The Strange World of Chinese Real Estate Liquidity

We are frequently so entranced with the meteoric price increases of Chinese real estate that seem to dwarf anything we’ve seen before that we frequently lose sight of details that really matter.  These details beyond price headlines really matter because many of these are the market dynamics that impact future changes.

One recently caught my eye and I will be honest in saying that I’m still trying to process the impact of what this data means.  The Chinese real estate market has extremely low levels of relative liquidity.

What I mean by that is that relative to the number of housing units in specific metro areas, there are actually a relatively small number of housing transactions.  Many times transactions will be reported for major metro areas in China, but there is frequently little context given that obscures the relative measure of liquidity.

To give you, some idea of what I mean the average home in the United States is sold roughly every 20 years.  This excludes housing units that are apartments, but gives us some idea.  Furthermore, this number does not fluctuate as much as you might think except in extreme periods like during and after the 2008 subprime crisis.

However, in China these numbers are astoundingly different.  As an example, in 2015 Shenzhen housing transactions relative to the number of housing units was under 1.9%. If each unit could only be sold once before being sold again, it would take Shenzhen 54 years before being resold.  What is amazing is that projected 2016 housing transactions in Shenzhen have gone down from 2015.  If current rates hold, unit turnover will fall to about 1.2% or approximately an 86 year turnover cycle.

In other cities, we see similar levels of low relatively real estate market liquidity.  2016 projections, estimate Shanghai unit turnover at under 2.7%, Tianjin at 3.4%, and Xiamen at 2%.  This would equal cyclical turn over rates of 37.6, 29.6, and 50.3 years.  To put these numbers in perspective, China only opened in 1979 so according to these numbers, each unit has yet to even sell a second time with many years to run.

Other cities show similarly low levels of relative unit turnover.  I’ll be honest in saying that giving some the peculiarities of the Chinese real estate market, I’m still trying to think through all the potential implications.  I have a couple of theories.

First, it is quite likely that Chinese real estate hit a tipping point a few years ago where real estate had been transferred on a preferential basis to existing urban residents under various schemes, the large majority of savings had been consumed to purchase housing units, and the only driver of the Chinese real estate market became channeled credit via state owned banks.  As I noted in a previous post, 2015 was the first year where the marginal new housing unit would have equaled a mortgage of greater than 50% LTV.  2016 is likely to be a significant increase from 2015 implying that the real estate driver has changed from a couple of years ago.  It is not longer unit transfer and savings consumption and credit growth of households.

Second, this seems to place a lot of pressure on Beijing to keep real estate prices elevated.  Many in China view this as a form of savings given the lack of other options and household income levels that are both under pressure and slower growing than real estate prices.  While households are little invested in the stock market, they are heavily invested in real estate implying much higher wealth dependence effect on real estate.

Third, it seems to imply much greater fragility in supporting the real estate market than is generally assumed.  Typically when prices go up, we see rapid growth in turnover volume but not only have we not seen this, in some cases we have see a decline in turnover.  The only apparent support of the real estate market is higher credit growth which means to support this, PBOC will need to pump even higher levels of credit to support price increases or stability.  This implies a lot more fragility in the real estate market given the relative lack of volume.

I think there are other implications but I’m still thinking through many of these implications.  However, I think it is pretty clear, there is a lot more going on here that requires thinking about the market microstructure of what is driving these dynamics.

One thought on “The Strange World of Chinese Real Estate Liquidity

  1. Christopher – As you ponder, a couple of things to keep in mind, some of which you probably already know –

    1. Those units sold in the 1980s and 1990s are very small, in poor condition, and are occupied by older retired people, if the units are occupied at all. They do not transfer until the government decides to knock the building down as part of some redevelopment plan. Many are still on the site of former danwei.

    2. Often, one of the “perks” of the transfer of farmer land for development is that farmers are entitled to buy a new apartment, part of the development plan, using the compensation given to them for their farm land. Farmers are transferred from relatively independent businessmen to mortgage holders, who need to struggle at service jobs to stay afloat. These units don’t transfer either, since new buyers don’t want to buy into a vertical community of former farmers.

    3. Job relocations are rare, particularly outside the city or province. And lots of SOE and government units and universities provide heavy subsidies to new employees so they can buy apartments, but those deals tie the buyer-employee to the employer for some years, or else they have to pay back at least part of the original subsidy. The deals with which I am familiar tie employees for 7 to 10 years.

    4. Chinese do “move” to get to a better school district, similar to the US practice. But that move may not entail a sale of the old unit, which might be required for hukou purposes or the subsidy purposes mentioned above. The “move” might entail simply buying a place, unfinished (undecorated) to provide the legal address required for the school. Again, no transfer of unit involved.

    5. There is, of course, the Chinese preference for new, rather than “previously owned.” If the old unit can function as a source of savings, as you point out, then one may as well wait another ten or fifteen years until the kid needs money for college in the US. Again, no transfer.

    6. There is some upgrading, people moving to a bigger newer closer apartment or villa. Sometimes, these involve a sale of the old unit, but again, if there is no other good place to save money, might as well keep the old place as well. If a couple has the cash flow, then the wife can buy the new place in her name. There are no multiple owners allowed on transfer documents.

    7. When the grandparents move in with the kids, they do not sell their old place. Might not be worth much, and may as well wait and see if the government buys the old place someday. Again, no transfer.

    8. As you know, the ownership records can be … fuza. Even in the current environment, there are plenty of vague ownership schemes. I don’t have any direct experience of this, but I would bet that there are transfers that are not otherwise recorded.

    9. I don’t know if this is still true, but for many years, businesses and government units were big buyers of blocks of apartments, perhaps entire buildings. Units were then sold to employees, possibly with the sort of subsidies mentioned above. This may have significant effects on reported prices and price increases.

    10. Real estate markets in the US, as everywhere, are lumpy – not such smooth curves in price changes or transactions. But markets in China are very lumpy. There are lots of state owned developers buying land from local governments, which need the revenue to pay for school teachers and policemen. Until there is a restructuring of fiscal relations between Beijing and the provinces (as I think did happen in Zhejiang last year), the pressure to keep land sales rising will keep real estate prices rising, even as it puts even more strain on households to come up with “first payments” and monthly mortgage payments. At some point there has to be an adjustment, but I keep waiting, and it hasn’t happened yet.

    My ability to read Chinese is poor, so I don’t have access to the published data, other than what shows up in English language sites, and my principal interest is not in real estate. I am offering you only anecdotal evidence, but that is based on some direct experience and lots of conversations with colleagues at school as well as my government contacts in Zhejiang and Liaoning. But I have no doubt that the low rate of transfers is a pretty good statistic. I do think that the reported percentage increases in prices, month to month and year on year, are fishy. I would like to see data by average square meters of unit, and location, and whether apartment or villa. And for Chinese, like American prior to 2007, real estate prices always go up. If there were a significant decline in prices, there would be a lot of unhappy owners, harmony would fail. So yes, propping up prices, and therefore credit, must be a part of macro strategy.

Comments are closed.