housing finance industry has to some degree suffered from certain government policy interventions in the sector. Initially, these policies seem to have had little real effect on portfolio decisions. In 1968, for example, the Government sought to promote lending for housing by requiring all commercial banks to invest a minimum of 50% of their savings deposits in the form of either longer-term government securities or housing loans. The regulation had no distorting effect on the banks' portfolios, however, since it was introduced shortly before the initial boom in housing prices and housing construction of the early 1970s when the commercial banks needed little additional incentive to expand that portion of their portfolios. 2.33 Other subsequent government restrictions have not been so benign, particularly those on the maximum rate of interest that can be charged on certain types of mortgage loans. Government assists buyers of lower to medium priced homes by placing an interest rate ceiling of 9.5% on the financing of units costing less than M$60,000, and 11% for units costing M$60,000 to M$100,000. Although interest rates on these loans are variable, adjustments are infrequent. In late 1987, when market rates for loans above the ceiling were 13% to 15%, the spread between uncontrolled interest rates for large mortgages and the lower rates for directed loans was 3-5%. Since housing finance institutions are often forced to borrow short and lend long, the controlled rates could jeopardize their financial strength if inflation (and the cost of bank borrowings) were to increase. To protect themselves from such risk, the banks have sometimes had to limit their lending for housing. In the late 1970s and early 1980s, for example, when housing prices were increasing most rapidly, the government-imposed - 15 - lending restrictions appear to have been a constraint on financial intermediaries, whose marginal borrowing costs during those years sometimes exceeded the stipulated maximum lending rates. The banks consequently limited their housing lending and, as illustrated in Figure 2.2, new housing loans lagged se-erely behind growth in the value of new housing in 1979 and 1980, the years of accelerating inflation and higher than average nominal interest rates. Funds are thus rationed by quantity, not by price. 2.34 At the same time, a number of recent policy developments in housing finance have been commendable. One has been the introduction of adjustable rate mortgage instruments, which provide lenders with a way of reducing interest rate risk and with the ability to continue to make mortgage loans during periods of higher than aver-ge inflation. An optional graduated payment mortgage was also introduced in connection with SLCHP; this enables borrowers to make smaller loan repayments early in the life of a mortgage and to make increasing payments over time as their incomes and repayment abilities presumably increase. These and similar lending instruments are preferable ways for ensuring that low-income families have access to housing finance than the interest rate restrictions still imposed. 2.35 Another improvement has been the establishment in 1986 of a secondary mortgage market institution, Cagamas Berhad, which provides a market in which primary lenders for housing can securitize their mortgage holdings so that they can be resold to other investors--such as other parts of the banking system, trust funds, insurance companies, and the general public. In so doing, primary housing lenders are able to sell some of their interest rate and liquidity risks to other investors, and thus to continue to make end-financing available to housing purchasers. The Cagamas Berhad was established with an initial capital authorization of M$200 million, of which M$50 million was to be paid in.5/ Shareholders of the institution include the Central Bank, commercial banks, finance companies, and merchant banks. The chairman is the Governor of the Central Bank. 2.36 The housing finance sector is thus in a good position to continue its recent rapid development, assuming that its financial viability is protected from the interest rate restrictions which in other countries have led to rapid decapitalization and financial collapse whenever inflation is high. It is therefore recommended that the existing interest rate restrictions be curtailed and that equity objectives of the Government be achieved through policy reforms discussed later in this report. 2.37 Housing Finance Subsidies. Commercial lending institutions bear the cost of housing finance subsidies provided through the Government's directed credit policies and the SLCHP, while the Government bears the cost of housing finar,.e subsidies provided under the PLCHE and the Treasury Loan Program for government employees. Table 2.8 shows the current terms of these programs as well as the subsidies involved, by loan type and in aggregate. 5. Cagamas made its first purchase of housing loans in October 1987 from three commercial banks, and at the same time placed a corresponding amount (M$110 million) In bonds in the money market. The Cagamas issues are attractive since the institution that sell the mortgages to Cagamas continues to bear responsibility for loan administration of the original mortgages and Is obligated to reimburse Cagamas should loan become non-performing. - 16 - Table 2.8: Present Valu of Selected Covernm nt Houston Loan Prosrams 1. REPRESENTATSVE LOANS finanelal SubsLdy a1 larket Value of harket If Lending Rate Unchane^i Sub*ldysbl FinancLal Valued Current Term And Harginal Cost If karket Rate io: Subsldy Subsidy of Repreaentative Loan of Funds Lot Os X u S of Loan of Loan Loan Loan Current Loan Grace * 6 a 8.5 9.5 11.5 I 6.0 I 9.5 Type nAount Rate Term Perlod Percent Percent Parcent Percent Percent Percent Percent Percent Treasury A 81.-; 00 *.02 25 0 ($0) $27,257 $47,503 $51,733 $59,382 $71,999 18X 40X Treasury 8 $70,000 4.02 25 0 ($O) $12,720 $22,148 $24,142 $27,712 $33,599 182 40o Treasury C $70,000 *.02 25 0 ($o) $12,720 $22,168 $24,142 S27,712 $33,599 18S 402 Treasury D $30,000 *.03 25 0 ($0) $5,451 $9,501 $10,347 $11,876 $14,400 lX *0X PLCHP $25,000 5.51 25 2 ($1,672) $3,738 S7,735 68.567 $10,067 $12,529 15S 4OX SLCHP $25,000 8.82 25 0 ($13,958) ($6,879) ($1,621) ($522) $1,465 $4,741 (282) 62 Directed Cdt $60,000 8.8X 20 0 ($27,741)($14,051) ($3,387) ($1.096) $3.106 $10,225 (232) 52 II. ESTIMATED AGGREGATE SUBSIDIES for 1987 (valued at market prices) Market-valued Share Aggregate Loan Subsidy Aggregate Percent of Program Of Loans Per Annum (Est.) Per UnLt Subsidy All SubsidLes Treasury A 122 $133 milLion 402 $52 milLion 10X Treasury B 82 $88 millon 402 $35 million 62 Treasury C 5ox $553 milLion 402 $219 million 412 Treasury D 302 $332 miliLon 402 $131 million 242 SUBTOTAL: 1002 $1.105 million $437 miliLon 81X PLCHP $100 miliLon 40X $40 million 72 SLCHP $200 milLion 62 $12 million 22 Dlrected Credit $1,000 million 55 $52 milion 102 TOTAL: $2,405 m illon $541 milLion 1002 Note: numbers in parenthesLs are negative values. a/ Financlal subsidies are calculated as the present value of the difference between the landing rate under the various programs and the Governmeant's marginal cost of funds. In this table the marginal cost of funds are assumed to be 42, 62 and 82. In practice, the funds raised have a variety of maturities and terms. In additlon to the direct transfers to the TLP from the Federal budget and repayments, which should be valued at the Treasury's margLnal cost of funds (presently 4-52), the TLP bas borroved from the Imploy ees Provident Fund (EPF) at 8.52 interest and from comnercial banks at somewhat higher rates. In 1988 CAGAMAS purchased M$750 milLion of TLY loans at a cost equivalent to 62 p.a. bI The market value of subsidies to the borrower are measured by the difference betvwen the lendLng rate under the specific program and the market rate of lnterest for sLmilar loans. The market rate measures the rate at which the Government could have lent the funds, i.e. the opportunity cost of the resources used in the progra's. ThLs is how economists define subsLdies, although *stimating financial subaidies Is also an lmportant exercLe. - 17 - 2.38 Loans provided under directed credit arrangements carry subsidies of about 5%. Since loans totaling about M$1 billion are annually extended under directed credit arrangements, the related subsidies total about M$52 million annually and represent about 1P% of all housing credit subsidies provided each year. SLCHP loans currently have a term of 25 years, with no grace period and an interest rate of 8.8%, indicating a subsidy rate of about 6%. Total annual SLCHP loans are estimated at M$200 million, including a subsidy element of M$12 million or 2% of annual housing credit subsidies. PLCHP loans have far more generous terms of 25 years, a two-year grace period and an interest rate of 5.5%. While the subsidy on individual PLCHP loans is very high, at about 40% of face value, they are not regarded as a major drain since the program is relatively small, at about M$100 million a year, the loans for M$25,000 are much smaller than those for all other programs except the SLCHP (which also offers loans for M$25,000) and the program itself is in its last stages. 2.39 The Treasury Loan Program (TLP). The TLP provides housing loans to civil servants at below-market rates. Like such schemes in a number of coun*lies, the TLP is considered part of compensation for government service, and nLt primarily as housing policy per se. The TLP began in the early 1970s as a replacement for the former practice of directly providing government housing to civil servants at no (or low) cost. It was introduced during a period when government wages and salaries were not keeping pace with private sector compensation. Some private sector companies also use below-market-rate mortgage schemes to compensate staff since such schemes reduce the income tax burden on employees (below-market financing is not taxed) and. help retain valued personnel. 