Scott Sumner goes hard after American political science:
Just one more reason why academics should pay no attention to “public opinion” polls. There is no such things as public opinion, there is only election results. No one knows what Americans would believe about Medicare if that sat down with all the government programs and tax revenues in a spreadsheet front of them, and told they had to equate the NPV of all future taxes with the NPV of all future spending. We simply don’t know. And anyone who argues otherwise isn’t thinking deeply enough about the issue.Sumner calls this post "Thinking like an economist", which reminded me of my past post on the problem with economists.
The UNC political science department is well-known in academic circles for the study of public opinion in American politics, and the consensus view in Hamilton Hall is not that public opinion doesn't matter, much less that there is "no such thing". I very much doubt that Sumner has any familiarity with this literature, but he could start here and here and here. Public opinion matters a lot for policymaking, especially on issues that are salient with voters*. It even matters for the judiciary, even those with extreme job security such as the Supreme Court (see first link). It's not about voters having policy expertise, or what they'd do if they had to match the NPV of spending and revenue. The public does not even have to be coherent to have a major impact on policymaking, and not just via elections.
Let's take an example. Sumner is frequently exasperated by the Federal Reserve. He notes that the economy remains depressed several years after the beginning of the recession. He notes that Ben Bernanke has said that the Fed has plenty of tools to boost nominal GDP even at the zero interest rate bound. He notes that Ben Bernanke did a lot of research on both the Great Depression and Japan's lost decade, and thus understands the situation we're in quite well. The Fed does not face elections and is considered one of the most independent central banks in the world. And yet despite possessing the requisite expertise and policy tools the Fed is nowhere near as activist as Sumner would prefer.
How can we explain this? It could be that the Fed are a bunch of idiots, but Sumner does not believe that to be true at least in Bernanke's case. Or it could be that the Fed has just witnessed a series of events that have made them cautious. The Tea Party has had a major effect on American politics, and one ideological leader of the Tea Party -- Ron Paul -- wants to abolish the Fed and is now chairman of the House committee that oversees the Fed. Paul's book *End the Fed* has 375 reviews on Amazon.com, and nearly all of them give the book four or five stars. A Nobel Prize-winning economist -- Peter Diamond -- was blocked from joining the Fed by this element of the contemporary GOP. The leading candidate for the GOP presidential nomination recently threatened Bernanke with bodily harm and insinuated that he was a traitor. And Bernanke is a fellow Republican who was appointed by a Republican president. He has also been criticized by the left for bank-friendly policies. All of these groups want a monetary policy that is tighter than Sumner's preferred policy, and that is what the Fed has done.
Given all of this, isn't it at least plausible that the Fed feels constrained by public opinion? To my knowledge no studies have focused directly on this question, but as a potential contributing factor to Fed policy choices it seems at least plausible. Even if public opinion doesn't affect the Fed, the finding that it affects the Congress, presidency, and judiciary is very robust. So to be so dismissive is really silly.
*My guess is that the specific issue Sumner is discussing -- the tax penalty for married couples -- is not highly salient for most people. When it is, couples can easily (and cheaply) get legally divorced or remain unmarried as Justin Wolfers and Betsy Stevenson have done.