Health economics.

Grossman, Michael

Health economists long have realized the importance of the distinction between health as an output and medical care as one input in the production of health. This distinction is reflected by two programs of health research at the NBER: the Program in Health Economics that I direct, and the Program in Health Care that Alan M. Garber directs. Current and recently completed studies by the members of my program focus on the determinants and effects of three unhealthy behaviors: excessive alcohol use, cigarette smoking, and the consumption of illegal drugs.

Alcohol Control Policies, Alcohol Use,

and Motor Vehicle Fatalities

Since the mid-1970s, federal, state, and local governments have been involved in a campaign to reduce motor vehicle fatalities by discouraging alcohol abuse. Much of this campaign has focused on teenagers and young adults because motor vehicle accidents are the leading cause of death of persons under the age of 35, and because alcohol is involved in over half these fatal crashes. The major element of the campaign against adolescent alcohol abuse has been the Federal Uniform Drinking Age Act of 1984. This act pressured all states into raising the minimum legal drinking age to 21 by withholding part of their federal highway funding if they failed to comply. Currently, all 50 states and the District of Columbia have a drinking age of 21.

Other measures of the antidrinking campaign have been directed at all segments of the population. For example, the Alcohol Traffic Safety Act of 1983 provides financial incentives for states to enact and enforce new, more stringent drunk driving laws. It resulted in approximately 500 new state and local laws. Public Law 100-690 required that, as of November 1989, all alcoholic beverages sold in the United States carry warning labels alerting consumers to the dangers of drunken driving and drinking during pregnancy.

Increased taxation, which results in higher prices, is another policy that might significantly reduce alcohol abuse among young people. Yet this approach virtually has been ignored in the antidrinking campaign. In January 1991, the federal excise tax rates on beer and wine were increased for the first time in almost 40 years. in addition, the federal excise tax rate on distilled spirits was raised in 1991, for only the second time since 1951. Like the federal government, state and local governments have raised taxes on alcohol modestly and infrequently, almost always with the intent of increasing revenues rather than discouraging alcohol abuse.

In part because of the stability of these taxes, the real prices of alcoholic beverages (their prices after accounting for the effects of inflation) have declined significantly over time. For example, between 1975 and 1990, the real price of distilled spirits fell by 32 percent; the real price of wine fell by 28 percent; and the real price of beer fell by 20 percent. Moreover, the recent federal beer and spirits tax hikes were much smaller than the fivefold increase for beer and the fourfold increase for spirits that would have had to occur to achieve the 1951 tax rates in real terms.

In previous research, my colleagues and I found that youth alcohol consumption, excessive consumption, and motor vehicle accident mortality rates were inversely related to alcoholic beverage prices and state excise tax rates on alcohol.(1) This research used microdata for the 1970s and a time series of states for 1975-81. It capitalized on substantial variations in the prices of alcoholic beverages among U.S. states, caused primarily by differences in state excise tax rates on these beverages.

Frank J. Chaloupka, Henry Saffer, Adit Laixuthai, and I have updated the studies on youth alcohol use for 1982 and 1989.(2) Our data come from the nationally representative surveys of high school seniors conducted in those two years by the University of Michigan's Institute for Social Research. We find that increases in beer prices or beer excise taxes lead to substantial reductions in both the frequency of youth alcohol consumption and in heavy drinking among youths. We use our estimates to simulate the effects of indexing the federal excise tax on beer to the rate of inflation since 1951. Based on the 1989 survey, it appears that indexing taxes would have caused a 20 percent decline in the number of youths who drank frequently (that is, on more than 30 occasions) in the past year (13 percent of the 1989 sample), an 11 percent drop in the number of youths who drank frequently (on more than 9 occasions) in the past month (10 percent of the sample), and a 7 percent fall in the number who drank heavily (at least one drinking episode during which five or more drinks were consumed) in the past two weeks (32 percent of the sample).

Chaloupka, Saffer, and I also have updated the research on motor vehicle fatalities of persons aged 18 to 20 in 1982 through 1988.(3) We consider the effects of beer taxes, legal drinking ages, and a variety of policies to deter drunk driving on motor vehicle fatality rates. We find that the lives of approximately 5000 persons of all ages, including approximately 1700 youths age 18 to 20, would have been saved in a typical year if the beer tax had been indexed to the rate of inflation since 1951.

A uniform minimum legal drinking age of 21, on the other hand, would have saved the lives of approximately 200 youths per year. Of course, the drinking age effect is small in part because many states already had drinking ages of 21 in 1982-8. Raising the legal purchasing age in all 50 states from 18 to 21 would have saved 664 lives. Measured in lives saved, this is approximately equal to the impact of doubling the beer tax, and only two-fifths as large as the effect of indexing the beer tax to the rate of inflation since 1951. As with the studies of youth alcohol use, our research on motor vehicle accident fatalities leads us to conclude that increases in alcoholic beverage taxes are an effective means of reducing youths' drunken driving and related deaths in traffic crashes.

Empirical Analysis of Cigarette

and Alcohol Addiction

Studies by Gary S. Becker, Chaloupka, Kevin M. Murphy, and me challenge the conventional wisdom that harmful addictions, such as cigarette smoking and alcohol abuse, are insensitive to price.(4) In our empirical work we allow for the possibility that addicts may be rational: that is, that they take account of the harmful future effects of their current decisions when they determine how much of an addictive good to consume. The demand function of a rational addict is symmetric: increases in past or future consumption (caused by reductions in past or future prices) cause current consumption to rise. Put differently, the reinforcement property of an addictive good, which is emphasized by psychologists, suggests that an increase in past consumption raises the marginal benefit of current consumption. By symmetry, an increase in future consumption also raises the marginal benefit of current consumption.