2.40 Unlike the PLCHP and SLCHP mortgage subsidy programs discussed above, the target recipients of TLP loans are middle class, although within the civil service staff all levels participate, and as Table 2.8 shows, about 80% of the loans (and 65% of subsidies) are for category C and D staff who are paid the lowest salaries. Of the roughly one million civil servants, military and police potentially eligible for such loans, about 25% have received them. 2.41 The TLP expanded rapidly through the early 1980s and, as shown in Table 2.9, in some recent years has represented as much as half of the total formal housing finance available. Thus, while primarily a compensation policy, it is also an important share of formal housing finance. Maximum loan amounts under the program vary with the civil st-vice rank of the recipient. Loans are made for 25 years at an annual interest rate of 4%; the rate is adjustable at the Treasury's option. Until recently, the interest rate structure was progressive; larger loans were made at 6% while loans for M$30,000 and M$70,000 had an interest rate of 4%. Current TLP contracts specify that interest rates are variable, and when (as last year) terms are changed, they are applied to pre- existing loans as well as new loans. However the one attempt at increasing the interest rate was short lived and was rolled back to previous levels within little more than a year. 'this has important implications for the government budgat. If the Government's marginal cost of funds rises, there is a potentially large contingent liability if the higher costs are not quickly and fully passed on to the final borrowers. - 18 - Table 2.9: Comparison of Treasury Housing Loans and Total Housing Finance CMS '000) Treasury Loans Total Housing Finance TLP as Net Net Cumulative Net Cumulative X of Total Year Housing Finance aI 1980 1,103 4,941 - 1981 990 2,093 2,011 6,952 49 1982 1,266 3,359 2,440 9,391 52 1983 871 4,230 1,876 11,267 46 1984 1,083 5,313 2,459 13,726 44 1985 1,110 6,423 2,739 16.464 41 ___--__--__________________________________________________________________________________________________-_ a/ Net is approximates no account is taken of repayments (few if any loans have been repaid). Source: J. Mohamed, Country Report of Malaysia (17th IUBSSA Conference, 1986), p. 5. 2.42 Table 2.8 presents estimates of the financial cost to the Government of subsidies provided under the TLP and the market value of the subsidies under alternative assumptions about interest rates. Assuming a marginal financial cost of funds to the Government of 6% and an opportunity cost of capital of 9.5%, on average, the loans contain a financial subsidy element of 18% and a market valued "economic" subsidy of 406 of the loan amount. With TLP lending at about M$l billion a year, the average market-valued subsidy is over M$400 million a year. The corresponding financial cost to the Government of the subsidies involved (18%) total about M$200 million per year. The subsidy does not appear explicitly in the government budget. Every time another Treasury loan is made, the Government must take on additional debt (or in surplus years fail to retire an equivalent amount of debt). Thus, the subsidies vary with the Government's marginal cost of funds and the opportunity cost of capital as long as the loans remain on the books. 2.43 Given the depth of the subsidy accorded TLP borrowers, it is not surprising that the analysis shows that the program may have contributed significantly to upward pressure on housing prices. This pressure is likely to have been most _cute during 1981 and 1982, the years when the volume of lending expanded most rapidly, and the years when the price of newly built housing increased by 23% and 9%, respectively, compared to consumer price increases of only 10% and 6%. Although efforts have been made in the last few years to direct the TLP to lower-income civil servants and to inierease interest rates under the loans, additional changes based on the housing sector impact of the TLP are worth further study. Perhaps the most politically feasible reform to the TLP would be to further reduce the risk of providing uncontrolled subsidies by making the interest rate adjustment automatic6', instead of an option which may be exercised by the Government. 6. A convenient formula might be to fix the ate in reference to the BLR. - 19 - 3. THE HOUSING MARKET A. Introduction 3.1 The Malaysian housing market seems to be operating satisfactorily. The supply ratio has increased to an estimated 90% (para. 2.4), and the quality of shelter is gocd (paras. 2.7-2.8). Although the ccst of purchasing a new house is relatively high and affordability seems to be an issue, rental housing is widely available and the government has intervened in an attempt to make the purchase of a new house affordable to the lower-income population through special st.bsidized housing development and finance programs. How well these programs ( and the policies they reflect) perform will partly depend on how well they were formulated. The key questions are (a) did the Government adequately understand the housing market and the fundamental problems limiting home ownership when it devised it policies and programs to expand low-cost housing and (b) did its policies and programs address those problems or did they respond to peripheral factors that superficially seemed to limit the production of low-cost housing? This chapter attempts to answer tlese questions by reexamining the data sources on which policy inferences were drawn in the past. In particular, it revisits the supposed decline in housing production in the early 1980s which prompted the use of housing as a countercyclical instrument of demand management. In addition the chapter looks at reestimates of housing investment data to ascertain the sector's contribution to economvwide performance and national housing objectives. In view of the findings of this analysis, that investment levels have been quite high while quantitative output has been somewhat low in comparison to countries at a similar stage of development to Malaysia's the chapter continues by examining the factors which seem to have directly contributed to the rapid increase in housing prices (and thus the inability of large segments of the population to afford new housing),, and highlights areas for possible reform. B. Housing Production and Investment Levels Housing Production 3.2 Government statistics on housing completions during the Third and Fourth Malaysia Plans (1976-80 and 1981-85, respectively) suggest that the number of housing units completed during the Third Plan was 484,190, of which 121,510 (25%) were built under the auspices of the public sector, and 362,680 (75%) were built by the private sector.1/ During the Fourth Plan, an estimated 391,924 units were built, of which 190,051 (48%) were publicly sponsored and 201,873 (52%) were privately built. These figures suggest that housing 1. The division between private and public sector is defined on the basis of the agency responsible for undertaking the production, but not necessarily the financing, of housing. Among the public housing programs included here are the PLCHP, various prograns of federal agencies and regional development authorities, institutional quarters of the Public Works Department and other departments, housing provided by the Aborigines Department and the Sabah and Sarawak Development Boards, and housing provided by the State Economic Development Corporations, the Government OMiEcers' Housing Company, and other mnor housing program. - 20 - < 3 A closer examination of the official figures reveals, however, that the data provided for t1;e two plan periods are not comparable and that the reported fall-off in production may not have occurred. In most reported statistics, completions during the Third Plan were based only on reported approved projects, with an assumption that approved units would be completed 18 months after project approval. In the Fourth Plan, "completions" were based only on those units reported by developers to have been completed. Because developers were under no obligation to report completions of approved projects, the latter procedure probably undercounts actual completions. Table 3.1 indicates that when a comparable method is used for calculating housing completions in the two periods, housing production during the Fourth Plan period is estimated to have Increased by about one tenth over the level of the Third Plan period (to 536,927 units). Table 3.1: Housing Completions 1976-86 Public Annual Private Annual Annual Private Date Sector Chg(Z) Sector Chg(S) TOTAL Chs(S) Sharem) Third Plan 1976 17,801 54,362 72,163 75.3X 1977 21,230 19.32 57,604 6.01 78,834 9.22 73.12 1978 21,460 1.1X 67,574 17.32 89,034 12.9X 75.92 1979 24,655 14.92 82,181 21.6Z 106,836 20.02 76.92 1980 36,364 47.52 100,95S 22.82 137,323 28.52 73.52 Subtotal 121,510 362,680 484,190 74.92 Fourth Plan 1981 31,009 -14.72 78,537 -22.22 109,546 -20.2Z 71.72 1982 43,474 40.2X 78,925 0.52 122,399 11.72 64.52 1983 35,056 -19.42 61,300 -22.32 96,356 -21.32 63.62 1984 43,482 24.02 53,064 -13.42 96,546 0.22 55.02 198S 36,030 -17.12 76,050 43.3X 112,080 16.12 67.92 Subtotal 189,051 347,876 536,927 64.82 Fifth Plan 1986 32,990 -8.42 63,974 -15.92 96,964 -13.52 66.02 Notes to the table: Private sector completion are based an approved plans for private developers, and assumes that approved units are completed in 18 months (until Juns 1982) and in 24 months thereafter. Source:Ministry of Housing and Local Development. 