1. Skip to navigation
  2. Skip to content
  3. Skip to sidebar
Source link: http://archive.mises.org/5572/stiglitz-is-wrong-on-government/

Stiglitz is Wrong on Government

September 6, 2006 by

Joseph Stiglitz shared the Nobel Prize in 2001 partly on the basis of an important paper of his (with Greenwald) that says interventions can make everyone better off. He is a prolific, outspoken, and outstanding spokesman for the pro-government school. Stiglitz sees market imperfections that are remediable by government everywhere he looks; and this paper is supposed to provide the intellectual and analytical foundation for government intervention. Here I argue that this important and oft-cited paper completely fails to prove the potential worth of government interventions. FULL ARTICLE

{ 19 comments }

Don Lloyd September 6, 2006 at 9:29 am

FULL ARTICLE has no link

Regards, Don

Ulrich Hobelmann September 6, 2006 at 11:07 am

Excellent article.

Government cannot optimize between taxes and externalities, for one simple reason: I’m not a price elasticity curve; I’m an individual.

There’s NO WAY for any organization (lacking the ability to read my thoughts) to find out what I’d buy if my taxes were lowered or raised.

*I* am the one person making my own decisions; and indeed, I claim they are optimal from my perspective, or at least so good that I wouldn’t delegate these decisions to anybody else (or otherwise I would!).

The fact that government ignores this, that its agents claim the power to overrule the decisions of millions–no, billions–of people, shows that it’s by definition psychopathic, arrogant, antisocial and totalitarian.

Not least, it’s unconstitutional, because even though it’s self-evident that all men are created equal, some are given (by brute force and arms) special powers over the rest of us. How can anything in society be optimized by giving some people power over others?

Brad September 6, 2006 at 11:52 am

Economics discussed in Stiglitz’ manner is manifestly collectivist. There is an a priori acceptance of government and its role that doesn’t have to be explained. It is the merging of collectivist paradigms as to human action and Faith that government is the instrument that will best lead the collective. Ultimately it seems ad hoc introductions of elements to his argument is otherwise known as philosophic circularity in reasoning.

That seems to be at the core of all collectvist arguments that SEEM to be eminently logical and linear in its argument, but there is bound to be circularity somewhere, well hidden perhaps.

Roger M September 6, 2006 at 12:55 pm

Great analysis of Stiglitz’s failed logic. If you’d like more of Stiglitz’s nonsense, check out his book, Globalization and Its Discontents. He has a new one that just came out, Making Globalization Work that should be just as ridiculous.

One source of neoclassical nonsense is their fetus for efficiency and utility at the expense of all else. Surely it may be true that a poor person will get more “utility” from an extra dollar than a rich person. The poor person may also use it more “efficiently.” But the logical leap from those propositions to the neoclassical proposition that the government should take a dollar from a rich man and give it to a poor man is like Eval Kneval jumping the Grand Canyon.

Paul Marks September 6, 2006 at 1:50 pm

Even if were possible to compare one person’s utility with another person’s utility and it could be shown that a government intervention (i.e. the use of aggressive violence or the threat of it’s use) increased the untility of some people more that it reduced the utility of others – this would in no way “justify” the aggression.

No more than “scientifically proving” that a rapist (or rapists) got more utility from a rape than the victim lost from the rape, would “justify” rape.

That someone (or a group of people) gain more pleasure from an aggression (for example a government intervention) than the victim or victims suffer pain, does not justify the aggression – it remains wrong.

However, for those people who are not interested in morals (or people who want it “proved” to them that murder, rape and robbery are wrong), there remains another point to be considered.

Even if there was a theorectical case in which a government aggression increased utility (overall – making the futher assumption that such a calculation could be made), such an intervention should still not be allowed.

This is because it would set a precedent for other interventions.

Once the nonaggression principle (so called “laisser faire”) is abandoned, there is no clear limit to government interventions. And anyone who believes that government will make a disinterested scientific examination and only engage in aggression when to do so will increase overall utility, is deluded.

