East Asia and Pacific 67883 Economic Management and Poverty Reduction an eye on east asia and pacific 5 Asian Housing Markets: Bubble Trouble? by Tehmina S. Khan Continued growth in Asia, combined with slow recovery and fiscal sustainability concerns in Western economies and ample global liquidity has fuelled large capital inflows into region. This has lead to growing concerns about asset bubbles, including in housing. This note takes a closer look at whether housing markets in the region are out of line with “fundamentals� and finds that, as yet, there is no immediate threat of a housing bust. However, caution is still warranted in the short term on a number of grounds: although monetary tightening has begun, interest rates are still extremely low; some countries are showing signs of a “quantity bubble�; and large capital inflows are expected over the coming year that should further feed asset prices. Over the long term, as private mortgage finance and asset-backed security markets grow, strengthened financial supervision and risk assessment closely coordinated with monetary policy are needed to minimize risks to financial stability. House prices in several key Asian economies recorded double digit growth in the third quarter of 2010, while rising inflation means that monetary conditions remain accommodative across the region. Meanwhile concerns about asset price bubbles have grown, fed by positive growth outlook of East Asia compared to Western economies, the start of the QE2 by the US Federal Reserve Bank and large capital inflows into the region. This note takes a closer look at housing markets in East Asia. It starts by looking at house price trends over the past decade. It then considers whether recent house prices movements are “fairly valued� or whether there is a potential risk of correction. Figure 1. Real house price growth in East Asia compared to western economies, selected economies and cities annual percent 25 20 15 10 5 0 -5 -10 -15 IRELAND THAILAND INDONESIA UK USA FRANCE TAIWAN, SHANGHAI HONG KONG, CHINA Asia SPAIN PHILIPPINES JAPAN KOREA MALAYSIA BEIJING CHINA AUSTRALIA TAIPEI SAR CHINA SINGAPORE GERMANY average Annual average 2002–07 2010 (average of 1st 3 quarters) Annual average 2008–09 Sources: CEIC, Haver and Thompson Datastream. Notes: Residential land price data for the 6 largest cities, the Halifax House Price, the Case Shiller 10 city index and residential building land price data are used for Japan, UK, US and China respectively. At the moment house prices in East Asia are not far above values implied by fundamentals, suggesting that there is no overheating as such, at least at the national level. However, housing affordability is under pressure in several of the key emerging economies, including China, Singapore, Hong Kong, SAR China and Taiwan, China, and there are also indications of a “quantity bubble� in China. In addition, price pressures may continue to build fuelled by capital inflows into the region, low interest rates and the fact that efforts to efforts to increase housing supply may likely to take a few years to come through the 1 pipeline. Together these factors may counteract government efforts to restrain “speculation� and to tighten loan conditions through capital controls and regulatory measures. Regional house price trends Figure 2. Nominal house price indices, selected Asian economies Growth in real house prices in Asia has been fairly muted over 2000=100 300 the past decade in comparison to large bubbles that built up in Western economies. For instance, in real terms house prices 250 grew at an annual average rate of less than 2% between 2002– 200 2007 in the region (excluding advanced economies such as Japan and Australia). 150 100 In nominal terms, house prices in several East Asian economies have only now recovered to (or are not far above) peak levels 50 witnessed at the time of the Asian Financial Crisis in 1997/98 0 (Figure 2). In part this is because the Asian Crisis resulted in a 1993 1995 1997 1999 2001 2003 2005 2007 2009 long period of high interest rates, rising defaults on property China Singapore Malaysia Indonesia Thailand Hong Kong Taiwan Korea loans and a period of relative economic stagnation that slowed Sources: CEIC, Haver and Thompson Datastream. Residential building land price data the growth of mortgage and housing markets. used for China. Economic and demographic fundamentals This is despite the fact that the economic and demographic fundamentals of the region as a whole imply much stronger growth in house prices over the past decade than experienced. Annual average GDP growth in East Asia has been close to 5.5% since 2000 while average, real GDP per capita has risen by 45% between 2000 and 2009 (Figure 3). In addition, with its population estimated at around 1.9 billion people, the overall magnitude of demand for