SPEAKER 1: The following is part of Cornell Contemporary China Initiative lecture series under the Cornell East Asia Program. The arguments and viewpoints of this talk belong solely to the speaker. We hope you enjoy.
SPEAKER 2: Professor Yongheng Deng is a Provost Chair Professor and Director of Institute of Real Estate Studies at the National University of Singapore. He's also head of department of Real Estate at NUS Business School. Prior to joining NUS, he was a full professor at the University of Southern California.
He has received numerous awards, and holds various titles. He currently serves on the World Economic Forum's Global Agenda Council for the Future of Real Estate and Urbanization. And he's also a special advisor to the Bank for International Settlements Representative Office for Asia and the Pacific.
So Professor Deng was the 50th president of the American Real Estate and Urban Economics Association, the first non-American ever to hold this position. Is
YONGHENG DENG: Well, good thing about I heard this invitation is, gee, I'm going to coming to Cornell, I see, because it's wintertime, right? I'm so excited. Because in Singapore, the government only put the fake snow during the Christmas on major shopping street.
I talked to my wife, I said, I'm going to send you some picture about the real snow in Cornell, right? But unfortunately, today I didn't see any snow. But good news is that means I should come back to Cornell next winter to give a talk to do some lecture or to do some research, collaborative research work. I should come more often.
In Beijing, if we ask family of three purchase 900 square feet condo units using a 30-year residential mortgage loans with the coupon rate, at that time, and 20% down payment, we calculate what's the percentage of monthly income the family need to pay for mortgages per month. Beijing, that is 6% of the household income need to pay for mortgage to buy this 900 square feet condo units. Shenzhen 112%. Shanghai, where I was growing up, 66%. And Hangzhou is 60%.
To put these numbers in perspective for those who never work in the banking industry or mortgage companies, I used to regulating mortgage-backed securities in Washington, DC, for regulating Fannie Mae or Freddie Mac. So that's back in the early '90s. And early '90s, we set up the capital regulation model for Fannie or Freddie. Any loans they are not allowed to purchase, which means high risk loans, there's one condition. One of the conditions is if the payment to income ratio is more than 38%, these are risky loans. Fannie or Freddie should reject any bank to issue this kind of loan.
So look at the chart. Many of those above exception to above the 38% were rejected according to the regulation US federal government set at the time. Of course, people will argue, say, China and US are different. Chinese people sort of pay more attention to investing in the housing market.
However, anyway, 38% thinks about that sort of that you buy a house, you need to use minimum about three years your income, you don't eat, and you don't do anything, then you can manage to buy that. So this is not a unreasonable number. So that's one situation shows how difficult the high price will cost a household in Beijing.
There's another related added risk regarding the global concern. And it's actually, in addition to the individual borrowing, the household borrowing too much and then cannot afford to pay, the government also borrow a lot. So that's the total growth of debt in China compare to GDP.
In 2000, the total debt to GDP is roughly about 120%. But 15 years later by 2014, the debt to GDP growing fast, became 282% of GDP. Around that, we can see the non-financial corporation, that means those real estate developers, other SOE companies borrowing money to build the housing, that increase from-- the share became much, much larger.
And then, also, we can see all those debt we break down into household debt, real estate sector debt and real estate-related sector debt, and government debt. Household debt is roughly about 8%. Real estate sector debt is about 10% to 15%. If we add this all together, it's real estate-related debt. We can add up with 40% to 45% of the debt are related to the real estate development in China.
What make especially the global concern even more is there's a new local government bank finance vehicle, the local government loan. It's in the middle chart here. It shows a local government financial vehicle-- that's another study I just did a presentation last Friday at MIT-- has increasingly sort of a risk imposed to the economy. Because by looking at the market share, it increased very fast.
And then those local government use the land as a collateral. They borrow the money from the bank. They say, don't worry, I have a lot of land. If you were concerned about whether I can be able to pay back or not, I sell a piece of land, I can pay you back. However, if the real estate and land are mispriced, there's a bubble. Then if there's a correction in the price of the land, whether those local government can pay back those loans, that's another question.
And based on our calculation, the compound annual gross of those local government borrowing has increased by 27% each year since 2007. So there's a huge increase in the local government debt.
On top of all those, there's also the political implication to those real estate booming. This is one of the reasons I work with my NUS colleague Bernard Yeung, who is the dean of the business school, who is a very famous corporate finance professor, and Wu Jing, my co-author in Tsinghua University.
