National Academies Press: OpenBook
« Previous: PREVENTION AND THE COMMUNITY
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 24
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 25
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 26
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 27
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 28
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 29
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 30
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 31
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 32
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 33
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 34
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 35
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 36
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 37
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 38
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 39
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 40
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 41
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 42
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 43
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 44
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 45
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 46
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 47
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 48
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 49
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 50
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 51
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 52
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 53
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 54
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 55
Suggested Citation:"TAXING AND SPENDING." National Research Council. 1984. Toward the Prevention of Alcohol Problems: Government, Business, and Community Action. Washington, DC: The National Academies Press. doi: 10.17226/18637.
×
Page 56

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3 Taxing and Spending INCREASING THE FEDERAL ALCOHOL EXCISE TAX PHILIP J. COOK, Duke University Inflation and the Erosion of the Excise Tax The prices of alcoholic beverages have been declining rapidly in recent years in comparison with prices of other consumer items. Con- sumer price index statistics indicate that since 1967 the real price of all alcoholic beverages has declined 27 percent. Distilled spirits have led the way with a 48 percent fall in price, followed by beer (25 percent) and wine (19 percent). The results of econometric and epidemiologic studies show that such reductions in relative price have the effect of promoting alcohol consumption and increasing the incidence of alcohol- related problems. The fact that liquor prices are currently at about one-half of the 1967 level can be attributed in part to congressional inaction with respect to federal excise taxes on alcoholic beverages. Current federal excise tax rates were set in 1951 but have been largely repealed by inflation since then. If the federal tax had been indexed to the consumer price index in 1967, the cost of a fifth of liquor would currently be about $3.50 higher. Indexing the beer tax would have had a smaller but still sub- stantial effect on beer prices. It can be argued that the failure of Con- gress to increase taxes in step with inflation constitutes the most impor- 24

TAXING AND SPENDING 25 tant feature of federal government policy in alcohol abuse prevention during the past 15 years. The recent wave of concern over mounting federal deficits has put the federal alcohol tax back on the national political agenda. There is an obvious parallel with the federal cigarette tax, which had also been left at its Korean War level—$.08 per pack—for 30 years and then was doubled by Congress in 1982. While the impetus for raising this tax was a concern with "revenue enhancement," the more important conse- quence will be to promote the public health by reducing smoking. Similarly, an important consequence of an increase in the alcohol tax would be to reduce alcohol abuse and its attendant personal and social costs. I summarize here the evidence on the preventive effects of alco- hol taxes and then discuss the distributional and revenue impacts of a tax increase. Alcohol Taxes and the Rate of Alcohol Abuse Alcohol abuse is a major public health problem. There are roughly 75,000 alcohol-related deaths each year (Gerstein, 1981:205). Chronic heavy drinking is responsible for a large fraction of the more than 27,000 deaths each year due to liver cirrhosis as well as about 5,000 deaths classified as resulting from "alcoholism." The remaining alcohol-related deaths involve failures in judgment, attention, and physical coordina- tion resulting from inebriation, including about 25,000 alcohol-related auto fatalities and a number of other accidents, homicides, and suicides. Alcohol-related morbidity and injuries impose a considerable burden on the health care system. A recent estimate (Schifrin, 1983) of health care costs stemming from alcohol-related problems is $20 billion for 1979—about 10 percent of total health care expenditures for that year. The cost of absenteeism and reduced productivity associated with employee drinking and premature death is even greater—roughly $78 billion in 1979 (Schifrin, 1983). For alcohol taxes to be an effective policy instrument for reducing these and other costs associated with alcohol abuse, it must be true that higher alcohol prices reduce the prevalence of chronic heavy drinking and the incidence of drunkenness. An increase in taxes on liquor, wine, or beer results in an increase in the average prices of these commodities and a reduction in alcohol consumption from these sources. (More precisely, consumption is less than it would have been without the tax increase.) This result has been established quite conclusively for data from the United States and Canada (Johnson and Okanen, 1977; Orn- stein and Hanssens, 1981; Cook and Tauchen, 1982). More important

26 PREVENTION OF ALCOHOL PROBLEMS and controversial is the question of how taxes and prices influence the consumption levels of the heaviest drinkers. It is logically possible that average consumption would fall as a result of a tax increase solely due to its effect on moderate drinkers; the heaviest drinkers are immune to economic incentives. An argument supporting this possibility can be stated by a chain of propositions: (1) a large portion of the heaviest drinkers are alcoholics, in the sense that they are addicted to alcohol; (2) alcohol addicts will drink something like the biological maximum every day, practically regardless of the cost of obtaining their drinks; (3) therefore, it must be the more moderate drinkers who adapt their drinking practices to the price of alcoholic beverages. This sort of argument may seem plausible to many. In reply, an economist would point out that a price increase has a greater economic impact on an alcoholic—who may already be spending a third or more of his or her income on alcohol—than on a moderate drinker, and that ordinarily this greater impact would be expected to yield, if anything, a greater response in consumption behavior. Furthermore, there is considerable clinical evidence that alcohol consumption by alcoholics is responsive to experimental manipulations of the costs of taking a drink (Mello, 1972; Nathan and Lisman, 1976). In any event, this issue can be better resolved through careful empirical analysis than through unsupported generalizations about the behavior in question. The Effect of Alcohol Taxes on Cirrhosis Mortality Statistics on the prevalence of chronic heavy drinking are not widely available. How- ever, there is a widely accepted proxy measure—the rate of deaths due to cirrhosis of the liver. These death rates have provided the basis for nearly all estimates of alcoholism prevalence rates (Seeley, 1960). Most people who die of liver cirrhosis, especially after age 30, exhibit a history of chronic intense drinking: Schmidt (1977) found that 80 percent of all cirrhosis deaths in Ontario in 1974 were alcohol-related. The typical victim of alcohol-related cirrhosis has consumed an enor- mous amount of alcohol; Lelbach (1974) estimates that a primarily healthy, 150-pound person who drinks roughly 21 ounces of 86 proof liquor every day for about 20 years has a 50 percent chance of contract- ing liver cirrhosis. Thus, for any one individual, there tends to be a long lag between the onset of heavy drinking and the prospect of death from liver cirrhosis. The cirrhosis mortality rate is therefore not a direct indicator of the current fraction of alcoholics in a population, but does give a good indication of the fraction that has been drinking heavily for two or more decades. It should be kept in mind that the cirrhosis mortality rate is of interest in its own right, in addition to being a proxy

TAXING AND SPENDING 27 for the prevalence of alcoholism. Cirrhosis is among the leading causes of death in the United States, Canada, and most European nations. Using this statistical indicator of the prevalence of chronic excess consumption, it is possible to explore the relationship between alcohol taxes and excess drinking. My first study of this relationship (Cook, 1981) was based on annual observations of 30 states for the 15-year period spanning 1960 to 1974. During this period there were 38 instances in which one of these states increased its liquor tax by a substantial amount (more than $.24 per proof gallon). I viewed each of these tax increases as a test case in a sort of "natural experiment." For each of these test cases, the percentage change in the state's cirrhosis mortality rate was calculated; the test statistic was defined as the mortality rate during the 3 years before the tax increase. The control groups for each of these test cases were the other states in the corresponding years. The result was that states that raised their liquor tax typically had a greater reduction (or smaller increase) in cirrhosis mortality than other states in the corresponding year (see Table 3-1). Indeed, 63 percent of all test cases were in the upper half of the distribution with respect to the test statistic—a result that would occur by chance alone with prob- ability .072. This result is fairly strong evidence that the tax increase reduced the cirrhosis mortality rate, at least in the short run. Why did some (38 percent) of the states with increased taxes expe- rience a relative increase in cirrhosis mortality? My interpretation is that cirrhosis mortality fluctuates from year to year for a variety of reasons besides changes in liquor prices. In some of the test cases, these chance fluctuations happened to be positive and large enough to more than compensate for the consumption-suppressing effect of the tax increase. The fact that in most cases (63 percent) the state exhibited a relative reduction in cirrhosis mortality suggests that this effect does exist. The principal challenge to the validity of this interpretation is that a state legislature's decision to raise the tax is influenced, directly or indirectly, by cirrhosis trends in the state. For example, if a sudden increase in cirrhosis mortality led to a tax increase and a natural regres- sion to the cirrhosis mortality trend subsequently occurred, then the tax increase would be followed by a reduction in mortality but would not necessarily have caused it. This possibility is tested in Cook (1981) and Cook and Tauchen (1982). The evidence from these tests does not support this interpretation of my result—tax increases apparently are largely exogenous. My quasi-experimental approach to studying the effect of liquor taxes on heavy drinkers has the virtues of simplicity and ease of interpreta-

