Network News

X My Profile
View More Activity
Posted at 10:25 AM ET, 03/10/2011

Common mistakes made by economists

By Ezra Klein

Tyler Cowen has posted lists of mistakes he thinks common to left-wing economists and right-wing economists. Both are worth reading, but they have a strong “from one economist to another” tone. So here’s a list of mistakes that I think economists and people who are heavily influenced by economists tend to make when they look at politics. I should preface this by saying I have, at one point or another, been guilty of literally everything on this list:

1. Political power matters. There are many outcomes that are economically efficient in the short term but lead to a dangerous imbalance of political power in the long term — which is, incidentally, not economically efficient at all. This has particular implications for how a lot of economists view unions.

2. Culture matters, as do the real ways that human beings behave. There are policies that fit with theory and evidence but not with communities and people. David Brooks is right about this.

3. If a policy makes sense only in the presence of a secondary compensatory policy — say, a regressive tax where low-income folks get some sort of refund — then you have to ask yourself whether the compensatory policy will pass. If the answer is no, then you need to come up with something that can pass or rethink your support for the policy. The fact that the losers of trade can theoretically be made whole doesn’t allow you to just assume they will be made whole.

4. Lots of policy problems can be solved with clever policy solutions. But Washington isn’t very good at passing or implementing clever. Simple programs and rules are often better in practice, even if they’re worse in theory.

5. Nationalism is a really, really, really powerful force, and you can’t make it go away by condescending to it.

6. “Theory implies” does not end arguments. Moreover, economic evidence should be treated with more humility. It’s often overturned later, or wrongly understood now. And a lot of the stuff you’ve told us in the past — particularly the recent past — didn’t turn out that well.

7. Listen to political scientists, sociologists, etc. They have perspectives, evidence and training worthy of consideration.

8. Policy arguments are often conscripted for political purposes. You may like Singapore’s health-care system, and a politician might find Singapore’s health-care system useful to invoke — usually incorrectly — in a speech against the Affordable Care Act, but before assuming the two of you are on the same side, try to figure out whether the congressman has introduced or co-sponsored legislation on this topic that you consider constructive. Nothing sadder than a policy expert who doesn’t realize he’s being played.

9. No one knows what the word “stochastic” means.

10. Odds are good that you primarily know one sort of person: highly educated, high-achieving, extremely cerebral, etc. Odds are also good that you give too much weight to feedback and ideas from this sort of person, while discounting arguments and complaints from people who don’t know the right way to persuade you. Try to keep that in mind.

Any more?

By Ezra Klein  | March 10, 2011; 10:25 AM ET
Categories:  Economics  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Was what Scott Walker did legal?
Next: Will Democrats overturn Walker if they get power back in Wisconsin?


I have a message to economists as well:

Not every problem is an economic problem. Being an economist does not confer upon you a general expertise applicable to everything under the sun. Even when economic analysis yields an answer, that does not make it the right answer or even a answer that makes sense. This includes problems ranging from baseball to education. The most recent example is an economic analysis that concluded students should be stuffed into the classes run by the best teachers because class size effect is small. Yeah, maybe economics says that but nothing else does. For one thing, how long do you think the best teachers will stay in education if their work environment is oppressive? If they're good, they have other opportunities.

And stochastic means random. It comes from physics and mathematics but economists use it to make guesses sound scientific.

Posted by: pj_camp | March 10, 2011 10:55 AM | Report abuse

#9 ftw

Posted by: arm3 | March 10, 2011 11:13 AM | Report abuse

Drat, I know what stochastic means, I guess I'm nobody.

But seriously, these insights are undoubtedly true (and valuable), and your ability to explicate Washington process is brilliant. Yet, IMO, the most troubling aspect of our political culture, is the willingness and ability of conservatives to discredit evidence with calculated misinformation. I'm thinking here specifically of climate change. Conservatives have destroyed any meaningful policy debate on this matter by abusing the implicit popular notion than in a two party system, at any given point, each "side" is about half right. This "feels" fair, but so what? The result is too many people believing they really don't have to inform their opinion and to just go with their gut. And so, the climate continues to become unstable and the fate of large swaths of peoples and civilizations are increasingly at risk of large scale disruption. Perhaps then, the character of American democracy makes it net net, unsuitable as a force for constructive world leadership.

Posted by: JohnL8 | March 10, 2011 11:19 AM | Report abuse

"9. No one knows what the word “stochastic” means."

You make the physicist in me use my neuronal connection between the lacrimal gland to shed salne (e.g. I shed a tear).


Posted by: chrisgaun | March 10, 2011 11:21 AM | Report abuse

I think it's worth re-stating number 7 in a much stronger way. As I understand it, basically all economic theory stems from a handful of axioms that are rooted in a fairly antiquated, modernistic, rationalistic ideas about human psychology.