3.4 Production under the Fourth Plan has been re-estimated, using approved units as the basis for calculating completions. For the first two years of the of the Fourth Plan period, the assumed completion time is 18 months -- the same period used in calculating Third Plan housing completions; from 1983 to 1985, the assumed completion time is 24 months to reflect the construction slow-down that occurred during this period. Because approved units may not always have been completed, the recalculated figures may be somewhat overestimated and therefore should be viewed as only indicative. This is particularly true for 1985-86 when it is generally believed that construction sector activity turned sharply downward, while the re-estimate shows increased or steady production. It is nevertheless likely that the fall-off of housing output during the Fourth Plan, as shown in the official figures, is at least overestimated and very possibly did not occur at all. The uncertainty in assessing production indicates the need for greater consistency and an improved methodology in calculating housing completions and compiling sectoral data. - 21 - 3.5 As indicated in both the official and unofficial statistics and illustrated in Figure 3.1, the share of public sector housing production has been increasing during the last decade. The largest share of this increase has been in the production of low-cost public housing, which increased from about 26,000 units during the Third Plan to over 72,000 units during the Fourth Plan; the other major area of increased public production has been in "medium- and high-cost" housing, a mix of government officers' housing, housing provided by State Economic Development Corporations, and other programs which, together, expanded from about 38,000 units in the Third Plan to above 58,000 units in the Fourth Plan. Figure 3.1: Housing Completions 1976-86 130 / 120- 100 1 190 19818 9218 9 70 so 40- 30- 20 - ¶0 975 19975 19Is0 ¶82 19184 9 Year a Publc Sector\ 4 Pflvata Sector ° Total 3.6 In assessing the level of housing production, consideration should also be given to. the considerable amount of informal sector housing built by individuals or small developers. The 1980 census reports that the number of housing units built in the five-year period between mid-1975 and mid-1980 was 798,000 or 65% above the reported Third Plan formal-sector production of 484,000 units. These figures siggest that for every ten units officially counted in housing production statistics, six units are not counted.31 It is presumed that many of these uncounted units are informal or possibly extra- legal units, although direct evidence of this is not readily available. S. The censu methodology differs from the methodology used to calculate official production figures In that It uses information on the age distribution of units and on the total number of units observed in different census periods to estimate rates of demolition and replacement (see General Renort of the Housini Census 1980. pp. 58). The official statistics reflect information furnished by developers and by localities which involved in the formal proces of appUcatioa for approval of new housing projects. - 22 - Housing Investment 3.7 The level of housing investment in Malaysia may also be underestimated in the official figures. 'Residential construction" is a line item in the national income accounts, and is calculated on the basis of the value of inputs to residential construction, such as labor, materials, and profits, without specific knowledge of either the number of housing units produced or of their unit value. However, the figures presented appear to be at variance with other evidence on the value of housing investment, which indicates a serious underestimation of the true economic importance of the sector. Table 3.2, Column 7 indicates the annual level of investment in residential construction as reported in the national income accounts, and Column 8 reports the value of residential construction as a percentage of GNP. Both figures appear low.4t 3.8 Two pieces of evidence raise questions about the reliability of these data. The first is a comparison of official data for Malaysia with data on housing investment in other countries at similar levels of economic development. The second is a comparison of the national income accounts data to a statistical series on housing investment that is ca'.culated using information on housing completions in Malaysia and unit valu 'f recently completed housing. 3.9 Figures 3.2 and 3.3 illustrate the way in which housing investment varies in relation to GDP per capita in countries at various levels of economic development. Such comparisons clearly depend on the adequacy of data used in constructing both national income accounts and housing investment statistics, and thus should be interpreted cautiously. As the figures indicate, the fraction of GDP that is invested in housing first rises with GDP per capita (to a per capita level of about US$8,000 in 1960 dollars) and thereafter falls. During the mid-1970s, for countries at Malaysia's level of GDP per capita, i.e., about US$1,200 (in 1980 dollars), the expected level of investment in housing relative to GDP (shown by the plotted curve) was about 3.6%, or from about 50% to 80% above the level of investment reported in Malaysia's national income accounts. By the early 1980s, at Malaysia's level of GDP per capita of about US$1,500, housing investment would have been expected to be about 3.5% of GDP, a level that is generally between 10% and 30% higher than the reported levels based on the national accounts data. A comparison of the official data for Malaysia with that for other Asian countries further suggests that the reported figures for Malaysia are on the low side. In 1976, for example, reported housing investment as a percentage of GNP was only 2.1% in Malaysia but 3.3% in Hong Kong, 3.6% in Korea and the Philippines, and 5.1% in aingapore. By 1981, Hong Kong invested 4.2% of GNP in housing, Korea 3.4%, the Philippines 4.0%, and Singapore 4.4%, all exceeding the reported figure for Malaysia of 3.2%. 3.10 The level of housing investment in Malaysia increases substantially when calculated on the basis of housing completions and their unit values. Column 1 of Table 3.2 gives the sale prices of new houses built between 1976 4. Over the period indicated, the ratio of GNP to GDP was about 0.96. To convert estimated ratios of houing Investment to GNP to ratios in terms of GDP, figures In columns 6 and 8 may be multiplied by 0.96. - 23 - and 1986, based on detailed Property Market Reports published by the Ministry of Finance.5' When these figures are multiplied by the reported completions, as adjusted according to the procedure described in para. 3.4 and given in Column 2, the gross value of housing output is obtained. This is shown in Column 3. As in the case of the national income accounts, the value of land has been netted out from the estimated value of housing output. Here it has been assumed, somewhat arbitrarily, that the cost of land comprises 40% of the value of completed housing. This figure is representative of Manila during roughly the same period, and may also be approximately correct for Kuala Lumpur and a few other cities; on average, however, the figure is likely to overstate the actual proportional contribution of land to total housing cost for most Malaysian cities and towns. It should also be recalled that the figures on housing completions on which these estimates are based consider only formal sector housing, and thus omit an estimated 35%-40% of all residential construction. 5. The housing value seris In Column 1 was constructed by taking an unweighted average of the sales prices of single- story terrace house for a sample of 6 and later 12 cities and towns. Such hou are not only the most common type of house for which prices awe reported, they are alo consistently the leat expensve type of housing for which prices are reported In the Pronertv Market Reports. Constructing a weighted average wa not posible ta consistent data were not available on the proportional distribution of unit types. In any ca", the method actually used gives . conservative estimate of new housing values since other types of units re on average more expenive than single-story terrace houes. It should also be noted that the figures reported for housing sales In the Pronertv Market Reports re based on reported transactions of both new housing and resales. 6. In principle, estimates of GNP originating in the housing sector based on the official statistics, which value contemporaneous factor inputs to housing construction, and baed on the procedure followed here, which reflects the value of contemporaneous sales net of land values, shouid on avinm produce similar reubts. When housing takes several years to complete, the value of housing investment estimated by the two methods wil be Identical if output levels are constant but will differ if output levels fluctuate. In periods of Increasing construction, the value of constructlon Inputs will exceed the value of sales; in periods of decreasing construction, the reverse wi be true. Tabte 3.2: Housing VaLues, Completed Units, and Value of Residential Output (1) (2) (3) '4) (5) (6) (7 (8) (9) Value of Value of Est.Housing Officiat Official Official HSg Inv Housing Housing Output Housing GNP Investment/ Hsg Const HsgConstGIGP Estimted Hsg In Date Value Coipletionns (billions)) (billions) (billions) GNP(percent) (biltions) (percent) (percent) 1976 N S28,800 72,163 2.078 1.247 26.988 4.6X 0.563 2.1X 45S 1977 N 130,600 78,834 2.412 1.447 31.064 4.7X 0.803 2.6X 552 1978 " S35,300 89,034 3.143 1.886 36.186 5.2X 0.808 2.2X 43X 1979 N 143,500 106,836 4.647 2.78t 44.354 6.3X 0.948 2.1X 34X 1980 N S59,700 137,323 8.198 4.919 51.390 9.6X MA NA KA 1981 N $73,600 109,546 8.063 4.838 55.602 8.7X 1.799 3.2X 372 1982 N $80,300 122,399 9.829 5.897 59.690 9.9X 2.153 3.6X 37X 1983 N 180,800 96,356 7.786 4.671 65.154 7.2X 2.088 3.2X 45X 1984 N 183,800 96,546 8.091 4.854 74.182 6.52 2.304 3.12 47X I 1985 N s79,700 112,080 8.933 5.360 72.039 7.42 2.083 2.9X 39X N) 1986 N $70,100 96,964 6.797 4.078 66.364 6.12 NA MA NA Notes to the tabte: Average 42X Col(1): Average vatues ofingte story terrace houses for a sample of cities. Source: Ministry of Finance, Property Market Report, Volumes for 1986-1986. Col(2):Public and private housing coqlpetions. Source:Ninlstry of Hog and Local Government Col(3):Col(1) times Col(2); the value of housing completed. Col(4):Value of output net of land, where land value is assd to be 402 of housing value. Col(5):Sources:Ministry of Finance,Economic Report,1987/1988 and World Sank Nalaysia: Industrializing a Primary Producer,Vol.2:Statistical Appendix. Col(6):Col(4)/Col(5):estimated housing investment to GNP. Col(7):Source:Department of Statistics,"Principat Statistics of Large Construction Establishments." Col(8):Col(7)/Col(5):"Official" Housing Investment/Estimated Housing Investment. - 25 - Figure 3.2: Housing Investment and GDP (mid-1970s) 11 - a 10 - 7~~~~~~~+ - +@ ++ **0+l% 3 - ",MO ,r O ° +* 1 +* 0 0 oqg 00 o Malaysia a~ ~ ~ ~~~~~~~~~~~~ - o Israel 2- 0 2 4 6 8 10 12 14 (Tho1usands) GOP per copBta (198OS) Figure 3.3: Houging Investment and GD? (early 1980s) 10 - 0Cyprus 8 0~~~~~ Israel a.