That aggression is wrong is the philosophical point against statism.

That, once such a power of intervention is accepted, there is no clear stop line for statism (and goverment will tax, spend and regulate as they wish), is the political case against statism.

But of course there is the economic case – Stiglitz is simply mistaken in holding that their are cases where statism is economically beneficial.

So it is a matter of P.P.E. – philosophy, politics and economics.

Paul Marks September 6, 2006 at 2:09 pm

Many people (including Stiglitz) argue that there is an important difference between the government using the threat of violence to get someone to do something (for example hand over money, or obey some command/regulation) and a person or persons doing so.

But, of course, “the state” is actually a group of human beings – politicians and administrators.

Nor is “society” an enity – society is the web of civil interactions between human beings, it can not will anything (it can not reason – it is not a thinking being), or give anyone anything (“society” owns nothing – not can it make a choice to give anyone anything, it is not an enity it is the web of civil interactions between enities).

Society is not an organization – but the state is.

But just because a group of people (politicians and others) organize themselves into a group gives them no right to use aggression against other people (otherwise churches, companies and even street gangs would have this right).

Nor does a “democratic vote” alter matters – 51% of a group of human beings do not have the right to murder, abuse, rob or order about 49% of a group of human beings simply because this group of people all happen to have been born (or have arrived in) the same area.

In any case (as stated above) the theoretical concept of a government that only uses aggressive force when “scientific experts” (as men like Stiglitz claim themselves to be)calculate it will improve overall utility has no connection to reality.

The theorectical concept is both false in theory (as no such calculated interventions are for the economic benefit) and absurd in practice (as it does not fit with how governments operate).

It is odd that “empirical” people tend to ignore the reality of how governments operate.

Som September 6, 2006 at 2:33 pm

Well there’s also the glaring problem that value is subjective, and the firm against the household (or “society” or whatever) will have different value assignations to the supposed externaility, but I don’t understand, even from a labor thoery of value point of view, how these supposed “externalities” can be added or subtracted by anyone, since they one cannot conclude the amount of labor (by one or more sources) to create that supposed externality. Certainly the government in the model can’t calculate that either, and would just impose it’s own values and costs onto these supposed externailities. So, as Padme would say, Stiglitz’ model “sounds alot like a dictatorship to me”.

eric September 6, 2006 at 3:52 pm

I’ve been programming computers for 40 years, and there is still no good way to validate programs except by testing. The program-proof fad seems to have been left entirely in academia, as it’s totally useless for any reasonablably sized program.

Therefore, even if the model was theoretically accurate, how on earth can the results be shown to be derived from a correctly written computer program (I assume this model is implemented on a computer).

To test a model, one must know before hand what the output should be based on certain inputs. Thus Stiglitz must have done some modeling by hand, or simply tweaked his model until it produced the results that he desired. Other modelers claim to test their models against past events (like the global warming folks) but in this case, where does Stiglitz get his output that he can compare against his model to verify his programming.

It’s simply a great way to make a living. Like being the court jester or wizzard.

Angelo September 6, 2006 at 4:10 pm

Great article. It is contemptible that people like Stiglitz think they are helping improve people’s lives by advocating government action.

Roger M September 6, 2006 at 4:18 pm

Eric, I think you’re giving Stiglitz too much credit. His models are just math on a chalk board, and as far as I know, modelers like him never even attempt to validate their models against the real world. Those modelers who dared, ended up sticking their head in an oven and turning on the gas. Any attempt at validation would prove most models of this type wrong. Some have proven been proven wrong and the creator ignores the results and continues using them because, to paraphrase Dan Rather, the models may be factually wrong, but conceptually correct.

Eric September 6, 2006 at 7:44 pm

Roger, I was wondering about that. Nowadays, when people discuss models, they usually mean computer models. And if they’re big government intellectuals, (of the kind that Rothbard would sneer at) they usually brag about using super-computers. I guess I’ve been listening to too much about modeling from the GW crowd.