Further, the economic viewpoint I think inherently tends to contain an unexamined bias towards a bizarre hybrid of libertarian deontological and utilitarian consequentialist ethical thinking.

These are crucial flaws in traditional economics, they are, I think, slowly being recognized and rectified, but could easily be helped by looking to psychology and political philosophy.

Posted by: Nemo1342 | March 10, 2011 11:26 AM | Report abuse

"Nationalism is a really, really, really powerful force, and you can’t make it go away by condescending to it."

I feel like the left is abandoning a very very powerful tool by ignoring nationalism. We should wrap more policies in the flag.

And stochastic isn't *just* random! It's random in a way that isn't deterministic! (Ex: a coin flip follows a deterministic random value. The probability it lands on heads or tails is known. A probability distribution, on the other hand, would be stochastic, that is, the probability of a certain even isn't known--it isn't a single value, rather, it follows a probability distribution)

Posted by: will12 | March 10, 2011 11:30 AM | Report abuse


- Putting the word economist behind your name, does not mean that the column you're writing is based on sound economics. It may mean that your political agenda is more important to you than the economics. (Krugman)

- writing about ecnomics, without ever having practiced economics in the real world is like designing beautiful castles on paper that could never be built in reality (Romer, Baker) Having actual money at risk in some way, tends to marvelously focus one's attention.

- the world is a zero sum place. no economic thoery can change that. cheaper prices for imported everyday goods means that fewer Americans will be employed in those fields for instance. There is no way to change the zero sum equation.

Posted by: johnmarshall5446 | March 10, 2011 11:33 AM | Report abuse

The left (fiscalists) seem not to understand the concept of 'unsustainable' when they think government intervention can return us to an unsustainable bubble growth path until the private industry comes back to keep us on the unsustainable trend line forever.

The left also demonstrates an unwillingness to acknowledge that increasing unemployment benefits puts an upward pressure on unemployment rate. Heck, Krugman noted that this was proven in one of his textbooks and then claimed it wasn't a factor in his column.

The left also never seems to consider that there might be differences in the implementation practicality of welfare state policies in a country with massive numbers of impoverished immigrants vs. implementing those same policies in countries with negligible immigration.

The right seems to have no useful understanding of employment, taxes, markets, prices, externalities, or the existence of utilitarian value beyond GDP growth rate.

Posted by: eggnogfool | March 10, 2011 11:49 AM | Report abuse

For the record, I do not put "economist" behind my name. I am not an economist. I'm informed by economics, as I am by political science, interviews with political practitioners and issue-specific experts, etc.

Posted by: Ezra Klein | March 10, 2011 11:49 AM | Report abuse

For those caught up on #9, please see #10.

Posted by: kungpow9960 | March 10, 2011 12:07 PM | Report abuse


Wasn't accusing you at all by putting "your" in that sentence. I was talking specifically about Krugman.

Posted by: johnmarshall5446 | March 10, 2011 12:09 PM | Report abuse


I think most would disagree about your "zero sum" statement. People would agree that some variant of Newton's 3rd law probably applies,

but the "sum" is ever changing.

Posted by: eggnogfool | March 10, 2011 12:19 PM | Report abuse

Ah, my mistake then.

Posted by: Ezra Klein | March 10, 2011 12:49 PM | Report abuse

I would put #10 on top in bold... we usually are completely unaware of how insular and self-referential the world we construct for ourselves is.

Posted by: kcar1 | March 10, 2011 12:54 PM | Report abuse

kungpow9960 :

You're right... :(

That's a hard one to fix, any suggestions?

Posted by: will12 | March 10, 2011 12:56 PM | Report abuse

As someone writing a doctoral thesis in mathematics about stochastic processes in biophysics, I would assert that there does exist people, myself included, who know what stochastic means.

However, in my attempts to refute will12's comments,* I realized that stochastic can be thought of more generally as a 'lack of information' necessary for deterministic inference in a Bayesian setting. That statement opens more questions than it answers. Thus, it is much simpler to label myself as nobody** and assert #9 holds.

*Probability distributions for most stochastic processes we are interested in are in fact deterministic. That is how we are able to make inferences about their properties.

**As a fraction of the population, the mathematicians, statisticians and physicists who study probability theory in enough depth to follow the arguments about the 'true' definition of stochastic, as it is far from agreed upon, are all effectively nobody.

Posted by: devpsychsmith | March 10, 2011 1:28 PM | Report abuse


Fair enough, could you give me an example of the win-win scenario where nobody loses anything?

Posted by: johnmarshall5446 | March 10, 2011 1:33 PM | Report abuse

11. Economic principles are of limited value in understanding how to run a business.

Businesses are not just economic entities. They are also communities (and often, the community is Peyton Place).

Economic principles don't and shouldn't guide leadership, relationship-building, or, for that matter, product development and marketing.

