哈佛大學：中國房地產業或已觸頂（45頁）（英文版）.pdf

1、NBER WORKING PAPER SERIES PEAK CHINA HOUSING Kenneth S. Rogoff Yuanchen Yang Working Paper 27697 http:/www.nber.org/papers/w27697 NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge, MA 02138 August 2020 The views expressed herein are those of the authors and do not necessarily

2、reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications. 2020 by Kenneth S. Rogoff

3、and Yuanchen Yang. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including notice, is given to the source. Peak China Housing Kenneth S. Rogoff and Yuanchen Yang NBER Working Paper No. 27697 August 2020

4、 JEL No. F39,G01,R3 ABSTRACT Chinas real estate has been a key engine of its sustained economic expansion. This paper argues, however, that even before the Covid-19 shock, a decades-long housing boom had given rise to severe price misalignments and regional supply-demand mismatches, making an adjust

5、ment both necessary and inevitable. We make use of newly available and updated data sources to analyze supply-demand conditions in the fast-moving Chinese economy. The imbalances are then compared to benchmarks from other economies. We conclude that the sector is quite vulnerable to a sustained aggr

6、egate growth shock, such as Covid-19 might pose. In our baseline calibration, using input-output tables and taking account of the very large footprint of housing construction and real-estate related sectors, the adjustment to a decline in housing activity can easily trim a cumulative 5-10 percent fr

7、om the level of output (over a period of years). Kenneth S. Rogoff Thomas D Cabot Professor of Public Policy Economics Department Harvard University Littauer Center 216 Cambridge, MA 02138-3001 and NBER krogoffharvard.edu Yuanchen Yang PBC School of Finance Tsinghua University 43 Chengfu Road, Haidi

8、an District Beijing 100083 China 2 I. Introduction It has long been argued that despite Chinas truly epic real estate boom over the past three decades (as documented in a number of previous papers),3 the central governments ability to tightly monitor the financial system, and to resolve insolvencies

9、 quickly in the event of a crisis, makes it relatively immune to conventional housing-related financial crises. At the same time, however, this view has been largely untested thanks to the countrys extraordinary growth performance, with expectations of further fast growth underpinning ever higher ho

10、using prices. How much of a vulnerability might the real estate sector prove in the wake of the Covid-19 shock? This papers contribution is threefold. First, we make use of newly available sources, to extend and significantly update earlier work on the size and scale of Chinas housing market. We esp

11、ecially take advantage of the digitization of Chinas statistics, which has helped provide both more extensive and more accurate data. Fang et al. (2015) document housing price growth in major Chinese cities from 2003 to 2013, while Glaeser et al. (2017) track changes in Chinas construction square fo

12、otage, household vacancy rates, etc. from 2002 to 2012. We find that significant changes, even before the Covid-19 pandemic, have taken place since. During 2013-2018, for example, real estate investment increased by 30% and household leverage ratio rose from 33% to 60%. Another important factor is t

13、hat China underwent a major change of leadership in 2013, and that has heavily influenced the countrys real estate policies. Thus, using data (some of which runs through 2020) gives a significantly more accurate picture of Chinas rapidly-evolving housing market. 3 See, for example, Fang et al. (2015

14、), Chivakul et al. (2015), Glaeser et al. (2017), and Koss and Shi (2018). 3 Second, we provide a quantitative characterization of the upstream and downstream contribution to GDP of Chinas real estate sector by exploiting newer input-output tables. Our updated measure, based on cross-industry correl

15、ations derived from Chinas most recent input-output matrix, and taking into account higher-order effects (in addition to first-order) is 29%, even higher than in earlier studies.4 Real estate activity involves both the production of property and the provision of services related to property. The for

16、mer refers to building, renovating, repairing and maintaining of property, while the latter includes buying, selling, renting and managing of property.5 In addition, real estate is closely linked to an array of industries through the supply chain. To complete a real estate project, various inputs ar

17、e needed, including intermediate inputs, such as steel and concrete, from the manufacturing sector, labor input from the construction sector, capital input from the banking sector, etc. Hence to estimate the sum effect of real estate on the economy, we consider not only the output of real estate sec

