Macroeconometric Studies of Private Consumption, Government Debt and Real Exchange Rates

Sammanfattning: Advances in time series analysis during the last two decades have stimulated research in a number of areas in macroeconomics. This thesis is a compilation of five essays using cointegrated vector autoregressive (VAR) models, unit root tests and regime switching models to investigate the behavior of private consumption, public debt and the real exchange rate. Chapter two explores the empirical evidence of the rational expectations permanent income hypothesis (PIH) in Sweden using a bivariate VAR model of labor income changes and saving. The data firmly reject the PIH and an extension of the model where a fraction of the consumers are supposed to use a rule of thumb: consume all the disposable income. The deviation from the behavior predicted by the PIH is, however, numerically small. Increased saving predicts a declining labor income and saving moves closely with “theoretical” saving. In order to explore the reason for the statistical rejection, a simple model of habits in consumption is introduced. The restrictions implied by this model of consumer behavior are generally not rejected. Chapter three is a short note on how to calculate permanent and transitory innovations in a vector time series from a cointegrated VAR model. This method is used in chapter four, where a structural common trends model of private consumption and disposable income growth is estimated and used to analyze the Swedish private consumption behavior. The conclusion is that Swedish consumers are forward-looking, adjusting consumption expenditures to the long-run level of disposable income very fast. Chapter five provides international evidence of the sustainability of the public debt. We test if the public debt-to-GDP ratio is stationary in six countries over the period 1885-1996. Standard unit root tests suggest that this is not the case, but when controlling for temporary government expenditures, e.g. military expenditures during wars, and temporary shortfall of taxes due to temporary low GDP, we find that the debt-to-GDP ratio is stationary in four out of the six countries. If we, furthermore, allow for a structural break in 1973, the debt-to-GDP ratio is stationary in all six countries. Chapter six investigates the behavior of the real exchange rate. We suggest that the real exchange rate between the major currencies in the post-Bretton Woods period can be described by a stationary, two state Markov switching AR (1) model. Based on the out of sample forecast performance, we find that this model out-performs two competing models where the exchange rate is non-stationary.