I finally read Gillian Tett’s Fools Gold, an account of the development of the derivatives industry centered on credit default swaps (CDS) and collateralised deposit obligation (CDOs) that collapsed so spectacularly last year. The discussion is excellent, but still, I think, too charitable to these instruments and their creators. Tett’s main source is the group at JP Morgan who pioneered many of these derivatives and, largely, got out before the crash. Their line, unsurprisingly, is that the problem was not with the concept as they developed, but its abuse by latecomers.
But a close reading of Tett’s account yields a different story. These innovations were defective from day one.
The crucial thing that made all these deals work was the so-called ‘super-senior’ tranche of debt that was supposed to be almost completely immune to failure (until, of course, it failed). This stuff was rated better than AAA, with the result that lots of banks were willing to carry it on their own books, using Enron-like special investment vehicles to skirt the Basel 2 requirements for capital adequacy. The alternative was to find a supposedly risk-free backer, such as an insurance company (AIG) or that contradiction in terms, a “monoline” municipal bond insurer willing to diversify into insuring exotic derivatives (Ambac and MBIA). The JP Morgan crew were never comfortable carrying huge volumes of debt, even allegedly riskless debt on their own books, and that’s why they ultimately left the field to others. But according to Tett, the very first deal that was done involved transferring the super-senior debt to none other than AIG, whose threatened collapse forced the Fed into the trillion dollar bailout of 2008. So, the worm was in the bud – there never was a sound basis for the whole idea.
Another important implication is that, thanks to the massive size and complexity of modern financial markets, fundamentally defective innovations need not be weeded out quickly, and can grow to astronomical size before they are. Bernie Madoff’s Ponzi scheme is a straightforward illustration of this. When it was exposed, quite a few commentators suggested that no one could run a Ponzi on this scale for nearly 20 years, as Madoff did. The alternative explanation, which was shown to be baseless, was that Madoff must have initially run a speculative strategy, turning to a Ponzi only when that ran into difficulties.
This is one instance of a more general point emerging from discussion of the financial crisis. As Felix Salmon observes, the extraordinary profitability of investment bank can most plausibly be explained by the hypothesis that risk is being shifted, without compensation, to someone else. Salmon focuses on the case of ignorant buyers, sold products they don’t understand. But, as Arnold Kling observes, an equally important source of investment banking profits is regulatory arbitrage at the expense of governments, and, ultimately, the public at large.
There are two main ways these problems can be resolved. To protect both ignorant buyers and the public, it would be necessary to regulate investment banks in the same way as other banks, and much more tightly than either was regulated pre-crisis. In particular, the idea of letting Goldman Sachs get the protection of a commercial banking license while operating as an investment bank is an obvious example of the kind of regulatory arbitrage that needs to be stopped. Properly done, regulation of this kind would kill off investment banking of the kind with which we are familiar.
The alternative is to assume that the buyers of investment bank products can look after themselves, and focus on protecting citizens from being made to repeat the bailout disasters of last year. The only way to do this is to reinstitute a much tougher version of Glass-Steagall, raising high barriers to all kinds of transactions (ownership, financing, joint venture) between investment banks and the core financial system guaranteed by government. Something of this kind will have to be done with respect to hedge funds and similar outfits if we are not to have a repeat of LTCM somewhere before long.
{ 20 comments }
Adam Hyland 10.06.09 at 1:17 am
I haven’t read Fool’s Gold but I wouldn’t be too shocked to see that Tett’s informants would tell him that the smart players got out early (as it were). I also don’t think a sentiment like that is completely self serving. Kindleberger spends some time walking through what he saw as the chronology of a bubble and the overeagerness of early adopters tends to give sway to the prevalence of latecomers and copycats. Donald MacKenzie’s explanation for LTCM’s failure in 1998 follows a similar path (rival hedge fudges duplicating trading strategies causes LTCM’s portfolio to be replicated many times).
We can believe both explanations. I agree that these instruments were suspect and I also agree that the alarm was sounded early. But a huge portion of this crisis was caused (at least in the proximate sense) by firms and investors chasing the tail of profits already captured by early firms.
