The federal tax on alcohol should be as much as five times greater than it is now to reduce alcohol problems in this country, where about 10 percent of adults are problem drinkers, according to a Cornell study.

This optimal tax, which would equal the price of the alcohol before tax, would significantly reduce the 100,000 deaths attributed to alcohol each year without overburdening consumers who do not abuse alcohol, says Donald S. Kenkel, Cornell associate professor of consumer economics and housing in Cornell's College of Human Ecology.

In a related study, Kenkel found that young men who abuse alcohol tend to earn 30 percent less and are 15 percent less likely to marry than their light-drinking counterparts, and young women who drink heavily are up to 45 percent less likely to marry.

To analyze the ideal tax for alcoholic beverages, Kenkel merged alcohol prices, statistics on drunk driving and laws, alcohol consumption and the rate of heavy drinking (defined as more than five drinks daily), and balanced the effects a heftier tax would have on heavy drinkers compared to its burden on moderate drinkers.

His analysis will be published in Economic Inquiry in early 1997. A version for consumers was published recently in Cornell CONSUMER CLOSE-UPS, a Cornell Cooperative Extension publication.

Kenkel also estimated the optimal tax if punishment for drunk driving were more certain and severe and consumers were better educated about the devastating health effects of alcohol. In this case, an effective and optimal alcohol tax rate would be about 42 percent of the net-of-tax price (i.e., 42 percent of the alcohol price before tax). The current rate is 20 percent net-of-tax.

"I believe that one of the most important things we can do to improve the health of Americans is not in the area of health care but in personal health habits," said Kenkel, a health economist with strong interests in public policy issues.

Kenkel said that studies by other researchers in the early 1980s had shown, surprisingly, that alcohol tax can have a significant impact on cirrhosis rates, traffic fatality rates, the purchases of alcoholic beverages and self-reported heavy drinking and drunk driving. "The tax rate I propose, which is significantly higher than those of most of my colleagues, would reduce the amount of heavy drinking, drunk driving and the costs of heavy drinking on society," Kenkel said.

Although the tax on alcohol was much heavier (more than 50 percent of the net-of-tax price) in the 1950s, Kenkel argues that it has become proportionately smaller because of inflation and lags far behind the tax on cigarettes.

In the past decade, tougher drunk driving laws have reduced the number of traffic-related fatalities from 50 percent of accidents involving alcohol to a current rate of 40 percent. "However, despite tougher laws and a national drinking age of 21, alcohol problems in this country remain a significant threat to public health," said Kenkel, who teaches evaluation of public policies and a new course on the economics of health behavior and policy.

In a separate study, Kenkel and David C. Ribar of Pennsylvania State University, examined the socioeconomic consequences of alcohol use. Analyzing data on 12,686 individuals, ages 14 to 21 in 1979, from the national Longitudinal (1979 through 1990) Survey of Youth, the health economists found that the likelihood of marriage among young men who were either alcohol dependent or alcohol abusive was reduced by 15 and 12 percent, respectively. Among women, the negative effects of drinking on earning were insignificant but on marriage, they were two to three times greater than the effects for men.

"It's clear that men and women in their twenties with drinking problems are much less desirable as potential spouses and that young men have a significantly lower wage growth potential and have worse job benefits than other men," Kenkel said. His findings were presented to the 1993 Microeconomics Panel Meeting for the Brookings Papers on Economic Activity and published in Brookings Papers: Microeconomics 1994.

Both projects were supported by the National Institute on Alcohol Abuse and Alcoholism.

In a third research project, Kenkel analyzed how companies can increase their profits and workers' productivity by offering on-site alcohol prevention and treatment programs and why some companies offer such programs and others do not. That work is under review at the Journal of Studies on Alcohol.

The toll of alcohol abuse in this country costs $98.6 billion to the U.S. economy, the lives of about 6,000 people killed by drunk drivers and hundreds of thousands of injuries from alcohol-related traffic accidents, not to mention the debilitating health effects of chronic alcohol abuse. About 26,000 people died of cirrhosis of the liver in 1988, making it the ninth leading cause of death in the United States. Alcohol also is involved in almost half of all fatal car crashes and believed to be a factor in many other accidents, homicides and suicides, Kenkel said.