Longwei Cheng

Poor service by insurers costs U.S. households billions of dollars a year. I examine whether financial shocks cause insurers to degrade the quality of their service. Natural disasters impose large and plausibly exogenous losses on the property and casualty (P&C) insurance groups exposed to them. Using the universe of customer complaints filed with state regulators as a measure of service quality between 2014 and 2023, I find that complaints against these groups' affiliated insurers rise by 4%. The increase concentrates in complaints the regulator resolves in favor of customers, suggesting a genuine decline in service quality. Capital constraints do not seem to explain the decline. The evidence is more consistent with the capital-market pressure to meet earnings targets. The findings show that financial shocks degrade the service insurers provide even where they stay solvent, a decline that solvency-based regulation does not detect.

Wealthier individuals earn systematically higher returns on their wealth, yet the channels driving this return gap remain unclear. Using a proprietary dataset of over 90,000 Indian retail investors, we document that average annual excess returns rise monotonically with financial wealth, driven almost entirely by higher risk-adjusted returns, not by differences in systematic risk. Exploiting India's 2015 reform of its insider-trading regulations, which sharply increased the cost of trading on private information, we show that the risk-adjusted return gap between the top 1% and the rest shrinks by 68%. The top 1% trade disproportionately in stocks during proven insider-trading violation windows, and their pre-announcement buying of M&A target firms disappears after the reform. To examine how wealthier individuals obtain private information, we draw on information-network theory and use two complementary proxies for investor connections to show that wealthier individuals have better-informed connections. Overall, private information can generate substantial return inequality even in stock markets commonly viewed as level playing fields.

We examine whether misconduct by insurance agents is disciplined by the labor market. Past misconduct strongly predicts future misconduct, making it an informative signal, yet the labor market response is limited. Complaints modestly increase industry exit and insurer-appointment terminations, but agents often replace lost appointments, leaving their access to insurers largely unchanged. The response to complaints comes almost entirely from Texas, the one state that posts complaints publicly, while complaints that are visible only to regulators generate little response. Regulatory sanctions increase exit only when they revoke a license, and even revoked agents usually remain in the industry, because licenses are issued state by state. Among agents registered as both insurance agents and brokers, insurance misconduct has stronger labor market consequences in the brokerage market than in insurance, even though the agent and the conduct are the same. The findings show that informative misconduct signals produce weak discipline unless institutions make the misconduct visible to the market or a firm is responsible for acting on it.