The problem seems simple: people consume more calories than they spend. Technically, reversing this is a matter of individual choice. Do I eat the doughnut or go for a run? But this masks the tangle of factors that influence weight.

People have evolved to favour salt, fat and sweets, gobbling as much as possible to store energy should they encounter famine. But human biology is ill-suited for the modern world. Junk food is cheap and delicious; firms spend billions advertising it. In America sugary drinks accounted for at least 20% of the weight gained from 1977 to 2001. In 2011 the average Mexican guzzled 728 eight-ounce servings of Coca-Cola, more than in any other country. About 30% of Mexican adults are obese. Meanwhile jobs require less physical exertion. Cars have reduced the need for walking and cycling. Add to these big trends a host of smaller factors—are there playgrounds nearby, do schools serve proper lunches, do parents have time to prepare meals?—and the balance between calories and exercise seems hopelessly tilted towards the former.

The Economist diet

This poses a grave challenge to health systems and public coffers. Obesity increases the risk of cardiovascular disease, diabetes, liver disease, joint troubles and some cancers. Worryingly, Asians develop health problems at lower weights than their Western counterparts. In America, obesity-related illnesses cost around $190 billion a year, or one-fifth of health spending, reckons the Institute of Medicine. In developing countries, obesity threatens to limit growth, as workers become less productive and wobbly health systems buckle under new demand.

One response is to do nothing. If an individual is fat, so be it. John Stuart Mill said the state may intervene only if a man’s actions harm not just himself but others. But medical costs are high and, in rich countries, are usually borne by taxpayers. Obesity squeaks by Mill’s test.

A second response is to punish those who are overweight. Japan has set a specific limit to citizens’ waistlines. If workers do not slim down, their employers face fines. This is overreach. Weight is hard to lose, and keep off, because of hormonal changes. And people may be fat for reasons—including their childhood and their work—which are not their fault.

A third response is somewhere in the middle: the “soft paternalism” favoured by behavioural economists. The idea is not to limit choice, but to make it easier for individuals to choose carrots over French fries. New York’s mayor, Michael Bloomberg, is veering from soft to hard paternalism: his ban on the sale of big soda bottles is due to take effect in March 2013.

What’s the right policy mix? Here’s some advice for the year ahead (call it The Economist diet). Governments should not force people to eat broccoli, but they can certainly change subsidies to make broccoli cheaper. They can ensure that school lunches are healthy and that children have time to run around. They can set clear standards for nutrition labels, so that educated consumers will shop wisely and demand healthier products. Governments should also consider imposing a hefty tax on soda. The syrupy stuff is a main driver of obesity and, unlike a hamburger, has no nutritional value. A soda tax is far less intrusive of an individual’s liberty than, say, forcing him to wear a seat belt.

More questionable is whether governments should force companies to make healthier food. Firms are trying to fend off intervention by acting first, for example by reducing salt. In 2010 companies promised to cut 1.5 trillion calories from their American offerings within five years. The first progress report will be published in 2013. Watch out, though, for firms that make food healthier in the West but keep peddling junk elsewhere.