Like Mr. Clean With Hair

by von

I share Hilzoy's concerns regarding the so-called "Geithner Plan"* (1, 2).  The Obama Administration's plan to save certain banks includes a host of incentives for bad behavior, waste, and general tomfoolery.  I don't know whether these problems can be fixed or whether the plan should be scrapped; however, it's not a promising start for the Administration.

The current stupidity also underscores what should be the broad, guiding concept in all our righteous interventions.  No matter how bad the disease, the cure can always be worse. (Hence the adage:  "First, do no harm.")  The cure currently on offer involves a great deal of faith in regulators.  This faith isn't well-placed.  The specific example of the AIG bonuses -- approved by the Bush Administration (Paulson), the Obama Administration (Geithner), and a key member of the Senate (Sen. Dodd) -- itself should cause one pause.  

Scott Sumner, a Bentley University economics professor, offers a more systematic criticism.  The efficient-market hypothesis ("EMH") basically asserts "that financial markets are 'informationally efficient', or that prices on traded assets, e.g., stocks, bonds, or property, already reflect all known information. The efficient-market hypothesis states that it is impossible to consistently outperform the market by using any information that the market already knows, except through luck."  A faith that more regulation will improve our lot is a statement that the markets have failed -- that the EMH is situationally false.**  Sumner explains problems with such faith:

So the anti-EMH argument for regulation must be based on the following; bankers are irrational and make lots of foolish loans.  Regulators are rational and can see that these loans are too risky, and can protect bankers from hurting themselves.  At a theoretical level this doesn’t even pass the laugh test.  But what happened in practice?  What position did the “regulators” take in this crisis?  First we need to define “regulators,” who are much more than just the low-paid Federal bureaucrats that oversee the banking industry.  Regulators are the watchmen, those who watch the watchmen, and those who watch those who watch the watchmen.  In other words:

1.  The President
2.  Congress
3.  The Fed
4.  The media
5.  Most academics
6.  Nouriel Roubini

Guess how many of these institutions warned us about the sub-prime crisis.  Now guess how many were encouraging banks to behave even more recklessly than they did.  Unless we plan on making Roubini dictator of the world, there is zero evidence from the sub-prime crisis that simply giving regulators more power would have helped.  And how do we know that even Roubini wasn’t just lucky, and might miss the next fiasco?

This is a good time to trot out my favorite philosopher, Richard Rorty.  In a recent book he quoted an old pragmatist maxim; “that which has no practical implications, has no philosophical implications.”  I would re-word that slightly for the current discussion:

That which as no practical implications; has no implications for economic theory.

....

(To forestall some objections based on a misunderstanding of my argument, let me reiterate that I am not saying the sub-prime crisis does not point to the need for tighter regulation, I am saying that any regulation that comes out of this crisis would have to be justified on grounds other than market inefficiency, unless you can convince me that future regulators will be able to predict markets better than future financiers.)

I, too, think that regulatory changes need to be made.  Some will have to be significant.  But a call for more watchmen and women -- without some explanation as to what harm they're trying to correct and why they might correct it -- ain't prudent.  Demanding that we just do something, that someone somewhere pay a price, is panic, even when wrapped up in justified anger. 

*Isn't it, truly, Obama's plan?  Where does the buck stop?

**This criticism is somewhat related to the standard, behavioral economics criticism of EMH, but it's not the same thing.  (Full disclosure:  My limited background in economics was focused on behavioral economics, and my sympathies lie in that direction.  ) 

(Source for title:  Weezy.)

Comments

[TJ]

You either think that banks are not lending because the economy sucks, or that the economy sucks because banks are not lending. If the former, like I do, then Geithner's plan is worse than useless.

Unless we're doing it to avoid armageddon. Then Obama better explain it that way, for future reference.

[Johnny Canuck]

"The specific example of the AIG bonuses -- approved by the Bush Administration (Paulson), the Obama Administration (Geithner), and a key member of the Senate (Sen. Dodd) -- itself should cause one pause."

