Oil Supply Shocks, US Gross Domestic Product, and the Oil Security Premium

Nov 30, 2017 | Alan J. Krupnick, Richard D. Morgenstern, Nathan Balke, Stephen P.A. Brown, Ana María Herrera, Shashank Mohan


This report examines estimates of and approaches to quantifying the costs of US dependence on oil, finding that the oil security premium is likely lower than reported in the existing literature.


The macroeconomic costs of unanticipated oil supply and oil price shocks remain the principal component of the oil security premium. A long history of academic papers have offered approaches to the estimation of such costs and the calculation of the oil security premium. Two relevant major changes have occurred in recent years: both the US economy and the world oil market are now more resilient, less dependent on oil in general, and (for the United States) less reliant on imports than a decade or two ago; and macroeconomic modeling has become more sophisticated, with advances coming from modeling dynamic economic relationships, using dynamic stochastic general equilibrium (DSGE) models, and extracting macroeconomic oil price shocks from time series data, using structural vector autoregression (SVAR) models. These advances suggest it is time to use sophisticated modeling tools to take another look at the macroeconomic effects of price shocks. In addition to using the DSGE and SVAR models, which are estimated directly from historical data, we also exercise the National Energy Modeling System (NEMS) model and perform a number of sensitivity analyses with all the models to check for the robustness of their estimates. This report develops new estimates of the relationship among gross domestic product (GDP), oil supply and price shocks, and world oil demand and supply elasticities; translates them into oil security premiums using a welfare-theoretic-based computation model; and compares all these estimates with those in the literature. The literature is divided into three categories: older studies, newer ones, and a mixture of old and new.