An old remedy for a new problem?

[Chaos]

Studying economics during a major financial crisis, and hearing what people have to say about it, I cannot help but think, "All right, you´ve been learning this for several years, now what does all this hard-won knowledge tell you?"

I think we can all agree that the crisis was not a good thing, and that any repetition of it should be avoided, as long as the effects of whatever it takes to avoid it are not worse than the effects of the crisis they are supposed to avoid.

From my point of view, it seems that at the core of the financial crisis was a problem of risk - specifically, the excessively risk-friendly behavior of (predominantly) financical institutions. Private households, of course, also took too many risks, which led to the sub-prime crisis, but even here it was the banks who enabled them to take such risks, by allowing them to take out mortgages in excess of what their homes were worth. And even these banks would perhaps not have accepted such risky mortgage applications if they had not been able to sell them to investors in the form of CDOs.

What caused this excessive risk-taking? I think it was a matter of incentives. Look at it from a game theory perspective: If you make a risky investment, then if it makes money, you get to keep the profit, and if it fails, you get bailed out; if you do not make the investment, you get no profit. It does not take a genius to tell what any rational investor will do. And predictably, now that the shock of the crisis is beginning to fade, financial institutions are going back to their old behavior as if nothing ever happened.

Is there an alternative, then? Alternative #1, simply not bailing out the institutions, doesn´t work. Even if we knew for certain that one round of mass bank collapses would cure everyone of excessive risk-taking, the side effects would just be too much. The entire banking sector, and with it the vital functions such as savings and loans, would be wiped out, and the crisis so far would in all probability seem like a minor hiccup in comparison.

So, what else can there be done? There was once, in the US, a piece of legislation called the Glass-Steagall Act (officially the Banking Act of 1933). One part of it was to, in a nutshell force banks to separate the retail banking from the investment business.
Taking a page from that book, I think one solution to the problem that caused the financial crisis would be to introduce the following new rules:
1) Any bank that is newly founded can conduct either retail banking (savings, loans, bank accounts, mortgages etc) business or investment, but not both. Existing banks would not be bound by this rule, except for the provisions below.
2) Any bank that receives bail-outs or a similar form of government aid, no matter how much, must reorganize according to rule 1, either by dissolving or selling the investment business or splitting into two completely separate institutions. There are no exceptions to this.
3) If a bank which refuses to reorganize when in trouble is deemed "too big to fail", but faces failure anyway, it is nationalized, bailed out, reorganized according to rule 1, and the separate institutions are then privatized
4) Existing institutions which currently conduct either only retail banking or only investing cannot branch out into the other area once the new rules are in effect; the same is, of course, true for institutions created through a reorganization per rule 1
5) Pure investment institutions (either pre-existing or created through rule 2 or 3) that are in danger of failing will never be bailed out; pure retail banking institutions that are in danger of failing will be bailed out, but this will have serious consequences for their management.

Anyway, these are my thoughts. Any comments?

 

[drkitten]

Studying economics during a major financial crisis, and hearing what people have to say about it, I cannot help but think, "All right, you´ve been learning this for several years, now what does all this hard-won knowledge tell you?"

Paul Krugman (New York Times economics columnist and Nobel Laureate) has a pretty good book out on exactly that subject, with the title "The Return of Depression Economics," and he lays out more or less this very point.

Basically, we forgot what we have learned.

(And by "we," he largely means "supply-side Republicans." The idea of demand-side control of the economy -- what he calls the "neo-Keynesian compact" had been so successful for so long no one really took it into their heads to defend it when the supply-side started to write their articles in the late 70s and early 80s. Thirty years later, the lunatics have taken over the asylum and Bad Things ensue.

point of view, it seems that at the core of the financial crisis was a problem of risk - specifically, the excessively risk-friendly behavior of (predominantly) financical institutions. Private households, of course, also took too many risks, which led to the sub-prime crisis, but even here it was the banks who enabled them to take such risks, by allowing them to take out mortgages in excess of what their homes were worth. And even these banks would perhaps not have accepted such risky mortgage applications if they had not been able to sell them to investors in the form of CDOs.

Not as much as you might think (according to Krugman), depending upon what you define as "financial institutions." The real problem was simply the insufficiency of regulation among the things (like hedge funds and auction-rate securities) that didn't qualify as "financial institutions" and therefore weren't subject to the various Depression-era regulations that were supposed to prevent this.

Basically, the regulators were asleep at the wheel and allowed a huge "shadow banking system" to arise, an unregulated collection of piles of money that promised higher returns than you could get from regulated investments, but didn't have enough capital to ride out the risks they were taking.

But it largely wasn't the banks' fault; it was the unregulated bank-like institutions who bore most of the risk -- and who ended up handling the majority of the money because they could offer better returns in good times.

What caused this excessive risk-taking? I think it was a matter of incentives. Look at it from a game theory perspective: If you make a risky investment, then if it makes money, you get to keep the profit, and if it fails, you get bailed out; if you do not make the investment, you get no profit. It does not take a genius to tell what any rational investor will do. And predictably, now that the shock of the crisis is beginning to fade, financial institutions are going back to their old behavior as if nothing ever happened.

Yes, he discusses this point at length as well.

But it's not simply a crisis of regulation; there are a number of technical issues at work here as well. Reinstating a stronger version of Glass-Steagall would be a good start (as you propose), but the bigger thing is simply recognizing that the government does have a role to play in stabilizing the demand side of the economy, and failure to actively stabilize causes bad things to happen.

 

[Tumblehome]

I think we can all agree that the crisis was not a good thing...

Not all, I think. I suspect that some well-placed people have the opposite view. If you knew what was going on--and there had to be some who did--you could have used it to your personal advantage, never mind what it would do to everyone else.

I find it hard to believe that every top-level money player forgot the lessons of the 1930s.

 

[Francesca R]

And by "we," he largely means "supply-side Republicans."

That would tend to undermine the credibility of the book, if it was true. Fortunately I never picked up any particular "anti Republican" slant. Of course, the book is not really specifically about the US recession, but a re-issue of Krugman's analysis of past crises in Japan, Mexico, Argentina and Asia.

 

[Francesca R]

Anyway, these are my thoughts. Any comments?

My briefest summary of this issue is that the problems with "the system" are:

1--Fear/greed
2--Correlated mistakes
3--Leverage that gets so high it creates negative externalities ("public bads")

And that (1) and (2) cannot be regulated or legislated away at all, since they are for one reason or another evolved into human nature, which leaves (3)--leverage limits within institutions and within the network of institutions as a whole--as pretty much the only accomplishment that can be achieved.

However (3) is difficult because as soon as regulators require tighter capital management in one area, the next bull market cycle finds a way of having it pop up elsewhere. This is exacerbated by special interests working their hardest (as it their mandate/incentive) to water down regulations.

To be pessimistic: separating "utility" banking from investment banking would achieve nothing IMO. The safety of some imagined "firewall" there is entirely illusory, and this was evident in Bear Stearns' rescue and Lehman Brothers' collapse (neither of these were regular banks and they had not one deposit between them)

To be similarly pessimistic, "too big to fail" is also a misnomer sometimes. If the errors in risk-taking are highly correlated enough, then lots of small leveraged balance sheets essentially become one large leveraged one anyway. Mathematically there is always some diversification in splitting them up, but if there correlation coefficient is almost 1 then that is not enough to save the system. Even if separate institutions are well enough diversified, the market is apt to doubt that is true in a crisis. This was evident from even Goldman Sachs having to accept some measure of public bailout (and some from Berkshire Hathaway).

 

[Francesca R]

I find it hard to believe that every top-level money player forgot the lessons of the 1930s.

Yes there are investors who did very well out of the 2008 market crash. That does not lessen the social ill of a huge avoidable income loss and a nasty sized call on public funds everywhere.

 

[SezMe]

My briefest summary of this issue is that the problems with "the system" are:

1--Fear/greed
2--Correlated mistakes
3--Leverage that gets so high it creates negative externalities ("public bads")

And that (1) and (2) cannot be regulated or legislated away at all, since they are for one reason or another evolved into human nature, which leaves (3)--leverage limits within institutions and within the network of institutions as a whole--as pretty much the only accomplishment that can be achieved.

I want to speak to your #1, especially greed. I agree that it cannot be regulated or legislated away, but the consequences of it can be changed. More specifically, if individual tax rates on high incomes are set to the pre-Reagan era levels of 70-90% then the effect is to not reward greed.

Consider two scenarios. First, let's look at the executive levels of compensation of large financial institutions. With high tax levels, individuals do not make much gain from incremental income and so high compensation packages are not motivational.

Second, let's look at the small/midsized business man. If he (ignore the sexism for the sake of the argument) takes a large compensation, much of it is lost to the government. If, however, he reinvests corporate income back into the company, he loses little and gains potential asset growth. And, by the way, boosts employment.

In summary, I agree that greed cannot be directly legislated away but its negative consequences can be addressed to effect.

 

[Tumblehome]

Yes there are investors who did very well out of the 2008 market crash. That does not lessen the social ill of a huge avoidable income loss and a nasty sized call on public funds everywhere.

Yes, that was my point. Some investors did very well to the detriment of the system as a whole, which makes me wonder if this was engineered to a certain extent. I say that based only on the observation that everything that is worth anything is controlled by someone or a group of someones, whether it's a song, a house, a corporation, a labour union, or a nation. Is the money market any different? Isn't there a small group at the top with the ability to pull strings and influence legislators?

 

[Francesca R]

I want to speak to your #1, especially greed. I agree that it cannot be regulated or legislated away, but the consequences of it can be changed. More specifically, if individual tax rates on high incomes are set to the pre-Reagan era levels of 70-90% then the effect is to not reward greed.

You equate greed with creating wealth, so your remedy is extremely blunt and has unintended consequences. Jealousy isn't perhaps the best tool kit to be equipped with if you want to rewrite policy for the better.

 

[Francesca R]

Yes, that was my point. Some investors did very well to the detriment of the system as a whole, which makes me wonder if this was engineered to a certain extent.

That seems to be quite a hair-trigger conspiracy button you have.

I say that based only on the observation that everything that is worth anything is controlled by someone or a group of someones, whether it's a song, a house, a corporation, a labour union, or a nation. Is the money market any different? Isn't there a small group at the top with the ability to pull strings and influence legislators?

Influence varies some but not to the point where financial markets are controlled by anyone, ex regulators (so you have to blame regulators for the crash--which many people do). Again you appear to lean towards paranoia and conspiracies (as I suppose many do also, but I only encounter them on the internet).

 

[Tumblehome]

Well, "hair-trigger" is an overstatement. It wasn't a definite theory, just a hypothesis. Given the cirumstances, I thought it was worth bringing up to see how it could be shot down, and you did that very effectively.

 

[SezMe]

You equate greed with creating wealth, so your remedy is extremely blunt and has unintended consequences. Jealousy isn't perhaps the best tool kit to be equipped with if you want to rewrite policy for the better.

Actually, I don't. If a greedy person does not accumulate wealth, then that greed is not manifest. If he/she does, my idea works.

Well, I can think of some forms of greed that might not lead to wealth but no public policy is perfect.

What unintended consequences can you envision?

 

[Chaos]

My briefest summary of this issue is that the problems with "the system" are:

1--Fear/greed
2--Correlated mistakes
3--Leverage that gets so high it creates negative externalities ("public bads")

And that (1) and (2) cannot be regulated or legislated away at all, since they are for one reason or another evolved into human nature, which leaves (3)--leverage limits within institutions and within the network of institutions as a whole--as pretty much the only accomplishment that can be achieved.

Greed and mistakes cannot be regulated or legislated away, that is true. However, by giving the right incentives through regulation and legislation, we can reduce their negative influence on the market. To use a deliberately overblown example, if your bank is nationalized the moment it buys a CDO, no bank will buy a CDO - so, greed would still exist, yet its effects on the market through CDOs would not.

However (3) is difficult because as soon as regulators require tighter capital management in one area, the next bull market cycle finds a way of having it pop up elsewhere. This is exacerbated by special interests working their hardest (as it their mandate/incentive) to water down regulations.

I think in the worst case, if we crack down on one incarnation of it, then the next incarnation will not be as strong and pervasive by the time it becomes a problem as it would have been without the crackdown.

To be pessimistic: separating "utility" banking from investment banking would achieve nothing IMO. The safety of some imagined "firewall" there is entirely illusory, and this was evident in Bear Stearns' rescue and Lehman Brothers' collapse (neither of these were regular banks and they had not one deposit between them)

Like I said in the OP, at least part of the problem was the expectation that any failure would be bailed out. If institutions had it in writing (in law, no less) that this will not happen again, they are bound to be more careful. If nothing else, when the next predictable crisis happens and investment bank after investment bank folds without a bailout, after that at least the survivors should have learned their lesson.

To be similarly pessimistic, "too big to fail" is also a misnomer sometimes. If the errors in risk-taking are highly correlated enough, then lots of small leveraged balance sheets essentially become one large leveraged one anyway. Mathematically there is always some diversification in splitting them up, but if there correlation coefficient is almost 1 then that is not enough to save the system. Even if separate institutions are well enough diversified, the market is apt to doubt that is true in a crisis. This was evident from even Goldman Sachs having to accept some measure of public bailout (and some from Berkshire Hathaway).

I don´t know. As far as I can tell, big companies that get into trouble are a lot more likely to be bailed out than smaller companies getting into trouble. If a hundred or a thousand firms collapse whose names nobody ever heard of, who gives a damn in Washington or London or Berlin? Whereas, if a big name is in trouble, the billions will start flowing sooner rather than later. I think Lehman was a fluke in that regard, a predictably failed attempt at sending a warning to Wall Street that they won´t be automatically bailed out, which of course is a weird thing to warn, against the background of a trillion dollar bailout orgy.

 

[SezMe]

Influence varies some but not to the point where financial markets are controlled by anyone, ex regulators (so you have to blame regulators for the crash--which many people do).

Not controlled as someone flying a model airplane but the four large banks formed an oligarchy which gave them effective control of some financial markets. The same situation existed with AIG and the other biggies in the insurance market.

I don't think these in any way constitutes a conspiracy, it is the natural evolution of a market segment that is basically unregulated.

 

[stevea]

I think we can all agree that the crisis was not a good thing, and that any repetition of it should be avoided, as long as the effects of whatever it takes to avoid it are not worse than the effects of the crisis they are supposed to avoid.

From my point of view, it seems that at the core of the financial crisis was a problem of risk - specifically, the excessively risk-friendly behavior of (predominantly) financical institutions. ....

What caused this excessive risk-taking? I think it was a matter of incentives. ....

Is there an alternative, then? Alternative #1, simply not bailing out the institutions, doesn´t work. .....

So, what else can there be done? There was once, in the US, a piece of legislation called the Glass-Steagall Act (officially the Banking Act of 1933). One part of it was to, in a nutshell force banks to separate the retail banking from the investment business....

Anyway, these are my thoughts. Any comments?

Generally a nice exposition Chaos'. I agree with your description of the 'current mess', tho' I disagree considerably with your analysis and proposed solution.

You use the term "excessive risk" which is a flawed concept. We want to encourage rational open-eyed, correctly evaluated risks and discourage reckless unprofitable risks. Very high levels of risk are perfectly acceptable and rational so long as the expectation value exceeds the cost AND we have the resources available to maintain the effort until a success is likely. If we have a venture with a 1% chance of success per try, but the payoff exceeds 100 times the cost, then there is, on average, advantage in the venture. If we only have resources to make one try at the project the chance of success is (usually) too low to risk the investment. If you have resources for hundreds tries then then the potential for net loss becomes quite small..

So "too much risk" is not the problem. The problem involves the incorrect assessment of risks and the fraction of total resources expended given the risk statistics.

