Benefits and Costs of the US Gasohol Program

The US Energy Security Act of 1980 establishes a federal gasohol program to encourage production of fuel alcohol from biomass equal to at least 10 percent of total US gasoline consumption by 1990. This implies a production of 10 billion gallons of fuel alcohol out of a total projected consumption of about 100 billion gallons of motor fuel.

Federal incentives to achieve this target include the following:

• Exemption from the 4-cents-per-gallon federal excise tax on gasoline for gasohol containing at least 10 percent alcohol, through 1992 (Crude Oil Windfall Profit Tax Act of 1980)
• A special 10 percent investment tax credit (through 1985)
• An entitlement (of about 5 cents per gallon) for fuel alcohol produced from biomass (program terminated in January 1981)
• Grants, loans, and loan guarantees totaling $1.05 billion (Energy Security Act of 1980 and Supplemental Appropriations Act of 1980). Loan guarantees allowed to be leveraged on a 3:1 basis (P.L. 96-514). The Reagan administration has proposed to rescind this financing authority.
• Mandated use of gasohol in federal motor vehicles when available at "reasonable prices" (Energy Security Act)
• Authority for the secretary of agriculture to grant alcohol producers preferential access to government-held stocks of corn, at the reserve release price (125 percent of support), instead of 105 percent of the call price (152 percent of support) applicable to other users (Agricultural Act of 1980)
• Authority for the secretary of agriculture to establish a gasohol feedstock reserve, using agricultural commodities acquired by the government as a result of export controls; these stocks may be released to alcohol producers at a price lower than 105 percent of the call price (Agricultural Act of 1980).
• Incentive payments to farmers devoting set-aside acreage to the production of crops for fuel (Amendment to Food and Agriculture Act of 1977)
• Priority allocation of natural gas (in the event of a shortage) for crop production devoted to fuel alcohol and for producers of fuel alcohol (Energy Security Act of 1980).

In addition, twenty-five states offer incentives for alcohol production from biomass, mainly in the form of a total or partial exemption of gasohol from state gasoline taxes (table 1).

By far the most important of these aids is the exemption from the federal gasoline tax: it amounts to 40 cents per gallon of the alcohol contained in gasohol. The value of the special investment tax credit has been estimated at 3 cents per gallon of alcohol (Williams, 1980). The subsidy equivalent of the other federal incentives is difficult to quantify.

The state tax exemptions average 5.4 cents per gallon of gasohol, or 54 cents per gallon of alcohol, in the twenty-five states where they apply. If the states granting no exemptions are included, the weighted average of state tax exemptions for the United States as a whole amounts to 25 cents per gallon of alcohol.

Ethanol versus methanol

The US gasohol program is based on the displacement of gasoline by ethanol, ethyl or grain alcohol, the same kind of distilled spirit that goes into whiskey or is used in scientific laboratories. Indeed, the very familiarity of ethanol production and use constitutes a powerful, if intangible, argument for its use as a motor fuel.

Why ethanol produced from corn and other feed grains and not some other alcohol? The general answer is that a blend of 10 percent ethanol and 90 percent regular gasoline—gasohol—can be burned in conventional automobile engines, with a slight loss in miles per gallon and a slight gain in octane rating.

The principal alternative, methanol, can be produced more cheaply (from wood and other cellulosic materials, and even more cheaply from coal), but it has major drawbacks as a motor fuel extender. In particular, whereas a 10-percent ethanol-gasoline blend can be used in the family car, admixtures of methanol require extensive engine modifications (Cecelski and Ramsay, 1979; Owens, 1980; OTA, 1980).

Ethanol is currently produced in two types of plants--dry-milling operations, which produce alcohol and distillers dried grain (DDG, 22 percent protein), and wet-milling operations, which turn out a range of co-products, including high-protein feeds, such as gluten feeds and meal, corn oil, and starch. Moreover, wet milling can process the starch into corn sweeteners or ethanol.

