The Top 20

An extremely distinguished committee of economists has selected the top 20 articles published in the last 100 years in the American Economic Review, widely recognized as one of the top journals of the profession. All 20 are publicly available, via links from the committee’s report.

The 20 choices are uniformly excellent, and taken together they give a good sampling of the ideas that have changed the way economists think. Of course, many equally influential articles were disqualified by virtue of appearing in journals other than the AER. (The first few that come to mind are Lucas on Expectations and the Neutrality of Money, Coase on The Problem of Social Cost, and Lucas again on The Mechanics of Economic Development).

Some of these are pretty technical. One that’s not is Hayek’s 1945 classic on The Use of Knowledge in Society, which is both one of the clearest and most profound essays in the history of economics. In fact, its clarity tends to mask its profundity; once you’ve read it, you feel sure you must have understood this stuff all along.

The Cliff Notes version comes down to this: Allocating resources efficiently requires vast amounts of information; prices (and only prices) can convey that information. Without prices, you’re sunk. It follows that when prices are distorted (through, say, “equal-pay-for-equal-work” laws) or when we choose to ignore their informational content (on the advice, of, say, a well-meaning locovore), we impoverish ourselves far more than is visible to the untrained eye.

As I said, all 20 articles are available for free, a price that reflects the fact that you can glance them over without consuming substantial resources — and that encourages you to act on that information. I second the encouragement.


35 Responses to “The Top 20”

  1. 1 1 Dave


    As a staunch anarchist libertarian, I am often faced with the dilemma of market failure however. Yes prices are vital but they also need to capture all externalities.

    When I first discovered the wonders of the pricing system (via many books including the Armchair Economist) many years ago, I started to cringe at the thought of any government involvement in ANYTHING. But how does one deal with externalities? The only way I could justify zero (non market) regulation was by dismissing externalities like global warming as highly unlikely, minute enough to be irrelevant to the calculation, and sometimes even positive (eg longer crop cycles in cold regions).

    I think that is somewhat satisfying but a bit of a cop out. It acknowledges the importance of pricing but somewhat pretending that they are largely all capturing.

    There is obvious danger in the market mispricing and ignoring a negative externality – to the point of global annhilation. The tragedy of the commons runs the other way in this instance.

    Not that I want government involvement in finding the solution. Those that promote regulation to capture externalities are usually blind to the real costs of having a government at all (which are usually elected by who can dazzle the electorate with bs the best) and therefore the externalities of regulation are dismissed.

    But trying to argue this is too hard. I would much rather risk being pushed around by some bureacrats to fill out forms all day long than have the world completely decimated.

    What to do?

  2. 2 2 William


    I don’t have the answer, but at least to help think about externalities I’d recommend Coase’s “The Problem of Social Cost” and Harold Demsetz’s “The Exchange and Enforcement of Property Rights”.

  3. 3 3 William


    Maybe I should sell my recommendations a bit more. Coase and Demsetz point out that “externalities” aren’t designated as such by God or some universal law. The fact that some effects remain “external” to market calculation reflects the fact that the cost of taking those effects into account exceeds the benefit. Unless government can do something to solve that problem, asking government to solve all externality problems can lead to less efficient outcomes.

  4. 4 4 Dave

    “The fact that some effects remain “external” to market calculation reflects the fact that the cost of taking those effects into account exceeds the benefit. ”

    Isn’t that a differentiation between the individual and the whole? And also across time? And region?
    I won’t be facing any pain if my great great great grandchildren drown or some Bangladeshi farmer chokes on my fumes.

  5. 5 5 Alan Wexelblat

    I can’t resist this. Does the fact that the articles are available for free mean they have no value? The price is known (zero) but what does that tell us about resource allocation and such?

  6. 6 6 JLA

    The Coase Theorem:

    “Parties internalize externalities efficiently unless something exists that prevents parties from internalizing externalities efficiently.”

    Isn’t this tautological? Perhaps it’s a useful tautology, but I’m not convinced. The Myerson-Satterthwaite theorem proves that because of asymmetric information, efficient trade does not always happen.

  7. 7 7 Silas Barta

    The dismissing of locavores is usually too hasty. While prices do convey important information, they require well-defined rights in the inputs for the goods that have prices. And that’s arguably what we don’t have for transportation — no market in roads, no market in consumption of atmospheric resources.

