Nobel in Economics Given For "Mechanism Design Theory"

As AP reports:

Americans Leonid Hurwicz, Eric S. Maskin and Roger B. Myerson won the Nobel prize in economics on Monday for developing a theory that helps explain situations in which markets work and others in which they don't.

The three researchers "laid the foundations of mechanism design theory," which plays a central role in contemporary economics and political science,

……..
Essentially, the three men, starting in 1960 with Hurwicz, studied how game theory can help determine the best, most efficient method for allocating resources given the available information, including the incentives of those involved.

Tyler Cowen, author of Discover Your Inner Economist and reason contributor, wonders if this was a great choice. Some in his comments thread think it was a better idea than Cowen did.

UPDATE: And Peter Boettke at the Austrian economists' blog sees some Hayekian background in Hurwicz's work:

Mechanism design theory ---- which seeks to find rules of the game so that the institutional structure operating under those rules will produce social optimum ---- was a by-product of Hayek's informational and incentive challenge to socialism and market socialism in the 1930s and 1940s.  Mathematical economists who took Hayek's challenge seriously set off on the path to provide the appropriate mechanism design that would meet Hayek's challenge.

Read his whole post for elaboration. 

Comments

[The Extispicator]

I'm torn here. I'm proud that Americans won the prize. Yet my libertarian brain finds it difficult to believe there is such a thing as a situation in which markets don't work.

[iih]

I heard only a little about "mechanism design theory" on NPR, something to do with determining when and how much government and other non-economic human factors can be used to intervene in free markets. But I am now very curious.

Brian (or someone else for that matter): Can you elucidate?

[Brian Doherty]

IIH---I'm not coming into this with any pre-existing expertise in the winner's field--but the links from my entry and from Tyler's blog post will give you plenty of background.

[iih]

Brian: Thanks. I will also do a quick library/Internet, but was interested in a "libertarian" take on the topic. I guess I will find that out for myself once I read about the topic.

[Neu Mejican]

iih,

The important summary paragraph:

In some cases, no market mechanism can ensure a fully efficient allocation of resources. In such cases, mechanism design theory can be used to identify other, more
efficient institutions. A classic example concerns public goods, such as clean air or national security. Paul Samuelson (1954) conjectured that no resource allocation mechanism can ensure a fully efficient level of public goods, because "it is in the selfish interest of each person to give false signals, to pretend to have less interest in a given collective activity than he really has..." (page 388 op. cit.). Mechanism design theory permits a precise analysis of Samuelson's conjecture. More generally, the theory can be used to analyze the economic efficiency of alternative institutions for the provision of public goods, ranging from markets and consensual collective decision-making through majoritarian decision rules all the way to dictatorship. An important insight is that consensual decision-making is frequently incompatible with economic efficiency. The theory thus helps to justify governmental financing of public goods through taxation. Applications of mechanism design theory have led to breakthroughs in a number of other areas of economics as well, including regulation, corporate finance, and the theory of taxation.

Follow the "wonders" link above for the full explanation of the theory...

The libertarian take will be interesting...the theory works from a "fact" ("In some cases, no market mechanism can ensure a fully efficient allocation of resources.") that many libertarians are uncomfortable with. I predict many arguments in this form:

Libertarians don't expect "fully efficient allocation of resources...markets provide "better allocation, not perfect allocation." Of course the theory that the Nobel committee honored disagrees with the "better" assertion in that statement.

[iih]

NM: Where is that excerpt taken from? Thanks.

[Reinmoose]

iih

here: http://nobelprize.org/nobel_prizes/economics/laureates/2007/ecoadv07.pdf

[Neu Mejican]

This mechanism is feasible only if participation is mandatory. If participation is voluntary and decisions to start the project must be taken unanimously, then the problem of free-riding becomes severe. Using techniques
developed by Myerson (1981), Mailath and Postlewaite (1990) show that the probability
of funding a public-goods project tends to zero as the number of agents increases. They
give an example where the asymptotic probability of funding the public project is zero despite everyone knowing that they can be jointly better off if the project is funded.