Do you have evidence for the 3 parts of your accusation? With respect to Dodd, he prepared an amendment, someone at Treasury reviewed it, told him attacking bonuses (generically) retroactively might not withstand legal challenge, Dodd acquiesced. According to Dodd no one flagged AIG. Do you know something more? Just paranoid? Believe Dodd knew and thought he could get away with it? Similarly for Geithner, unless you are counting his feeble protests to AIG to as functional equivalent to approval, what evidence is there he knew and approved? I wasn't following the story back in the days of Paulsen? Did he actually make a conscious decisiion about the AIG bonuses?, or just generically, govt should not regulate compensation?

[Scott de B.]

You seem to define 'regulation' as somebody in charge giving orders on a case-by-case basis. That makes little sense. If we simply had stronger reserve requirements, or prevented rating agencies from lying in bed with the agencies they regulated, or limited what kind of financial activities banks could engage in, the crisis could have been averted or mitigated.

"A faith that more regulation will improve our lot is a statement that the markets have failed -- that the EMH is situationally false.** "

Well, no. If we choose to regulate the market in labor by outlawing slavery, that doesn't amount to a statement that the market was mispricing the value of slaves.

[Yarrow]

A faith that more regulation will improve our lot is a statement that the markets have failed -- that the EMH is situationally false.

Well, yes. That's pretty obvious, isn't it? I'll let John">http://crookedtimber.org/2009/01/03/refuted-economic-doctrines-1-the-efficient-markets-hypothesis/">John Quiggan do the heavy lifting:

Even the strongest advocates of the EMH would not seek to apply it to, say, the Albanian financial sector in the 1990s, which was little more than a series of Ponzi schemes. They would however want to argue that the massively sophisticated global financial markets of today, with the multiple safeguards of domestic and international financial regulation, private sector ratings agencies and the teams of analysts employed by Wall Street investment banks is not susceptible to such systemic problems, and is capable of correcting them quickly as they arise, without any need for large-scale and intrusive government intervention. I’ll leave it to readers to make their own judgements

[TheWesson]

The EMH *cannot work* in a bubble or a panic - where is the "information" which drives a bubble market?

The price there is what the market expects the expectation of the market to be - in my opinion hopelessly self referential. In this case either a positive or a negative or a neutral expectation on future prices is a valid expectation!

What does the EMH do in the presence of imaginary information? Generates imaginary prices, I suppose.

So you can't outperform a bubble/panic market (never know when bubble/panic state will end) but nonetheless you can assert a bubble or panic currently exists relative to fundamentals (e.g. price/rent ratio in the case of housing.)

[Johnny Canuck]

Wasn't the problem that AIG got to choose its regulator, an organization originally designed to regulate thrift institutions, and that on staff was a grand total of one person in the insurance department?

[von]

Well, no. If we choose to regulate the market in labor by outlawing slavery, that doesn't amount to a statement that the market was mispricing the value of slaves.

Huh? I mean, leave aside the part where you take my statement completely out of context. Or that you're flirting with proof of Godwin's law way, way too early in this thread.* How are you equating regulating a market because you believe that its informationally inefficient with eliminating the market?

By all means, if you view the market for home loans and credit as morally akin to slavery, you should argue that we should end it.

*I consider references to slavery to be roughly equivalent to references to Nazis as rhetorical bombs. Godwin's law is: "As a Usenet discussion grows longer, the probability of a comparison involving Nazis or Hitler approaches 1."

[Ara]

To be honest, I'm not following the argument of this post. I think everyone pretty much agrees that regulation for the sake of regulation is a horrible idea and that what makes this crisis so tricky is that a regulatory solution that does more benefit than harm is hard to come by.

BTW, I think the idea that a government regulation should do no harm whatsoever is too stringent a test. In fact, I'm not sure that you'll ever really get a regulation that passes that test, since it seems that a regulation that does no harm just moves us to a Pareto superior position. Well, *why do we need the coercive force of government* to move us to Pareto superior positions?

Also, pretty funny to use Richard Rorty, the old New Dealer himself, and pragmatism to justify EMH and its attendant laissez-faire attitude.

[von]

The EMH *cannot work* in a bubble or a panic - where is the "information" which drives a bubble market?

