The Singapore Solution Part II: time for Singapore-on-Liffey?

The Week in Housing 05/11/21

As noted last week, while in many respects successful, Singapore’s housing system is not without its challenges. One of the major challenges that dominated the evolution of the housing system in the ‘80s and ‘90s was the development of a secondary or resale market for HDB dwellings. This was necessary to allow residential mobility and to ensure the housing system is responsive to the changing needs of households over their life course. Prior to 1971, the HDB required owners who wanted to sell their flats to return them to the HDB at the original purchase price. This ensured that households did not obtain a ‘windfall gain’ by obtaining a HDB house and then selling it on privately at market prices, or flipping it, so to speak.

However, from ’71, households who were a minimum of three years in their dwelling could sell their dwelling at market prices, provided the purchasers met the eligibility criteria for HDB housing. Households who sold in this manner, however, were debarred from applying for another public housing dwelling for three years period, later increased to five. This was a huge deterrent as most housing was (and is) public housing. In 1979, this system was abolished and replaced by a 5% levy on the transaction price, allowing the HDB to recoup some money and limiting the extent of the ‘windfall gain’.

This has led to the development of parallel housing systems. On the one hand, new HDB dwellings are provided at non-market prices to eligible households. On the other hand, ‘resale’ HDB units are available on what is essentially a private market at market prices.

The development of a secondary market has been paralleled by the deregulation of the financial system. The main forms of deregulation included allowing HDB dwellings to be bought with private mortgages and allowing the CPF to be used to obtain private mortgages – in short, the development of a private mortgage market. Moreover, many of the eligibility criteria for HDB housing were removed which, in particular, meant high income households were able to purchase HDB dwellings, driving up prices in the resale market. Finally, in the mid-1990s a number of grants were also introduced as demand side measures to support purchase of HDB dwellings. This has given rise to some challenges which are interesting in terms of understanding some of the tensions between decommodification and a modern housing system.

These policies led to a surge in the price of housing in mid-1990s, partially due to the fact that they supported demand and partially due to the fact that housing began to be seen as investment asset, as is common in most housing markets. The initial government response was to radically increase housing supply (which it was able to do because of the HDB), in order to moderate house prices. This failed (perhaps an interesting case study of the limited potential for supply to reduce prices in a speculative housing market), and the Government later introduced a number of specifically anti-speculative measures, including capital gains taxes, stamp duty, and limits on LTV ratios. The introduction of these measures coincided with the financial crises in South East Asia in 1997, and house prices collapsed.

Since then house prices have recovered and the Government has continued to pursue anti-speculative policies, while at the same time seeking to avoid a crash in house prices. The HDB continues to provide new dwellings at below market rates to support affordability.

Today, Singapore’s housing system is somewhat mixed, with some market and some non-market elements. Nevertheless, it is the central role played by the non-market elements that explain the main characteristics of the housing system, even today. Most importantly, these include the fact that state ownership of land has allowed a stable supply of affordable (compared to market prices) housing, that 90% of the housing stock is publicly built, and that 90% of the population are homeowners. Many commentators view the Singapore model as an overwhelming success, and indeed it is a unique achievement:

“Within three decades of its existence, the HDB had resolved the housing shortage problem and had progressed to providing larger and better quality flats for upper middle income households…, the redevelopment of older estates, and retrofitting existing flats” (Phang, 2007).

As Phang and Helbe note, ‘establishing an integrated land, housing supply and mortgage finance framework can deliver dramatic increases in housing supply and improvements in homeownership affordability’, and the provision of affordable housing has contributed to ‘social stability, economic growth, and the development of communities’ (Phang and Helbe, 2016).

One of the challenges of such a system, however, is around the distribution and allocation of housing. A public model is very good at getting housing to those in need. It is not so good, however, at supporting residential mobility. For example, what do households do as their needs change over their life course (e.g. they may require smaller or larger house over time, or housing in a different location)? The market, because it consists of deregulated interactions between individuals on a ‘peer-to-peer’ basis, is very effective on this front, allowing the system to adapt in real time to changing needs. To meet this challenge, Singapore has developed a kind of parallel housing market largely based on HDB public housing units being resold. But in developing this model they have run into the classic problems of housing markets – unaffordable house prices and speculative housing demand. There is an interesting lesson here about the challenges associated with both market and non-market systems.