28 PREVENTION OF ALCOHOL PROBLEMS TABLE 3-1 Effect of State Liquor Tax Increases on Cirrhosis Mortality Rates, 1960-1974 Number of Percentage of Rank Order Test Cases Test Cases 1-5 9 6-10 9 63.2 11-15 6 16-20 3 21-25 9 > 36.8 26-30 2 Note: For each year during the sample period, the 30 states are rank ordered with respect to percentage change in cirrhosis mortality rate. The state with the largest reduction is ranked first. States that raised their taxes were usually at the low end (with the greatest reductions) of this distribution in the year of the tax increase. Source: Cook (1981:277). tion. It does not generate a usable estimate of the magnitude of the effect in question, however. Primarily for this reason I undertook a second study (Cook and Tauchen, 1982), which applied an estimation technique (analysis of covariance estimated by generalized least squares regression) to annual data from the same 30 states for the period 1962- 1977. Before undertaking this task, we refined the annual state-level cirrhosis mortality data; our variable was the age-adjusted mortality rate for state residents ages 30 and over. Our principal result can be stated this way: other things being equal, a $1 per proof gallon increase in a state's liquor tax will reduce the state's cirrhosis mortality rate by 1.9 percent in the short run. (The tax variable in the regression was adjusted for inflation, as measured by the consumer price index. The statement of our results given here is converted to October 1981 dollars.) The 95 percent confidence interval for this estimated reduction is 0.4- 3.5 percent; thus our parameter estimate is statistically significant by the usual standards of social science. Our parameter estimate suggests further that the tax effect is far from trivial—according to this estimate, a doubling of the U.S. federal liquor tax would reduce the nation's cirrhosis mortality rate by a figure in the neighborhood of 20 percent. Given the normally long lag between the onset of heavy drinking and death from cirrhosis, it may not be obvious how an increase in the liquor tax could cause an immediate reduction in cirrhosis mortality. The reason is that the cirrhotic process is interruptible—if at any time an alcoholic should stop drinking, his or her liver would cease to deterio-

TAXING AND SPENDING 29 rate. If the rate of consumption slows, then the deterioration process also slows. At any one time, there is a reservoir of people who are within one year of death from cirrhosis at their current rate of consump- tion. If some of them reduce their consumption in response to a tax increase, then not all of them will die in that year—i.e., the mortality rate will decline in the first year. What about the trend in mortality over the long run? The mortality rate will gradually decline after the initial drop, as the size of the reservoir gradually shrinks. The total effect of the tax increase will not be realized for many years, but clearly the ultimate reduction in mortality rates due to such an increase will exceed the initial reduction. Thus, our figures underestimate the full effect. In conclusion, there is considerable statistical evidence that a liquor tax increase causes an immediate and substantial reduction in cirrhosis mortality. If cirrhosis mortality rates are a reliable indicator of the prevalence of alcoholism, then it can be inferred that alcoholics' drink- ing habits are quite sensitive to the price of liquor. The Effect of Liquor Taxes on Auto Fatalities There is also some direct evidence that accident rates are responsive to changes in the liquor tax. I tested the effect of state liquor tax increases on the auto fatality rate in my 1981 paper, using the same sample and technique as those in the cirrhosis study. Between 1960 and 1974, most of the states (25 of 38) that increased their liquor tax subsequently experienced a below-average change in auto fatality rates relative to states that did not increase their tax. Unfortunately there is no comparable evidence on the effect of beer prices on auto fatalities. Beer is particularly important because it is the beverage of choice for the demographic group posing the greatest risk on our highways—young people. Conclusion Available evidence suggests that increases in alcoholic beverage taxes cause reductions in per capita consumption, consump- tion by chronic heavy drinkers, and the incidence of drunken driving. It seems safe to conclude that the sharp decline in alcohol prices during the last 15 years has exacerbated alcohol-related public health prob- lems. Fortunately there appear to have been countervailing forces at work during this period, such as increased emphasis in the popular culture on health and fitness, that have prevented a large increase in drinking from occurring. Indeed, adult per capita consumption has changed very little since 1970. If alcohol prices had kept pace with inflation, the per capita consumption would probably have declined substantially during this period.

30 PREVENTION OF ALCOHOL PROBLEMS Equity Considerations in Alcoholic Beverage Taxation Alcohol control policies, including taxation, have been criticized by some as excessively blunt instruments, reducing the enjoyment of the many for the sake of curtailing the alcohol-related problems suffered by the relatively few. In the imagery of Gusfield (1976:275), these policies fall "like sober rain from heaven upon the problem and prob- lem-free drinkers alike." There are two comments to be made in response to this critique. First, much of the social cost of excess drinking also falls on "the problem and problem-free drinkers alike," as well as the abstainers. Our government social insurance tax rates and private auto and health insurance premiums reflect, in part, the costly consequences of alcohol abuse. The drunk driver puts all of us at increased risk of an injury or death on the highway. Thus, it can be argued that an effective alcohol control measure will indirectly benefit the problem-free drinkers (as well as those who abstain) by reducing the collective costs generated by problem drinking. Second, the direct costs of alcohol control measures such as taxation are more or less proportional to the amount of alcohol an individual consumes—the "sober rain" falls on all drinkers, but with much greater intensity on the chronic heavy drinkers than on others. In the United States it is estimated that the heaviest-drinking 10 percent of the adult population consumes about 57 percent of all beverage alcohol sold each year (Gerstein, 1981:193). To the extent that alcohol taxes are propor- tional to ethanol content, then, this top group of drinkers will also pay 57 percent of the taxes. Thus, the relatively small group of drinkers who have the highest incidence of alcohol-related problems also pays the bulk of the alcohol taxes. To summarize, alcohol taxes, if effective in reducing the costly con- sequences of excess consumption, reduce the burden alcohol imposes on society at large. Furthermore, whether or not alcohol taxes are effective in reducing the costly consequences of excess consumption, they have the characteristic of exacting payment in proportion to con- sumption and hence (very roughly) in proportion to the social costs generated by their consumption. If we believe that the drinker should pay (at least in an actuarial sense) for the social costs related to his or her drinking, then the excise tax on alcohol is reasonably efficient. A second equity issue concerns the burden imposed on poor house- holds by alcohol taxes. Such taxes are of course not adjusted to the household's ability to pay (as is the income tax), and by a traditional measure alcohol taxes appear regressive. However, it is certainly pos-

TAXING AND SPENDING 31 sible that an increase in alcohol taxes will actually prove beneficial to the children of the heavy drinkers in poor families. For example, if a household's demand for alcoholic beverages is elastic (price elasticity greater than 1.0), an increase in price will cause a reduction in total expenditures on drinking, thereby leaving more money for other house- hold expenditures. Surely poor households vary considerably with respect to price elasticity of demand. However, the available evidence suggests that the average household's demand for liquor is quite elastic; further- more, poor households would tend to be more elastic than higher- income households. Therefore, for some fraction of poor households, an increase in alcohol tax rates would reduce expenditures on alcoholic beverages. And, it is quite possible that a tax-induced reduction in TABLE 3-2 Revenue Effects of Changing Federal Alcohol Excise Taxes: Proportional Increases Liquor (86 proof fifth) 1.81 3.94 7.89 6.35 14.86 8.90 Beer (six-pack^ .16 1.49 2.98 2.82 5.61 4.84 Wine (gallon, 12% alcohol)' .17 .21 .41 .41 .78 .76 Increase in revenues 0 5.64 3.94 15.62 8.86 Current "Doubling" Option "Restore to 1951 Levels" Option" Tax Tax Tax Tax Collections Collections Collections Collections Excise Tax Zero Unit Zero Unit Tax Collections* Elasticity Elasticity Elasticity Elasticity Rate (billions) (billions) (billions) (billions) (billions)' The price level in March 1983 exceeded the 1951 price level by a factor of 3.77. "From 1979. These calculations assume that the own price elasticity of demand is unity for all three beverage types, with zero cross-price elasticities. Assumed prices were $7.50 per fifth for liquor, $2.80 per six-pack, and $8 per gallon of wine. •The federal excise tax on beer is $9 per 31-gallon barrel, implying a $0.16 tax per 72-oz. six-pack. 'Wine tax rates are higher for fortified and sparkling wines.

