The Secret Consensus Among Economists

If you follow the economic policy debate in the popular press, you would be excused for missing one of our best-kept secrets: There’s remarkable agreement among economists on most policy questions.  Unfortunately, this consensus remains obscured by the two laws of punditry: First, for any issue, there’s always at least one idiot willing to claim the spotlight to argue for it; and second, that idiot may sound more respectable if he calls himself an economist. 

How then can the quiet consensus compete with these squawking heads?  A wonderful innovation run by Brian Barry and Anil Kashyap at the University of Chicago’s Booth School Initial on Global Markets provides one answer: Data.  Their “Economic Experts Panel” involves 40 of the leading economists across the US who have agreed to respond on the economic policy question du jour.  The panel involves a geographically and ideologically diverse array of leading economists working across different fields.  The main thing that unites them is that they are outstanding economists who care about public policy.  The most striking result is just how often even this very diverse group of economists agree, even when there’s stark disagreement in Washington. 

That observation is the starting point for my latest column with Betsey Stevenson

Let’s start with Obama’s stimulus. The standard Republican talking point is that it failed, meaning it didn’t reduce unemployment. Yet in a survey of leading economists conducted by the University of Chicago’s Booth School of Business, 92 percent agreed that the stimulus succeeded in reducing the jobless rate. On the harder question of whether the benefit exceeded the cost, more than half thought it did, one in three was uncertain, and fewer than one in six disagreed.

Or consider the widely despised bank bailouts. Populist politicians on both sides have taken to pounding the table against them (in many cases, only after voting for them). But while the public may not like them, there’s a striking consensus that they helped: The same survey found no economists willing to dispute the idea that the bailouts lowered unemployment…

How about the oft-cited Republican claim that tax cuts will boost the economy so much that they will pay for themselves? It’s an idea born as a sketch on a restaurant napkin by conservative economist Art Laffer. Perhaps when the top tax rate was 91 percent, the idea was plausible. Today, it’s a fantasy. The Booth poll couldn’t find a single economist who believed that cutting taxes today will lead to higher government revenue — even if we lower only the top tax rate

The point here isn’t that the panel of economists have all the answers.

Rather, they agree on the best reading of murky evidence. The folks running the survey understand this uncertainty, and have asked the economists to rate their confidence in their answers on a scale of 1 to 10. Strikingly, the consensus looks even stronger when the responses are weighted according to confidence.

I’ve never seen the disjunction between the political debate about economics and the consensus of economists be as large as it is today.  And I think this is incredibly damaging.  Instead of having a serious discussion about how best to end the current economic slump, Congress is gridlocked, as one of the major parties is blocking every effort to improve the economy, using arguments which are so far outside the mainstream that it is hard to find a single economist to agree with them.  The result is that Congress is doing nothing in the face of the tremendous suffering wrought by high unemployment.  And it’s also doing nothing in the face of a longer-run budget problem.  Indeed, fear that continued inaction will lead the US to walk off the “fiscal cliff,” may already be dragging the economy down. 

The optimist in me thinks that the new Economic Experts Panel may be a (small) part of the solution.  My hope is that journalists will come to see this as a useful resource—a way of checking whether political debates reflect serious disagreement, or something more cynical.  And hopefully this will lead us to focus on the more serious debates, rather than convenient and disingenuous talking points. 

You can read our full column, here.


Yes, if only we would just let all the "smart people in the room" run things for us the world would be a better place and we'd be happy.

The problem isn't people ignoring consensus. "Consensus" is a classic headcount fallacy, hallmark of an argument lacking in real substance. The problem is economists such as yourself who are willing to sell their soul to shameless shill for a political party.

It is sad that the man who said this died so young:

"Let's be clear: the work of science has nothing whatever to do with consensus. Consensus is the business of politics. Science, on the contrary, requires only one investigator who happens to be right, which means that he or she has results that are verifiable by reference to the real world. In science consensus is irrelevant. What is relevant is reproducible results. The greatest scientists in history are great precisely because they broke with the consensus." -- Michael Crichton.

You sir have revealed nothing more than which side of the politics vs. science line you are on.



Okay, I hate to engage in this ridiculous discussion, but let me tell you what an actual hard scientist (theoretical physicist) thinks of these kinds of discussions. (One who has no real political opinion, is not a fan of any party, and doesn't generally find politics all that interesting. But does have an advanced background in math, has read many economics textbooks, as well as graduate texts and papers.)

1. The idea of a "consensus" representing anything scientific is *insane*. As a physicist, if someone asks me to prove relativity is correct, I don't take a poll of people I claim are leading scientists. This is what crackpots do. To prove it, I write down the equations that define the theory, and preform experiments demonstrating that they are correct.

