His plan on climate change shows the problem. He has a sensible idea -- putting caps on greenhouse gas emissions and letting companies buy and sell the right to pollute. That would discourage harmful activity while leaving market forces to find the most efficient means to that end.

Alas, Obama isn't content to leave it there. He unpacks an array of bright ideas to reduce carbon emissions -- demanding higher fuel economy from automakers, showering money on clean coal technology, giving consumers tax credits for plug-in hybrids, and on and on.

This belt-and-suspenders approach reflects a familiar liberal vice: the insatiable urge to meddle. It's like the team owner offering the coach a generous new contract if he wins the championship -- and then dictating the starting lineup and the play selection for the entire season. It presumes that the government knows in advance the right mix of changes to achieve cleaner energy use at the lowest cost, which neither it nor Stephen Hawking nor anyone else does.

Obama also seems to regard the nation's productive sector as a laboratory for well-intentioned policymakers. In his "60 Minutes" interview, he praised Franklin Roosevelt for his "willingness to try things. And experiment in order to get people working again." What he overlooks is that experimentation creates uncertainty, and uncertainty discourages businesses from doing what they are supposed to do.

During the 1930s, as economist Robert Higgs showed in a 1997 essay, the effect of all the experimentation was the opposite of what Obama assumes. The endless fear of what FDR might do caused net business investment to fall, year after year, prolonging the very catastrophe he was trying to end.

Obama exhibits blithe confidence in the government's power to take economic problems and make them better. He will fare better if he keeps in mind its unbounded capacity to make things worse. For that you don't need socialism.