    So yes, many people are revealed to be ignorant when they forget to look at what prices are telling them in terms of what’s wasteful. But low prices need not indicate non-wastefulness. For example, if a lake is being unsustainably depleted of fish (due to lack of property rights), then naive economists (ahem…) may say, “Hey, prices are low enough that people keep buying the fish, so they must not be wasting any.” And they continue to be low until that fish becomes unavailable (and yes, this has happened many times).

    Rather than focusing the more foolish locavores, you should engage the ones with a more nuanced position — specifically, the one making an analogous argument to the fish example: that the lack of property rights related to food transportation puts us in the position of overfishing. The prices don’t reflect the fact that we’re overusing a given resource and thus we’ll eventually hit a shock and/or have to impose drastic measures later to repair the damage.

    Also, what’s with “*only* prices convey the relevant information”? If that were true, no one would ever have to talk to their suppliers, negotiate, discuss alternate arragments, etc. They could just post prices, adjust them, and get to work. Also, that position would imply that no one ever benefits from non-price market research (e.g. focus groups) or is able to launch a fundamentally new products. (iPods — the Apple ones — were not “too pricey” before Apple introduced the first model. Rather, the price didn’t exist — maybe not even the demand.)

  8. 8 8 Steve Landsburg

    Alan Wexelblat:

    Does the fact that the articles are available for free mean they have no value?

    No more than the fact that water is cheap means we can largely live without water.

    The price reveals that it costs almost nothing to provide these articles to you. The fact that their value exceeds their price is a particular case of the general phenomenon that allows us to keep getting richer.

  9. 9 9 Steve Landsburg


    “Parties internalize externalities efficiently unless something exists that prevents parties from internalizing externalities efficiently.”

    Is that really all you took from Coase’s paper?

  10. 10 10 JLA

    I think there’s great value in Coase’s paper. His claim that people are incorrect to obsess about the notion of fault is spot on. I just think that what has become known as the Coase theorem is either tautological or incorrect (a la Myerson-Satterthwaite).

    It’s certainly possible that I’m incorrect. I just don’t see where my mistake is. Could somebody point out my error?

  11. 11 11 Ricardo Cruz

    Dave, one problem with having the government fixing things is the one we are experiencing today. Educated voting is an externality, so most people will just vote based on the politician’s demeanor or whatever.
    (But if you’re into market anarchism, you probably already know those arguments from David Friedman and Bryan Caplan.)

    In the case of global warming, given there are many countries in the world, it will be tough to fight that externality. Even if governments come to an agreement, it will always pay off to do a bad job regulating their own negative externalities.

    Anyhow, I don’t disagree with you. I’d love to see, for instance, my city taxing car drivers (for their air pollution, noise, hazard, the space they take, etc). I just think that will never happen. (That will only happen if the number of drivers shrink for some reason, in which case the benefit won’t be as large, though it will be probably be positive still.)

  12. 12 12 Ken B

    Every time I read Hayek I get the feeling he has articulated my own ideas not just more clearly than I can articulate them, but more clearly than I can think them.

  13. 13 13 Ken B

    @JLA: A fluid will reach thermodynamic equilibirium unless something prevents it. This is not a tautology because 1) it might logically be the case that there is no thermodynmaic equilibrium 2) some fluids might never reach equilibrium by their very nature but others might 3) context tells you what the somthing is: an energy source.

  14. 14 14 Silas Barta

    Is that really all you took from Coase’s paper?

    No, he also got the idea that you can sound clever by saying, “Let’s get rid of the notion of fault” and then re-justify its use under a different name (“Hey, it would be great if rights were stable too, so maybe women shouldn’t have to buy out rapists.”)

  15. 15 15 neil wilson

    I find it amazing that anyone is willing to defend paying a black man less than paying a white man for the same work: or hiring a white man when a qualified black man would work for less.

    There is a strong emotional benefit to me to hire someone who looks like me. Therefore, I am making a rational decision to hire the white guy.

    We have had over 100 years where blacks were hurt by the rational view that my customers would rather deal with higher prices or worse service because they felt the emotional benefit of working with a white man were worth it

    My kids care far less about race than my grandparents.

