For many years I had students in my classes write a paper using what they've learned in class to solve some of society's problems. One of the options was to solve the problem of Americans being so fat. Many of them chose to put some kind of tax on food that is bad for us, and the better papers would then use the money to subsidize food that is good for us (which is, by the way, the opposite of what our government food policy does!).
The "Economics Focus" column in last weeks Economist speaks to this issue. It is a rather interesting examination of the idea, but if falls short in a couple of ways. First, it fails to recognize that not only do we not now tax food that is harmful to our health, as would make sense if we are trying to make Americans healthier, but in this country we actually subsidize its productions. For an entertaining look at this, get the video "King Corn". Corn is not good for us, at least not the way we eat it in this country. But our government encourages it's production, instead of encouraging production of, say, broccoli, which is good for us.
Also, the column talks about a food tax as being as ineffective as the taxes on tobacco have on decreasing tobacco use. Again, they ignore a couple of important facts: 1) Tobacco addicts in this country have dropped from almost 50% of the population in the 1960's to under 25% today (far more successful than the War on Drugs!), and 2) That cheeseburgers, while delicious and often craved, are no where hear as addicting as cigarettes.
Still, a story worth reading, and an idea worth talking about. I am strongly in favor of Pigouvian taxes, as in Gregory Mankiw (see link at right). Seems to make sense to tax bad stuff and subsidize the good. Too bad our politicians are too gutlesss to make the right call. And, by the way, if you haven't noticed the dollar menu at McDonalds doesn't have any salads on it!