housing in East Asia (exc. Japan, Australia and New Zealand) is enormous. For instance, it is estimated that in Indonesia—the fourth most populated country in the world—some 800,000 housing units are required annually to accommodate new household formation1. Moreover, housing demand is also being spurred by a shift of populations from rural areas to cities as countries have developed. Urban populations in the region have grown at an annual average rate of 3% between 2000 and 2009. Figure 3. Per capita income and GDP growth Figure 4. Urban population characteristics 140 12 120 7 120 10 100 6 100 5 8 80 80 4 6 60 60 3 4 40 40 2 20 2 20 1 0 0 0 0 China Indonesia Philippines Vietnam Korea Taiwan, Cambodia Lao China Indonesia Malaysia Singapore India Malaysia Thailand Hong Kong, Singapore China Vietnam Thailand East Asia & Pac. Philippines Korea SAR China Increase in real income per capita, 2000–09 Average GDP growth 2000–09, RHS Urban population (as % of total) in 1980 Urban population (as % of total) in 2009 Source: IMF. Annual average growth 2000–09, RHS Sources: World Development Indicators, World Bank. 1 UN Habitat (2008). Asian housing markets: bubble trouble? 2 Omnipresent governments Furthermore, government intervention has also been a ubiquitous feature of Asian housing markets. Governments have regularly undertaken measures such as altering loan to value ratios, imposing taxes and increasing the supply of public housing to avoid sharp and potentially destabilizing swings in house prices, and to increase housing affordability. This has helped to restrain house price growth in several economies, but has resulted in a stop-go cycle in others. For instance, between 2005 and 2007, Korea introduced 5 policy packages 2 to dampen “excessive� house price growth in Seoul and to prevent it from spreading to other cities. On the supply side, measures included the construction of an additional 2.6 million housing units for rent by 2017 to boost public rental housing from 6% of the country’s total housing stock to 20%. On the demand side, tighter loan-to-value ratios and debt service-to-income ratios were introduced. Consequently, nominal house price growth in Korea slowed from 11.6% in 2006 to just 3.1% In Q4 2007 and has averaged 2.2% since then. In Malaysia on the other hand, government attempts to increase homeownership for ethnic Malays has reputedly led to an Figure 5. Recent government interventions in oversupply in some property segments such a condominiums, property markets and nominal house price growth helping to restrain price increases. 3 % yoy 100 China raises down- payments for homebuyers 80 In China the authorities have also maintained a close eye on China restores sales lock up period eligible for favourable transaction housing markets, intervening regularly to dampen speculation. For 60 Korea raises loan to value ratio and taxes from 2 to 5 years announces increase in public housing example in 2010, it raised down-payment requirements by 10% 40 stock for first- and second home buyers to 30% and 50% respectively. 20 Similar measures to restrain house price increases have been 0 adopted in Singapore, Hong Kong, SAR China and Taiwan, China -20 Singapore increases China: land appreciation as well (Figure 5). Nevertheless, despite these interventions, -40 tax subjecting developers land supply Hong Kong, SAR China tightens loan to value ratios to a 30–60% levy on residential house prices in Singapore, and Hong Kong, SAR China capital gains and introduces Special Stamp Duty on properties sold -60 within 2 years and land prices in China were some 20% higher in real terms in 3/05 3/06 3/07 3/08 3/09 3/10 the first three quarters of 2010 compared to prices in 2009 and China Hong Kong Singapore Korea Source: World Bank. nearly 10% higher in Taiwan, China. Asian housing markets: “fairly valued� or overheating? The question that arises then is to whether house prices in the region are currently “fairly valued� or not. Two key rules of thumb are house price to disposable income ratios and gross rental yields, although their main shortcoming is that they ignore changes in underlying fundamentals such as borrowing costs and demographic changes. They are also likely to be affected by the institutional arrangements. For example, both the Philippines and Taiwan, China maintain rent controls. House prices to income/wage ratios provide an indication of affordability or roughly the number of years that it would take a median/average household to pay off its loan. Since such data (on either actual median incomes or median house prices) is generally lacking4 , housing affordability can be assessed indirectly by looking at the trend in house price to wage/income ratios. As Figure 6 shows, since 2001 prices had been rising at a slower pace than incomes in China until 2009 when affordability seems to have declined. There are also marked declines in affordability in Hong Kong, SAR China and Taiwan, China while Singapore too has experienced a deterioration over the past year. An alternative metric is gross rental yields, which is similar to a price to earnings multiple for stocks and reflects the relative cost of owning versus renting. In a fairly valued market, there should be little difference between renting and buying, since if 2 OECD (2007). 