Essays on a new Keynesian perspective for Japan

Mode of access: World Wide Web.Thesis (Ph. D.)--University of Hawaii at Manoa, 2005.Includes bibliographical references (leaves 77-82).Electronic reproduction.Also available by subscription via World Wide Webix, 82 leaves, bound ill. 29 cmThis study looks at dynamics in Japan's aggregate demand and supply over the period, 1972-2003, using the New Keynesian/New Neoclassical Synthesis model. This model incorporates sticky prices to the optimization problems of households and firms within a context of rational expectations. Overall, the results are similar to those using U.S. and Euro area data. The aggregate supply model, also known as the new Keynesian Phillips curve predicts inflation depends on next period's inflation and a measure of real marginal cost and allows for the estimation of the degree of rigidity and subjective discount rate, both structural parameters. The results suggest the new Keynesian Phillips curve is a reasonable approximation of Japan's inflation, especially when the model is extended to allow for inflation inertia. Specifically, the results predict Japan firms are forward looking and adjust prices every 2-3 quarters. Structural stability tests indicate parameter breakpoints occur in 1990 or 1996; near the end of the asset price bubble or start of the prolonged deflationary period. Aggregate demand is determined by future real output and real interest rates with the latter permitting evaluation of the impact of monetary policy. However, dissimilar to the new Keynesian Phillips curve, the Euler equation for output does not have much explanatory power as a theory of aggregate demand for Japan. Although allowing for output inertia and an open economy influence through the exchange rate channel predicts forward looking behavior dominates, the impact of real interest rates is effectively zero. Throughout this thesis, the empirical evidence suggests the Generalized Method of Moments is sensitive to the way the orthogonality condition is written and this is consistent with evidence on the U.S. and Euro area. However, for Japan, the variation in parameter magnitudes across equivalent moment conditions is much greater