Now, this is a citation, a quotation by the London Telegraph that in 2013, they say, well, based on Professor Wu Jing, Professor Deng, Randall Morck, Bernard Yeung study, city government spending on environmental improvement is actually more difficult to get promoted. Their title pick is green politician is less likely to be promoted in China. How should you get promoted? If you sell more land, you build more real estate development, and then you push up the GDP, you got promoted.
And then even up to the last week, The Economist quote back our study again. The January 30th this year, The Economist has a article called Grossly Deceptive Plans. They cited our 2013 study. Say, well, those GDP goals targeted actually is sort of deceptive, because local government tried to use the land, sell more land, and than try to build up GDP by those kind of growth is not sustainable.
This actually based on our these studies. We study about 283 major cities' mayors and party secretary in China. Party secretary is the number one boss. Mayor is number two boss. We look at these two government officials over the 283 cities over past 10 years. We know, for example, their track record, their education degree, their salary, their background, their income, and their performance.
And then we all say, what kind of government official get promoted faster than others? And then we study if the city mayors, the party secretary in a city, they can sell more land, do more real estate development, their GDP goes up faster, about one standard deviation faster than the rest of the mayors. Their promotion odds, mayor, will be increased by 10% than the rest. Party secretary increased by 5%.
However, we also have the data during the term, their term, in the past 10 years, how much money they spend on clean air, water treatment, sewage treatment. We found, unfortunately, if they spend one standard deviation more on those environmental-related expense, their promotion odds drop by 8.5%.
So that's the quote after [INAUDIBLE] based on Professor Deng's study in China, green mayors got difficult to got promoted.
So because all those concerns, why there's a debate, say, we don't understand whether China is risky or not risky, there's a ghost town or not ghost town or China has crash landing or not. The major concern is the data quality. Because in China, not like the in US, I talk to [INAUDIBLE]. I talk to [INAUDIBLE]. They do a lot of fantastic high quality studies in US, because they have high quality data.
But in China, you ask the government, what's the supply/demand, how much vacancy, no one can give you a straight face answer.
So we thought there's not only we concerned about that, that actually, this report, Financial Times in March 2013, fears about Chinese property bubble. However, there's doubts over increasing property price.
And then Wall Street Journal also cited one of the studies by Joe Gyourko and myself. That's a NBER working paper. We argue, say, well, actually, what Chinese statistic bureaus reported datas is not trustworthy. So that's immediately sort of quoted by not only the Wall Street Journal and New York Times, but also Chinese referenced the [NON-ENGLISH]. So that became a great concern about a quality of data.
So we don't have reliable data, and now we cannot make sort of confident judgment whether there's sort of the oversupplied, whether there's a risk of a crash landing, whether prices are overpriced. So the discussion in remaining after the session, I will try to list out what we can say we know about China's housing market and land market, or what we need to know about the market.
The first step is we try to make sure we can provide some contribution to provide that data, which the policymaker, the economist, the planners, or even individual household can trust about those data. And there we can debate whether there's an oversupply on demand, whether it's overpriced or not in a more confident way.
So what we do in a first step is that we create this Chinese Residential Land Price Index. This is a joint work with Wharton and Tsinghua University and our NUS Institute of Real Estate Studies. We collect about 35 major cities in China, not just East Coast. That 35 major city cover the East, Middle, and West. We have all those transaction-based data. It's not the appraisal, not the government statistic bureau's data.
Government statistic bureau's data there was a very well-known common joke. Say, well, they always take the average report. And then the local bureau report the average, and then they report to provincial. Provincial say, OK, what's the central government target? And if it's not, they massage the data a little bit, report update.
So when you aggregate to the state-- country's bureau of statistics, the data is already, I think-- it's not accurate, right? right?
And then there's another joke. Say, OK, since it's a average. Average is a mix between the low-end price and higher-end price. And the government want to cooling down the housing market, say, next week the price has to be dropped by 5%. If your city don't have a price drop, the mayor, you're gone, fired.
The mayors now all got a PhD degree. Very smart. They pick up a phone, call the specific bureau or sort of housing authority. They say, in the next month, you can only sell the affordable housing, the low-end housing. So what they do is they sell more low-end housing, because the prices average is total sales revenue divided by total square meters. So you see more low-end housing that are reported. OK, we achieve the affordable housing goal. The house price drop. But everyone live in the city knows the price is not dropped.
