Does China Have a Housing Bubble?
China’s housing market is not generating the kind of financial system risks that developed in the U.S. during the decade prior to the Global Financial Crisis, in part, because China’s regulators have learned from our mistakes.
Now that fears of capital flight and dramatic currency devaluation have receded, the prospect of a housing bubble has become the latest leading cause of concern for China’s economy. Some recent headlines include:
China’s Property Bubble Keeps Getting Bigger—Wall Street Journal
Here's the Smoking Gun That China Has a Huge Housing Bubble—Bloomberg
China Property Bubble Could Cause $600 Billion in Bad Debts – Bloomberg
Worries Grow That China Faces a Perilous Property Bubble—Wall Street Journal
In China, Property Frenzy, Fake Divorces and a Bloating Bubble—New York Times
In this Sinology, however, I will explain why a residential property bubble is not among the long list of significant problems facing the Chinese economy.
China’s housing market is not generating the kind of financial system risks that developed in the U.S. during the decade prior to the Global Financial Crisis, in part, because China’s regulators have learned from our mistakes.
Chinese banks have not been permitted to make irresponsible mortgage loans, and homebuyers are required to put down a lot of cash. There is very little mortgage securitization and most banks hold mortgages through maturity, so they have a clear incentive to avoid lending to risky borrowers.
Sharp increases in home prices have been limited to a small number of cities, and overall, prices have risen in line with income.
Next year, the property market is likely to be much softer, but that is unlikely to create significant problems for China’s state-controlled financial system.
China does face a serious property problem, but it is a social and political problem. In many cities, most residents are priced out of the market and may never be able to afford to own a home. This problem— shared by cities such as San Francisco, New York and London—is a long-term challenge, and it’s one with consequences that are very different from a housing bubble.
Most Buyers are Owner-Occupiers, Not Speculators
Let’s start by addressing one of the biggest misconceptions about China’s property market: that most buyers are speculators. In fact, the residential market is driven by owner-occupiers.
CLSA, the Hong Kong-based brokerage firm, collects data from sales managers at new residential projects across China each month, and reports that between 2012 and 2015, the share of buyers who were investors was only 6% to 10%. That share rose to 18% in September, a level it last reached in 3Q10, but is likely to return to the mean in response to new government measures designed to curb speculative buying. Beijing, Shanghai and Shenzhen, for example, just raised the minimum cash down payment for a high-end investment flat to 70% of the purchase price.
One reason many people imagine that speculators dominate the market is that if we, or the analysts we listen to, know people who live in China, they tend to live in one of the handful of major cities where real estate investors are concentrated. But that is no more of a representative sample than the property markets in New York and San Francisco are of the entire U.S. market.
While the four so-called Tier 1 cities of Beijing, Shanghai, Shenzhen and Guangzhou get most of the attention, those cities account for only 4% of national new home sales, and 9% of China’s urban population. Most urban Chinese live in cities that most of us have never heard of, and that’s where 96% of new home sales take place. And in most of those cities, the share of speculative buyers is very low.
Moreover, China has an amazing number of large cities: there are more than 150 Chinese cities with a population of at least 1 million, while the U.S. has only nine cities and 50 metro areas of that size.
Homeowner Leverage is Low
In my view, the most important precondition for a bubble in any asset class is a high level of leverage because in the absence of high leverage, the consequences of a sharp price decline are limited.
In China, there is extremely low leverage among homebuyers because about 10% of buyers pay all cash, while for those using mortgages a minimum cash down payment of 20% is required by government regulation. In a September survey by the brokerage CLSA, property developers reported that 65% of buyers that month used cash down payments of 20% of the purchase price, while another 28% of buyers put down 30% or more of the purchase price.
And those numbers were for first-time homebuyers. For those purchasing their second home—including buyers who were upgrading their primary residence to a nicer place—the survey found that 56% made down payments of 30% and another 42% of buyers put down cash equal to 40% or more of the purchase price.
This is far from the 2% median cash down payment in the U.S. in 2006, which was the primary cause of the financial crisis here.
While there are lots of rumors about homebuyers using P2P loans to make their down payments, this isn’t supported by evidence, and seems unlikely to be occurring on a large scale, given the very high interest rates for P2P loans and the very small scale of total P2P loans. (All P2P loans are equal to less than 1% of total bank loans outstanding in China.)
