We use arbitrage activity in equity, fixed income, and foreign exchange markets to characterize the frictions and constraints facing intermediaries. The average pairwise correlation between the 32 arbitrage spreads that we study is 22%. These low correlations are inconsistent with canonical intermediary asset pricing models. We show that at least two types of segmentation drive arbitrage dynamics. First, funding is segmented—certain trades rely on specific funding sources, making their arbitrage spreads sensitive to localized funding shocks. Second, balance sheets are segmented—intermediaries specialize in certain trades, so arbitrage spreads are sensitive to idiosyncratic balance‐sheet shocks.

When should private companies allocate public resources? In our model, delegation is attractive when delay is costly, the impact of funds is similar across firms, and government and private objectives are aligned. We use novel firm-level survey data to measure heterogeneity in the impact of the Paycheck Protection Program and to assess whether banks targeted loans to high-impact firms. Banks did target loans to their most valuable pre-existing customers. However, we find that treatment effect heterogeneity is moderate, suggesting that delegation was likely superior from the government’s perspective to delaying loans to improve targeting.

We study links between the labor market for loan officers and access to mortgage credit. Using novel data matching mortgage applications to loan officers, we find that minorities are underrepresented among loan officers. Minority borrowers are less likely to complete mortgage applications, have completed applications approved, and to ultimately take up a loan. These disparities are reduced when minority borrowers work with minority loan officers. These pairings also lead to lower default rates, suggesting minority loan officers have an informational advantage with minority borrowers. Our results suggest minority underrepresentation among loan officers reduces minority borrowers’ access to credit.