It should also be noted that the Singapore system is not focused on delivering cheap housing. The ‘median multiple’ (median house price divided by the median household income), in Singapore is 4.6. This is similar to Dublin, at 4.7, but much lower than London, at 8.2. However, there are some key points to keep in mind here. First, while price to income ratios are not particularly low, the accessibility of mortgages and the stability of supply means that homeownership continues to be accessible for most Singapore citizen households. Second, and most importantly, a median multiple of 4.6 is actually an incredible achievement in the context. Singapore’s GDP per capita is the third highest in the world - so it is a rich country. More importantly, it faces incredible constraints on the availilbity of land. Demographia, who produce an international survey on housing affordability, argue that:

“No major metropolitan area in the high-income world faces the housing affordability challenge of Singapore. Singapore’s six million people live on a a fully developed island nation so small that it could fit into one-half of Tokyo Bay. As a result, Singapore lacks the “supply vent” of low-cost suburban or exurban land that moderates house prices across an urban area. Further, Singapore is by far the most densely populated sovereign nation outside the microstate of Monaco. Indeed, Singapore is approximately 75 percent as dense as the core city of New York and 50 percent more dense than London”

The fact that Singapore has achieved a similar median multiple to Dublin, a small city in one of the least densely populated countries in Western European, demonstrates the potential of their model. The median multiple in Hong Kong, to take a very comparable case, is 20.8.

Third, and finally, the system is designed to deliver steady, stable supply over the long term. The median multiple has remained stable for several decades, unlike many countries where it has been worsening. In my view, achieving stable, predictable supply over the long term is much more important than delivering low cost housing in the short term.

There are, however, other challenges associated with the system. At the outset of this discussion of Singapore housing, I noted that land/real estate ownership is related to political power. When the state has total dominance of land and housing markets, it bestows enormous power on the state. In Singapore we see this, for example, in the Ethnic Integration Policy, which places quotas on the proportion of different ethnic groups that live in HDB blocks and neighbourhoods to support social integration. While the Government consider this a success (historically there was a lot of conflict between Chinese and Malay communities, for example), many would baulk at such a level of state intervention. More worryingly, the HDB is empowered to evict households for late payment of bills, illegal renting out of entire flat, breaking immigration laws by harbouring ‘illegal’ immigrants, or being convicted of extreme litter offences. This is of course related to the particularities of Singapore’s wider political culture, but it is salutary to recognise that publicly provided housing does not mean that power inequalities in housing disappear, but rather that those that arise from market relations are replaced with others, in this case of a paternalistic and potentially authoritarian variety. Moreover, Singapore’s housing system is highly exclusionary with regard to much of the migrant population, who are excluded from HDB housing altogether.

On the transferability of the Singapore model, Phang and Helbe (2016) have this to say:

“The transferability of Singapore’s experience to other countries needs to be juxtaposed with the local political and social context. In the housing policy sphere, a housing provident fund [i.e. state savings scheme] is relatively simple to set up, designed as a savings and payment institution. The more complex institution to replicate is the HDB, in particular its resettlement, town planning, and estate management capabilities, as well as attention to developing good quality affordable housing at a large-scale. Moreover, the tactics on which Singapore relies - compulsory savings, state land ownership, and state provision of housing - can easily spawn widespread inefficiency and corruption in other socio-political contexts.”

This is an important and much under-appreciated point. Over the passage of time, institutions such as the HDB (or the limited-profit housing associations in the Austrian context), build up enormous institutional capacity and expertise which is essential to the efficient provision of housing. It is not easy to develop these capacities over the short term.

Perhaps the biggest obstacle, however, is political opposition to the decommodification of land. But this brings us back to the fundamental political questions we started with: why is land private to begin with and what is the case for a market model as the most efficient, effective and fair form of allocation and distribution? To my mind, the case of market-base land system is not a strong one (although I am, of course, open to being convinced otherwise). There appears to be a lot to be gained from the decommodification of land. Crucially, the associated gains are only likely to become more salient as demand for urban land intensifies due to population growth, demographic change, financialization, and economic concentration in urban areas. We need much more research and much more discussion about the politics and economics of land, but to my mind that discussion should be broad enough to encompass seemingly radical, but perhaps entirely practical, alternatives such as the whole scale decommodification of land in urban areas. Singapore shows it can be done.

Events

The concept of ‘rentier capitalism’ has been a big part of research on the financialization of housing, e.g. in the work of Brett Christophers. On the 24th of November, the Equality Research Centre in UCD is organising a public webinar on the issue with some leading international academics. Also on the 24th of November, Housing Rights are hosting this event on the PRS in the north. The Centre for Sustainable Solutions is organizing this conference looking at the built environment and the goal of reaching net zero emissions.

What I’m reading

The whole ‘supply’ question has been much discussed on Irish housing twitter in recent weeks/months, here’s an interesting analysis from the Bank of England blog that looks houses as assets rather than goods and explores what this means for prices. Part II of Jennifer Harris and Kim McKee’s research on the PRS and wellbeing is now available. And this blogpost presents new evidence on the disproportionate impacts of Covid-19 on minorities. Finally, the Nervous States programme on Dublin Digital Radio have put out this interesting piece on institutional landlords (for which I was interviewed).