The optimum distilling capacity for a single plant is 50 to 100 million gallons annually (Meekhof and coauthors, 1980a, table 7). Estimated dry-milling production costs of ethanol in a modern grain alcohol distillery are shown in table 2. Total net costs of ethanol produced as part of a wet-milling operation appear to be similar: higher by-product credits are approximately offset by higher investment and operating costs. The wet-milling process is more attractive, primarily because it offers more flexibility in response to changing market conditions, a quality that will be particularly valuable in the event that crop-based alcohol production loses government support or becomes unprofitable because of high feedstock prices.

Market value of ethanol

With the federal subsidies, ethanol in late 1979 was competitive with regular non-lead gasoline, which then sold for 85 cents at the refinery. Whether ethanol is, in fact, exactly comparable in value with regular nonlead gasoline is a matter of debate. For example, ethanol contains about 38 percent fewer Btus per gallon than does gasoline, and, compared with gasoline, laboratory tests show a mileage loss for gasohol varying between zero and 4 percent (OTA, 1980), or zero to 40 percent if attributed to ethanol alone. The mileage penalty tends to be more severe in late model cars (Austin and Rubenstein, 1980; Owens, 1980).

On the other hand, ethanol raises the octane rating of gasoline, with estimates varying rather widely from 0.8 to 5.0 points (OTA, 1979). At 0.8 cents per octane point, the value of the octane-enhancing property of ethanol amounts to 0.64 to 4.0 cents per gallon of gasohol, or 6.4 to 40 cents per gallon of ethanol. On balance, in the absence of more certain information, gasohol seems worth neither more nor less than regular nonlead gasoline.

Prospects through 1992

Five months after enactment of the extended federal tax exemption, the total ethanol production capacity announced or under construction was approaching 1 billion gallons—roughly double the Carter administration's original expectation. Investments totaling at least another 1 billion gallons seem to be under active consideration. Whether this pace is sustained depends on developments that are difficult to foresee, including, in particular, the future course of crude oil and corn prices, federal and state support, and the cost of borrowing.

Probably it is safe to assume that the real cost of crude oil will not fall in this decade. Rather, it is likely to rise further, so that the profitability of ethanol production is unlikely to be threatened from that side.

Grain prices have risen sharply in 1980-81, but this is largely the result of temporary factors, including the 1980 drought in the United States and two successive crop shortfalls in the Soviet Union. Looking further ahead, real grain prices had been projected to rise somewhat during the decade, even before gasohol entered the picture: US grain exports, which tripled during the 1970s to over 100 million tons, could rise to 150 million tons by 1990. With domestic demand rising moderately, to about 180 million tons, the total demand for US grain—excluding grain for fuel—may be projected at 330 million tons by the end of the decade, an increase of about 58 million tons (21 percent) over the current trend rate.

Assuming a continuation of long-term productivity growth at the rate experienced in the past forty years, an increase of this magnitude could be met largely out of increased productivity per acre and per unit of input, without a significant increase in production costs. This conclusion no longer holds, however, if on top of this demand, another 100 million tons of grain—the quantity needed to produce 10 billion gallons of ethanol—is claimed by the gasohol program. Even allowing for the soybean acreage that would be released because of the increased availability of protein feeds accruing as a by-product of ethanol production, and the possibility of bringing additional land into cultivation, the gasohol program will put increasing pressure on US agricultural resources and drive up all agricultural prices, especially those of feedgrains.

The principal objective of the gasohol program is to reduce US dependence on petroleum imports, a goal of self-evident importance. Whether alcohol—and crop alcohol in particular—is a rational and cost-effective option for accomplishing the objective is open to question.

A related economic objective is the strengthening of the US balance of payments, now burdened by petroleum imports that will approach $100 billion in more than one-third of total imports. What matters here is the net effect, after allowing for possible adverse effects on, exports. Again, is gasohol a rational and cost-effective answer to the balance-payments problem?

Assume that the target of 10 billion gallons of alcohol (ethanol) from biomass is reached by 1990. Assume, further, that motor fuel consumption stabilizes at 100 billion gallons per year. Then ethanol will replace 10 percent of US gasoline consumption; and gasohol will be the predominant motor fuel.

