The Singapore Model of Housing and the Welfare State
"While Singapore is not generally regarded as a welfare state, the provision of housing welfare on a large scale has been a defining feature of its welfare system...
The state owns four-fifths of the land and determines the deployment of substantial domestic savings...
Despite the very visible presence of the state and government-linked companies, provision of welfare as is generally understood in Europe has not been a feature of the economic system...
The government relies almost exclusively on the Central Provident Fund (CPF) scheme, a mandatory savings scheme to finance a range of different welfare services: housing, healthcare, insurance, tertiary education and retirement (Asher, 2004)...
From 1964, the HDB began offering housing units for sale at below market prices, on 99-year leasehold basis, under its Home Ownership Scheme (HOS). The HDB was able to price its units below market prices mainly because HDB flats are built on state owned land, much of which had been compulsorily acquired from private landowners at below market prices (Phang, 1996). This was made possible by the Land Acquisition Act, enacted in 1966, which abolished eminent domain provisions.
The political and economic motivations for the HOS are perhaps best understood in the words of the then Prime Minister, Mr Lee Kuan Yew:
My primary preoccupation was to give every citizen a stake in the country and its future. I wanted a home-owning society. I had seen the contrast between the blocks of low-cost rental flats, badly misused and poorly maintained, and those of house-proud owners, and was convinced that if every family owned its home, the country would be more stable (page 116)...I had seen how voters in capital cities always tended to vote against the government of the day and was determined that our householders should become homeowners, otherwise we would not have political stability. My other important motive was to give all parents whose sons would have to do national service a stake in the Singapore their sons had to defend. If the soldier’s family did not own their home, he would soon conclude he would be fighting to protect the properties of the wealthy. I believed this sense of ownership was vital for our new society which had no deep roots in a common historical experience (Lee, 2000, p. 117)...
From the perspective of public policy, there was early concern that given the then general housing shortage, HDB dwellings should not become a vehicle for speculation by allowing the price subsidies to be capitalized on a secondary market. Resale regulations were therefore extremely onerous in the early days of the housing program. These regulations were eased as the housing stock increased over time and the housing market became more mature (Phang, 1992, pages 92-4).
Prior to 1971, there was no resale market for owner-occupied HDB dwellings. HDB required owners who wished to sell their flats to return them to the HDB at the original purchase price plus the depreciated cost of improvements. In 1971, a resale market was created when the HDB allowed owners who had resided in their flats for a minimum of three years to sell their flats at market prices to buyers of their choice who satisfied the HDB eligibility requirements for homeownership. However, these households were debarred from applying for public housing for a year. The debarment period was increased to two and a half years in 1975. The minimum occupancy period before resale was increased to five years in 1973 and has remained in place since.
The debarment period, a great deterrent for any household considering sale of its dwelling, was abolished in 1979 thereby greatly facilitating exchanges within the public housing sector. This was replaced by a five percent levy on the transacted price of the dwelling to `reduce windfall profits’. A system of graded resale levy based on flat type was introduced in 1982, and rules regarding circumstances under which levies could be waived were fine-tuned in the 1980s (Phang, 1992, page 93). The resale levy system ensures that the subsidy on the second new flat purchased by the household from the HDB is smaller than that for the first flat.
Between 1968 and 1981, CPF savings could only be withdrawn for purposes of down payment, stamp duties, mortgage, and interest payments incurred for the purchase of public-sector-built housing. In 1981, the scheme was extended to allow for withdrawals for mortgage payments for the purchase of private housing. From 1984, rules governing the use of CPF savings have been gradually liberalized to allow for withdrawals for medical and education expenses, insurance, and investments in various financial assets (Asher, 2004)...
Loan financing prior to 1993 was based on 80 percent of 1984 HDB new flat (posted) prices. As both new and resale prices rose (see Figure 2.3), households purchasing resale flats had to pay an increasing larger proportion of the price in cash. In 1993 HDB moved its mortgage financing terms closer to market practice by granting loan financing of up to 80 percent of current valuation or the declared resale price of the flat, whichever is lower. In 1993, the CPF Board also began to allow withdrawals of CPF savings to be used to meet interest payments on mortgage loans for resale HDB and private housing purchases. Before this, CPF members were allowed to withdraw only up to 100 percent of the value of these properties at the time of purchase.
