Chinese Housing Markets: What We Know and What We Need to Know

Joe Gyourko

Wharton School, University of Pennsylvania Paper presented at the International Symposium on Housing and Financial Stability in China. Hosted by the Chinese University of Hong Kong, Shenzhen Shenzhen, China─December 18-19, 2015

The views expressed in this paper are those of the author(s) only, and the presence of them, or of links to them, on the IMF website does not imply that the IMF, its Executive Board, or its management endorses or shares the views expressed in the paper.

Chinese Housing Markets:What We Know and What We Need to Know

Yongheng Deng, Joe Gyourko & Jing WuHousing and Financial Stability in China

Shenzhen ChinaShenzhen, ChinaDecember 18, 2015

1

Documentation—Prices and QuantitiesQ

• What Do These Data Show?– Strong trend price growth in aggregate in land and house values

• Economically large variation about that trend

– Substantial heterogeneity across markets• Extremely strong real price growth in Beijing, but real land values have fallen 

substantially the past two years in Dalian

– Transactions Volume• Has fallen in recent quarters both in terms of land parcel sales and in the amount of 

new housing soldnew housing sold

2

Documentation:  A Cl L k t L d P i A M j M k tA Closer Look at Land Prices Across Major Markets

• Chinese Residential Land Price Indexes (CRLPI)( )– Wharton‐Tsinghua‐NUS collaboration (Gyourko, Wu, Deng)– 35 major markets, not just east coast or East region cities

• See next slideSee next slide

– Transactions‐based, not appraisal‐based– Full samples of land sold by local governments to private residential 

developersdevelopers– Constant quality price indexes created

3

城市覆盖范围城市覆盖范围 City CoverageCity Coverage

•目前CRLPI指数覆盖全国35个大中城市国35个大中城市。Currently CRLPI covers 35 major cities in mainland Chinamainland China.

•这些城市占据了全国新 75%•这些城市占据了全国新建商品住房市场约50%的市场份额。These cities contribute

60%

65%

70%

75%

Market Share by Floor Area

Market Share by Total Value

These cities contribute to about 50% of the new home market in the

h l t40%

45%

50%

55%

whole country.30%

35%

2004 2005 2006 2007 2008 2009 2010 2011 2012 20134

Figure 1: Chinese National Real Land Price Index35 Markets, Constant Quality Series

(Quarterly: 2004q1 – 2015q3)(Q y q q )

500

400

300

200

100

2004q3 2005q3 2006q3 2007q3 2008q3 2009q3 2010q3 2011q3 2012q3 2013q3 2014q3 2015q3

5

Figure 2: Number of Parcels Sold35 Markets (Quarterly: 2004q1 – 2015q3)

800

600

400

200

0

2003q3 2004q3 2005q3 2006q3 2007q3 2008q3 2009q3 2010q3 2011q3 2012q3 2013q3 2014q3 2015q3

6

Chinese Housing Unit (Really Space) Growth2007 20142007‐2014

Figure 6: Year-Over-Year Growth in Floor Space Sold,

Newly-Built Housing Units, 2007(1)-2014(4)Newly Built Housing Units, 2007(1) 2014(4)

60%

80%

20%

40%

-20%

0%

-40%

2007

Q1

2007

Q2

2007

Q3

2007

Q4

2008

Q1

2008

Q2

2008

Q3

2008

Q4

2009

Q1

2009

Q2

2009

Q3

2009

Q4

2010

Q1

2010

Q2

2010

Q3

2010

Q4

2011

Q1

2011

Q2

2011

Q3

2011

Q4

2012

Q1

2012

Q2

2012

Q3

2012

Q4

2013

Q1

2013

Q2

2013

Q3

2013

Q4

2014

Q1

2014

Q2

2014

Q3

2014

Q4

National Level 35 Major Cities

7

Figure 3: Chinese Regional Real Land Price IndexEast, Middle and West Regions, Constant Quality Series

(Semi‐annually: 2004h1 – 2015h1)(Semi annually: 2004h1  2015h1)

5.00

4.00

50

3.00

1.00

2.00

2004h1 2005h1 2006h1 2007h1 2008h1 2009h1 2010h1 2011h1 2012h1 2013h1 2014h1 2015h1

East MiddleWest

8

1000

11002004=100

Beijing

700

800

900

1000

400

500

600

100

200

300

0

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

9

300 2004=100Dalian

200

250

100

150

0

50

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 20142004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

10

700 2004=100Chongqing

500

600

200

300

400

0

100

200

0

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

11

What Do We Need to Know?

• Better measures of price and quantity are not enoughp q y g– Cannot tell all that much just by looking at P and Q

• Intersection of supply and demand

• What do local market supply and demand fundamentals look like in Chinese housing markets?