Often it is asserted that every gallon of ethanol produced will save one gallon or more of gasoline. This claim must rest on rather sanguine assumptions regarding one or more of the following: (1) crude oil savings in oil refineries made possible by the octane-enhancing property of ethanol when it is blended with low-octane gasoline; (2) the fuel efficiency (miles per gallon) of ethanol; and (3) the possibilities of minimizing the use of liquid fuel and natural gas in the production of ethanol.

Since the objective is to reduce imports of liquid fuel, it may appear at first sight that it does not matter what quantities of nonliquid fuels are used in the production of ethanol. This reasoning is valid for coal and other solid fuels because they are in ample supply in the United States and because they are not, at this stage, generally acceptable substitutes or feedstocks for liquid fuel. Natural gas, on the other hand, must be treated on the same basis as liquid fuel because it is relatively scarce, and because it is a highly acceptable substitute for oil in residential and industrial heating. In what follows, therefore, liquid fuels and natural gas are considered together as premium fuels.

Several factors affecting the premium fuel economy of ethanol production are so uncertain that even the most thorough expert studies usually provide a wide range of estimates. Accordingly, any attempt to arrive at a single set of estimates inevitably involves a large element of judgement. The estimates presented in table 5 generally lie between the extremes found in the literature, but the conclusions of this study would not be invalidated if more generous estimates had been adopted.

The estimates in table 5 are based on the assumption that no premium fuel is used as process fuel in the distillery. This is not the case at present; indeed, natural gas is commonly used in existing plants. Although ethanol produced in this way is eligible for the gasoline tax exemption, it appears that future plants will use non-premium fuels (mainly coal) in anticipation of the scheduled decontrol of natural gas prices.

The estimates in table 5 allow for a 20 percent credit for crude-oil savings in oil refineries made possible by the octane-enhancing value of ethanol. The Office of Technology Assessment of the US Congress allows a 40-percent credit (OTA, 1979), but this seems excessive for several reasons.

First, it is questionable whether the addition of 10 percent ethanol to gasoline results in a three-point gain in octane, as assumed by the OTA; two points may be more realistic.

Second, the crude oil savings estimated by the OTA assume that future cars typically will require 91-octane fuel; at lower octane ratings, the savings would be smaller. As the OTA report points out, however, "in the 1980 there could be an increased use of automobiles that do not require high octane fuels. With these engines, the octane-boosting properties of the alcohols are essentially irrelevant. There are numerous (other) factors which can lower the actual savings below that which is technically possible."

Third, oil refineries may "convert to more energy-efficient processes for producing high-octane gasoline." In the light of these considerations, even a 20-percent credit may be too high.

Based on table 5, it would appear that the premium fuel savings resulting from the production of 10 billion gallons of ethanol will be 6.5 billion gallons, gasoline equivalent. Allowing for a refinery loss of 10 percent, this would be equivalent to 170 million barrels of crude oil per year, or 470,000 barrels of crude oil per day. This represents 2.7 percent of total US consumption of liquid fuels, and 7 percent of oil imports. [In 1980, US imports of crude oil and products averaged 6.7 million barrels per day (DOE, 1980b)].

In discussions of the balance-of-payments benefits of the gasohol program, attention usually is focused on oil-import savings. At the current price of about $35 per barrel, a net saving of 170 million barrels amounts to $6 billion. If the real price of crude oil should double again by 1990, the savings would double, in turn, to $12 billion.

The effects of the program on US agricultural exports are more difficult to assess. In calendar year 1979, exports amounted to $35 billion, of which $7.8 billion was accounted for by feedgrains, $5.5 billion by wheat and flour, and $7.9 billion by soybeans and soybean products. Projections by the author, assuming no gasohol program (table 6), imply that export growth will continue in the 1980s but at a slower rate than in the period since 1972.

The gasohol program will have an immediate adverse impact on the volume of feedgrain exports, but this is likely to be more than offset by higher export prices. This also applies, though to a lesser extent, to wheat. Exports of protein feeds (soybean, soybean meal, and protein feed by-products of ethanol production) should increase in volume and possibly also in price, but this result is particularly sensitive to trade policy reactions by US trading partners. The most significant adverse effects will be on American exports of commodities such as cotton, tobacco, peanuts, and animal products, which are more vulnerable to foreign competition.