That's one of the key criticisms of behavioral economists -- a criticism with which I agree. But, as I note in the second footnote, it's not the criticism addressed here. The criticism here is that regulators would (or did) have better information than the market. Thus, regulation would prevent this particular bubble/crash, and would prevent similar crashes in the future. I don't see any evidence of that. It's certainly not an idea supported by behavioral economics.

(I had the pleasure of both participating in and designing experiments studying the bubble/crash phenomena as an undergrading. Interestingly, the irrationaility of the bubble/crash is itself predictable -- and, no, I don't mean merely the crash bit. But that is a post for another day.)

[TheWesson]

Anyhow, we already interfere with the market when it comes to economic collapse. Why shouldn't we interfere with a collapse while it's still in a bubble stage?

Throttling back during a bubble (raising capital reserve requirements, for example) would have a systemic tendency to discourage fraud by reducing the free (yet imaginary) wealth floating around.

Another thought: Forcing the exposure of information by regulation is surely something free-marketeers can agree with (the more information the better the EMH will work, right?) Part of the problem before this last collapse was various "shadow entities" (shadow banking, shadow markets, shadow insurance) which displayed a great lack of public visibility (not publically traded.)

In this case the individual players don't want visibility, but the market as a whole could benefit. Hence regulation.

[Whammer]

Unfortunately, as an MBA who had the EMH pounded into me consistently, I must say that the EMH has probably cost me more personal $$ than anything....

That being said, I think a lot of the problems at the moment are due to lack of information and lack of transparency. This is a good article by one of the premier experts on the history of regulation in the US:

http://www.tnr.com/politics/story.html?id=686e3d61-aa6f-4431-acac-d067fe28ed8e&p=1

[Bernard Yomtov]

So the anti-EMH argument for regulation must be based on the following; bankers are irrational and make lots of foolish loans. Regulators are rational and can see that these loans are too risky, and can protect bankers from hurting themselves. At a theoretical level this doesn’t even pass the laugh test.

This just seems wrong to me. Bankers can be quite rational and still make harmful loans. What is rational behavior for one individual or organization may be irrational from a broader point of view. It all depends on who gets the winnings and who pays the losses.

This is more or less what all "principal-agent" problems are about, and of course it is exemplified whenever you have a system with private profits and socialized losses.

Let's take a much simpler example. A judgment-proof individual may quite rationally decide not to carry auto liability insurance. Is it stupid for the law to require carrying such insurance as a condition of driving? Of course not. The regulation prevents the driver from making a private profit - saving insurance premiums - if things go well, while making others pay the cost of an accident he causes - socializing losses.

[TheWesson]

The criticism here is that regulators would (or did) have better information than the market.

First of all, if the market is actually reflecting no information other than 'animal spirits', then it should be trivial for regulators to do better in the long term. Just hire regulators who have a complete bloodless lack of animal spirits :)

Second (to state that in a different way) they don't *need* to have better information than the market. Their job is not to beat the market - their job is to assume that the market eventually returns to a historical pricing basis and act accordingly.

Every bubble seems to have come up with rationalizations why history does not apply in this case; it would be the job of regulators to ignore these frothy sentiments.

I don't know if coming up with some 'historical pricing basis' is the be-all end-all, but regulators could do well coming up with any basis other than 'animal spirits' if you will.

If you assume EMH is broken (for long term price setting) during a bubble/panic, then some smart people can surely find a way to look for and detect speculation (changes in buy/sell patterns perhaps?) and counterweight that.

You can't necessarily make money by selling short in a bubble ("the market can remain irrational longer than you can remain solvent") but you can still know there is a bubble.

In other words, requiring in effect that regulators "beat the market" (do better than price setting via the EMH) is not appropriate. Just have them stupidly assume a long-term return to the long-term trendlines.

By the way - this reminds me - can you actually buy an 'indefinite short' in any form? I would personally love to be able to short a bubble on an 'eventually it will pop' basis.

Then your regulators could pay attention to the 'indefinite short' market and see what it is saying ...

[woody]

The job of regulators is to assume that, when there is enough money involved, everyone will cheat. No exceptions. They must also assume that the cleverness of the cheating increases logarithmically with the amount of money available to be stolen. Regulators must ASSUME the richest will cheat the most because they have the most to gain. But they cannot assume any predisposition for 'honesty,' ever. People, but especially the people motivated to make money with money, are dishonest charlatans and mainly should be in stocks...

