All Commentary
Tuesday, May 26, 2015

Don’t Listen to the Public (Even When They’re Right)

Public "opinion" is a dangerous unicorn


Voters don’t have the foggiest idea what they’re talking about, and they are almost never right about any complex subject except by accident.

Here’s a classic example: the polling company Rasmussen Reports surveyed people’s opinions of the Federal Reserve.

Rasmussen asked: “Do you favor or oppose auditing the Federal Reserve?”

The result? 74% support auditing the Fed.

But what does the public know about the Fed? Nothing — or worse.

Pew Research asked people to name the current chair of the Federal Reserve Board.

The result? 76% don’t know who is in charge of the Fed.

Only 24% correctly picked Janet Yellen from a multiple choice list of John Roberts, Sonia Sotomayor, and Alan Greenspan.

48% volunteered that they didn’t know, and 28% picked the former Fed chairman or one of the Supreme Court justices.

Even though half the public concedes that they don’t know who the chair of the Federal Reserve Board is (while a fourth only thinks they do, and the rest are probably just guessing), only 16% will admit that they “aren’t sure” about whether we should audit the Fed.

Given their inability to answer basic factual questions, what are the odds that the public is expressing an informed, rational opinion about complex meta-monetary policy? Zero.

Why the discrepancy between knowledge and opinion? Because, as Scott Sumner notes, there is no such thing as public opinion.

The odds are that 999 people out of 1,000 had no opinion at all about the Fed before a pollster phoned and asked them, “A proposal has been made to audit the Federal Reserve and make the results available to the public. Do you favor or oppose auditing the Federal Reserve?”

They think, “Hmm… it’s not audited… that sounds bad. Audit will be available to me the public… yep, sounds good.” Voila. Opinio ex nihilo.

Of course, the voters don’t know what they’re expressing an opinion about. They don’t know what the Fed does, or is supposed to do, or how it affects the economy, or even who runs it. They don’t know the tradeoffs or uncertainty or political pressures it faces.

People asked that question surely don’t know that the finances of the Board and the twelve Federal Reserve Banks are already audited by the Office of Inspector General, Government Accountability Office, and outside accounting firms — much less how the proposed “policy audit” would differ from these.

The Fed audit currently exempts transactions with foreign central banks or governments, actions on monetary policy, “open market” operations, or discussions about any of these things; a bill proposed by Ron (and later Rand) Paul would open up these areas to scrutiny.

There’s a strong argument that more transparency on these matters would pressure the Fed to behave more responsibly.

There’s also a somewhat plausible argument that such a review would increase congressional pressure on the Fed to do damaging, short-sighted, politically expedient things. Would we rather Janet Yellen direct monetary policy or Senators Elizabeth Warren and Bernie Sanders?

The answer suddenly isn’t as obvious as you might think. And there’s a whole other debate about just how independent from political pressure the Fed really is or ought to be.

It’s absurd to believe that voters know enough to decide this question. Yet, the public is expected (and all too happy) to have an opinion about messy, complex topics without the commensurate expectation that they know anything about it.

This isn’t an accident. Polling is not meant to measure public opinion — it is meant to create it.

Pollsters face a choice:

  1. Ask a question totally bare of information or context and (as a result) get the least informed opinion possible.
  2. Ask a question loaded with information, conditions, or implications and (as a result) change opinion merely by asking the question.

Polls typically break along these lines, with big polling companies like Gallup electing more for Option 1, and ideological organizations that do “advocacy polling” electing for Option 2. Then pundits pick and choose whatever polls that show that the public is on their side.

It’s tempting to dismiss conditional polling — such as “Would you rather more government services with higher taxes, or fewer government services and lower taxes?” — as less legitimate. But in the real world, nobody makes a decision absent any information at all. To insist that reality not be allowed to affect public opinion is just to confess that raw public opinion is useless.

So what does the public really believe? The answer, often, is nothing — yet.

This situation isn’t confined to monetary policy. The public is wildly misinformed about everything, from foreign policy and government spending to the Supreme Court and healthcare policy. Yet, if you ask them who should be in charge of these things, voters will agree: it should be them.

The danger of enlisting the political public to support your agenda is that you are validating ignorance and giving credence to the idea that everyone else should get to decide how you live your life — what kind of drugs you can buy, what kind of insurance you can have, what kind of car you can drive. It’s also quite likely to backfire when interest groups start to form around the issue and begin asking the question in ways to suit their agenda.

Costless, thoughtless public “opinion” is not only not a sufficient justification for doing something, it does not even provide an additional reason for it. If anything, we should become extra cautious when we suddenly find ourselves in a crowd. 


  • Daniel Bier is the executive editor of The Skeptical Libertarian.