Skip to main content

Housing boom and non-housing consumption: evidence from urban households in China


We study the response of non-housing consumption to housing price movements in urban China, which has been witnessing a real estate boom ever since 2003. Using Urban Household Survey data, we estimate an elasticity of consumption with respect to housing price of 0.06–0.07 for homeowners. Moreover, we find that the average marginal propensity to consume out of housing wealth is 0.025–0.03. We employ a novel instrumental variable associated with higher-education expansion to ensure that these estimates are causal effects. As for renters, we show that their consumption response to housing shocks is insignificant. We further reveal that the marginal propensity to consume is larger for homeowners who are more credit constrained.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6


  1. 1.

    The equivalence is built on the fact that the average share of housing wealth in total net worth was around 56% in 2002 for homeowners (multiplying 0.12–0.13 by 0.56 produces 0.06–0.07).

  2. 2.

    Aladangady (2017) adds to Mian et al. (2013) with a stable relationship between consumption and housing price starting from 1986, covering both the U.S. housing boom and bust periods. However, he does not compare the housing–consumption nexus across housing boom and bust periods.

  3. 3.

    See the online appendix for more details on the linkage between housing markets and land supply, as well as how China managed the massive higher-education expansion.

  4. 4.

    More details on the UHS are presented in the online appendix.

  5. 5.

    The comparison is implemented for both the starting and ending years to further demonstrate that our sample can represent China’s broad economic changes during 2002–2009.

  6. 6.

    More information and discussion on variable construction can be found in the online appendix.

  7. 7.

    In China, communities constitute counties, counties constitute prefectures, prefectures constitute provinces, and provinces constitute the nation. We choose county as the baseline to reflect two concerns, more perceived randomization of the sample and less potential endogenous influence on the construction of our instrumental variable that has a province-level component.

  8. 8.

    We also have detailed information on home characteristics such as home age, size, and architectural styles.

  9. 9.

    We present more detailed discussion on the construction of household wealth in the online appendix.

  10. 10.

    We also use the same criterion for prefectures and obtain 59 observations for the prefecture-level sample that is employed in the robustness checks.

  11. 11.

    We discuss the legitimacy of housing leverage ratio as a proxy for credit constraints in the online appendix.

  12. 12.

    We present summary statistics for key county-level variables. Household-level summary statistics are available in the online appendix.

  13. 13.

    Even the true value is supposed to be lower than the U.S. counterpart because the Chinese households save much more and only borrow externally when outlays surpass savings. Young adults in China are also more likely to get financial supports from parents or extended relatives, probably due to differences in culture and financial development.

  14. 14.

    More discussion on the household-level regression specification can be found in the online appendix.

  15. 15.

    This includes shocks to trend growth in major economic variables, as emphasized by Aguiar and Gopinath (2007) for large emerging economies like China.

  16. 16.

    Our instrument is virtually a variant of the typical Bartik instrument as recently discussed by Goldsmith-Pinkham et al. (2020). To demonstrate robustness, we also update the original instrument with the product of the following two components: the share of the number of colleges located in a county (among all colleges in the province) in 1998 and the expansion of college enrollment during 1998–2005 at the province level. The updated instrument is a typical Bartik instrument. It produces comparable IV estimates as our original instrument. The results are available upon request.

  17. 17.

    We also provide a detailed timeline of the higher-education expansion and endogenous housing price movements in the online appendix to better display the 4-year gap between them.

  18. 18.

    College here means a broad set of post-secondary educational institutes, including universities, academies, colleges, seminaries, conservatories, and institutes of technology. It also contains short-term college-equivalent institutions, such as vocational schools, trade schools, and other career colleges that award academic degrees or professional certifications. Moreover, we are cautious about exogeneity when constructing the college enrollment expansion shock. As such, we use the initial number of colleges in 1998, prior to the higher-education expansion starting from 1999.

  19. 19.

    This is supported with the cultural phenomenon that Chinese parents prefer college graduates to work in a location that is close to home (but no need to be in the same county). Hence, if a college graduate comes from a faraway place, she is more likely to migrate back to hometown after the graduation to align with parents’ preferences.

  20. 20.

    Additionally, we need an underlying assumption that housing supply is relatively inelastic in the short run. This is probably true in China because real estate developers cannot freely increase housing supply in anticipation of a surge in housing demand. To build more houses, they first need to obtain more use rights of land from local governments, which typically takes a long time due to institutional hurdles such as red tapes.

  21. 21.

    We provide details about the construction of the permanent income shocks in the online appendix.

  22. 22.