KY Leong September 6, 2006 at 9:01 pm

Great article! Just love this in the conclusion:
“Stiglitz…has given us an inadequate and incomplete theory with inconsistent and ad hoc assumptions.”
That pretty much sums up the lot of the Anglo-Saxon economics tradition for the last couple hundred years.

Suresh Moktan September 7, 2006 at 12:01 am

Great article by Michael!
Yes, Paul, I salute your observation. You are right in concluding that ‘…”empirical” people […like Stiglitz] tend to ignore the reality of how governments operate.’
We must discern that any government is not a separate entity per se. Government constitutes “individuals”, and any self-centered and self-seeking “Tom, Dick and Harry” with vested interest may want to dictate and prescribe their selfish wishes on others, who they find weak. There is, indeed, no guarantee!

Fiacha McAsey September 7, 2006 at 8:36 am

It would appear that the author read Stiglitz’ paper with a bias equal but politically opposite to the bias found interpreted in the paper. Government has an obligation to attend sensibly to externalitites, some of which may not be terribly apparent. Those of use who seek to increase our own utility dolloar for dolloar, are not disposed to do so. Who will see that a STOP sign is errected or zebra crossing laid outside your child´s school. Not you, no the school but government. I understand the point of Stiglitz article and i understand the conccept of utility. But I don’t believe individuals or society would be entirely better off ignoring the role of government.

TGGP September 7, 2006 at 9:52 am

The phrase “All men are created equal” appears in the Declaration of Independece, not the Constitution. Of course, if you’re a follower of Jaffa the Declaration somehow got added to the Constitution somewhere.

Fiacha: There are plenty of signs that aren’t put up by the government. Schools and roads used to be private as well. You can try to make an externality argument, but that’s not what you’re doing right now.

Francisco Torres September 7, 2006 at 11:04 am

Government has an obligation to attend sensibly to externalitites, some of which may not be terribly apparent

Rozeff’s point is that if the homes and firms try to maximize their utility/profit, then what externality can exist that is NOT terribly apparent? The contention here is that Stiglitz, like you, simply assume externalities that require government, without indicating their nature. It is a classic case of Begging the Question.

Who will see that a STOP sign is erected or zebra crossing laid outside your child´s school. Not you, no the school but government.

Again, the point is that such cannot be infered directly from Stiglitz’s model, only that you assume it is so and move on. Assuming not ONE person will place a stop sign or a few lines in the street before a school is also begging the question – you just assume this to be true.

[...]I don’t believe individuals or society would be entirely better off ignoring the role of government.

Please explain why. Stiglitz’s model does not explain this – Stiglitz only assumes the role of government.

billwald September 8, 2006 at 12:21 pm

“In the model, the households and the firms optimize, yet somehow they are still not as well off as they can be.”

“Well off” means what? Is money in the bank the true measure of man? Are households that make charitable contributions “not as well off as they can be?”

Am I not as well off as I could be because I blow $3/week on my state lottery? (“A voluntary tax,” Geo. Washington.)

P.M.Lawrence October 23, 2008 at 10:41 pm

“If Stiglitz’s pro-state view is correct, it provides justification for the interventionist state. If Rothbard’s anti-state view is correct, then it removes the justification for all government activities, including those supposedly justified by externalities.”

Both statements are incorrect. On the one hand, if Stiglitz is right (as I think he is), that still does not supply justification, both because it does not address the moral question of justifying yet other aspects of the state, and because there may be yet other methods of addressing the same problems (which is my view). On the other hand, if Rothbard is correct, there might still be justification for the other aspects; that is, it is arguable for those with certain values, though I do not share that view or those values myself.

Now an overview of Rozeff’s questions:-

“Why don’t households and firms fully optimize, that is, take into account the externalities?”