Henry 10.06.09 at 1:50 am
See also “Cosma’s review”:http://www.cscs.umich.edu/~crshalizi/reviews/fools-gold/.
Joshua Holmes 10.06.09 at 2:45 am
I wholly agree that the worm was in the bud. What I don’t understand is how the repeal of Glass-Steagall caused it. From what I’ve read, most of what went wrong was investment banks making bad investments, not commercial banks making bad investments or investment banks making bad commercial loans. AIG is neither. The bailout of the investment banks was done outside the commercial banking system, through TARP and the Federal Reserve’s “extraordinary measures”. Perhaps Glass-Steagall’s repeal was a bad policy, but I don’t see how it shoulders the blame here.
Thorfinn 10.06.09 at 3:17 am
I’m with Joshua. There’s no evidence that Glass-Steagall had anything to do with this crisis; if anything, by allowing a legal route for bad investment banks to grab commercial deposits, it made things better. Nor has there been any trouble from Hedge Funds; even LTCM didn’t cost the government anything (AFIAK).
An alternative is to force investment banks back into unlimited liability partnerships. Surely it’s not a coincidence that the moment they all went public they started piling up tons of risk. Or, we could break up their banking and brokerage arms, which we can’t help but bail out, from their proprietary trading side. Banks will fail from time to time; they just need to be structured so that they don’t take down everyone with them.
John Quiggin 10.06.09 at 4:07 am
With full separation between investment and commercial banking, the bailout would not have been necessary. AIG could have been liquidated, and taken down Goldman Sachs and the rest of Wall Street with it, without posing a systemic risk or killing off the flow of credit to business. The same in LTCM (BTW, Thorfinn, the fact that there was no cash payout doesn’t mean this rescue was free – it paved the way for the trillions being paid out now).
Thorfinn 10.06.09 at 4:51 am
I’m confused. Bear Stearns would still have failed; Lehman would still have failed; and I don’t see any mixing of commercial and investment banking at AIG either. This shadow banking system would still be important without Glass-Steagall, and would still require a bailout. At the core of all three failures is the mismatch between short-term borrowing and long-term lending, and a horrible mismatch of financial intermediation combined with excessive risk-taking. I don’t see Glass-Steagall as helping any of these issues.
John Quiggin 10.06.09 at 5:31 am
The point is that, if these institutions weren’t closely intertwined with the commercial banking system, they could be allowed to fail – people who dealt with them would lose money in just the same way as people lose their money in the stock market. Of course, the barriers between the two had been broken down long before the final repeal of Glass-Steagall.
Adam Hyland 10.06.09 at 6:39 am
Before this gets derailed, I should point out that John’s point about regulatory arbitrage was broader than Glass-Steagall and we would do well to look past that heavily trafficked debate.
I will note that regulatory arbitrage is an exceptionally broad category of trading strategies. And a category which is difficult to clamp down on. Governments have been writing tax law for thousands of years, with little progress in eliminating tax avoidance. We have to face the struggle between writing strict legislation which could open up many opportunities for arbitrage with empowering strong regulators, which allows for the possibility of capture.
Why not attempt something approaching Gary Gorton’s solution (Slapped in the Face by the Invisible Hand, hyperlink removed so as not to offend the post gods) where the government treats all short term informationally insensitive debt as equivalent to demand deposits, rather than walling off banks and investment banks?
Phil 10.06.09 at 8:29 am
I like Cosma’s review, which (a) gives due prominence to what Tett says the problem was (not CDOs but over-correlated CDOs) but (b) doesn’t quite buy it. I particularly like the conclusion (which is essentially CDOs don’t kill banks!), although I’m not sure I can gauge the level of irony.
