This article investigates the effect of bank lending frictions on employment outcomes. I construct a new data set that combines information on banking relationships and employment at 2,000 nonfinancial firms during the 2008–9 crisis. The article first verifies empirically the importance of banking relationships, which imply a cost to borrowers who switch lenders. I then use the dispersion in lender health following the Lehman crisis as a source of exogenous variation in the availability of credit to borrowers. I find that credit matters. Firms that had precrisis relationships with less healthy lenders had a lower likelihood of obtaining a loan following the Lehman bankruptcy, paid a higher interest rate if they did borrow, and reduced employment by more compared to precrisis clients of healthier lenders. Consistent with frictions deriving from asymmetric information, the effects vary by firm type. Lender health has an economically and statistically significant effect on employment at small and medium firms, but the data cannot reject the hypothesis of no effect at the largest or most transparent firms. Abstracting from general equilibrium effects, I find that the withdrawal of credit accounts for between one-third and one-half of the employment decline at small and medium firms in the sample in the year following the Lehman bankruptcy.

We document the importance of covenant violations in transmitting bank health to nonfinancial firms. Roughly one‐third of loans in our supervisory data breach a covenant during the 2008 to 2009 period, allowing lenders to force a renegotiation of loan terms or to accelerate repayment of otherwise long‐term credit. Lenders in worse health are more likely to force a reduction in the loan commitment following a violation. The reduction in credit to borrowers who violate a covenant can account for the majority of the cross‐sectional variation in credit supply during the 2008 to 2009 crisis.

We provide evidence of the stock market consumption wealth effect by using a local labor market analysis. An increase in local stock wealth driven by aggregate stock prices increases local employment and payroll in nontradable industries and in total, with no effect on employment in tradable industries. In a model of geographic heterogeneity in stock wealth, these responses imply an MPC of 3.2 cents per year and that a 20 percent increase in stock valuations, unless countered by monetary policy, increases the aggregate labor bill by at least 1.7 percent and aggregate hours by at least 0.7 percent two years after the shock.