China’s property bounce is unsustainable

Advertisement

Courtesy of Also Sprach Analyst.

Standard Chartered put out a note last week, suggesting that the Chinese “residential housing sector is coming back”.

Their view is based on the fact that sales volume growth has been back in positive territory on a year-on-year basis, and growth in floor area under construction is also bouncing back. The other observation is that new home inventory is declining from a peak to a seemingly normal levels, and developers are now more active in bidding for land.

Advertisement

Of course, the note was out before they got a taste of sluggish golden week sales, so perhaps they might not have said that if the report was dated after the holiday. But anyway:

image[13]

image[14]

Source: Standard Chartered

But can we conclude that this recovery is for real on this basis? Not in my view.

In the short-term, although prices are holding up, the positive momentum since the summer is losing steam, as I’ve noted.

Advertisement

The table below from Credit Suisse shows the sluggish sales during the Golden Week. Although transaction volume year-to-date is higher this year than last year, this Golden Week saw a 28% yoy decline in primary market sales in major cities, and 71% decline compared with the previous week.

image

Perhaps a more fundamental reason for disagreeing the analysis by Standard Chartered is their view on supply and inventory. It is a very common to see analysis on the real estate which refers “supply” to the flow of newly constructed homes. If real estate is like other products such as food, then it is fine to think of “supply” as the newly constructed homes which are being sold in a given period, and the StanChart’s inventory period analysis is relevant.

However, real estate is not food. Real estate does not disappear after being sold from a developer to a new homeowner, unlike food, which is usually eaten after being sold. Instead of just looking at the “flow”, we also need to look at the “stock”. We all know that many Chinese cities have a phenomenon that real estate investors and speculators purchase more than one apartment, and leave most of them empty. For some places like the famous ghost city of Ordos, empty houses are everywhere. For top tier cities, you are quite probably able to find a lot of empty apartments as long as you know where to look for them.

Advertisement

Ultimately, we imagine most of the owners of these apartments would like sell and cash in at some point, otherwise there is no point in purchasing more flats than one actually needs. In other words, we believe that some of this massive yet unknown number of flats and houses will be on the market sooner or later. What we do not know is when this part of the housing stock will be out on the market. As Jinsong Du of Credit Suisse pointed out late last year (emphasis ours):

Many popular supply/demand models implicitly assume homebuyers will never sell vacant units. Ministry of Housing data indicates that developers’ existing land banks can produce almost 100 mn new housing units. Our calculation shows that some potential resale of vacant units plus developers’ existing land banks should already be enough to satisfy 10 to 20 years’ housing demand.

Once one considers the fact that owners of these vacant homes will one day put their apartments onto the market, the level of inventory currently held and being developed by real estate developers is made much less relevant. Property developers may be destocking now, and they may feel tempted to go and get land and build more after destocking. However, the apparently “normal” inventory of new homes means nothing as far the the total housing stock is concerned. In other words, the analysis by Standard Chartered, which based only on the current level of inventory and the average pace of sales of new homes is completely flawed.

Advertisement

Then there is the problem with assessing the data. I, for instance, only rely on National Bureau of Statistics’s data for the direction of home prices, not for the actual prices. The mostly watched NBS’ data for new home prices reflects the prices of flats being sold in that month, without any adjustment for location, quality, etc. In other words, the number for August 2012, reflects prices of a certain sample of properties which happens to be on sales during the period, which is completely different from the sample of properties in August 2011. In other words, they are actually not comparable.

Jinsong Du of Credit Suisse also points out that as most Chinese cities are expanding, new home sales are happening increasingly in places further away from the city-centre. With the composition of the NBS’ index being located in increasingly remote areas, it is not surprising that the index does not capture the actual price increase. And unfortunately, even the commonly used 100-city price index compiled by Soufun does not use a fixed sample of properties.

So the most popular price indices available suffer from the same problem, namely that they have most likely understated the price increase. As a result, one could spin that Chinese people’s income has been increasing faster than home prices based on official data. However, that clearly does not match the reality as experienced by people on the ground.

Advertisement

image

Facing this problem, Jinsong Du of Credit Suisse used some fixed samples of properties in a few Chinese cities, computed the price increase for the fixed samples of properties, and compared that with income growth of those cities for the same period.

The result? In the last few years, the increase of home prices for the selected properties in major cities have far outpaced the increase in income. This is exactly what people on the ground actually experience:

image

Advertisement

image

Looking back even longer, the divergence is more significant:

From a long-term history perspective, housing prices in the downtown area of tier-one cities have increased more than 20 times since around 1990:

Fangzhuang Estate, one of earliest commodity house projects in Beijing was sold at an ASP of ~Rmb1,500/sq m around 1990, and now the ASP in Beijing downtown has exceeded Rmb30,000/sq m;

In Shanghai, the ASP for newly built commodity houses in Xianxia area, Changning district was around Rmb1,600/sq m in 1990, and now the ASP in this area has reached nearly Rmb39,000/sq m;

In Guangzhou, Donghu New Estate, one of earliest local commodity house projects was sold at Rmb900/sq m in late 1980s, and now in Guangzhou downtown it has exceeded Rmb22,000/sq m.

By contrast, the disposable income in Beijing, Shanghai and Guangzhou only increased 828%, 743% and 591% over 1993~2011, respectively. (disposable income data earliest available since 1993).

image

So unfortunately, Chinese real estate prices got ever more unaffordable in the past decade. Even though wages are rising at double-digit rates (despite economic slowdown), home prices remain unaffordable, simply because prices have gone up far quicker than income for too long. With large sold-but-vacant inventory within the total housing stock, it is hard for us to understand how the current “recovery” in prices is actually turning into a sustainable recovery.