18、tor alone, but also the output share of closely related industries. Using the input-output table, we find that a 20% fall in real estate activity could lead to a 5-10% fall in GDP, even without amplification from a banking crisis, or accounting for the importance of real estate as collateral. Third,

19、 we present a substantial body of evidence from the last 20 years using a variety of Chinese and international datasets that strongly suggests the existence of a housing bubble in China. Extant literature has largely concentrated on China alone (Fang et al., 2015; Ding et 4 Previous studies using na

20、rrower measures of housings impact typically put the size of the sector around 16% (e.g., Glaeser et al (2017), Liu and Xiong (2018). Our measure is closest in spirit to Cook et. al. (2018) who use an earlier (2012) input output table and arrive at a 22% contribution of housing. See the Technical Ap

21、pendix. 5 According to the International Standard Industrial Classification (the SIC codes), real estate is classified as a tertiary industry, encompassing only the provision of property-related services, which is merely part of housing activity. 4 al., 2017), with a few papers drawing comparison wi

22、th the United States (e.g. Glaeser et al., 2017). We take a cross-country perspective, starting with classic housing booms from advanced economies while zooming in on Chinas current boom, to enable comparison on a global scale. Despite the repeated argument that China is different, we note that it s

23、hares striking similarities with other boom episodes in the run-up of housing prices, the scale of the construction sector, the debt accumulation etc. Indeed, given the severity of many economic indicators, Chinas decades-long housing boom shows many signs of having hit a potentially precarious peak

24、.6 The structure of the paper proceeds as follows. Section II briefly reviews the most closely-related literature. Section III presents recent developments of Chinas housing market up to the Covid-19 pandemic (though we include some very recent 2020 data just before the conclusions.) Section IV anal

25、yzes housing supply-demand dynamics. Section V examines real estates linkage to the economy and its resilience to the Covid-19 stress. Section VI concludes. Appendices provide data description and technical details. II. The literature The most closely-related studies to ours include earlier papers s

26、uch as Fang et al. (2015), Chivakul et al. (2015), Glaeser et al. (2017), and Koss and Shi (2018). Using data ending in 2013, 2014, 2012 and 2016 respectively, all of these papers explore the unique features 6 A considerable body of research has shown that housing sector is closely linked to busines

27、s cycles (Leamer, 2007; 2015), Mian, Sufi and Verner (2017) and to post World War II financial crises (Reinhart and Rogoff, 2009; Jorda, Schularick and Taylor, 2016). 5 characterizing Chinas housing market, and, like ours, analyze the possibility and macroeconomic severity of potential bubble bursts

28、. Employing detailed information on the profile of mortgage borrowers, Fang et al. (2015) find that housing price appreciation is largely matched by household income growth. The participation by bottom-income mortgage borrowers in the housing market remains steady despite onerous financial burdens,

29、and their mortgage loans are protected by down payments usually in excess of 35%. Therefore, an imminent financial crisis triggered by the housing market is unlikely to happen in China. Chivakul et al. (2015) identify Chinas housing oversupply based on various datasets, and consider a number of scen

30、arios to restore supply-demand balance. The paper concludes that the process of adjustment implies negative growth impacts, with the adverse effect being stronger in smaller cities, as well as cities in the northeast region. Glaeser et al. (2017) examine a variety of factors that help sustain Chinas

31、 soaring housing prices, highlighting, among other factors, the capacity and determination of the Chinese government to regulate housing prices. The authors also warn about the potentially significant social costs associated with the overemphasis on housing price stability, and propose the alternati

32、ve solution of accommodating high levels of housing construction and supply. Koss and Shi (2018) discuss the history of Chinas property market and then assess its resemblance to classic housing bubbles. In the event of a price drop and bubble burst, the authors propose a battery of near-term policy

33、prescriptions, ranging from the tightening of 6 mortgage requirements to the improvement of legal infrastructure, to help China withstand the impact. . III. Recent Developments in the Rapidly Evolving Landscape 1. Chinas Real Estate Boom Chinas real estate market started to form in the year 1988, wh