Deregulation of the HDB resale market has been accompanied by an increase in the number of transactions. The transaction volume of resale HDB flats increased from fewer than 800 units in 1979, to 13000 units in 1987, 60000 units in 1999, and 31000 in 2004 (HDB Annual Reports)...
In cities of developed countries, new construction of housing is a small percentage of existing stock and comprises mostly high quality housing. Even as the construction of the basic 1 to 3-room HDB flats have been phased out, the construction of 4-room HDB flats may eventually meet with the same fate. The housing board’s ongoing modernization of older estates and its selective en-bloc redevelopment scheme (under which old apartment blocks are repurchased, demolished, and new estates built) will be even more important then. Owner-occupier subsidies (which almost all new households and a large proportion of existing households have come to expect as a right of citizenship) will, as a matter of economic efficiency if not political efficacy, has to be increasingly in the form housing grants for the purchase of existing housing rather than subsidized prices for the purchase of a new unit...
Foreigners are prohibited from owning private landed properties and private flats in buildings of less than six storeys without government approval. Foreign demand for housing assets in Singapore is thus effectively confined to the private flats and condominiums...
The housing approach adopted in Singapore has undoubtedly increased the savings and homeownership rates, mobilize resources for the housing sector and contributed to increase in housing loans and the development of the primary mortgage market. However, the approach is not without its detractors. Singapore’s housing strategy is inherently policy driven and centrally controlled, with major decisions on savings rate, savings allocation, land use, housing production, and housing prices being largely determined by the government. It is, in other words, a neo-classical economist’s nightmare. Pugh (1985), in the context of providing a set of operating guidelines for a good housing system, and advocating Singapore’s strategy as a good model, writes:
... do not be too perturbed if some orthodox (neo-classical) economists argue that housing is over-allocated by subsidy. Show them that `subsidy’ is a concept which cannot be fitted easily to housing, and produce counter arguments, which are respectable in economics, and which are readily available.
However, two decades and 35 years later, it may no longer be `respectable’ to continue to argue for the continuation of a system that has outlived certain aspects of its usefulness. While the CPF-HDB housing scheme had its merits in the past, the objectives that the scheme set out to achieve have been surpassed and the policy problem has become one of how to reduce its dominance with minimal upsets to asset values, household wealth and lenders’ balance sheets. This section reviews a number of problems that have become associated with the large scale directed credit to the housing sector, and may serve as guidance for other countries with, or contemplating the adoption of similar schemes.
Crowding out
Singapore’s housing strategy has been criticized for over-allocation of resources to housing, resulting in crowding out of consumption, as well as human capital and corporate investments. Despite widespread homeownership and rapid increase in housing wealth, Phang (2004) found no evidence that house price increases have produced wealth or collateral enhancement effects on aggregate consumption. Instead, due to the mandatory nature of the CPF as well as households’ inability to withdraw housing equity to finance consumption, households in Singapore face strong liquidity constraints. In addition to the welfare loss from consumption denied, Bhaskaran (2003) is of the view that the low percentage of disposable income spent has hurt the development of the retail sector in Singapore.
The CPF has also been blamed for a weak domestic corporate sector (since potential entrepreneurs are unable to access their savings for start-ups), and the crowding out of domestic private sector investments. The corporate sector in Singapore is dominated by MNCs and government-linked companies -- a recent study by Bhaskaran (2003) confirms that indigenous firms earn lower returns than foreign-owned firms within Singapore, and lower returns as compared to listed companies in Hong Kong, Japan Korea, Taiwan the US. Krugman (1994) and Young (1992, 1995), for example, have questioned the basis and sustainability of Singapore’s economic growth in a series of studies as far back as the early 1990s. Pointing to the low contribution of total factor productivity (TFP) growth, Krugman referred to the Singapore miracle as having been based on ‘perspiration rather than inspiration’ – ‘All of Singapore’s growth can be explained by increases in measured inputs. There is no sign of increased efficiency.