~ ~ ~ ~ ~ ~ ~ ~~~~~~~ 0 0 +400 0+ 4 - 0 *400 o0*4. N+ + C3 *4e 000 2 - a 0 1 00 0 - , I I I I I I I I I 0 2 4 a 1 0 1 2 1 4 CThousands) GOP per capita (1980$) - 26 - 3.11 Adjusting the estimates of housing value of Column 3 for land costs, estimates of the value of housing investment are shown in Column 4. The fraction of GNP represented by housing is estimated in Column 6, which indicates that prior to 1980, 4.6% to 6.3% of GNP went to the housing sector. In 1980, housing investment jumped to 9.6%, and since that time has fluctuated between 9.9% (in 1982) and 6.1% (in 1986). With the exception of 1986, in no year since 1980 has housing's share of GNP been below the highest fractional level attained during the 1970s. In comparison, the average housing inveLtment reflected in the national income accounts data (column 9) is only 42% of the estimates provided here.71 3.12 A comparison of the revised investment figures with other countries' housing investment shares as shown in Figures 3.2 and 3.3 indicates that while Malaysia's level of investment in 1976 was relatively close to its predicted level, by 1981 housing investment was well above the level that would have been expected, considering its level of income. In fact, among the 39 countries evaluated, only one, Cyprus, appears to have invested a higher fraction of its GDP in housing in the early 1980s. Thus, Malaysia's level of housing investment relative to GNP has been consistently above the level that would have been predicted on the basis of international experience since the mid- 1970s. 3.13 To a certain degree, such performance can be explained by two factors that would lead one to expect higher levels of housing investment for Malaysia than for other units Guideline (1) (2) (3) (4) Plot Area m2 68 68 68 68 68 Floor Space per plot m2 42 42 48 53 53 % Land Salable % 44% 55% 47% 55% 55% Site Floor Area Ratio 0.23 0.29 0.33 0.41 0.42 House+Plot Sale Price M$ 19,200 17,100 19,300 19,300 19,300 Profit per hectare increase or decrease as compared to the profit/ha for middle income plots selling -15% -4% 2% 17% 20% for M$ 42,000 Note: See Anrex 3, Tables 3 & 4. Figures 1 & 2 in Annex 3 shov layout for SLCHP and iteration (3). - 79 - 5.28 The selection of. low-cost land-use standards. The target values for the floor-area ratio and for the percentage of salable land should be selected before deciding on the value of the other parameters. The most appropriate values are the ones which would insure that low-cost housing would have a safe profit margin over alternative residential uses in areas where the demand is high. When these values are established, say 55% for salable land and 0.37 for the floor-area ratio, it is possible to set detailed standards for roads, utility areas and community 'facilities. Trade-offs would have to be made among various standards, and an infinite combination of values is possible, th.e only constraint being that the surn of the non-residential use should not be more than 45% (in the case where 55% has been selected for salable area). Hard choices might have to be made in allocating land between parks, schools and roads. But those choices would also stimulate the use of ininovative designs for site layout. The new guidelines have already shown the way in that direction in advocating that part of the area for school sites and buffer zones of oxidations ponds should be designed to allow their use as public recreational spaca. 5.29 The selection of approRriate engineering standards. The reduction of unit cost would also require a systematic review of all engineering design standards, to insure that no unreasonable trade-off is made between capital cost and maintenance costs. Local authorities should be encouraged to explore ways to improve the productivity of their maintenance tasks. For instance, the use of modern equipment for the maintenance of sewer pipes can result in important savings in land (reduction in the width of back lanes) and civil works. If necessary, a program should be set up to help finance the modernization of maintenance tasks. 5.30 Creating incentives for local authorities. The development of new low-income areas within a local authority's boundary should produce additional resources. Those additional resources should come from, (a) a modified road grant system, (b) property tax, and, eventually, (c) a maintenance fee. To insure the support of local authorities, low-income areas should produce the same amount of revenue per unit of land as middle-income areas, while the amount of tax collected per household should of course be smaller. This would be possible if the density in low-income areas were significantly higher than in middle- income areas, i.e., if the percentage of salable area and the floor-area ratio were higher. The formulae through which the road grant system is calculated and the property tax collected should be reviewed. It would be necessary to insure that the formulae for the grant and the tax result in the same revenue whether the area is low-income or middle-income. 5.31 Improving and disseminating better site Rlan design Rractices. Private and public developers would have to improve the quality of site plan designs to fully capture the potential benefits (lower costs and higher profits) which the amended standards would provide. This process could be acceli-rated if schemes could be designed to demonstrate that higher land-use efficiency which could be achieved, and the consumer acceptability of different design patterns. Simultaneously, the local officials responsible for site plan approval would bave to familiarize themselves with these new designs in order to avoid penalizing any design innovation by a lengthier approval process. Better site planning methods, linking design, planning and costs should be disseminated through - 80 - professional associations such as the Malaysian Institute of Town Planners. The participation of a qualified professional familiar with the current land subdivisior. legislation should be mandatory for the subm'ssion of planning approval for large site plans. 5.32 The financing of land for large community facilities. The financing of land for large community facilities (schools, main places of worship, etc.) should not depend on scheme size, but on a development tax on all land in the process of being developed. All residential developments should pay a fair share of the cost of community facilities irrespective of their size or density. This would avoid the penalization of residential schemes V.,ere low-income plots are dominant. The tax could be made progressive by taxinig the larger consumers of land at higher rates. This method would have the added benefit of giving a '!igher cost to raw land and therefore to encourage the use of innovative site design techniques to save land. The cost recovery and the provision of land for small community facilities, such as parks, playgrounds and small worship places, should be made using tht existing formula, although the value of the parameters setting the space standard c ld be reviewed to stay within the agreed target for salable land and floor area ratio. - 81 - Table 5.5: Adjusted Standards to Achieve 55% Salable Land Under Present Possible Way of SLCHP guidelines Reaching 55% salable m2/plot % m2/plot % A. Residential Area (salable area) Plot Size 68 44.26% 68 55.11% B. Community Facilities Parking 3.72 2.42% 0 0.00% Surau 7.33 4.77% 7 5.67% Kindergarten 1.29 0.84% 1.29 1.05% Playlot 7.51 4.89% 5.5 4.46% Oxidation Pond 11.59 7.54% 10 8.10% Total Com. Facilities 31.44 20.47% 23.79 19.28% C. Road Area Roads serving community facilities 12.86 8.37% 7.45 6.04% Roads serving plots 41.32 26.90% 24.14 19.57% Total Roads 54.18 35.27% 31.59 25.61% Total Area Required per Plot 15' 62 100.00% 123.38 100.00% Detailed calculations of road arsa for 100 plots Roads Serving Community facilitirs Street width 10 m 7 m Block width 40 m 40 m Block length 79 m 59 m Street Area 1,286 N2 745 M2 Road area as % of Community 29.03% 23.86% Facilities Roads Serving Plots Road area as % of plots 37.80% (1) 26.20% (2) N6ze: (1) s,e Figure 1 Annex 3 , (2) see Figure 2 Amnex 3 - 82 - Figure 5.4: Relationshig between Profit and Salable Area Cost/Unit when % Salable Varies 120- ; 110 - - - - range implied 'recommended range \00 F:SLCHP rules for low income 90 . hou, ing - i3so - 80 40 - a 30 1K,_---- medium standard plots 20 a * - ~~~-~ - - @ @ _ minimum standard plots lox 204 3S 3;X 401X 45X3S GOS S 70X 75x ax a COdVU"1t Co t ° COa w U d.Cwt .Profit/ha when % Salable Vories 210 _-i medium standard plots I - - -S * * ISO ,__> , >g, ^ aminimum standard plots 170 - 160 - 120 -- 0- - - 140 170 120- _ _ _ _ __ :__ _,_ _ ._ _. 100'- go. - .S, 10X I.;X 2;X .2 X 3;X 4X 4X 5SX $ GM% 70 7;xS S a;S asx t Of Sk A o PIOta LOW Coo Md.Cft - 83 - G. Sumary 5.33 This chapter has examined how the project level standards and practices in Malaysia affect the cost and output of new housing. The analysis concludes that land-use is the most importa-t aspect because standards, in combination with present practices, lead to inefficient use of land, which in turn, makes low-cost projects either risky from a marketing point of view because prospective buyers find the sites poorly located or less profitable than other uses for which the developer could put the land. 5.34 These standards and practices, which increase the cost per dwelling by using more land than necessary, also increase the capital and maintenance costs of infrastructure and ultimately drive up the price of all urban land for all users. The effect of the standards is to make the cost of land the principal determinanit in selecting a site for low-cost housing rather than to weigh various combinations of costs and benefits to determine the most attractive overall housing project attractive to prospective buyers and financially attractive to the supplier. 5.35 In identifying the standards which need to be changed, the key factor is not floor space or minimum lot size, both of which are highly valued by prospective buyers. Instead, this chapter finds road space and set-back regulations to be the major causes of the extremely small portion of salable land which determines developer costs and ultimately profits. High infrastructure standards in respect to maintenance and the threshold system of financinb -*mmunity facilities also provide incentives against the provision of low-cost housing for poorer households. 