While there are enormous remaining oil resources that could be made available using advanced tertiary recovery techniques, the OTA forecasts that, largely because of technical problems associated with the development of CO2-enhanced recovery technology, oil production via tertiary recovery will increase only relatively modestly-from .4 million barrels per day in 1979, to 0.5 to 0.9 million barrels per day in 1985, and 1.5 to 2.5 million barrels per day by 2000.

As far as imports are concerned, there has been a dramatic reduction recently, but the availability of oil on the world market will probably be less in the future. One factor involved is OPEC's stated intention to cut back on oil production to conserve its finite oil resources. Another is the expectation that OPEC internal oil consumption will increase dramatically over the next decade as a consequence of rising incomes and astonishingly low oil prices.

What is particularly troublesome is that oil supplies available for export from major non-Arab OPEC sources are not likely to increase over the next decade and may well decrease. Table 1 attached to my written testimony shows that in the first half of 1980 nearly 40 percent of our oil imports came from relatively secure sources-Canada, Mexico, Nigeria, Venezuela, and Indonesia. The increased oil that will probably be available from Mexico in the future could well be more than offset by reduced oil availability from the other countries in this group.

Now I would like to turn to the prospects for domestically produced synthetic fuels. The major supply initiative that has been taken to offset the prospect of diminished petroleum availability is the passage of the National Energy Security Act of 1980, which contains provisions for stimulating synthetic fuels production. This act established a federally owned Synthetic Fuels Corporation that would provide industry with $20 billion in immediate subsidies and up to $68 million in additional future subsidies in the form of purchase agreements, price guarantees, loans, loan guarantees, and government-industry partnerships.

This act targets the production of synthetic fuels in the amounts 0.5 million barrels of oil equivalent per day by 1987 and 2 million barrels per day by 1992. Many analysts believe that these targets are upper bounds on what is practically achievable by 1990; even if these targets could be achieved, however, synfuels production would probably fall far short of closing the oil demand/supply gap that is likely to develop over the next 5 to 10 years.

In addressing our oil supply problem it is important to recognize that our petroleum crisis is primarily a supply crisis for high quality liquids-mainly gasolines, distillates, and liquefied petroleum gases.

Among end uses, the transportation sector, which accounts for about 70 percent of the total demand for high quality liquids, is particularly troublesome because of the limited opportunities for fuel substitution. And in the transport sector automobiles and light trucks are dominant, accounting for nearly 75 percent of the high quality liquids consumed in transportation.

Because of the central role of the automobile in our petroleum problem, Congress took action in 1975 to curb automotive fuel demand with the automotive fuel economy standards established in

the Energy Policy and Conservation Act of 1975, EPCA. The EPCA mandated that the average fleet fuel economy for new cars rise to double the 1975 level by 1985.

However, because it takes time to turn over the automobile fleet, because of the rising number of cars, and because the on-the-road performance is not as good as the fuel economy in the mandated EPA test, the actual average fuel economy for all autos and light trucks is expected to be only 18 mpg in 1990, and light vehicle fuel use in 1990 is expected to be only 200,000 barrels per day less in 1990 than in 1980.

Thus, the major effect of the 1975 fuel economy standards will be to keep automotive fuel requirements from rising. The automobile will remain particularly vulnerable to the prospect of sharply diminished petroleum supplies that are likely over the next 5 to 10 years.

One of the major attractions of a gasoline tax is that it could be immediately effective in reducing gasoline demand. Consumers would respond to the tax by cutting back on unnecessary automobile use, by carpooling or vanpooling, by using buses more, and when buying new cars, by choosing more fuel-efficient models. Since the purpose of an added gasoline tax is to curb the demand for gasoline-not generate increased revenues-the gasoline tax revenues should be returned to consumers. There are a wide range of possibilities for doing this. I will not discuss the merits of the various alternative approaches here. But among the many considerations that should bear upon the selection of an appropriate rebate scheme two are especially important: first, the effect of the rebate on petroleum demand; and second, the fairness of the rebate.

Since the purpose of a long-term tax is to bring about a smooth transition to a long-run situation where gasoline demand is considerably lower than today, the rebate should be designed to encourage new transportation habits consistent with this shift. For a given level of taxation, a rebate that increases with the capacity to consume gasoline would be less efficient in changing habits than one that is independent of this capacity.

