And so perhaps this is part of the political legacy of the Bush years: a subtle shift in what makes you a progressive, and how liberal lines are drawn. The debates among Democrats in the nineties mostly followed established ideological lines—some believed that Clinton’s centrism was whitewashing liberalism, others that he was modernizing it. But there is less ideological divergence among Democrats now. The deep liberal disappointment with the president has a different source: He believes in politics more than they do. If you believe, like Obama, that politics is what brings the social compact to life, then the only way to build an equitable society is to make sure everyone has an equal say. Or if you believe, like Krugman, that the data can show you the shape that the fairest society should take, then politics might not always make good on the social compact. Politics, instead, might make it impossible.

Krugman’s purism is partly tactical, his way of correcting for the inevitable dilutions of legislative negotiation. “You want to have a pretty clear vision of what it is you want even though you know what you’re going to get is only a small fraction of that,” he says. When he pushed for a stimulus so large that it seemed implausible it would pass Congress, “I was making a political calculation of my own, that a policy that only did half of what was necessary would lead to political disaster.” His online commenters are more absolute, and they have sometimes turned on him when he has acquiesced to political realities (when, for instance, he eventually endorsed Obama’s health-care reform after having attacked its compromises). But Krugman insists he is playing a more sophisticated game than his supporters and critics give him credit for. “Do we know that if they’d really gone bigger in their proposals, they would have gotten more and it would have worked out better? Of course we don’t,” he says. “But I don’t think it was just me being an outsider and not having a grasp of the realities. I think in some ways I had a better grasp than the insiders had.”…

But talking about Merrick prompts Krugman to think about how that moment might have been extended. “Suppose an alternative history in which big-box stores, Wal-Mart and others, were unionized,” he says. “You could easily imagine that you could have a large number of service-sector workers who were, if not like autoworkers, like ­manufacturing-sector union workers in the golden age of private-sector unions.” He thinks for another minute. It might not have been Utopia, he says, but it could have been France. But now these possibilities seem further away than ever. Part of the basic loneliness of economic study is that you are always looking back, at data sets that are already completed. And so you realize your vision of a perfect society just as it disappears from view.