Superstar founders—those who take a firm public through a true IPO—represent a substantial share of the extreme right tail of the US wealth distribution. This paper provides the first comprehensive analysis of their wealth accumulation, from the IPO through the firm’s public lifecycle. Using hand-collected data on founder and venture capital ownership from 1996 to 2021, we document three key findings. First, we quantify a persistent trade-off: while VC backing enables firms to reach a larger scale at IPO, it is associated with a significant reduction in founder ownership and wealth. Second, we introduce a novel and comprehensive wealth measure that incorporates both the accumulated proceeds from post-IPO share liquidations and the gains from underpriced acquisitions. We show that relying solely on retained ownership understates founder wealth by 50% for firms ten years post-IPO. Additionally, founders' acquisition of new shares below market price contributes 5% to total wealth. Third, we demonstrate that path-dependent ownership dynamics, predictable from the initial IPO stake, explain 40% of the variation in founders’ wealth growth rates. Ultimately, extreme wealth is achieved by founders who maintain substantial stakes in the selection of firms that persistently outperform the market and achieve massive scale over time. Our findings provide the microfoundations for the rise in top wealth shares, highlighting the IPO as a key entry channel and post-IPO ownership dynamics as critical amplifiers of founder fortunes.

This paper identifies a new channel of informed insider trading: insiders can exploit delayed market reactions to public information. Using U.S. insider transactions, firm-level news, and daily stock-market data from 2000 to 2024, we show that after favorable public news, the usual negative turnover--return relation reverses before insider sales. Higher turnover predicts higher subsequent returns, especially when price adjustment is delayed, suggesting that investors require compensation for unresolved price-discovery risk. Insiders appear to trade strategically during these windows. The turnover-related premium is concentrated in opportunistic trades and exercise-and-sell transactions, and insiders are more likely to bundle sales with same-day option exercises after favorable pre-sale returns. The effect is stronger among executive officers and shifts around role transitions. These findings show that insider trading advantages extend beyond undisclosed information and that regulation focused only on trade timing may miss an important margin: how insiders structure sales while public information is still being absorbed by the market.

This study investigates whether increased diversification in insurance-linked securities (ILS) hedge funds improves their performance and identifies the underlying mechanisms. We develop a new measure of diversification based on cross-class return dispersion, which captures ILS funds' flexibility in adjusting their catastrophe risk exposure, financial market timing, and factor selection. Our findings show that more diversified ILS funds achieve higher returns without increasing risk, with performance driven primarily by factor returns rather than excess returns. These funds also outperform their counterparts during catastrophe market downturns by dynamically reallocating risk and exploiting market timing opportunities. An event study reveals that these funds actively manage their risk exposure before negative shocks, but adopt a risk-averse stance afterward, potentially sacrificing gains. Our results reconcile the conflicting findings in the literature, demonstrating that strategic and dynamic diversification enhances ILS fund performance through factor-driven mechanisms, rather than excessive specialization or passive risk allocation.

We investigate how niche assets (e.g., catastrophe bonds) affect fixed-income hedge fund portfolio performance. Using a portfolio choice model with adjustable weights for both niche and existing assets, we demonstrate that niche assets provide diversification or substitution benefits depending on time-varying correlations and relative Sharpe ratios between the asset types. We further find that diversification and substitution benefits are equally important when incorporating niche assets but are often neglected. We propose improving fixed-income funds by dynamically adjusting the portfolio allocations of niche and existing assets according to market conditions or including niche strategy funds in a fund-of-hedge-funds approach.

I propose that the nonfinancial component of financial firms' assets, in particular the growth opportunities associated with business operations, drives most of the variation in their equity valuation. I document this fact for a large class of intermediaries: life insurance companies. In particular, I decompose insurers' market equity returns into net financial asset returns and net business asset returns and show that these two components have very different risk exposures and are negatively correlated outside of the 2008-2009 financial crisis. The variation in life insurers' net business asset returns drives 81% of the aggregate time series variation and 100% of the cross-sectional variation in their market equity returns. For this reason, the current intense regulation on life insurers' net financial assets may be insufficient as a great deal of risk is derived from their net business assets which are comparatively under-regulated.

Capital reallocation is procyclical, despite measured productive reallocative opportunities being acyclical, or even countercyclical. This paper reviews the advances in the literature studying the causes and consequences of capital reallocation (or lack thereof). We provide a comprehensive set of capital reallocation stylized facts for the US, and an illustrative model of capital reallocation in equilibrium. We relate capital reallocation to the broader literatures on business cycles with financial frictions, and on resource misallocation and aggregate productivity. Finally, we provide directions for future research.

This paper studies whether firms' innovation talk is supported by subsequent innovation output, and whether innovation washing spills over across peer firms. We measure innovation talk using FinBERT-classified positive innovation-related sentences in the MD\&A sections of annual reports, and we measure realized innovation using subsequent invention patent applications. Using Chinese A-share listed firms from 2008 to 2024, a setting in which innovation is strongly encouraged by policy and capital markets, we examine whether firms’ innovation talk is supported by subsequent innovation output and whether innovation washing spills over across peers. We first document a wedge between innovation talk and innovation output. Firms with more positive innovation disclosure subsequently invest more in R&D, but they do not necessarily produce more invention patents. High-talk firms therefore exhibit lower conversion from innovation investment into realized patent output. We then construct a firm-year measure of innovation washing as the extent to which a firm's innovation talk exceeds what is supported by its subsequent invention patent output and observable characteristics. Using this measure, we find strong peer effects in innovation washing: firms increase their own innovation washing when their industry peers engage in more washing. Economically, a one-standard-deviation increase in peer innovation washing is associated with a 3.31% increase in the firm's subsequent innovation washing.

Whether algorithmic pricing facilitates tacit collusion and harms consumer welfare has become a central antitrust concern in digital markets. We argue that demand elasticity is a key determinant of how algorithmic pricing affects competition. When demand is relatively inelastic, pricing algorithms can more easily sustain tacitly collusive outcomes. In contrast, when demand is highly elastic, consumer price sensitivity constrains coordinated price increases and shifts the role of algorithms toward improving operational efficiency. We test this hypothesis using high-frequency operating data from more than 70 electric-vehicle charging operators and over 7,000 charging stations in Chongqing, China, between 2025 and 2026. We identify AI-based dynamic pricing using textual information and observed pricing patterns, and address endogenous adoption using the local market share of non-AI charging stations as an instrumental variable. We find that, in the highly elastic EV-charging market, AI-priced stations achieve significantly higher utilization and more efficient load management, without evidence of systematic price coordination. A theoretical model further shows that collusive equilibria become difficult to sustain as demand elasticity increases. By contrasting the charging market with the less elastic gasoline market, our analysis identifies demand elasticity as a critical boundary condition governing whether algorithmic pricing promotes collusion or efficiency. These findings provide a basis for more targeted regulation of algorithmic pricing across markets with different demand characteristics.