I don't care if the sample size was 100% of all economists. I don't care if you psychically contacted all economists from across time and space in the entire universe and polled every single one of them. It would not matter if they all completely agreed or disagreed on every single issue. It would still be totally meaningless.

Correct science speaks for itself, it does not need your help.

2. Scientifically, if you need to solve a problem, write down your theory, make a prediction, act according to the prediction, and then later find results which disagree with that prediction THEN YOUR THEORY IS WRONG.

You do not get to say that doing something else would have been more wrong.

You do not get to speculate on how, based on the theory that made these predictions, things changed to make it wrong; that's logically fallacious (in the technical sense, not the colloquial one). You don't know how these changes effected your system *because they were not included in its description*.

I mean, physicists recently discovered the Higgs boson, which was predicted as a general effect in the 1960s, and refined to specific predictions in the '70s and '80s. If we had not found the Higgs boson at the LHC, we could not have claimed, "well, if we hadn't made this prediction, the theory would've been even worse!" That's crazy. All you could do is say "whoops, that theory was wrong," then throw it out and try again.

If you predict "doing x will cause a y% change in the unemployment rate" then, simply, it had better, or you were wrong. This applies to *all* sides of course, the Republicans, Democrats, and Economists, have made predictions, based on their models which were shown to be radically incorrect. You simply can't plow on after your model has made an incorrect prediction, no respectable science works that way.

3. Real science has clear, meaningful, definable, margins of error.

If a physicist published a paper saying "I measured the Higgs mass to be 125 GeV using the LHC" they would be punched in the face by every physicist who read it for wasting there time.

A real model makes statements like: "These results are consistent with a Higgs-like particle of mass 125 GeV +/- 2 GeV with a combined 5 sigma confidence over all decay channels." That *still* doesn't mean anything, so they go on to explain in detail how these details were all calculated. Read the papers involved with things like this, no one talks about the measurement. The papers are almost entirely devoted to excruciatingly discussing how margins of error were obtained and to statistics.

No one cares about the actual value of the measurement, it was just a meaningless detail, we care that the measurement was done *correctly*.

I should point out, the only reason 5 sigma (about 1 in a million chance of a signal being due to random noise) is good enough, and we don't need something like 10 or 100 sigma or more, is because of the huge amount of effort put into correctly doing the statistics, the *thousands* of experts working on it, and because of the amazingly sophisticated mathematical structure behind the models describing, e.g., the detectors and machinery, in addition to the theory.

Any of the other hard sciences adhere to these kinds of standards: chemistry, engineering, math, etc. This is what it means to be a hard science. (This is also one reason, of many, that hard scientists mock and ridicule results of soft science.)

4. It's not hard.

I often hear economists complain that successfully modeling "complicated" systems like the economy is hard. There are too many unknown factors. There are too many unknowns. There are too many random elements.

You know what? Boo-freaking-hoo. Let me break out my tiny violins for you.

Every scientist's job is hard! Do physicists complain that "ohhh quantum mechanics is so hard how can I ever build you a liquid crystal display out of it!" No, they shut up and do it. If they don't know how to do it, they say "I don't yet know how to do this," and then they get back to work.

They do not make ludicrous predictions about how it might kind of work if only their job weren't so hard, and then, when shown to be wrong, conduct a freaking poll of other scientists to see who agrees with them.

Let me tell you, real scientists simply do not care if anyone agrees with them, what they care about is if they are demonstrability correct, about if they can make predictions, and about how to make their models even cooler.

And, really, not to be condescending, but, I can literally tell you how the universe started. Maybe not *all* the details, such as how this particular particle spectrum came about, but I can tell you about how it started from a process similar to bubble nucleation, how the early universe came to be in that state, how it tunneled to be at that value of the potential, how that potential energy was released as particle radiation ("the Big Bang"), how that process slowed down, how nuclei formed, how the CMB and large-scale cosmic structure formed, how this expanded and cooled to become galaxies and stars, how those stars used up their fuel, created large clouds of heavy elements, out of which more stars formed, out of which more heavy elements and stars formed, out of which planets like the Earth and stars like the Sun formed.

I understand how quantum mechanics really works (despite the mystical B.S. you read about it), how the universe really has more than 4 dimensions, and I even understand why and what that means!

I can write down physics problems that can't be solved because no known math exists that can solve problems like it.

This is what I live with. This is the level of quality and standards that are a part of my every day life.

And what do I see from economics?

A poll?

