Estimating the Demand for Energy in Jordan: A Stock-Watson Dynamic OLS (DOLS) Approach

Recent developments in econometrics provide robust estimators for cointegrated series where sample sizes are small. We estimate the demand for energy in Jordan over the period 1968-1997 using the dynamic OLS method developed by Stock and Watson, and compare the results with conventional forms of cointegration and error correction estimation. Results are found to be robust to various departures from standard regression assumptions and to be stable of the rapid structural changes in the Jordanian economy over the period of the study. Income, construction activity, and political instability are found to impact significantly on consumption, while real price has only a neutral or weak effect. Changes to energy prices on their own are unlikely to achieve current goals for energy conversation.