This can happen when, e.g., they cannot get to grips with them. To see that this can happen, just consider that the state itself can cause them, and then – precisely because it exists – no-one else can get at the problems. For instance, unemployment benefits and the burden of state funding for them is like that. However, it goes deeper: in times and places where that doesn’t happen, there is still an externality, “Vagrancy Costs”, unless local conditions allow everybody to fend for themselves adequately or charitable institutions provide non-state support. State benefits came in to mitigate Vagrancy Costs.

“[W]hy is government able to recognize the externalities and compensate for them by a series of taxes, subsidies, and transfers while households and firms cannot and do not?”

But that is not the only possible obstacle – see above.

“What does the entity called government have to know in order to effect the taxes, subsidies, and transfers?”

Surprisingly little – see below.

Now Rozeff’s detailed objections:-

“A genuine theory would explain how the externalities arise even while households and firms optimize. It would introduce a theory of costs or some other factors that cause externalities. This is essential if it is to be argued that government can recognize and overcome these externalities… I do not say that there should never be ad hoc theorizing. There is a cost and benefit to it, and there will be times when it is worth doing. But in a case where the central message is to promote government and downplay markets, it just will not do to introduce the reason for government in an ad hoc manner.”

Well, no, it isn’t essential to have a theory, just helpful. All that is necessary is that there should be some externalities and that enough of their features should be recognisable that something useful can be done about them – even on an empirical, ad hoc, case by case basis, though of course the more understanding the better. I shall use the Vagrancy Costs/unemployment benefits issue to illustrate this. It appears that Stiglitz took it as so obvious that there were externalities that he simply proceeded to the next stage; I will provide an actual example (please bear in mind that even if I get it wrong, that only means a problem with the particular example unless the objections apply to a wider range of possible cases – and only apply to all externalities if the objections are that comprehensive).

“The Stiglitz model… assumes that actors who otherwise are absolutely punctilious about optimizing utility and profits ignore the potentialities of taking externalities into account”.

No – see above, about difficulties of getting to grips with things. I myself have advocated Professor Kim Swales‘s approach to addressing unemployment, but I am in no position to do anything beyond that about the problem in that or any other way.

“Households and firms ignore the externalities, but the optimizing government does not. Why is government so perspicacious?”

Again, the problem is not solely one of understanding.

“Stiglitz never describes what this government is or how it comes about. He never even describes its basic properties. Government in the model is basically ad hoc. It is motivated to produce taxes, subsidies and transfers to achieve optimality in a manner that is nowhere described. It simply does it. There is no warrant for making this assumption and there is no warrant for taking it to be characteristic of actual government.”

This is a misunderstanding of the kind of result Stiglitz is offering. He is not assuming that actual government is like that; he is simply pointing out that actual government has the scope to do those things. Whether it does or not (usually not, in my view) is an entirely separate sort of question. For those who believe that it can be influenced or reformed to do these things or keep doing them if it already is, it is a challenge of a civics sort, to make the government do what it should – to work within the system. To those like me who think that it is physically but not morally capable of doing the right sort of thing short of duress, it is yet another piece of evidence that it is deeply flawed and must be driven even while it is being eroded. That governments can but usually do not do the little good that lies within their reach is a criticism of them, not of Stiglitz.

“What does government have to know in the model in order to accomplish its task? The list is staggering [list omitted - refer to article]. If we again stay within the confines of the model itself, there is nothing in it that tells us how the actors who govern are supposed to acquire this information or what the acquisition costs are.”

