Given its mounting impact on China’s macroeconomic growth and financial stability, the real estate sector attracts enormous attention from both policymakers and private investors. The focus has been on residential housing prices, which rose at an unprecedented pace during 2009-2010, when monetary policy was eased in the wake of the global financial crisis, and a substantial number of banking credits found their way to the property market. Worried about a looming property bubble and the social unrest caused by the worsening housing affordability, authorities have implemented several rounds of tightening measures since 2010, and these appear to have somewhat cooled down the market. There are still many questions as to how far the price correction in the housing market should go, however, since the quality of the official data has made it very difficult to evaluate the disconnect between housing prices and economic fundamentals. This report discusses recent efforts to more accurately measure housing prices in China and the likelihood of a nationwide housing market collapse in the near future.
Background: China’s Housing Market
China didn’t have a truly private real estate market until 1998, when the government ended its practice of allocating housing units and private housing ownership became widespread. After more than a decade of development, the real estate sector has become a major growth engine for China’s economy. On the activity side, fixed asset investment (FAI) in property in recent years has accounted for 20 percent of total FAI and is estimated to have made up around 10 percent of China’s GDP. Property construction has led to growth in demand for commodities, automobiles, household appliances, furniture, and heavy machinery and equipment. On the financing side, revenue from land sales became the most important off-budget income source for local governments, and real estate loans—including both mortgages and loans to property developers—accounted for about one-fifth of total loans, sharply increasing the exposure of the banking sector to the property market.
By far, the private residential market is the predominant component of China’s real estate sector, explaining the majority of the property FAI and more than 80 percent of total commodity property sales.
The Puzzling Nature of China’s Residential Property Prices
Since the private housing market has a short history, the bulk of China’s housing stock is relatively new and hasn’t been traded multiple times. Therefore, accurately measuring the price change on new units should capture most of the variations in housing values. The relevant Chinese statistical agency regularly publishes two sets of price data for new residential homes. One is the average selling price (ASP) for all new homes, calculated by dividing total sales in value by total sales in floor space at the national, provincial, and city level. The other, called a 70-city home price index, computes the average transaction price for all units sold in the same sampled building complex and then uses the volume-weighted average price for all complexes to measure the sales price of new homes in each city. Based on these two series, China’s housing prices showed relatively stable growth over the past decade, and there hasn’t been a great boom that could lead to substantial overvaluation, especially since 2008 (Figure 1).
Both series tend to understate the true price appreciation, however. The ASP series makes no attempt to control for the diluting effect of the rising share of new home supply in inland provinces, which tend to have lower home prices than coastal cities. The 70-city index does not take into consideration that new homes within the same city can be further away from the downtown area, nor does it adjust for the potential quality deterioration over time caused by the increasing density of new housing complexes, both of which could create downward bias on the average home price.
The “repeated sales” methodology, which uses only properties sold at least twice in the historical sample, could largely overcome the location and quality problem, and has therefore become a widely accepted standard across many countries to determine price trends in a housing market. This approach cannot be employed in a rapidly growing and urbanizing country such as China, however, where home sales are dominated by new homes, and repeated sales records are very limited and of questionable quality.
Recently, several researchers adopted the hedonic regression approach to construct housing price indexes that can directly measure housing quality adjustment over time by investigating the relationship between housing prices and the structural and locational attributes of housing. They find that at the national level, the real housing price appreciated at an annual rate of 16.7 percent between 2006 and 2010, significantly higher than that suggested by the ASP index (7.7 percent) and the 70-city index (4.1 percent). Similar trends can be found in city-level data (Figure 2). Given that construction costs in this period were fairly flat and wages for construction workers accounted for only a small share of housing prices, it appears that large growth in land values was the major driver behind the house price appreciation.
Empirical studies on 35 major Chinese cities show that between 2004 and 2011, the inflation- and quality-adjusted land price rose at more than a 20 percent annual pace in all but three years (Figure 3). The dramatic price appreciation during 2009-2010 can be at least partially explained by the loose monetary policy and the support from local governments, which obtained a major portion of their revenues from the sale and leasing of state-owned land. The sharp increase in bidding activities by several major state-owned enterprises may have played some role as well, although it remains an open question as to why they became so much more active in the housing market over the past few years.
A Nationwide Housing Market Collapse is Unlikely
Despite the rising uncertainties in the global economy, policymakers have vowed to maintain the tightening measures on the real estate sector. Although it is still unclear how long the housing price correction will last, especially in mega-metropolitan areas such as Beijing, Shanghai, and Chongqing, where current price levels appear to have significantly disconnected from fundamentals, a nationwide housing market collapse is unlikely to happen.
On one hand, a large portion of funding for residential property purchases comes from a household’s own savings, so the mortgage-to-GDP ratio is still relatively low when compared with other developed countries, implicating that Chinese households have a higher tolerance of housing price drops and the delinquency rates for mortgage loans should be low (Figure 4). On the other hand, the ongoing weakness in the housing market is largely driven by the government, and its goal is to restore affordability while not crashing the market.
The underlying demand for housing continues to be strong thanks to the high savings rates, limited investment options, and lack of a property sales tax. The double-digit income growth has helped push up housing prices as well. The rapid urbanization process, which moved 470 million people from rural to urban areas over the past three decades, is forecast to bring another 160 million people to cities by 2020, and this will continue to be the main growth driver in China’s residential housing market in the near future.
 In 2010, new home sales accounted for 87 percent of total sales in value terms and 64 percent of sales in floor space terms.
 Even for the small share of existing homes, owners have incentives to underreport the real value of their homes to avoid transaction taxes and capital gain taxes.
 Starting in 1997, the series covered 35 major cities, and expanded to 70 cities in 2005. In 2011, an overall home price index for all 70 cities was suspended following widespread criticism of data quality.
 The method was initially proposed by Bailey, Muth, and Nourse in the 1960s and was further developed by Case and Shiller in the 1980s.
 Again, owners of existing homes have incentives to underreport the real transaction value of their homes to avoid transaction taxes and capital gain taxes.
 Jing Wu, Yongheng Deng, and Hongyu Liu, “House Price Index Construction in the Nascent Housing Market: The Case of China,” National University of Singapore and Institute of Real Estate Studies Working Paper Series, IRES2011-017, June 2011; Yongheng Deng, Joseph Gyourko, and Jing Wu, “Land and House Price Measurement in China,” National Bureau of Economic Research, NBER Working Paper 18403, September 2012.
 Economist Intelligence Unit, “The Sustainability of China’s Housing Boom,” 2011.