Just because a business uses money to keep score, that doesn't make it economics.

Posted by: RobertDLewis | March 10, 2011 2:19 PM | Report abuse


The scenario that I'd expect to be convincing toward the more conservative folks might be the 'above the Laffer point' scenario.

If in my community, we have a 100% tax rate on everything produced, and then we just divide it up evenly between citizens, Friedman might argue there's a good chance that you'll operate below full production.

Dropping that tax rate down to 50%, still dividing all tax revenues evenly, if production increases by at least 100%, everyone wins.

If your point is that where there are winners, there are almost always losers, that's fine; I don't disagree. But there can be more Charlie Sheening than Michael Dukakising or more losing than winning, which don't add up to 0; N + 10 - 5 does not equal "zero sum".

Posted by: eggnogfool | March 10, 2011 2:30 PM | Report abuse

A rational, information-based argument will never change the mind of someone who is arguing based on ideology.

Posted by: rashomon | March 10, 2011 2:45 PM | Report abuse

I should note that the reverse is also all too true. Which explains a lot about the current political polarization in this country.

Posted by: rashomon | March 10, 2011 2:47 PM | Report abuse

-Whatever you say as an economics expert will be dehydrated to a kitchen table campaign slogan that undercuts your original idea's validity.

Posted by: mrloucks | March 10, 2011 3:25 PM | Report abuse

One problem with economists is that money is not a "unit of account." Money is a vote. Money is a direct, legally backed claim upon the allocation of scarce social resources. That's what a vote is. The value of money is the quantity of money divided by the amount of scarce social resources.

Voting is merely an indirect claim upon scarce social resources, and voting fails due to both a principal-agent tension and the speed/relative strength of money.

This is why we live in any oligarchy and not a "representative democracy" - the rich simply have more say as to the allocation of scarce social resources, and the poor have almost no say. A positive feedback loop makes the rich richer, and the poor have no vote; and because the poor cannot financially compensate those who help them, altruism toward the poor is punished in our society...yet, each of us WANTS to be altruistic. That is human nature.

The reason giving is so satisfying is because way back when, we needed to help each other in order to survive. If you have food now, you might not have it tomorrow, so you'd better share. Now, wealth is permanent, so there's no need to share and no incentive to share. There is, however, an incentive to hoard or cheat or pollute to obtain and keep money.

Therefore, Money is a Law, because it governs how we must live. We have created a society in which capital = fitness, and thus we are socially selecting at the upper echelons of power psychopaths who will do anything for money. There is no cost to their fitness if they exploit people or pollute or waste or hoard, and so their insane behavior continues.

The legal system (1) protects the wealthy, who can afford the best attorneys, buy politicians, and manipulate the tax code, and (2) takes violence off the table so the poor can't retaliate, when the rich commit daily acts of violence against the poor. Yet, game theory suggests that both punishment and reward are necessary for cooperation to exist.

So, can we change the law by redesigning money to create "digital financial karma" to capture the "externalities" of people's behavior? Could we each digitally destroy (not take) money, at the same rate (we could have the option to pay $1 per day to destroy $2 of someone's ill-gotten money, depending on how much deflation the Fed wants)?

Could we decentralize the Federal Reserve's creation of money (votes) so that it is no longer created for central banks and lent back to the public? Could we digitally give everyone, say, $3 (votes) per day to buy FOOD so they can have healthy brains?…

Why not?

"There is a sort of poverty of the spirit which stands in glaring contrast to our scientific and technological abundance. The richer we have become materially, the poorer we have become morally and spiritually. We have learned to fly the air like birds and swim the sea like fish, but we have not learned the simple art of living together as brothers." -MLK

Posted by: functionalneuroanatomy | March 10, 2011 4:06 PM | Report abuse

I'm not familiar with a zero-sum rule, but my background makes me ask: at what time scale does this rule act?

I think Ezra's #9 was meant to suggest that people rarely acknowledge the constant presence of the unknown rather than a "i'm better than you" type comment. Stochasticity is the bane of modelling efforts, if the goal is to be correct (but it shouldn't be, see Thomas Kuhn).
Reductionism, leading to the creation of deterministic models, is ingrained in economics and the sciences. In fact, it is the fabric of its being. Throughout these fields, a more holist, system-based approach is needed.

Effective policy decisions cannot be made without the recognition that the economy is not a machine, it is a system. Thus, it is obeys the rules of the ecosystem, and is driven by variation at the level of the community (see RobertDLewis comment). In turn, the community is driven by variation in individual behavior... so on and so forth). Stochastic events occur at every level. Welcome to the infinitely complex. Don't fight it, embrace it.

I think economics and political science should chat with the outcasts over in ecology. Who knows? There might be might be some wisdom in the interactions within the much larger non-human world.