The fundamental flaw leading to the current problem involved Investment ranking firms and insurers (AIG deserves some special blame here), given the task of assessing risks of CDO arrived at a figure that was far too low. It makes perfect sense that if a bank can sell low quality mortgages at a premium price, then they will create these mortgages and sell them; that's not illegal or immoral. Eventually FNMA got involved and became an easily accessible market for these (in retrospect) overpriced low quality mortgages, creating a government guaranteed market for junk loans. Many retail banks did not participate and are as a result are in fine shape. A few of the large investment banks also saw the problem (Goldman's for example) and began reducing their exposure early. In interviews AIG claims their assessment of risk was rational given the information available at the time. The primary fault was they were only considering selective risk, not systemic risks.

My assessment is that the banking industry created a reflexivity condition - their assessment that risk was low, in itself made more money available and the increased credit supply creating new risk.

===

The second topic is the Federal/International reaction to this problem. There are US Federal safeguards in terms of FDIC insurance which protects the vast majority of retail banking accounts. So, yes the Federal government SHOULD HAVE HAD restrictions in place on use of insured assets. The Federal governemnt has proven to be entirely incompetent at managing this banking insurance and risks. In the 1980s the Keating S&L crisis was another smaller scale example of the same Federal error of underestimating risk and failing to require risk management.

So yes, IF the government insures investors assets they have every right to set conditions on use of those assets, and that there be some separation between the various asset management since the insured assets cannot underwrite the uninsured without introducing unlimited risk. Government also has a role to play in standardizing the type of risk disclosure all banks (insured or not) make to customers, to ensure transparency in a free market. They have absolute no place in telling uninsured investment banks what level of risk or reserves they should incur and no place in reducing that risk ex post facto by propping up uninsured investors whose investments fail. Retail/Investment is the wrong dichotomy; If you want to keep your personal in the uninsured hi-risk account, that's your decision. You should have access to good disclosure of risk when you invest, but you should not be prevented from taking risk.

The idea that there would be some huge problem if more investment banks and their insurers closed, is nonsense thinking. These banks (well many divisions of them) have failed in actuality. Pushing then thru reorganization would not create any greater problem than already exists EXCEPT the investors rather than non-investor taxpayers would pay for their incorrect risk assessment. I'll admit the Fed has a place in creating an orderly market place so stabilizing these organizations long enough to reorganize their operation in a non-catastrophic manner. That's quite different from the current Bernanke "fiasco on the 10 year plan" scheme.

The other woo concept in play today is "too big to fail". Again read Soros on reflexivity. They are only too big to fail because we don't allow them the shrink into deserved oblivion. It's a self-fulfilling prophecy that funnels public funds into hopeless risky causes. This also (the public risk underwriting) is already appearing in fiscal risk assessments ! The public penchant to pay-off investment errors is being calculated into the odds.

==
As for institutional changes ....

The only problem then is to correctly assess the risks of the banking investments; and that is not the trivial problem that Krugman (who won a Nobel for hating Bush, not economic achievement) and the current administration imagine. Hundreds of billions of dollars managed by very bright and highly motivated investors went into CDO and MBS because they all failed to see the risk. The idea that the Fed or some other population of G-15s will recognize the risk that Lehman's and Warren Buffet failed to see is extreme hubris, not rational thinking,; it's kookie talk.

Even if some panel of geniuses could see bubbles forming early based on some econometric model, then we must still consider the advantages of a successful bubble determination versus the losses. This panel may have have also stalled the formation of the plastics boom, the semiconductor industry, the PC, the pharmaceutical, GMO, cell phone industry advances too. Not all booms go bust !

The other flaw in the administration plan is that the government agency assessing the systemic risks will, if they take any action, itself enforce a "herd mentality" behaviour. that is potentially bubble-creating, and certainly exploitable by investors. Instead of 300 regional banks independently assessing risks and making decisions there would be a Federal agency restricting selections at various times, thus concentrating the credit. The agency would actually increase the chances of another systemic error ! It's centralized risk management is just another form of central planning and have the same types of exploitable inefficiency.

At this point I'd advise that you look into the history of market bubbles, since this is mostly a mortgage credit bubble phenomena. These bubbles not generally NOT recognizable by intelligent observers until the excess well under way. The first parties to recognize the excess may reduce or remove their exposure, but that very recognition causes the bubble to burst. If you have a tool to modify human nature then you might be able to address this; otherwise ... diversify and study MPT.