Anticorruption Enforcement and Sale Mechanism Choice in China's Land Market

Upon taking office in late 2012, Chinese President Xi Jinping launched one of the most intensive anticorruption campaigns in the history of the People’s Republic of China. Prior to the campaign, China’s land market suffered from corruption, particularly surrounding sale method selection (auction versus listing). Listing is a two-stage sale mechanism that prior research has identified as more susceptible to corruption, leading to lower prices. This paper examines the campaign’s impact on land allocation, focusing on whether corruption influences the choice of sale method and, in turn, land sale prices. This paper is the first to utilize Blackwell and Yamauchi (2021, 2024)’s marginal structural model with fixed effects in the inverse probability of treatment weighting model; absorbing time-invariant unobserved confounding and utilizing a set of time-varying covariates as controls, this model can estimate causal effects in the land sale case. I find that indictments in a prefecture cause a statistically significant drop in the probability that land is sold via listing$\unicode{x2014}$an effect that is further compounded when indictments occur in consecutive months. Sensitivity analyses indicate that any violations of the identification assumptions would bias estimates towards zero, confirming the negative effect. A second marginal structural model shows that both mean and median land sale prices increase in the presence of indictments. Together, these results suggest that the anticorruption campaign not only deterred actual corrupt allocation practices, but also impacted the discretionary use of listings.

💡 Research Summary

This paper investigates how President Xi Jinping’s sweeping anti‑corruption campaign, launched in late 2012, altered the behavior of Chinese local governments in the land‑sale market, focusing on two outcomes: the choice between “listing” (a two‑stage auction) and standard English auctions, and the resulting land sale prices. The author argues that listings are especially vulnerable to rent‑seeking because the first stage can be manipulated by officials, often resulting in a single bidder and a low final price. By contrast, English auctions are more competitive and tend to generate higher revenues.

To test whether anti‑corruption enforcement changed these practices, the study combines a prefecture‑level monthly dataset of corruption indictments with a newly constructed, cleaned database of land transactions scraped from the Chinese government’s land‑transaction system. The author first attempts a conventional two‑way fixed‑effects specification but finds that it fails to satisfy key identification assumptions, prompting the adoption of a more sophisticated causal framework.

The core methodology is the marginal structural model (MSM) with fixed effects in the inverse‑probability‑of‑treatment‑weighting (IPTW) estimator, as developed by Blackwell and Yamauchi (2021, 2024). In the first stage, a logistic IPTW model predicts the probability of an indictment in a given prefecture‑month, conditioning on a rich set of time‑varying covariates (e.g., local GDP, land‑supply quotas, population growth, policy shocks) and including prefecture fixed effects to absorb time‑invariant unobserved heterogeneity. The resulting weights are then applied in two separate MSMs: one with the binary outcome “any listing used” and another with continuous outcomes (mean and median price per square meter). This approach simultaneously controls for both observed time‑varying confounders and unobserved time‑invariant confounders, addressing the shortcomings of standard fixed‑effects models.

The first MSM yields a statistically significant negative effect of indictments on the probability of using a listing. A single month with an indictment reduces the probability of any listing by 1.16 percentage points on average. When indictments occur in consecutive months, the effect compounds; a five‑month window surrounding an indictment is associated with a 7.78‑percentage‑point decline in listing usage. Sensitivity analyses show that violations of the identification assumptions would bias the estimates toward zero, suggesting that the reported effects are conservative.

The second MSM demonstrates that indictments increase land sale values. In months when a prefecture records an indictment, the mean price per square meter rises by approximately 6.78 %, and the median shows a comparable increase. Given a baseline mean price of about 2,200 yuan/m² (≈ US $350), a 7 % uplift translates into several hundred thousand yuan per parcel, a material gain for local governments. The author interprets this as evidence that the reduction in listings—combined with a shift toward more competitive English auctions—enhanced market efficiency and lifted revenues.

Beyond the substantive findings, the paper contributes a high‑quality dataset that corrects known inaccuracies in the widely used Chen‑Kung (2019) compilation. By meticulously cleaning indictment records and matching them to transaction‑level land data, the author provides a valuable resource for future research on Chinese real estate and policy evaluation.

In sum, the study provides robust causal evidence that the anti‑corruption campaign not only curtailed overt corrupt allocation (fewer listings of “favorable” parcels) but also altered discretionary behavior, nudging officials toward more transparent, competition‑driven sale mechanisms. The methodological innovation—applying Blackwell‑Yamauchi’s MSM with fixed effects in an IPTW framework—offers a template for rigorous policy impact assessment in settings where both time‑invariant and time‑varying confounders are present.