Models that are "sometimes" right, if you allow for margins of error of 2, or 5, or 10, or more? And other times fail even with those error levels?

Claims that economics uses sophisticated math? Like calculus! Oh my god, you can write down integrals, it must be right! A subject nearly all of my colleagues were experts at by 15 or 16, is surely the pinnacle of math!

Here is my challenge to economists if you want hard scientists to believe anything any you say. Do one of the following two things:

1. Make at least one nontrivial, *quantitative* prediction of a meaningful, *dimensionless*, real-life macroeconomic observable, complete with a proper statistical analysis, careful Bayesian analysis of how this prediction compares to other models, including several candidate null hypotheses, as well as several other models. Then, show me that these predictions agree with future data, made after the model was published, to at least 5 sigma for an appreciable period of time.

Then, you may claim: "our model for predicting the observable X is consistent with the observed data in the 5 sigma confidence interval."

Do this because this is the *minimum* standard for calling yourself "correct" in science.

Or, alternatively,

2. Say "we don't know, our models are not good enough to make a definitive claim about any observables yet."

We'll believe that too. And there's no shame in making claims like that. Science is all about making that exact claim, analyzing, in the above manner, what is not known, and repeating this until you can make a claim.

Or, you know, keep making polls and claiming you all agree on the wrong models, and those who don't agree with you agree with each other about other wrong models.

This kind of thing really irritates me, because it damages the public's understanding of real science. When they see inane, and huge volumes of sure-to-be-wrong predictions by actual economists (disregarding the claims from politicians and political lackies), they come to the conclusion that it's too hard to make predictions about anything. That science, in general, isn't sure of things. That "you can't really *know* things."

When people live in an environment like this, they don't trust science, they do not support science, and they do not make correct scientific judgments. For example, they believe that doctors don't really understand how vaccines works, or that maybe compound X in their food or water really is bad/good even though it is definitively fine/bad.

This kind of discourse unnecessarily diminishes the public understanding of science and their confidence in science. You should be ashamed of yourselves for participating in it.



And to think that economics is the only 'hard' social science.

Fellow Traveler

The concept of "policy" implies that some group of people are using coercion to force the rest of us to do things we otherwise wouldn't do, in a free market.

Perhaps the "policy" itself is the whole problem?

Isn't our entire monetary system a form of policy? And our banking system?

Policy == force.


Let me explain why this panel is just about useless.

Imagine you belong to a group of firefighters whose job is to put out all forest fires in your area. This group has been going strong for about a 100 years, and doing a pretty good job putting out all fires. Then one day in the middle of a hot, dry summer a fire starts. It starts out normally, but you can't seem to put it out. Nothing seems to work. So you bring in top experts in firefighting from around the country to map out a plan. You try this and that, but nothing really seems to solve the problem. In the end the entire forest is wiped out.

Why were the experts unable to help?

By the time the experts were brought in it was too late. It was impossible to fix the problem. The real problem started a hundred years earlier with the decision to put out all fires. You see, over time dead wood and bushes built up. Then gaps between areas filled in. Slowly the entire forest positioned itself for collapse. Small fires would automatically fix these problems, but they were heavily suppressed by the firefighters. Over time the build-up of dead wood and bushes became too great, and the forest system collapsed.

Forests and societies both work the same way. They both move into the future with a positive feedback loop that automatically positions them for collapse after a long period of time. The seeds of destruction are sown during the good times. Economic collapses and wars have the same (or similar) distribution as forest fires - a power-law distribution.

Over time societies collapse because a build-up of bad decisions and corruption becomes too great. Suppressing economic recessions (The focus on stability) allows these bad decisions and corruption to grow. If you thought the global financial crisis was about finance or economics, then you would be wrong. It's about bad decisions and corruption. Other areas will be affected by this as well.

The expert panel in the article will not be able to fix our problems because it's not just about finance and economics. At this point there is only one solution - let the collapse happen.



Click on the above link showing the participants. Every single one is listed with a university affiliation. I guess you can only be a real "economist" if you are a professor.


Half of all economists think the benefits of the stimulus outweighed the costs? So we spent $800 billion on something that in the end only half of economists think was worthwhile. On smaller policy questions a 50/50 shot may be a risk worth taking, but on an $800 billion program? Geez...


And if someone surveyed "scientists" in the year 1500, most would agree that the sun revolved around the earth. Certainly made that true!


Well, I don't like. A few problems I see with this.
1) Who chooses who is on the list? What sort of qualifications does one need to be on it? There have been 20 econ nobel prizes recipients in the last decade and only one of them made the cut. I am wary of the opinions of a subjectively selected panel. Even if they were chosen in some subjective manner (i.e. highest impact factor of publications in the last decade) there are always some subjectives controlling those whether big journals like what they are saying).

