chapter  3
Oil Supply Disruptions and the Optimal Tariff in a Dynamic Stochastic Equilibrium ModelZvi Eckstein and Martin S. Eichenbaum
Pages 29

The importance of petroleum as a source of energy to the United States and other industrialized economies is by now a matter of common knowledge. Partly in response to the Mideastern oil cutoffs since 1973, as well as the recognition of the U.S. economy’s dependence on foreign sources of oil, many economists have advocated a variety of government policies aimed at mitigating the effects of disruptions in the international oil supply network. Examples of such proposals, which range from tariffs on imported oil to government-held strategic petroleum reserves, may be found in Nichols and Zeckhauser (1977), Nordhaus (1974), Teisberg (1981), Tolley and Wilman (1977), and Wright and Williams (1982), among others. In essence, all of these proposals revolve around the notion that the private sector of the economy is somehow incapable of dealing in a socially optimal way with shocks to the supply of oil.The purpose of this chapter is to examine the welfare-enhancing role of government interventions in decentralized economies that are subject to uncertainty regarding the price of oil. In order to do so, we first construct

a stochastic general equilibrium model of an economy that imports energy. The economy consists of a finite number of infinite-lived con­sumers and firms. Consumers maximize the expected discounted utilityderived from leisure and consumption, while producers maximize the expected present value of profits. Output of this economy consists of a single good, which may be used for consumption or capital. The produc­tion function of the good in question depends on three inputs, labor, capital, and energy, and is subject to costs of adjustment in capital. In order to emphasize the transmission of oil shocks into domestic labor and capital markets, we restrict uncertainty in this economy to the pricing of oil.The nonintervention equilibrium is characterized for a fairly general parameterization of the stochastic process that governs the evolution of energy prices over time. For the model under consideration, the rational expectations competitive equilibrium is optimal if, and only if, at each time period the conditional elasticity of world energy prices with respect to domestic imports is equal to infinity. Put somewhat differently, there is no room for welfare-enhancing government interventions if, and only if, aggregate domestic imports have no effect on current and future world energy prices. Essentially, this finding follows from the fact that competi­tive agents view those aggregate domestic state variables that influence the distribution of energy prices as being beyond their ability to control or influence. As a result, the latent ability of the country as a whole to influence energy prices goes unexploited. A social planner, however, when determining optimal contingency plans, would take into account the country’s ability to influence the equilibrium distribution of energy prices. Thus the standard nonoptimalities involved in static optimal tariff arguments can be viewed as a special case of the nonoptimalities in the present dynamic, stochastic environment.Given the general nonoptimality of the competitive equilibrium, we proceed to analyze the set of interventions that support an optimal equilibrium. When only current aggregate imports affect energy prices, a dynamic optimal tariff that exploits the country’s ability to influence both the first and second moments of the energy price process is sufficient to support an optimal equilibrium. When current energy prices depend on lagged as well as current imports, however, an optimal tariff must be combined with an import tax for which firm-specific tax rates are fixed functions of lagged import levels. Thus by searching over alternative sets of policy instruments, it is possible to support a decentralized equilibrium that reproduces the unique time-consistent optimal solution of the rele­vant Pareto problem. This approach is in contrast to the situation that typically emerges when optimal policy is conceived as corresponding to the best use that can be made of a set of prespecified policy instruments.