5.36 While the revised stzadards issued for the SLCHP are a step in the right direction, until local authorities allow the standards to be widely implemented, they will have 1%ttle effect. This chapter ends by suggesting that a revision of standards and practices aimed at reducing costs without reducing benefits to buyers should Fe formulated along the following lines. First, make it at least as profitable for developers to build low-cost housing as other types of housing by allowing more flexibility in the regulations for a number of project characteristics or standards to reduce costs on a given piece of land. Second, set standards using two general criteria, the percentage of salable land and the floor-area ratio *rather than the more than forty separate standards presently applied. Third, provide incentives to local authorities through the Federal Government's system of grants to encourage the approval of projects which have appropriate standards and for which mair.cenance and capital costs have been properly balanced. Fourth, finance community facilities in an equitable fashion so as not to distort the housing market. .h4~ ' - Annex 1 Page 1 of 7 MALAYSIA THE HOUSING SECTOR: GETTING THE INCENTIVES RIGHT Estimating the Long-Run Price Elasticity of Housing SuDRlY 1. The Basic Model Estimating the parameters of the long-run housing supply function can provide useful information on the degree to which the housing supply system is able to respond to shifts in demand. The model that we use to estimate housing supply parameters is derived from a simple competitive model of the housing market, in which equilibrium quantities of housing produced and housing prices are determined simultaneously by the intersection of housing demand and supply curves. The system that determines housing output and prices consists of three equations: a demand curve,a supply curve, and an identity. Following convention, one may specify both the stupply and demand equations in terms of loglinear functions. Housing demand is, in the simplest model, determined by household income, housing prices, and the general price level. Housing supply is determined by housing prices, end by input costs. In equilibrium, housing demand and supply ate equal to each other; this establishes an identity between the supply and demand equations. The simplest such system consists of the following equations: Demand ln Qd - a + Ey ln y + Ephd In Ph + Epod ln p0 (1) ln Qs - b + Eph. ln Ph + EpG ln pc ;2) Eguilibrium ln Qd ln Qs where, Qd - quantity of housing demanded Qs - quantity of hcusing supplied y - household income Ph - price per unit of housing po - price of "other" goods pc - price of construction inputs and the major parameters to be estimated are the constants a and b and the following behavioral parameters: Ey - Income elasticity of demand for housing Ephd - Price elasticity of demand for housing Ephs - Price elasticity of supply of housing. - 85 - Annexj Page 2 of 7 Eltimating garaMeters of the model If one observes directly the variables in the system, it is posLtible to estimate its parameters using standard econometric techniques for the estimation of systems of simultaneous equations. Equations (1) and (2) can be estimated using two-stage least squares, for example, where the quantity and price of housing are endogenous variables. In certain circumstances, however, it will be advantageous to estimate the model's parameters from a reduced fiom of the system rather than the two structural equations. The reduced form of the system is gotten by setting Equations (1) and (2) equal and solving for the observable variable, Ph, the price of housing. Doing so results in the following specification, which expresses nominal housing prices as the dependent variable, and includes other nominal prices and nominal income as independent variables: ln Ph b - a + Ec ln pc - By ln y - ___Rod ln po (3) Ephd-Ephs Ephd-Ephs Ephd-Ephs Ephd-Ephs In principle the results of estimating the structural equations and the reduced form should give comparable results, although the reduced form will not permit the underlying parameters of supply and demand to be estimated without bri-.tging to bear additional outside information. In the case of the current exercise, there are reasons for preferring to estimate the parameters of the system based on the reduced form equation, despite the fact that underlying parameters cannot be directly estimated. There is, for example, a general problem with developing country housing production statistics in that production magnitudes are often not well known. Sometimes this is a result of underreporting of informal housing which is undocumented. This is likely to '.e a particular problem in Malaysia and Thailand, for which equations are estimated below, though somewhat less of a problem in Korea. Moreover in the case of Malaysia, while the government reports statistics on formal housing production, the figures are calculated on the basis of housing starts which are then assumed to result in housing completions and thence sales after a stipulated lag. For figures reported in the main text, for example, the government has typically assumed that houses are completed in 18 months (later assumed to be two years) from their reported initiation. Even with relatively thorough reporting of housing starts, and because of the high rate of construction of informal housing in Malaysia this is far from the case, the assumption of a fixed period between the beginning and the end of construction is likely to be erroneous, and erroneous in such a way that attempts at estimating the structural equations of the housing market would almost certainly produce biased parameter estimates. This is so because the time period between initiation of construction and completion is likely to be an endogenous variable, such that any errors in measuring the actual rate of dwelling zompletions and/or sales will be correlated with the other variables included in the system. For example, during periods when the price of housing is particularly high relative to household incomes, sales may be slow and dwelling completions will lag. Thus the rate of completion, or the level of throughput of the system, will tend to be positively related to income and negatively related to housing price, each of which is a variable in - 86 - Annex 1 Page 3 of 7 the system. But becausa the system by which dwelling completions are estimated assumes a fixed period of construction, errors in measuring completions will be correlated with levels of the variables included in the system. When this is the case, parameter estimates will be biased. Notwithst:anding this problem, unless outside information can be brought to bear on the estimated reduced form equation, it will not be possible to estimate the parameters of the structural equations. Fortunately, the literature concerning housing demand in developing countries has become quite advanced in the past several years, so that it is possible to estimate the price and income elasticities of housing demand in most developing countries with considerable confidence on the basis of broad and robt.st findings concerning demand parameters in a variety of countries. Consequently, the procedure followed here is to estimate the reduced form equations, and to subsequently identify the key underlylng parameter of this investigation, the price elasticity of housing supply, based on parametric estimates of housing demand parameters in the literature. Specifically, if one estimates Eq.(3) in terms of the following equation: ln Ph- A + B ln pc + C ln y + D ln po (4) an estimate of the price elasticity of housing supply is obtained from the following relationship: A A AA Ephs EX + CEphd (5) A C where the parameter C is estimated econometrically as the elasticity of housing price with respect to household income and the two parameters Ey and Ephd are estimated on the basis of external evidence. In thin report, information on the price and income elasticities of demand for housing is taken from a cross '-untry study of housing demand conducted by Malpezzi and Mayo at the World Bank.' The simplest of their cross-country models is a log-linear demand model which expresses housing expenditure by renters and owners as a function of income and the relative price of housing, where the latter variable was constructed using data from Kravis, Hcston and Summers (1982). Defining R as rent, y as household income, and PH as the relative price of housing, Malpezzi and Mayo estimated the following models for renters and owners in developing countries: See Stephen J. Malpeazi and Stephen K. Mayo, Housintc Demand in Develoinir Countries, World Bank Staff Working Paper No. 733, 1986, which reviews the literature on housing demand and presents empirical evidence on housing demand parameters in some 15 developing country cities. - 87 - Annex 1 Page 4 of 7 Renters: ln R - - 5.39 + 1.60 ln y + 0.15 in PH (0.18) (0.15) - 0.90 d.f. - 13 (standard errors in parentheses). Qwnffers: ln R - 3.57 + 1.38 ln y + 0.65 ln PH (0.35) (0.50) R2 - 0.76 d.f - 11 where rent, and income are city means converted to 1981 U.S. dollars2, and pH is the Kravis-Heston-Summers price index, with the U.S. relative price normalized at ene. The implications of these models, which were confirmed with alternative specifications, are straightforward. In the very long run, housing consumption is income elastic. Price elasticities are smaller in absolute value than income elasticities, although confidence intervals are quite wide for the former. Long- run income elasticities are estimated to be slightly higher for renters (1.60) than for owners (1.38). This means that as cities' economies develop over the very long run, that owner and renter consumption patterns increase at a similar pace, ceteris paribus. However, because renter price elasticities are estimated to be higher than owner elasticities, the net effect of both incomes and prices rising with economic development is that owners' consumption increases