Unless the government curtains new construction in lower tier cities, it is hard to see how the housing boom will sustain.
Between 2003 and 2013, housing prices almost quadrupled in China’s most populous cities (Fang et al. 2016). This growth dwarfs the 50% increase in US home prices between 1996 and 2016. A massive construction boom has accompanied China’s price boom. According to China’s Statistical Yearbook, Chinese annual construction reached 12 billion square feet in 2013, which is triple the US figure.
China’s housing boom is hardly unique. Real estate booms and busts in Thailand, Indonesia, and South Korea precipitated the 1997 Asian Financial Crisis. Home values have increased dramatically in Mumbai and Delhi since 2009. After years of robust price growth, Brazilian housing prices have been falling since 2015.
The ubiquity of housing bubbles
In a previous study (Glaeser 2016), I argue that housing bubbles are ubiquitous because real estate is a particularly fungible form of capital. An investor who acquires a Kolkata jute factory through a default may struggle to eke value from the enterprise. Managing or reselling an apartment building or a retail mall is far easier. Investments in Chinese apartments are far less subject to insider expropriation than investment in companies listed on the Shanghai stock exchange. Consequently, arm’s-length lenders have fuelled the leveraged real estate industry for centuries. Dutch and British backers supported the colossal success and colossal bankruptcy of Robert Morris’s multi-million acre land empire in the 1790s.
Moreover, investment euphoria is understandable during epochs of extreme economic growth. Chinese real estate optimism is easily understandable given the tenfold growth in Chinese income per capita over a generation. The housing boom in the US and Europe that occurred from 2001 to 2006 is far less comprehensible because it was accompanied by generally lacklustre economic performance. China’s immense housing boom is far more reminiscent of the earlier US real estate bubbles that accompanied domestic economic growth spurts, such as the 19th century transportation revolution or New York’s skyscraper frenzy of the 1920s.
Sensible models could easily justify the high property prices that were paid during most of the past real estate booms in the US (Glaeser 2013). The high prices paid for Alabama cotton land in 1819 or Iowa wheat land a century later seem almost sensible given the high global prices paid for those crops. High commercial rents in 1929 seemed to justify sky-high construction. Similarly, Fang et al. (2016) argue that there is no Chinese housing bubble since prices are rising about fast as incomes.
Forecasting the housing market
The great flaw in housing market forecasting is to focus on demand rather than supply. Forecasting demand is nigh impossible. Who knows what future billionaires will be willing to pay for homes in Belgravia or apartments in Hong Kong? However, when supply is unconstrained and land is abundant, elastic supply should ensure that prices hew closely to construction costs. The Las Vegas bust of 2007 was quite predictable given the region’s few building regulations and vast supply of desert. Similarly, the land buyers in historic Alabama and Iowa would have avoided losses if they had recognised how easy it is to grow cotton and wheat in different parts of the globe, such as Egypt and Russia.
In a recent paper, my co-authors and I attempt to forecast future Chinese housing prices, assuming that in the future China will become a ‘normal’ country where buyers are willing to pay about 10 times their incomes for housing (Glaeser et al. 2016). By combining this assumption with forecasts of future urbanisation and income growth, we predict future housing demand. We then predict future prices by looking for the intersection of this demand with future supply.
Our estimates suggest that the high prices paid in tier-1 cities, such as Beijing and Shanghai, could easily persist. The high prices paid in lower-tier cities are far less sustainable, unless the central government radically restrains the current rate of housing supply growth. The growth rate of housing is actually higher in lower-income cities in China, and that is also true in the US. This flood of new homes in less economically successful places makes it hard to imagine how prices will stay high.
Housing bubbles: The costs outweigh the benefits
Given the enormous role of Chinese governments in the Chinese housing market, we conclude that China could make a housing bust far less likely by radically restraining new construction. But restraining construction would also cause short-term construction unemployment and economic losses from reduced urbanisation. More generally, overly optimistic real estate bubbles may even have larger economic benefits (Glaeser 2016).
The existence of agglomeration economies or coordination failures can mean that cities are normally too small. Real estate bubbles can solve some of these problems. A remarkable number of the tallest buildings at the heart of Manhattan’s hyper-productive agglomeration were built during the overly optimistic 1920s. Real estate bubbles also appear in the development of Chicago, Los Angeles, and Tokyo.
While it is possible that overly optimistic real estate investors compensate for other market failures, there are reasons to be sceptical about such a rosy view. In China, housing construction may be occurring at the wrong quality level, in the wrong places, and at the wrong time. About 20% of China’s urban housing is vacant, often held empty by investors (Glaeser et al. 2016). This fact suggests that the homes were built too early and at too high a quality level to be appropriate for current urbanites. Moreover, the forests of empty skyscrapers in China’s lower-income interior cities suggest that development has often targeted the wrong metropolitan areas.
The largest downside of real estate bubbles was shown in the financial market meltdown that the US experienced after 2007. The costs of such meltdowns are welfare rectangles, while the benefits of more urban development are at best welfare triangles (Glaeser 2016). Consequently, the downsides of bubbles gone wrong can be far higher than the benefits of extra building.
Urbanisation is the natural partner of economic development, and real estate development is a major part of urbanisation. The tendency of the real estate sector to boom and bust therefore creates financial and real risks that can bedevil any economy. As more developing economies urbanise, development economists will have to think more about the causes and consequences of housing bubbles.
Fang, H, Q Gu, W Xiong and L Zhou (2016), “Demystifying the Chinese housing boom”, NBER Macroeconomics Annual 30(1): 105-166.
Glaeser, E (2013), “A nation of gamblers: Real estate speculation and American history”, The American Economic Review 103(3): 1-42.
Glaeser, E (2016), "Real Estate Bubbles and Urban Development", NBER Working Paper No. 22997.
Glaeser, E, W Huang, Y Ma and A Shleifer (2016), "A Real Estate Boom with Chinese Characteristics", NBER Working Paper No. 22789.
Photo credit: Jumay Designs and Getty Images.