Search

We find that, when estimated, a two sector computable dynamic stochastic general equilibrium open economy model of the U.S. that formally admits energy into the production process can generate plausible parameter values that can be applied to deal with a broad range of economic issues. As a benchmark, we require that the model fits the data for output, real exchange rate, energy use, and consumption: output because it serves as a measure of a country's total income,real exchange rate because it serves as a determinant of a country's relative competitiveness,energy use because it serves as an indicator of special inputs into a country's production process,and consumption because it serves as a yardstick for evaluating a country's standard of living. Finally, we argue that this model, with appropriate extensions, some of which we also propose, can help future modelers to tackle other research questions.

This thesis represents largely the two sides to both theory and econometrics of dynamic\ud macroeconomics, namely stationary and non-stationary models and data. The stationary\ud part concludes with Chapter 3 and in Chapter 4, I look at the non-stationary side.\ud More specifically, I preview the thesis in Chapter 1 highlighting the modelling and\ud econometric approaches commonly found in the economics literature; also I report some\ud key results. In Chapter 2, I provide a comprehensive, but certainly far from being exhaus-\ud tive, review of the literature dating back to the publication of Stanley Jevon's (1866) The\ud Coal Question, but with the main discussion beginning with Harold Hotelling's (1931) The\ud Economics of Exhaustible Resources. I develop a two-sector open economy extension to\ud the Kydland and Prescott (1982), Long and Plosser (1983) and Kim and Loungani (1992)\ud models in Chapter 3 and estimate it on H-P �ltered annual U.S. data covering 64 years,\ud with the main purpose of discovering how energy price along with other supply-side and\ud demand-side shocks (imported and domestic) impacts on the U.S. economy. The model\ud presented only contains the current account and I restrict trade to balance in every pe-\ud riod. I find that model fits the data for my benchmark variables of interest in the auxiliary\ud model: output, real exchange rate, energy use, and consumption. When more variables\ud and in particular sectoral variables are added, meanwhile, to the auxiliary model, I find\ud that the model's performance especially as it relates to this estimated model parameters\ud did not fit. What I take from this is that the estimated structural parameters are not\ud globally applicable within this economic environment.\ud This model is then further extended by including the capital account in Chapter 4\ud before re-estimation, but now also on non-stationary data, which I suppose is more repre-\ud sentative of reality. I focus on the fit of the model to output and the economy's measure\ud i\ud of competitiveness: the real exchange rate. I find that the energy price and technology\ud shocks have major e¤ects on the U.S. output and relative competitiveness. The mecha-\ud nisms by which these e¤ects are transmitted are two-fold. First is via the terms of trade\ud occurring as a resource drain on the economy as the U.S. would need to find extra resource\ud to commit to the import of crude oil. The second is via household's reduced investment\ud activity. Both channels can be explained by the fact that the substitution away from oil\ud is happening at too slow a pace because of low estimated elasticities parameters. This\ud agrees with Hamilton who argued that oil shock works via demand contraction. I have in\ud this thesis verified his conjecture via a well-motivated and detailed microfounded dynamic\ud stochastic general equilibrium (DSGE) model.\ud Finally, I review the thesis speculating on possible future extensions in Chapter 5.