These results provide a rigorous foundation for Samuelson's (1954) negative conjecture
about public goods cited above (Section 1). They give a plausible explanation for
observed failures to provide public goods. For example, the fact that English villages
were much earlier than French villages in deciding on public goods such as enclosure of
open fields and drainage of marshlands can arguably be ascribed to the fact that French
villages required unanimity on such issues whereas the English did not. This may at
least partially explain why the productivity growth in English agriculture outstripped
that of French agriculture in the period 1600-1800 (Grantham, 1980; Rosenthal, 1992)

More from the link Reinmoose gives...

The crux of the debate...

[iih]

Reinmoose, NM: Thanks.

[Neu Mejican]

A related current event:

BEIJING (Reuters) - China's Communist Party must stay firmly in charge as the nation embraces economic and social change, President Hu Jintao said on Monday in an agenda-setting speech vowing tightly controlled political reforms.

In a "state of the nation" report to the 17th Party Congress, Hu said that the country he has led for five years would pursue an increasingly open economy but also had to surmount social fissures and an environment battered by breakneck growth.

"Our economic growth is realized at an excessively high cost of resources and the environment," he said, drawing dutiful applause from carefully chosen delegates.

But Hu said the country's future was promising -- and even some political loosening was possible -- as long as the Communist Party maintained its long-unchallenged domination.

"China is going through a wide-ranging and deep-going transformation. This brings us unprecedented opportunities as well as unprecedented challenges," Hu told over 2,200 delegates -- one of them his predecessor Jiang Zemin, who appeared to doze through stretches of Hu's recital of slogans and goals.

"We must uphold the Party's role as the core of leadership in directing the overall situation and coordinating the efforts of all quarters," Hu said in the speech lasting more than two hours.

The question, of course, is whether China's Communist party is using mechanism design theory to assure that their use of force maximizes social benefits, or if they are just maximizing benefits for the ruling class.

[x,y]

The theory thus helps to justify governmental financing of public goods through taxation.

IOW, the theory helps justify the forcible taking from one even if he actually has less interest in a given collective activity.

[Neu Mejican]

x,y,

theory helps justify the forcible taking

Yes.
It makes an outcome-based argument to justify a means.

So the argument against it is not based on outcomes but on means.

Debate on policy always boils down to: "which ends justify which means?"

Right?

[joe]

But...Demand Kurv!

This year's winners of the Nobel Prize in Economic Science obviously need to take Econ 101.

[robc]

Debate on policy always boils down to: "which ends justify which means?

Thats an easy debate. Ends never justify means. Immoral means are immoral. We should choose amongst the moral means to reach the ends we desire.

[Neu Mejican]

robc,

Yes. It is always easy to debate the absolutist position. Libertarianism, however, is not an absolutist philosophy.

Immoral means are immoral. We should choose amongst the moral means to reach the ends we desire.

This statement does not, actually, change the nature of the debate. It gives a parameter upon which we can make a determination as to whether the particular means being considered is justified by the particular end desired, but it does not change the fundamental nature of the debate. To do so would assume that there is an objective moral criteria that can be used to exclude certain means.

[Neu Mejican]

"assume" = "require"

[Neu Mejican]

fwiw, imho, it seems that libertarianism is on stronger ground making the "immoral means" argument than the "better outcome" argument, but this has the habit of leading to the anarchist's position.

If government is an immoral means due to the use of force required to implement it, then government is always immoral and never justified (since outcomes don't matter). But the libertarian position doesn't go this far. It requires only that government intervention is minimized to minimize the negative outcomes that adhere to the use of force. As such, it is an outcome-based philosophy, at least to a degree.

If this is true, then outcomes matter and taxation to address the free rider problem may be justified in certain cases. So the debate moves towards defining those cases where the outcome justifies the means.

[iih]

NM: What about the "free-market" argument. Don't interfere and the invisible hand will take care of everything?

[joe]

How can we judge the morality of any means, if not by the ends it produces?

Torturing someone for information about terror cells, for example, is immoral because of the ends it produces - the agony, fear, and physical/psychological/spiritual harm it does to the prisoner.