2) Consensuses are a funny way of determining truth. As stated in other comments, the consensus in 2007 was that the economy was fine. The consensus around 50 years ago was that smoking wasn't bad for your health, even though good evidence had shown it caused cancer. We can find these examples in every field today, and we can find them hundreds of years ago. What is the case is you have correct people and incorrect people, and a majority doesn't change what is correct. 99.9% of people believing the earth was flat doesn't change the fact that Columbus was right.

3) We can't eliminate direct looks at exactly which question was asked these economists and what their responses are. While a 10 might show confidence, I usually trust people with some doubts a bit more, as they are making decisions after seeing both sides.

While this forum has definitely pulled together some big people, I do think more needs to be considered here. Economists in general have been wrong before, and will be again.



Problem is the squawking journalist reporting the results who doesn't understand that the one squawking pundit on the opposite side of the debate from the consensus is a lone quack, particularly if the quack produces a shiny, informationally thin chart to "prove" their point. The real scientist, and it the case of economists I use the term very loosely, knows it's impossible to "prove" much of anything, you can only disprove it. Every year journalism schools produce more journalism graduates than there are positions in journalism, a tribute to why you should never trust a journalist with a number. Or a "expert opinion."


When I left University with an Economics degree in hand I figured that Keynesian Theory was an opportunity for economists to get out of the ivory towers of the university and into fat government jobs showing politicians how to fiddle. And fiddle they have.
Just because a bunch of 'economists' agree - consensus - does not make them right.
And so far they have been very wrong.
And its not going to get any better just because someone has polled a bunch of them and got a 'consensus'. They are still just as wrong.
The best thing to do is get ready for the debacle coming up as the economists try to prove they are right in fiddling.


I looked up the composition of the "Economic Experts Panel". Here is a very telling statement on the site: "The panel members are all senior faculty at the most elite research universities in the United States." This tells me that they are clearly not ideologically diverse since the universities are probably the single-most intellectually homogenous institutions in the country. I would be more impressed with the findings of an "Expert Panel" if there was also strong representation from people with the most hands-on experience with an economy - business people, especially entrepreneurs.

Young Economist

Nice post, except for the slight bias: why does Wolfers only mention the Republican talking points that are at odds with the "economist consensus?" There seem to be plenty of examples on the other end of the political spectrum. Heck, I actually thought of Paul Krugman and Brad DeLong when he first mentioned the "squawking heads!"

john werneken

The divinity is in the details. First, any spending anywhere is likely to support employment. It does not follow that all employment or even employment in general is worth supporting. Downstream effects of policy could affect innovation, investment, infrastructure, education, and economic incentives, perhaps in ways that REDUCE the longer-term income and employment of the people as a whole. The odds are that ANY public policy will be less wise than any alternative private policy, if there is one, as the private policy IN GENERAL seeks net benefits whereas the public policy only claims to do so, there is no test or reality check in the future on the wisdom of past public policy, whereas a dumb business can go broke.

Second, tax collection depends more on economic and population growth and the exemption or inclusion of various incomes in taxable income than it does on rates. That said, raising tax collections is only a good thing in a couple of cases: if what it will pay for provides broad and enduring benefit, which is doubtful for the majority of public spending in a democracy, or if the possibly better fiscal condition of the government leads to stable currency values. A large difference in marginal rates may penalize those with more money and choices more than it benefits those less money and with fewer choices, and may encourage the better off to preserve wealth rather than to invest to try to increase it.

Also there is more to be said for possible benefits to decreasing taxes on wages and on incomes and increasing taxes on consumption, than there is for rate changes per se for any one type of tax.
Finally, the public tends to be largely of one mind on many things without getting into any grasp of the details – in favor of peace, employment, the environment, security, debt relief, and growth all at once; I doubt ANY economist could identify an economic policy equally supportive or even a bit supportive of ALL of those values.

So our leaders cloth arguments about what works (which economics should be able to discuss) and about who wins and who loses (not an economic subject; that is politics!) as if they were disagreements about HOW to achieve simultaneously a set of objectives which may not be compatible with each other.



"The Booth poll couldn’t find a single economist who believed that cutting taxes today will lead to higher government revenue – even if we lower only the top tax rate. "

I do not believe the conservative view is that cutting taxes will bring more money into the government. The view is that cutting taxes will bring more money into the private sector. Republicans are against more growth of government and believe in building private business that produces products.