facter than re-.,ers' consumption over most of the range of the data. Price and income elast'city estimates from these models are used below in a parametric fashion to identify the price elasticity of housing supply based on Equation (5). The Data Data for estimating the models described above were collected for Malaysia, Thailand, and Korea. In the case of each country data were collected on the following variables, whose sources are listed: - Price of housing Malaysia - price series on single family terrace houses 'including both new and existing houses),constructed by the mission 2 Note that In a log-linear eipendituxe equrtlon the coefficient of price Is equal to one plus the price elasticity; thus the price elasticity Is the estimated coefficient minus one, or -0.86 and -0.35 for owners. - 88 - AnDAx x Page 5 of 7 based on data presented in volumes of the Property Market Reports published by the Ministry of Finance. Thailand - shelter price series, a component of the consumer price index; Department of Business Economics, Ministry of Commerce. Korea - price series on sales prices of housing; Ministry of Construction. - Income Malaysia - based on national income accounts estimates of personal consumption expenditures divided by estimates of household size; from Department of Statistics. Thailand - same procedure as Malaysia; from Department of Statistics. Korea - same procedure as Malaysia; from Department of Statistics. - Construction materials index Malaysia - based on the producers' price index for "crude materials"; from Department of Statistics. Thailand - index of building materials cost; from Department of Business Economics, Ministry of Commerce. Korea - index of building materials cost; from Department of Statistics. - Consumer price index Malaysia - Self explanatory; from Department of Statistics. Thailand - Self ex-,'anatory; from Department of Business Economics, Ministr) f Commerce. Korea - Self explanatory; from Departmsnt of Statistics. Results of the Model Tab'.e A.1 presents the estimated parameters of Equation (4), from which the housing supply price elasticity may subsequently be calculated. As the table indicates, the results of the reduced form equations indicate a uniformly high degree of statistical goodness fit. In the case of Malaysia and Korea, the two countries believed a priori to have more restrictive regulatory environments affecting the housing market, the coefficients of income are highly significant in either equation, indicating that controlling .or construction costs and for - 89 - Anney 1 Page 6 of 7 the general price level, income shifts are associated with increases in the price of housing. In Thailand, by contrast, the coefficient of income is insignificant. These results are consistent with the arguments presented in the text concerning the unresponsiveness of the housing supply system in Malaysia. The fact that similar results obtain for Korea and radically different results for Thailand should be seen as giving further support to the arguments advanced in the text concerning the unresponsiveness of the Malaysian supply system and the likely cause -- that of a restrictive and time consuming regulatory environment. While there is room for improvement in the statistical estimation of these reduced form equations, particularly with regard to including additional variables characterizing housing inputs and taking account of the possibility of serially correlated error terms in the estimating equations, it is believed that the results presented here provide strong support for the hypotheses advanced in the text. Table A.1 - Results of Reduced Form Supply Elasticity Estimates (Equation 4) Variable Malaysia Tha.and Korea Constant -1.778 0.556 5.367 LN Income 1.563 0.140 1.673 (0.320) (0.134) (0.334) LN Construction -0.074 -0.600 -0.808 Costs (0.138) (0.087) (0.493) LN CPI (Cther -0.120 1.343 -0.280 Costs) (0.492) (0.213) (0.733) 0.989 0.996 0.988 Degrees of freedom 1' 13 12 Note: Standard errors are shown in parentheses The results presented in the table may be used to derive housing supply elasticities, based on substituting the coefficients of the income variables into Equatton 5, along with estimates of the price and income elasticities of housing demand. As indicated above, the literature provides strong evidence of the similarity of these parameters in a wide range of developing countries. Consequently supply elasticities have been estimated here on the basis of what the literature suggests is a reasonable set of values for long-run price and income elasticities of housing demand. Point estimates of the housing supply elasticity are presented in Table A.2, based on an assumed value for the pri e - 90 - Annex 1 Page 7 of 7 elasticity of demand of 0.5 and upply elasttlty for two values of the income e6 elasticity that are representative 14 - of the Malpezzi/Mayo results presented above (1.0 and 1.5). For 12 , comparison, supply elasticity 10o estimates for the United States based on a similar reduced form estimation procedure are also 6 provided in the Table. Figure A.1 4 illustrates the sensitivity of estimated housing supply 2 elasticity estimates to a wider o range of estimates of housing demand parameters -- price -21 elasticities ranging from 0.0 to - 0 0.5 I 1.5 2 2.5 1.0, and income elasticity Assumed Income elsticity estimates ranging from 0.5 to 2.0. - al41ys(m. EghE"e Kornt. EphdtO.S*. ThIllandEphdO.0 As the figure indicates, the conclusions that parameters for Malaysia and Korea are similar, and that parameters for Thailand are quite different than those of the other two countries are extremely robust over a wide range of plausible demand parameters. Table A2 - Estimated price elasticity of housing suRply* Ey-l.0 Ey-1.5 Restrictive regulatory environments Korea 0.10 0.40 Malaysia 0.14 0.46 NonrestXictive regulatory environments Thailand 6.64 10.21 U.S.A. 22.03 40.04 The source of estimates and their derivation is given in the discussion above in this annex. The source of estimates for the U.S. in calculations from Table 4 In James R. Follain Jr., 'The Price Elasticity of the Long-Run Supply of New Housing Construction," Land Economics. 56, May 1979, pp. 190-199; and is based on a somewhat different functional form than that used to estimate parameters for the three Asian countries. PRESENT VALUE MODEL -- SUMMARY OF INPUT DATA LOCATION: Selancor Selansor Selangor Selangor X.L. PerIls Selangor Selengor Selangor TSst Selangor K.L. Slacngor Selangor PROGRAM: PLCHP PLCBP SLCbP SLCBP SLCBP SLCHP SLCHP PrLvate Private Renter SLCBP SLCHP IndividuaSLCBP LANDt PubiLc Public Public Private Private Private Private Prl ate Squatter Cse Private Private Private Private DEVELOPER: PINS Private Private Private Private Private Private Private Private Private Private Prlvate PrLvate Versions 1.1 STRUCTURE: Low Cost Low Cost Low Cost Low Cost Low Cost Low Cost Mod Cost Low Cost Low Cost Low Cost Low Cost Low Cost Ned-El Co Run Date: 22-Nov TENURE: Sales Sales Sales Sales Sales Sales Sales Sales Squatter Sales Sales Sales Sales Run TLme: 04:34 PM PLANNING STDS: Current Current Current Current Current Current Current Current Current Current Low Std Current UnplannedCurrent Sq. Clear INPUTS AND INTERMEDIATE CALCULATIONS Target Income $750 $750 $750 $750 $750 $750 $750 $750 $750 $750 $750 $750 $750 $750 Selling Price $25,000 $25.000 $25,000 $25,000 $25,000 $20,000 $70,000 $25,000 $5,000 $25,000 $25,000 $25,000 $30,000 $85,000 Market Value $20,000 $30,000 $30,000 $30,000 $50,000 $20,000 $70,000 $30,000 $5,000 $30,000 $30,000 $50,000 $30,000 $85,000 Annual Market Rent $2,538 $2,538 $2,538 $2,538 $2,538 $2,538 $3,600 $2,538 $600 $2,538 $2,538 $2,538 $2,538 $3,600 Rant to Value (-) 0.127 0.085 0.085 0.085 0.051 0.127 0.051 0.085 0.120 0.085 0.085 0.051 0.085 0.042 Controlled to Market Rent 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Legal Sales Price to Market Price () 1.25 0.83 0.83 0.83 0.50 1.00 1.00 0.83 1.00 0.83 0.83 0.50 1.00 1.00 Background Inflation 5.01 5.0X 5.02 5.01 5.01 5.01 5.01 5.01 5.01 5.0X 5.01 5.01 5.01 5.01 Real Increase in Land PrIce 0.01 0.0S 0.0S 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.02 Real Increase in Structure Prices 0.0Z O.0X O.0X 0.01 0.02 0.01 0.0S 0.02 0.01 0.01 0.01 0.01 0.01 0.01 Real Increase in Rents 0.02 0.01 0.01 0.01 0.01 0.02 0.0S 0.02 0.02 0.01 0.01 0.02 O.O 0.01 Real Discount Rete 10.01 10.01 10.01 10.02 10.01 10.0S 10.01 10.01 10.01 10.01 10.01 10.01 10.01 10.01 Deposit Amount $0 $0 $2,500 .0 $0 $0 $0 $0 $0 $0 S0 $0 $0 So Deposit Period 0 0 18 J 0 0 0 0 0 0 0 0 0 0 Value of Condesmed Unit $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $5,000 $0 $0 Fixed Search & Move Costs $1,000 $1.000 $1,000 $1,000 $1,000 $1,000 $1,000 $1,000 $1,000 $1,000 $1,000 $1,000 $1,000 $1,000 SIK Costs: S of Income 3.01 3.01 3.01 3.01 3.02 3.02 3.01 3.01 3.O 3.01 3.0S 3.01 3.01 3.01 Total Transactions Cost (-) $1,023 $1,023 $1,585 $1,023 $1,023 $1,023 $1,023 $1,023 $1,023 $1,023 $1,023 $6,023 $1,023 $1,023 Extra Transactions Costs (.) $0 $0 $563 $0 $0 $0 $0 $0 S0 $0 $0 $5,000 S0 so Plot Size 800 800 800 800 800 800 1100 800 400 800 800 o00 800 1400 Total LandlPlot Size 2.50 2.50 2.50 2.50 2.50 2.50 2.20 2.50 1.20 2.50 1.50 2.50 1.20 2.00 Acquisition Price of Landlaq ft $0.00 $0.00 $0.00 S4.00 $8.00 $2.00 $4.00 $4.00 $0.00 $5.00 $4.00 $8.00 $5.00 $4.00 Market Price of Larndlsq ft $64.00 $4.00 $4.00 $4.00 $8.00 $2.00 $4.00 $4.00 $4.00 $5.00 $4.00 $8.00 $5.00 $4.00 Conversion Premiumlsq ft $0.50 $0.50 $0.50 $0.50 $0.50 $0.50 $0.50 $0.50 $0.00 $0.50 $0.50 $0.50 $0.00 $0.50 Premdum Reductlon (S) 1001 100S 601 601 601 601 601 01 01 0S 601 601 02 601 Economic L-nd Cost (-) $9,000 $9,000 $8,600 $600 $600 $600 $726 $0 $1,920 $0 $360 $600 $0 $840 Finencial Land Cost (-) $0 $0 $400 $8,400 $16,400 $4,400 $10,164 $9,000 $0 $11,000 $5,040 $16,400 $4.800 $11,760 Slte Development Costs $3,000 $3,000 $3,000 $3,000 $3,000 $3,000 $3,500 $3,000 $1,000 $3,000 $3,000 $3,000 $1,000 $5,000 Infrastrw:ture Costs $5,600 $5,600 $2,000 $2,000 $2,000 $2,000 $2,000 $2,000 $1,000 $2,000 $2,000 $2,000 $500 $3,000 Site Devliopment Subsidy (1) 01 0S 01 02 01 0S 0S 01 O0 0S O0 02 0S O0 Infra,tructure Subsidy (S) 101 102 10S 10S 10S 102 101 101 02 101 101 101 01 102 Pro~f. Services Subsidy (S) 01 02 S 