The Volokh Conspiracy - Price Matters:

There is a "national effort to reduce alcohol related problems," so says the mission statement of the federal agency charged with researching these problems and disseminating the findings. I'm willing to believe it, but would like to pose a question: in pursuing this national effort, are lower prices of liquor and beer likely to help or hurt? Because for better or worse, price reductions are the de facto policy of the federal government.

Here's the story. Federal alcohol excise taxes have been an important component of alcohol prices, and particularly prices of distilled spirits, since Repeal (or really since the Civil War, with time out for Prohibition). The federal tax is set as a flat amount per unit of ethanol, regardless of value.

For example, in 1951 Congress set the tax at $1.68 per fifth of 80 proof liquor. In today's dollars, that's the equivalent of $13.50 per fifth. But Congress has only succeeded in raising the tax twice since 1951, and by meager amounts, so that instead of $13.50, the current tax is just $2.16 per fifth.

Even if there were no markup on taxes (and there in fact is), the result is that the current price of a bottle of spirits is over $10 lower than it would have been if Congress had simply indexed the liquor tax to the Consumer Price Index in 1951 and then left it alone.

The states have also been slow to raise their nominal tax rates on liquor and beer. The result — inflation has deeply eroded the value of alcohol excise tax rates. It's not surprising, then, that the price of spirits (adjusted for overall inflation) has been falling over time. Just in the last 25 years, the price for package sales has declined 12%. The price of a 6-pack has also declined, by about 8%.

Meanwhile, the price of cigarettes has been following a quite different trajectory, thanks in large part to state and federal tax increases since the Master Settlement Agreement of 1998. Over 40 states have raised their tax by more than 25 cents, and 11 have raised their excise tax by more than one dollar per pack. What's going on here? Why is one "sin tax" so popular politically, while another has been largely neglected?

These days, increases in the cigarette tax are being touted as a public health measure, sure to reduce initiation by teens and help adults to quit or cut back. While economists have long supported this conclusion with direct empirical evidence, the idea has been difficult to sell to the public and the legislators. Smoking is addictive, and addicts will get their "fix" regardless of the cost, right? But now it seems that everyone is a convert to the power of taxation and price in the campaign to reduce smoking.

As it turns out, the evidence that higher prices discourage alcohol consumption and abuse is strong, and of just the same sort as the evidence that supports the conclusion that higher prices reduce smoking.

The most persuasive evidence comes from the laboratory of the states. Since 1981, I, my collaborators, and other economists and epidemiologists have analyzed the effect of state tax increases on alcohol sales but also on other outcomes, such as mortality due to alcohol-related causes. The results consistently favor the view that taxes (through their effect on prices) matter.

The legislators and the public are not buying it. For some reason, there is profound resistance to the idea that beverage alcohol is a commodity with the usual downward sloping demand curve. The alcohol industry knows better -- they must believe that higher taxes can't be passed on to consumers without a reduction in sales and consumption, or else why would they fight tax increases so fiercely? But when I make the argument to non-economists, I get a skeptical hearing.

The more refined skeptics accept the premise that higher prices lead to a reduction in sales, but speculate that that reduction is entirely due to the behavior of moderate drinkers — those who do not drive drunk or abuse their children or lose productivity to hangovers or do long term damage to their organs.

In this scenario, the abusers have zero elasticity, while moderate drinkers are price elastic. To which I could point out (and do) that in fact it is these heavy drinkers who should be most likely to respond to price, since they're the ones for whom drinking places a real dent in the household budget.

But ultimately my belief in the efficacy of tax as a basis for controlling abuse does not rest on such qualitative arguments -- it is based on the empirical evidence. In the 1960s there were actually some experiments on the effects of price, including several conducted in alcoholism-treatment clinics. Token economies were set up that offered the patients drinks at a price -- either a certain amount of "work" or loss of privileges.

These experiments demonstrated that alcoholics were price sensitive. But they have not been replicated, since offering drinks to alcoholics is a tough to sell to the Institutional Review Boards these days.

Much more directly relevant are the studies based on "quasi-experiments," including the dozens of instances in which states have changed their tax rates. We can observe what happens to an outcome variable (for example, a mortality rate) in states that raise their tax compared with states that don't raise their tax in a particular year.

Thirty years of peer-reviewed research has documented that even small increases in alcohol excise tax rates have desirable effects. Among the specific findings are that tax increases: • Reduce alcohol sales and binge drinking • Reduce highway fatalities (stronger effect for youths) • Reduce the rate of STD transmission (stronger effect for adolescents) • Reduce youth suicide rates (under age 24) • Reduce the cirrhosis mortality rate • Reduce rates of robbery and rape

Economists are sometimes defined as people who, when told that something works in practice, want to know whether it works in theory. In my own experience that does not just apply to economists -- evidence that contradicts ones own theoretical perspective tends to be ignored or discounted.

Unfortunately the preconception that leads to skepticism in this area is false. It is not true that alcohol abusers as a group are so highly motivated to get their "fix" that price is no object.