Getting back to the OP, I’m not convinced that offloading the super-secret-special tranche to AIG was such a smoking gun. “We have reduced the risk of default to a historically unprecedented 0.0001%” (or whatever) isn’t incompatible with “We don’t like carrying monster debts even if the risk of default is a historically unprecedented 0.0001%”. Or (I haven’t read the book) were Tett’s young bucks going around telling people the risk was zero – not effectively zero, actually *zero* zero? If that was the case, well…
Ginger Yellow 10.06.09 at 12:04 pm
My usual quibbles:
This stuff was rated better than AAA, with the result that lots of banks were willing to carry it on their own books, using Enron-like special investment vehicles to skirt the Basel 2 requirements for capital adequacy
It was not rated better than AAA. There’s no such thing. It was considered safer than a “normal” AAA, and understandably so – in order to suffer a loss, another AAA tranche would have had to be completely wiped out. The problem being, of course, that it turned out the risk of the subordinate AAA tranche being wiped out was a lot less than AAA (plus the correlation of underlying assets was much higher than assumed, rendering subordination much less effective). Also, Basel II meant they did not need to use “special investment vehicles” (not actually a term in use – do you mean “structured investment vehicles”, “special purpose vehicles” or something else?) to hold the risk, because the capital requirements for super-senior tranches of securitisations was extremely low, especially under the IRB. It was even more cost-effective for insurance companies to hold these exposures, which is one reason why monolines and multiline insurers (especially reinsurers) ended up with so much of it.
I’m not sure why transferring the super senior risk on the first deal to AIG demonstrates there was never a sound basis for it, per se. Certainly AIG’s outsized market share in the later years should have been a warning sign, but there many players in the intervening years. The returns on super senior tranches were very low indeed, commensurate with the perceived risk. Consequently it didn’t make sense for institutions subject to gross leverage limits to hold such exposures, nor those with large exposure limits. But for institutions built around a low risk, low return model, with the right (or wrong, as it turned out) capital adequacy rules, it seemed to make sense.
Or (I haven’t read the book) were Tett’s young bucks going around telling people the risk was zero – not effectively zero, actually zero zero?
Clearly not, or if they were, only a complete idiot would have believed them. There is no such thing as a zero risk investment, including a US government bond maturing tomorrow.
DLJ 10.06.09 at 1:07 pm
In the mid 90s investment banks wouldn’t have been able to keep the super senior tranches on their books because their cost of borrowing as non-banks was too high. They were not even viable counterparties for many derivative transactions for the same reasons as their funding restrictions.
Since then the size of investment bank balance sheets has ballooned which demonstrates a significant change of business model. There are probably many reasons for it: reduced revenues from broking client transactions, competition for market inefficiency profits and for staff from hedge funds are plausible reasons. However it is clear that crossover institutions like Citi and JP Morgan had an advantage in the very large mortgage transactions that would have knock-on benefits for other business. Glass Steagall and other regulatory loosening might well have facilitated these changes.
It is also clear that investment banks were not used to keeping large institutionally significant risks on their books for extended periods and financial institutions have demonstrated again and again that they are no good at managing big new things. There is a fundamental difficulty with managing something that is new and if it is going wrong may not go wrong straight away and in that sense CDOs and sub-prime were no different to, say, Latin American debt or internet stocks.
The CDO affair has many similarities with the far smaller split capital investment trust debacle. These similarly split pools of investments into different parts for sale to different investors. The first issues were excellent but as the market developed appetite for one part was outstripped by the tax advantage fuelled demand for the other which left the industry sitting on what were by the original criteria undervalued slices and as a result rocketing leverage.
To me this does suggest that something plausibly achievable and useful could have been done if someone had insisted that the investment banks show good reasons for the changing nature of their balance sheets and for keeping large slices of their own issues on their books — something that used to be seen as a sign of failure. I can’t see that a conservative presumption here would be harmful. Glass Steagall reform may have prevented that being done effectively because a homogenisation of balance sheets would be only to be expected.
In summary abolishing Glass Steagall and related reforms may have done damage by unleashing fundamentally new business models, muddying the waters for regulators and removing a firewall. I can’t see that there was anything remarkably dangerous resulting from CDOs and imagine that they will return in smaller but still significant volumes without repeating this disaster. The worst damage they will have caused is the legacy of increased concentration and decreased diversity in financial services institutions. Agency issues are real and as intractable in practice as they are in theory.