'S 5S S 2 5X 0S 02 02 52 52 01 51 Economic Cost of Plot (-) $17,600 $17,600 $14,000 $14,000 $22,000 $10,000 $16,390 $14,000 $3,920 $16,000 $10,400 $22,000 $6,300 $20,600 FLnanclal Cost of Plot (-) $8,040 $8,040 $3,950 311,950 $19,950 $8,200 $11,964 $13,800 S2,00C $15,8.0 $8,590 $19,950 $6,500 $15,210 Ix 0 PRESENT VALUE MODEL -- SUMMARY OF INPUT DATA LOCATION: Selangor Selangor Selangor Selangor KL. Perlis Selangor Selangor Selangor Test Selengor ..L. Selangor Selangor PROGRAM: PLCBP PLCBP SLCBP SLC8P SLC8P SLCHP SLCHP Private Prtvate Renter SLCHP SLCBP Ind1vJduaSLCSP LAND: Public Publie Public Private Private Private Private Private Squatter Case Private Private Private Private DEVELOPERt PKNS Private Private Private Private Private Private Private Private Prlvate Prlvate Private Private Version: 1.1 STRUCTUREI Low Cost Low Cost Low Cost Low Cost Low Cost Low Cost Med Cost Low Cost Lov Cost Low Cost Low Cost Low Cost Mad-8i Co Run Date: 22-Nov TENURE: Sales Sales Sales Sales Sales Sales Sales Sales Squatter Sales Sales sales Sales Run Time: 04:34 PM PLANNING SMDS: Current Current Current Current Current Current Current Current Current Current Lov Std Current UnplannedCurrent Sq. Clear Unit Size (sq ft) 705 705 705 705 705 705 1000 705 250 705 705 705 705 1000 Structure Cost/sq ft $20 $20 $20 $20 $20 $20 $25 $20 $5 $20 $20 $20 $20 $30 Materials Subsidies $500 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 Construction Finance Subsidies $1,000 $0 $0 $0 $0 $0 60 $0 $0 $0 $0 $0 $0 $0 Other Construction Subsidles $2,500 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 Economic Cost of Structure (-) $14,100 $14,100 $14,100 $14,100 $14,100 S14.100 $25,000 S14,100 $1,250 $l1,100 $14,100 $14,100 $14,100 $30,000 Fna ncial Cost of Structure (-) $13,600 $14,100 $14,100 $14,100 $14,100 $14,100 $25,000 $14,100 -',250 $14,100 $14,100 $14,100 $14,100 $30,000 Depreciation Rate 2.0X 2.01 2.0Z 2.0X 2.01 2.01 2.01 2.02 S.01 2.02 2.01 2.01 2.02 2.01 Maintenance (S of structure cost) 4.OX 4.01 4.ox 4.0X 4.o0 4.01 4.01 4.0S 5.01 4.0S 4.02 4.02 4.01 4.0S. Managoment Fees (2) 2.01 2.01 2.0X 2.0X 2.01 2.01 2.01 2.0X 0.01 2.0X 2.01 2.0S 2.01 2.0X Annual Real Infrastructure Costs $150 $150 $150 $150 $150 $150 $150 $150 $50 $150 $150 $150 $150 $150 Annual Financial Infra Costs $50 $50 $50 $50 $50 $50 $50 $150 $0 $50 $50 $50 $50 $50 Additional Infra Reduction (C) 25.02 25.01 25.02 25.02 25.01 25.01 25.0X 0.02 0.0X 25.0X 25.01 25.01 25.01 25.0X I Recurrent Infra Subasdies C-) $113 $113 $113 $113 $113 $113 $113 $0 $0 $0 $113 $113 $0 $113 Cost of Land Use RestrLctions $6,000 $6,000 $6,000 $6,000 $6,000 $6,000 $6,000 $6,000 $0 $6,000 $3,000 $6,000 $1,000 $6,000 r Cost of Building Standards $0 $0 $0 $0 $0 $0 S0 $0 $0 $0 S0 S0 $0 S0 I Cost of Permits $1,000 $1,000 $1,000 $1,000 $1,000 $1,000 $1,000 $1,000 $500 $1,000 $1,000 $1,000 $250 $1,000 Direct Cost of Sale Restrictions 41,625 $1,625 $1,625 $1,625 $1,625 $1,300 $4,550 $1,625 $0 $1,625 $1,625 $1,625 $0 $5,525 Loan Term 25 25 25 25 25 25 25 25 25 25 25 20 20 25 Loan to Value Ratio 100.01 100.0X 95.0X 95.0X 95.01 95.02 95.01 90.01 0.01 90.02 95.01 95.01 90.01 95.01 Interest Rate 5.5X 5.51 8.5X 8.5X 8.52 8.52 8.52 8.52 11.01 11.01 8.51 8.52 8.51 .51 Market Fixed Rate 11.01 11.02 11.01 11.02 11.02 11.01 11.02 11.01 11.02 11.02 11.02 11.02 11.01 11.01 Grace Perlod 2 2 0 0 0 0 0 0 0 0 0 0 0 0 Grace Interest Capitalized? Stamp Tax (X of SP) 0X 0X 1X 3X 3X 3X 32 32 0X 32 32 31 31 3X Investment Tax Credit $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 Tenant Income Tax Rate 30X 30S 12X 30X 30X 302 301 302 302 301 30X 302 301 302 Capital Galns Tax Type Nominal NomLnal Nominal Nomlnal Nominal Nominal Nominal Nomlnal Nominal Nominal Nominal Nominal Nominal Nominal Capital Gains Tax Rate 5X 5X 52 5X 5X 52 52 52 0X 52 5 S2 5X 5X Property Tax Rate 0.4005 0.0001 0.000S 0.000S 0.0001 0.000X 0.000X 0.000X 0.0001 0.0002 0.0002 0.0001 0.000X 0.000X Quit Rent $50 $50 $50 $50 $50 $50 $50 $50 $0 $50 $50 $50 $50 $50 LAndlordlDeveloper Tax Rate 30X 302 302 302 301 301 30X 301 301 301 301 301 30X 301 OQ~ m m 0 PRESENT VALUE MODEL -- SUMMARY OF INPUT DATA LOCATIONa Selangor Selangor Selangor Selangor .L. Perlis Selengor Selangor Selangor Test Selengor KL. Selangor Selangor PROGRAMs PLC9P PLCBP SLCBP SLCBP SLCHP SLCBP SLCBP Private Private Renter SLCB? SLCBP individuaSLClP LAMD: Public Public Public Privete Private Private Private Private Squatter Case Private Private Prlvate Private DEVELOPERt PMNS Private Prlvate Private Private Private Private Private Private Private Private Private Private Versions 1.1 STRUCTURE: Low Cost Low Cost Low Cost Low Cost Low Cost Lov Cost Med Cost Lov Cost Low Cost Low Cost Low Cost Lov Cost Med-Hi Co Run Date& 22-Nov TENURE: Sales Sales Sales Sales Sales Sales Sales Sales Squatter Sales Sales Sales Sales Run Time: 04:34 PR PLANNING STDSt Current Current Current Current Current Current Current Current Current Current Low Std Current UnplannedCurrent Sq. Clear REY MODEL OTPUTS: Incentives and DLsincentives Land Subsidy $9,000 $9,000 $8,600 $600 $600 $600 $726 $0 $1,920 $0 $360 $600 $0 $8O Infrastructure SubsLdy $560 $560 $1.450 $1,450 $1,450 $1,200 $3,700 $200 $0 $200 $1,450 61,450 $0 $4,550 Recurrent Infra. Subsidy $958 $958 $958 $958 $958 $958 $958 $0 so $0 8958 $958 $0 $958 Total afra. Subsidy $1,518 $1,518 $2,408 $2,408 $2,408 $2,158 $4,658 $200 $0 $0 $2,401 $2,408 S2 $5,508 Materials Subsidy $500 S0 $0 $0 $0 $0 $0 $0 $0 $0 $0 SO $0 So Labor & Other Con. Subsidy $2,500 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 o0 Con struction Finance Subsidy $1,000 $0 $0 $0 $0 $0 $0 $0 $0 S0 $0 $0 $0 $0 Total Constructlon Subsidies $4,000 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 60 60 $0 Building Stsndards $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 Land Use end Infrastructure Stds ($6,000) ($6,000) ($6,000) ($6,000) ($6,000) ($6,000) ($6,000) ($6,000) $0 ($6,000) ($3,000) (86,000) ($1,000) ($6.000) Planning Permission ($1,000) ($1,000) ($1,000) ($1,000) ($1,000) ($1,000) ($1,000) ($1,000) ($500) ($1,000) ($1,000) ($1.000) ($250) ($1,000) 'D Sale Restrictions ($1,625) ($1,625) ($1,625) ($1,625) ($1.625) ($1,300) ($4,550) ($1,625) $0 ($1,625) ($1,625) ($1,625) $0 ($5,525) Rental Regulations $0 Total Regulatory Costs ($8,625) ($8,625) ($8,625) ($8,625) ($8.625) ($8.300)($11.550) (88,625) ($500) ($8,625) ($5,625) ($8,625) ($1,250)($12,525) End User Financing Subsidy $11,962 $11,962 $4,017 $4,017 $4,017 $3,213 611,247 $3,805 $0 ($218) $4,017 83,485 $3,962 $13,657 Acquisltlon Tax $0 $0 ($250) ($750) ($750) ($600) ($2,100) ($750) $0 ($750) ($750) ($750) ($900) ($2,550) Registration Tax $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 S0 $0 $0 Investment Tax Credit $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 $0 So $0 Property Tax ($951) ($951) $0 ($951) ($951) ($761) ($2,664) ($951) $0 ($951) ($951) ($951) ($1,142) ($3,235) Cepital Gsins Tax ($0) ($0) ($0) ($0) ($0) ($0) ($0) ($0) $0 ($514) ($O) (SO) ($0) ($0) Tax on Rental Income ($4,321) Total Taxes ($952) ($952) ($250) ($1,702) ($1,702) (61.361) ($4,764) ($1,702) $0 ($6,537) ($1,702) ($1,702) ($2,042) ($5,7 5 Extra Transactions Costs (Tenant) ($0) ($0) ($563) ($O) ($O) ($o) ($O) ($0) ($0) ($0) (SO) ($5,000) (SO) ($O) Economic Cost of Unit: $31,700 $31,700 $28,100 $28,100 $36,100 $24,100 $41,390 $28,100 $5,170 $37,302 $24,500 $36,100 $20,400 $50,600 AffordabilLty at Economic Cost of the Unlt $13,597 $13,597 $11,695 $11,695 $15,995 $9,652 $18,977 $11,695 $1,409 $49,444 $9,853 $15,995 $7,837 $24,395 Current Selling Price of Unlt: $25,000 $25,000 $25,000 $25,000 $25,000 $20,000 $70,000 $25,000 $5,000 $2,538 $25,000 $25,000 $30,000 $85,000 Affordability at Current Selling Prices $9,319 $9,319 $11,603 $11,603 $11,603 $9,283 $32,489 $10,993 $0 $19,902 $11,603 $12,548 $14,266 $39,451 Prlce of Unit if Developer Sells at Breake $33,465 $37.465 $30,025 $38,525 $54,525 $35,300 $22,004 $40,375 $3,920 t4,089 S28.56 $54,525 $12,950 $25,885 Affordability at Breakeven Price: $14,550 $16,755 $12,705 $17,350 $26,783 $15,554 $8,615 $18,397 $997 $44,071 $11,937 $26,783 $4,441 $10,554 Karket Price of Unit: $20,000 $30,000 $30,000 $30,000 $50,000 $20,000 $70,000 $30,000 $5,000 $30,000 $30,000 $50,000 $30,000 $85,000 Affordability at Current Market Prices (I. $10,687 $16,030 $16,030 S16,030 $26,717 SLO,687 $37,403 $16,030 $2,672 $19,902 $16,030 $28,255 $16,953 $45,418 PC i CDX 0 w -94 A_NNEX 3 Table 1 Page 1 of 2 .MALAYSIA THE HOUSING SECTOR: GETTING THE INCENTIVES RIGHT Land Use and Desi8n Standards for Low-cost Housing MINIMUM LAND USE AND DESIGN STANDARDS FOR LOW COST HOUSING From the Special Low-Cost Housing Programme Single Double Single Double Storey Storey Storey Storey Feet Feet meters meters PLOT STANDARDS Width 16 14 feet 4.88 4.27 meter Length 55 52 16.76 15.85 Area 880 728 Sq.F 81.75 67.63 m2 Minimum Floor Space 450 450 Sq.F 41.81 41.81 m2 .Number of floors 1 2 1 2 Hi* Mn. of plot built-up 51.14% 30.91% 51.14% 30.91% Set Back on ped Way 7.5 7.5 feet 2.29 2.29 m Set Back on Roads 15 15 4.57 4.57 m * Buildable area for plots facing Roads & Ped W. 520 413 Sq.F 48.31 38.37 m2 * Maximum Floor Space 520 826 Sq.F 48 77 m2 * Standard Plot Max.FAR 0.59 1.13 0.59 1.13 * Minimum Width of Corner plots * 31.00 29.00 feet 9.45 8.84 meter * Minimum size of Corner Plots 1705 1508 Sq.F 158.40 140.10 m2 * Corner plot Max FAR 0.30 0.55 0.30 0.55 STREET STANDARDS Back Lanes 15 15 feet 4.57 4.57 meter Road ROW Main Access 66 66 feet 20.12 20.12 meter Distributor 40 40 12.19 12.19 Internal 30 30 9.14 9.14 Pedestrian was 15 15 4.57 4.57 Maximum length of Ped.Way 250 250 feet 76.20 76.20 meter SAMPLE BLOCK Boundaries Roads right of ways: North road: Int. 30 feet 9.14 meter South Road: Main 66 " 20.12 East Road: Dist 30 n 9.14 n West Road: Dist. 40 n 12.19 Inside Roads are Pedest.Way 15 " 4.57 Back Lanes 15 " 4.57 Number of rows: 6 Block dimensions: 6 6 rows 6 6 rows * Length 268 270 feet 81.69 82.30 meter * Width 510 492 155.45 149.96 * Total area 136 680 132 840 Sq.F 12 698 12 341 * Total Area (Acre, Hectare 3.14 3.05 Acre i.27 i.23 Hectare * Number of plots 78 90 78 90 * Number of Corner plots 12 12 12 12 * Number of Minimum plots 66 78 66 78 * % of corner plots 15.38% 13.33% 15.38% 13.33% - 95 - ANNEX 3 .... ... ... ... ... ... ... ... .... ... ... ... ... ... ... ... ... Table 1 COMMUNITY FACILITIES STANDARDS Page 2 of 2 Parking Plot/parking space 4 4 4 4 Area per parking space 160 160 Sq.F 14.86 14.86 Parking space/piot 40 40 Sq.F 3.72 3.72 Religious Facilities Muslim resident/surau 800 800 pers 800 800 Proportion of Muslims 50% 50% 50% 50% Area of one S6.