VI. Effects of Energy Price Shocks on the United States Economy1
1. The recent volatility of world energy prices has led to concern regarding the potential adverse effects on the U.S. and world economies. Geopolitical and other factors helped cause world oil prices to roughly double between late 2001 and March 2003, and prices remained elevated into June. Natural gas prices also have risen sharply, nearly tripling from levels in 2000, amid concerns regarding supply constraints, including in pipeline and storage capacity, and rising demand. The already tentative nature of the current U.S. recovery, as well as the adverse effect of previous energy price shocks—including oil price shock of the 1970s, have led many analysts to worry that high energy prices pose continuing risks to U.S. growth prospects.
2. This chapter examines the impact of energy shocks using the IMF’s Global Economy Model (GEM). The GEM model is particularly useful because it permits the analysis of supply as well as demand effects, reflecting the use of energy as both intermediate and final consumption good. The simulation results suggest that the impact of energy price shocks tends to be moderate, especially if price hikes are short-lived and monetary policy responds appropriately.
A. The Modeling Framework
3. GEM is a new open-economy macroeconomic model based completely on a choice-theoretic framework.2 A two-country version of GEM is considered here, comprising the United States and the rest of the world. The model includes four types of goods: (1) energy (oil and natural gas) as a tradable intermediate input; (2) a traded intermediate good; (3) a non-traded intermediate good; and (4) a non-traded final consumption/investment good. Energy is used in the production of both the traded and non-traded intermediate goods and consumed directly in the final good. The model incorporates a distribution sector that uses non-traded goods to deliver energy to its final users. This implies that the retail price of energy changes by less in percentage terms than the producer price of energy.3
4. Energy prices are market determined under monopolistic competition, implying energy firms charge a markup over marginal cost. Energy is produced with capital, labor and land—a fixed factor—using a constant elasticity of substitution (CES) technology. Price shocks can originate from two sources on the supply side: changes in the quantity of land available for use in energy production, and changes in the markup charged by energy firms. With an extremely large elasticity of substitution between domestically produced and imported energy goods, changes in the quantity of land, or in the markup in the rest of the world, lead to identical changes in energy prices for both foreign and U.S. producers.
5. With two central roles for energy in the economy, each with different frictions, energy price shocks affect volumes and prices with different speeds. Because the short-run costs of switching to more energy-efficient production processes are high, profit-maximizing firms respond to increased energy costs by reducing the labor input. The impact on production output is therefore felt relatively quickly. However, competitive pressures are assumed to impose costs on firms that change output prices too rapidly, and the effect on non-energy goods prices is therefore only felt over time. By contrast, energy price shocks have an immediate impact on the consumer price index, and therefore on real wages and household welfare, because energy is consumed directly in the final consumption good.
6. The model was calibrated to reflect U.S. oil and gas usage in 2000. Valued in terms of real producer prices, the consumption of oil and natural gas was set at 2.4 percent of GDP, of which 1.1 percentage points are produced domestically and the remainder are imported. The calibration assumes that roughly half of the energy consumed is used in the production of intermediate goods, and another half in the final consumption good.
B. Model Results
7. The impact of energy price shocks is illustrated by a 50 percent increase in the price of oil and gas. Three alternative durations for the shock are considered: the first alternative has a duration of only one quarter, the second has a half life of one year, and the final alternative has a half life of five years.4 These shocks are induced by changing the markup charged by the energy producers in the rest of the world.5 The model incorporates rational expectations, so that future energy price paths are completely understood by all agents. The responses of several key variables are presented in Figure 1.
Figure 1.United States: Impact of a Fifty Percent Increase in the Producer Price of Energy 1/
In percent or percentage point deviation from baseline 2/
8. Output losses are relatively moderate, including in the case of an energy price shock lasting over several years:
- The simulation results suggest that an energy shock lasting for one quarter reduces real GDP by roughly ¾ percent relative to baseline in the first quarter. However, as the return to the baseline GDP level is almost instantaneous, the long-term impact is negligible. By contrast, effects on output and inflation are longer-lived under a more persistent price shock.6 In the case of the longest lasting shock, the maximum effect on GDP is a drop of roughly 1 percent occurring after three quarters. Over the long-term, however, this impact gradually eases, with GDP 0.4 percent below baseline after ten years.
- Under the long-lived increase in energy prices, the brunt of the adjustment to energy price shocks is borne by consumers, which own companies and therefore hold all external debt in the model. The initial increase in the current account deficit leads to a higher stock of foreign debt, the servicing and eventual repayment of which depresses consumer spending relative to baseline for a sustained period. Investment initially falls because of an increase in the user cost of capital, reflecting tighter monetary policy, as well as a decline in the return to capital due to increased costs and limited ability to raise output prices. With the capital stock below baseline as the oil price shock dissipates, the gap between the return to capital and its user cost reverses and investment spending will increase above baseline until the two are re-equilibrated. Households supply additional savings to fund this investment, further constraining consumption spending.7
- The response of monetary policy is determined by an inflation-targeting monetary policy reaction function, which firmly anchors inflation expectations. Inflation stabilization is aided by the model structure, which assumes nominal wage stickiness (rather than real wage stickiness), i.e., workers do not attempt to maintain real wages even under persistent energy price increases. As a result, CPI inflation increases for a short period initially, but thereafter returns to baseline relatively quickly.
9. However, there are several reasons why these simulations may understate the effects of energy price shocks especially when compared to historical episodes:
- The model assumes that the deterioration in the U.S. current account would be financed by increased borrowing from the rest of the world. However, if the ability to borrow is constrained, including by shifts in confidence or portfolio preferences, U.S. consumption and investment may decline more than these simulation results suggest.
- If the monetary authority was more accommodative of the inflationary impact of the shock, attempting instead to mitigate output effects and, at the same time, workers bargained aggressively to maintain their real consumption wage, the persistent shock could lead to more persistent CPI inflation and longer-lived real output effects.8
- In the simulations, energy-exporting countries—which are included in the rest-of-the-world block—increase consumption in proportion to higher energy revenues. If energy exporters’ saving rates temporarily increased, however, as was the case during past oil price shocks, demand for U.S. exports could be weaker than simulated.
- Historically, large shocks to energy prices have often coincided with significant geopolitical events that may have impacted on investor and consumer confidence; however, the energy price shocks considered here have no such effects.
10. The simulation results presented here suggests that the impact on U.S. growth of temporary energy price shocks should be mild. A short-lived spike in energy prices, such as occurred during the past year, would have a modest and short-lived impact on growth. The simulations also suggest that even in the face of expectations that a shock would be longer-lived, the impact on growth would be only marginally larger. However, the results have to be interpreted with caution, in part because the analysis does not incorporate confidence effects of the kind that have accompanied oil price shocks in the past.