34、en major urban housing reforms intended to commercialize housing according to the demands of a socialist planned economy were initiated. Since then, real estate has quickly emerged to be a “pillar” industry and housing prices have taken off (Figure 1). Figure 1. Residential Housing Prices and Annual

35、 Growth Rates This figure presents average residential housing price and year-on-year percentage change at the national level from 1991 to 2018. -10% -5% 0% 5% 10% 15% 20% 25% 30% 35% 40% 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 0

36、7 08 09 10 11 12 13 14 15 16 17 18 National Average Housing Price (yuan/sq.m)Year-on-year Percentage Change 7 There are three important growth episodes in Chinas modern real estate history. The first one originated in 1989 from Hainan province, the nations largest and southernmost island featuring c

37、ultural tourism. Within three years time, housing prices went from just 300 yuan per square meter to around 7,500 at the peak. When interest rates began rising in 1993, housing prices plummeted to 1,000 yuan per square meter and remained there until the early 2000s. The second growth episode came on

38、 the heels of the global economic crisis in 2008. Banks set interest rates barely above inflation and easy money boosted investment demand for housing. Home prices rose rapidly from 2009 until significant policy tightening starting in the second half of 2010. After a moderate softening in 2014, the

39、most recent growth spurt began in early 2015, driven by a combination of loose monetary policy and expansionary fiscal policy. Since 2002, housing prices in Chinas tier one cities have risen more than six-fold compared to the 80% overall national increase in US housing prices between 2000 and 2005,7

40、 or the increase in Ireland of 100% and Spain of 230% in their respective housing booms.8 With the crash of Japans real estate market in the 1990s and the bursting of the U.S. housing bubble during 2006-2007 as precedents, Chinese authorities have long been wary of the risks associated with overheat

41、ing housing prices and, at various times, have taken measures to clamp down on speculation. Nevertheless, housing prices in China have risen at unprecedented rate (Figure 2). Until Covid-19, whenever the Chinese government looked to 7 Source: S 2) clothing; 3) residence; 4) household facilities, art

42、icles, and services; 5) transport and communications; 6) education, cultural and recreation; 7) health care and medical services; 8) miscellaneous goods and services. Housing related consumption here refers to household expenditure on residence. If we also include expenditure on household facilities

43、, articles, and services, the figure rises to 30%. 0 5 10 15 20 25 30 35 40 45 50 Beijing Shanghai Hong Kong Shenzhen Singapore London Tel Aviv Guangzhou Paris Vancouver Munich Barcelona Tokyo New York San Francisco 11 Chinas economic rise largely relies on an investment-driven growth model, and the

44、 ratio of total fixed assets investment to GDP has reached 70-80% in recent years, of which a significant proportion is in the real estate sector. In 2018, real estate investment composed 13% of Chinas GDP, while historically in the United State, this figure has been about 5% (Figure 5). Figure 5. C

45、hina and U.S. Real Estate Investment This figure shows annual real estate investment in China and the United States from 2002 to 2018. Real estate investment is placed on the left axis, and the ratio of real estate investment to GDP is placed on the right axis. Source: U.S. Department of Commerce Bu

46、reau of Economic Analysis Real estate is naturally related to the construction industry in that developers basically make profits by managing the construction process and turning raw land into salable property. In 2016, real estate and construction industries combined accounted for around 29% 0% 5%

47、10% 15% 20% 0 500 1,000 1,500 2,000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 China Real Estate Investment (billion $)U.S. Real Estate Investment (billion $) China Real Estate Investment/GDP - RightU.S. Real Estate Investment/GDP - Right 12 of Chinas GDP,13

48、 comparable only by pre-crisis Spain and Ireland. Real estate and construction are also crucial for job creation, making up around 20% of urban non-private employment (Figure 6).14 Figure 6. Contribution of Real Estate and Construction to Urban Employment This figure shows the contribution of real e

49、state and construction sectors to urban employment. Total employment in real estate and construction sectors is placed on the left axis, and the ratio of real estate and construction employment to total urban employment is placed on the right axis. Although our later analysis of an impact of real estate activity decline is entirely on the real side and does not build in amplification from a financial crisis, o