Essays in monetary and fiscal policy

The dissertation topic is "Three Essays in Monetary and Fiscal Policy". The three articles are: (1) Interaction between Monetary and Fiscal Policy in the U.S.; (2) Modeling and Forecasting the U.S. Federal Budget Deficit; (3) Causality between Government Expenditure-Revenue: Evidence from Symmetric and Asymmetric Modeling.; Our research in the first essay has examined the dynamic interaction between monetary and fiscal policy in the U.S. The empirical work undertaken in this essay is based on estimating the Unrestricted Vector Autoregression (UVAR) model and plotting the generalized impulse response functions (GIRF). Four macroeconomic variables macroeconomic are used. They are: Inflation rate, real output growth, federal fund rate, and government budget deficit. By tracing the dynamic effects of contractionary monetary policy and expansionary fiscal policy, the research concludes that the conduct of monetary policy has been more responsive to fluctuation in output and inflation and less responsive to fluctuation in deficit. Additionally, the estimated results obtained here provide no evidence for the proposition that budget deficit has been an important contributing factor to inflation and interest rate.; In the second essay we have examined the efficiency of out-of-sample forecasting performance of several time series models for the U.S. federal government deficit. The empirical work undertaken in this paper is based on modeling the federal government deficit as a univariate autoregressive moving average (ARMA) process as well as a combination of autoregressive moving average process plus autoregressive conditionally heteroscedastic disturbances processes (ARMA-ARCH-type). Our empirical research has specified the deficit process as an ARMA (3, 0). With the presence of ARCH effect, our empirical research arguing that improving ARMA forecasts for the deficit may be possible through the use of ARMA-ARCH-type models.; The combinations of ARMA-EGARCH models did appear to improve the forecasting performance over single ARMA and random walk models. In fact, relative to single ARMA models, the ARMA-EGARCH models produce the forecasts with the lowest in all forecasts error criteria and at all forecast horizons. The random walk model when estimated with EGARCH (1, 1) has the strongest performance in short-run forecasting.; Our research in the third essay takes a closer look at the directional causality between government revenues and government expenditures. The main focus is to provide new evidence by applying the asymmetric modeling. Thus, we relaxed the implicit assumption of a symmetric adjustment process underlined in the conventional cointegration and Error Correction Model (ECM). The threshold autoregression and momentum threshold autoregression developed by Enders and Granger (1998) and Granger and Siklos (2001) are the underlying models for such asymmetric process.; Our research concludes that only revenues respond asymmetrically to departures from the budgetary balance, i.e. revenues respond to negative not to positive budgetary disequilibrium. Additionally, a significant long-run causal effects running from revenues to expenditures is also found.