32 PREVENTION OF ALCOHOL PROBLEMS consumption by heavy-drinking household heads may benefit other family members insofar as reduced drinking yields improved health and higher earnings. Federal Revenues and Tax Rates Currently federal excise tax collections on alcoholic beverages are about $6 billion. Table 3-2 provides estimates of the revenue effects from increasing excise tax rates uniformly on all three beverage tax rates. For example, if Congress were to double alcohol excise taxes (as they did cigarette taxes in 1982), a reasonable estimate of the revenue increase is $4 billion in the first year. The assumption behind this estimate is that the demand function for each beverage type is charac- terized by a constant elasticity of 1. (Revenue estimates for beer and wine are not sensitive to changes in this assumption.) Under the same assumption, revenue would increase by about $9 billion if Congress were to return alcohol beverage tax rates to 1951 levels in real terms. Instead of increasing all three excise tax rates proportionately, Con- gress might be well advised to standardize tax rates per ounce of ethanol across all beverage types. Under the current tax structure, liquor is taxed at $.164 per ounce of ethanol, whereas the tax on beer is equiv- alent to about $.057 per ounce, and table wine about $.011 per ounce. Raising beer and wine taxes to the level of the liquor tax would bring in more revenue than doubling all tax rates and would have the impor- tant advantage of giving official recognition to the principle that ethanol is the problem agent, regardless of the type of beverage in which it is contained. Conclusion The decline in the real value of the U.S. federal excise taxes on alcoholic beverages has benefited heavy drinkers in some ways but has had the effect of increasing the prevalence of alcoholism and its atten- dant costs. Raising the excise tax would be a rather well-targeted response to the social burden that heavy drinkers as a group impose on the rest of society. There are a number of alternatives for reducing the U.S. budget deficit. Few of them have the substantial beneficial side effects that would result from raising the alcohol excise tax rates.

TAXING AND SPENDING 33 MORE DATA ON TAX POLICY JEFFREY E. HARRIS, Massachusetts Institute of Technology My comments are fivefold. First, I show that the tax burden on alcohol has been declining at all levels of government. Second, I attempt to shed some additional light on the possible relation between alcohol excise tax rates and cirrhosis mortality rates. Third, I demonstrate that alcoholic beverage taxation is not a highly regressive policy, as is sometimes claimed. Fourth, I explain why the current design of differ- ential taxes on beer, wine, and liquor is a complex problem. Finally, I suggest that ad valorem taxation of alcoholic beverages is a practical alternative to current excise tax policy. Local, State, and Federal Tax Burdens on Alcoholic Beverages The tax burden on alcoholic beverages has declined considerably in the past two decades. From 1965 to 1980, as shown in Table 3-3, federal alcohol tax revenues declined from 3.6 to 1.4 percent of total federal tax receipts, while state and local government alcohol tax revenues declined from 1.8 to 1.2 percent of total receipts. Tax payments from alcoholic beverages also constituted a declining proportion of excise tax revenues. As shown in Table 3-1, combined tax revenues from alcohol for all levels of government fell from 14.0 percent of excise tax receipts in 1965 to 6.3 percent in 1980. When corrected for inflation, annual alcohol tax payments per adult fell by 50 percent during 1965-1980. Cirrhosis Mortality and Taxes: Cook's Argument Reconsidered Cook and Tauchen (1982) have estimated that an increase in the liquor excise tax of $1 (in current dollars) per proof gallon would reduce the mortality rate from cirrhosis of the liver by 1.9 percent in the short run and perhaps by twice that amount in the long run. With the current federal excise tax on liquor at $10.50 proof per gallon, a doubling of the federal tax rate would reduce the cirrhosis mortality by a predicted 20 percent. The effect measured by Cook and Tauchen was not estimated too precisely. Thus, the predicted 20 percent short-run decline in cirrhosis mortality has an overall 95 percent confidence range of about 5-40 percent. Still, the effect estimated by Cook and Tauchen is extraordi- narily large. A 20 percent decline in the U.S. cirrhosis mortality would be about twice the proportional decline in mortality experienced during 1968-1978. (The age-adjusted death rates for cirrhosis of the liver were

34 PREVENTION OF ALCOHOL PROBLEMS TABLE 3-3 Tax Burden on Alcoholic Beverages in the United States Year 1965- 1970- 1975- 1980- Category 1966 1971 1976 1981 Alcohol tax revenues ($ billions) Federal 3.7 4.8 5.4 5.7 State and local 1.0 1.6 2.2 2.8 All government 4.7 6.4 7.6 8.5 Alcohol tax revenues as percentage of total tax receipts (percent) Federal 3.6 3.5 2.7 1.4 State and local 1.8 1.7 1.4 1.2 All government 2.9 2.7 2.1 1.3 Alcohol tax revenues as percentage of sales and excise tax receipts (percent) Federal 25.3 24.6 24.9 11.7 State and local 5.3 4.8 4.0 3.3 All government 14.0 12.1 9.9 6.3 Per capita alcohol tax payments ($) 36 45 49 50 Per capita alcohol tax payments in constant dollars (1967) 38 38 29 19 Notes: Sales and excise taxes include general sales and gross receipts taxes; custom duties; motor fuel taxes; alcoholic beverage, tobacco products, and public utilities taxes; and other excise taxes including (in 1980-1981 only) the windfall profits tax. Per capita quantities based on noninstitutional population ages 16 years and over. Adjustment to constant dollars used the consumer price index for all items. Source: U.S. Bureau of Commerce, Government finances, selected years. Economic Report of the President 1983, Appendix B. Washington, D.C.: U.S. Department of Commerce.

TAXING AND SPENDING 35 13.9 per 100,000 in 1968 and 12.4 per 100,000 in 1978.) Thus, a 20 percent decline would mean a postponement of about 6,000 deaths annually. Is an effect of this magnitude conceivable? A doubling of the federal tax rate, if fully passed on to consumers, would constitute an approxi- mate 24 percent increase in the retail price of liquor. (For example, the increased tax would raise the retail price of a fifth of 86 proof liquor from about $7.50 to about $9.30.) Hence, according to Cook and Tauchen, each 1 percent increase in the current retail price of liquor is estimated to reduce cirrhosis mortality by about 0.8 percent in the short run alone. Social scientists have estimated that a 1 percent increase in price reduces overall alcohol consumption by about 1 percent (see the studies cited in Cook and Tauchen, 1982). If the cirrhosis mortality rate is a reason- able index of the extent of heavy drinking, then the estimates imply that heavy drinkers are about as price-sensitive as the general population. Since more than half the alcohol in the United States is bought by about one-tenth of the population (Moore and Gerstein, 1981), this may not be unreasonable. A decline in the mortality rate from cirrhosis of the liver could reflect either a decline in the prevalence of cirrhosis or a reduction in the death rate among established cirrhotics. Powell and Klatskin (1968) found the 5-year survival of cirrhotics to be 40.5 percent among those who con- tinued to drink and 63.0 percent among those who stopped using alco- hol. The differential survival in the first year after alcohol cessation, however, was only about 1 percent. The estimated difference in survival is necessarily sensitive to the stage of the disease at the time of cessation of alcohol. (Powell and Klatskin included only liver biopsy-proven cirrhotics who came to their attention as a result of certain clinical signs.) Still, the data suggest that an effect of the magnitude estimated by Cook and Tauchen would be mainly the influence of price on the prevalence of cirrhosis. Several additional lines of research are needed to fill in the gaps in the taxation-cirrhosis story. First, it would be desirable to study the effects of price at the individual level, as has been done for cigarette use (Harris, 1980, 1982; Lewit and Coate, 1982). In the case of ciga- rettes, increased taxes affect smoking participation rates more than they affect the number of cigarettes smoked by continuing smokers. In particular, a high cigarette price may also dissuade young people from starting to smoke (Lewit et al., 1981). Whether an alcohol price increase might affect behavior in these or other ways is worthy of serious study. Second, the cross-responsiveness of demand for various types of beverages needs to be studied. If an increase in the liquor tax alone would result in a substantial shift in demand to beer and wine, then the

