Knowledge that Transforms

The importance of skilled labor and the inalienability of human capital expose firms to the risk of losing talent at critical times. Using Swedish microdata, we document that firms lose workers with the highest cognitive and noncognitive skills as they approach bankruptcy. In a quasi‐experiment, we confirm that financial distress drives these results: following a negative export shock caused by exogenous currency movements, talent abandons the firm, but only if the exporter is highly leveraged. Consistent with talent dependence being associated with higher labor costs of financial distress, firms that rely more on talent have more conservative capital structures.

How do investors perceive dependence between stock returns; and how does their perception of dependence affect investments and stock prices? We show experimentally that investors understand differences in dependence, but not in terms of correlation. Participants invest as if applying a simple counting heuristic for the frequency of comovement. They diversify more when the frequency of comovement is lower even if correlation is higher due to dependence in the tails. Building on our experimental findings, we empirically analyze U.S. stock returns. We identify a robust return premium for stocks with high frequencies of comovement with the market return.

We show that aversion to risk and ambiguity leads to information inertia when investors process public news about assets. Optimal portfolios do not always depend on news that is worse than expected; hence, the equilibrium stock price does not reflect this bad news. This informational inefficiency is more severe when there is more risk and ambiguity but disappears when investors are risk‐neutral or the news is about idiosyncratic risk. Information inertia leads to news momentum (e.g., after earnings announcements) and is consistent with low household trading activity. An ambiguity premium helps explain the macro and earnings announcement premium.

We develop a model in which customer capital depends on key talents' contribution and pure brand recognition. Customer capital guarantees stable demand but is fragile to financial constraints risk if retained mainly by talents, who tend to quit financially constrained firms, damaging customer capital. Using a proprietary, granular brand‐perception survey, we construct a firm‐level measure of the inalienability of customer capital (ICC) that captures the degree to which customer capital depends on talents. Firms with higher ICC have higher average returns, higher talent turnover, and more precautionary financial policies. The ICC‐sorted long‐short portfolio's spread comoves with financial constraints factor.

I develop a structural model of mortgage demand and lender competition to study how leverage regulation affects the U.K. mortgage market. Using variation in risk‐weighted capital requirements across lenders and mortgages with different loan‐to‐values (LTVs), I show that a 1‐percentage‐point increase in risk‐weighted capital requirements increases lenders' marginal cost of originating mortgages by about 26 basis points (11%) on average. I use the estimated model to study proposed leverage regulations. Counterfactual analyses show that large lenders exploit a regulatory cost advantage, which increases concentration by about 20%, and suggest that banning high‐LTV mortgages may reduce large lenders' equity buffer.

An entrepreneur borrows from a relationship bank or the market. The bank has a higher cost of capital but produces private information over time. While the entrepreneur accumulates reputation as the lending relationship continues, asymmetric information is also developed between the bank/entrepreneur and the market. In this setting, zombie lending is inevitable: Once the entrepreneur becomes sufficiently reputable, the bank will roll over loans even after learning bad news, for the prospect of future market financing. Zombie lending is mitigated when the entrepreneur faces financial constraints. Finally, the bank stops producing information too early if information production is costly.

We study risk premiums in the U.S. Treasury bond market from the perspective of a Bayesian econometrician who learns in real time from disagreement among investors about future bond yields. Notably, disagreement has substantial predictive power for yields, and 's risk premiums are less volatile than those in the analogous model without learning. 's forecasts are substantially more accurate than the consensus forecasts of market professionals, particularly following U.S. recessions. The predictive power of disagreement is distinct from the (much weaker) one of inflation and output growth. Rather, it appears to reflect uncertainty about future fiscal policy.

We provide a theoretical framework to uncover in a model‐free way the relationships among international stochastic discount factors (SDFs), stochastic wedges, and financial market structures. Exchange rates are in general different from the ratio of international SDFs in incomplete markets, as captured by a stochastic wedge. We show theoretically that this wedge can be zero in incomplete and integrated markets. Market segmentation breaks the strong link between exchange rates and international SDFs, which helps address salient features of international asset returns while keeping the volatility and cross‐country correlation of SDFs at moderate levels.

We study how parent liability for subsidiaries' environmental cleanup costs affects industrial pollution and production. Our empirical setting exploits a Supreme Court decision that strengthened parent limited liability protection for some subsidiaries. Using a difference‐in‐differences framework, we find that stronger liability protection for parents leads to a 5% to 9% increase in toxic emissions by subsidiaries. Evidence suggests the increase in pollution is driven by lower investment in abatement technologies rather than increased production. Cross‐sectional tests suggest convexities associated with insolvency and executive compensation drive heterogeneous effects. Overall, our findings highlight the moral hazard problem associated with limited liability.

We estimate real losses arising from the cross‐sectional misallocation of financial liabilities. Extending a production‐based framework of misallocation measurement to the liabilities side of the balance sheet and using manufacturing firm data from the United States and China, we find significant misallocation of debt and equity in China but not the United States. Reallocating liabilities of firms in China to mimic U.S. efficiency would produce gains of 51% to 69% in real value‐added, with only 17% to 21% stemming from inefficient debt‐equity combinations. For Chinese firms that are large or in developed cities, we estimate lower distortionary financing costs.