So that's why we do this. We do a sort of traditional way. It's called constant quality house price sort of index. That's by control all the hedonic factors. Because we know the land value is not only measure by this piece of land, but also attached amenities, for example, whether it's close to the highway or there's a shopping mall nearby will affect the land value. So we control all those hedonic factors and how we calculate a pure constant quality price index.
Actually, this is the first constant quality price index we created in China. And also, it will be the first constant quality land price index in the world. Because only in China, you observe the separate land transactions. In US, the land value is included in the bundle of the housing sales.
So every year, you'll pay a sort of property tax. The government will say, well, this is your property value. Government assume about 40% is the land value, 60% is the price value, but that we don't observe the transaction dollar value for the land bought in China, because all the land are owned by the government. Government transfer the land use rights to the developers to build. For residential, it's 70 years. And then commercial, it's 50 years. Industry is 40 years.
So that during each transaction, we have a transaction price. And now we use that transaction price, we build this land index. That actually is very useful. A lot of studies, we try to understand the breakdown of the land and housing value. Only the Chinese data can allow us to do that. US data, you cannot do that, right?
So what these data tells us, after we collect all those data, what we found is a strong trend grows in China-- it's a national trend-- if price goes up, all the cities, all the provinces, the land price goes up. However, we also find there's a substantial heterogeneity. Heterogeneity, for those who are not econ students, means this city price movement are different from other cities. It's called heterogeneity.
So there's a substantial heterogeneity across the markets. For example, there's extremely strong real price growth in Beijing, I mentioned. Over the 10 years, there's over 1,000% price growth in Beijing. However, in Dalian, price actually is falling. So different cities, the price movement are quite different.
And then, also, we found the transaction volume has dropped in the recent years. But in US, if then we observe a transaction value drop, there's a strong indicator there will be a future price adjustment for, actually, down the road. But in China, because the market's still new, so we'll wait and see whether those transaction value drop has some indication or not.
Regarding the sample, we see only 35 major cities in China, but actually those major cities contribute about half of the total market shares.
For those who are not familiar with China, this is a map of China. The green part is the east coastal cities. And so Tier 1 city, it's very expensive, sort of like New York City, Boston, San Francisco, Los Angeles, Chicago.
And then second tier city, the blue one, see in the middle. And west in the third tier cities.
From that we can see real price since 2004, first quarter, has a huge increase from our calculation. This is sort of the index we calculate. It's a constant quality. We can see, it's starting from 2004, goes up all the way to 2007. That's right before the subprime crisis. And then it hit by the global recession, the price goes down.
But what happens, see, at the end of 2008 and beginning of 2009, Chinese government issued 4 trillion RMB stimulus package like the US government. Obama government also issued a $100 billion US stimulus package. But in US, it doesn't work, because banks get the money they did not want to lend out.
In China, looks like the stimulus package worked very efficiently. The price immediately shooting up until 2011. And by that time, government say, uh-oh, it's too much overprice. They worry about a bubble, so they have a lot of macro sort of cool-down policy regulations. So price goes down a little bit.
But that's the problem with Chinese policy and central government policy. They find, OK, the real estate development is the important factor to boost up the GDP. They don't want to see the slowdown in growth rate, so they relax the policy, and it goes up again.
And then however, our data is quite different. We saw a real price appreciation, and it's over about 370% since the 2004, about calculate compound annual growth rate of 14% each year. But according to statistic bureau a Ministry of Land and Resource of China, they suggested the annual gross compound rate is only 4%. So the government data and our using the transaction, real transaction data calculation are so different. That's also reflect the Wall Street Journal first mention, say, whose data we should trust, right?
I also mention, say, these heterogeneities. This is how we separated by our land market by three different regions. Coastal region, that's Shanghai, Beijing, Guangzhou, Shenzhen, Hangzhou, We see price growths are very fast. And then blue line is the middle region, west, like Xian, [INAUDIBLE], Chongqing, that's green lines.
We can see, even there's a common trend. The common trend is very similar like [INAUDIBLE] before hit by the financial crisis, and that stimulus package boost again, and then the macro prudential cooling down policies, and that boost again. However, we can see different region, actually, especially in the last two years, like West region, they start to drop, price has a drop.