36 PREVENTION OF ALCOHOL PROBLEMS TABLE 3-4 Alcohol Use According to Family Income, Persons Ages 17 and Older Alcohol Use (Percentage) Annual Family Income Never Use Light Use Moderate Use Heavy Use Less than $5,000 43 42 12 3 $5,000- $9,999 32 49 16 3 $10,000- $14,999 24 53 19 4 $15,000- $24,999 19 54 23 4 $25,000 or more 13 55 27 5 All incomes (incl. unknown) 25 52 19 4 Notes: Never Use: Denies current use of wine, beer, or liquor. Light Use: Up to 0.5 drinks per day. Moderate Use: Up to 2.0 drinks per day. Heavy Use: Exceeding 2.0 drinks per day. Total drinks computed from the number of drinking days per week and the number of drinks per sitting. Drinks of beer, wine, and liquor weighted equally. Source: U.S. National Center for Health Statistics, unpublished data from 22,842 respondents, Health Interview Survey, 1977. effect measured by Cook and Tauchen (which, I assume, depends only on liquor tax increases) would be even more remarkable. Third, we need to know what is happening to the prevalence of non-alcohol-related cirrhosis in the United States, particularly those forms of the disease caused by non-A, non-B hepatitis virus and by certain hepatotoxic drugs. We also need to know more about recent trends in the survival of cirrhotics, particularly with the increased avail- ability and quality of medical care since 1965. Does the cirrhosis death rate still accurately track the extent of heavy alcohol use? Finally, the relations between taxes and other alcohol-related end- points need to be studied further. Alcohol-related traffic casualties and hospital discharges, admissions to alcohol rehabilitation facilities, and deaths from alcohol overdose come to mind.

TAXING AND SPENDING 37 Tax Equity Revisited Table 3-4 shows the relation between alcohol use and family income among adults responding to the 1977 Health Interview Survey, con- ducted by the U.S. National Center for Health Statistics. My method of measuring the extent of alcohol use from these data may differ from those used in other surveys. Moreover, respondents to such surveys probably understate the extent of their alcohol use (though it is unclear whether the degree of underreporting is related to income). Still, the evidence shows that alcohol is a luxury good, that is, one for which consumption markedly increases with income. As in the case of tobacco (Harris, 1982; Harris, unpublished data), the reduced consumption at very low incomes reflects to some degree a large number of older, low- income abstainers. Even when age is taken into account, however, alcohol use increases with income. In view of the cited data it is hard to make the case that alcohol taxation is highly regressive. As in the case of cigarette taxation, an increase in alcohol taxes would be a means of raising revenues with minimal burden on the elderly. With respect to the economic burden of taxation on the family of an alcoholic or heavy drinker, Cook has noted that when a consumer's use of alcohol is sufficiently sensitive to price, an increase in price actually causes a net decrease in his total expenditure on alcohol. Therefore, among very heavy drinkers, if higher prices dissuade a sufficient quan- tity of drinking, then there is a net saving in the family budget. Taxing Beer, Wine, and Liquor Differentially The tax per unit alcohol varies widely across beverage types (Moore and Gerstein, 1981). Per unit weight of alcohol, the federal excise tax on wine is about one-fifteenth that of liquor, and the federal tax on beer is about one-third. At current prices, equalization of the tax per unit alcohol would require about a $0.30 increase in the price of a six-pack of beer and about a $2.40 increase in the price of a gallon of wine. If the damage from alcoholic beverage use depends solely on the quantity consumed, then shouldn't the tax rates per unit alcohol be equalized across beverage types? The answer is complicated. Let me give three examples: Suppose that alcohol enhanced the risk of traffic accidents most notably among those who preferred their alcohol in the form of beer. Then, other things being equal, the tax per unit alcohol on beer should be higher. Similarly, if alcohol is most damaging to health among heavy users, who preferred liquor, then the tax per unit

38 PREVENTION OF ALCOHOL PROBLEMS alcohol on liquor should be higher. Finally, suppose that the purported salutary effects of alcohol use on the risks of coronary heart disease accrue mainly to light drinkers of wine. Then the tax per unit alcohol on wine should be lower. The issue here is not whether the harm from alcoholic beverages depends, for any one person, solely on the quantity of alcohol. Instead, we need to know the correlation across consumers between alcohol- related damages and alcoholic beverage preferences (Harris, 1980). The Ad Valorem Alternative The prospect of large future budget deficits even if the current eco- nomic recovery should accelerate has heightened interest in the use of excise and sales taxes to raise new revenues. In such an economic environment those favoring the use of increased alcohol taxes to reduce alcohol-related damage should propose some form of ad valorem tax- ation. A tax levied as a fixed proportion of the wholesale price of alcoholic beverages would avoid the need for periodic new tax increases to keep pace with inflation. COMMENTS ON TAX POLICY MARY HEVENER, U.S. Department of the Treasury Dr. Cook made a very good point: much of the legislative interest in the area of alcohol excise taxes originates because they are a source of revenue. The federal liquor excise tax raises $4 billion, which is not an insignificant amount of money. But in the 2 years I have been in the U.S. Treasury's tax policy office, there has been very little legislation to review in this area. The only substantive legislation that has been subject to hearings on Capitol Hill has been a bill to repeal the existing excise tax on liquor "stills," which we supported because it cost $90,000 to administer the tax and only raised about $70,000. Still, in working on various alcohol excise tax proposals, I have been a little troubled that the major consideration in reviewing the proposals is simply dollars. One proposal that surfaced briefly in 1982 was to increase by a tiny amount the existing excise taxes on liquor in order to fund a nationwide anti-drunk-driving program. I received a call from the Hill with questions about the current level of alcohol consumption and what the effect of a truly minuscule increase in alcohol excise

TAXING AND SPENDING 39 taxes—pennies a gallon—would be in an effort to raise $30 or $40 million to fund the anti-drunk-driving program. It was an interesting proposal because it was not geared simply to raising revenue, but to funding a social program. It was very difficult to find any information on three hours' notice; as a tax attorney, my sources are largely the Internal Revenue Code and regulations thereunder, which do not yield much information about alcohol consumption or the effects of excise tax increases on cirrhosis of the liver. When congressional consideration was given in 1982 to increasing excise taxes in an effort to raise revenues, I understand that Congress briefly considered increasing alcohol instead of tobacco excise taxes (which were ultimately increased). Consideration was given to exempt- ing beer from any increase: in the opinion of the staff—formulated statistically by inquiring what people in the room thought would be the impact—the beer tax was deemed more regressive than the others, beer being "the elixir of the working man." Not much research was done by the tax lawyers who were putting together the statute; in the press of legislation, there is not time to hold hearings and elicit studies. In my opinion—not to be taken as an official position of the admin- istration nor of the U.S. Treasury—the focus in presenting recommen- dations for legislative change should be wider than simply balancing the budget; it should reflect the health and social concerns voiced here. The materials from this conference could be quite useful in preparing future revenue estimates and testimony, which, as a result, might reflect the estimated social and medical effects of any proposed alcohol excise tax change. THOMAS TURNER, Johns Hopkins Alcohol Research Center We have been hearing about tax matters here from two standpoints: revenue and health. I assume that it is not our business, as a group, to solve this country's fiscal problems. We should concentrate on health effects. It is an attractive but not proven idea that health benefits can flow from an increase in taxation. Relatively few taxes have been directly imposed for health reasons. I believe in other countries—especially Norway and Australia—it has occurred that as taxes increased, alcohol production became a cottage industry, so what is measured there might not be total consumption at all. For a decade there has been an epidemic of viral hepatitis as well as a fair amount of drug-induced liver damage in this country. These cases are now declining. There is evidence that moderate drinking has some beneficial effects. We therefore have to be careful in measuring presumed alcohol effects, including the cirrhosis

