In a recent working paper, Washington University economist Charles Courtemanche finds a negative association between gasoline prices and obesity, supporting his hypothesis that paying more at the pump encourages people to walk instead of drive, which leads to weight loss. Courtemanche projects that a $1 increase in the price of gas would reduce the prevalence of obesity in  the U.S. by 13 percent, which he says would prevent 15,000 obesity-related deaths and save $16 billion per year. He also estimates that "13% of the rise in obesity in the U.S. between 1979 and 2004 can be attributed to declining real gas prices during the period." He therefore sees a "silver lining" in recent gas price increases, which have "the potential to significantly improve public health," and suggests that gas taxes should be set with an eye toward a slimmer America.

Using data from the Behavioral Risk Factor Surveillance System, Courtemanche finds that gas prices are correlated with self-reported exercise levels. Oddly, though, his measure is all forms of exercise, rather than the walking he posits as the cause of greater calorie expenditure. He suggests that "as people become accustomed to additional walking, physical activity becomes more pleasant for them, and they may increase other types of exercise." It seems at least as plausible that they would cut back on other forms of exercise, reasoning that they are already getting a workout by walking so much. It is also worth noting that Courtemanche's projection of health care savings seems to hinge on the assumption that excess weight per se is responsible for the diseases associated with obesity. Even assuming he's right, the savings amount to just 15 percent of the additional spending on gasoline that he estimates would result from a $1 increase in gas prices.

Maybe higher gas prices do affect how much people weigh. But using that connection as a rationale for raising gas taxes means overriding the preferences of people who would rather drive than walk, even if it means they weigh a bit more than they otherwise would. The same argument could be used to justify high taxes on labor-saving devices such as dishwashers, remote controls, and gas-powered lawnmowers, not to mention TV sets, video games, books, and other products associated with a sedentary lifestyle. Courtemanche takes it for granted that fining people for avoiding exercise, which is what taxing gasoline to encourage walking instead of driving effectively does, is an unobjectionable policy, provided it works. This approach, all too common in treatments of sloth, gluttony, and other "public health" issues, transforms what should be a debate about the value of liberty and the proper function of government into a technocratic discussion about which forms of social engineering are most efficient.