-au 0.5 0.5 acre 2023 2023 m2 Sureau area/plot 79 79 Sq.F 7.33 7.33 o2 Muslim Resident/Mosque 2600 2600 pers 2600 2600 pers Area of one mosque 1.2 1.2 acre 4855 4855 m2 Mosque area/plot 58 58 sq.F 5.42 5.42 m2 Non muslim resident /place of worship 2600 2600 pers 2600 2600 pers Area of place of worship 1.2 1.2 acre 4855 4855 i2 place of worship area/plot 58 58 Sq.F 5.42 5.42 m2 Kindergarten Residents/kindergarten 2500 2500 pers 2500 2500 Area of kindergarten 6000 6000 Sq.F 557 557 m2 kindergarten area/plot 13.92 13.92 Sq.F 1.29 1.29 m2 Playlots Resident/Playlot 500 500 pers 500 500 pers Playlot area 0.16 0.16 Acre 647 647 m2 Playlot area/plot 80.85 80.85 Sq.F 7.51 7.51 m2 Playground Resident/playground 2000 2000 pers 2000 2000 pers Area of playground 2.00 2 Acre 8092 8092 .2 playground area/plot 253 253 Sq.F 23.47 23.47 m2 Multipurpose Community Hall Resident/community hall 200 200 pers 200 200 pers Area of Community Hall 0.25 0.25 Acre 1012 1012 m2 * Community Hall/plot 316 316 Sq.F 29.33 29.33 m2 Primary school Number of student/school 800 Population student ratio: 0.16 * Resident/p.School 5000 Area of primary School 6 6 Acre 2.43 2.43 Ha * p.school area/plot 303 303 Sq.F 28.16 28.16 m2 Secondary school Number of student/school 1000 Population Student ratio 0.1 * Resident/S.School 10000 Area of secondary school 9 9 Acre 3.64 3.64 Ha * s.school area/plot 227 227 Sq.F 21.12 21.12 m2 Oxydation ponds Number of residents: 2000 Standard:B * Number of plots 345 * Area required for pond 0.99 0.99 Acre 0.4 0.4 Ha * Area per plot 125 125 Sq.F 11.59 11.59 m2 ---_--- ------ ------ ----- --_--- ------- --..-- ----_- ------ ---_ _ - 96 - ANNEX 3 Table 2 SUMMARY LAND USE FOR A SCHEME CONTAINING ONLY MINIMUM PLOTS AND CORNER PLOTS Large Scale Medium Scale Small Scale >2500 persons about 1000 about 500 peoples Residential 24.76% 37.02% 44.26% Parking 1.19& 1.78% 2.13% Community facilities Surau 2.35% 3.51% 4.20% Mosque 1.73% 0.00% 0.00% Worship 1.73% 0.00% 0.00% Kindergarten 0.41% 0.00% 0.74% Playlots 2.41% 3.60% 4.30% Playgrounds 7.52% 0.00% 0.00% Community Hall 9.40% 14.05% 0.00% Primary School 9.02% 0.00% 0.00% Secondary School 6.77% 0.00% 0.00% Oxydation Pond 3.71% 5.55% 6.64% TOTAL 71.01% 65.52* 62.27% Streets serving plots 19.16% 28.66% 34.26% Streets serving facil. 9.83% 5.83% 3.46% TOTAL AREA 100.00% 100.00% 100.00% Population density 186 people/ha 278 people/ha 332 people/ha Plot density 32 plot/ha 48 plot/ha 57 plot/ha Floor Area Ratio (FAR) 0.13 0.20 0.24 -- - - - - - - - --_ - -- -_-- - - - - - - - - -- - - - - - - - - - - - - - - 97 - TABLE 3: DEVELOPER PROFITABILITY AND LAND USE STANDARDS ANNEX 3 Table 3 ..................................................................................................................................... 11-Apr-89 I Middte I SLCHP Guidelines iteration (1) iteration (2) I Iteration (3) Incom ....................I.................... ................ I... ... Alternative I Standard Corner Standard Corner Standard Corner Standard Corner I Plot Type I plot Plot I plot Ptot | plot Plot I plot Plot A. Unit Sale Piice (units)| A | B C O E | F G | H i ............. .....,,,,,,,,,,,,, I z I Sate Price of Construction NIuSm2 f 225 200 200 I 200 200 I 200 200 200 200 Sale Price of Land MS/m2 I 30 I 30 30 I 30 30 j 30 30 I 30 30 Infrastructure Sale Price S/m2 I 45 I 40 40 I 40 40 I 40 40 I 40 40 S. PLot Geonmtry Regulations | | 1 I i ........ .......................... I I Plot area m2 I 139 I 68 4142.37 jl 68 < 142.37 | 68 < 99.85 S' 68 4 99.85 BS Plot Frontage mi 6 j 4.27 < 8.94 'l 4.27 ' 8.94 'I 4.27 < 6.27 ', 4.27 ' 6.27 I Plot length m | < 23.17 )]1 15.93 3- 15.93 3I1 15.93 '< 15.93 'I1 15.93 >4 15.93 )I1 15.93 " 15.93 ol Front Set back m 5 5 4.67 4.67 1 4.67 4.67 1 4.67 4.67 I 4.67 4.67 I Back set back ml 3 | 2.5 2.5 | 2.5 2.5 | 2.5 2.5 | 2.5 2.5 Side set back m| 0 0 4.67 0 4.67 0 21 0 2 Max.Ground Buitdable Area t2 | 91.00 >f' 37.38 " 37.38 >14 37.38 " 37.38 ;,I 37.38 >4 37.38 >Ij 37.38 " 37.38 'I Actual Builded Area m2 I 82 1 21 21 I 21 21 24 25 j 26.5 30 I Nui*er of ftoors 1 2 21 2 21 2 21 2 2| Floor space/ plot m2 82 >1 42 - 42 'I< 42 " 42 3l1 48 > 50 'I 53 " 60 'I Plot Floor Area Ratio I 0.59 'I1 0.62 " 0.30 'I1 0.62 '4 0.30 >j< 0.71 >4 0.50 >'I 0.78 >< 0.60 'I C. Site Land Use Regulations | | | | | | .................................. Percentage of corner plots I I 15.00% I 15.00% I 15.00% I 15.00% I % of Salable land | 45% 44X % 55s% 49X | 55% I Site Floor Area Ratio | 0.27 '| < 0.23 '| C 0.29 >| C 0.33 >1 0.41 1 C. Costs per Dvelling Unit | I I I I .. ................... Cost of Superstructure MS I 4 18,450 Ic 8,400 '- 8,400 '1' 8,400 '4 8,400 'I1 9,600 'c 10,000 'I 10,r! : 12,000 >1 Cost of Plot NS 4 23,167 'I1 10,755 '4 22,517 'I 8,655 >' 18,120 'I 9,714 ' 14,264 2I< 8,655 >' 12,708 '- ------ I......-----..... ............I........ Total Cost per Dwel.Unit NS | 41,617 >j< 19,155 ' 30,917 'l 17,055 - 26,520 >I- ;9,314 ' 24,264 'I 19,255 > 24,708 I . ..I.................. ..... . ............ D. Profits Percentages Assuwptionl . ................................ Profit/m2 of Construction I 15% | 15% I 15% 15X I 15X | Profit/m2 of Land I 10%X 10% 10% I0% I 10% | F. Average Profit/ha by Site | I I I I ............I............ I I I Profit/Ha MS < 164,595 '| '140,453 >I '157,549 |I '167,559 'I '192,541 3I Profit incentive(disencentive) I | (24,142P| 4 C7,046)0> 2,964 >1 4 27,946 > as a % of alternative profit | -15X < -4.4%) 2%>I 17%) G. Additional Indicators | I | | i .......................... Profit per Plot j I 2,335 - 3,512 >[c 2,125 > 3,072 'I 2,411 " 2,926 'I 2,455 " 3,071 ) number of plots/ha j I 47.53 " 8.39 >j< 59.06 "< 10.42 )]' 57.23 " 10.10 >j< 64.24 ' 11.34 'I Total ntmrber of ptot/ha | | < 55.92 )] < 69.48 | * 67.33 I 75.57 'I Average household size | 5 | | | | Population Denity People/Ha IC 280 'I 347 >1 < 337 3>1 378 ). ..................................................................................................... notes Calculated outputs are shown as 4...'. =zn Indicates a change In parameter over previous iteration Parameters in column B and C line 18 to 46 correspond to minimum standards explicitly or ifflicitly contained in SLHCP guidelines - 98 - TABLE 4: DEVELOPER PROFITABILITY AND LAND USE STANDARDS ANNEX 3 Table 4 11 -Apr-89 IMiddle ISLCHIP GUidelines I Iteration (4) 1 Iteration (5) 1 Iteration (6) lincome ...............I........... . ............... . ..... Alternative IStandard Corner IStandard Corner IStanidard Corner IStandard Corner Plot Type I plot Plot ptot Plot plot Plot I plot PLot ..................I........... ....... . .............. . ..... A. Unit Sale Price (units)I A I B C I D E I F 6 I H I I ... ... .. ... ... .. ... I I I I I I Sale Price of Construction MS/M2 I 225 200 200 j 200 200 I 200 200 I 200 200I Sate Price ;;Land NS/M2 ~ 30 I 30 30 I 30 30 I 30 30 I 30 30 infrastructure Sale Price NS/a2 I45 40 40 j 40 40 40 40 I 40 40 S. Plot Geesmtry RegutationsII II Plot area m2 139 I 68 l 42.37 3I 68 ~c90.67'!- 68 oc 8.00 ;I 68 -c68.00 3I Plot Frontage ml 6 I 4.27 -c 8.94 2I 6 < ( > 6 -c 6 :1 6 c 6 ;, PLot length m < 23.17 )+c 15.93 - 15.93 )[1( 11.33 v4 11.33 :-1 11.33 >, 11.33 ;,I< 11.33 ~'11.33 2I, Front Set back ml S 4.67 4.671 3 31 3 31 3 31 Back set back ml 3 I 2.5 2.5 I 3.5 2.5 I 3.5 2.5 I 3.5 2.5 Side set back ml 01 0 4.671 0 21 0 01 0 DI Maxc.Grounid Buitdabte Area m2 < 91.00 ),I 37.38 3' 37.38 'Ic1 29.00 >< 35.00 21< 29.00 - 35.00 >j< 29.00 " 35.00 3'I Actual Builded Area m2 82 I 21 21 I 26.5 33 I 26.5 33 I 24.5 33 NurTer of floors I 2 21 2 21 2 21 2 2! Floor space/ plot m2 < 82 'Ic 42 3' 42 ,14 53 '- 66 'IJ 53 >< 66 >14 49 3- 66 I- PLot Floor Area Ratio I 0.59 >1I 0.62 2- 0.30 2'1< 0.78 >4 0.73 3-1c 0.78 3' 0.97 >jc 0.72 >c 0.97 >1 C. Site Land Use Regulations III Percentage of corner plots 15.002 15.002 15.002 15.002 % of Saabte land I4521 442 5521 5521 5021 Site itoor Area Ratio 0.27 3- -C 0.23 'I - 0.42 3'1 0.44 'I1 < 0.38 >1 C. Costs per DwetLing Unit IIIII Cost of Superstructure MS$ 183,450 [jc 8,400 >< 8,400 Jl 10,600 ", 13,200 >I-' 10,600 >< 13,200 >[-C 9,800 "< 13,200 '1 Cost of Plot Ns Ic 23,167 3-i 10,755 >" 22,517 >1' 8,655 >< 11,539 3'1 8,655 " 8,655 >14 9,520 >< 9.520 j- ......I....... .... I.... .. ....I.... ....I..... .... . I Total Cost per Dwel.Unit NS < 41,617 I19,155 >. 30,917 'I< 19,255 >. 24,739 >I< 19,255 '21,855 >[c 19,320 >< 2*2,720 3'1 0. Profits Percentages AssumiptionIIIIII ..I I.. .. .I...I...I...I Profit/m2 of Construction I132 152 152 152 152 Profit/m2 of Land J10j 10ox 1021 1021 102J F. Average Profit/ha by Site IIIII .......... ......I I......... Profit/Ha NS I 164,595 :-I '140,453 >1 c196,985 >1 c203,335 '1 <183,713 'j Profit incentfve(disencentive) II c(24.142)>l < 32,390 '1 < 38,739 >1 -C 19,118 'I1 as a 2 of alternative profit I C 52 202's 242'1 I2~ G. AdditionaL Indicatc's IIIII Profit per Plot j14 2,335 'o 3,512 ,1< 2,455 3- 3,134 3,1 2,455 ' 2,845 'j1 2,422 3'. 2,937! 3'j number of plots/ha I14 47.S3 *< 8.39 '-I' 65.48 >< 11.55 '[c 68.75 >' 12.13 .1< 62.50 3' 11.0! 3-1 Total nuisier of plot/ha I 55.92 >'I 77.03 >1 80.88 >I < 7.53 3'I Average househotd size I5 I I Population Density PeopLe/Ha I C 280 'I1 < 385 'I1 c 404 :1I- 368 >1 --to: Calculated outputs are shown as 'c.*..',. was- Indifcates a change in parameter over previous iteration Parameters fn coiumn B and C line 18 to 46 correspond to miniman stanidards explicitLy or iopllcitty contained in 0LHCP guidelines - 99- ANNEX 3 Figure 1 Site Plan Corresponding to SLCHP Guidelines -!a 44 X of Salable Land 20.5 % Community Facilities 35.5 % Roads FAR: 0.23 1' 1: 1 1 I r I I . * = 1U 1- 11[ I Ia Discussed in1text para. 5.27 and Table 5.4. /a Discussed in text para. 5.27 and Table 5.4. - 100 - ANNEX 3 Figure 2 Site Plan Corresponding to Iteration (3) ia 55 % of Salable land 19.4 % Community Facilities 25.6 % Roads FAR: 0. 41 t IaI i Isu Is in I par 5 an Tal 5 an 5.5. I f I IffL1iL _r IL iLI_ /a Discussed in text para . 5 .2 7 and Tables 5 .4 and 5 .5 .