The Determinants of the World Price of Crude Oil: An Empirical Analysis

Crude oil prices have recently increased considerably due to enhanced global demand and political instability in the Middle East and North Africa. This paper examines the factors that contribute to the short-run and long-run trends in the price of crude oil with particular emphasis on those trends that can be assessed quantitatively. The world price of crude oil is fundamentally determined by the interaction between the market forces of supply and demand of crude oil. This price can also be influenced by the market power of major oil-producing countries and political instability and conflict. 

In this study, an econometric analysis of the determinants of crude oil prices is performed. First, a simultaneous equation model, comprised of models of supply and demand in the market for crude oil, is utilized to establish the price as an independent variable and test for its elasticity. The US crude oil price (US dollars per barrel) is a function of many variables such as days of forward consumption of the Organization for Economic Cooperation and Development (OECD) crude oil stocks, the Organization of the Petroleum Exporting Countries (OPEC) production quotas, the OPEC’s tendency to cheat on its quota, and the capacity utilization of OPEC.  The quantitative impact of these variables is then examined by utilizing a regression model as well as a co-integration test. 

All independent variables were expected to have a significant impact on the dependent variable, the world crude oil price. We expected a negative estimated sign for the variables of OPEC quota and tendency to cheat, a positive sign on capacity utilization, while the other variables’ signs were ambiguous. The method of ordinary least squares (OLS) was used to estimate the regression model. Further, the Augmented Dickey Fuller test was performed to see whether a unit root is present in our time series sample. A co-integrating relationship was found between the variables in the first specification.  The existence of cointegration between the variables indicated biased estimates in the regression model.  Thus, the model was transformed accordingly, and another OLS regression was conducted.  Estimation of the new equation provided that all signs were as hypothesized, and the coefficient for the days of forward consumption was negative.  The empirical results of the model were consistent with those in the literature; the variables included in the model significantly influenced the crude oil prices.