Additionally, China doesn’t have subprime or no-doc loans, and it is extremely difficult for speculators to obtain a mortgage, so most sales to investors are all-cash. (As we noted earlier, Beijing, Shanghai and Shenzhen just raised the minimum cash down payment for a high-end investment flat to 70% of the purchase price.) In any event, about 90% of new home sales across the country are for primary residences.
Chinese banks have not been permitted to offer subprime mortgages, option adjusted-rate mortgages and other loans that allow borrowers to defer interest or principal payments. Home-equity loans are rare, so Chinese cannot use their house as an ATM.
A February 2016 survey by CLSA of middle-class Chinese families found that 96% of households owned one or more properties, but just over half of the home-owning households had no outstanding mortgage on any property. Of the 48% of households with a mortgage, the average outstanding mortgage debt was equal to less than two years of average annual family income. Only 11% of all of the middle-class homeowners reported an outstanding mortgage balance of greater than RMB 500,000 (US$74,000).
These high levels of cash create a low loan-to-value (LTV) ratio. A low LTV means the owner’s equity stake in his home is high, providing an equity buffer that absorbs the initial losses from a fall in house prices. For example, a low LTV meant that when Hong Kong property prices fell by almost 70% during the Asian Financial Crisis, mortgage delinquencies peaked at only 1.4%.
Prices Up Sharply Only in a Few Cities
Media reports have highlighted extremely sharp jumps in new home prices in a few Chinese cities, but that exaggerates the overall market. For example, while prices rose 83% in the four Tier 1 cities, between the start of 2011 and September 2016, prices rose by 21% during the same period in the Tier 2 cities, which account for 32% of new home sales (by volume). And in the Tier 3 cities, which account for 64% of new home sales, prices rose by only 4% during that time.
It is also worth noting that average income is significantly higher in the cities with higher housing prices. The CLSA survey of middle-class families cited earlier found that average annual disposable income in the four Tier 1 cities was 28% higher than in the Tier 2 cities, and was 51% higher than average income for households in the Tier 3 cities.
Incidentally, in the same survey, the middle-class families reported that, on average, they put 41% of their annual household disposable income into savings or investments, up from a 37% share in a comparable 2013 CLSA study.
Overall Prices Up in Line with Income
From an American or European perspective, China’s 7% average annual growth in residential property prices over the last five years may appear to be crazy. But this assessment would fail to put that data point into the context of similarly rapid income growth—a rise in wealth that may be difficult for residents of developed countries to comprehend.
With average annual nominal urban income growth of 10% over the past five years, the 7% average annual appreciation in new home prices is far less worrying. (Contrast that with the pre-crisis period in the U.S., when new home prices rose 11% per year between 2001 and 2006, while income rose 5% annually.)
Unprecedented income growth not only supports China’s remarkable consumption story, it also underpins a healthy property market. Over the past decade, inflation-adjusted urban income rose by 130%, compared to increases of about 11% in the U.S. and 2% in the U.K.
Demand Driver: Expectations for Further Income Growth
One factor driving strong demand for residential property is expectations of further strong income growth.
Seventy-three percent of middle-class families surveyed by CLSA reported that they expected their financial situation to be better in three years, with only 3% expecting to be worse off. Those families said they anticipate 7% annual growth in their family’s disposable income over the period 2016 to 2018. As a result, 54% of the households said they expect to be living in a bigger home over the next five years.
And Chinese families appear to be optimistic about their longer-term futures: 82% of those surveyed this year told the Pew Research Center they expected their children to be better off financially than their parents. In a similar survey by Pew last year in the U.S., only 32% of Americans said they expected their children to be better off financially.
We’ve Seen Similar Asset Price Inflation Cycles Before
China’s residential market—which has only existed since the mid-1990s when the government began to allow private, urban homeownership—has been very cyclical. Over the last decade there have been several periods when the market was very hot, as well as several when the market was very soft. None of those cycles resulted in financial crisis.
In 2007, for example, new home sales rose 27% YoY by volume, only to decline by 16% in 2008. The following year, sales jumped by 45%. Then, from 2010 through 2012, sales rose by 2% to 8% each year, before another hot year in 2013, with sales up 18%. This pattern continued, with a 9% fall in sales in 2014, and then a 27% jump during the first nine months of 2016.