[tgirsch]

woody @ 02:50 PM wins the thread.

[Donald Johnson]

"woody @ 02:50 PM wins the thread."

Agreed, but when he says

"people motivated to make money with money, are dishonest charlatans and mainly should be in stocks..."

is he referring to financial instruments or wooden ones? Works either way for me.

[TheWesson]

I want to second Bernard here - it is quite likely that in a complicated system, one or more parties have an agency problem (their agendas are different than their stated agendas or the overall interests of the group) and therefore quite possibly it makes sense to have an organization represent the collective interest (e.g. the government) ironing out the agency problem. (E.g. act in the interests of the stockholders or be fined/jailed.)

In addition to agency problems, there are also time horizon problems ("I'll make a lot of money in semi-fraud here and then move on when the house of cards comes down.")

In short, assuming that EMH regulates the conduct of bankers assumes a tightly coupled system across time and parties involved. We assume, for example, that Bankster McFraud will say, "I'm not going to do this deal because in five or ten years it could quite possibly ruin the value of the company for the stockholders."

This assumes the banker cares what happens (a) to the stockholders (b) in five or ten years.

Now, in some societies, Bankster McFraud could be immediately coupled to eventual outcomes happening to other parties by his moral sense ("I feel bad about this deal because it seems semi fraudulent and therefore is probably going to screw the shareholders in the long run".)

However, in the recent abundance of free market fundamentalism ("I must act solely in my own interests") this moral underpinning is absent.

Therefore it's reasonable for the collective (e.g. gov't) to step in and try to accomplish this coupling more tightly.

In other words -- what woody said.

[cleek]

is he referring to financial instruments or wooden ones?

or perhaps... they should be made into stocks.

carrot
onion
celery
water
carcass of one dishonest charlatan.

simmer for two hours.

[TJ]

Fat cats make good stocks? Does it taste like chicken?

[TheWesson]

I think you should securitize bankers for soup by slicing and dicing them first. It completely eliminates the risk of getting a bad tasting soup from a rotten banker.

[publius]

i think one casualty of the past few years is the EMH. It's a useless concept that explains virtually nothing about how the world really works.

so to the extent you're relying on that as a premise, i'm not sure it's a solid foundation

[publius]

to build on BY, you don't need to show that bankers are "irrational" to justify regulation. you just need a market failure.

here, the biggie is externalities. Super-leverage and super-risk led to extreme short-term gains at enormous cost (costs borne by the public as a whole). that's all you need to justify regulation

[Sebastian]

"Their job is not to beat the market - their job is to assume that the market eventually returns to a historical pricing basis and act accordingly."

Why would this be a good rule? Oil prices SHOULD be going up as more people want more oil and as it becomes more scarce.

Food prices SHOULD go down as better farming techniques become available.

Food prices SHOULD go up if these techniques are oil intensive.

etc. etc. etc. Historical pricing SHOULD change when real-world circumstances change. The whole problem is knowing which times are due to underlying real changes, and which ones are due to weird bubble/crash changes.

[Sebastian]

publius: "to build on BY, you don't need to show that bankers are "irrational" to justify regulation. you just need a market failure."

The existance of a market failure might justify good regulation, but it doesn't necessarily justify the regulation you actually get. See for example the stupid way Congress is going after the AIG bonuses.

BTW can you identify any clear government failures (without resorting to they didn't regulate banks the way I wanted). I mean cases where the government took clear action, and it was wrong? Because if you can't, I have trouble believing that you are analyzing closely enough to strike a good balance between government and market.

[TheWesson]

> The whole problem is knowing which times are due to underlying real changes, and which ones are due to weird bubble/crash changes.

That's right, sometimes you don't have a clear idea of what price the underlying "fundamentals" should dictate, and sometimes you do.

It's as if regulators should be using a model which is exactly the same as the market, but with speculation/fear taken out.

Sometimes not possible, sometimes possible. For real estate, price/rent ratios have been historically very stable, so it makes sense to use that as model of 'fundamentals'.