[joe]

NM,

The most clever answer to that question I've seen came from a righty commenter on these very threads, on the subject of "taxation = theft."

Theft is defined as the unauthorized taking of someone else's property. The Constitution authorizes spending tax dollars on an army, but not only Medicaid.

Ergo, the taxes collected for Medicaid are unauthorized, and therefore theft; whil the taxes collected in exactly the same manner which go towards the Pentagon budget are authorized, and therefore are not theft.

[kohlrabi]

Neu Mej.

It seems your position takes for granted that taxes will we levied to account for a free rider problem. In the case that gov't funding is voluntary, or user-fee-based, you have free riders, sure, but you can also justify a libertarian position while eschewing outcome based justifications.

[kohlrabi]

In other words, it is possible for the government to protect individuals rights and property with police and courts without the immoral means of taxation. Feasible? We can argue all day, but I don't believe libertarianism necessitates any amount of coercion, even for the minimal state.

[Michael A. Clem]

Of course there are situations where the market doesn't work (or at least, doesn't work as well). Situations involving coercive interference in the market, and situations involving undefined or poorly defined property rights. The real question is whether we're better off simply creating or protecting the situations where the market will work, or if there's some alternate way of handling those situations. I know what I prefer, but perhaps this will help prove it.

[Justin Rietz]

RE: public goods

As far as clean air / pollution, we have a mechanism in place already - courts and class action lawsuits. If a factory pollutes, and it negatively affects others, the factory may be taken to court and sued for damages.

RE: taxation as theft

A national sales tax would not be theft, as one would have the choice not to purchase anything. In Minnesota, there is a sales tax that excludes "necessities" (basically food and clothing), thus allowing individuals to avoid the tax altogether yet still purchase the essential goods needed to survive.

Another benefit is that a sales tax does not punish hard work (as an income tax does) and creates an incentive for saving (which the U.S. sorely lacks).

[robc]

To do so would assume that there is an objective moral criteria that can be used to exclude certain means.

Yes.

Um, I fail to see the problem here.

[robc]

If government is an immoral means due to the use of force required to implement it, then government is always immoral and never justified (since outcomes don't matter).

Lets apply some basic logic here

A->~B
B
Therefore: ~A

If use of force implies government is always immoral
and
government is sometimes moral
therefore
our premise that use of force is always immoral must be wrong

[joe]

Justin Reitz,

What if six factories pollute a certain town, causing its brain cancer rate to triple.

Which cancer patients get to the be "excess" ones caused by the pollution, and which ones get to be "ordinary" people with brain cancer who would have gotten it anyway?

Once we've settled that question, which set of patients has their cancer caused by Plant 1, which by Plant 2, et al.?

RE: taxation. In practice, person has just as much choice not to earn income as he has not to buy goods.

[joe]

robc,

Why did you just restate N.M.'s argument?

[R C Dean]

Debate on policy always boils down to: "which ends justify which means?

This can only be true in a country where there is no concept of limited government and no separation of powers (IOW, in the Total State).

Otherwise, debate on policy always begins with "who does this problem belong to?"

[R C Dean]

We can argue all day, but I don't believe libertarianism necessitates any amount of coercion, even for the minimal state.

I do. In a minarchy, there are laws that are enforced by the state. You cannot enforce a law without coercion.

[joe]

RC Dean,

Arguing that something goes beyond the proper limits of government is an argument about the ends, too. In this case, one would be arguing that the means would result in an undesired end - the expansion of govenrment into an improper sphere, and the harms that result from that.

[prolefeed]

The theory appears to assume that it is possible to quantify optimal outcomes for many people, and then to achieve them via non-market processes -- both really shaky assumptions, especially if you throw politicians into the mix.

It takes a great deal of chutzpah to say "I know what each person desires more than they do, and I can accurately sum all those desires for a crowd and weight them all accurately -- without using the price mechanism utilized by markets -- oh, and I will be fair and disinterested and not use the process for my own personal gain."

Markets aren't perfect, but the price mechanism takes into account specialized, local information that individuals possess. Ironically, this theory will likely be seized upon by statist politicians to push agendas that demonstrably produce bad, suboptimal outcomes.