40 PREVENTION OF ALCOHOL PROBLEMS rate, in this country. I'm not certain that taxes are going to deter such effects, and the report is careful not to say "this is a proven fact." CHARLES CRAWFORD, Gallo Winery The baseline on the statement that excise taxes have not been raised was 1951. But we had no tax at all on wine during Prohibition. Between 1940 and 1951 the federal tax on table wine increased 240 percent, on dessert wine 570 percent, and on sparkling wine 750 percent. If one took 1940 as the baseline, excise taxes would appear to have kept up with other costs. I have gone over the work presented by Dr. Cook, and I believe his conclusions require more data and a longer lag time. Dr. Cook indicates that his is a "quasi-experimental approach"; I think it should not be used to make decisions on tax changes. I think we need to put far more emphasis on individual initiative and social responsibility. JOHN VASSALLO, JR., State of New Jersey It is my belief that consumption can be cut down among those who are not abusive users of alcohol, but that taxes and prices, no matter how they are raised, will not affect the drinking habits of alcoholics. Increased prices and taxes will only cause alcoholics—and for that matter social drinkers, occasional users, or persons who purchase for use at parties but not for themselves—to purchase less expensive brands. It may have some effect on the consumption of those who are relatively indifferent toward alcoholic beverages, but I do not believe it will have an appreciable effect on alcoholics. At the same time, the decrease in consumption that does occur will affect the overall tax picture. Raising taxes may not in fact create additional revenue for a state. It could even create a negative tax impact, and alcoholic beverage taxes or profits (in the control states) are very important to the economy of almost every state. ROGER THOMSON, S & A Restaurant Corporation There must be some sort of balancing. The restaurant and tavern business employs many people. The income taxes that these employees pay are substantial. Most states and the federal government, to which my own firm paid several million dollars last year, are more than happy to tax our receipts. The revenue that excise taxes raise may have some side effects, such as decreasing sales, loss of some employees, and

TAXING AND SPENDING 41 corresponding loss of revenue. There may not be as many restaurants built, and those property taxes will be lost. It is just not quite as simple as raising the taxes to gain extra money. DAVID KRAFT, University of Massachusetts, Amherst I have reluctantly been persuaded that an increase in excise taxes is a reasonable approach and will cut down on alcohol problems. My question is: How long is the effect? There may be an initial immediate effect with a tailing off, depending on how the tax legislation is written. DAN BEAUCHAMP, University of North Carolina That is true if it is a flat excise tax, not ad valorem. I think part of our charter as individual citizens is to think about whether this approach is feasible. I am impressed with how difficult it is to get a state legislature to raise alcohol taxes unless there is some kind of emergency. Various reasons are given to not raise taxes: people might cross the border to other states; legislators are concerned that the poor should not be penalized. But as long as alcohol is available, as I believe it should be, we are never going to be without alcohol problems in this society. Since we cannot eliminate alcohol problems, we should think about making them somewhat smaller. I think an excise tax increase is attractive, and it could be the most important measure in the preventive realm. It appeals to me because it is a measure for which we have some clear evidence, in an uncertain world, that two good things would happen: revenues would increase, and there would be important health benefits. This evidence is con- vincing for the improvement in cirrhosis, more tentative in the case of fatal highway crashes. Taxation has the clear advantage of working at a distance from people and accomplishing more than one end, which in politics is always good. Taxation has one other effect that we should all hold in mind: A full slate of programs that are purely educational in focus will no doubt do some good, but if a six-pack of beer in a supermarket costs roughly the same as a six-pack of Coca-Cola, all the educational programs you can design fly in the face of that fact. Soft drinks and beer now sell at roughly the same price; this was not true 15 years ago. The cost of nonalcoholic beverages has more than tripled in the same period that the cost of alcoholic beverages has less than doubled, and we now see price convergence of the two. This has much to do with government policy on sugar and alcohol. We have seen great increases in the cost

42 PREVENTION OF ALCOHOL PROBLEMS of essential goods, partly as a result of government policy, while alcohol stays cheap. Is this sound public policy? I think not. In spite of the earlier history that Mr. Crawford alluded to, the tax level in the days of Eisenhower was roughly three times what it is now. If I changed nothing else, I would change that. Prevention is a zero-sum game, in the sense that the alcohol beverage industry will have to lose something if these measures work. I can understand perfectly why the industry does not want that, but there is not much more to say than that it looks like it works. We should just confess the fact that it is going to hurt a very important, influential industry—not as much as they fear, I think, but it is going to hurt. FRANK KENEL, American Automobile Association We have done that with the automotive industry. Like it or not, auto safety has cost jobs. I have no quarrel with it; I favor what has happened in terms of the life-saving benefits. I think other industries will have to do the same thing. LAWRENCE WALLACK, University of California, Berkeley Tax policy is one piece of an overall prevention strategy that would deemphasize the use of alcohol in our society. The direction of current excise and other tax provisions actually encourages greater use of alcohol. An effective tax policy would say, "Alcohol has a special status in society because it is a drug, and the Surgeon General says it contrib- utes to (not necessarily causes) as many as one out often deaths." An effective policy would condition the environment to make it more sup- portive of those who want to drink moderately, more supportive of people who do not want to drink at all, and more supportive of parents' interests. One part of our policy dealing with alcohol should support the other parts of our policy. For example, it does not make sense to have enormous amounts of misinformation in much of the advertising we see, and then to have people working on a voluntary basis to con- vince kids that the multi-million-dollar effort telling them that drinking is good is not quite right. I think we have to look at it from many different vantage points. Tax policy is not the single answer. Anyone who thinks that there is a single answer has missed the whole point of the National Research Council report.

TAXING AND SPENDING 43 TAX POLICY AND LEGISLATIVE COALITIONS MICHAEL Fox, Ohio General Assembly As a politician and a legislator who for nine years has been trying to sort out data from groups like the National Research Council and trans- late them into public policy, it is refreshing to start with some very specific information about the consequences of actual policy options. I think the next question to be addressed as we look at tax policy, whether at the state or federal level, is: What is the context in which the battle will occur? The Context of Policy Formation The context in Washington, D.C., is a little different from the context in Ohio or Connecticut or any of the states. The policy goals are differ- ent. There is a difference between control (or monopoly) and noncontrol states. I am from one of the control or monopoly states, where we control the wholesale and retail distribution of spiritous liquors. That changes somewhat the building of coalitions in the legislative process for any policy affecting alcohol. Within the state there are certain subgroups with different objectives, which sometimes make for strange bedfellows in the pursuit of rational public policy. So I speak from my own bias and experience in a control state in discussing from a practical standpoint some of the options and some of the problems in translating research data into public policy. As I reflect on the comments of Dr. Cook and others regarding the use of tax policy to affect consumption, I have to note that two months ago I discussed the very same question with a lobbyist representing Miller Brewery. The lobbyist argued against a certain tax proposal in Ohio, saying that it would affect Miller adversely; since he was talking to a legislator who represents a county in which Miller Brewery had just built about $500 million worth of brewing facility, he was a little distraught that I would suggest taxing his product. The brewers had conducted a survey of beer consumption and public policy in 17 states. Their study found the most pronounced effect of policy on consumption in Michigan. Two events occurred simulta- neously there that resulted in roughly a 6 percent drop in consumption: a 10-cent deposit tax and an increase in the drinking age to 21. Indiana recently increased its excise tax on beer and wine and found that in the 2-3 month period immediately following the increase in the excise tax, there was a demonstrable drop in consumption. But after that period,