Everybody could suddenly decide they'd like to spend 90% of their income on a house, but they usually don't.

Obviously if a bubble/panic goes on long enough then it, too, will be wrapped into the historical baseline (if that's your model.) However, bubbles are characteristically not very long lived.

Why is the current crash (an ongoing decline YOY of 6% of GDP) not 'normal' and needs 'action' to 'return to normal'? Because it doesn't conform to some model you have of a 'normal' economy.

Use some common sense here.

[Roger Moore]

Another thought: Forcing the exposure of information by regulation is surely something free-marketeers can agree with (the more information the better the EMH will work, right?) Part of the problem before this last collapse was various "shadow entities" (shadow banking, shadow markets, shadow insurance) which displayed a great lack of public visibility (not publically traded.)

I think that this highlights one of the big problems with the market. The EMH says you can't beat the market by using information the market already knows. The corollary is that you can beat the market by dealing in opaque products that deliberately hide vital information from potential buyers, or by poisoning the well with false information.

The whole point of those "shadow" entities was to avoid regulations that enforce transparency. Some other areas- like real estate- were rife with fraud. The worst offenders, like CDOs, had both.

[TheWesson]

I would also like to point out that bond rating agencies were using rather simplistic models to rate the safety of housing-related securities.

A reasonable, relatively non-coercive means of intervention would be for the government (a somewhat disinterested party) to be developing their own models for rating investment assets.

This would at least be better than the models being developed by parties who have a financial interest in optimistic models (i.e. the bond rating agencies.)

We don't let corn-futures speculators predict the weather; the government does that.

[magistra]

Most countries that I know of have speed limits on the roads. We do not just assume that even experienced drivers will know how fast they can safely drive. Sometimes that means that very skilled drivers are restricted by arbitrary limits. It also means society gets protection from the effects of overconfident idiot drivers.

The people who set speed limits don't need to be the world's best drivers. They just need to know what limits are likely to keep the average driver safe and impose those. There's room for altering speed limits cautiously if there are new developments in car safety etc. But to assume that speed limits are intrinsically a bad idea seems to me to be rash. Particularly when it currently seems that many more than 50% of bankers are below average.

[Sebastian]

I'm all for speed limits if you think that means something like leverage control limits. I'm much more skeptical if you think it means government workers figuring out what assets are in a bubble and which ones are legitimately higher prices than last year.

[TheWesson]

I'd be fine with leverage being reduced in boom times (and expanded in bust times). Capital requirements for banks being regulated by the Fed in an anti-cyclical way, for example.

I'm just objecting to the notion that "no one can detect a bubble" due to the EMH.

I also don't believe it would be a problem if the government did do some research (and publish the results) to try to detect bubbles.

I also suspect the market may have some trouble pricing bubbles because of the lack of an 'indefinite short' - the short equivalent of "buy and hold". I am ready to be illuminated if someone can point me to such a thing.

[russell]

financial markets are 'informationally efficient', or that prices on traded assets, e.g., stocks, bonds, or property, already reflect all known information.

I think this is, by and large, true of the current situation. The problem is that if the paper in question is sold at the 'informationally efficient' price the sellers will be bankrupt.

Ergo, deadlock.

My guess is that, when the big complex web of sliced and diced securities is unwound, they will not turn out to be worth much more than what investors are willing to pay for them now.

All Geithner is doing is transferring a lot of the risk to the public sector.

We're taking one for the team.

The job of regulators is to assume that, when there is enough money involved, everyone will cheat.

here, the biggie is externalities.

Thank you, and thank you.

It's entirely possible to be a completely rational actor and be a greedy, conniving, larcenous SOB.

That's why we have regulation.

Arguing that the regulation we've had for the last couple of decades has been piss poor is an argument for better regulation, not less.

[liberal japonicus]

I'm not pinning this on any particular comment or commentator here, but there seems to be this notion that if Dow goes up, it is indicative that the fix is in, but if the Dow doesn't, Obama is doing the wrong thing. I tend to think that there is a puritanical streak in Americans that think someone, somewhere, has to suffer, and if we don't see some sort of suffering we aren't actually on the right path to getting out of this. Something along the lines of no pain, no gain.