    Aggregate data from the NBS and Ministry of Land and Resources of the People’s Republic of China show that land revenues accounted for 40.4% of total fiscal revenues collected by local governments during 1998-2015. Between 2002 and 2009, it even reached 45.1%. Moreover, it went beyond 50% for some cities lacking other alternative sources of revenues such as the value added tax.

  23. 23.

    Notice that the asymptotic variance covariance matrix for estimators barely hold in small samples, we instead utilize bootstrapped SE in county-level regressions.

  24. 24.

    For the Chinese currency, 1 yuan equals to 100 fen.

  25. 25.

    We also consider household-level regressions using the sample of households from the 59 prefectures that exist throughout our sample years. The results are robust.

  26. 26.

    95% confidence interval is specified for calculating the upper bounds. To save space, we do not present the IIV estimation results. However, they are available upon request.

  27. 27.

    We also examine the heterogeneous impacts of housing shocks on different categories of consumption, which are differentiated by durability and income elasticity. The results are reported in the online appendix.

  28. 28.

    Since we have a relatively small sample size at the county level, we employ prefecture-level income and wealth distribution to implement the division. It generates a meaningful number of households for each income and wealth bin.

  29. 29.

    Correlation patterns and first-stage regression results for the prefecture-level sample are available in the online appendix. They tell consistent stories as those from our county-level sample.

  30. 30.

    One concern might be that why we do not treat the estimates with error-in-variable correction as baseline results. The reason is that averaging across households within cities tends to reduce measurement error of variables at the household level while it introduces a new type of measurement error at the city level by construction when we employ sample averages to replace population means. Since we are not sure which measurement error is more severe, we treat the error-in-variable estimation as a robustness check rather than the baseline result.

  31. 31.

    More detailed discussions on the comparison of our housing prices and the hedonic ones are presented in the online appendix.


  1. Abel AB (1990) Asset prices under habit formation and catching up with the joneses. Am Econ Rev Pap Proc 80(2):38–42

    Google Scholar 

  2. Aguiar M, Gopinath G (2007) Emerging market business cycles: the cycle is the trend. J Polit Econ 115(1):69–102

    Article  Google Scholar 

  3. Aladangady A (2017) Housing wealth and consumption: evidence from geographically linked microdata. Am Econ Rev 107(11):3415–3446

    Article  Google Scholar 

  4. Arellano M, Bond S (1991) Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Rev Econ Stud 58(2):277–297

    Article  Google Scholar 

  5. Berkowitz D, Ma H, Nishioka S (2017) Recasting the iron rice bowl: the reform of China’s state owned enterprises. Rev Econ Stat 99(4):735–747

    Article  Google Scholar 

  6. Brandt L, Van Biesebroeck J, Zhang Y (2012) Creative accounting or creative destruction? Firm-level productivity growth in Chinese manufacturing. J Dev Econ 97(2):339–351

    Article  Google Scholar 

  7. Carroll Christopher D, Wendy E, Dunn. (1997). “Unemployment Expectations, Jumping (S, s) Triggers, and Household Balance Sheets”. Chapter 4 in NBER Macroeconomics Annual, (2012) volume 12. In: Ben S (ed) Bernanke and Julio J. MIT Press, Rotemberg

  8. Carroll CD, Kimball MS (1996) On the concavity of the consumption function. Econometrica 64(4):981–992

    Article  Google Scholar 

  9. Che Y, Zheng L (2018) Human capital, technology adoption and firm performance: impacts of China’s higher education expansion in the late 1990s. Econ J 128(614):2282–2320

    Article  Google Scholar 

  10. Cochrane J (1991) A simple test of consumption insurance. J Polit Econ 99(5):957–976

    Article  Google Scholar 

  11. Deaton A (1985) Panel data from time series of cross-sections. J Econ 30(1–2):109–126

    Article  Google Scholar 

  12. DeFusco AA (2018) Homeowner borrowing and housing collateral: new evidence from expiring price controls. J Finance 73(2):523–573

    Article  Google Scholar 

  13. Disney R, Gathergood J, Henley A (2010) House price shocks, negative equity, and household consumption in the United Kingdom. J Eur Econ Assoc 8(6):1179–1207

    Google Scholar 

  14. Fang, Hanming, Gu, Quanlin, Xiong, Wei, Zhou, Li-An (2015) “Demystifying the Chinese Housing Boom.” Chapter 2 in NBER Macroeconomics Annual 2015, volume 30, edited by Martin Eichenbaum and Jonathan Parker. University of Chicago Press

  15. Gan J (2010) Housing wealth and consumption growth: evidence from a large panel of households. Rev Financ Stud 23(6):2229–2267

    Article  Google Scholar 

  16. Giroud X, Mueller HM (2017) Firm leverage, consumer demand, and employment losses during the great recession. Quart J Econ 132(1):271–316

