Document Type
Dissertation
Date of Award
1976
Keywords
Petroleum industry and trade, Mathematical models, Econometrics
Degree Name
Doctor of Philosophy (PhD)
Department
Economics
First Advisor
V. Kerry Smith
Second Advisor
K. V. Greene
Third Advisor
E. O. Olsen, Jr.
Abstract
The designated role of the economist is that of determining the efficient allocation of scarce resources. Assuming profit-maximizing behavior on the part of producers and utility-maximizing behavior on the part of consumers, the economist can determine the sensitivity of each to economic incentives. Criteria may thus be provided for the evaluation of the speed and impact of policies designed to influence the pattern of future production and consumption by accounting for producer and consumer motivation.
The purpose of this study, therefore, is to develop an econometric model of the petroleum industry in order to determine the output and pricing behavior of petroleum producers, as well as the economic behavior of petroleum consumers. Once these behavioral patterns have been ascertained, the effectiveness of various national policy options on future oil production and consumption can be gauged.
Chapter II establishes the basic framework within which the model is developed. Through the caprice of energy resource distribution around the world, the oil industry is in a strategic position to affect the foreign policy of the nation. It must, therefore, be determined if the practices of the oil industry in any way restrict the availability of its products so as to endanger the national interest. An investigation of the structure of the petroleum industry focuses attention on certain economic and financial practices that may be anticompetitive or otherwise in violation of the spirit, if not the letter of the law.
Several antitrust litigations have been initiated over the past fifty years, some of which were conspicuously successful such as the dissolution of the Standard Oil Trust in 1911, and others which had but a patina of success, such as the famous “Mother Hubbard” case of 1940. The House of Representatives conducted a study of the concentration of the petroleum industry within the energy market in 1971, and the Justice Department has been investigating joint venture pipelines and other market linkages among the majors. Senate sub-committees have investigated gasoline marketing practices, and the Federal Trade Commission recently completed an investigation of the market behavior of the vertically-integrated firms in the petroleum industry. Over 300 formal investigations have been made during the last fifty years, two trade practice conferences have resulted, and there have been 22 formal complaints against the industry involving restraint of trade and deceptive marketing practices.
It is impossible to define an industry in a binary fashion as either competitive or non-competitive. There are degrees of competition, and industries may be distinguished only by relative competitiveness using the axioms of perfect competition as a benchmark for comparison. Chapter II is deemed vital to this study for two reasons: 1) doubt has been cast upon the ability of the petroleum industry to behave as “workably” competitive, and 2) the structure of the industry, whatever it is ascertained to be, is basic to the formulation of an economic model purporting to explain producer behavior in output and pricing. Restrictive assumptions concerning the structure of the industry serve to restrict the predictive capability of the model based on these assumptions.
Chapter III provides a review of the literature of previous attempts to model the petroleum industry. These attempts are relatively new, the first being formulated only in 1964. They have been burgeoning, however, during the past five years as has the interest in the scarcity of our fossil fuels. In general, these attempts may be classified into two or possibly three basic approaches which serve to provide a mode of presentation. Attempts to model the refining sector have been confined to the linear-programming or process-analysis approach. Considerably more attention has been paid to modeling the producing sector, but relatively few approaches have been utilized.
With few exceptions the bulk of the models, to date, have been based upon the assumption of a perfectly competitive oil industry. Such an approach has no doubt stemmed partly from the nature of the tools of the economist. Economics is a comparatively new social science; it is only the theory of perfect competition which has received complete treatment in the development of production theory in economics. The theory of imperfect competition covering the spectrum of slightly imperfect to complete monopoly is in its infancy.
To the extent, however, such an approach has been motivated by sincere conviction of the existence of a competitive petroleum industry, one must examine in depth the historical development of the models to glean insight into the how's and why's of the current state of the arts in petroleum modeling. It is by appreciating where we have been that we are able to determine where we are going and the best technique for getting there.
Chapter IV sets out the theoretical underpinnings of the model to provide a rationale for its structure. Two plausible theories of producer behavior are hypothesized: 1) the industry is perfectly competitive, and 2) the industry is characterized by the existence of a few dominant firms, and numerous smaller firms representing a continuum of vertical integration. The imperfection of competition suggested by the latter theory implies the use of less restrictive assumptions in the development of the model.
