Marriage entry, divorce and reconciliation: The unintended consequence of the home purchase restriction policy in China

The study addresses the rising trends of delayed marriage, increased divorce, and the growth of cohabitation, often termed the "second demographic transition." It leverages Becker's economic theory of marriage, which posits that marital decisions are rational choices based on socioeconomic factors like wealth and education. The researchers examine how the Chinese HPR policy, implemented to curb rising property prices by limiting property purchases per household, impacts family formation. While existing research focuses on HPR's effect on property prices, this study explores its largely unexplored influence on marriage and divorce patterns. The authors posit that HPR might influence marriage in two ways: by encouraging marriage due to potentially lower house prices (making homeownership, a prerequisite for marriage in Chinese culture, more attainable) or by encouraging divorce to exploit a policy loophole allowing divorced individuals to purchase more properties. The study's importance lies in highlighting the unintended consequences of non-family-oriented policies on family demographics and the need for policymakers to consider the broader social ramifications of their decisions, especially given China's demographic challenges.

The literature review examines existing research on the relationship between house prices and marriage stability. Some studies link falling house prices to higher divorce rates for owner-occupier couples, particularly young, low-income ones with high mortgage debt, suggesting that reduced wealth diminishes the perceived benefits of marriage compared to divorce. Conversely, other studies find that declining house prices decrease divorce risk due to high transaction costs during market downturns. Yet others find no statistically significant correlation between marital stability and housing market fluctuations. The inconsistency of findings motivates this study's in-depth examination of the Chinese context, focusing on the HPR policy's influence. The authors note the limited research on the policy's impact on family formation and the scarcity of studies examining the causal relationship between house prices and marital status, highlighting a gap that the study aims to fill. They mention one related study which used indirect measures of divorce (Baidu keyword searches) rather than official data.

The study employs a difference-in-differences (DID) approach to evaluate the causal impact of the HPR policy on family formation. Initially, a sample of 125 first, second, and third-tier Chinese cities from 2007 to 2016 was used. A logit regression model (Eq. 1) was used to identify socioeconomic variables that predicted a city's adoption of the HPR policy. This included variables such as GDP, population, fixed asset investment, municipal revenue, and a housing affordability index. The 2010 affordability index emerged as the strongest predictor, enabling the researchers to create treatment (cities implementing HPR) and control groups with comparable initial socioeconomic conditions. The final sample consisted of 71 cities. A two-way fixed effects (TWFE) DID model (Eq. 2) was then used to assess the impact of HPR on several outcome variables: housing prices, marriage entry (total and disaggregated by age group), divorce, remarriage, and reconciliation. City-fixed effects, year-fixed effects, and city-specific linear trends were included to control for time-invariant and time-variant city-specific factors. Equation (3) was used to assess pre-policy parallel trends assumption. The authors acknowledge the potential biases due to heterogeneous treatment effects and use HTE-robust estimators proposed by De Chaisemartin and D'Haultfoeuille (2020) and confirm that all individual treatment effect weights are positive, thus eliminating any negative weighting bias in the TWFE DID model. Data were obtained from multiple sources including municipal government websites, China Civil Affairs statistical yearbooks, and China City Statistical Yearbooks.

The study finds that the HPR policy effectively reduced house prices, validating the policy's primary goal. The magnitude of the price reduction varies depending on the model used, ranging from a 5.8% reduction in the basic model to a 9.1% reduction in the model with the affordability index criteria for control group selection. The study shows that the HPR policy led to a statistically significant decrease in the number of marriages among the 20-24 age group (9.9% reduction) and the 25-29 age group in Tier 1 cities (13% reduction). In contrast, there's a statistically significant increase in marriages among the 30-34 and 35-39 age groups in Tier 1 cities (18.4% and 14.5% increases respectively). While the overall impact on divorce and remarriage was not statistically significant, a significant increase in marital reconciliations (55.4% in first-tier cities and 39.4% in new first-tier cities) suggests that couples engaged in strategic divorce to circumvent HPR purchase restrictions. This indicates that the policy incentivized such behavior, particularly in large cities with higher potential returns on housing investment and more favorable down payment ratios.

The findings support Becker's economic theory of marriage, demonstrating that housing affordability significantly impacts marriage decisions. The results highlight the unintended consequences of housing policies on family formation and stability. The policy's effect on marriage entry varies across age groups and city tiers. The delay in marriage among younger individuals and the increase in strategic divorces indicate the policy's unintended negative impacts on family stability and demographic trends. The findings are more consistent with studies linking negative house price shocks to increased divorce rates than those showing decreased divorce risks associated with declining house prices, as the potential gain of increased family wealth through strategic divorces appears to have overridden other factors influencing marriage decisions. The study contributes to the literature by highlighting the overlooked social ramifications of housing policies designed primarily to address economic concerns.

This study demonstrates that China's HPR policy, while successfully curbing house prices, has had unintended and significant negative consequences on marriage patterns, particularly among younger individuals and in higher-tier cities. The finding of strategic divorce to exploit policy loopholes further emphasizes the need for careful consideration of wider societal effects when designing housing market regulations. Future research should investigate the issue at a micro-level to better understand households' motivations and refine policy implications.

The study's reliance on city-level data limits the ability to explore individual household-level factors influencing marriage decisions. The inability to account for unobservable factors at the household level could lead to some omitted variable bias. While HTE-robust methods were employed to address potential heterogeneity in treatment effects, the impact of residual confounding factors cannot be completely ruled out. Future research incorporating micro-level household data would allow a more in-depth understanding of the mechanisms driving marriage decisions in response to housing policies.