Research

This paper explains and quantifies the secular decline of financial covenants in private debt markets. Using a panel of U.S. syndicated loans from 1996 to 2023, I document that the annual incidence of covenant violations declines from 20% to below 5% over two decades, while the creditor intervention rate following a violation rises from 40% to over 70%. Crucially, this decline is accompanied by a reallocation across the micro-channels of creditor control: managerial discipline has weakened considerably while investment conservatism have become more important. I develop a micro-founded optimal contracting model that endogenizes covenant design through moral hazard, endogenous signal acquisition, and the contingent allocation of control rights. I structurally estimate the model using the Simulated Method of Moments, targeting causally identified moments from the reduced-form literature. The estimates reveal three forces behind the secular decline: a moderate easing of moral hazard frictions, improved signal informativeness, and a shift in the relative importance of the micro-mechanisms of creditor intervention. Together, these shifts characterize financial covenants as a fading tripwire: an alarm mechanism that is increasingly rarely set and functioning differently when triggered.

This paper uncovers a novel contractionary channel through which quantitative easing (QE) affects corporate financing: the loan renegotiation channel. Using loan-level data from 2006 to 2024, we exploit cross-sectional variation in banks' pre-QE MBS holdings interacted with Federal Reserve purchase flows to identify causal effects on existing credit relationships. Within a firm-quarter design that absorbs borrower demand shocks and lender heterogeneity, we find that banks most exposed to QE purchases significantly increase lender-favorable renegotiations, tightening credit terms through commitment reductions, spread increases, maturity shortening, or additional collateral, while borrower-favorable amendments remain unaffected. High MBS exposure raises the probability of lender-favorable renegotiation by approximately 2.8%. Aggregating renegotiation exposure to the firm level, IV estimates indicate that a one-unit increase in exposure-weighted lender-favorable renegotiation reduces capital expenditure by 12.2% of lagged total assets. These findings establish loan renegotiation as a distinct intensive-margin mechanism through which unconventional monetary policy propagates to firm investment.

Why do financial contracts leave entrepreneurs exposed to aggregate risk even when contracts can be made fully state-contingent? I develop a financial accelerator model with a moral hazard problem in which private benefit of low effort is non-monetary and incentive compatibility holds state by state. Since the marginal value of entrepreneurial net worth is countercyclical, the incentive constraint tightens procyclically. Entrepreneurs take advantage of this tradeoff by offloading more idiosyncratic risk at the cost of taking on more aggregate risk. Imperfect aggregate risk exposure thus emerges endogenously from the interaction between state-dependent incentives and risk sharing, rather than from contractual restrictions. This exposure amplifies aggregate shocks: entrepreneurial net worth becomes more sensitive to macroeconomic conditions, compressing borrowing capacity procyclically. Quantitatively, state-contingent contracting does not eliminate the financial accelerator, and the model generates substantial and persistent amplification of investment and output.