Job Market Paper

This paper disentangles the roles of capital regulation and credit ratings in mitigating insolvency risk in the U.S. property insurance market. I first investigate the mechanism through which capital requirements affect the insurance market. Using reduced-form evidence and an instrumental variable approach that exploits a 2017 policy change as a quasi-experiment, I find that a $1 million increase in required capital leads insurers to hold $3.34 million more in capital and to raise insurance prices by 0.218 percentage points. These results reveal a direct trade-off between financial stability and consumer affordability. To further explore the underlying mechanisms, I develop a structural model in which insurers make capital and pricing decisions in a competitive market with limited liability and exposure to catastrophic risks. Counterfactual analyses show that tightening capital requirements improves solvency but raises prices. In the absence of capital regulation, the model predicts that the insolvency rate would increase by 0.09 percentage points, while insurance prices would decline by about 5.1%, accompanied by greater risk-taking and market concentration. A third counterfactual scenario examines a market without capital regulation but with high credit rating salience. When consumers place greater emphasis on credit ratings, the insolvency rate decreases; however, intensified price competition reduces profitability and increases market concentration. Overall, the findings underscore that capital regulation remains crucial for sustaining market stability, as heightened rating salience alone cannot fully substitute for its stabilizing effects.

This paper studies the impacts of different types of post-secondary education on education and labor market outcomes. I first use variations of distances to institutions and exogenous variations of colleges that upgraded into universities to investigate the value added to university education. Then I separately estimate the impacts of the transformed universities and traditional colleges and universities. I compared results from OLS and IV regressions. To address the treatment heterogeneity, I adapt the locally linear specification from Mountjoy (2022). Results suggest that the difference between university and college graduates is marginal. However, university entry improves labor market outcomes, such as employment and earnings compared with cohorts without post-secondary education. Graduates from transformed universities obtain a higher probability of being employed and higher earnings compared with people without post-secondary education. Nonetheless, transformed university graduates may have worse performance in the labor market, compared with graduates from colleges or traditional universities. Furthermore, graduates from transformed universities are more willing to register for training when they are unemployed compared with the other three groups.

(Draft available on request)

This paper studies how employer mandates and health insurance affect labor market outcomes and health. I use staggered difference-in-differences research design and variations in the Affordable Care Act to learn how employer mandate affects labor market outcomes. I use doubly robust difference-in-differences in my main specifications to reduce selection bias. Results in the full sample suggest that the employer mandates in the Affordable Care Act increased hourly wages and did not have significant impacts on employment and part-time employment. Employer mandates stimulate a larger increase in employer-sponsored health insurance coverage rates among low-income workers. However, low-income workers are more vulnerable to involuntary part-time employment if employers reduce work hours to circumvent employer mandates. Firms prefer to reduce work hours to circumvent employer mandates instead of firing workers. Using doubly robust estimators and staggered difference-in-differences research design, I find evidence that providing health insurance improves workers' health. The employer mandate may increase productivity by improving workers' health status. Still, it may widen income inequality in the long run because low-income workers are more vulnerable to work hours losses.

This paper examines the policy debate on transitioning from static, formula-based capital regulation in the U.S. property insurance market to a dynamic, model-based approach. I propose a theoretical framework in which regulators set a target insolvency probability, compelling insurers to adjust capital holdings to remain below this threshold. To operationalize this framework, I develop a learning-and-verification methodology that uses reinforcement learning to identify optimal regulatory policies, formal reasoning to verify compliance with safety constraints, and decision-tree extraction to produce interpretable policy rules. This hybrid approach enables regulators to adapt to changing market conditions while maintaining accountability and transparency. The results suggest that a model-based regulatory regime can enhance the resilience and adaptability of the insurance sector in the face of growing uncertainty.

Stringent capital regulation in property insurance markets creates a tension between insurer solvency and affordability. This paper demonstrates that adaptation to natural catastrophes can mitigate this trade-off. I develop and estimate a model of the property insurance market with interdependent loss structures, competition, and capital regulation. My counterfactual analysis shows that adaptation measures, by reducing the correlation of losses, would decrease the market insolvency rate and lower market concentration. Furthermore, I find a net positive impact on social welfare when the costs of insurer insolvency are repurposed to subsidize adaptation. My results indicate that policy should focus on incentivizing adaptation as a key tool for maintaining sustainable insurance markets.

This paper investigates the complex relationship between industrial agriculture, plant biodiversity, and economic welfare. While industrial agriculture has been credited with increasing food production, it has also been linked to a significant decline in agrobiodiversity. This research disentangles two opposing economic forces: the homogenizing effect of industrial agriculture's focus on a few high-yield, storable crops, and the potential for market competition to foster product variety. I develop and estimate a structural model of demand and supply for fresh produce, employing a random coefficients logit model (BLP) to capture heterogeneous consumer preferences for a wide range of crop attributes, including variety. On the supply side, I model farmers' and food companies' decisions to offer different cultivars, considering the influence of production costs, market structure, and agricultural policies. Using detailed market-level data, I quantify the welfare effects of changes in plant biodiversity on consumers and producers. Furthermore, I extend the traditional welfare analysis to incorporate the non-market value of agrobiodiversity as a source of resilience to climate shocks. My findings aim to provide a more complete picture of the true social costs and benefits of industrial agriculture and to inform policies that promote a more diverse and sustainable food system.