3 Developers in Malaysia are required to reserve 30% (and up to 50% in some states) of apartments in new housing schemes for bumiputra or ethnic Malays. Although these apartments are also eligible for additional government subsides, the ethnic Malay tend to be reluctant to purchase them because of resale restrictions (they can only be sold to other bumiputra). With developers finding it hard to sell the reserved amount, this has resulted in downward pressure on the prices of new apartments. Also see: http:// www.asiabuilders.com/asiabuilders/NewsSingle.aspx?rec_code=19802&ind_ctry_code=conMY , and http://www.penangwatch.net/node/2406 4 In Taiwan, China however, house prices at the end of the second quarter of 2010 stood at 7.7 times income, and were 11 times higher in Taipei. Asian housing markets: bubble trouble? 3 Figure 6. House price to wage/income ratios 2001=1 2001=1 1.2 2.5 1.1 1.0 2.0 0.9 0.8 1.5 0.7 0.6 1.0 0.5 0.4 0.5 0.3 0.2 0 2001 2003 2005 2007 2009 2001 2003 2005 2007 2009 China Beijing Shanghai Malaysia Korea Indonesia Singapore Hong Kong Taiwan Sources: Datastream, Haver, CEIC,World Bank. With the exception of Singapore for which median household income is used, average monthly income is used for the remaining countries. prices are too high then potential homebuyers will choose to rent, Figure 7. House price to rent ratios exerting downward pressure on house prices. If however price to rent ratios remain high for a prolonged period of time, then this 2004=1 2.0 suggests that prices are being held up by expectations of future price gains, suggesting the presence of a “bubble.�5 As Figure 7 shows, this appears to be the case for Hong Kong, SAR China and 1.5 Taiwan, China and to a lesser extent for Singapore and Shanghai. 1.0 Finally, to better account for underlying economic fundamentals, an econometric model is estimated 6 . The log of real house prices is 0.5 modelled as a function of log of real income per capita, population and housing supply as measured by housing construction 0 completed or coming on stream. Frictions in housing markets such 9/04 9/05 9/06 9/07 9/08 9/09 9/10 as supply lags, are accounted for by including lagged real house China Shanghai Hong Kong Beijing Singapore Taiwan Sources: Datastream, Haver, CEIC, World Bank. prices as an explanatory variable and the real cost of borrowing through the use of real mortgage interest rates or commercial Figure 8. Gap between actual house prices and bank lending rates7. those implied by fundamentals Q1–Q3 2010 The gap between actual prices in 2010 and those predicted by Thailand 3.8 the model is shown in Figure 8. While this can reflect omitted Taiwan, China 3.0 variables, such as ease of obtaining housing credit, it can also be Singapore 3.2 interpreted as a measure of the degree to which prices deviate Malaysia 5.3 from macroeconomic fundamentals. The figure shows that house Korea 4.2 prices in East Asia are currently not far above predicted values based on fundamentals, suggesting that there is no substantial Indonesia 3.1 overheating as such, at least at the national level. The largest Hong Kong, 0.6 SAR China degree of “overvaluation� is in Malaysia (albeit of around 5%), China 1.0 where prices have been steadily increasing over the past decade 0 2 4 6 despite a doubling of the housing stock since 2000. Source: World Bank. 5 Himmelberg, Hummel and Sinai (2005). 6 Typically, two approaches are followed. In the first case, price rent ratios are compared with time varying discount factors that are determined by the user cost of owning a house, which take into account mortgage rates and interest rate deductibility, property taxes, maintenance costs (e.g Himmelberg, Hummel and Sinai, 2005). However such data is not available for most Asian countries, so we resort to the second method which compares actual house prices with those that would be predicted by underlying macroeconomic and demographic fundamentals, as for example in Abraham and Hendershott (1996) and Glindro et.al (2008). 