Thus, for example, while a rebate to owners of registered vehicles might be effective in minimizing dislocations in an emergency supply disruption, such a rebate would encourage consumers to hold onto a number of old clunkers. A tax rebated equally among adults, on the other hand, would be relatively efficient in fostering gasoline conserving habits.

In practice, a rebate should be designed to achieve a balance between efficiency in promoting a change of habits and fairness. In general, there would be no conflict in the design of a rebate between the goals of efficiency in promoting gasoline conserving habits and fairness to the poor, simply because a typical poor household consumes less gasoline than the average household. Because nearly half of poor households have no cars and because those that do typically drive only about half as much as the average household, poor households would generally benefit from a tax that is rebated on a per adult basis.

But what about those groups and individuals especially dependent on the automobile? Even with a rebate, the gasoline tax would

70-601 - 81 - 9

of course be harsh for these people, but the impact could be cushioned by phasing in the tax. If it is decided, for example, that the tax should be as high as $2 a gallon, the tax might be phased in over, say, a 4-year period, with increments of 25 cents a gallon every 6 months. An announced preliminary schedule of phased tax increments would enable consumers to anticipate and plan for eventual very high gasoline prices.

Aside from the problems of special groups, a gasoline tax rebated equally among adults would be fair in the broader sense that the system would offer broad opportunities for beating the tax. To illustrate this feature of the tax-rebate system, it is useful to introduce the concept of the effective price of gasoline, which I define as the total net expenditure for gasoline-the actual expenditure minus the rebate-divided by total gasoline consumption.

Now, suppose that the gasoline tax is raised to $2 a gallon. By introducing the tax/rebate system, the pump price of gasoline would increase from $1.10 to $3 a gallon, but the effective price would be less and would depend on the level of consumption. I have estimated that the rebate would amount to about $730 per adult in the first year after a tax of this magnitude is levied.

Figure 5 attached to my written testimony shows the effective price with this rebate for two-adult households at different consumption levels: the typical poor, carowning household would pay an effective price less than the pretax price-that is, the rebate would be greater than the amount paid in taxes. The typical lower middle-income carowning household would pay an effective price about equal to the pretax price. But the average carowning household would have to reduce consumption 25 percent to reduce the effective price to the pretax level-that is, to beat the tax.

This reduction could be achieved either by driving less or using a more fuel efficient car. The average household, driving 14,000 miles per year, could beat the tax with a fuel economy of 19 miles per gallon or more, while a poor household, driving 8,000 miles a year, could beat the tax with a fuel economy of 10 miles per gallon or

more.

Clearly, no tax-rebate scheme will be perfectly fair. But a tax/ rebate scheme decided by consensus in the U.S. political process is infinitely more fair than continuing tax increases by OPEC, for which the revenues cannot be rebated.

Because the question of fairness is key to the political acceptability of a gasoline tax, the rebate scheme should be worked out and perhaps legislation specifying how the gasoline tax revenues would be refunded should be passed before setting the level of the gasoline tax.

How large should the gasoline tax be? The level of tax depends on the desired level of oil demand reduction and how effective price will be in suppressing demand. If a demand reduction on the order of 12 to 2 million barrels of gasoline per day is sought by the mid to late 1980's, then a tax in the range of $1 to $2 a gallon appears to be needed.

Unfortunately, economists do not know with precision how consumers will respond to higher prices. With a phased in tax, this uncertainty could be dealt with in practice by adjusting the levels of successive tax increments up or down from the originally pro

posed schedule in light of experience with various increments, so as to keep the oil import reduction goal on target.

The question of how large a demand reduction we should be seeking is much tougher to deal with. There is general agreement that domestic oil production will fall, but there is uncertainty about how much. The world oil supply situation is even cloudier. I believe there is no purely analytical answer to the question of whether our imports should be reduced by 1 or 2 or 3 million barrels per day over the next several years. In the final analysis, the import reduction goal and the associated tax should be set in the political process, taking into account both quantifiable and unquantifiable costs and values relating to our energy system.

Clearly one of our goals should be to reduce our vulnerability to oil supply disruption. A related political consideration is that the levy of a stiff tax would enable the United States to demonstrate the seriousness of its commitment to dealing with the world oil crisis to its allies, many of whom already have stiff gasoline taxes. As of July 1980, the tax on regular gasoline was $1.23 a gallon in West Germany, $1.68 a gallon in France, and $2.16 a gallon in Italy, compared to 14 cents a gallon in the United States.

