May 2019

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31  
Blog powered by Typepad

« Unemployment: What and Why | Main | New Blog to Bookmark --- The Sociological Imagination »


Feed You can follow this conversation by subscribing to the comment feed for this post.

According to Jeffrey Friedman, banks did not "willingly, knowingly, and repeatedly [take] huge amounts of risk, hoping they'd get bailed out by everyone else," as Jeff Miron claims, and I am inclined to agree with Jeffrey.

I was quite impressed with Friedman's opinion that ignorance--and not greed--better explained why so many bankers took such large risks.

It seems to me that ratings agencies misled bankers into believing that MBSs were safe; capital requirement regulations, along with the activities of Fannie Mae and Freddie Mac, made MBSs more attractive to hold; and the Federal Reserve's easy money policy kept the financial insensitive to the problems brewing within.

Eventually, when the Federal Reserve stopped pumping the economy full of anesthetic, the reality of these products was revealed, and everyone felt a very sharp and painful pinch.

Although the "Greenspan Put" was no doubt harmful, I think it was possibly the most minor villain in whole play.

To any who missed it (for shame), Jeff Miron gave a talk on the subject posted here:

For my money, there is no one writing better commentary on this subject than Jeff Miron, without question.

And his latest academic work, on the mortality effects of the minimum drinking age (Economic Inquiry), continues his long tradition of good empirical papers that make the drug prohibitionists look exceedingly silly.


I my previous comment, I said that I agree with Jeffrey, but both Friedman and Miron are called Jeffrey. It should be obvious enough who I mean, but just in case: I meant Jeffrey Friedman.

And the later sentence is supposed to read: "and the Federal Reserve's easy money policy kept the financial *system* insensitive to the problems brewing within."


I think you should look at the paper by Jablecki and Machaj "The Regulated Meltdown" --- especially the section "The Basil Rules with Players Too Big to Fail." I agree that Friedman's introduction is excellent, but I do think part of the language issues with respect to motivation/incentives cause Jeff to miss some of the main economic points. However, let me state clearly that in my opinion monocausal explanations of the crisis are flawed as it was a perfect storm of government perversity and distortion.



Unfortunately, I do not have access to Jablecki and Machaj's paper.

Could you briefly explain what you mean by, "I do think part of the language issues with respect to motivation/incentives cause Jeff to miss some of the main economic points"?

In any case, I used to be quite sold on the idea that the "Greenspan Put" was big cause of the financial crisis, but after reading Friedman's piece, and considering some other comments I have read, it seems to me a less and less important part of the puzzle. I think Jeff Miron may be attributing too much blame to the "Greenspan Put", and, in the process, he may be allowing other more guilty causes to avoid suspicion.

Miron says:

"When people try to pin the blame for the financial crisis on the introduction of derivatives, or the increase in securitization, or the failure of ratings agencies, it's important to remember that the magnitude of both boom and bust was increased exponentially because of the notion in the back of everyone's mind that if things went badly, the government would bail us out."

Friedman says:

"They, like the other investors, and the rating agencies, were simply mistaken—or so it seems, with the luxury of hindsight. And the reason,
apparently, is that they were ignorant of the true risk of the securities they bought."

Are these two views compatible? I don't think so, or at least not without notable modification on either side.

Although Miron's stress on the Greenspan Put fits well within economic logic, I suspect few people in the financial industry knew much about it or gave it much thought. Moreover, Friedman provides evidence that many high ranking investors and bankers were also buying risky securities; even with Geenspan's implied guaruntee, it doesn't make sense unless they too were mistaken about the risks. How were bankers to know that the failure of their own investments would coincide with a financial crisis?

Of course, I haven't studied these matters in nearly the detail of Miron or Friedman, but that's why I am writing this in a blog and not an academic journal.

In my view, that is precisely the (relatively) weakest part of the otherwise excellent Friedman´s paper. He draws too sharp distinction between incentives / knowledge problems in banks. You´ve got a third category there: bank employees could have had weak incentives to discover / obtain the proper knowledge about the real risks in the first place.

Matěj Šuster,

"bank employees could have had weak incentives to discover / obtain the proper knowledge about the real risks in the first place."

But is that not exactly what Friedman argued was the case with ratings agencies?