Many observers have forgotten that we’ve seen similar gyrations in new home prices over the past decade:
In Tier 1 cities (which account for 4% of new home sales), prices fell for 7 consecutive months in 2014; prices are up 26% during the first nine months of 2016; prices are up 83% since the start of 2011.
In Tier 2 (32% of total sales), prices fell for 13 consecutive months in 2014-15; prices are up 12% during the first 9 months of 2016; prices are up 21% since the start of 2011.
In Tier 3 (64% of total sales), prices fell for 21 consecutive months in 2014–16; prices are up 4% during the first nine months of 2016; prices are up 4% since the start of 2011.
This pattern of sharp cycles is likely to continue for the near future, in part because of a herd mentality among Chinese buyers (of many asset classes), and in part because the government tends to intervene with measures designed to boost or cool the market. We are in the midst of yet another government cooling period, so it is likely that by this time next year, the market will be fairly soft (again).
Financial System Risk is Low
There are several reasons why I believe that despite continued volatility in residential sales and prices, the risks to China’s financial system are low. In fact, because homebuyers are required to use a lot of cash and because banks have not been permitted to make irresponsible mortgage loans, mortgages may be among the safest of bank assets.
In an earlier section of this report we explained that Chinese homebuyers who use a mortgage must put down at least 20% cash, in contrast to the 2% median cash down payment in the U.S. in 2006. It is also worth noting that Chinese banks have not been permitted to engage in the “financial engineering” that precipitated the U.S. crisis:
- Nearly one in 10 American mortgage borrowers in 2005 and 2006 took out “option adjustable rate mortgage (ARM)” loans, which meant they could choose to make payments so low that their mortgage balances rose every month.
- Nearly one-quarter of all mortgages made in the first half of 2005 in the U.S. were interest-only loans.
- In 2007, roughly 9% of outstanding mortgages in the U.S. were low- and no-doc loans.
As one academic paper* explained, “Standards for mortgage loans were fairly consistent in the decades prior to the development of the housing bubble. Most mortgages were 30-year fixed rate loans requiring a down payment of at least 20% or mortgage insurance if the 20% down payment requirement were not met. The borrowers also had to prove that their income was sufficient to ensure that the monthly mortgage payments would be manageable.”
This changed, of course, exemplified by the U.S. Federal Department of Housing and Urban Development’s (HUD) 2004 “Zero Down Payment” initiative, which HUD said was designed to “remove the greatest barrier facing first-time homebuyers—the lack of funds for a down payment on a mortgage.”
Low, or no down payments amplified the impact of falling house prices. Home prices peaked in 2Q06 and then declined just a bit, falling by less than 2% from 2Q06 to 4Q06. Despite that small decline the foreclosure rate rose by 43% over those two quarters and increased by 75% YoY in 2007. “This implies that mortgage default rates began to rise as soon as home prices began to fall. Speculators who bought homes (often with no money down) simply walked away from the property when the home price fell. Many never made even the first monthly payment.”*
Chinese banks, in contrast, are only allowed to issue the plain-vanilla mortgages that American banks used to rely upon. No subprime (which, in the U.S., had a foreclosure rate about ten times higher than prime mortgages), no no-doc loans, and no “option ARMs.”
There is little mortgage securitization in China, which means that almost all mortgages are held through maturity by the originating bank, which provides a clear incentive to underwrite carefully and consider the borrower’s capacity to repay. In contrast, when American banks often originated mortgages to sell, for securitization or other purposes, they faced no risks if the loan defaulted.
And, while the use of mortgages in China has been rising rapidly, context is also important here. Private homeownership, and therefore mortgage lending, has only existed in China for about 20 years, so both continue to expand. The level of mortgages-to-GDP rose to 19% last year, from 10% in 2007 and 3% in 2000. Still, this compares favorably to mortgage-to-GDP levels of 77% in the U.S. last year (and 100% in 2007) and of 69% in the U.K.
Moreover, because Chinese save so much, total household debt is equal to only 50% of household bank deposits—modest, compared to the U.S., where household debt is equal to 170% of household bank deposits, down from 237% in 2007.
Demand Drivers: Falling Mortgage Rates
One of the reasons that new home sales have been so strong this year (up 27% YoY by square meter through September) is significantly lower interest rates for mortgages (down about one-third from two years ago).