	[Philip Cook, guest-blogging, January 6, 2009 at 9:31am] Trackbacks Price Matters: There is a "national effort to reduce alcohol related problems," so says the mission statement of the federal agency charged with researching these problems and disseminating the findings. I'm willing to believe it, but would like to pose a question: in pursuing this national effort, are lower prices of liquor and beer likely to help or hurt? Because for better or worse, price reductions are the de facto policy of the federal government. Here's the story. Federal alcohol excise taxes have been an important component of alcohol prices, and particularly prices of distilled spirits, since Repeal (or really since the Civil War, with time out for Prohibition). The federal tax is set as a flat amount per unit of ethanol, regardless of value. For example, in 1951 Congress set the tax at $1.68 per fifth of 80 proof liquor. In today's dollars, that's the equivalent of $13.50 per fifth. But Congress has only succeeded in raising the tax twice since 1951, and by meager amounts, so that instead of $13.50, the current tax is just $2.16 per fifth. Even if there were no markup on taxes (and there in fact is), the result is that the current price of a bottle of spirits is over $10 lower than it would have been if Congress had simply indexed the liquor tax to the Consumer Price Index in 1951 and then left it alone. The states have also been slow to raise their nominal tax rates on liquor and beer. The result — inflation has deeply eroded the value of alcohol excise tax rates. It's not surprising, then, that the price of spirits (adjusted for overall inflation) has been falling over time. Just in the last 25 years, the price for package sales has declined 12%. The price of a 6-pack has also declined, by about 8%. Meanwhile, the price of cigarettes has been following a quite different trajectory, thanks in large part to state and federal tax increases since the Master Settlement Agreement of 1998. Over 40 states have raised their tax by more than 25 cents, and 11 have raised their excise tax by more than one dollar per pack. What's going on here? Why is one "sin tax" so popular politically, while another has been largely neglected? These days, increases in the cigarette tax are being touted as a public health measure, sure to reduce initiation by teens and help adults to quit or cut back. While economists have long supported this conclusion with direct empirical evidence, the idea has been difficult to sell to the public and the legislators. Smoking is addictive, and addicts will get their "fix" regardless of the cost, right? But now it seems that everyone is a convert to the power of taxation and price in the campaign to reduce smoking. As it turns out, the evidence that higher prices discourage alcohol consumption and abuse is strong, and of just the same sort as the evidence that supports the conclusion that higher prices reduce smoking. The most persuasive evidence comes from the laboratory of the states. Since 1981, I, my collaborators, and other economists and epidemiologists have analyzed the effect of state tax increases on alcohol sales but also on other outcomes, such as mortality due to alcohol-related causes. The results consistently favor the view that taxes (through their effect on prices) matter. The legislators and the public are not buying it. For some reason, there is profound resistance to the idea that beverage alcohol is a commodity with the usual downward sloping demand curve. The alcohol industry knows better -- they must believe that higher taxes can't be passed on to consumers without a reduction in sales and consumption, or else why would they fight tax increases so fiercely? But when I make the argument to non-economists, I get a skeptical hearing. The more refined skeptics accept the premise that higher prices lead to a reduction in sales, but speculate that that reduction is entirely due to the behavior of moderate drinkers — those who do not drive drunk or abuse their children or lose productivity to hangovers or do long term damage to their organs. In this scenario, the abusers have zero elasticity, while moderate drinkers are price elastic. To which I could point out (and do) that in fact it is these heavy drinkers who should be most likely to respond to price, since they're the ones for whom drinking places a real dent in the household budget. But ultimately my belief in the efficacy of tax as a basis for controlling abuse does not rest on such qualitative arguments -- it is based on the empirical evidence. In the 1960s there were actually some experiments on the effects of price, including several conducted in alcoholism-treatment clinics. Token economies were set up that offered the patients drinks at a price -- either a certain amount of "work" or loss of privileges. These experiments demonstrated that alcoholics were price sensitive. But they have not been replicated, since offering drinks to alcoholics is a tough to sell to the Institutional Review Boards these days. Much more directly relevant are the studies based on "quasi-experiments," including the dozens of instances in which states have changed their tax rates. We can observe what happens to an outcome variable (for example, a mortality rate) in states that raise their tax compared with states that don't raise their tax in a particular year. Thirty years of peer-reviewed research has documented that even small increases in alcohol excise tax rates have desirable effects. Among the specific findings are that tax increases: • Reduce alcohol sales and binge drinking • Reduce highway fatalities (stronger effect for youths) • Reduce the rate of STD transmission (stronger effect for adolescents) • Reduce youth suicide rates (under age 24) • Reduce the cirrhosis mortality rate • Reduce rates of robbery and rape Economists are sometimes defined as people who, when told that something works in practice, want to know whether it works in theory. In my own experience that does not just apply to economists -- evidence that contradicts ones own theoretical perspective tends to be ignored or discounted. Unfortunately the preconception that leads to skepticism in this area is false. It is not true that alcohol abusers as a group are so highly motivated to get their "fix" that price is no object. Related Posts (on one page): Last Call: The Amethyst Initiative A Free Lunch? Price Matters: An Insurrection and 2 Constitutional Amendments: Philip Cook Guest-Blogging on Alcohol Control Policy: