Anmerkungen:
In: Ph.D. Dissertation, University of Wisconsin – Madison, 2001
Nach Informationen von SSRN wurde die ursprüngliche Fassung des Dokuments November 11, 2001 erstellt
Beschreibung:
The first two chapters of this thesis propose new time-series methods and apply them to macroeconomic problems, while the third chapter evaluates the predictions of a dynamic general equilibrium model. The first chapter develops a practical log-linear aggregation procedure, which is applied to the heterogeneous growth problem in the U.S. The second chapter presents a simple nonparametric long-run correlation estimator with optimal lag-selection and alignment criteria, and uses it to measure interconnections between American and Latin-American stock returns. The third chapter uses a dynamic general equilibrium model to analyze the effects of bank account debits taxation. Time-series techniques are employed to empirically evaluate the model predictions. In the first chapter, a practical aggregation method for heterogeneous log-linear functions is presented. Inequality measures are employed in the construction of an exact representation of the aggregate behavior of an economy formed by heterogeneous log-linear agents. The exact aggregate representation is relatively simple and intuitive. It can be used thereafter in applied issues and in teaching, easing the solving and understanding of aggregation problems. Three macroeconomic applications are discussed: the aggregation of the Lucas supply function, the time-inconsistent behavior of an egalitarian social planner facing heterogeneous discount rates, and the case of a simple heterogeneous growth model. The latter application, which leads to a decomposition of growth rates of the mean into means of growth rates plus inequality changes, is explored empirically. Aggregate CPS data is used to show that, when inequality changes are taken in consideration, the slowdown that followed the first oil shock appears to be worse than usually thought. Additionally, the “new economy” growth resurgence seems less impressive when compared to the growth performance of the period that preceded the first oil shock. In the second chapter, a simple consistent nonparametric estimator of the long-run correlation between two variables is proposed, based on the estimation of the bivariate k-lag difference correlation. It is shown that the estimator is asymptotically equivalent to the Bartlett kernel spectral estimator of the complex coherency at frequency zero. The asymptotic distribution is derived, with a test for the absence of long-run correlation. Optimal lag-selection and alignment criteria are presented. Monte Carlo experiments show that the asymptotic approximations are satisfactory, sometimes even for small samples. They also reveal that the lag-selection and alignment criteria are effective. Long-run correlations between American and Latin-American stock returns are considered. The estimates increase substantially in the second half of the nineties. The results could indicate the presence of a correlation component common to Latin-American markets, which was important in the second half of the period but not in the first. The significant development of investment funds specialized in Latin-American markets and the much-improved foreign access after capital account liberalization in the region may be among the explanations for these patterns. The third chapter uses a dynamic general equilibrium model to study the economic effects of bank account debits (BAD) taxation. Australia and various Latin-American countries have levied or levy BAD taxes. Theoretical aspects such as tax cascading, financial disintermediation, market illiquidity, impacts on dividend and interest rates, tax revenue, government deficit, and effective rates on final transactions are considered. The Brazilian BAD tax (CPMF) experience is evaluated. The empirical analysis shows that revenue productivity appears to be very sensitive to the tax rate, engendering a Laffer curve. It is also shown that there may be impacts on real interest rates. Part of the BAD tax revenue can be lost due to increased interest payments on government debt. Furthermore, the deadweight losses seem to be significant if compared to revenues. Theory and evidence indicate that the BAD acronym is perhaps more than a witticism