I think people make the worst cases for their policy proposals when they write or speak as if they think there is no credible intellectual argument against their position. A case in point is a column in today’s New York Times by regular columnist Thomas Friedman. In it, Friedman makes his case for an increase in the federal tax on gasoline.

Friedman writes that we’re in a play

where gasoline prices go up, pressure rises for more fuel-efficient cars, then gasoline prices fall and the pressure for low-mileage vehicles vanishes, consumers stop buying those cars, the oil producers celebrate, we remain addicted to oil and prices gradually go up again, petro-dictators get rich, we lose.

Then he adds, “It always ends the same way — badly.”

Notice two interesting things. First, by using the “addiction” terminology, he neatly avoids having to counter the claim that we use as much gasoline as we do because it’s low-price. Using a resource intensively when it’s price is low is hardly the sign of addiction.

Second, to say that it ends badly, Friedman needs to know when it “ends.” It’s obvious that he thinks it ends when prices are high, but the reality is that it never ends or, at least, hasn’t ended yet. Moreover, in the last 25 years, the price of gasoline has been at or near $2.00 per gallon more often than it has been at a price of $3.50 or more, all inflation-adjusted.

Friedman doesn’t seem to understand that to make a case for a tax on the grounds he wants to make it, that is, that it would enhance our prosperity, he needs to make one of two arguments. Either he would need to argue that using gasoline imposes an externality that is so large that the current disproportionately high gasoline tax is too small or that for a given amount of revenue raised, the deadweight loss per dollar raised from an increase in the gasoline tax is less than the deadweight loss per dollar from raising other taxes. Friedman doesn’t even try.