Taxes on harmful substances like tobacco and alcohol are often scoffingly referred to as "sin taxes," but that does not mean they are a bad idea. My home state of Minnesota is considering increasing these taxes to get revenue for better budget balance over the long term. They incorporate some interesting economics, so it is useful to think about them a bit.
"Sin taxes," is a term economists would not use. Such taxes date back centuries and largely were introduced for pragmatic reasons, rather than for moral reasons.
A theoretical basis for these taxes had to wait until British economist Arthur Pigou's work 90 years ago on reducing the economic inefficiency that inevitably occurs when external costs are not controlled.
External costs are those borne by society, but not directly by either the producer-seller or the buyer-consumer. Pollution is a common example. External costs like pollution often are thought of as unfair, since they hurt third parties, but, at least as importantly, they cause resources to be wasted and society ends up with fewer goods and services to meet people's needs.
Pigou argued that the optimal way to halt waste caused by such "market failure" was to impose a tax on the product activity producing the external cost, or harm to others, equal to the amount of that external cost. That way, the purchaser of the good or service would face its full cost to society when deciding whether to buy it or not. This would not reduce production of such harmful goods to zero, but it would restore efficient use of resources.
Smoking and drinking produce pleasure for some people and both have external costs.
The greatest harm caused by inhaled tobacco is to users themselves in terms of lung cancer and other diseases. There is harm to third parties who inhale secondhand smoke, but these damages are small compared to those incurred by smokers themselves. So the scope for restoration of economic efficiency via Pigou's tax theory is limited.
More importantly, because of the way we socialize medical spending via insurance or tax-supported programs like Medicare and Medicaid, the costs of treating both drinking - and smoking - induced diseases effectively becomes an external cost to those paying premiums or taxes.
The health care costs of smokers themselves would be "internalized" in a free market that did not have government-paid health care and that allowed insurance companies to charge sharply higher premiums to drinkers and smokers.
All this aside, the external costs to society as a whole of alcohol consumption are now many times higher than that of tobacco.
There is a plethora of studies of costs going back 30 years showing annual social costs of about $5 billion in 2011 dollars in my state with a population of a bit over 5 million. That $900 plus cost per capita ain't chump change. It is about 17 times the amount collected in alcohol taxes. Doubling the taxes, as currently proposed still would leave a large gap.
However, alcohol-related externalities differ from externalities like pollution, in that the harm is not linear. Someone drinking a beer or two a week may cause zero external costs for society. Someone getting plastered every other night may create enormous costs. Why, some would argue, should a majority of moderate drinkers pay a tax when the majority of damage is caused by a minority of the population who drink a lot?
One answer is that moderate drinkers pay very little in tax. I drink about a beer a week, on average. My wife doesn't drink at all, but we occasionally buy a bottle of wine. The proposed increase will cost us somewhere between $6 and $10 per year. We may lean toward the abstemious end of the scale, but, given the skewed pattern of little alcohol use by many and heavy use by a few, a large majority of households would probably pay less than $50 additional a year.
Alcohol taxes are not indexed for inflation and were last raised here in 1985. The general consumer price index has increased by a factor of 2.17 since then, so one could double the tax and still would leave the overall inflation-adjusted tax levels a bit below what it was when last raised.
There is a complication in that my state, like many others, taxes the alcohol content of beer at a lower rate than that of wine and distilled spirits. There are political reasons for that, but not economic ones except for the fact that dangerous problem drinkers can get plastered quicker on hard liquor than on beer. The increase proposed here would raise the tax on a 12-ounce bottle from 1.4 cents to 8.4 cents per bottle. Even at the higher rate, it isn't much of a deterrent to drinking.
We face a further complication. A state can set its alcohol taxes at whatever level it wants. If its major population centers are far from the state border, most people won't have the choice of driving to an adjacent state with lower taxes. With beer-loving Wisconsin smack-dab against 60 percent of my state's population, the threat is real and liquor distributors are quick to point it out. One can have all the public finance theory one wants, but practical considerations like this end up driving lots of policy.
Then there is the issue of tobacco taxes. Economists who study the question find that alcohol generally remains under-taxed by Pigou's criteria, but cigarettes are overtaxed, with smokers paying well above the actual external cost of their activity. But in a sharp reversal of attitude over the last 20 years, smoking is now socially scorned, and few are willing to speak up for them in politics.
Economist Edward Lotterman teaches and writes in St. Paul, Minn. Write him at email@example.com.