The weighted average rate for a mortgage was 4.55% in June, compared to 5.53% a year ago and 6.93% two years ago. And, while the central government has encouraged local officials in cities with hyper-hot residential markets to take some steps to cool off sales, Beijing has yet to raise mortgage rates, and I do not expect them to take that step.
In addition to boosting sales, lower mortgage rates have (along with higher prices) encouraged a higher share of buyers to use a mortgage (as opposed to paying all cash). This year, 90% of new home purchases were made with a mortgage, up from 67% in early 2011.
Demand Drivers: History, Upgrading and Demolition
In order to understand China’s property market today, it is important to understand a bit of history. The development of China’s commercial housing market over the past two decades is one of the world’s greatest and least recognized privatization success stories.
In just 20 years, China went from having one of the world’s lowest urban homeownership rates to one of the highest. Back in 1958, 86% of urban workers were employed by state-owned enterprises (SOEs) and government departments, and that share remained above 70% through 1989. At that time, most workers lived in public housing, which was provided at minimal cost by their government work unit. But between 1995 and 2001, the Communist Party laid off 46 million state-sector workers—equal to sacking about 30% of today’s U.S. labor force over six years! As a result, the state share of urban employment fell from 59% in 1995 to 28% in 2002. (As of last year, the state share was only 15%, with 85% of Chinese workers at private firms.)
The Party realized that it couldn’t throw all of those laid-off workers out onto the street, so they allowed most of them—as well as most remaining SOE and government staff—to buy their government-provided flats at a steep discount to the market value (which was, at that point, very low). This was the largest one-time transfer of wealth in the history of the world, as most of China’s urban-housing stock was handed over to its occupants, and it helped create the liquidity to fuel China’s brand-new commercial-housing market.
Because many of the old government-provided flats were of poor quality, upgrading or replacement of those units has been an important driver of demand for new, commercially-built apartments. For example, as recently as 2010, one-quarter of urban apartments did not have private kitchens or bathrooms.
And widespread demolition of old units is important context for the extraordinary building boom over the last 20 years. So many old apartment blocks were torn down, that we estimate housing space demolished in China was equal to 40% of new construction between 2001 and 2010. Another way to describe it is that the residential space demolished in China during that period was equal to one-quarter of the total U.S. housing stock in 2014.
Unsold Inventory Coming Down
Inventory levels of unsold apartments remain high in many cities, but stronger sales this year have had a significant impact. In the Tier 3 cities which account for 64% of new home sales, unsold inventory as of September was equal to 21 months of sales (calculated on a six-month moving average basis), down from the peak inventory level of 37 months in June 2015.
In the Tier 2 cities which account for 32% of new home sales, unsold inventory was equal to seven months of sales in September, down from the June 2015 peak of 17. And in the four Tier 1 cities which account for four percent of new home sales, unsold inventory was equal to nine months of sales in September, down from 13 months at the peak.
Recognizing that unsold inventory levels remain high, developers have responded cautiously to a hot sales market this year. Floor space under construction rose by only about 2% YoY during the first nine months of the year, after falling by about 1% last year. Similarly, residential real estate investment rose only about 5% this year, after rising by less than 1% last year.
Some cities that had very high levels of unsold inventory at the start of this year, such as Wuxi and Zhengzhou, have seen those levels come down to the extent that they recently announced new purchasing restrictions, in an effort to cool off hot markets.
One of the most frightening stories about China’s property market is that the country is full of “ghost cities” of empty apartment buildings and tumbleweed, with Lon Chaney, Jr. lurking around the corner. These stories are widely believed, but are plagued by the absence of hard data and a misunderstanding of the Chinese market.
The “ghost city” reports we’ve read are usually based on observations along the lines of, “we drove past buildings that appeared to be completed yet unoccupied.” This is problematic, unless those analysts have x-ray glasses that are far better than the pair I bought as a kid from Boy’s Life magazine.
It is important to understand how China’s property market differs from that of the U.S. and other developed countries. For example, about 80% of new apartments are sold on a pre-sale basis, which means the contract is signed from one to two years before the building will be completed. And the majority of new apartments are sold unfinished, which means the owner takes possession of a bare concrete shell.
Finishing the apartment—everything from floors to kitchens and bathrooms—can take another three to six months after the new owner takes possession, and many people wait to move in until most other owners have finished their decorating, to avoid living in the midst of a construction site.
Another unique factor is that most new apartments are being built outside of city cen¬ters, because the traditional downtown area is too congested and land is either unavailable or too expensive. The government is encouraging developers to focus on new areas away from downtown by planning infrastructure—everything from subways to light rail and hospitals and schools. But these projects take time to complete.
You may have seen a report by CBS News’ 60 Minutes program titled “China’s Real Estate Bubble.” It was first broadcast in March, 2013, and focused on the new Zhengdong district of the city of Zhengzhou.
In the Zhengdong district of Zhengzhou, as in many other Chinese cities, apartment buildings were completed several years ahead of the opening of the subway line to the area. Zhengzhou, with an urban population of 6 million (about the size of the Washington, DC–Arlington, Virginia Metro Area), started constructing a subway system in 2009 planned for six lines, covering a total of 125 miles. The first line opened in late 2013. Many people in Zhengzhou decided to buy apartments in the new area with the intention of not moving in for several years, based on the view that housing prices would rise after the subway lines were completed.
60 Minutes host Lesley Stahl introduced her report by saying, “So this is Zhengzhou. And we are on the major highway, or major road. And it’s rush hour . . . and it’s almost empty.”
Just two years after that report was first broadcast, I visited Zhengzhou and went to some of the same places . . . and I spent much of a day sitting in traffic. There was one key difference: I visited soon after the new subway line opened. The commute from the new residential district of Zhengdong to the downtown business center was more convenient, so families who had bought newly built apartments over the prior few years finally moved in. Zhengdong was no longer ghostly—a pattern common to many new housing developments that I have visited across China. A five-minute video of our visit to Zhengdong is on the Sinology page of the Matthews Asia website. In fact, unsold inventory in Zhengzhou fell so low over the summer that new home prices rose 16% just during the third quarter of this year.
Affordability is a Serious Problem, but it’s Not a Bubble
Based on average income for Beijing, almost no one in that city can afford to buy a new home. While true, that data point is as helpful as stating that based on average income for San Francisco, almost no one there can afford to buy a house, or that based on average global income, Porsche should not be able to sell a single car worldwide.
But, the housing markets in Beijing and San Francisco (and Porsches) are not targeted at average income groups. (In fact, according to the California Association of Realtors, only about 11% of San Francisco households can afford to buy a home at the median sales price, assuming a 20% down payment.)
In China, the market for commercially built urban homes, which has been in existence for only two decades, is focused primarily on the middle class and wealthy, who are a relatively small share of the nation’s population but number in the hundreds of millions of individuals. Some of the data we have cited earlier in this report—notably that about 10% of recent first-time homebuyers paid all cash; that the minimum cash down pay¬ment for those using a mortgage is 20%; and that about half of middle-class homeowners have no mortgage balance—reflect that for the target population, housing is relatively affordable.
And, as we noted earlier, 64% of new home sales take place in Tier 3 cities that most of us have never heard of, where average prices are 77% below those in the Tier 1 cit¬ies, such as Shanghai and Beijing.
The fact that the average residents of the largest Chinese cities, like the average residents of most major cities around the world, cannot afford to buy a home in or close to downtown does represent a significant, long-term social and political problem, but is not a bubble or a financial sector risk. This problem does explain why the Chinese government has recently accelerated spending on subsidized housing, and why 27% of total new residential land supply has gone toward affordable housing projects between 2010 and 2015.
What Could Go Wrong?
As I’ve explained in this commentary, I do not think China’s residential property market is a bubble. It is expensive, which is a social problem, but homeowner leverage is low and irresponsible mortgage lending is not allowed, limiting risks to the financial system.
So, what could go wrong?
The biggest risk is continued government intervention in the market - one day to cool it down, and the next day to heat it up—efforts which distort demand and prices and which are prone to overshoot. The Chinese government needs to reduce significantly its level of intervention in the market, which should reduce the frequency of policy-induced volatility.
At the same time, the government needs to establish the framework for a well-regulated, mature and broad capital market, providing investment options other than real estate for an increasingly wealthy population.
Finally, and most importantly for the long term, the government needs to take steps to significantly increase the availability of affordable housing, especially for rental units.
* The Journal of Business Inquiry, 2009: A Summary of the Primary Causes of the Housing Bubble and the Resulting Credit Crisis: A Non-Technical Paper by Jeff Holt