| This dissertation focuses on the consumption-based asset pricing models developed by Lucas (1978) and others in the area of financial economics. The first chapter studies the effect of a change in aggregate economic risk and uncertainty on the prices of bonds and stocks. A decomposition method of the dividend and discount rate effect, which is similar to the traditional income and substitution effect, is defined and utilized. While it unambiguously raises bond prices, an increase in risk does not necessarily decrease stock prices. Sufficient conditions on preferences are specified such that an increase in risk does guarantee a fall in the prices of stocks. In the second chapter, an empirical study is carried out to examine whether Bayesian learning can help the Lucas-type models predict the low levels of short-term real interest rates. The results do not support the hypothesis that parameter uncertainty alone can resolve the risk-free rate puzzle. In particular, the learning process ends so fast that the uncertainty about parameters hardly plays a role in affecting short-term interest rates. The third chapter investigates whether the downturns of business cycles have caused the falls of real interest rates, based on the observation that real interest rates typically experienced troughs during historical recessions in the US economy. A standard consumption-based asset pricing model, with an added feature that investors have to learn about the unobservable alternation of business cycles, is examined and calibrated. The simulation technique of the Markov Chain Monte Carlo is used to estimate the model and compute real interest rates. Although it appears to match the dynamics of real interest rates prior to Year 1980, the model completely misses the drastic increase in real rates in that year, in which it was believed that the quick tightening of the monetary policy by the Federal Reserve was responsible for this rate jump. It is therefore concluded that asset pricing models without a monetary perspective are difficult to capture the dynamics of the real interest rates in the data of the US economy. |