    The law is a good law and you are nucking futs to think otherwise.

  16. 16 16 Ken B

    @Neil Wilson: I get the feeling you posted here in response to a converstion elsewhere. Who’s talking about anti-discrimiantion laws here?

    Worth noting though that not just blacks suffered for a century (more in fact) but so did whites and everyone else. Say I was a poor white who’d have preferred the better service. I was denied it. This is part of a general phenomenon. This is not a minor point.

  17. 17 17 JLA

    Ken B:

    I don’t know anything about thermodynamics, but that sounds valid. I guess my problem is with the definition of “transaction costs.” To be honest, I don’t have the slightest clue what counts as a transaction cost and what doesn’t.

  18. 18 18 Bennett Haselton

    Ken B I assume Neil Wilson is responding to this part of Steve’s post: “It follows that when prices are distorted (through, say, “equal-pay-for-equal-work” laws)… we impoverish ourselves far more than is visible to the untrained eye.”

    Myself, I think the pure-free-market argument against antidiscrimination laws, ignores the fact that the law itself might change people’s preferences by changing the social climate. Before civil rights laws, people discriminate against blacks even at the cost of not hiring the most efficient workers, because that’s their preference. After civil rights laws help to stigmatize discrimination, people *want* to discriminate less than they used to want to, so the laws are not abridging people’s preferences as much as they would seem. Thus, paradoxically, a law that restricts people’s behavior (by outlawing discrimination), can actually leave people more satisfied with their choices than they were before, if the law *changes* people’s preferences by normalizing new choices and stigmatizing old ones.

  19. 19 19 Ken B

    @Bennet: Yes you are probably right about what Neil meant, so I can see some connection. Your thought about laws creating preferences is an interesting one. I wonder if it really works though. The choice to go to church was restricted in Poland and Hungary and other iron curtain countries. Is the church stronger there or in countries like England and Canada? While Brown v Board was the right decision I think it certainly made tensions worse. Young adults binge drink more in places where they can’t drink until 21 than in places (like France) where the laws are less restrictive. So I think your suggested possible benefit is … possible but unlikely, unproven, and possible minor in effect.

  20. 20 20 Al V.

    Re. Hayek, I don’t understand how, if the market sets prices efficiently via a network of effects, banks were making so much money off of financial constructs like Credit Default Swaps and Collateralized Debt Obligations. I would have thought that competetion would have shrunk the margins on those activities to the point where (a) there was much less money to be made, and thus (b) less money to be lost when the CDO market cratered and took CDS’s with it. Is it that there are such high barriers to entry that it isn’t a truly efficient market?

    Hayek’s paper reminded me of the first time I read Dawkins’s The Selfish Gene, in that both describe a system optimized by a multitude of independent actors optimizing their own benefit.

  21. 21 21 Ken B

    Al V: Not that I know much about the peculiar beasts you mention but as far as I can tell the market did drive some of these guys out — hard. There is no guarntee that actors in a market will respond smoothly or smartly, only that eventually they cannot buck the tide. Say there is not profit in buggy whips. Bill Gates could still set up a buggy whip factory and pour billions into it. He can either notice the lack of sales and stop or eventually be forced to stop when his money runs out. Several banks were forced to stop — but then given money by the govt specifically to evade the market’s “instructions.”

  22. 22 22 Ken B

    @Al V: In other words (since rereading my note I see it’s not a masterpiece of clarity): markets don’t make you smart and they don’t make you rational. They just ensure that in the long run the survivors are forced towards smarter more rational choices.

  23. 23 23 Ken

    Alan Wexelblat,

    The articles are not free. They take time to read and much more time to understand. As we all know, time is one of the scarcest of all resources, and one of the most valuable. I’d say the cost of reading and understanding any of the articles posted are more expensive than the computer on which I’m typing.


  24. 24 24 Steve Landsburg

    Ken: The *cost* of reading the articles is non-zero, but the *price* is still zero.

  25. 25 25 Ken


    Fair enough. I know the difference, but sometimes I think of them as the same. A residual of bad thinking habits from when I was younger that I’m trying to kick.