The extent of these changes is highly uncertain because little is known about medium- and long-term demand-and-supply responses to price increases as large as those that may be expected to flow from the gasohol program. The Department of Agriculture (Meekhof and co-authors, 1980a), in trying to assess the probable balance-of-payments effects of a $2-billion-gallon gasohol program using 20 million tons of corn by 1984-85, found only modest effects on export quantities and export revenue (see table 7, column 3). (These results are partly explained by the large drawdown of stock projected by the authors.)

The cost of the gasohol program to the consumer will be primarily in the form of higher food prices.

Unfortunately, no fully adequate estimates are available of the effects on the food system of ethanol production, on the scale envisaged in the legislation. The two more thorough studies (Schnittker Associates, 1980; and Meekhof and coauthors, 1980a, 1980b) are for programs of more modest scope (2 to 4 billion gallons of ethanol) whose agricultural effects could be expected to be limited.

The Schnittker study includes a brief discussion of the effects on corn production, utilization, and prices of ethanol production at the presumed "maximum profitable" level of 8.9 billion gallons, which puts the effect on corn prices at 39 percent. But Schnittker Associates' conclusion that the diversion of 89 million tons of corn to ethanol would educe other domestic consumption and exports by a total of only 25 million tons (11 percent) seems unduly optimistic. Other estimates of the price effects, at both the 4-billion-gallon level and the 7- to 10-billion-gallon levels of ethanol production are more than twice as higher as Schnittker's and point to a doubling of feedgrain prices when the 10-billion-gallon level is reached (table 3).

A projection based on simplified assumptions (table 8) may help to explain why a 10-billion-gallon program is certain to have severe effects on the food system. Without the ethanol program, the domestic and foreign demand for grains is here projected to increase by 58 million metric tons by 1990. This increase would come largely from a 16 percent increase in yields and would require only a slight expansion of grain acreage, by 8 million acres. Most of the growth of demand (49 million tons) is for feedgrains. The demand for soybeans is projected to grow by 12 million tons (23 percent); because of a more modest increase in yields (10 percent), soybean acreage would have to be expanded by 8 million acres (12 percent).

Now assume that on top of this demand, another 100 million tons of feedgrains are claimed by ethanol. This would require an enormous expansion of feedgrain acreage, some of which would come about by rotating corn less frequently with soybeans. Because yields will be lower in these circumstances, this expansion probably would preclude the increase in average yields that would otherwise take place. Taking the yield effect into account, the acreage planted to corn and other feedgrains would have to be expanded by an additional 54 million acres. The demand for soybeans would be reduced by 18 million tons because of the availability of 28.5 million tons of DDG (or of equivalent protein by-products of ethanol). This would release 23 million soybean acres that could be planted to corn, but it would still leave a new requirement of 31 million acres over and above the 16-million-acre increase projected without ethanol. Smaller increases in acreage would be required assuming greatly increased fertilizer use, but this option also would raise production costs.

Since food accounts for 18 percent of consumer expenditures, the direct effect of a 12 percent rise in retail food prices on the CPI would be 18 percent of 12 percent, or 2.2 percent. To this should be added such factors as the effects of higher cotton prices on clothing and other nonfood components of the CPI.

The inflationary effects do not stop there, however. Higher food prices will be locked into the structure of nonagricultural wages and prices. Higher costs of nonagricultural goods and services, in turn, will feed back into agricultural production costs. By accelerating the inflationary spiral, sharply higher food prices ultimately could result in an increase in the general price level twice as great as can be accounted for by their direct (pass-through) effects. The total effect of the gasohol programs on food prices may, therefore, be estimated at about 25 percent and the effect on the cost of living at about 5 percent.