Federal protection from competition--partly by giving the ratings of some agenices legal weight--established a oligarchy in the ratings industry. Even if these agencies acted as thought competition were a pressing concern, it was impossible for it to materialise. In consequence, at least one agency did not update its relevent models even as the housing bubble was in full swing. Incentive failure led to knowledge failure, and mistakes followed.

Is it possible something like this happened in the banks, too? I think so, especially with all the "regulatory" interference and vested interests in the financial industry. But I am inclined, after reading Friedman's paper, to put less stock in that as an explanation of current events--too risky.


I would argue that Matej has hit the nail on the head. Jeff's piece is outstanding as I said, but it fails to consider this third path of explanation --- which as Matej points out, would incorporate the incentive structure point and the ignorance point (alla Kirzner). And in your third paragraph above you capture the point as well. These issues are linked, rather than separated.



I also think we are being a little loose with the idea of the Greenspan put. That is basically the idea that the Fed policy under Greenspan was that any economic troubles would be met with credit expansion to keep the system liquid. This is best captured by John Taylor's paper in the CR collection, or in his book Getting Off Track.

They are related, but slightly different than the moral hazard issues associated with bailout, or insurance, etc. Combine an implicit promise to provide the financial means with an explicit (or implicit) promise to in fact bail you out if things go bad, and your betting practices will change. Especially if you think you have discovered instruments which have already diversified your risk portfolio.

The bottom-line in all of this is that in order for this crisis to turn into a crisis a "perfect storm" of government perversity and distortions had to be in place. Businessmen didn't just wake up one day and were incredibly stupid. They didn't just check their rational self-interest at the door for half a decade. No, each step along the way they were engaging in the best choice possible for them given the incentives and information that they face at that time. This is not a psychological crisis, it is an economic crisis caused by economic policies and economic structures that emerged because of those policies.

Fiscal and monetary policy (especially monetary policy) fueled the process, but the particular manifestation that crisis took was a function of particular distortions to the information and perverted incentives. In other words, the reason why the crisis took shape in the housing boom/bust was a function of a set of policies that directed the credit expansion in that direction, and that financial practices to micro-manage risk evolved due to structures in the industry that enabled mortgages to be treated in ways they weren't in the past, and the rating agencies faced incentive structures and information acquisition and utilization issues, etc., etc.

How we unwind all of this is different set of questions.

But one of the things we need to get straight is the multi-causal factors in the perfect storm. The Austrian theory of the trade cycle is a component of the story, but not the only component. Just as it wasn't the only component in explaining the origin, length and severity of the Great Depression --- credit expansion, trade restrictions, tax hikes, and regime uncertainty to name a few all work to explain why the system got derailed. Same is going on now --- and lets remember just how important regime uncertainty is to this story of the failure of getting business back on track.

Thus, despite Matej's objections, that is the reason why the piece Ordinary Economics for Extraordinary Times serves the purpose for which it was written, just as my piece Whatever Happened to Efficient Markets attempts to serve the purpose for which it was written. Not every article one writes can answer all questions, but instead tries to tackle usually one issue at a time. But when you add it all together there is more than one issue involved in a true explanation.

Gentlemen--an incentive structure does not exist apart from the perceptions of those it is supposed to influence. If it does not manifest itself in motives that lead to actions, it is like a tree falling in a forest--it may as well be a fiction.

It is possible that the doctrine of "too big to fail" actually motivated bankers to take what they *knew* to be undue risks, but as Lee points out, they often took these risks with their own money and had no reason to think they would be bailed out personally. Moreover, the types of bailouts we have seen are unprecedented and surely could not have been likely in the eyes of the bankers.

But what motivated bankers is an empirical question, and as yet there is no evidence that they were motivated by the idea that they would be bailed out. NO EVIDENCE. Economists who favor this theory might actually want to investigate it, not only through archival and interview research but simply by comparing the purchases of mortgage-backed securities by large (too big to fail) and small (not too big to fail) banks, yet nobody has bothered to do this: it's so much easier to assert that "incentives" that *in retrospect* the economist knows about (omnisciently) *must have* motivated the parties in question. (This apriorism is the biggest target of the Colander et al. paper in Critical Review's special issue on the crisis.)