So whenever the Wall Street Journal or New York Times or some Western professor say, OK, China has a bubble, that China will crash, we ask, say, which part of China? You are talking about a coastal city, Beijing, Shanghai? Or you are talking about Chongqing or Chengdu? Or you are talking about Xinying? China is big. So different city, the answer is different.
So that's what I earlier mentioned, say, there's a common trend. This is a first observation, looks like move together. Also, there's a heterogeneity. Different city, different region, the price movement are different.
This is another chart shows that Beijing, the scale is 1,000. The index scale is 1,000. However, Dalian, the maximum scale is 300. Even it goes down. So different city, if we go to detail, there are some quite different movement.
So what we need to know with all those data collected, I say there's a common trend. Whenever our central government say, OK, I issued a stimulus package, now price goes up. And then our cooling down price goes down. However, we also say they are heterogeneity.
But what we need to know, we actually, we argue, say, we need a better measurement of price and quantities. We have to understand the intersection of demand and supplies. We want to look at the market demand/supply fundamentals, whether the Chinese-- those fundamental can support those price goes down that fast, and whether there will be sustainable growth in the housing market. That's the concern what we want to explain.
So what we tried to do here is that's actually, for those of you who are not in the Econ class or are less familiar with the housing sectors, this is a good set of the measurement you may-- whenever if you go back home, your parents ask, you, say, whether we should invest in the housing market, you say, well, Professor Deng from Singapore give us a cup of measure. We can use this measure to find out whether it's worse or makes sense to invest in the housing market in whatever, Hangzhou, Xuzhou, [INAUDIBLE] or not.
When as we look at the annual new construction of relative to the market size. The other is we tried to look at the unsold inventory. If the market has too many units unsold, developers still try to put more units, then chances are, price were going down, because not so many people are going to buy that.
And then we also tried to look at the vacancy rate. And then there's a more technical for those of you who may be going to the major in finance, you look at the price to rent ratio, price to income ratio, or break down real appreciation expected from user cost and equation model. That's more technical.
There's a very famous professor whose name is Jim Poterba. He is the president of National Bureau of Economic Research. That's his famous article in 1985. So every PhD student study finance using this model. So we try to adapt this model.
So are the rough measure we try to provide. And if you use those measure as a ruler to say whether it's worse for me to buy a house, or rather I go to another town to buy a house.
So first is look at a aggregate price growth. We calculate a annual house price growth. And now we implied a compound growth rate, that Beijing average annual growth is over 20%, Shanghai, it's about 13% each year price growth, and then Chongqing is about 11%. So most of the price growth is more than 5% each year.
So this is a huge growth for any investment. That's partly explain why so many Chinese want to put their hardly saved money into the housing market, because they expect those growths will continue for the future years, right?
And then on average, if we average those cities, we can get to almost 12% gross each year.
And then we compare with-- because we have the land price growth, we have the house price growth. That's the beauty of China data. We say in US, there's a urban theorists argue, say, well, the housing bubble actually is a land bubble, right? And for housing, like any durable goods, like the refrigerators, like TV set, after first year, there's depreciation. The price will drop down.
Only housing is different. Housing, maybe you buy a house. Next year, housing price double. But the theorists argue, say, no, the value of the structure is the building on top of land, it's just like any durable goods, like a car, a refrigerator, TV set, all those things, price will go down. It's the land underneath the structure, the price value.
And varied. Why? There's a very easy explanation, say. If a piece of farmland growing potatoes, you can calculate, say, how the land value is, right, because you calculate how many annual gross production is. Now you calculate what the market value for the potato is. And then you can figure out new discount factor. You can calculate what this piece of land worth.
However, if you put the structure, put a building, put a housing on top of this piece of land, then just something happened. The price can go anywhere, right? So there's the magic of your connection to the housing, to the construction, to the building, to a piece of land. So that's the part make it difficult.
However, in the US or in other European country, you cannot separate the land price with the structure price. But only in China we see it, because we observe the transaction value of the land, and now we can separate the land price to the housing price, and then we can figure out.
So this is what we do. We calculate the housing residential land price, total growth rate. We included annual growth rate. In Beijing is annual growth rate is 27.5%.
And then we also calculate house price growth. In Beijing, annual house price growth rate is only 11.5%. So we say land price growth over the house price growth. This is the last column. Land price growth is much faster than house price growth. Like in Beijing, it's almost five times faster. In Shanghai, it's 3.4%.
