| Summary: | 博士 === 臺灣大學 === 國際企業學研究所 === 98 === The Ph.D. dissertation is a collection of three essays on money and the term structure of interest rates. Chapter 1 examines the predictive power of the term structure on real output growth. We first incorporate 900 models into the alternative set, and then test the predictive power and identify superior models by a new large-scale multiple test, the stepwise superior predictive ability test. The result shows that the term structure has its inherently predictive power in predicting the U.S. real GDP growth rates. In addition, all identified superior models consist of a short-term rate and a term spread. This suggests that both short-term rates and term spreads contain useful information about future economic activities. In particular, for predicting annual growth rates, the best model depends on the short-term rate of 3-month Treasury bill and the term spread between this short rate and the yield of 5-year Treasury note.
In Chapter 2, we develop a search model to explain the popular bartering activities of transitional economies in the 1990s. By introducing a simple concept of consumption interdependence into the model, we show that an economy with higher consumption interdependence would be in a barter equilibrium during hyperinflation. In contrast, an economy with an extremely low degree of consumption interdependence is always in a monetary equilibrium. The intuition is that when individuals strongly need others’ goods, they are better-off with direct bartering during hyperinflation. On the other hand, when individuals only need few others’ goods, they always accept money in order to trade some basic goods for living, even though the purchasing power of money is extremely low. Moreover, the model also contains a possibility that agents may switch from using money to bartering as the inflation rate is low, and this is the case of Russia during 1995 and 1997. In addition, the gains from using money can also be documented in the model by reducing the bargaining power of individuals.
In Chapter 3, we construct a term structure in a bank run model to investigate the strategic behaviors of depositors under different real interest rates. As a bank consists of some dominant depositors, one depositor’s consideration for others’ action could be properly modeled by the symmetric mixed-strategy Nash equilibrium. By allowing depositors to use the symmetric mixed strategy, the probability of early withdrawing could be constructed in the equilibrium. Our simulation shows that a low short-term real interest rate, a high long-term real interest rate, or a small number of depositors would raise the probability, and hence the banking system would be more unstable. This result is consistent with recent empirical evidences. Therefore, when the monetary authority cuts short-term interest rates to stimulate the economy, it should also consider the adverse effect on raising the probability. On the other hand, policy makers could reduce the probability by inducing banks to diversify their liabilities.
|
|---|