7 The Within-Groups estimator is used, which takes into account unobserved heterogeneity across sample countries. The consistency of the estimator depends on a large T and Monte Carlo studies (Judson and Owen, 1999; Bruno, 2005) show that when N is small relative to T, the within estimator outperforms IV-GMM type estimators in terms of bias and root mean square error. This is because the asymptotic properties of IV-GMM type estimators rely on large N when T is fixed so that their use can yield highly biased and imprecise point estimates in small samples. N equals 8 countries while T equals quarterly data from 2000 onwards till 2010 Q3 (or 43) in our sample. Note that in our model the calculations on the gaps for 2010 are conditioned on the real house price level in 2009; this might lead to some underestimation—in absolute value—of these price gaps. Asian housing markets: bubble trouble? 4 Sources of risk Generally the results suggest that there is no immediate risk of a housing bust or financial instability, although there are signs that house price increases are starting to outpace income growth in several major economies and that price to rent ratios are also too high Caution is warranted on a number of grounds, including the fact that interest rates in the region are currently being raised from extremely low levels, that some countries are exhibiting signs of a “quantity� bubble, and that large capital inflows are expected over the coming year which should feed into asset prices. These factors are considered below a) Loose monetary policies for now, what next Figure 9. Real borrowing costs (%), selected economies 20 The recovery in real estate markets in East Asia has been driven by a number of factors. First, with central banks cutting policy 15 rates sharply, mortgage or borrowing costs fell to extremely low rates. Although base rates have been raised gradually over the 10 past year as the region’s economies have recovered and output gaps closed, real borrowing costs have stayed low as inflation 5 has climbed (Figure 9). 0 In Hong Kong, SAR China, interest rates have remained at rock -5 bottom due to its peg with the US dollar. Singapore on the other 11/05 11/06 11/07 11/08 11/09 11/10 hand targets its exchange rate against a basket of currencies. Malaysia Thailand Hong Kong Korea Singapore Taiwan Sources: Haver, CEIC, Datastream, World Bank. Base policy rates used for Malaysia and Although in October it increased the pace of currency appreciation Korea, commercial bank rates for Thailand and Hong Kong, and mortgage or housing loan rates for Singapore and Taiwan. to limit imported inflation, domestic interest rates have remained relatively flat: 15 year mortgage rates are currently at their lowest levels in roughly 20 years. Meanwhile, in China increased bank lending was a key part of the massive monetary stimulus introduced in 2009 to mitigate the impact of the external slowdown. Mortgage lending rose by more than 40% in 2009, to 14% of GDP from 11% the previous year, while the share of new bank lending to households, notably mortgages increased from 14% to 26%8 . To the extent that central banks are laggards in raising policy rates in line with inflation (in Hong Kong, SAR China’s case, this is not even an option), then the extremely low real cost of borrowing should continue to counteract government efforts to restrain “speculation� and to tighten loan conditions through capital controls and regulatory measures. On the other hand, there is the risk that policy makers raise rates too quickly in the face of a weakening/still weak economic environment in Western economies. Widening interest rate differentials between emerging market economies and advanced countries (where rates are likely to remain low for some time to come) could also exacerbate volatile capital flows into the region. b) Not just prices but quantities matter too Another area of concern is that, notwithstanding the positive outlook for income and urban population growth rates, a construction bubble may be building up. This is worrisome because when housing markets turn, housing oversupply can keep prices depressed for a long time and help delay a recovery in bank and consumer balance sheets. Vacancy rates in newly built properties are one indicator of oversupply, but such data are hard to come by in East Asia. Alternative benchmarks include residential investment to GDP ratios as well as other indirect measures such as electricity meter usage. In the case of China, the residential investment to GDP ratio has grown steadily since the start of the decade (Figure 10). The upward trend is not worrying per se, given the intense urbanization pressures in China. Rather it is the rate of increase which matters: between 2000-06 the ratio climbed by 2.7 percentage points (pp) to 7.3 in 2006. By 2009, however it had reached 10.8, an increase of 3.5 points (Figure 11). This is not much lower than the 4.7 pp increase in Ireland in the four years prior to when its housing investment to GDP ratio peaked (in 2005) but much higher than in other countries that suffered large housing price and construction busts such as Spain, the USA and Japan. 8 Schroders (2010). Asian housing markets: bubble trouble? 