But that is demanding too much. All that is actually necessary for this sort of action to be constructive is, that it be an improvement – not that it be optimal. This list is of what it would take to be optimal. Taking the Vagrancy Costs case as an illustration, buying off the problem with unemployment benefits – like the Elizabethan Poor Law or Bismarck’s policies in the area – was an improvement. It is also clearly sub-optimal, from the funding burden and the incentives for unemployment, but it is an improvement. Equally clearly, Professor Kim Swales‘s approach is an improvement on unemployment benefits, because we can easily see that tax breaks that matched unemployment benefits would have zero net cost to the economy and would raise productive employment as well. We can see that even without looking into his modelling, and even without knowing that tax breaks at that level would actually be optimal; it is sufficient to know that some levels exist (short of being in a Malthusian crisis) that would be an improvement over what is in place now. As it happens, I believe that there are yet better ways of handling the matter, in the long run, but that this would be a good first step in a transition both because it matches where we are coming from (more in some countries like Australia, where I am) and because it is faster acting, which makes it fit well as part of a transition to those long run methods.

“The real economic questions concerning externalities are why they arise, how they relate to property rights, what to do about them, or when it pays for interested parties to do something about them. In a sense, they are no different than any other kind of inefficiency. Therefore, the presumption is that free exchanges can address the perceived costs and benefits associated with them whereas government is ill-suited to the task.”

To stop there is the same sort of error as the criticism that Rozeff supposes Stiglitz has made, of feeding in assumptions. We should of course make that presumption, pending more information. But as, when and if candidates for externalities that are not like that come up – and I have given an example, in Vagrancy Costs/unemployment benefits – then we can and should substitute the actual facts of the case for presumptions, even if we only have some knowledge of them and even if we got there in an ad hoc way. Anything else is “my mind is made up, do not confuse me with the facts”.

Michael Rozeff October 24, 2008 at 6:52 am

Thank you, Mr. Lawrence, for your extensive comments.

Your first error is the “head-on-a-pin” fallacy. This occurs when an analysis is criticized for omitting features that are not within its chosen realm. All analyses are partial, so all analyses can be criticized for not placing all thousand angels on the pin’s head. Neither Stiglitz, Rothbard, nor I were considering moral questions within the discussion of externalities. They are important and we discuss them elsewhere, but this particular discussion is not about them. The fallacy Mr. Lawrence engages in at the outset is to drop the context of the discussion. Note his reference to our omitting the “moral” perspective. My statement stands: “If Stiglitz’s pro-state view is correct, it provides justification for the interventionist state.” Rothbard’s work provides justification for no interventions. These are not on moral grounds, but on the economic grounds of the entire discussion. Mr. Lawrence implicitly agrees, for he goes on to mention a number of economic arguments.

I asked: “Why don’t households and firms fully optimize, that is, take into account the externalities?”

The context of that question is within a competitive and free market model in which there is no government. It is Stiglitz who introduces government in an ad hoc manner. I do not introduce it at all. The question is, sans government, why don’t household and firms optimize, for Stiglitz assumes that they do not.

Lawrence again drops the context and misapprehends the analysis when he asserts that this can happen when he replies: “This can happen when, e.g., they cannot get to grips with them. To see that this can happen, just consider that the state itself can cause them, and then – precisely because it exists – no-one else can get at the problems.”

If I had wanted to analyze the optimal behavior of households and firms in the face of violent conflict with third parties like government, I would have. I didn’t. Lawrence, like Stiglitz, is introducing the State into this question. This begs the question, which is how does he or Stiglitz justify that households and firms need the State in the first place to take care of externalities when they are free to take care of these externalities themselves?

To my next question: “[W]hy is government able to recognize the externalities and compensate for them by a series of taxes, subsidies, and transfers while households and firms cannot and do not?”, Lawrence has no answer except his very first one, which is that they face the obstacle of government. Again, that assumption violates the context and assumptions of my analysis. Stiglitz assumes that such externalities exist and that households and firms, who otherwise optimize, can’t recognize them and/or can’t deal with them. Lawrence’s answer, which is that the State prevents it, violates the entire premise and context of my analysis, which is the ability to optimize freely. His answer says that the State prevents that optimization. (Moreover, it is incongruous that the State first prevents optimization and then is supposed to help households and firms attain it.)