[hilzoy]

von: first, about this:

"But a call for more watchmen and women -- without some explanation as to what harm they're trying to correct and why they might correct it -- ain't prudent."

I don't think anyone is proposing regulation for regulation's sake, or regulation for which no explanation can be given. Some commenters might not feel up to giving that explanation, or might not have the time, but I assume that generally they imagine that one exists, or else they would abandon their views. So I don't know who you're arguing against here.

The EMH is more interesting. As I understand the EMH, it says that markets are very efficient at disseminating information, that prices tend to reflect what information there is, and thus that it is not a good idea to try to outsmart the market. But I think there ought to be a crucial next clause here, namely: it is not a good idea to try to outsmart the market by making better investments, within the financial system that exists. You'd do better to pick a diversified portfolio and stick with it; prices diverge from the value that would reflect the available information only in random ways, so trying to predict things is foolish.

But regulators are not investors. They are not trying to pick better stocks than the market. Regulators have powers that ordinary market participants do not have: e.g., to set reserve requirements, to require certain forms of financial disclosure, to prohibit certain kinds of transactions or instruments, etc.

The fact that market participants have settled on the most efficient price for an asset given the current set of rules does mean that if I am a market participant, I should not bet against them. I have the same basic powers that they have; they have already exercised those powers optimally; I should ot try to outsmart them. However, it in no way means that if I were given a set of powers that normal participants in the market do not have, it would be irrational for me to use them. New powers completely alter the set of possibilities open to me, and it does not follow from the fact ( supposing it is one) that investors have exercised their powers optimally (on average) that there is no rational scope for mine, if mine are quite different.

Moreover, the aims of a regulator and the aims of market participants differ enormously. Regulators are not trying to set prices, only better ones than market participants. If they're good regulators, they are trying to set up rules that on balance make the actions of basically self-interested people work together for good, and prevent catastrophe.

If regulators set up financial disclosure requirements, they are not assuming that they have greater knowledge than those market participants who try to conceal the state of their balance sheets. Likewise, if they outlaw theft as a means of enrichment, they are not assuming that they're smarter than thieves. They just play a very different role than your standard economic actor, and have different powers and different goals.

[hilzoy]

Seb: "BTW can you identify any clear government failures (without resorting to they didn't regulate banks the way I wanted). I mean cases where the government took clear action, and it was wrong?"

Easy. Agricultural subsidies. The regulation of the airlines until Carter changed it. (Nb: I have no idea when that started, or whether it might have been a good idea very early on. If it started in WWII, for instance, I can imagine that given the overall level of regulation in the wartime economy, the need for aircraft, etc., I could be convinced that it was a really good idea. But surely not by, say, the 60s.) Nixon's wage and price controls.

[Bernard Yomtov]

BTW can you identify any clear government failures (without resorting to they didn't regulate banks the way I wanted). I mean cases where the government took clear action, and it was wrong? Because if you can't, I have trouble believing that you are analyzing closely enough to strike a good balance between government and market.

Easy. I believe that many trade protectionist measures are bad policy. I also believe that much agricultural policy is unwise, though I'm not familiar enough with the details to be specific.

[Bernard Yomtov]

I'm all for speed limits if you think that means something like leverage control limits.

That's one thing. Size limits on financial institutions and the positions they can take is another. Exchange-type derivative trading, to make sure traders can meet their obligations and the pricing is transparent. Modification of accounting rules so that there are not perverse incentives to refuse to sell assets at market prices. (BTW, how's that efficient market working with the CDS?) I'm sure others have good suggestions.

[Turbulence]

The current stupidity also underscores what should be the broad, guiding concept in all our righteous interventions. No matter how bad the disease, the cure can always be worse. (Hence the adage: "First, do no harm.")

I don't understand what you mean here. Of course we should refrain from harmful actions that have no potential upside. That's obvious; it is beyond trite. But real world interventions always carry the possibility of harm. If your rule is really "do no harm" then you will do absolutely nothing: even an aspirin can kill you.

Put another way: would you rather drive a car without seatbelts and airbags? Both of those technologies have killed people. Under the do no harm principle, we should not employ them, even though the probability that they will save your life is much higher than the probability that they will kill you.

[von]

Publius, when you write things like the current economic downturn proves that the EMT is wrong, I just scratch my head. The EMT has flaws. But it's a model. It's useful, and predictive, even now.

[von]

That's obvious; it is beyond trite. But real world interventions always carry the possibility of harm. If your rule is really "do no harm" then you will do absolutely nothing: even an aspirin can kill you.

Nahh, all I mean is that we should act rashly ("For we are lovers of the beautiful in our tastes and our strength lies, in our opinion, not in deliberation and discussion, but that knowledge which is gained by discussion preparatory to action. For we have a peculiar power of thinking before we act, and of acting, too, whereas other men are courageous from ignorance but hesitate upon reflection." yada yada yada*)

Frankly, this post isn't as pointy as I would have liked -- which is to say, it's not good at conveying, well, information.

*I'm bringing it back, baby!

[Lurker]

Hilzoy makes a very good point. The regulator is not (and should not be) a player of the market. The regulator esssentially creates and maintains the market.

As far as I understand, much of the financial regulation actually aims to create an effective market environment. For example, financial disclosures and bans on insider-trading are supposed to equalize the information possessed by the players. This is done to maximize the effectiveness of the market. Similarly, an effective market requires that the no player is able to dominate the market monopolistically. Thus, the anti-trust legislation.

Banning certain complex financial products could actually improve the effectiveness of the market. This is because the market players are not ideal but human, with limited information-processing capability. Most traders were professionally unable to make independent evaluation of different mortgage-backed securities. These products were designed by guys with theoretical physics backgrounds to be non-transparent. Imagining you could have an effective market of informed players for these is ridiculous. It would be the same as trying to build an effective stockmarket with 99% of the traders being five-year-olds.

[von]

Hilzoy, I agree with the distinctions you're drawing between the regulator and the regulated. I agree that we may need more process hurdles (disclosures, etc.), although, having seen Sarbanes-Oxley from the inside, I will say that even well-intentioned process hurdles aren't always a good idea. The EMT doesn't have much to say regarding such process requirements.

The kinds of regulations that are addressed by the EMT are "thou shalls" and "thou shall not" regulations. Regulations that govern trading activity, not the process of trading (e.g., who you need to disclose what to). It's these kinds of regulations that Professor Sumner criticizes because they presume that regulators have more information than investors. Sumner rightly notes that the evidence in support of this point is thin.

The fault for the confusion lies, I fear, in my post. It ain't stellar work.

[Bernard Yomtov]

The kinds of regulations that are addressed by the EMT are "thou shalls" and "thou shall not" regulations. Regulations that govern trading activity, not the process of trading (e.g., who you need to disclose what to). It's these kinds of regulations that Professor Sumner criticizes because they presume that regulators have more information than investors.

I think you've missed my point, von. EMH deals with the pricing of a single asset. It says nothing about trading volumes, ability to cover obligations, etc. No one is saying that if you want to buy a CDS at some price the govt should tell you it's a bad deal. What they are saying is that the govt has a right to keep you from making so many deals, or taking on so many obligations, that you threaten the financial system.

That has absolutely nothing to do with the price of a single transaction, which is what the EMH deals with. The argument is a total red herring.

Nor does the EMH have anything to say about the wisdom of dealing with certain counterparties. That's why exchanges have clearinghouses. Look, in a perfect market as envisioned by Sumner, those buying CDS from AIG would have known the state of its finances and ability to pay. Presumably they would refuse to deal with them at some point. Didn't happen, did it?

[von]

I think you've missed my point, von. EMH deals with the pricing of a single asset. It says nothing about trading volumes, ability to cover obligations, etc. No one is saying that if you want to buy a CDS at some price the govt should tell you it's a bad deal. What they are saying is that the govt has a right to keep you from making so many deals, or taking on so many obligations, that you threaten the financial system.

"EMH deals with the pricing of a single asset" - that's not the way I'd put it ("single asset"??). EMH concerns how well a market price for a given asset (or group of assets, or even asset classes) reflects a hypothetical world of perfect information. The strongest version of the hypothesis states that a market price reflects perfect information; weaker versions of the hypothesis allow imperfections to creep in.