Becker, Murphy, and I estimate the demand for cigarettes using state data, and Chaloupka performs a similar exercise with micro panel data. All four of us collaborate on a study of excessive alcohol consumption, measured by the mortality rate from cirrhosis of the liver. We find evidence of rational addiction, since an increase in future consumption causes an increase in current consumption. We also find that responses to price are larger in absolute value in the long run than in the short run. Moreover, our estimates of long-run price elasticity (-.75 for cigarettes and -1.06 for liver cirrhosis) are higher than those in research that ignores addiction. The message from our studies is that it is a mistake to assume that the behaviors at issue are insensitive to price.

Other Alcohol Studies

Previous studies have found that alcohol advertising has modest effects on consumption. Given the small variations in alcohol advertising in these data and the possibility of a nonlinear advertising effect, these studies cannot be used to predict the impacts of a ban on broadcast advertising. Saffer capitalizes on the existence of such bans in some OECD countries but not others during 1970-83, and concludes that bans may have substantial effects.(5) Countries with bans had 16 percent lower alcohol consumption and 10 percent lower motor vehicle fatality rates.

There is controversial medical evidence that moderate alcohol consumption reduces deaths from coronary heart disease. Using the First National Health and Nutrition Examination Survey and its ten-year followup, Coate shows that this effect disappears when persons with heart problems at baseline are omitted from the analysis.(6) Thus, his study underscores the importance of controlling for causality from heart problems to abstinence.

Determinants and Consequences

of Illegal Drug Use

During the past year, we have undertaken five new projects dealing with the determinants and consequences of illegal drug use. Saffer and Chaloupka are studying the effects of fines, penalties, and probabilities of punishment for the use of illegal drugs on crime, drug use, and motor vehicle mortality. Chaloupka, Charles C. Brown, and I are estimating the demand for illegal drugs using drug prices obtained from the Drug Enforcement Administration of the U.S. Department of Justice. Theodore J. Joyce, Andrew D. Racine, and I are conducting a cost-benefit analysis of a prenatal substance abuse treatment program run by Kings County Hospital in Brooklyn, New York. In a related project, the three of us are investigating the neonatal care costs of maternal substance abuse using data on women who gave birth in New York City municipal hospitals. Hope Corman and H. Naci Mocan are studying the effects of drug use on crime in New York City using monthly data on crime rates and drug-related deaths (a proxy for drug use) for 1970-90.

Low birthweight, defined as less than 2500 grams (5.5 pounds), is the most important proximate cause of infant mortality. Moreover, low-weight infants who survive the first year of life are more likely to encounter health problems in childhood and adolescence and impaired cognitive development and school performance. Therefore, it is alarming that the number of black low-weight births rose by 19 percent from 1984-9 in New York City. Some have attributed most of this increase to cutbacks in public programs during the 1980s. Using detailed New York City birth certificates for 1980-9, Joyce, Racine, and Mocan tell a very different story.(7) They find that more than half of this increase can be explained by the 175 percent rise in maternal drug users from 1984-9.

(1) M. Grossman, D. Coate, and G. M. Arluck, "Price Sensitivity of Alcoholic Beverages in the United States," in Control Issues in Alcohol Abuse Prevention: Strategies for States and Communities, H. D. Holder, ed. Greenwich, CT: JAI Press, 1987, pp. 169-198; D. Coate and M. Grossman, "Effects of Alcoholic Beverages Prices and Legal Drinking Ages on Youth Alcohol Use," NBER Reprint No. 1054, October 1988, and Journal of Law and Economics 31, 1 (April 1988), pp. 145-171; and H. Saffer and M. Grossman, "Beer Taxes, the Legal Drinking Age, and Youth Motor Vehicle Fatalities," NBER Reprint No. 913, December 1987, and Journal of Legal Studies 16, 2 (June 1987). pp. 351-374. (2) F. J. Chaloupka, H. Saffer, and A. Laixuthai, "Alcohol Price Policy and Youths: A Summary of Economic Research," Journal of Research on Adolescence, forthcoming. (3) F. J. Chaloupka, M. Grossman, and H. Saffer, "Alcohol Control Policies and Motor Vehicle Fatalities," NBER Working Paper No. 3831, September 1991, and Journal of Legal Studies, forthcoming. (4) G. S. Becker, M. Grossman, and K. M. Murphy, "An Empirical Analysis of Cigarette Addiction," NBER Working Paper No. 3322, April 1990, revised March 1993; F. J. Chaloupka, "Rational Addictive Behavior and Cigarette Smoking," NBER Reprint No. 1668, December 1991, and Journal of Political Economy 99, 4 (August 1991), pp. 722-742; and F. J. Chaloupka, M. Grossman, G. S. Becker and K. M. Murphy, "Alcohol Addiction: An Econometric Analysis," presented at the annual meeting of the American Economic Association, January 1993. (5) H. Saffer, "Alcohol Advertising Bans and Alcohol Use: An International Perspective," NBER Reprint No. 1685, January 1992, and Journal of Health Economics 10, 1 (May 1991), pp. 65-79. (6) D. Coate, "Moderate Drinking and Coronary Hearth Disease Mortality," American Journal of Public Health, forthcoming. (7) T. J. Joyce, A. D. Racine, and H. N. Mocan, "The Consequences and Costs of Maternal Substance Abuse in New York City: A Pooled Time-Series, Cross-Section Analysis," NBER Working Paper No. 3987, February 1992, and Journal of Health Economics 11, 3 (October 1992), pp. 297-314.