Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)

Graduate Group


First Advisor

Joao Gomes

Second Advisor

Michael Roberts


This dissertation is motivated by the 2008 financial crisis and consists of three chapters. In the first chapter, I show that the cooperative objective of credit unions enabled them to lend significantly more than profit-maximizing banks during the Great Recession. Loan growth rates were higher for the $1.3 trillion credit union industry by as much as 10 percentage points at the peak of the crisis. Using a newly constructed database containing balance sheet information and loan-level activity, I compare institutions that faced identical borrowers in the same local credit markets and control for crises exposures to show that the effect is supply-driven. Further, the lending difference was sustained by 15-20 percent lower profit margins. Loan pricing, informational advantages, taxes, or the regulatory environment do not explain the results. Rather, member-oriented objectives precluded the slow economic recovery of credit unions after the financial crisis.

In the second chapter, which is joint work with Yasser Boualam, we document that higher measures of liquidity risk on banks balance sheets are associated with lower expected stock returns. We first calculate a measure of liquidity risk which reflects how much of a bank's volatile liabilities are covered by its stock of liquid assets. We show that the standard factor models do not fully explain the cross section of bank stock returns. A portfolio that is long in low liquidity risk banks and short in high liquidity risk banks delivers a statistically significant alpha of 6 percent annually. This effect is not driven by bank characteristics such as size, profitability, leverage, or asset quality, but appears to be partly connected to the degree of bank complexity and potential valuation errors pre-crisis.

In the third chapter, I compare traditional banking and shadow banking based on the type of security that funds them: money funds the former while money-like liabilities fund the latter. I show the following key findings. First, using a parsimonioius two-sector model of non-balanced growth that captures structural changes within the financial sector, I measure the time series relative productivity of shadow banking to traditional banking. In the 1960's shadow banking sector productivity started at around 0.6 relative to traditional banking, which peaked to 1.2 starting in the 2000's. Second, growth in money-like liabilities lag growth in output, contrary to the well-known leading relationship between traditional monetary aggregates and output growth.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."