"What they are saying is that the govt has a right to keep you from making so many deals, or taking on so many obligations, that you threaten the financial system." - But this is suggesting that regulators are better placed to know where this leverage tipping point is than private actors. The evidence adduced so far is that regulators missed the boat on this one.

Oh, sure, in hindsight, everone knows that these firms were overleveraged. But, prior to this point, there were not many voices among the watchman or those who watch the watchman. It was pretty much Nouriel Roubini among established names.

That's Sumner's point: why are we confident that the regulators will get it right the next time?

[Bernard Yomtov]

"EMH deals with the pricing of a single asset" - that's not the way I'd put it ("single asset"??). EMH concerns how well a market price for a given asset (or group of assets, or even asset classes) reflects a hypothetical world of perfect information. The strongest version of the hypothesis states that a market price reflects perfect information; weaker versions of the hypothesis allow imperfections to creep in.

Yes. I do understand. And no one believes the strong version - "reflects all information, public and private." What this has to do with the the behavior of financial institutions is unclear.

But this is suggesting that regulators are better placed to know where this leverage tipping point is than private actors. The evidence adduced so far is that regulators missed the boat on this one.

I suggest that from a regulator's point of view the appropriate limit to leverage almost surely is, lower than the limit the private actor regards as best. Neither may be perfect, but the regulator is concerned with much broader risks than the private actor, who offloads risk onto society.

Did regulators miss the boat? That may be an overstatement. How much control do they have over, say Bear Stearns, as opposed to commercial banks? And if they did maybe that's an argument for better regulation. If you refuse to regulate CDS you can't blame incompetent regulation when they generate a crisis. The "private market" argument, for example, is that the ratings agencies have every incentive to get it right, etc., so they are reliable guides. Hmm.

That's Sumner's point: why are we confident that the regulators will get it right the next time?

Again, what in the world does the EMH have to with this? Of course we're not sure they'll get it right. But so what. By that argument you'd look at every problem and say, "Well, no point in trying to address it because we can't be sure we'll get it right." Should we not have put in the system of bank regulation that was created in response to past systemic financial problems, such as those in the Depression, because of that lack of certainty?

[hilzoy]

von: ""EMH deals with the pricing of a single asset" - that's not the way I'd put it ("single asset"??). EMH concerns how well a market price for a given asset (or group of assets, or even asset classes) reflects a hypothetical world of perfect information. The strongest version of the hypothesis states that a market price reflects perfect information; weaker versions of the hypothesis allow imperfections to creep in."

Hmm. I thought it said that information -- the information available at the time -- was transmitted efficiently through markets. I.e., that it's information transmission that's perfect 9i.e., perfectly efficient), not that there is perfect information. (Nb: I get my understanding of the EMH from Malkiel.)

This matters. One case in which one might argue that the EMH implies no regulation is with regard to things like leverage. The point of limiting leverage, one might think, is to prevent bubbles -- but if the EMH is true (in a suitably strong version), then there are no bubbles!

However, another reason to limit leverage would be: new information can reveal that an asset's old value is wrong, not relative to the old information (if the super-strong EMH is right, then this can't be true), but relative to the new, amended information. And specifically, new information can reveal that the price of an asset should be lower. Maybe a lot lower. If this happens, then limits on leverage can limit how much trouble this causes people who own this asset.

If the EMH held that information is always perfect, then this could not happen, and there would be no surprises, and the EMH would be obviously absurd. But if the EMH holds only that what information there is is transmitted with perfect efficiency, then the possibility of new info forcing a revaluation of some asset, possibly a dramatic one, exists, and there's still a good reason for limits on leverage.

[Title loans]

With my experiences in the loan industry, I believe it is safe to say that the banks aren't lending because there are less customers to lend to. The credit card companies over saturated the market with credit. Most college kids had credit cards at their disposal before they left college. With some of them being financially illiterate - their credit gets ruined. Other casualities happen - foreclosures, defaulting student loans, and delinquent car payments.

The banks have milked consumers dry. I am surprised their hasn't been mergers instead of bailouts. There is too much competition in the banking system and not enough customers. The government needs to offer bailout money, but let the industry cut out the fat!