12

Metrics on Supply‐Demand (Im)Balancespp y ( )

• Annual new construction relative to market size• Unsold inventory relative to sales volume in market• Vacancy rates in nine provinces• Price‐to‐rent ratios• Price‐to‐income ratios• Breakeven real appreciation expectations from Poterba user 

cost equation

13

Annual New Supply As Share of 2010 Stock

8.00%

9.00%

10.00%

8.00%

9.00%

10.00%

3.00%

4.00%

5.00%

6.00%

7.00%

3.00%

4.00%

5.00%

6.00%

7.00%

0.00%

1.00%

2.00%

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Beijing Tianjin Shanghai Chongqing

0.00%

1.00%

2.00%

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Changsha Guangzhou Chengdu Xi'an

6 00%

7.00%

8.00%

9.00%

10.00%

1 00%

2.00%

3.00%

4.00%

5.00%

6.00%

14

0.00%

1.00%

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Nanjing Dalian Hangzhou Wuhan

Unsold Inventory Held by Developers As Share of T ti V l i M k tTransactions Volume in Market

300%

350%

300%

350%

150%

200%

250%

150%

200%

250%

0%

50%

100%

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

B iji Ti ji Sh h i Ch i

0%

50%

100%

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Changsha Guangzhou Chengdu Xi'anBeijing Tianjin Shanghai Chongqing Changsha Guangzhou Chengdu Xi'an

250%

300%

350%

50%

100%

150%

200%

15

0%2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Nanjing Dalian Hangzhou Wuhan

Aggregate Space Delivered35 M j M k t35 Major Markets

million sq.m.

450

500

300

350

400

200

250

300

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

16

Longer‐Run Supply/Demand Trends2001 20142001‐2014

• Supply at least 10% below our projected demand– Beijing (87%), Hangzhou (79%), Haikou (79%), Jinan (86%), Ningbo (85%), 

Shanghai (70%), Shenzhen (73%) 

• Supply at least 30% above our projected demandpp y p j– Chengdu (230%), Chongqing (193%), Guiyang (162%), Harbin (160%), 

Hohhot (178%), Lanzhou (143%), Qingdao (144%), Shenyang (154%), Taiyuan (148%), Tianjin (132%), Xian (130%), Xining (32%), Yinchuan (193%), Zhengzhou (191%)(193%), Zhengzhou (191%) 

• Other Major Markets Somewhere in Between, with Most Looking Modestly Oversupplied

h h ( %) h h ( 9%) li ( %) h ( 9%)– Changchun (111%), Changsha (119%), Dalian (115%), Fuzhou (119%), Guangzhou (93%), Hefei (125%), Kunming (120%), Nanchang (102%), Nanjing (104%), Nanning (115%), Shijiazhuang (113%), Urumqi (94%), Xiamen (115%), Wuhan (129%) 

17

18

19

20

Quarterly Price‐to‐Rent Ratios

45

50

55

60

45

50

55

60

25

30

35

40

45

25

30

35

40

45

20

2009

Q3

2009

Q4

2010

Q1

2010

Q2

2010

Q3

2010

Q4

2011

Q1

2011

Q2

2011

Q3

2011

Q4

2012

Q1

2012

Q2

2012

Q3

2012

Q4

2013

Q1

2013

Q2

2013

Q3

2013

Q4

2014

Q1

2014

Q2

2014

Q3

2014

Q4

Beijing Tianjin Shanghai Chongqing

20

2009

Q3

2009

Q4

2010

Q1

2010

Q2

2010

Q3

2010

Q4

2011

Q1

2011

Q2

2011

Q3

2011

Q4

2012

Q1

2012

Q2

2012

Q3

2012

Q4

2013

Q1

2013

Q2

2013

Q3

2013

Q4

2014

Q1

2014

Q2

2014

Q3

2014

Q4

Changsha Guangzhou Chengdu Xian

45

50

55

60

20

25

30

35

40

21

20

2009

Q3

2009

Q4

2010

Q1

2010

Q2

2010

Q3

2010

Q4

2011

Q1

2011

Q2

2011

Q3

2011

Q4

2012

Q1

2012

Q2

2012

Q3

2012

Q4

2013

Q1

2013

Q2

2013

Q3

2013

Q4

2014

Q1

2014

Q2

2014

Q3

2014

Q4

Dalian Nanjing Hangzhou Wuhan

Annual Price‐to‐Income Ratios

16

20

16

20

4

8

12

4

8

12

0

4

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Beijing Tianjin Shanghai Chongqing

0

4

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Changsha Guangzhou Chengdu Xian

12

16

20

4

8

22

0 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Dalian Nanjing Hangzhou Wuhan

Conclusions

• Need more effort on measurement of prices in particular– China needs a S&P/Case‐Shiller price index– Ability to gauge land market is a real advantage

• Keep context in mind when examining prices and quantitiesChi i t t d f l b– Chinese prices started from very low base

– China is a high growth and high volatility market

• Do not see much value in trying to put the label ‘bubble’ on• Do not see much value in trying to put the label  bubble  on these markets– Very short time series, with imperfect data– Still, no doubt Chinese housing markets are very riskyStill, no doubt Chinese housing markets are very risky

• ‘Priced to perfection’ in the following sense:  even in fundamentally strong Tier 1 cities, relatively small changes in expectations, absent countervailing rent increases, will lead to large negative changes in price levels per Poterba’s user cost framework

23

Conclusions

• Beyond ‘rich’ pricing, supply appears to have outpaced y p g, pp y pp pdemand over the last decade, not just the last year, in various markets

Primarily but not exclusively in the interior of the country– Primarily, but not exclusively, in the interior of the country – Any negative demand shock will occur in an environment of weak 

fundamentals in these places—this combination always leads to large price declines in any durable goods market and housing is a durableprice declines in any durable goods market, and housing is a durable good

24