5 Figure 10. Investment in construction/housing to Figure 11. Investment in construction/housing to GDP ratios, selected economies GDP ratios, selected economies 20 9 18 15 6 12 9 10.8 10 3 6 8.7 3 8.0 5.3 0 0 4.7 -3 3.4 -3 -6 0 Thailand Singapore Japan* US* Hong Kong Philippines Spain* China Ireland* 1995 1997 1999 2001 2003 2005 2007 2009 (2009) (2009) (1988) (2005) SAR China (2009) (2009) (2006) (2009) (2005) China Singapore Philippines Hong Kong Thailand Taiwan Increase 2006–09, * percentage points Increase 2000–06, * percentage points Average in 1997 Housing/construction investment to GDP ratio, RHS Sources: Datastream, Haver, World Bank. Investment in construction to GDP shares Sources: Datastream, Haver, World Bank. Investment in construction to GDP shares used for the Philippines and Taiwan, investment in residential construction for China used for the Philippines and Taiwan, investment in residential construction for and Singapore, construction including real estate development for Hong Kong, and China, Singapore, USA, Japan, Ireland and Spain, construction including real estate construction for private buildings for Thailand. development for Hong Kong, and construction for private buildings for Thailand. Notes: The dates in parentheses indicate the year for which residential/construction investment ratios data is shown in the chart. This is 2009 for East Asian economies or the year in which these ratios peaked in the other economies. For countries marked with an asterisk, the 2006–09 increase is for the 4 years up to the date at which the ratio peaked (e.g. from 2002–05 for the USA) while the 2000–06 increase is for the seven years prior to that (1996-2002 for the US). Meanwhile, media reports suggest that some 64.5 million urban electricity meters are reported to have registered zero consumption over a six-month period in China. In the absence of data on vacancy rates, this provides an upper limit on the amount of new residential property that lies empty and may have been purchased for speculative purposes.9 While growing housing needs are one possible driver, these trends suggest that expectations of further price appreciation are also driving housing markets. c) Rising tide of global liquidity lifting asset prices... Figure 12. Capital flows to emerging markets Furthermore, capital flows into the region are up sharply compared to a year ago (Figure 12). More inflows are 2008 2009 2010 Total H1 Q3 Total H1 Q3 Sep Oct YTD expected in the coming year on account of the second Total 390 110 99 353 189 169 89 46 403 round of quantitative easing worth US$600bn by the Bonds 65 36 33 115 87 59 27 21 166 US Federal Reserve Bank that started in November. Banks 257 43 34 129 46 45 16 5 96 Equity 68 32 32 109 57 65 47 20 141 Capital inflows to Hong Kong, SAR China for instance, Lat. America 90 37 37 137 64 65 48 11 140 were up sharply in Q4 which is estimated to have raised Bonds 20 15 18 62 36 31 15 8 75 total liquidity stock in the banking system to a new E. Europe* 157 22 21 72 53 34 14 15 103 record level of US$170 billion, exceeding the previous Bonds 35 13 11 33 31 17 7 9 57 peak in late 200910. With the currency pegged to the Asia 98 44 33 122 58 59 23 18 135 US$ and no capital controls, this has also meant that Bonds 7 6 3 16 14 8 3 3 26 Others 45 7 9 22 14 10 4 1 25 capital inflows readily translate into capital price gains. Source: World Bank. * Including Poland In China, although capital inflows may increase the pressure on the Renmibi to appreciate, they may be less problematic than is generally thought. This is because the amount of net financial inflows is relatively modest compared to other flows and domestic credit creation, due to relatively effective capital controls. More than 100% of the increase in foreign reserves in 2006–09 was driven by the current account surplus and net FDI while in 2009, the first year with positive net financial inflows since 2004, such net inflows were equivalent to around 4% of domestic credit creation11. 9 Shttp://www.marketwatch.com/story/chinas-overhang-of-empty-apartments-2010-08-03?siteid=rss&rss=1. 10 Morgan Stanley (2010). 11 China November Quarterly, World Bank, 2010. Asian housing markets: bubble trouble? 