While we were out some people who keep to a less rigorous vacation schedule than we do wrote some stuff about complexity in finance. Lisa Pollack at FT Alphaville started the ball rolling by attributing increasing complexity in finance to increasing smartness in the financial industry. This is a delightful theory if you are or were in the financial industry, particularly if you were tasked with developing financial instruments, so we'll go ahead and endorse it.
Others, however, disagree. Here is a very nice thing at Interfluidity that various financial pundit types found life-changing:
Like so many good con-men, bankers make themselves believed by persuading each and every investor individually that, although someone might lose if stuff happens, it will be someone else. You’re in on the con. If something goes wrong, each and every investor is assured, there will be a bagholder, but it won’t be you. …
If the trail of tears were truly clear, if it were as obvious as it is in textbooks who takes what losses, banking systems would simply fail in their core task of attracting risk-averse investment to deploy in risky projects. Almost everyone who invests in a major bank believes themselves to be investing in a safe enterprise. Even the shareholders who are formally first-in-line for a loss view themselves as considerably protected. The government would never let it happen, right? Banks innovate and interconnect, swap and reinsure, guarantee and hedge, precisely so that it is not clear where losses will fall, so that each and every stakeholder of each and every entity can hold an image in their minds of some guarantor or affiliate or patsy who will take a hit before they do….
This is the business of banking. Opacity is not something that can be reformed away, because it is essential to banks’ economic function of mobilizing the risk-bearing capacity of people who, if fully informed, wouldn’t bear the risk. Societies that lack opaque, faintly fraudulent, financial systems fail to develop and prosper. Insufficient economic risks are taken to sustain growth and development. You can have opacity and an industrial economy, or you can have transparency and herd goats.
This is delightful too, and features goats; my guess is that it's mostly wrong but I don't count that against it. One objection to it - though not a decisive one - is that it is wrong about the psychology of the people creating the financial complexity. People who design swaps and reinsurance and guarantees and capital structures actually want to know, really really clearly, what happens when things go bad. Part of the proliferation of complexity - not opacity, complexity - is the desire to have that definition. That's what a CDO is. It's slicing the world into possible future states and clearly defining who gets what payouts in those future states. Simple banking is "I will put equity into a bank, you will put a deposit in, I'll loan out the money and whatever we get back goes first to you then to me." Complicated banking is "that, but also if we don't get the money back because of reason X, insurer A pays; and if we don't get it back because of reason Y, noteholder B pays; etc." Complexity is a move toward greater definition, greater clarity about where losses will fall, not less.
Of course it is popular to equate complexity with opacity and mendacity, and sometimes that's true. Surely my credit card agreement is long and complicated because it's intended to trick me. And many transactions really are designed mainly to obfuscate bank balance sheets. My sense, though, is that when you do things like build a CDO, your goal is certainty and transparency of payouts. The documents are long not to create confusion, but to clearly set out what happens in any likely event. And my sense is that when you do things like buy a tranche of a CDO, your goal is actually to get exposure to the set of risks that the CDO is designed to give you exposure to. Perhaps that's naïve.
In any case you can easily find places where this thinking gets you into trouble. Here is one: MBIA and Countrywide/Bank of America are in a lawsuit about some mortgage-backed bonds that Countrywide sold to investors and that MBIA insured. MBIA wrote its insurance based on reps and warranties from Countrywide, which it says are false, presumably because Countrywide repped things like "we underwrote these loans carefully and the people who got them had jobs and credit scores and appraisals and whatnot" when in fact they were just giving out 120% LTV mortgages to unemployed corpses or whatever. A New York court just ruled that MBIA doesn't need to prove that the wrongness of the reps caused its losses - that is, if Countrywide did crap underwriting and lied about it, and then the mortgage defaulted not because of the bad underwriting but because home prices fell, um, country-wide, then it has to pay MBIA back for the insurance it wrote. Is this a big deal? I don't know, maybe. Probably not. But if similar reasoning applied to bond investors - and it probably doesn't, this ruling is about the specific language of the insurance contract - then basically all of the risk of the housing downturn, which banks thought they'd offloaded to MBS investors and monolines, could come back to them. That would suck. For the banks. Not for lawyers. Or monolines.
So was Countrywide's contract with MBIA designed "precisely so that it is not clear where losses will fall"? Surely not, or at least, no one at Countrywide or MBIA thought that it was. Instead it looks like a pretty straightforward case of two sides trying to choose certainty-maximizing rules: Countrywide wanted the certainty of "MBIA is on the hook for losses on these bonds no matter what"; MBIA wanted the certainty of "if the reps are wrong we're off the hook no matter what, without having to prove that the reps caused our losses." Those desires for certainty were irreconcilable, and the conflict wasn't resolved probably because no one noticed it at the time (what could possibly go wrong with mortgages?), and for now the New York court has favored MBIA's desire not to be lied to over Countrywide's desire not to have any credit exposure to its own mortgages. But they didn't want the uncertainty. If they'd had a magic system for determining whether the reps were accurate at inception, they'd have used it. Because Countrywide really did want MBIA to take on the default risk, and MBIA really was willing to get paid to do so.
But, lacking magic, all they could fall back on was contract drafting. And for all their efforts, their contract couldn't cure what seems to have been a real ambiguity in what they thought was the deal. Countrywide is certainly not the first bank to be done in that way, and won't be the last. Clarity isn't always easy. For all the appeal of the increasing smartness theory of financial complexity, the intellectually capacity of financiers remains finite - at least at Countrywide - and can't encompass the infinite ways that things can go wrong. That is not, though, reason to believe that they're not trying.