Society of Petroleum Engineers 6200 North Central Expressway Dallas, Texas 75206

THIS PAPER IS SUBJECT TO CORRECTION

American Institute of Mining, Metallurgical, and Petroleum Engineers, Inc.

Introduction

Perhaps my early exposure to forecasting as an Air Force meteorologist was habit forming, but for one reason-or another I have been involved in long-term mineral forecasting almost continuously for nearly thirty years. During that time I have progressed from a master's thesis on forecasting future mineral demand, to a post-doctoral study at Penn State on future energy sources for Pennsylvania, to seven years in Denver as Director of the Future Gas Requirements Agency, to my last three years with Resources for the Future. I should check my high school yearbook and see if it listed me as a young man with a future.

I would like to spend a little time reflecting upon a few interesting characteristics I have noticed in the various types of projections of future supply and demand, both for energy as well as for the individual fuels. Following that, I will examine the current type of energy "outlook" and discuss why it is no longer popular to produce traditional "forecasts." You will note in the process that, along with my colleagues among the energy economists, I too will avoid making forecasts.

TYPES OF ENERGY FORECASTS
The Next-Year Forecast

Although I do not want to devote much attention to extremely short-range energy forecasts, I should at least mention how they fit into the forecasting scheme. In this classification we find the trade press or industry association forecasts which link the production and consumption of energy or the individual fuels to the business outlook for the next year. These may be simple composites of expert opinion or they are the product of fairly complex analyses where expected changes in energy prices and supply are applied to many-sectored input-output models of the economy to determine what effect energy will have on demand, employment, GNP, or other parts of the economy. Or, in the reverse, they start with certain assumptions as to next year's GNP, the industrial production index, number of cars to be manufactured, and other variables, to determine what energy products will be required month-by-month. products will be required month-by-month. These are interesting exercises for use by the energy executive, the financial community, or government officials. However, as we have found, our short-term business forecasting has its share of problems and frequently misses the mark.

This content is only available via PDF.
You can access this article if you purchase or spend a download.