Likewise the other "incentives structure" story that economists like to tell about the crisis: that the structure of executive compensation made bank employees take undue risks. It's possible, but there zero evidence for it, and considerable evidence against it. I present some of this evidence in my paper, and having now finished Michael Lewis's article on AIG in the August Vanity Fair (I know, I know!), I'd urge interested parties to read that, too. Lewis's point is that even the executive responsible for *insuring* these assets had no idea what they consisted of (and when he found out, he stopped insuring them)--even though he had every incentive to know.

Anyway, I leave you with this thought: We have two "incentives stories," the too-big-to-fail story and the executive-compensation story. The first is favorable to free marketeers, the second to interventionists (e.g., Posner, Obama). But neither has a shred of evidence in support of it. Eventually the evidence will come in--maybe someone on this site will actually do the research. But in the meantime, what, other than political ideology, would make us pick the too-big-to-fail story over the executive-compensation story?

Jeffrey Friedman
Editor, Critical Review

Matej--I agree that that is possible, but it could be true only of already-known facts, of the sort that you can look up in a book. Otherwise, how would a person with every conceivable incentive go about discovering "proper knowledge"? Everyone thinks that his opinions are true. And when we're discussing risks of an unpredictable future, we are talking about opinions--not "information" for which a sufficiently incentivized agent needs merely to "search" (as in a book).

A nice aspect of the Lewis article that's consistent with everything else I've read about the crisis is that he tells about how the AIG people actually met with the investment bankers and heard over and over again that the securities were safe because housing prices had never gone down in a big way across the whole United States since the Great Depression. This is what they considered "proper knowledge" of the risks. They thought they already had possession of the truth. They were wrong.

But looking back over human history, when has having a high incentive to be right ever produced wisdom?



Look as I have said to your repeatedly, your article does a great service. But in your response to Matej you miss the essential Kirznerian point about sheer ignorance. Discovery is not search --- that is what you try to equate Matej point to. No, Kirzner's point (and Matej I would argue) is precisely linked to your point about ignorance. The incentive structure alerts us to that which is in our interest to be alert to --- if we are not so incentivized we will not be alert, i.e., we will remain ignorant of the relevant information.

And as I have said to you on email and on this post, the story about incentives and the story about ignorance can be linked without much damage, the only thing that is lost is your understanding of what the economists are arguing. BTW, this is why i think the Colander et al piece way overstates the claims about the indictment of modern economics as well. They have a notion of rationality which overstates what economists actually believe and model. In this regard, I do think it significant that your collection does not have space for the voices not of White, Selgin, Garrison or Horwitz, but Cochrane, Zingales, and Mulligan. I think if you look at their work, especially Zingales, you will see details in facts, and logical argument, that is impressive and it is from a more or less standard economic approach. Nothing exotic required.

Thus, the title of my essay for the Kates book on the crisis, Ordinary Economics for Extraordinary Crisis. Back to first principles. This is why Jeff Miron has been very good to read throughout this debate as well. Simple economics need not be simple-minded.

But in those first principles is the question of coping with our ignorance. That is why we have a price system. And when the policy-regime distorts information, and provides perverted incentives, our ignorance is not eradicated, but instead compounded. You have made that argument very well. Congratulations. But it is an argument completely consistent with an economists story of incentives, and the problem with the Posner slant is that story is not told to the end and thus misses the critical factor --- just as all greed (or psychological) based stories do.

Your own issue has important contributions to understanding the consequences of perverse incentives and distorted information --- Taylor, but also Smith, and of course the piece by Jablecki and Machaj. The crisis is a consequence of a perfect storm of policy failures manifested in massive errors that require correction.

I would just add that "too big to fail" is not the only incentive issue that free marketeers have focused on. The very existence of artificially low interest rates distorts the incentive structure in ways that matter. The politicization of the mortgage market changed incentives in ways that matter.

And before Jeff tells me this is all apriori speculation, he's welcome to have a conversation with my former college roommate who has been in the mortgage business for 20 years and who worked at Fannie, Countrywide, and IndyMac, and who also helped develop some of the zero-down instruments.

