Christina Romer, President Obama's first chief of his Council of Economic Advisers, is now advocating tax hikes, saying that federal spending cuts would be worse for the economy than tax hikes. Coming from just any Democrat, this would not be news. This is news because Ms. Romer did an academic study, with her husband David that showed how harmful a tax increase would be to the economy.

I wrote about her original study, in which she concluded that a tax increase of 1% of GDP would cause a recession and then a permanent decrease in real GDP of 1.84%. She had also done previous academic studies which concluded fiscal stimuli are powerless in getting us out of recessions or depressions.

These Romer studies were solid academic studies. Her recent one on taxes was published in the American Economic Review, a peer-reviewed journal. It was based on years of data and accepted methods of economic modeling and statistical inference. This kind of study is the way economists earn PhDs.

Now Christina is trying to walk back her conclusions. But note she uses the forum of the New York Times for that, not a peer-reviewed academic journal.  Look at how she justifies her new story.

"Our study, which examined only federal tax policy, found that conventional analysis underestimates the effect of tax changes on the economy substantially...  If there were a similar study on government spending, it would likely show that spending cuts also have larger effects than conventionally believed."

You've got to be kidding. "If there were a similar study on government spending"? She is basing her new opinion on a study that was not done! A non-study trumps a study!

She did academic studies that showed that a "stimulus" would not stimulate. Then she advocated for a stimulus. She did an academic study showing how harmful a tax increase would be on the economy, then she advocated for a tax increase.

This shows us how the world works. These economists get credentialed by doing legitimate academic work. But once they enter the policy and political world, that all goes out the window. Their policy recommendations have nothing to do with evidence or accepted methods of inference. Their policy recommendations are all about politics. In short, they simply sell out.

Who has more influence? An academic economist who gets published in a peer-reviewed journal, or a credentialed economist who says in the New York Times exactly what Democrats want to hear? I think Paul Krugman could provide the answer.