6 d) ...but an unwinding of capital flows could hit asset markets hard With global financial markets remaining jittery and given the Figure 13. Household mortgage debt and banking potential for a protracted slowdown in Western economies, sector real estate loans 80 there is the risk that investor, business and consumer sentiment 70 could also take a hit in East Asia, with ramifications for property markets. Research suggests that housing markets in Singapore 60 and Hong Kong, SAR China are more vulnerable to price 50 movements in other asset markets (notably stock markets) than 40 other economies in the region.12 30 20 Moreover although bank balance sheets are generally healthy 10 across the region, in several countries banking sector assets are 0 Hong Kong, Taiwan, heavily concentrated in real estate loans, notably in Singapore, Korea Indonesia Philippines China Thailand SAR China Malaysia China Singapore Malaysia and Taiwan, China. Furthermore, the relatively high Hhs mortgage debt as % of GDP, 2008 Total hhs debt as % of GDP, 2009 Real estate loans as % of total, 2010 levels of household debt in these economies suggests that a Sources: ADB Asian Economic Monitors, Dec 2010 and July 2009. Household debt data sharp fall in house prices could deal a strong blow to consumer for Thailand is from the Bank of Thailand and is for 2009. spending. As Figure 13 also shows, in Singapore, Taiwan, China and Hong Kong, SAR China, mortgage debt has accounted for roughly two-thirds or more of household debt in recent years. With mortgage lending in the region underpinned by variable rates (Table 2) interest rate risk is mainly borne by borrowers. There is accordingly a risk that as policy makers tighten monetary policy, then interest rate increases will affect household balance sheets, particularly in countries where household debt levels are high. Table 1. Real house price growth in East Asia compared to western economies, selected economies and cities, annual % Mortgage Debt Mortgage Type Average Length of Average (Maximum) Country as % of GDP* (Mostly Fixed or Variable?) Mortgage Contract (Years) Loan to Value Ratio* China 11.6 Variable 15-Oct ...('80) Indonesia 2.5 Variable 15 75–80 (80) Thailand 8.529 Variable 20-Oct 70–80 (100) 3–5 (private lenders), Korea 23.4 Variable 30 (Korean Housing Fin. Corp 56 (70) set up in 2004) Malaysia 26 Variable 30 (Maximum) …(80) Philippines 2.1 Variable 30 (Maximum) 70–80 (…) Taiwan 38.4 Variable 20 60–70 (80) Hong Kong 38.8 Variable 20 50–60 (60) Singapore 34.8 Variable … …(70) US** 69 Fixed/ Variable 30 75–85 (100+) UK** 75.4 Fixed/ Variable 25 80 (100+) Australia** 74 Fixed/ Variable 15–20 90–100 Canada** 43 Fixed/ Variable 25 70–80 Ireland*** 65 Variable 20 90–95 (100+) Sources: Various media, ADB, BIS, V. C. Warnock and F. Warnock (2008), H Zhu (2006), IMF (2010), Ahearne et. al (2005). **Data for 2004, *** Data for 2007 from The Economist and Cardarelli et. al (2009). Added to this there has also been an influx of buyers within the region which is helping to push up property prices in Hong Kong, SAR China and Singapore. In the latter, sales of new private residential units in 2009 to foreigners were more than 300% higher on 2008 levels. These have continued to rise since then, with residential unit sales to foreigners in the first 3 quarters of 2010 exceeding total 2009 sales by some 12% and accounting for some 20% of sales year to date. 12 H. Zhu (2006) “The structure of housing finance markets and house prices in Asia.� BIS Quarterly Review. Asian housing markets: bubble trouble? 7 One silver lining here is that Singapore already has some experience of this. At the time of the Asian Crisis, many investors in the region, in particular Chinese-Indonesian families viewed Singapore as a “safe haven� thereby boosting private property demand. Consequently, if any intra-regional capital flight were to occur, then Singapore may likely remain attractive as a destination for regional property investors (possibly due to its relatively transparent regulatory regime and strong property rights protection). Finally, mortgage markets in both Singapore and Hong Kong, SAR China are very different to, for instance, US and UK mortgage markets which were characterised by the use of under-regulated and opaque asset backed security lending to sub-prime borrowers and extremely high loan to value ratios. Mortgage-backed securitisation is still in its infancy in Hong Kong, SAR China and Singapore, maximum loan-to-value ratios range between 70 to 80% (see Table 1 overleaf) and �teaser-rate� mortgages, whereby mortgage rates increase dramatically a year or two after a loan is taken out, and which were used to reel in sub-prime borrowers in the US, are unheard of. Looking ahead At the moment Asian housing markets do not seem overvalued, although there are considerable risks that price pressures may build over the coming year. On the flipside, with policymakers alert to the risks of asset price bubbles including in housing markets, they may also be successful in containing house price increases, although efforts to increase housing supply are likely to take a few years to come through the pipeline. Policy is likely to be further tightened over the coming year as well and growth is expected to soften in the region and this may lead consumers to postpone the purchase of big-ticket items such as housing. Over a longer-term horizon, rising prosperity and urbanisation imply a considerably favourable outlook for Asian housing markets with the region expected