Although the circumstances are not exactly comparable, this conclusion is consistent with the experience during the "food crisis" of 1972-74. In that instance corn prices approximately doubled, in real terms, between mid-1972 and mid-1974; soybean prices rose even more. Food price inflation leaped from 4 percent in 1972 to 18 percent from January to December 1973. Largely as a result of the food price increase, the general inflation rate, as measured by the CPI, accelerated from 4 percent to 9 percent (this was before the oil price explosion). Food price inflation averaged about 15 percent through 1974, adding to the inflationary pressure exerted by the doubling of oil prices in 1974.

The experience of 1972-74 suggests that the effet of the doubling of grain and soybean prices on retail food prices may have been on the order of 14 percent in 1973 and 4 percent in 1974, or a total of 18 percent; the effect on the CPI was about 3.5 percent in the two years combined. Additional delayed effects occurred after 1974, particularly in the meat sector, caused by the biological lags in hog and cattle production.

Destabilizing effects on food supplies and prices

As pointed out earlier, the gasohol program will place additional pressures on world grain markets that already are increasingly volatile. Because of the profitability of subsidized ethanol production and the high proportion of its fixed costs, the grain-for-fuel demand will tend to remain strong and inflexible even in years in which feedgrain prices are high. As a result, the burden of adjustment will fall even more heavily than in the past on feed consumption in the United States, which is sensitive to price fluctuations. Inflationary shocks emanating from the grain sector will become both more frequent and more painful.

There are several ways of dealing with this problem, but all involve costs to our society. If nothing is done, the United States will have to cope with recurrent bursts of inflation which will be irreversible. And export controls will erode the American position in foreign markets.

Grain reserves are a better solution, but they are costly to maintain. For example, if government-owned or -subsidized feedgrain reserves are increased by 20 million tons—a conservative figure, considering the increased vulnerability of feed supplies to production shortfalls domestically or abroad—the annual bill for storage and interest could add up to more than $600 million. These costs are not to be found among the budgetary costs of the gasohol program, but they are nonetheless real.

Resource cost to the US economy

Some of the cost to taxpayers and consumers is in the nature of transfers of income and wealth to grain growers and distillers. Much of this will be capitalized in higher land values.

But there are resource costs to the economy as a whole. Most of these will be in the form of higher real costs of food production (inputs per unit of output). No attempt has been made here to estimate these costs, but they are likely to be not far below the $30 billion estimated increase in food costs to consumers.

There is an additional resource cost (or opportunity cost) to the economy that can be attributed directly to ethanol production. One way of estimating it is to compare the cost of production of ethanol with the cost of the gasoline it replaces. More relevant is the comparison between the cost of production of ethanol with that of alternative ways to produce or conserve premium fuels.

Ethanol versus gasoline

Even before the 1980 drought, when corn sold for $2.50 a bushel, it cost an extra $1.110 to replace on gallon of gasoline refined from imported oil by ethanol (table 9). Almost $2.00 per gallon would have to be spent on ethanol for 85 cents' worth of oil imports saved. The refinery price of gasoline would have to more than double in real terms—and the crude oil price almost triple—before ethanol would yield a positive social return.

Another way of looking at the problem is to compare the cost of replacing oil imports by ethanol with the costs involved in alternative ways of reducing petroleum imports.

Several approaches are more cost-effective than ethanol and can make a greater contribution toward energy independence. In the short run, the most effective steps will be conservative and increased use of coal for stream generation in power plants and in industry. For example, the fuel economy standards mandated for American passenger cars and light trucks are estimated to save 4 billion gallons in 1980, 21 billion gallons in 1985, 37 billion gallons in 1990, and 46 billion gallons in the year 2000 (DOT, 1981). A 10-percent tax on gasoline is estimated to induce, over a period of about ten years, a 7- to 10-percent drop in gasoline consumption (Pindyck, 1979)—about equal to the import savings that could be realized from a 10-billion-gallon gasohol program. About 30 billion gallons of oil are still being used for stream generation in electric power plants and industrial plants that could be converted to coal. These conservative measures involve much lower costs per gallon of imports saved than ethanol.

