incidence of costs on refiners-by unequal non-product and imported crude oil costs the same refined product may have many different prices. Customers of suppliers with access to low-cost products would tend to take more of their allocation than purchasers from suppliers with high cost products, who would be expected to shop around.

These factors, and the simple scheduling problem of finding surplus product before deciding whether to take an allocation, may inhibit purchasers from moving around.

EVALUATION OF 1974-75 CONTROLS

Phase IV controls were instituted during a world crude oil shortage and developed under the severe pressure of the Arab oil boycott. A shortage of refinery capacity existed simultaneously. Under these conditions the extension of price controls and mandatory allocation to all levels of the industry could be necessary to hold down consumer prices and to provide access to petroleum products on a basis judged equitable by policy makers.

However, by the end of 1975 refining capacity was adequate, and unlimited supplies of crude oil were available from foreign sources (at a high price). Under these conditions prices charged consumers could be controlled effectively through the crude oil price control and entitlements system without refined product controls. Although effective in controlling prices, with attendent consumer and macroeconomic benefits, crude price controls and entitlements were responsible for a decrease in the efficiency of use of energy resources.

The remaining structure of product price regulation and allocation may be justified in efficiency terms if there is a significant lack of competition in the refining, wholesaling, and retailing sector. To be sure, that structure could serve other purposes: to squeeze the profits of specific sectors of the industry, to preserve existing market structure as an independent goal, or to provide price advantages or penalties to specific classes of energy users. For example, during Phase IV and the embargo consumption of gasoline was seen as less meritorious than heating of homes. Consequently higher retail prices and lower refinery yields of gasoline were encouraged to provide adequate supplies of heating oil at low cost.75 However, if competition in the industry is adequate, any aggregate consumer benefit from across the board lowering of domestic oil prices could, at least theoretically, be achieved with the crude oil program alone. Retaining that program alone would also remove many of the individually minor but pervasive and in total probably large-inefficiencies created by product price and allocation rules.

The plan of this Chapter is to identify and evaluate the consequences of crude oil and refined product price control and allocation programs that existed prior to 1976.76

75 "History of Petroleum Price Controls," p. 1315.

76 After this section was completed, the Federal Trade Commission published a "Staff Report on Effects of Federal Price and Allocation Regulations on the Petroleum Industry." by Calvin T. Roush, Bureau of Economics, December 1976. The author develops, independently, a theoretical analysis of crude oil price controls similar to that presented here he also evaluates those controls under the alternative assumption that the oil industry is not competitive.

A theory challenging the conclusion that crude oil, since controls and entitlements hold down refined product prices, is also described in this chapter. This theory claims that as long as there is free trade in refined products, and the U.S. importer has the ability to import foreign refined product as the marginal barrel of consumption, domestic refiners will target their prices at the level of world market supplies. Although theoretically possible, there is no definitive statistical evidence to support this view. Although the price of some products in some regions may be determined by the price of imports, the price of the bulk of domestic refined products appears more closely related to the controlled average cost of crude oil.

CRUDE OIL PRICE CONTROLS

Factors affecting crude oil production

Exploration for oil is characterized by uncertainty. After a series of geological investigations designed to identify the strong probability that oil will be found in a given location, test wells are drilled. Between 1973 and 1975 only 25 percent of all exploratory wells drilled were successful."

Drilling statistics include wells drilled in the hope of extending the boundaries of or increasing production from fields which are already in production. These wells are more often successful than are exploratory wells, so that on average about two-thirds of all wells are successful. All exploration and drilling activity, including that in proved fields, carries a risk of failure: the incentive to engage in those activities is provided by the expectation that revenues from productive wells will sufficiently exceed cost of production to compensate for expenditures on attempts that failed to find oil.

The decision regarding exploration, then, is based on expectations of profits to be earned from successful wells. Once a well is drilled, however, all exploratory and development costs are sunk and no longer relevant to decisions about how the well should be used. A well will be kept in operation as long as revenues are adequate to cover its variable costs of operation. Wells would continue in operation as long as total revenue remained above operating cost, which will be less than the sum required prospectively to justify investment in finding the wells.

