The last five years have seen an ongoing intellectual battle in academia over the proper response of government in an economic downturn. On one side, Keynesian economists have advocated government stimulus as a way to prop up demand. On the other, so-called Austerians have pushed for government belt tightening in the form of reduced spending and increased taxes. Sound familiar?

The easy way to represent this complex academic divide would be to quickly peg Keynesians as liberals and Austerians as conservative — but that would require cherry picking the theories of John Maynard Keynes, the early 20th century economist whose name is now almost used synonymously with government spending. While Keynes was undoubtedly the intellectual backbone behind FDR's New Deal, he actually saw the value in government spending and tax cuts to stimulate the economy — but you wouldn't know it from the way most liberals talk about him now.

In theory, the Keynesian school and the austerity-minded Austerian school — typified by the work of Ludwig von Mises and Friedrich Hayek — needn't line up with America's political divide. Conservatives, for example, tend to favor tax cuts during a recession (which is also supported by Keynesian theory) but want to pair those tax cuts with spending cuts (which Keynesians oppose). By contrast, liberals generally favor increased government spending (which is anti-austerity) but also support increased taxes, particularly on "the rich" (which is a form of austerity).

There are, of course, some Keynesians, like Greg Mankiw, who are conservative. For the most part, however, many of today's prominent Keynesians are really liberals first, Keynesians second. They preach the gospel according to Keynes when the subject is the need for increased government spending, but when the subject turns to taxes, their Keynesian convictions tend to melt away.

Consider, for example, the case of America's most prominent disciple of Keynes, Paul Krugman. In a recent column, Krugman chastised French President Hollande for giving in to "the grip of austerity fever." Hollande's crime? Giving a speech announcing his intention "to reduce taxes on businesses while cutting (unspecified) spending to offset the cost."

From a Keynesian point of view, this makes little sense. Tax cuts are not a form of austerity — they're actually a form of stimulus, according to Keynes. And even if the tax cuts are offset by "unspecified" spending cuts (which is doubtful), that would simply involve maintaining the Keynesian status quo, rather than a turn to austerity. What's more, when Hollande first came into office promising to raise France's top tax rate to 75 percent, this earned not Krugman's ire, but his admiration.

Nor is this an isolated incident. During the debate over the fiscal cliff at the end of 2012, Krugman argued that the Obama administration ought to insist on raising top tax rates, even if it meant raising taxes on everybody. This despite the fact that, by his own admission, doing so would harm the still fragile economic recovery.

Krugman isn't alone, of course. Bruce Bartlett claimed in early 2013 that further Keynesian stimulus was "desperately needed." Yet shortly after the fiscal cliff deal, he mused about how tax increases were associated with improvements in the stock market. And, of course, liberal politicians from Obama to the Democrats in Congress have advanced Keynesian arguments to support more government spending while simultaneously supporting anti-Keynesian tax increases.

There's nothing surprising about a politician using arguments selectively — but a Nobel prize winner like Krugman? That's more surprising.