44 PREVENTION OF ALCOHOL PROBLEMS observed consumption started matching over time what their normal projections would have been. If the main, publicly stated objective from a policy standpoint is to reduce consumption through taxation, I believe it will create substantial problems in building coalitions in the legislature. A lot of folks make a good living selling alcoholic beverages, and they have for years con- tended that tax policy affects consumption. I spent four years on the state government committee in Ohio, and all we heard in that committee were what we affectionately called "booze bills." One after another, the tavern owners and the wholesalers would come before us and say, "You are putting us out of business, this is affecting our sales." The traditional position of the advocacy groups in our state during this period was, "No, it will not affect your business, you will still be in business, but it will affect our ability to direct resources into programs for pre- vention and intervention." I mention this as a word of caution. There is a real world out there in which legislators get beat over the head. If a bill goes to a state government committee or a similar committee in one of the legislatures, it will have a hard time getting out of committee because such committees are dominated by people who are generally friendly to the industry. A second problem is the public's perception. It has been mentioned that beer is considered the poor man's drink. When tax proposals are presented in the legislature, those who sell the product quickly team up with those who consume the product to champion the poor working person. Strange coalitions develop; when the United Auto Workers, the AFL-CIO, and other unions interested in policy development team up with industry, they seldom lack funds for getting their message communicated. The ability to influence public perception is an impor- tant consideration in translating research findings into public policy. A Coalition-Building Proposal I am involved in a proposal in Ohio to get the state out of the retail liquor business, give up its retail monopoly, and go to a "license" or "package store" concept for distribution. The legislative committee of the regional councils on alcoholism in Ohio now endorses this proposal, which they had traditionally opposed. The enticement was to put a 2- cent-a-drink tax increase before the legislature and earmark portions of that tax for treatment and prevention programs. Forty percent of that 2-cent-a-drink revenue stays local as a block grant to deal with the results of abuse of the product, a plan that brought support from the municipal league and county commissioner groups. These are the peo-

TAXING AND SPENDING 45 ple struggling to deal with legislation we passed a year ago saying, "We want drunk drivers off the road and in jail." The local governments did not have the resources to carry out that legislation, and right now it is like a meat market—"Take a number and in a year or so we will tell you if there is a jail cell for you." We put another 10 percent of the revenue directly into alcohol abuse programs, another 10 percent into school and education-based prevention, another 10 percent exclu- sively into drug prevention, and 30 percent into the state's general revenue fund. I see this proposal first as a budget and fiscal issue. The cost of running the retail distribution portion of the Department of Liquor Control is currently $35 million a year; inventory management is another $30 million a year. The net effect of getting the state out of the liquor business and imposing the 2-cent-a-drink tax—not counting the dedi- cated revenue or the revenue that would stay local or the revenue that we would earmark to alcohol prevention and intervention programs and drug prevention and intervention programs in the schools—would mean a net revenue increase for the state over a 2-year period of $221 million. We knew that tavern owners and retailers would oppose the 2-cent- a-drink tax. They have an armlock on the state government committee, and no proposal could get through that committee unless it mobilized equal firepower. So we went to the grocery chains and said, "If you want to sell this stuff, what we have to offer is the combination of a 2- cent-a-drink tax and getting the state out of the retail business." They were very cautious. They did not want the negative public relations that might arise from being out front on the issue unless they knew where Mothers Against Drunk Drivers (MADD) was going to stand. The president of the state MADD organization has endorsed the pro- posal and the other councils on alcoholism have followed suit, so we may now have the firepower to get it through. We have formed a coalition of the retail merchants with an interest in selling the product and the activist and civic official groups that are attracted to the shared revenue arrangement. Without that firepower I do not think we could dream of getting this proposal through the legislature. We have at least 40 lobbyists in Ohio who meet with legislators to tell them we ought to leave this product alone. Our coalition has reduced the opposition basically to the tavern owners, and I would fight that battle politically any time. The odds are still against the proposal, but close enough to be worth fighting for. In conclusion, policy objectives and public polls have to be translated into coalitions, which can be built on diverse bases. Revenue is one common concern. Another is the rising cost of health care. A third is

46 PREVENTION OF ALCOHOL PROBLEMS criminal legislation: chiefs of police, local sheriffs, fraternal orders of police, judges, and so forth are very interested in penal legislation. There is tremendous coalition-building potential over the concern about young people's involvement with alcohol. The opportunity to greatly expand government partnerships with the private sector, with respect to alcohol and substance abuse, has never been better. The visibility given this issue by groups like MADD and by campaigns to make the public aware of the social costs and consequences of the abuse of this product have created an opportunity to move and move quickly. If you can design a coordinated strategy to incorporate all these factors, you will have built the base, the numbers, the power to confront opposing forces in the legislature. Now is the time to gather banners that have not traditionally been used together as effectively as they could be. I do not know how long the opportunity for building coalitions will last, but I suggest that this opening be used to the fullest extent to create prevention-oriented public policy. MORE COMMENTS ON THE POLICY PROCESS DAN BEAUCHAMP, University of North Carolina As I understand it, Mr. Fox, Ohio might become the first control state since Prohibition to deregulate. I hope that does not happen. Putting together a coalition of people who mainly represent the need for money in these dire times—prevention interests and segments of the industry who have an interest in deregulation—may be smart politics in the short run, but in my opinion, it would be detrimental to the public in the long run. The first reason is revenue generation. James Mosher and I have done a study indicating that, generally, control states make much more money than they think they do from the monopoly system. They not only recoup the costs of running the warehouses and the distribution system, but also retain all the dollars that would go into profits and other things if the enterprise were totally private. In other words, they get more than just the tax; they also get the profits and other returns to private purveyors. Second, while we do not have absolutely overwhelming evidence that the control states do a better all-around job of protecting the public than the license states, I think we will find, if we develop more precise instruments, that control states do a better job of controlling underage

TAXING AND SPENDING 47 drinking. Underage purchases and falsified IDs are almost never found in the monopoly stores in my home state, North Carolina. I realize that is partly because monopoly stores do not carry beer. But I think that if you are concerned about teenage drinking, you do not want to lose the advantage you have in the control state on underage purchases. As far as I can tell, if the deal in Ohio goes through, you may have a one-time political coalition that makes sense. But in the long run you are going to lose more control of the problem. I do not know whether in the details of the bill you are going to retain your fine system in Ohio of permitting only so many outlets per capita; if you are victorious, that system is still a good idea—I hope you do not throw it out. That is one of the more sensible policies we have in the United States. So my comment is: I hope you lose. I understand your motivations politically in the short run, but in the long run the one interested group that is not at the table is the public. I have given testimony on taxing alcohol at legislative committees in North Carolina and generally I have observed three kinds of people there. First, there are many, many people from the industry. Second, there is the alcoholism constituency, which seeks money. Third, there are researchers interested in the effects of the total system on the health and safety of the public. The public itself is simply not paying attention to this issue, and you cannot mobi- lize them. I'm afraid a deregulatory environment is going to destroy 20 good systems in this country for controlling alcohol by making short- term deals that are politically smart but in the long term are going to be bad for the public. MICHAEL Fox, Ohio General Assembly I understand your goal and I think that if you read the legislation you would find that we agree rather than disagree. I cannot go into all the details, but let me note some of them. First, on the revenue side, we will technically keep Ohio in the wholesale business. We will do that for reasons not related to alcohol policy, but to our liquor profits bonding program for economic devel- opment. We will retain—and the legislation requires that we generate in addition to the tax revenues—an amount of profit equal to the last fiscal year prior to the transition. Last year, that was a little over $103 million. We will retain that profit factor. The impact, in terms of the goal we are talking about here, will be a price increase, on the shelf, of about 20 percent. For the consuming public, who may think that pack- age stores mean lower prices, that will not be the result in Ohio. That is the net effect when we add a 2-cent-an-ounce tax on liquor and