So land price growth is much faster than house price growth. That suggest all the house price movement in Beijing, Shanghai, Guangzhou, Shenzhen, actually, it's driven by the land price.
And then that goes back to the previous argument. The local government, because they want to push up their KPI, they want to move up the GDP, and then they sell more land, push the land price going higher. However, the consequences, once the land price goes higher, the local house price will be pushed higher, because of these ratios.
So actually, what local government official did is good for their promotion, but local people will have to pay much, much higher house price for their promotion.
Now, we also look, for those who study finance, the volatility is always a concern. Volatility is risky. From this you can see the land market transaction is highly volatile. Even though you can, the pattern is similar, like the before crisis going up. And then hit by the crisis. And there's stimulus. And there is-- it's highly volatile.
And then this is the same for the housing transaction volume is highly volatile.
For those who are-- in reference, in the US housing market, Case-Shiller, they calculate it. Also, there's volatility in housing market. But that's every four or five years, there's sort of a mean reverting.
But in China, every six months, price goes up, goes down, goes up, goes down. That's for investment purpose. This is highly risky. That's why you need a high return for that. So because China, only the real estate is a good investment, but you also need to realize it's a highly risky investment.
And then next question, we try to establish some facts about the data. We create a constant quality measure of the land price, house price. And then we argue, say, whether it's the land pulling the price up, housing price up, all those factors. And as it turns out, more and more people trust our data more than the government statistic bureau data. That's fine.
But next question is what those data tells us, whether the Chinese price growth can be possibly explained by fundamentals. That's for those who study econ, and say, well, if the price going up, there's a fundamental support price going up.
There are some studies say there are superstar cities in the US. Namely, it's Boston, New York, San Francisco, and Chicago. Those cities, the prices always going up. However, fundamental support going up, that's fine. That means in the short-term, there's no crash landing. The price will drop down. However, we will check whether there's a fundamental, the price growths are supported by fundamental.
So first is look at the demand and supply. Housing start. How many supply put on the market? This is-- we saw starting from 2004 until the 2012, there's a huge increase in housing start. Because a lot of government official find a secret, and if they move the GDP up faster, they will get promoted. And typically, within three or four years, they need to get promoted to another place. How to make three or four years, get promoted, get the GDP fast? Sell more land, build more house, and then GDP, because there's a transaction going on, GDP going fast.
And if you put the money into the environmental clean air, it may take about 10 years or 15 years to work, but you already miss your chance, because you have the guys who are after that, because every three or four years, there's a turn, right? You have the third or fourth generations mayor to build up that GDP. So that's why we see a huge bump in housing starts. A lot of buildings going up.
And now we look at annual new supply. Try to see what the number of new supply compare with the current stock in the market, which means if the market, the new supply adding, there's a existing stock not sold out. 9% means, typically, the new stock adding to additional 9% to the market. So that will take much longer for the market to absorb.
So as before, we can see there's huge heterogeneities among different. For example, the red line in the first chart is a Tianjin. Tianjin, there's a lot more sort of the new stock adding to the market. It will make the supply stay even longer hanging over there. So the price will tend to be drop much faster.
But if you look at the city like Wuhan, Hangzhou, they are in the 3% or 4% of that. So they are quite different compared with different cities. And there we also compute those.
And so inventory by developers, we survey 119 major developers. We ask, say, what's the unsold inventory compared that each year's sales? That chart, if it's, say, here, it's 350 means the unsold inventory. If no new housing add to the market, it will take three and 1/2 year to clean up the inventory. That means the higher the number, the more vacancy likely, the oversupply will be.
Again, different cities. Even as the common trend is on the rise, however, different city are quite differently, especially for, like, this red is Dalian. Dalian is way oversupplied.
And then we did a quick calculation. We cumulated where there are unsold inventories. It turns out the inventory amount measured by dollar increase over the years. Especially by 2013, there are more and more inventories in the city built up. We normalized by the developers [INAUDIBLE] Still it's increased dramatically very fast.
And then we look at the demand and supply. We calculate, say, how many new household are formed during those years? And then how many new units are putting into the market? And then we try to use that to calculate demand and supply. And if it's oversupplied, we will see the market is more prone to a crash, right? The prices will be decline.
So we did a calculation. And surprisingly, because we read so many Wall Street Journals, New York Times, they say, OK, there's a ghost town. There's China is going to crash next week or next month.