    Article  Google Scholar 

  17. Goldsmith-Pinkham P, Sorkin I, Swift H (2020) Bartik instruments: what, when, why, and how. Am Econ Rev 110(8):2586–2624

    Article  Google Scholar 

  18. Groen JA (2004) The effect of college location on migration of college-educated labor. J Economet 121(1–2):125–142

    Article  Google Scholar 

  19. Han J, Liu R, Zhang J (2012) Globalization and wage inequality: evidence from urban China. J Int Econ 87(2):288–297

    Article  Google Scholar 

  20. He Z, Ye J, Shi X (2020) Housing wealth and household consumption in urban China. Urban Stud 57(8):1714–1732

    Article  Google Scholar 

  21. Hsieh C-T, Klenow PJ (2009) Misallocation and manufacturing TFP in China and India. Q J Econ 124(4):1403–1448

    Article  Google Scholar 

  22. Kaplan G, Mitman K, Violante GL (2020) Non-durable consumption and housing net worth in the great recession: evidence from easily accessible data. J Public Econ 189:104176

    Article  Google Scholar 

  23. Khandelwal AK, Schott PK, Wei S-J (2013) Trade liberalization and embedded institutional reform: evidence from Chinese exporters. Am Econ Rev 103(6):2169–2195

    Article  Google Scholar 

  24. Knight J, Deng Q, Li S (2017) China’s higher education expansion: the labour market consequences of a supply shock. China Econ Rev 43:127–141

    Article  Google Scholar 

  25. Li H, Zhou L-A (2005) Political turnover and economic performance: the incentive role of personnel control in China. J Public Econ 89(9–10):1743–1762

    Article  Google Scholar 

  26. Li J, Liu S, Yifan W (2020) Identifying knowledge spillovers from universities: quasi-experimental evidence from urban China. mimeo, New York

    Google Scholar 

  27. Manova K, Wei S-J, Zhang Z (2015) Firm exports and multinational activity under credit constraints. Rev Econ Stat 97(3):574–588

    Article  Google Scholar 

  28. Mian A, Rao K, Sufi A (2013) Household Balance Sheets, Consumption, and the Economic Slump. Quart J Econ 128(4):1687–1726

    Article  Google Scholar 

  29. Mishkin FS, Gordon RJ, Hymans SH (1977) What depressed the consumer? The household balance sheet and the 1973–75 recession. Brook Pap Econ Activ 1977(1):123–174

    Article  Google Scholar 

  30. Nevo A, Rosen AM (2012) Identification with imperfect instruments. Rev Econ Stat 94(3):659–671

    Article  Google Scholar 

  31. Saiz A (2010) The geographic determinants of housing supply. Q J Econ 125(3):1253–1296

    Article  Google Scholar 

  32. Waxman A, Liang Y, Li S, Barwick PJ, Zhao M (2020) Tightening belts to buy a home: consumption responses to rising housing prices in urban China. J Urban Econ 115:103190

    Article  Google Scholar 

  33. Winters JV (2011) Why are smart cities growing? Who moves and who stays. J Reg Sci 51(2):253–270

    Article  Google Scholar 

Download references


The author received no specific funding for this work.

Author information



Corresponding author

Correspondence to Dong Cheng.

Ethics declarations

Conflict of interest

The author declares that have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants or animals performed by the author.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

I thank two anonymous referees for insightful comments that greatly improve the quality of the paper. I am indebted to my advisor Mario J. Crucini for his invaluable encouragement, guidance, and support. I am grateful to Gregory W. Huffman, Atsushi Inoue, Hyunseung Oh, David C. Parsley, and Joel Rodrigue for providing valuable comments and feedback. I would like to thank participants at the Midwest Macro Fall Meetings 2018, Midwest Macro Fall Meetings 2017, and seminar participants at the Vanderbilt University, Union College, College of William and Mary, and Eastern Connecticut State University for their helpful comments and discussions. The author also benefits from insightful conversations with Jess Benhabib, Anton Braun, Morris Davis, Grey Gordon, Nick Lei Guo, Federico Gutierrez, Narayana Kocherlakota, John Leahy, Yue Li, Mattias Polborn, Kim J. Ruhl, Pedro Sant’anna, Eric Sims, Matthew Turner, Jun (Bean) Zhao, and Kai (Jackie) Zhao. All errors are my own.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary material 1 (pdf 250 KB)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Cheng, D. Housing boom and non-housing consumption: evidence from urban households in China. Empir Econ (2021).

Download citation


  • Consumption
  • Housing prices
  • Wealth effect
  • Collateral constraint

JEL Classification

  • D15
  • E21
  • E32
  • R21
  • R31