I argued: “…in a case where the central message is to promote government and downplay markets, it just will not do to introduce the reason for government in an ad hoc manner.” Stiglitz did this. Lawrence defends Stiglitz by saying: “Well, no, it isn’t essential to have a theory, just helpful. All that is necessary is that there should be some externalities and…that something useful can be done about them – even on an empirical, ad hoc, case by case basis.”

Lawrence again drops the context of the discussion and introduces a straw man here. Stiglitz provided abstract theory. His message was a theoretical message in toto about justifying government economically. In that context, Stiglitz cannot get away with introducing government in an ad hoc fashion.

To my next observation, for the third time Lawrence provides the same wrong answer, which is that the State gets in the way of households and firms. My observation was: “The Stiglitz model… assumes that actors who otherwise are absolutely punctilious about optimizing utility and profits ignore the potentialities of taking externalities into account”.

Lawrence’s reply: “No – see above, about difficulties of getting to grips with things.”

We are analyzing Stiglitz’ justification for government in the first place. We can’t assume that government already exists without begging the question. Lawrence has joined up with Stiglitz in question-begging.

In his next criticism, Lawrence yet again drops the context entirely. The interchange goes like this:

Rozeff: “What does government have to know in the model in order to accomplish its task? The list is staggering…If we again stay within the confines of the model itself, there is nothing in it that tells us how the actors who govern are supposed to acquire this information or what the acquisition costs are.”

Lawrence: “But that is demanding too much. All that is actually necessary for this sort of action to be constructive is, that it be an improvement – not that it be optimal.”

The context could not be more clear, as I said “in the model”. Stiglitz’ whole contribution is supposed to be a demonstration within a rigorous economic model of the good that government can do. Lawrence has no grounds for jumping outside the model to let Stiglitz off the hook and say “Well, we really do not need this model. Stiglitz really is not trying to show optimality. If an action of government is constructive, that’s enough.” Give me a break! Stiglitz’ whole point is to show that government works to improve welfare within processes of optimization. Take that away, as Lawrence does, and he throws out Stiglitz altogether!

Lawrence then abandons Stiglitz altogether in order to provide an instance where he thinks government works to reduce an externality, namely vagrants. Why bother? Stiglitz has examples in his paper where he thinks liberty fails and government succeeds. Why beg the question, which is to demonstrate theoretically that government can do better than liberty? Lawrence thinks that poor laws and unemployment benefits work well, on balance. He writes: “It is also clearly sub-optimal, from the funding burden and the incentives for unemployment, but it is an improvement.” That’s his opinion, but how does he know? How does he know that matters are not better handled in liberty? How does he know how these tradeoffs are valued? He cannot know without having access to household and firm information that neither he nor Stiglitz possess or can obtain, even in theory.

Next Lawrence supposes that I am in error to point out that “In a sense, they [externalities] are no different than any other kind of inefficiency. Therefore, the presumption is that free exchanges can address the perceived costs and benefits associated with them whereas government is ill-suited to the task.” I am said to be “feeding in assumptions.” We are now outside Stiglitz’ model, it should be noted. There is an inefficiency. Who perceives it? Who perceives its costs and benefits? Who perceives the costs and benefits of organizing or not organizing to deal with the inefficiency? Costs and benefits are aggregated and submerged within government as in any freely-made organization, but because of the powers of government and the poor control mechanisms that people have over government, the presumption is that other ways of handling problems take priority. If we do not know this both from the theory of the state as an organization and from the empirical experience that Lawrence alludes to, then we know very little.

Lastly, Lawrence mentions a proposal of Professor Swales to improve on unemployment benefits with a tax break. Assume that the proposal is sound and that it reduces the losses produced by government. Stiglitz will have no objection. I will not object. Rothbard would not, I believe, have objected. A change at the margin that reduces government in some way and improves welfare is welcome. This, however, does not justify government in the first place, and it does not prove that government has what it takes to produce optimal results that households and firms could not themselves have produced by other means including other organizations that maintained the liberty of the participants.

Comments on this entry are closed.

Previous post:

Next post: