Using a comprehensive dataset containing a detailed nine-level political hierarchy classification for all state-owned enterprises (SOEs) in China from 1998 to 2013, we find that low-ranked SOEs have 14% higher productivity than private firms. The result is robust to propensity score matching, instrumental variable approach, and within-firm analysis. Further findings lend support to the theory of local government's protection as opposed to local information advantage. Macro-level results suggest that growth benefits might lead local government to resist the central government reform of ending low-ranked SOEs.

This paper addresses police resource allocation across multiple locations, aiming to minimize the overall cost of potential crimes. Unlike previous literature focused on reactive police tasks, we propose a proactive approach that emphasizes crime prevention through deterrence. To account for the deterrence effect of police resources on crime, we employ the multinomial logit model to calibrate the distribution of crime locations. Our model sheds light on two facets of the deterrence effect in proactive policing—crime control diffusion and crime displacement—relevant to modern crime patterns from both criminology and economics perspectives. We establish the NP-hardness of our problem and provide mixed-integer linear/conic reformulations solvable with standard optimization software. Additionally, we extend our results to a dynamic model over multiple time periods. Finally, we showcase the efficacy of our model through a data-driven case study on the allocation of surveillance cameras in New York City.

A recommendation platform sequentially collects information about a new product revealed from past consumer trials, and uses it to better guide later consumers. Because consumers do not internalize the value of information they bring to others, their incentive for trying out the product can be socially insufficient. Given such a challenge, I study how the platform can improve social welfare by designing its recommendation policy. In a model with binary product quality and general trial-generated signals, I find that the optimal design features a U-shaped sequence of recommendation standards over the product’s life, and the optimal learning dynamic can involve temporary suspensions following negative consumer feedback when the product is young. Comparative statics and extensions explore how the optimal design adjusts under changes in trial informativeness, consumer arrival rates, and platform bias. My analysis also illustrates the usefulness of a Lagrangian duality approach for dynamic information design.

The intensifying rivalry between major global economies highlights the need to better understand the politicized nature of international business. This study introduces a new political factor—community political ideology (i.e., the dominant political ideology along the liberalism-conservatism spectrum among a community’s members)—and examines its influences on cross-border M&As between countries that are deemed as economic and political rivals. Building on the literature on political ideology and cross-border M&A, we argue that conservative communities tend to perceive greater threats and uncertainty posed by cross-border M&As from rival countries and therefore exert stronger resistance to these deals. Using Chinese cross-border M&As in the U.S. as a research context, we predict that these M&As are less likely to be completed in U.S. communities with a higher proportion of conservative residents. Moreover, the negative effect of community conservatism on the completion of Chinese M&A deals in the U.S. is further enhanced when the target firms are in sensitive industries or when the target communities suffer from greater economic distress. An analysis of 267 Chinese cross-border M&As in the U.S. from 2002 to 2021 supports these arguments. Our study contributes to research on geopolitical rivalry, political ideology, and cross-border M&As.

This paper studies the impact of trade openness on welfare through alterations in workers' skill acquisition. Guided by empirical evidence, we integrate endogenous choices of learning investments into a multisector Eaton–Kortum model. Our model reveals that trade openness influences skill acquisition by two channels: (1) reallocating labor between sectors with varying skill intensities and on-the-job learning opportunities and (2) allowing producers in each country to source varieties from more cost-effective suppliers in other countries, thus reducing costs of material inputs for learning. Our quantification indicates that the gains in skill acquisition account for 5% of the total gains from trade.

This paper studies inference in two-stage randomized experiments under covariate-adaptive randomization. In the initial stage of this experimental design, clusters (e.g., households, schools, or graph partitions) are stratified and randomly assigned to control or treatment groups based on cluster-level covariates. Subsequently, an independent second-stage design is carried out, wherein units within each treated cluster are further stratified and randomly assigned to either control or treatment groups, based on individual-level covariates. Under the homogeneous partial interference assumption, I establish conditions under which the proposed difference-in-“average of averages” estimators are consistent and asymptotically normal for the corresponding average primary and spillover effects and develop consistent estimators of their asymptotic variances. Combining these results establishes the asymptotic validity of tests based on these estimators. My findings suggest that ignoring covariate information in the design stage can result in efficiency loss, and commonly used inference methods that ignore or improperly use covariate information can lead to either conservative or invalid inference. Then, I apply these results to studying optimal use of covariate information under covariate-adaptive randomization in large samples, and demonstrate that a specific generalized matched-pair design achieves minimum asymptotic variance for each proposed estimator. Finally, I discuss covariate adjustment, which incorporates additional baseline covariates not used for treatment assignment. The practical relevance of the theoretical results is illustrated through a simulation study and an empirical application.

Marketers are increasingly using unconventional design tactics to visually disrupt consumer expectations, like turning brand logos upside down. Across four experiments, this research examined how inverted logos influence consumer brand responses. In two binary choice tasks (Studies 1A and 1B), participants exhibited a lower preference for an inverted logo than a standard logo for branded products. Study 2 determined the psychological mechanism underlying this effect: inverted logos increase perceived unexpectedness, which increases perceptions of brand rebelliousness and, ultimately, reduces purchase intentions. Study 3 demonstrated that political ideology moderates this effect: more conservative, but not liberal, consumers respond negatively to inverted logos. Finally, we discussed the theoretical and practical implications for logo design and visual branding strategies.