48 PREVENTION OF ALCOHOL PROBLEMS maintain the policy that we retain wholesale profits equal to retail profits. Second, no study indicates that the mode of distribution affects con- sumption—at least none that I have seen; if you have one, I want to see it. With regard to sales to minors in state-controlled stores, I do not know how things are arranged in North Carolina, but in Ohio the reason we do not have instances of violation in state-controlled stores is that in the last decade the Bureau of Liquor Control and the agents inves- tigating false sales or sales to minors have been in the same department and have a private arrangement whereby they just do not "set up" state stores by sending in underage purchasers, which is the standard practice to get evidence solid enough to make a bust. There has not been one arrest in 10 years in our state system. Finally, with respect to that sector of the public you think is not listening, I suggest that with the attention this issue has gotten nationally the public is now waiting to be mobilized, and MADD and other advo- cacy groups have already demonstrated their ability to do this. To achieve the goals we are talking about here, as a realistic matter, one has to be ready to compromise. To go into those committees and not have that coalition built, no matter how worthy the cause, is to lose. The issue of increasing tax revenues is essential at this point for one key reason. Policies adopted by state legislatures across the country to get drunk drivers off the road will have sustained effects only to the degree that funds start to flow to the local communities to make sure that the policies can be carried out. We are seeing the short-run impact now in Ohio, where we recently enacted mandatory jail terms for drunk drivers. There is a fear that drunk drivers are going to be caught, and a lot of publicity about the courts breaking down under the load. But it is starting to seep through that we have a waiting list for jail cells. The pendulum will swing back to the other side, to water these efforts down and go back essentially to business as usual unless resources can be raised. ALFRED MCALISTER, University of Texas Health Sciences Center I want to illustrate that point further. In Texas, because of decreasing oil revenues, we have a mad scramble to find new revenues for the first time in many years. Our governor has seized on what we in Texas call sin taxes, referring of course to alcohol and tobacco taxes. It has become very clear that the political acceptability of such a tax increase depends on whether the public perceives some connection, an earmarking or diversion of a portion of the funds, to prevention programs. It is in a

TAXING AND SPENDING 49 sense similar to the way that the administration in Washington ration- alized the increased gasoline tax as a user fee. We are finding good evidence that not only the treatment interests, for example, that seek this money, but also the average taxpayer sees some political accept- ability in it. Are there survey data to that effect? GEORGE HACKER, Center for Science in the Public Interest A number of surveys show that, without any prior education, a clear majority of the public favors increased, earmarked revenues from tax- ation of alcoholic beverages. Several surveys—one in South Carolina, one done by Gallup for the National Council on Alcoholism in New York, and the Harris polls—show that there is now growing public awareness, and non-alcohol-related organizations are developing an interest in alcohol tax and advertising issues. For example, we at the Center for Science in the Public Interest are now coordinating the National Alcohol Tax Coalition, which involves the National Council on Alcoholism, drunk-driving prevention groups, and dozens of children's, elderly, family, consumer, health, church, and other non-alcohol-related groups. We have begun to act at the federal level to take advantage of deficit-reduction opportunities and the heightened public awareness about drunk driving. I think it is impor- tant that we not sit too long without recognizing that there are oppor- tunities to mesh these diverse interests around a common interest. ALFRED MCALISTER, University of Texas Health Sciences Center Of course, the fate of these issues often boils down to what the head of the PTA and legislative lobbyists for health or medical associations and other pressure groups think, as opposed to what the voters think. People have a naive perception that the more the government earns on drinking, the less interest it may have in controlling the problem. There is a problem of voter education, of educating people and influencing their attitudes on these issues. One possible way of broadening the discussion would be to agree that if taxes are raised there is some logic to earmarking or diverting a portion of them to prevention. We could use a portion of any tax increase for a nationwide anti-drunk-driving campaign, for example. If a tax increase is not packaged right, the industry will try to make the drinker feel that it is onerous to him or her. In Texas we had a smoking campaign; the newspapers reported, "Health Group Wants Money from Smokers." It was being presented in the form of someone robbing someone else.

50 PREVENTION OF ALCOHOL PROBLEMS JANE SMITH PATTERSON, State of North Carolina I am very concerned with legislators' perceptions of the alcohol issue. When dealing with a legislature, it is important to get across in a short amount of time, to people who come from very different backgrounds, the information necessary to make public policy. The task is something like talking to the child who returned from a prayer meeting and told her mother not to worry, the minister had said her quilt was on the way. The mother was puzzled, but the child insisted that that is what the minister had said. The next day, her mother learned that the text of the minister's lesson was, "Fear not, for thy Comforter is coming." Per- ception is critical. One of the critical things is directing alcohol taxes into prevention efforts. In our state, we require at least 25 percent of the funds used on the alcohol effort in the total budget to go directly into prevention efforts. We would like to direct more into prevention efforts. This emphasis on prevention must come directly from the funding source. Substantial responsibilities are now shifting to the states; therefore, substantial numbers of lobbyists are shifting to the states. In our current legislative session, more than 20 lobbyists are paid by liquor, beer, and wine interests. That is a considerable number to spend time persuading legislators of their particular points of view. I think this is especially relevant to the proposition for equal taxation of alcohol products based on the amount of ethanol. This is appealing in that it recognizes that ethanol itself is the problem with these bev- erages, and therefore appears more equitable than the way we address the issue at present. But there is an even more important point for anyone working with a legislature: whatever the reasons for proposing this legislation—revenue, prevention efforts, or both—it splits the var- ious lobbying interests from each other. Divide and conquer is some- thing to use in trying to get any bill through a legislature. To talk about dedicating federal excise taxes to the states to be spent on prevention is one thing. But at the state level, legislatures and governors across the country flatly dislike the earmarking of state rev- enues and will fight you even though they agree with what you want to do. They do not want funds to be dedicated in the actual excise tax act. They want the funds raised first; then the legislature will say, for exam- ple, that they want to spend $25 million for alcoholism and drug abuse and 45 percent of that for prevention.

TAXING AND SPENDING 51 LORNE PHILLIPS, State of Kansas But in that case you will not get the money. It becomes part of the social health dollar. When advocacy groups and agencies start lining up their handicapped and retarded children and all the other serious health problems for consideration, and we line up our alcoholics, we will not get any dollars. JANE PATTERSON, State of North Carolina You will just have to sell your state legislature on prevention. CHRIS LUBINSKI, U.S. Senate I think we need to emphasize that taxes are a way of raising money to deal with the problem. As a Senate staff member, I deal primarily with human services. You need not work for the Senate to know that there are fewer funds available to deal with any kind of social problem. Despite the large, recent increase in the alcohol research budget, treat- ment and prevention funds have been cut severely. It takes money to prepare and print pamphlets, to disseminate information from research- ers and physicians in a way that is comprehensible to parents and children. Whether alcohol taxes decrease consumption overall is less important than that we might be successful in getting a 10 or 15 percent set-aside from the funds raised. We should not set treatment against prevention. We need to treat alcoholics, including youthful alcoholics. We do not want to rob Peter to pay Paul, to take money away from people who are being damaged in order to educate kids. The only way we are going to increase funding substantially on a local or federal level is to generate revenues through user fees. I cannot think of a more appropriate way to do it. If we don't do that, the funds will be taken from another critical area of the federal budget, such as AFDC or food stamps. The issue of excise taxes, rather than being seen mainly as a political means to reduce consumption, should be viewed as a way to raise necessary funds to deal with this problem on all levels. LORNE PHILLIPS, State of Kansas In talking about taxes, it is important to think about the trade-offs that might have to be made, which may look good in the short run but do damage in the long run. When we sought new taxes a few years ago