Jeffrey Friedman:

Those banks certainly seemed to be plagued by a tremendous amount of very short- term thinking on the part both of their CEOs and lower- level employees. I find the extent of their ignorance about the risks of those MBS and CDOs incredible. How do you explain this fact? Is it really just a hindsight bias on my part? This from Lewis´ article on AIG is truly shocking:

“That’s when Park decided to examine more closely the loans that A.I.G. F.P. had insured. He suspected Joe Cassano didn’t understand what he had done, but even so Park was shocked by the magnitude of the misunderstanding: these piles of consumer loans were now 95 percent U.S. subprime mortgages. Park then conducted a little survey, asking the people around A.I.G. F.P. most directly involved in insuring them how much subprime was in them. He asked Gary Gorton, a Yale professor who had helped build the model Cassano used to price the credit-default swaps. Gorton guessed that the piles were no more than 10 percent subprime. He asked a risk analyst in London, who guessed 20 percent. He asked Al Frost, who had no clue, but then, his job was to sell, not to trade. “None of them knew,” says one trader.

That is not just an “ordinary”, excusable ignorance. The fact that they did not even know what exactly they were insuring screams for a further explanation.

As for empirical evidence, there is some evidence that employees of rating agencies did know that their “models” were inadequate and unsound in many respects. [“Let's hope we are all wealthy and retired by the time this house of cards falters,'“ one e-mail from an S&P employee said. ].

Also, from Michael Lewis´ Vanity Fair article on AIG:
“Two months after their meeting with the investment bank, one of the A.I.G. F.P. traders bumped into the Goldman guy who had defended the bonds, who said, Between you and me, you’re right. These things are going to blow up.)

Nevertheless, I completely agree with you in one respect: more empirical research of this particular issue is definitely needed.

Pete: Kirzner and I debated this question in Critical Review vol. 18, no. 4. I stand by what I said there: It is magical thinking to believe that profit opportunities are somehow automatically perceived by entrepreneurs. But without that perception, there is no incentive. So Kirzner's entrepreneur brings in through the back door the omniscience theory Kirzner tried to throw out the window.

Calling omniscience "alertness" doesn't help one bit.

I am not qualified to have an opinion about Miron's work, and I allowed that the too-big-too-fail story, which (like the executive compensation story) has advocates in some of the articles in our special issue, might be true. But if anyone has done research showing that it *is* true, I'm ignorant of it, so please point it out. I mean research that provides empirical *evidence,* not a priori economic theorizing about incentive structures that *might* have been at work.

Posner's story is definitely not psychological, nor is mine. In fact, he brilliantly refutes the psychological story ("irrational exuberance").

I agree with you, Pete--I never said that a story of ignorance is incompatible with a story of incentives, and the story I tell in my introduction is, in fact, a story of both. The Basel rules changed the incentives facing bankers. HOWEVER, different bankers responded to these incentives in different ways--depending on their different perceptions of how risky holding the Basel-privileged AAA or AA securities would be. Citicorp thought there was little risk in holding them, so it bought many; J.P. Morgan Chase thought there was a lot of risk, so it bought few. In retrospect, then, we can say that Citi was ignorant of the risk, and that this radically affected its behavior--even though Citi and Morgan faced the same "incentive structure," from an omniscient perspective.

My complaint about incentives stories is not that they are necessarily wrong, but (1) that they are usually under-researched, because they are easily modeled and are congruent with the economists' bias in favor of seeing self-interest as the only thing that really matters, empirically, in human affairs; and (2) they tend to leave out cognitive considerations.

Cognitive considerations are *not* disposed of by invoking the price system, because entrepreneurs have different theories about what a given array of prices signifies about profit and loss opportunities. That is why *competition* is necessary--to test these theories against each other. I think Lavoie came close to saying this in *Rivalry and Central Planning,* which is the book that led me to my views. Alchian says it directly in "Uncertainty, Evolution, and Economic Theory." But they were rarities.

A problem with contemporary economics is that since its focus tends to be "incentives matter," even Austrians, in practice, tend to go for theories that homogenize all market participants' behavior when the participants all face the same "incentive structure." But the reason competition is useful is that it pits entrepreneurs' (a) heterogeneous theories against each other, even though they all generally face (b) a homogeneous incentive structure. To be a professional economist is, as a rule of thumb, to see the entire world as consisting of (b). No wonder there is so little real appreciation of the imperfect but important function of capitalist competition.

Yes, I am saying that the lessons of the socialist calculation debate still haven't fully been plumbed.

Let me close by pointing out that I never took a single economics course. I've read Hazlitt, some Mises, a lot of Hayek--that's about it. This means that I don't know the current economics literature. Yet I somehow managed to produce what people seem to find a compelling account of the crisis without the "benefit" of a graduate degree in economics--which in my experience tends simply to indoctrinate students with the tacit assumption that incentives are *all* that matters. So, young people reading these words: You've already read Hazlitt. If you don't want your mind distorted to the point where *all* you can think about is incentives, consider graduate study in a field other than Economics.