The model is sketched out in basic structural form with each equation specified and explained in terms of the theoretically relevant variables. Estimated at the aggregate rather than firm level, the model provides maximum flexibility in terms of specifying the nature of the underlying production function and the scale economies which characterize production in the industry. The basic framework thus allows for the possibility of either a perfectly competitive or imperfectly competitive industry.
A trade-off is always involved in determining the proper level of aggregation of a model. To aggregate at the national level overlooks differences between states and regions and can result in the attribution of typical behavior to non-typical producers. To focus on a small production unit, however, is to fail to appreciate the national character of the major oil producers. Oil produced in one state is generally consumed in another, and numerous exchange agreements exist between the firms and across states to insure profit maximization for each firm. Moreover, the thrust of this study is the evaluation of national oil policy for the United States as a whole. To the extent the majors are both aware of and affected by the output/pricing policies of one another, the microscopic examination of the producing behavior of a single firm would produce unreliable myopic predictions of how it behaves as part of a group of firms, firms whose activities are confined to neither state not regional boundaries.
Chapter V discusses the estimation techniques utilized and their justification, and presents the final estimates of the 41-equation model. The results of estimating two alternative versions of the model are compared and the predictive performance of each is evaluated with respect to single equation goodness-of-fit tests. The first version of the model specifies the structural form of the producing sector explicitly. It is assumed that drilling activity results in additions to reserves, and that these crude reserves are then used to produce crude. Net imports of crude are determined as an endogenous variable in the block-recursive producing sector. The refining sector is characterized by the simultaneous determination of refinery outputs and prices and is linked to the producing sector via the price of crude. The demand for crude is specified as that of refiners, and crude is deemed an input into the production of refinery outputs. The price of foreign crude enters into the determination of the demand for crude, the price of crude, and the prices of four major refinery outputs which comprise approximately 82% of the refiner's total receipts. The alternate model leaves the producing sector unspecified, stipulating only that supply is a function of output and input prices. A single equation thus replaces the block-recursive producing sector of the first version of the model. As in the first version, refinery outputs and prices are assumed to be determined simultaneously by the market forces of supply and demand. This treatment is consistent with output and price determination over the sample period. The alternate specification of the model is based upon the acknowledged lack of insight as to what the correct structural form of the producing sector of the petroleum industry is. The first version provides a hypothetical structure in the hopes of providing some information as to the nature of the correct structural form.
The reliability of any model is based upon an evaluation of its performance. Credibility is established by observing the ability of the model to correctly track the data, discerning turning points as well as trends. Estimated as a simultaneous-equation model, such that the relationships postulated are interdependent, the relevant question is how the model behaves as an integrated whole in tracking the data. Chapter VI presents the results of simulating the model over the sample period (1946-1973) using both single-period and multi-period predictions. The model is then evaluated as to its ability to predict outputs and prices accurately as well as to predict turning points.
Since the basic purpose of developing the model was to provide criteria for evaluating various policy options for coping with the oil shortage, the model was simulated from 1974 to 1985 and predictions made based upon the estimated sensitivity of both outputs and prices to a number of economic incentives. The model was first adapted to reflect the elimination of the oil depletion allowance as of 1975, and then simulated for the following policy options: 1) an increase in crude prices by OPEC consistent with the world rate of inflation, 2) gasoline rationing, 3) increased gasoline taxes and 4) the Ford energy proposals of deregulation of the price of crude coupled with windfall profit taxes on the industry.
The final chapter develops the policy implications of each of these proposals and the predicted impact on future production and consumption of petroleum and petroleum products. Avenues for future research in petroleum model-building are also suggested.
It has been predicted that oil imports could increase from 3.2 mmb/day in 1973 to as much as 19 mmb/day by 1985 and that the deficit in balance of trade in petroleum liquids could increase from almost $3 billion to almost $30 billion annually by 1985. Long term planning to provide for increased domestic production of crude and adequate refining capacity for processing both domestic and foreign crude supply has become increasingly more difficult. Sound governmental policy guidelines based upon the economic realities underlying both the production and consumption of petroleum and petroleum products are essential if the oil industry is to maximize self-sufficiency, given the remaining ten years of oil reserves. Consumers must be motivated to minimize domestic consumption to help alleviate the excess demand. Both supply and demand policies are critical to assuring long-run economic growth for our energy-intensive nation.
Recommended Citation
Rice, Patricia L., "An econometric model of the petroleum industry" (1976). Graduate Dissertations and Theses. 395.
https://orb.binghamton.edu/dissertation_and_theses/395