[robc]

Why did you just restate N.M.'s argument?

My point is subtly different than his. Apparently too subtle.

[squarooticus]

If the goal were maximum efficiency, the state would simply drug or brainwash us all into submission and then control us like cogs.

Uhhh....

And of course, even if (even though?) the government does this, that doesn't mean they'll achieve maximum efficiency anyway. 😉

For me, the goal is freedom. If that comes along with lower-than-optimal efficiency, so be it. We are not cogs in a machine, so coercion should be used only to prevent violence against others, not to maximize efficiency at the expense of personal freedom, which I define very broadly.

[robc]

Let me try again, with more quoting:

imho, it seems that libertarianism is on stronger ground making the "immoral means" argument than the "better outcome" argument, but this has the habit of leading to the anarchist's position.

I agree with NM on first part but not where it leads to.

If government is an immoral means due to the use of force required to implement it, then government is always immoral and never justified (since outcomes don't matter). But the libertarian position doesn't go this far.

This is where I used logic, we are in sorta agreement here but I assumed that there was an unstated result from my logic.

It requires only that government intervention is minimized to minimize the negative outcomes that adhere to the use of force.

Wrong. It requires that some use of force not be immoral. Outcomes still dont matter.

As such, it is an outcome-based philosophy, at least to a degree.

Wrong. Still means based, just have to accept that use of force is sometimes a moral means.

So the debate moves towards defining those cases where the outcome justifies the means.

No, it moves towards defining what use of force means are acceptable.

I can see NM's point, that some might say that it is acceptable when a certain outcome is at stake. That is where I subtly disagree, but I guess it wasnt obvious that that is where the ~A logic leads us.

[Neu Mejican]

robc,

our premise that use of force is always immoral must be wrong

See joe's question.

Um, I fail to see the problem here.

The problem is that I (perhaps) think your moral criteria stink, so you and I will choose different criteria to use to measure a means moral standing...stated differently, moral criteria are subjective, not objective.

kohlrabi | October 15, 2007, 2:19pm | #

It seems your position takes for granted that taxes will we levied to account for a free rider problem. In the case that gov't funding is voluntary, or user-fee-based, you have free riders, sure, but you can also justify a libertarian position while eschewing outcome based justifications.

If by my position you mean this

"If government is an immoral means due to the use of force required to implement it, then government is always immoral and never justified"

This is not a statement regarding taxes. Government, by definition, involves regulation of behavior that would force individuals to behave in ways they otherwise would not. If everything is voluntary, there is no need for government (think Proudhon, or syndicalist schemes, or anarcho-capitalism).

iih,
The invisible hand argument is about outcomes, and the theory which won the Nobel gives arguments to the effect that non-market approaches out perform markets in some cases.

[robc]

For me, the goal is freedom. If that comes along with lower-than-optimal efficiency, so be it.

My measue of efficiency is how much freedom we have. Its hard for freedom to lead to lower-than-optimal efficiency in that case.

[robc]

The problem is that I (perhaps) think your moral criteria stink, so you and I will choose different criteria to use to measure a means moral standing...stated differently, moral criteria are subjective, not objective.

Just because we disagree (perhaps) on moral criteria, doesnt make it subjective. It just means one (or both) of us is wrong.

[Neu Mejican]

robc,

Thanks for the expanded version.

just have to accept that use of force is sometimes a moral means

What objective criteria makes the moral distinction? My guess is that it depends upon a judgment of the ends produced once you boil it all down to basics.

[Neu Mejican]

Just because we disagree (perhaps) on moral criteria, doesnt make it subjective. It just means one (or both) of us is wrong.

Ok. I'll accept that.
But in the end, where is this objective moral code that determines which of us is right or wrong?

[joe]

Wrong. Still means based, just have to accept that use of force is sometimes a moral means.

And how do we make such a distinction, apart from looking at the ends they produce?

[Neu Mejican]

joe,

There is no rational reason that some criteria other than the ends produced can't be used to make the moral judgment...

