Tax Policy Analysis in a Flexible Computable General Equilibrium Model: Applications to Energy and Gross Receipts Taxation
Barbe', Andre' Jean-Curtis
Zodrow, George R.
Doctor of Philosophy
In this paper, I construct a new general equilibrium model of the United States economy that is better able to analyze energy and gross receipts taxes than previous models. Existing models in the energy literature fall into two groups: general equilibrium models of the entire economy with exogenous energy resource supply and partial equilibrium models of the energy sector with endogenous resource supply. I combine the main advantages of these two strains of the literature by incorporating endogenous resource supply in a computable general equilibrium model with highly disaggregated and flexible industry cost and consumer expenditure functions. My new model is able to analyze all the major inefficiencies caused by energy taxation, i.e. those related to production, consumption, resource rents, and externalities. In addition to its application in energy, my model is also ideal for looking at gross receipts and retails sales taxes. Gross receipts and retail sales taxes are important revenues sources for most US states and share many of the same issues as energy taxes. Retail sales taxes are commonly viewed as more efficient than gross receipts taxes because the latter apply to intermediate goods and thus result in production and consumption inefficiencies. However, in reality the retail sales taxes used by the US states are not pure consumption taxes, but tax many intermediate inputs while exempting many consumption goods. My model determines whether retail sales taxes are still more efficient than gross receipts taxes when these realistic factors are included. As an application, I use the model to analyze two tax reforms for energy or gross receipts taxes. First, President Obama's 2014 budget proposes to reform energy taxation by eliminating fossil fuel tax preferences. I find that the budget's tax increases for fossil fuels increase household welfare if the social cost of carbon emissions is over $15 per ton but otherwise reduce welfare. Second, I also use the model to examine a tax reform that replaces a typical retail sales tax with a generic gross receipts tax. Contrary to the conventional wisdom, I find that the gross receipts tax is more efficient than the retail sales tax, with an efficiency cost that is 6.8 percent of revenues less than that of the retail sales tax. These results demonstrate that the predicted impacts of the tax reforms are significantly altered by the features included in my model: general equilibrium effects, flexible substitution, resource rents, and externalities.
Computable general equilibrium; Tax policy; Resource rents; Translog; Fossil fuels