Also, the decisionmaking process should take into account longrange as well as shortrange issues in setting an import reduction target and in determining the appropriate tax level. A stiff excise tax on gasoline could powerfully influence the future course of automotive transportation.

When buying a new car, the typical consumer would respond to a gasoline tax by buying a model with high fuel economy. The consumer who responds to a $2 a gallon tax by trading in his 14 mile per gallon gas guzzler for a new car with a 45 mile per gallon fuel economy would pay no more per mile for gasoline taxed at $2 a gallon than he paid for pre-taxed gasoline to run his old car. But can the average car buyer purchase such a car?

The VW Rabbit Diesel, with an interior volume only 20 percent smaller than that of the average U.S. car, has an on-the-road fuel economy of 42 miles per gallon. Numerous studies have concluded that it is technically and economically feasible to raise the average on-the-road fuel economy of all new cars in the United States to 40 or 50 miles per gallon in the 1990's.

And it is feasible to go much further than this. Volkswagen is developing a four-passenger car with a small three-cylinder supercharged diesel engine which shuts off under coasting or idle conditions and is expected to have an EPA composite fuel economy of over 65 miles per gallon.

Thus, if there were a market for high fuel economy cars they could be produced. Some combination of fuel economy standards and a gasoline tax could bring about this shift to cars with high fuel economy.

New post-1985 fuel economy standards are desirable to spur the development and introduction of truly high fuel economy cars-40 to 80 miles per gallon-while a complementary tax would aid Detroit by reducing the uncertainty in the future demand for fuel efficient cars. A legitimate fear in the auto industry is that higher fuel economy standards by themselves may reduce sales, because

many consumers would hold onto their old gas guzzlers longer than they otherwise would.

But a tax rebated to adults would mean instead increased total car sales and an accelerated transition to cars with high fuel economy. This effect can be seen in figure 6, attached to my written testimony, showing the effective gasoline price as a function of automotive fuel economy in the first year after raising the gasoline tax to $2 a gallon.

Note that for a two-adult household driving the average amount at a fuel economy of 45 miles per gallon, the effective gasoline price is minus $1.50 a gallon-that is, the rebate is much greater than the total amount paid for gasoline. As more and more people shift to fuel efficient cars, this incentive would diminish, because the amount of the rebate would decline. Thus a gasoline tax rebated to adults would accelerate a transition to fuel efficient cars because it would powerfully reward those who first shift to fuel efficient cars.

The logic of using a gasoline tax to foster a transition to fuel efficient cars, while at the same time boosting auto sales, has not been lost to the leadership of the automobile industry, as indicated by Henry Ford's proposal for a 50-cent-a-gallon tax. A major question that should be addressed is whether Detroit could meet this challenge without some form of economic assistance to enable them to make the necessary investments.

If automotive fuel economy is not improved beyond the 1985 level mandated by Congress in 1975, then fuel consumption by automobiles and light trucks in 2000 would be only slightly less than in 1980 (see table 3 attached to my written testimony) doubling (tripling) of the fuel economy of new cars between 1985 and 1995 would lead to fuel savings of 2.6 (3.4) million barrels per day by the year 2000. Clearly, a policy to foster a twofold to threefold increase in automotive fuel economy would be an effective response to the prospect of declining oil supplies.

A gasoline tax could also be used to bring about a transition to methanol as a major fuel for automotive transportation. Methanol is particularly attractive as a synthetic automotive fuel for a number of reasons:

Methanol is the fuel of choice in automotive racing, because it would be safer in a car crash than gasoline, and because of the increased power obtainable from the same engine compared to gasoline. Use of methanol in today's automotive engine can result in an energy efficiency improvement of 25 percent or more-largely because the fuel/air mixture can be leaner and because the methanol can be used at a higher compression ratio. While the toxicity of methanol is roughly comparable to that of gaoline, there would be net air quality gains in shifting automobiles from gasoline to methanol.

Methanol will probably be used as an automotive fuel first in fleets. At present, there is considerable fleet testing of methanol cars in the United States. GM has announced that it could be ready to market straight methanol-fueled cars that use gasoline for starting in as soon as 5 years. To facilitate wider use of methanol, automobiles capable of operating on methanol or gasoline or diesel fuel could be brought to market in a few years.