Posted by: JDE906 | March 10, 2011 4:24 PM | Report abuse

Eggnofool wrote:

"Dropping that tax rate down to 50%, still dividing all tax revenues evenly, if production increases by at least 100%, everyone wins."

Sorry I was hoping for real world scenarios. I should have specified. I realize that not everyone would agree. Maybe that's my GOP side in conflict with my Dem side? My Dem side believes in win-win, but my GOP side says if I win somebody has to lose!

Posted by: johnmarshall5446 | March 10, 2011 5:50 PM | Report abuse

@johnmarshall, I might not be understanding what you're saying, but aren't most (or at least many) trading relationships based on win-win?

If I grow tomatoes and have many more than I can eat before they go bad, and you build fences but have already built all you need, don't we both win if I pay you in tomatoes to build a fence for me?

Posted by: dpurp | March 10, 2011 10:07 PM | Report abuse

A fundamental common mistake made by self-proclaimed wonky-type economics-leaning policy-interested bloggers-cum-journalists: overreliance on politically-active economists with stellar academic standing (at least at some point) to interpret for them technical literature they (the bloggers-cum-journalists) can’t understand on their own. Among other things, Milton Friedman was famous for asking, ‘How do you know?’ Well, how do these bloggers-cum-journalists ‘know’? Because someone else told them it was so?

Posted by: rothmanp | March 10, 2011 10:19 PM | Report abuse

Alan Krueger recently gave a talk at the Barcelona Graduate School of Economics about economic policymaking and the role of academic research in that process. Highlights of the talk are available on YouTube:

Posted by: barcelonagse | March 11, 2011 5:07 AM | Report abuse

Economic statistics are valid? It often seems that such stats are like the drunk searching for car keys under the streetlight, they're measuring what's easy to collect and measure, or rather what someone sometime in the past was able to measure and not what is important.

I suppose this goes hand in hand with assuming that all that is economic is important and vice versa--if money's not involved, as in the benefits of the Internet or the costs of homemaking, it's not economic activity and not important.

Posted by: bharshaw | March 11, 2011 8:57 AM | Report abuse

I'd like to hear an elaboration of how exactly a lack of inefficient unions would lead to a "dangerous imbalance of political power in the long term."

Posted by: Hansonresponse | March 11, 2011 9:55 AM | Report abuse

I think that economist do not account for the absence of perfect information enough in their analysis when it comes to implementing policies. For example, in the debate regarding unions versus regular private employees: Shouldn't there be some consideration put forth regarding the increase in information that unions, whether they be "bad" or "good", are able to gather in employment negotiations?

Posted by: dnquiggs | March 11, 2011 11:23 AM | Report abuse

"If a policy makes sense only in the presence of a secondary compensatory policy — say, a regressive tax where low-income folks get some sort of refund — then you have to ask yourself whether the compensatory policy will pass."

I'd go even further and say that the compensatory policy will always and everywhere be under threat as to its very existence and will always be underfunded.

Posted by: georgepurcell1 | March 11, 2011 12:49 PM | Report abuse

@will12, I actually do have a suggestion for fixing #10, sign up to mentor a kid in a DC public school. You will learn a lot about the rest of the world that way.

@Hansonresponse, the choice is not between "inefficient unions" and some better, more rational voice for the interests of worker. It's between unions in all their inefficiency, cronyism and imperfection and a workforce completely at the mercy of Walmart, the Koch Brothers, etc.

Posted by: wrybread | March 11, 2011 1:35 PM | Report abuse

Dear economists,

People are not human calculators. Most of us do not have the time, resources, and know-how to assess the thousands of competing options and products available to us on the market. Often we use heuristics, or judgmental shortcuts, to make purchasing decisions. That means private solutions to retirement and health-insurance plans are not feasible, if the market offers 1,000 different plans written in legalese. Most consumers will be overwhelmed and forced to make uninformed decisions on highly important and very, very expensive matters. Bottom line: Keep in mind the practical constraints of being human when you offer policy solutions.

Posted by: thrustmoney | March 11, 2011 5:44 PM | Report abuse

"Simple programs and rules are often better in practice, even if they’re worse in theory."

This from one of the Obamacares greatest cheerleaders.

Posted by: invention13 | March 12, 2011 1:48 AM | Report abuse

Post a Comment

We encourage users to analyze, comment on and even challenge's articles, blogs, reviews and multimedia features.

User reviews and comments that include profanity or personal attacks or other inappropriate comments or material will be removed from the site. Additionally, entries that are unsigned or contain "signatures" by someone other than the actual author will be removed. Finally, we will take steps to block users who violate any of our posting standards, terms of use or privacy policies or any other policies governing this site. Please review the full rules governing commentaries and discussions.

characters remaining

RSS Feed
Subscribe to The Post

© 2011 The Washington Post Company