Barry 10.06.09 at 1:23 pm
I’ve read Tett’s book, and she’s obviously both relying on the Morgan bankers for too much information, and overly sympathetic to them.
Glen Tomkins 10.06.09 at 1:48 pm
Rent-seeking
When looking at an “investment” product like these securitized mortgages, do you really have to analyze them and find fundamental flaws in the internal structure of their risk allotment, to be able to say that the worm was in the bud? Wasn’t that obvious just from the externals, from the fact that these products offered something for nothing?
The whole idea behind the senior tranches was that the “investor” would make money by doing nothing but forking over some of his cash, which would then multiply by breeding with itself, with no need for the “investor” to exercise any sort of oversight or control over the thing invested in, real estate. You can characterize these products as complex all you want, and yes, adding another layer of abstraction between the homeowner and the mortgage holder does create complexity, but that abstraction itself just makes the moral standing of these products as rent-seeking opportunities, not investments, all the more clear and simple. I imagine that the people pushing these safe tranches made the pitch very simple, “Look, you’re going to be able to sit back and clip coupons on an investment for which the sap originator did all the work, and those saps over there buying the junior tranches are taking all the risk.” I’m sorry, but anyone past the fifth grade would understand the moral hazard in this pitch, of loan originators maybe not doing the same due diligence for mortgages they planned to unload on the saps down the road, that they did when the business was still honest, and the lender was expected to retain oversight of the security for the loan. Anyone who read the papers would have understood that the overheated market in homes, created in no small part by the effect of securitizing mortgages, had, by many sober observers’ estimation, inflated home prices to the point that even homeowners in the safe tranches would be underwater should prices ever deflate to sane valuations. Anyone paying any attention at all would have understood that the junior tranches, in such an inflated market, and with the problem exacerbated by the related refi industry encouraging people to use their inflated home prices to leverage “free” ATM money to finance their lifestyles, would be exposed to if home prices even leveled off, and that a fall in prices that started in the unior tranches would trigger a general deflation in prices in all tranches. Anyone with any moral sense would have questioned the ethics of a scheme for drawing profits from a scheme in which other did all the work and assumed all the risks, and, having a moral sense, would have been circumspect, and looked around at the whole complex of lies and treachery on which this “investment” was built, have seen that the structure was rickety, and not a good idea just in terms of enlightened self-interest. But, you know, everything is hopelessly complicated and opaque, however inherently simple, to people who just aren’t paying any attention at all. If we set up a system in which people are not just allowed, but positively encouraged, to throw something as powerful as large sums of money around without requiring of them any circumspection, of course it will end as badly as that beginning, original sin, that had that worm of wilfull ignorance in the apple.
I don’t disagree with any of the proposed regulatory remedies. But I think one more measure is needed, a measure to address directly the real problem this financial crisis has revealed — that we have far too many people with far too much excess money out there seeking sinecure rents. There is much, much more money than common sense and common decency out there, and that imbalance is deadly. It’s not just a matter of malallocation, that they could be directing this money to actual capitalization of enterprises that would produce needed goods and services, instead of throwing it at onanistic something-for-nothing schemes. There is a huge danger to the real economy of leaving so much money that it’s no longer money but has grown to be the Wrath of God, out there unsupervised, in the hands of a class of folks who have neither the inclination nor the ability to oversee its workings in the real economy. Let Capital slip its leash of human oversight, and it gathers itself naturally into bubbles and pyramids. Add to the suggested regulatory oversight the confiscation of excess income that individuals cannot allocate safely, much less wisely, and you have the reasonable, sane, moderate solution. If people won’t suprervise their excess money, then the people must.
SamChevre 10.06.09 at 6:32 pm
AIG could have been liquidated, and taken down Goldman Sachs and the rest of Wall Street with it, without posing a systemic risk or killing off the flow of credit to business.
I can’t see how, unless businesses are funded by commercial banks only. Most largish businesses in the US are funded by various forms of market borrowing (auction-rate securities, commercial paper, bond issues), not by bank loans.
