Abstract: This paper presents and tests a theory in which firms manage capital structure using simple rules, largely ignoring leverage until it approaches critical thresholds. We formalize this behavior in a structural agency model in which incentive conflicts gain priority only outside a safe leverage range. We structurally identify firm-specific thresholds averaging 8.7% and 76% debt-to-assets, with variation linked to model-predicted determinants including free cash flow,  governance and risk. The model achieves predictive accuracy between 68% and 74%. Our findings help reconcile theoretical predictions with real-world financial decision-making.

Abstract: Empirical corporate finance often relies on samples of publicly listed firms because data on private firms are limited. Since listing is endogenous, such samples may be subject to selection bias. This paper studies the implications of ignoring private firms for empirical inference. We propose a tractable semiparametric approach that corrects this bias without requiring data on unobserved firms. Applied to investment regressions, the method materially changes estimated sensitivities: the investment–cash flow relation weakens substantially, while the sensitivity of investment to q increases markedly. These results suggest that excluding private firms can bias estimates obtained from commonly used empirical specifications.

Abstract: This paper measures the cost of greening the economy from the customer’s perspective using nearly six million US federal procurement contracts from 2007 to 2024. Green contracts are on average 18 to 43 percent more expensive than comparable non-green contracts. Accounting for endogeneity with a Bartik instrument yields an even higher cost premium. The premium increases with public concern about climate change, declines with experience, and rises with regulatory complexity. Green contracts also involve greater administrative effort, including more modifications and delays. Overall, the green transition imposes substantial but partly transitory costs shaped by public sentiment, learning, and regulation.

Abstract: Productivity dispersion across firms remains large and persistent, in part because technology diffusion is slow and uneven. Prior work attributes these patterns to skill gaps between firms and workers, but largely treats these gaps as exogenous and abstracts from the forces that shape them. This paper presents evidence that skill gaps evolve endogenously over the technology life cycle. We construct granular measures of skill gaps using matched data on job postings and worker résumés and document three sets of findings. First, skill gaps follow a U-shaped pattern over the technology life cycle: they are highest at the onset of adoption, decline as firms and workers adjust, and rise again as technologies approach obsolescence. Second, skill gaps reflect shortages in combinations of technical and nontechnical skills, rather than technical skill deficits alone. Third, skill gaps are systematically smaller in firms that are larger, older, and have higher stocks of organizational capital, consistent with these firms having greater capacity to align skill demand with worker capabilities. The results imply that skill gaps are not merely exogenous frictions to technology adoption, but rather, endogenous outcomes of the diffusion process itself. Ignoring this endogeneity obscures the interpretation of observed skill gaps, biases inference about their effects on productivity, and may result in policies that amplify, rather than mitigate, productivity dispersion.

Abstract: A growing literature suggests that managers learn from the financial market and react to it, leading to a "feedback effect" on firm investment decisions that is distinct from the effect of firm fundamentals. Identifying this effect is complicated by the unobservability of fundamentals and the challenge in finding exogenous shocks. We develop a firm investment model and a novel identification strategy to disentangle the feedback effect from the effect of firm fundamentals. The model yields an extended q-theory investment equation in which Tobin's q mediates the feedback effect while also proxying for marginal q. We deploy novel econometric approaches to bound the investment equation coefficients, obtain point estimates based on higher-order moments, and analyze the sensitivity of the estimates. Our estimates admit a nonzero feedback effect that is substantially larger than previously estimated. The more precise the signal from the financial market is, the more investment reacts to it.

Abstract: This paper develops and tests a Q-theoretic framework for government investment. We present a model in which a welfare-maximizing government chooses public investment while subsidizing private investment, predicting that government investment increases with government Q but decreases with private-sector Q due to local resource constraints. We construct novel regional measures of government Q from municipal bond prices and aggregate firm-level Qs to measure private-sector Q. Consistent with the model, government investment is positively related to government Q and negatively related to private-sector Q, with stronger effects in regions with larger subsidy programs. The results link corporate investment theory and public finance through market-based valuations, and offer a unified approach to studying economy-wide resource allocation.

Abstract: We provide the first structural assessment of the U.S. green transition using the $q$-theory of investment. Distinguishing emissions-free green capital from emissions-intensive brown capital, we develop a transparent identification strategy that infers the replacement cost of the brown capital stock directly from observed emissions, avoiding concerns about greenwashing. Structural estimation yields the market values and adjustment costs of green and brown capital. A pronounced divergence emerges: by 2022, brown capital accounts for 37% of replacement costs but only 22% of market value, indicating that markets value green capital at a premium and discount brown capital. Green investment entails large adjustment costs— 16.7% of output—reflecting both its intensity and the frictions involved in scaling up green capital.

Abstract: Although bond insurance intermediaries are frequently relied upon by local governments for external financing, there is significant debate about the value that bond insurers provide. In this paper, we study U.S. drinking water to estimate the real effects of bond insurance on public infrastructure. We show that exogenous reductions in municipalities' access to bond insurance cause local governments to face higher borrowing costs, reduce external bond issuance, decrease investment in water infrastructure, and experience greater drinking water pollution. The evidence supports the view that well-functioning insurance markets for municipal debt have significant real effects on public infrastructure.

Abstract: Local governments often seek credit ratings from multiple agencies for external financing. Rating shopping and catering, the incentives of the rating agencies, and the career incentives of the analysts issuing the ratings may render the ratings interdependent. This paper estimates the feedback among the credit ratings as well as the impact of the ratings on a bond's yield. To this end, we put forward a simultaneous equations model with three features. First, the model allows the bond's latent fundamental (i.e. credit worthiness) and observed characteristics to directly influence its ratings and yield. Second, the model allows each credit rating to influence the other ratings. Third, the model allows each credit rating to influence a bond's yield. To tackle the simultaneity and endogeneity in the model, we make use of higher order moments. We report simulation results that support our novel econometric framework. We report several important empirical findings. First, ceteris paribus, an agency increases its rating in response to another agency increasing its rating. Second, in full equilibrium, the feedback improves the overall quality of the ratings, measured by the reliability ratio. Third, a ceteris paribus increase in a credit rating decreases the bond yield, corroborating the literature's previous findings. Last, we document a separate substantive effect of the latent fundamental on the yield.

Abstract: Debtor rights vis-a-vis creditor during bankruptcy  can evolve over time due to changes in the nature of the prevailing bankruptcy law and its practice. I empirically study such a time-series trend in debtor rights using comprehensive sample of U.S. firms. To this end, I develop a dynamic model to estimate debtor rights and  examine its implications. Larger debtor rights make  debtors more willing to default, which increases borrowing costs. In response,  firms  lower leverage ex-ante. This channel helps to match  joint distributions of  leverage and default probabilities. Structural estimation reveals a  change over time:  debtor rights increased from 1.2% to 4.4%  around the Bankruptcy Reform Act of 1978, and has gradually decreased back to zero in the subsequent four decades.