Asset Prices, Banking and Economic Activity

Abstract

This dissertation examines the role of asset prices to act as a transmission and amplification mechanism. Specifically, it looks at how changes in asset prices can help transmit and amplify technology shocks through the credit channel by changing the supply of loanable funds, or changing the supply of deposits, or both. Using a modified version of the Kiyotaki-Moore credit cycles model with concave utility and decreasing returns to scale production function, the dissertation illustrates that asset prices can as a credible amplification and transmission mechanism. Using concave utility and decreasing returns to scale production function allows the incorporation risk aversion into the credit cycles model. The model can help explain the gap between observed magnitude of shocks, and the corresponding changes in economic activity. The behavior of a heterogeneous agent economy in response to a technology shock is simulated using computer programs. The simulations show that a one percent technology shock translates into a more than four percent change in capital held by the constrained agents by moving capital from one agent type to the other. This moves the economy away from a first-best equilibrium. If the technology shock is positive there is an increased demand of capital from the more productive agents, and thus a more than proportionate increase in output. If the technology shock is negative, the opposite path is followed, and economic activity falls more than proportionately. There are credit constraints built into the model. Agents' access to credit is determined by the value of collateral on oer, which in turn depends on asset prices. Technology shocks change demand for assets, their prices, their value as collateral, and hence agents' access to credit. Further, since prices are forward looking, a shock in one period propagates through time. These simulations show that the effects of the shock can be felt up to 13 periods after it has hit. An event analysis with housing price data from 18 countries spanning a period of more than four decades is also performed. It shows that there is strong co-movement of housing prices and economic activity. In particular, larger changes in housing prices have been accompanied by qualitatively similar changes in economic activity. The period leading up to the peak of a real estate cycle is accompanied by a more than proportionate increase in private sector lending, and once the peak has been crested, there is a more than proportionate fall in nominal private sector lending. This evidence is in sync with the earlier observation that changes in asset prices influence agents' access to credit and contribute to the persistence of the effects of the shock far into the future. Further, the preferred measure of economic health, the rate of inflation, sees no measurable change in periods leading up to a real estate peak, and beyond. This throws up the need for some other measure of economic health that is better able to capture the events in asset markets. Policy makers have been paying more attention to this channel in the aftermath of the sub-prime mortgage crisis in the United States. There have been multiples changes in regulatory policy across the world, and specific steps are being taken to dampen exuberance in the real estate market. Only time can tell if these measures turn out to be effective, but at least a step has been taken towards realizing that housing market can lead to a wider economic and banking crisis.