Housing sector impacts
Given the nature of the real estate industry, the tilt of resources to the housing sector is not easily matched by corresponding increases in supply. Ceteris paribus, one can infer that housing and land price levels in Singapore are higher with the CPF housing scheme than without. In the first decade (1968 – 1981) when the CPF savings was directed only toward public housing, administered prices, stringent eligibility criteria, and long waiting lists were used to allocate the HDB’s supply of new flats. Moreover, in the absence of a secondary market for HDB flats during that period, the inflationary impacts on a tightly regulated housing sector were not immediately apparent.
However, such mechanisms could not be used when savings were similarly directed to private housing beginning in 1981. The 1981 liberalization as well as the 1993 liberalization of HDB and CPF regulations for HDB resale flat housing loans had significant impacts on housing prices, contributing to the development of speculative bubbles that subsequently burst (Phang and Wong, 1997).
When the CPF contribution rates were used for macroeconomic stabilization, increased to mitigate inflationary pressures from higher wages (1978 – 1984) or cut to reduce wage costs and preserve jobs (1985 and 1998), the effect was to exacerbate the housing price boom and bust (see Figure 2.3 for housing price index) by channeling resources into real estate during an inflationary period and reducing resources to the sector during a recession.
Consumption inefficiencies
Singapore’s mandatory savings and housing policies have very substantial impacts on household’s consumption and investment patterns. Savers’ and consumers’ rights in decision making are constrained by numerous CPF and HDB restrictions and regulations. Consumption inefficiencies arise when households value the in-kind transfers/subsidies at less than the costs of providing them, or alternatively, at less than an equivalent cost but unrestricted cash grant. Moreover, numerous regulations to prevent profiteering and speculation as well as restrictions in housing location choice resulted in inefficient location and commute patterns for households (Phang, 1992).
Over the years, both the CPF and HDB have found it necessary to become more saver/consumer responsive and have liberalized regulations in order to reduce distortions and provide more investment as well as housing options. There is now a wider range of investment and merit good related consumption (education and health) options for CPF members. The shift toward demand-side subsidies in the form of CPF housing grants to subsidize the purchase of resale HDB housing since 1994 has been very well received. This has improved households’ ability to optimize with regard to their housing options.
Retirement financing
The typical household in Singapore has the bulk of its wealth invested in housing (Phang, 2001). Despite the high savings rate, overinvestment in housing and over exposure to the risk of a decline in housing price affecting the retirement (and healthcare) financing of an aging population has become issues of policy concern, especially since the bursting of the real estate bubble in 1997 (see Figure 2.3). Lim (2001) projects that 60 to 70 percent of the 50-55 years age group will not have sufficient funds in their account to meet the government stipulated minimum sum needed for retirement of S$80,000 in 2003. Analyzing CPF data for 2000, Asher (2004) finds the average balance for active CPF contributors was $53,000, equivalent to 1.27 times the per capita GNP – inadequate to finance retirement of more than 20 years duration on the average.
McCarthy et.al. (2002) show through simulations that the average worker in Singapore is likely to be `asset-rich and cash-poor’ upon retirement with 75 percent of his retirement wealth in housing asset, provided housing values continue to rise in real terms. In contrast, an American elderly household would have only 20 percent of their retirement wealth in housing asset. If the housing market were to take a downturn and remain depressed for years (as in Japan), this could reduce retirement asset accumulation for the Singapore worker substantially. This raises the problematic issue of over concentration of household assets in housing resulting in a risky under-diversified portfolio at retirement.
A report by the government appointed Economic Review Committee in 2002 arrived at a similar conclusion that CPF members were `asset rich and cash-poor’ and made recommendations to limit CPF withdrawals for housing, and for the government to explore ways for homeowners to monetize their property. However, the committee was also cognizant of the need to implement changes gradually, in order not to further destabilize the fragile post Asian-crisis real estate sector. Agreeing with the committee’s recommendations, the government moved to cap CPF withdrawals for housing at 150 percent of the value of the property, with the cap moving down gradually to 120 percent over five years for new private housing loans.