But what we look here is we calculate, say, for the entire nation, we break down from 2001 to 2010, and the last five years, I know, so the last column is the past 15 years, right? We first calculate the supply. We calculate each city's and nation's total number of supply and where. Then we calculated the population growth, what each city sort of new, age getting to marriage age, and then will get a house, or there's a migration to the city, get a house. We calculated demand. And we estimated demand and supply.
For the nation, for the past 15 years, roughly balanced, right? Supply over demand is under 100%. It's not oversupply.
And now we look at different cities. Beijing is way undersupplied, right? Right For the past 15 years, supply only meet about 87% of demand, which means if you live in Beijing, you want to wait for a price drop. In the next few years, it's less likely. Because still there are another 13% of people has the cash in hand waiting to buy a house. The price may still go up.
Guangzhou, 92% of supply to meet the demand.
And then Hangzhou, we say the prices are way too high. It's only less than 80% of supply to meet the demand. Shanghai, 70% [INAUDIBLE]. So I always try to say, well, my wife and I, we need to buy some house to upgrade my in-laws. They are still living Shanghai. It looks like next few years, we still cannot afford to buy. It's too-- their price.
And then, of course, Shenzhen, where price is very high, but undersupply. However, there are certain cities, like I earlier mentioned, Chongqing, in the last five years, supplies have almost 300 times, three times of-- 300% of demand. So if there are 300 times of supply on the market, demand is only 100 times, the price war goes down. If no one buys, developer have to cut the price, right? Even over the past 15 years, as average supply over demand is 230%.
Chongqing growing very fast. Supplies are way above the demand. So those places, price tend to drop very fast, because of oversupply.
Now we summarize. At least about eight cities' supplies are sort of less than demand. However, there are another eight cities' supplies that over 50% of demand. But remaining sort of in between are the majority as the supplies outpace the demand. So depends on which cities. Most of the city, there's an oversupply going on. However, [INAUDIBLE] or Hangzhou, those coastal cities, supply is under demand. So it depends on which city you are talking about.
And then we come to the highly debated question about the vacancy rate. Even government cannot give us a straight face answer about whether there's vacancy or whether there's no vacancy.
There's a 60 Minutes sort of program interview. Mr. Wang Shi, who is the chairman of the Vanke-- the 60 Minutes and cast say, Mr. Wang, do you think China is a ghost town? Mr. Wang say, yes. But because his English is a bit slow, he say, yes, but. And the camera cut. Say, OK, Mr. Wang shi say, there's a ghost town. There's a vacancy.
But actually, even academic, they debate it. For example, my colleague, Professor Gan Li, at University of-- Southwestern University of Finance Economics, he ran this very famous household survey of 28 provinces. Each provinces, he survey about 1,000 household.
He sent a student go to the household, asked how many housing units you have? They say, maybe three or four. How many you are sort of stay in? How many you lease out? So which means he calculates it.
And if you say three housing units, you only need to stay in one unit, right? So there are two vacant. So that's his calculation. So by that calculation, he estimated a very high national-wide vacancy rate at 22%.
And then there's another investment corporation in China, International Capital Corporation, they estimated roughly about 18% national vacancy rate.
What we did is we were more careful microlevel demand/supply, and then we estimated. Well, actually, it still shows different cities has different results. Our estimation of the vacancy rate, it's, roughly, in 2009, it's at 5%. Much lower than Professor Gan Li's survey data.
Because what their problem is, their surveys say how many units they have. For example, my in-laws live in Shanghai. They have one resident units the company assigned to them. And then because of the housing reform, the company give them another small storage place. They kind of has two place. But second place cannot lease out for rental, right? And then if you count it as a one vacancy, then it's overcounted. So that's the survey issue.
So what we use is we use a microdata population growth. We calculate what the household growth, and how many units you put into the market, and we include a vacancy rate. It's much lower. It's 5%, so the Chinese government would love to see our data.
However, the bad news is we also estimated what's the vacancy rate movement trend. It looks like the vacancy rate is moving up. Around 2014, the vacancy rate has reached to about 8%. And then also it depends on which cities.
So there are a lot of disagreement about the vacancy rate. And then we also look at the price-rent ratio. Price-rent ratio, here we can see also there's high variations across different cities. For example, Beijing, the price-rent ratio can reach about 55. In Hangzhou here, also 55 at one point.