to remain on a high growth trajectory. In addition, rising populations and high urbanisation rates in the relatively less developed economies of the region (where on average less than 50% of the population lives in cities) suggests that demand for urban housing and related infrastructure is set to increase dramatically over the next decade. Indeed, in some countries, urbanisation still has a long way to go: in Vietnam and China, urban residents comprise only 29% and 40% of their total populations respectively, compared with 67% in Malaysia and 80% in Korea. Furthermore, the role of private financial institutions in the provision of mortgage finance (which has traditionally been the remit of government supported housing agencies) should also grow. Although housing markets have been gradually liberalised since the Asian Financial Crisis, housing finance systems are considerably smaller than in other developed economies (again with the exception of Hong Kong, SAR China and Singapore) and the range of mortgage products more limited suggesting considerable scope for growth in mortgage finance13 . Indeed, the first mortgage backed securities in China, Hong Kong, SAR China and Thailand were only issued after 2004. However governments across the region are actively encouraging the development of asset-backed security markets primarily in order to deepen capital markets. But this suggests that as private mortgage finance and asset-backed security markets grow, the need for strengthened financial supervision and risk assessment closely coordinated with monetary policy will also increase. Otherwise the absence of long- established credit histories in the region and relatively poor financial and regulatory supervision (with the exception of Hong Kong, SAR China and Singapore) will pose a growing risk to financial sector stability. 13 Lee and Park (2009) and Arner, Lejott and Zibell (2008). Asian housing markets: bubble trouble? 8 References 1. Ahearne, A., Ammer, J. ,Doyle, M., Kole, L. and Martin, R. (2005). Monetary policy and house prices: a cross-country study, International Finance Discussion Papers 841, Board of Governors of the Federal Reserve System 2. Abraham, J. and Hendershott, P. (1996): Bubbles in Metropolitan Housing Markets, Journal of Housing Research, vol 7, pp 191-207. 3. Arner, D., Lejot, P., and Schou-Zibell (2008) L, Securitization in East Asia. Regional Economic Integration Working Paper No. 12. 4. Bruno, G., 2005, “Monte Carlo Analysis for Dynamic Panel Data Models,� United Kingdom Stata Users’ Group Meetings 2005–06, Stata Users Group. 5. Cardarelli, R. Monacelli, T., Rebucci A., and Sala, L (2009) Housing Finance, Housing Shocks, and the Business Cycle: 6. VAR Evidence from OECD Countries. mimeo 7. Glindro, E., Subhanij, T., Szeto J., and Zhu, H. (2008): “Are Asia-Pacific housing prices too high for comfort?� BIS Working Papers 8. Himmelberg, C., C. Meyer and T. Sinai (2005), Assessing high house prices: Bubbles, fundamentals, and mis- perceptions, Staff Report No. 218, Federal Reserve Bank of New York, September 9. International Monetary Fund (2010), Consolidating the recovery and building sustainable growth, Regional Economic Outlook, Asia and Pacific, October 2010, IMF 10. Judson, R., and A. Owen, 1999, Estimating Dynamic Panel Data Models: A Practical Guide for Macroeconomists,� Economics Letters 11. Lee, J. and Park, C. (2009), Global Financial Turmoil: Impact and Challenges for Asia’s Financial Systems-super, Asian Economic Papers, 8, issue 1, p. 9-40 12. Morgan Stanley (2010) New Property Measures to Dampen Speculation, but Not Prices, Denise Yam, http://www. morganstanley.com/views/gef/archive/2010/20101123-Tue.html 13. OECD (2007). OECD Economic Survey for Korea, Chapter 3 http://www.oecd.org/document/10/0,3343,en_2649_3373 3_38756234_1_1_1_1,00.html, OECD 14. Schroders (2010), China’s stop-go residential property market, Schroder Property Investment Management Limited 15. UN Habitat (2008) Housing Finance Mechanisms In Indonesia, Human Settlement Finance Systems Series, http://www. unhabitat.org/pmss/listItemDetails.aspx?publicationID=2548 16. Warnock, V., and F. Warnock (2008). Markets and Housing Finance. Journal of Housing Economics 17: 239-251. 17. Zhu, H (2006), The structure of housing finance markets and house prices in Asia, BIS Quarterly Review, December An Eye on East Asia and Pacific is a series of short notes related to the East Asia and Pacific Region prepared by the region’s economists. These notes are not peer reviewed or edited. They do not represent the official position of the World Bank Group, its directors, or management. Contact the authors for discussion on the individual notes, or Ivailo Izvorski (iizvorski@worldbank. org) about the overall series or to have a note considered. www.worldbank.org/eapeye Asian housing markets: bubble trouble? 9