Lack of unemployment insurance
With recent changes in the macroeconomic environment, unemployment rates above five percent for 2002-2004, and greater volatility in economic growth, there is growing concern over structural unemployment associated with the restructuring of the economy. Various programmes exist to help the unemployed find jobs, upgrade their skills or retrain for employment, so that they can become economically independent again. These programmes may be considered short termunemployment benefits under a different name. However, the lack of unemployment insurance and limited public social assistance available when structural unemployment is expected to be long term has led some to question the viability of the current welfare system. Yap (2002) has proposed that an employment insurance scheme be introduced as another component of the CPF. Asher (2004) makes a strong case for the introduction of tax financed schemes to address the needs of the lifetime poor, and social risk pooling to address longevity and inflation risks. According to him:
This case has become stronger due to the unilateral alteration of the implicit social contract by the government. This contract provided for acceptance of government’s socio-economic engineering and political control in return for job security and full employment. The government is not able to fulfill the latter element, but still wants to continue to undertake socio-engineering and maintain political control. This is a disequilibrium situation which will need to be resolved.
However, Asher acknowledges that the prospects for the required mind-set change by policymakers are not encouraging.
Financial Sector development
As required by the CPF Act, the CPF channels members’ deposits to the purchase of government bonds. The government through the Government of Singapore Investment Corporation invests the bulk of the funds directly abroad, thus bypassing Singapore’s financial markets. Bhaskaran (2003) contrasts the `remarkably poorly developed’ status of Singapore’s savings industry against its high savings rate. `Financial planners, unit trusts, stock brokers, pension funds, pension advisors, wealth management associated with middle-class households, financial journals, etc are all under-represented in Singapore compared to say Hong Kong.’ It has also been observed that the debt market in Singapore has remained relatively unsophisticated and illiquid due primarily to the cash rich public sector and the dominance of MNCs that typically do not depend on domestic sources of capital (Committee on Singapore's Competitiveness, 1998).
A similar observation can be made with regard to the mortgage sector. While the housing loans to GDP ratio has exceeded 70 percent, more than half of outstanding housing loans are originated by the HDB whUBirminghamich (sic) in turn obtains mortgage funding loans from the government. The HDB is not a financial institution while the CPF is described as a non-bank financial institution. The size of the mortgage market as such is smaller than one would expect, given the high loan to GDP ratio. The mortgage sector’s linkages with capital markets are weak and mortgage instruments relatively unsophisticated – there is no secondary mortgage market...
Governance issues
Pension fund governance issues relating to the responsibilities of trustees, accountability, and transparency are important areas of concerns for all stakeholders. In the US for example, the principles and procedures required of pension fund fiduciaries are legislated under the Employee Retirement Income Security Act. In the Singapore context, the CPF balances due to members at S$110 billion (in September 2004) are considerable. Their ultimate investment by the Government Investment Corporation of Singapore has however been described by Asher (2002) as `by law non-transparent and non-accountable, leading to a highly regressive large tax on provident fund wealth and to low replacement rates’. These concerns as well as that of ensuring the core administrative tasks are performed satisfactorily are major challenges for any mandatory pension scheme...
The housing welfare strategy is itself a legacy of the post-World War II years, when `only governments could marshal the resources necessary to rebuild devastated and dislocated nations’ (Yergin and Stanislaw, 2002). In the last two decades, the focus of attention has shifted from market failure to government failure – with privatization and deregulation as the preferred strategy generally, as well as in the development the housing sector. The present concerns faced by Singaporeans, in particular the lack of unemployment safety nets and the possible inadequacy of personal resources for retirement and healthcare in the future, serve to highlight the risks of overemphasizing housing in the welfare system for too long. Policymakers now have to grapple with tradeoffs and difficult decisions in trying to reduce the dominance of housing welfare without adversely effecting housing asset markets."