Farther down the road, nuclear power and coal-based synthetic fuels could replace imported petroleum. Shale oil also looks more promising than alcohol, although it does pose some difficult water resource and environmental problems. True, the investment costs involved in these alternatives are likely to be as high or higher per gallon of oil imports saved as those involved in the ethanol program, but the investments can be expected to pay off without resort to government subsidies. Unlike ethanol, these alternatives do not require large amounts of premium fuels in their production, and this helps to make them more cost-effective per gallon of oil imports saved.

Finally, whatever the resolution of the social and environmental questions surrounding nuclear power and coal-based synthetic fuels, these technologies do not exert severe pressure on scarce resources: uranium reserves are relatively ample and can be stretched by reprocessing and breeder technology; and 10 billion gallons of coal-based synthetic fuels would require no more than 10 percent of current US coal production.

Many advocates of the gasohol program admit that crop-based ethanol is a high-cost substitute that can make only a modest contribution toward energy self-sufficiency. In their view, the costs are justified to meet short-term needs until more cost-effective alternative sources of energy become available. As for other approaches, "we already are doing all we can."

But this line of reasoning begs the basic question, Is this the most effective use of society's resources? The evidence suggests that even in the short run, more modest additional resources directed toward alternative approaches would yield benefits several times greater, in terms of oil imports saved, than the gasohol program.

Phase-out costs

As matters stand, the exemption of gasohol from the federal gasoline tax probably assures the profitability of crop-based alcohol production through 1992. By that time, crop-based fuel alcohol is likely to become noncompetitive with alternative synthetic fuels whether or not the legislation is extended. Prudent investors will have recovered their investments before all or part of their plants become obsolete.

It is, of course, possible that the subsidy will be phased out before 1992, at least as far as crop-based alcohol is concerned. An early phaseout could take the form of an amendment to the legislation which would limit the tax exemption to production from distilleries constructed before a specified date, and, possibly, to new plants using biomass sources that do not compete with food production. The first provision would avoid the need to compensate investors for losses on investments made on the basis of present legislation, but it would leave the American taxpayer saddled with subsidies for the ethanol capacity already in place.

Dealing with the agricultural consequences of the termination of the program will be more difficult. Whether or not termination occurs ahead of schedule, grain growers will resist it strongly and press for alternative support programs that maintain grain prices above market-clearing levels or compensate farmers by direct payments.

The longer the delay, the more serious the adjustment problem will become as high grain prices sustained over a long period tend to be locked into land values. The resulting adjustment costs are likely to burden American consumers and taxpayers for many years after the gasohol program has run its course.

A costly stopgap

Gasohol is appealing to politicians and the public because it uses familiar technology and can produce results in the next few years. And it is especially popular among farmers because it is certain to raise prices of grains as well as other crops. In fact, the farm vote doubtless led a previously hesitant Carter administration to come out in support of massive subsidies for grain-based ethanol at a time when feedgrain prices were low and falling because of the partial embargo on grain sales to the USSR. The gasoline lines of 1979 and the subsequent oil price boosts by OPEC also spurred congressional approval of the program.

Gasohol's benefits seemed obvious and its costs rather modest. By replacing 10 billion gallons of gasoline with ethanol, it was thought that by 1990 the United States could save about $10 billion if real crude oil prices remained where they were, and twice as much if they doubled. The principal cost—the federal tax exemption—would then run at only $4 billion annually.

More careful analysis reveals, however, that the benefits are not nearly so great as may appear at first sight, and the costs much higher. What has been generally overlooked is the far-reaching impact of the program on our food system as heavily subsidized fuel alcohol competes for an ever greater share of our agricultural resources. At its target level of 10 billion gallons of ethanol in 1990, the program would absorb 100 million tons of grain, only one-third of which would be returned to the food system in the form of by-product protein feed. The net drain on the food system would be about 67 million tons grain equivalent—roughly one-third of all the grain and soybean meal fed to livestock in the United States.

Can the program be justified on the ground that it provides a useful stopgap for the eighties—until other, less costly synthetic synthetic fuels become available? Hardly, because even in the eighties, a dollar invested in alternative approaches—notably in conservation measures—will yield much higher returns, in terms of oil imports saved, than will fuel from biomass.

It is time for second thoughts.