52 PREVENTION OF ALCOHOL PROBLEMS in Kansas, it looked good initially because we were historically a dry state. Then we went through a process of licensing private clubs, and that bill included a 10 percent drink tax, which was a real gain. Then we sought a beer tax, and the beer industry wanted certain things in return that the legislature was willing to give them: longer hours, Sunday sales, etc. People in the alcohol field finally backed away from the new tax because what would have to be given up in the way of reduced restrictions on availability was more than we felt we could afford. This is something to look at: Is a tax of so-many-cents-a-drink going to be enough in the long run to pay for what we might end up losing in controls? DAN BEAUCHAMP, University of North Carolina I have never understood the alcoholic beverage industry on this point. If I were in the industry, I would encourage this earmarking practice, because it will always lead to a very small tax. A little tax increase raises a lot of money in this field. If we doubled the excise tax and dedicated it all to alcohol-related programs, we would have to put alcoholism treatment centers in everybody's house, we would have so much money to spend. Money for research on alcoholism has gone up dramatically, and the alcoholism constituencies in the states are fairly strong; they are hanging on to their share. I am not so much concerned that we are not going to get enough for them. CHARLES CRAWFORD, Gallo Winery Funds need not flow from a government in order to do good. There are other ways. Ernest Gallo has put $3 million into the Ernest Gallo Clinic and Research Center at the University of California, San Fran- cisco, and the San Francisco General Hospital. The research center is trying to develop treatment for withdrawal symptoms, and they are trying to find markers for alcoholism. I think that $3 million will be followed by a lot more money because Mr. Gallo is extremely interested in the question—why do some people become alcoholics and not oth- ers? If, through research, a biological marker can be found to identify potential alcoholics before they ever take a drink, we could save billions of dollars and we would not need tax money. Raising excise taxes is not necessarily prevention. Since 1975 the National Institute on Alcohol Abuse and Alcoholism (NIAAA) and other alcoholism groups have spent a lot of money. To raise excise taxes to spend more money in the

TAXING AND SPENDING 53 same direction and not achieve any results could be unfortunate. But if industry itself becomes socially responsible and spends the money, they can get a dollar's worth for every dollar; when governments collect excise taxes they get five cents on every dollar. BARRY SWEEDLER, National Transportation Safety Board I do not think we should discount the excise measure. The Swedes, for example, definitely feel that the increasing price of alcohol there has kept consumption down. The Swedish Council for Information on Alco- hol and Other Drugs (1982:14) states: "The taxation of alcoholic bev- erages . . . has also served as perhaps the most powerful instrument of a temperance policy designed to keep consumption within reasonable limits." The user fee analogy has been effective in the drunk-driving area; some states require drunk drivers to pay for their own treatment programs. FRANK PADAVAN, New York State Senate In New York State we doubled our fines, and half the money goes back to the counties for education and prevention programs. MARILYN GOLDWATER, Maryland House of Delegates I think the key to raising the tax is to get a set-aside. On the federal level that can be done. At the state and county levels, it depends on policy, on what the constitution allows, and on history. Once a few states pass new laws, other states will follow. The first few states have all the difficulty in doing the research necessary to argue effectively for innovative legislation. JOHN NOBLE, National Institute on Alcohol Abuse and Alcoholism Tax monies already support treating people who are alcohol abusers and alcoholics. One contribution of identifying tax revenues that deal with consequences is to fuel a trade-off argument: we could stop paying so much for treating the consequences of alcoholism if we could spend some of that money on having fewer people who need such treatment. This cost argument is often made in employee assistance programs. NIAAA can supply baseline data, such as county-level information regarding alcohol consequences.

54 PREVENTION OF ALCOHOL PROBLEMS ROBERT Ross, State of New York Everyone would probably agree that, in making policy decisions, it is better to have more research than less. But from a practical point of view, decisions are sometimes made, by a state legislature for example, without the ideal data. To give a practical example: for a number of years we have been trying to get the New York state tax on alcohol raised, with some portion of that tax dedicated to alcohol prevention and education. Many argu- ments were used against it: "We don't have enough data, we don't know this, we don't know that." Suddenly we had a $1.8 billion gap in the budget, and this year a $32 million increase in the tax on alcoholic beverages was passed. But it was passed without answering such ques- tions as: On what beverages will it be placed? What is it going to be used for? Is it only going to be general revenue or is it going to be general revenue plus usage for prevention activities? Are we going to attempt to evaluate the impact on consumption? It may not have been a bad decision in terms of revenue, but a wonderful opportunity was lost when the decision was made to increase the tax on alcohol by $32 million and do nothing on those issues, instead of raising it $36 million and putting $4 million more into the prevention system. One could not say that the beverage patrons could accept a $32 million increase but not a $36 million increase, without going out of state for purchase and consumption. There is no rational argument that anything more than $32 million would have been overbearing. The legislature made that decision without benefit of a convenient body of information that said: "If and when you get ready to pass this increase, here are the questions that you should address, here are the kinds of things you should do." They have heard all the arguments individually over a period of time from the Division on Alcoholism and Alcohol Abuse and other people who are interested in alcoholism issues in New York, but they did not have a nice compact 8- or 10-page statement with appropriate references that said, "If you are going to change policy, here is what you ought to do, here are the people you ought to consult." These kinds of opportunities are very frustrating, because the tax may not be changed again in the near future. We have all lost an opportunity, and I do not think anyone has gained anything on the other side—it was done for expediency and everyone understood that. But with the right kind of support, some sort of prevention study to see what the impact would be could easily have been provided for.

TAXING AND SPENDING 55 HENRY KING, United States Brewers Association That example is not unique. I have directed the legislative lobbying activities of the brewing industry for 21 years, and I cannot recall many instances of states raising the tax on alcohol in which either our oppo- nents' or our arguments were fully considered on their merits. We have all done the best for our constituencies according to what we believed was our proper role, but it always comes down to economic budget considerations and the interplay of political forces. That is the way the process works. We live in a political environment in which deals go on in a legislature that have nothing to do with the ultimate merits of an issue. The industry, incidentally, is not at all opposed to many things that have been suggested over the years; for example, the industry supports the interim recommendations of the Presidential Commission on Drunk Driving and is actively trying to generate legislative support at the state level. I have sat for 14 months on the Presidential Commission on Drunk Driving and have seen no data that would give me a conscientious reason to vote against a 21-year-old minimum purchase age, for exam- ple. I find this conference extremely helpful, to talk to one another directly. I would have been afraid to sit across the table from someone who suggested that we should limit the number of licenses. Maybe we should, but we need a lot more research, information, and dialogue. BEN MASON, Adolph Coors Company I have some concern that we may be on the wrong track. Let me give you an analogy. I worked in education policy for a number of years and saw educators seduced into addressing tax policy issues at the local level, becoming terribly embroiled in issues of equal access, equal opportunities, equal burdens, almost to the exclusion of focusing on the purposes of education. Now we articulate rather well the tax policy issues related to education, though decisions are made to a great extent as Henry King described. However, we do not focus very well on what we expect education to accomplish. As a person interested in primary prevention, I am concerned that the debates over tax policy at the state level are long, terribly complicated, and politically brutal, and dilute the focus of what we want out of primary prevention programs. I do not think there is enough collective wisdom at this conference or most others to be persuasive in the area of tax policy, whereas we could be

56 PREVENTION OF ALCOHOL PROBLEMS persuasive in drawing conclusions about primary prevention program accomplishments. Certainly we should develop a variety of approaches to prevention. But given the amount of data we have about tax policy, I am concerned that we not get sidetracked by something that takes energies away from the real objective of reducing the harmful effects of abuse. SHEILA BLUME, National Council on Alcoholism I am not moved by the argument that we should not talk about taxes because we should talk about something else. It seems to me that prevention policy is too often based on "either/or." "If you are inter- ested in school education, then you cannot possibly talk about taxes." Or "if you are training servers then you are forgetting about the people who are drinking at home." There is no single way to take hold of everything at once. If we ever develop a rational prevention policy, it will have hundreds of items in it, some more and some less effective, but they will all have to be done.

Next: ENGAGING THE BUSINESS SECTOR »
Toward the Prevention of Alcohol Problems: Government, Business, and Community Action Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!