Look we are debating over an issue of economic thinking and the culture of economists, when in fact your contribution is one of application. The world that you describe --- homogenous incentives exclusively --- is not a world that research economists (as opposed to textbok economics) would recognize.

How did Kirzner respond to your summary dismissal of his position and the claim that omniscience is brought in through the backdoor? I am guessing he was "confused" not agitated. Talking past each other, not engaging.

But lets move beyond this issue of semantics, and focus on the implications of your claim that the lessons from the socialist calculation debate haven't been fully plumbed --- I agree. That is what I argued (with Pete) in the pages of CR as well as elsewhere (including a 9 volume reference work on the debate).

Finally, for the political scientists etc. that Jeff is recruiting to work on public ignorance and cognitive questions. FANTASTIC. Please do keep in mind that along with Timur Kuran I am the editor of a Cambridge University Press series entitled Studies in Cognition, Economy and Society, and we would be more than thrilled to consider for publication quality work along these lines for possible publication.


I agree that the financial crises--and the following recession--were the consequence of a "perfect storm" of disregulation by governments: primarily the expansionary monetary policy of central banks. I suspect we disagree very little, if at all, on these matters, and my relatively shallow understanding is probably what separates us most.

My previous comment was concerned with disentangling the multi-causal factors, and identifying a point of contention among thinkers who are otherwise in agreement. (And I think it led to very interesting discussion.)

Please understand, I did not mean to suggest that one day businessmen woke up "incredibly stupid"--that is a gross distortion. I agree that, "each step along the way [businessmen] were engaging in the best choice possible for them given the incentives and information that they face at that time."

I think Friedman has a point, even if it is inapplicable to most of the contributors of this blog. Some economists may need to be reminded that just because some fact would have, in hindsight, provided a good incentive for some action is irrelevant unless the people acting *actually* took it into consideration. I am inclined to say that the Greenspan put was such an unconsidered fact; the implicit guarantee of Fannie and Freddie by the Federal Government seems to have been more important.

Was the mistaken knowledge of bankers due to a prior incentive failure? Perhaps. Considering the degree of disregulation in the financial industry, I am pretty sure things were quite screwed up already, but I am also inclined toward the view that banks were less to blame than the ratings agencies on who they were forced to rely.

I want to stress that while incentive failure can lead to a knowledge failure, a knowledge failure does not necessarily mean there was an incentive failure. With regard to Matěj's comments about ignorance within banks themselves, it would be interesting to read a study comparing the knowledge of banks who held risky assets against those who did not, or the same bank from different periods to see if their knowledge of the products they held deteriorated.

In any case, apparently, many bankers personally invested in the same kind of risky products that were on the banks' balance sheets; it pushes credulity to suggest they were counting on the Greenspan put, or a Federal bailout, if their personal investments turned sour, because how were they to know that would coincide with a financial crises? And how were they to predict the unprecedented Federal bailouts?

This may be of interest to you:

Regulatory Use of Credit Ratings: How it Impacts the Behavior of Market Constituents
(Viktoria Baklanova, CFA, is senior director at the Fund and Asset Manager Rating Group of Fitch Ratings)

From the conclusion:

"To conclude, we would like to note a growing dependency of all market constituents on the CRA ratings as a common measure of creditworthiness, especially in the world of less transparent structured credit securities. The behavior of market constituents, including investors, issuers, and regulated entities has been affected by such dependence. The SEC proposal came about to address the perceived failure of the CRA to accurately indicate riskiness of structured credit securities. Still, the feedback to the SEC proposals to eliminate references to NRSROs’ ratings in its rule indicates that the market participants are not ready to accept responsibilities for an independent credit risk assessment. We infer that investors, fiduciaries, and regulated entities are looking to regulators to offer a common measure of risk, accurate and free of conflict of interests. At the very minimum, the market participants expect the SEC to assume a more important role in controlling the integrity of the credit rating process."


A very depressing read. :-( Leaves one wondering how many genuine entrepreneurs / investors are actually there ... Those "market participants" sounds more like people dependent on welfare ("welfare bums"). :)

The comments to this entry are closed.

Our Books