  26. 26 26 Nick

    How do we interpret that not a single one of these papers was published in the past 30 years (with the most recent author famously departing from the world of rational expectations)? What does it say about the slowdown of genuine discovery, increased fragmentation / discordance among opposing schools of thought, and the ephemeral nature of economic models that vigorously realign themselves to explain prevailing phenomena?

  27. 27 27 Steve Landsburg

    Nick: This is a fascinating observation. A more benign explanation might be that nobody wants to go on record endorsing a paper as fundamentally important until the paper’s continued influence has been proven for at least a few decades. But I don’t know how much weight to put on that explanation as opposed to yours.

  28. 28 28 Chicago Methods


    What makes you believe that Lucas has abandoned the idea of rational expectations? This relatively recent article in The Economist seems to suggest that he thinks the criticisms of the DSGE model (I think) and the EMH are not substantial – with an undertone for these criticisms as nothing more than a model of politically-driven economics.

  29. 29 29 Chicago Methods

    PS. Personally, I tend to agree with Kocherlakota’s assessment more than anyone.

  30. 30 30 Chicago Methods

    Whoops. I have to learn to read a bit better. Sorry nick. Steve, go ahead and remove my comments.

  31. 31 31 Bill

    Alan Wexelblat:

    Does the fact that the articles are available for free mean they have no value?

    Value in use vs. value in exchange

  32. 32 32 Andrew B

    @Neil W. “I find it amazing that anyone is willing to defend paying a black man less than paying a white man for the same work” (re: “equal-pay-for-equal-work laws”)

    The problem with your observation, (and it’s the same problem that many feminist’s share when thinking about the inequality of pay), is that, unless you are producing widgets on a factory floor, it is extremely difficult, if not outright impossible, to equate the work of two different employees – *even when they have the same position/title*. There is no such thing as “equal work” really; employees, as a rule, are paid based on the contribution they make to the firm – employers must make judgement calls based on the information at hand, as inadequate as that may be.

    Your normative presumptions assume not only that two (or more)employees can have identical contributions, but that the employers will be able to tell when that’s the case and compensate them accordingly. It’s a nice notion – equal pay for equal work – it just doesn’t reflect reality.

  33. 33 33 Andrew B

    @JLA: on Coase – Pay special attention to Coase’s insights on the reciprocity of externalities. By way of example, “Sure, the sludge I pour into the river is destroying where you swim, but, if you didn’t swim there, we wouldn’t have that problem” – it takes two-to-tango.

  34. 34 34 Harold

    I found Coase’s paper very interesting, and as you say very clear and easy to follow (although not without some effort). Having been introduced to the ideas through your writing it was not totally new, but I can imagine the effect it may have had when it came out.

    He seems clear that to asses the optimum outcome, one must consider all the costs and benefits of the choices. I am not sure if he has neglected one effect. The idea that the liability could fall on either party, eg. farmer or railway, will be a disincentive for the farmer to invest in the first place. If he may become liable for the costs of someone coming along later and starting a rabbit farm or building a railway, he might not bother to get farming at all. I may have missed the point here. Also there is very great difficulty in assesing the costs. In the railway example, it was better for the farmer to bear the costs, allowing the railway to operate profitably. However, it is not possible for the farmer to separate all his land from the railway, whereas it may be possible for the train to be better enclosed at small cost, thereby preventing the sparks. This solution will not become apparent unless the train company has an incentive to look for it, which liability for crop damage would provide.

  35. 35 35 Steve Landsburg

    Harold: Re Coase, you are on to an important point, and there’s a vast literature that addresses this point, reaching different conclusions under different assumptions.

    One argument that has been made is the following: If ranches impose externalities on farms, then farms will never locate next to ranches in the first place, except in those extraordinary circumstances where a particular location is particularly conducive to both farming and ranching. The amount of farming and ranching that takes place in close proximity, then, is determined primarily by the number of such locations, rather than by any private investment decisions — so those investment decisions are not an important factor.

    Of course this argument has spawned many counter-arguments and counter-counter-arguments. Also, it applies only in those cases where two firms must be located near each other in order to impose externalities (as opposed to, say, the steel factory whose smokestack imposes costs on all the laundries for 50 miles around).

    There’s a great article by H.E. Frech showing that under plausible assumptions, the considerations you’ve raised must lead to inefficient outcomes regardless of the choice of liability rule.

Comments are currently closed.