But roughly, what the right price-rent ratio is, when I taught MBA at USC, a group of developers, they push me, say, Professor, tell us what the right price-rent ratio we should get out of the market. I told them, say, based on my judgment, 16 years, the magic number, if go to 18, you need to consider. If go beyond 20, you definitely don't want to invest in the market.
The reason is the reverse after price-rent ratio in finance is called cap rate. So if it's a 52 or a 55 or a 60, the cap rate will be less than 1%. If you want to invest anything, you tell them, your return is less than 1%, no one will do that, right? If it's at 60, return is about roughly about 5% or 6%. So 5% or 6% is the margin I say it's a reasonable investment.
But look at Shanghai or Beijing or Hangzhou, it's a high price-to-rent ratio, which means, actually, those investment in the housing market, the Shanghai, Beijing, the return opportunity is not great. But why people want to do that? Well, that's the Professor Poterba's model.
We put into the user cost model, they calculate, say, Poterba told us, say, well, in order for Beijing to survive that high price-rent ratio, price keep going high, there's only one magic number. That magic number is the market expectation, everyone in Beijing has to believe the future growth, if it's more than 7.3%, the price will keep going up.
And then we ask, we check the data, we say, what's the actual annual per year growth in Beijing? Lot more than what Professor Poterba ask the [INAUDIBLE]. Actually, it's almost 20% a year.
So people in Beijing, don't worry, say. Well, Professor Poterba, famous finance professor say, if the expected market expectation is about at 7.3%, market will not crash. But actually, in the past 15 years, the real annual growth is at 20%, so we can still pull the money into the Beijing market. We never worry about.
This is only one story. We also ask, say, what happens if the expectation changed by one percentage in Beijing? Because the expectation is very vulnerable. If government announced some regulations, the market opinion will change. If the market opinion change from 7.3% or 6.3%, that's only a 1% adjustment in the expectation we put back into Professor Poterba's model, which implied price overnight will drop by one third, 30%. So that's a huge risk.
If market [INAUDIBLE] opinion fluctuated by 1%, the Beijing equilibrium price will drop by 30%. That's for investment jargon, there's a huge risk. So we say, well, the fundamentals still support those Beijing, Shanghai. However, it's also very risky. That's our argument.
The conclusion. So I hope my discussion today can allow you to bring home some news for measure, say whether my hometown is in Beijing or in Suzhou, in Jinzhou, or in Wuhai. My mother, my parents are consider to put all the money into the next housing investment. I can-- Professor Deng showed me some judgment about whether it's worth to do that or not.
In Beijing, Shanghai, Guangzhou, Shenzhen, or Hangzhou, fundamentals still support those prices are going very high. However, at the same time, we say the risk is also high. Because even one standard deviation adjustment in the market opinion, the price will jump overnight. This is highly risky.
However, we also want to bring your attention. You read a lot of Western media, Wall Street Journal, New York Times, and The Economist say market will crash, because they just buy the Western statistics. We make the argument, say, China's price start from very low base. And then we know, for the ratio, if the base is low, the ratio tend to be very high, right? It's [INAUDIBLE] be very high.
Also, China is facing a high growth, but highly volatile market. But we don't want to put any label whether there's a bubble or no bubble in this market or that market, this city or that city. The main reason is the time series is very short. Data measurement, we believe our data is better than the Chinese government data, but the data measurement issue is challenging.
The last thing we want to measure is beyond the rich pricing, I mean, very high pricing. The Chinese fundamental still seem to be support those high pricing. However, in many cities, the supply is overpaced the demand. Even in those Beijing, Shanghai, Guangzhou, Shenzhen, I say, the fundamental, say, supply is under the demand.
However, if there's external shock, for example, if there's government policy change or whatever, the growth rate change, the consumption behavior change. Even those city, we argue, say, where there looks like there's a fundamental support, the high price, they're still highly risky.
So that's why if you have extra money your parents want to put into the housing market, you need to think more carefully. Thank you. They have a list of extra references if you want to take a look.
We've received your request
You will be notified by email when the transcript and captions are available. The process may take up to 5 business days. Please contact email@example.com if you have any questions about this request.
Yongheng Deng, Provost Chair’s Professor and Professor of Real Estate and Finance at the National University of Singapore, proposes strategies for understanding the state of the Chinese housing market and its many risks and misconceptions. Recorded February 8, 2016 as part of the East Asia Program’s Cornell Contemporary China Initiative Lecture Series. Co-sponsored by the Cornell Institute for China Economic Research (CICER).