However, opportunities for investing in an oil well or reservoir are not exhausted when the well is first put in operation. Oil is a nonrenewable resource, in that every barrel of oil produced leaves one less barrel underground. In many types of field, reduction in the amount of oil left in the ground reduces the pressure which forces oil up the well. Consequently, for any amount of effort supplied by the manager, less oil will be forthcoming over time.

77 FEA "Quarterly Report: Energy Information, Second Quarter, 1976," p. 15. 78 FEA "Monthly Energy Review," July 1976, p. 50.

The natural decline in oil production can be overcome by actions open to the oil field manager. For example, it is possible to drill another well and pump water into the reservoir to drive oil out. Many of these actions involve significant initial investments, which are repaid when oil revenues are increased sufficiently by additional output. Investments which change the characteristics of the oil reservoir to increase production are often referred to as "enhanced recovery." Some change in production rates can be achieved without going to secondary or enhanced recovery. These quickly achievable changes are limited in magnitude, but can be undertaken rapidly in response to temporary price condition. Enhanced recovery projects, because of their long life, are like exploration projects more responsive to expected future prices. Consequently the effect that price controls have on expectations of future prices is likely to be as important as their effect on current prices.

A model of oil production

No simple account of how crude oil production decisions respond to regulatory policy can deal adequately with all the complex forces at work. However, some elementary economic concepts can serve to give an orderly and broadly accurate description of the effects of oil price regulations. Three sources of complexity must be kept in mind:

(1) Oil fields differ drastically from one another;

(2) The behavior of a field changes over time, as a function of the amount of oil taken from the field and the method of production; and

(3) Large, discrete investments are required to make significant changes in a field's production levels.

The decision problem facing a manager is similarly complex. The manager must take into account not only current prices and costs, but also expectations of future prices and the effect which current decisions will have on the future relation between production costs and crude oil output.

To describe these decisions, we posit the existence of a marginal cost curve like that drawn in Figure 6. The horizontal axis measures the production to be achieved from a well in the current year. The vertical axis measures dollars per barrel. The marginal cost curve shows how much it would cost to increase output by 1 barrel per year when output is at the level measured on the horizontal axis.

5.03

FIGURE 6.-Marginal cost curves in crude oil production.

The position of the marginal cost curve changes over time, as the oil reservoir is depleted. The cost of maintaining any level of output increases over time, and when prices are constant the most profitable output level declines. In terms of Figure 6, this effect is represented by the shifting of the marginal cost curve to the left (from MC' to MC") over time.

Enhanced recovery can be represented by shifting the marginal cost curve to the right. The curve MC"" might be obtained if reinjection were applied to a well that previously had marginal cost curve MC". Marginal cost curves can be used to predict how much oil will be produced from a well at any given price. Consider, for example, a $5.03 per barrel price. The marginal cost curve reveals that when output equals Q1, changing output by one barrel will change costs by exactly $5.03. If production were increased much above Q1, each additional barrel of oil would cost more than the price of $5.03 at which it could be sold. Consequently when prices are $5.03 per barrel, Q, will be produced. The marginal cost curve is the supply curve, showing the prices that would induce various rates of production.

0

Dollars per barrel-14.50

Technically, the marginal cost curves discussed thus far are short run marginal cost curves, since they state the relation between cost of increasing production and level of production on the assumption that a specific type of investment has taken place (None for MC", reinjection for MC""). It is also possible to construct a long run marginal cost curve that will represent the relation over time between an existing output level and the cost per barrel of increasing output when the cost of investing in enhanced recovery is taken into account. The long run marginal cost curve is also the long run supply curve and can be used to predict how in the long run oil production will respond to price levels that are expected to be stable over time. At some points in the discussion that follows short term behavior is important, and at other points long term behavior. The distinction will be noted when it

matters.

Criteria for evaluating crude oil price controls

To characterize an efficient allocation of resources in oil production and consumption it is necessary to examine the market for crude oil. Supply and demand for crude oil is depicted in Figure 7, based on July 1975 prices and quantities.

FIGURE 7.-Supply and demand for crude oil.