It is possible to reject ends-based arguments outright, but doing so tends to lead to absolutist positions or a great deal of mental gymnastics, imho.

[robc]

But in the end, where is this objective moral code that determines which of us is right or wrong?

And how do we make such a distinction, apart from looking at the ends they produce?

To both of you, I can give two reasonable answers:

1. How the fuck should I know?
2. It comes from God.

I lean towards 2, but my interpretation depends heavily on 1.

[Neu Mejican]

robc,

1. How the fuck should I know?
2. It comes from God.

Which is why, in practice, policy debates boil down to "which ends justify which means?"

[robc]

It is possible to reject ends-based arguments outright, but doing so tends to lead to absolutist positions or a great deal of mental gymnastics, imho.

Ends based is easier, I admit. But, I get that queasy internal feeling of wrongness when I catch myself making arguments that way. Then I either perform the gymnastics or become an anarchist for a short while. After about 2 beers, I become more anarchist with each.

[robc]

NM,

Which is why, in practice, policy debates boil down to "which ends justify which means?"

I didnt say applying God's standards was easy.

[Neu Mejican]

robc,

I didnt say applying God's standards was easy.

It is not applying them that is difficult.
It is knowing what they are that is difficult.
:^)

Applying some version of God's standards is always the easiest approach. All you have to do is look up the answer in the book. Crib notes don't work with the empirical approach.

[robc]

Applying some version of God's standards is always the easiest approach.

The tricky bit is figured out which is the real one.

[joe]

OK, robc, you got me.

I was asking in terms of reasoning to a moral conclusion.

NM,

How? Even in the case of "God said so," the argument against the action is based on not wanting to accomplish the End of "people disobeying God."

[joe]

I guess what I'm asking is, isn't all moral reasoning based on looking at ends?

The old saying about the ends justifying the means seems to assume that only the desired ends, and not the other ends that come about through the application of the means, are to be considered.

[x,y]

RE: taxation as theft

A national sales tax would not be theft, as one would have the choice not to purchase anything. In Minnesota, there is a sales tax that excludes "necessities" (basically food and clothing), thus allowing individuals to avoid the tax altogether yet still purchase the essential goods needed to survive.

Another benefit is that a sales tax does not punish hard work (as an income tax does) and creates an incentive for saving (which the U.S. sorely lacks).

A national sale tax would still be legalized theft. According to your logic, an income tax is not theft either because people have the choice to not work.

And who defines what the necessities are?

[Justin Rietz]

Re: x,y & taxation as theft:

True, one could choose to not work. But most people could not survive if they didn't work. I'm not providing the "perfect" system, just one that is perhaps more fair and provides better incentives.

As far as determining what are necessities, food and clothing are included. Again, it isn't perfect, but does provide some leeway for the poor (versus a straight sales tax) for whom necessities are a large portion of their spending.

Re: Joe & pollution:

This is currently how are court system works and therefore no new regulations or laws are required. Both sides present their case (scientists providing evidence regarding the effects of pollution, how much pollution each factory creates, how far that pollution can travel, etc.) and the jury decides.

[joe]

Justin,

This is currently how are court system works... And in the case of environmental pollution, our court system works very, very poorly. Both from the point of view of those on the receiving end of the pollution, as well as for industry, owing the problems I described.

In the example I gave, even if it was definitively proven that it was the plants' waste that increased the brain cancer rate, it would be impossible to show a causal link between Plaintiff A's harm and Plant 3's emissions.

[Justin Rietz]

In tort cases, juries are allowed to portion out blame. So if Plant 3 provided 10% of the pollution, and Plant 10 provided 90%, damages could be enforced at the same ratio.

[Mr. Nice Guy]

"Markets aren't perfect,"

Heresy! Heresy! Drive this man from the Temple!!! Markets will get rid of discrimination, raise all boats, provide safe food and drugs with no policing, and deflect meteor showers when prayed to correctly, every libertarian knows that!

[Mr. Nice Guy]

"Of course there are situations where the market doesn't work (or at least, doesn't work as well). Situations involving coercive interference in the market, and situations involving undefined or poorly defined property rights."

