Credit boom or credit crunch? - A general-to-specific approach to modeling bank lending

Abstract: In this thesis factors affecting bank lending in the United States are sought for. For this purpose variables that can be derived from prevailing theories and previous research are used. These include for example the interest rate, the nonperforming loans ratio and the capital to asset ratio. The data series cover the time period 1988Q1-2009Q3. Regression models in a-general-to-specific framework together with other econometric methods are applied to attain the results. The nonperforming loans ratio followed by the interest rate, excess reserves to total reserves and house prices are found to be the most important variables in explaining bank lending. Another finding is that banks allocate more loans to loan categories that are less risky than the average bank loan.