Chris 10.06.09 at 7:13 pm
I already commented on this over at Brad DeLong’s, but if you’ll allow me to cut and paste:
The point of the ignorant-buyer phenomenon is not that all buyers are ignorant, or even that most buyers are ignorant, but that if you create an investment that is only attractive to the ignorant, then only the ignorant will invest in it.
Everyone wants more return and less risk. The non-ignorant understand that there is a tradeoff between these variables, and that you can’t have more return without there being more risk somewhere. There is no perpetual profit machine.
So if you offer something with high return and low apparent risk, the non-ignorant will understand that there is hidden risk somewhere, probably hidden in the exact instrument you are selling, and they will go elsewhere. It’s the ignorant who will buy in to risk-hiding instruments expecting a free lunch.
So, for a certain kind of social Darwinist or Objectivist, who doesn’t pity the fools and instead believes that they deserve to be parted from their money, everything is fine, right? Wrong. Everyone is ignorant about something. The purpose of financial innovation is to produce new tricks to fool investors and/or regulators who know the old tricks. As the complexity and variety of financial instruments increases, the finance sector finds and exploits new deposits of untapped ignorance.
P.S. The Quiggin I thought I knew had more sense than to write that last paragraph. Starting out by assuming something you *know to be false* based on recent events is no way to craft a policy.
Stuart 10.06.09 at 10:02 pm
The purpose of financial innovation is to produce new tricks to fool investors and/or regulators who know the old tricks. As the complexity and variety of financial instruments increases, the finance sector finds and exploits new deposits of untapped ignorance.
I suppose this is similar to how a con-man only needs to know a couple of tricks to make a good living, as there are always plenty of people that don’t know how it works. And as Madoff proved, even a time honoured and well known con can be made to work (for a time) if you dress it up right.
Barbar 10.07.09 at 2:45 am
The non-ignorant understand that there is a tradeoff between these variables, and that you can’t have more return without there being more risk somewhere.
I don’t think is true. First off, although it is a staple of textbook finance it is not true that higher returns necessarily equals higher risk; a good argument can be made that in reality investors often pay good money to chase after risk (and the possibility of getting better-than-normal returns), depressing the returns. And if higher returns are earned through informational advantages, then higher-return investments are actually more of sure things.
In the case of the super-senior tranches being discussed here, I don’t believe that the promised returns were remarkably high given the level of risk assumed (i.e., little to none).
david 10.07.09 at 5:23 am
And the higher returns from regulatory arbitrage, and from screwing Simmons matress, etc. The risk-return relationship is a complicated one, and the idea of a perfect tradeoff is one of the things that got us into this mess.
Cranky Observer 10.07.09 at 4:56 pm
> P.S. The Quiggin I thought I knew had more sense than to write that
> last paragraph. Starting out by assuming something you know to be
> false based on recent events is no way to craft a policy.
This is the most amazing financial and economic crash in human history: no one did anything wrong, no one committed any fraudulent/criminal acts, there was no bezzle, none of the changes to regulations made in 1998 had any affect on later events, similarly the gutting of the financial regulatory apparatus 2001-2006 had no effect, and not a single widely-held theory of Economics(tm) [word spoken in deep booming voice] was proven incorrect. Amazing.
Cranky
John Quiggin 10.07.09 at 8:10 pm
@Chris, I intended this to be taken more snarkily than you read it, but I stand by the substantive point.
There are a wide range of investments where buyers are not protected against poor judgement or defective products, beyond some limited disclosure requirements and laws against outright fraud. Hedge funds are one example, but the most important case is that of the stockmarket – there’s nothing to prevent you setting up a company to sell dog food over the Internet and selling shares for whatever people are willing to pay, and no one is suggesting that this is going to change.
If we are going to continue with investment banking in anything like its current form, it will have to be on a similar basis to stockmarkets and hedge funds, which means it needs to be kept clearly separate from banking in the ordinary sense, both in terms of financial connections (to prevent regulatory arbitrage) and in terms of consumer protection, so that no one is under the impression that they are dealing with a bank backed by some kind of guarantee.
Comments on this entry are closed.