Of course magic does not always work, like when the laws of magic prevent it from working.

Jeez, just say that the field of economics, which has of late expanded to more than simply the work of Hayek and Mises, does not treat your ideological positions as empirical facts...

[Neu Mejican]

joe,

Even in the case of "God said so," the argument against the action is based on not wanting to accomplish the End of "people disobeying God."

I believe you are over thinking this.

If you believe that the means are moral or not based on some criteria or principle, then their outcome is orthogonal to the moral calculus. The fact that the means violates the principle or criteria is not in any sense an outcome (i.e., "end") of the means. It is an aspect or feature of the means before the means is implemented (or before a rational prediction of the outcome is made).

[Neu Mejican]

joe,

not the other ends that come about through the application of the means, are to be considered.

To expand a bit.
If, let's say, force is immoral by definition (not because it creates an undesired end, but because it is unequivocally immoral based on an axiomatic assumption or belief), then a means that includes an aspect of force is immoral without consideration of other factors. The immorality is a constituent of the means so it can be rejected without the ends even coming up for consideration.

If by convention all things that bark (use force) are by definition dogs (are immoral) then as soon as barking (force) is recognized that thing (means) can be identified as a dog (immoral).

[Neu Mejican]

robc,

I just saw this one...

My measue of efficiency is how much freedom we have. Its hard for freedom to lead to lower-than-optimal efficiency in that case.

Right. All you have to do is figure out how to properly distribute that freedom.

;^)

[Mr. Nice Guy]

NM-why is the use of force so special (I'm not saying you are arguing this, but even if you are not you may know more about how libertarians think about this than I)? I've never seen why this is given such special treatment. I mean, if it is because it often causes pain, then other things fit (how many of us as kids would take a spanking over a grounding any day, or even a guilt trip from mom). And if it is because it limits autonomy then so does other things, such as the withholding of love and affection by certain significant others in ones life or the withholding of needed resources (pay, food, housing). It might be said "well in non-force situations you agree to limit your autonomy in order to avoid the sanction." Well my reply is that this is true in force situations as well. Just as the person could just take the pain of having love withheld, he could also take the pain of being kicked in the stomach every time he does x...I could also add that the ability to withold some of the things I mention is of course ultimately backed up via force in libertopia (I can refuse you certain resources no matter how much you need them because we have defined them as "my property" and libertopia allows the police to come womp you if you try to take them).

[iih]

Ron Paul wins NV straw Poll!

[iih]

And a great piece on RP in the WaPo.

[Neu Mejican]

Mr. Nice Guy,

Although a pretty basic concept when discussing moral issues, force is not special in my view, but it is one of the pillar principles for libertarians.

From Wiki:
The non-aggression principle (also called the non-aggression axiom, anticoercion principle, or zero aggression principle) is a deontological ethical stance associated with the rights-theorist school of the libertarian movement (consequentialist libertarians do not base their ethics on it[1]). It holds that "aggression," which is defined as the initiation of physical force, the threat of such, or fraud upon persons or their property, is inherently illegitimate. The principle does not preclude defense against aggression.

Now, as to how legitimate it is to define force as physical force (and to include your property as part of your being), that can be contentious. Physical force seems hardly the only kind implied by this principle (particularly if you think of it in relation to liberty or autonomy, force being the opposite).

http://en.wikipedia.org/wiki/Non-aggression_principle

[Neu Mejican]

FWIW, the central feature of ethical discussions, it seems, is an attempt to create dichotomous categories (moral vs. immoral). This is typically attempted by listing definitional features of each category (e.g., if it includes force it is immoral). The problem with this approach is that it ignores the basic nature of natural categories, which are not feature driven, but created through analogy. Because they are not feature driven, not list of features will completely define any natural category (including categories of moral or immoral behavior).

It is the fuzzy nature of categories that creates the conflict between different feature driven systems.

[Michael A. Clem]

Mr. Nice Guy, why did you cut off the rest of my comment? Did you think the question I presented as being irrelevant or inconsequential? Isn't the whole point of this thread a question of what people should do, and how, or if, Mechanics Design Theory is helpful in that regard?