Sometimes quants have their heads in the sand (or some other dark place). This post on the beauty of market segmentation in insurance floated by recently. Another case of ignoring Your Good Mother's Warning: "what would the world be like if everybody behaved as you do?" Here we find the few profiting from the exploitation (largely legal, alas) of the many.
When I was in school, there was a class called "Risk and Insurance". I think I took it, because I remember the text was an Irwin book, notable for the standard Irwin livery of baby blue binding. From the point of view of the macro-economist, the point is shared risk. From the point of view of the micro-economist, it is risk avoidance (and claim payment). It's no accident that the largest criminal in The Great Recession was AIG, an insurance company. It exploited failures in law to purvey unfunded insurance.
Charpentier makes the micro-economist's argument:
If we consider two companies, one segmenting, while the other one has a flat rate, then older drivers will go to the first company (since insurance is cheaper) while younger ones will go to the second one (again, it is cheaper). The problem is that the second company implicitly hopes that older drivers will compensate the risk.The issue, of course, is that policy (the political structure) permits such Darwinist commerce. Taken to the extreme, as The Great Recession proved, capital will always seek to socialize cost and privatize profit. Production of widgets is secondary to the true goal: make capital relatively richer than labor. Being short-sighted, capital cares not for raising all boats (which provides output which is not sustainable if sold only to the .1%). In the case of insurance market segmentation, the inevitable end is transforming shared risk (insurance) into pre-paid consumption.
"Medicare for All" was a name given to an effort, failed of course, to broaden the base of health insurance. Disallowing of segmentation, arguably segmentation falls under anti-trust laws, not only broadens the base, but in doing so level the playing field, making all insurers equal. Much like Adam Smith's pin manufacturers. Not what they want, of course.
Reporting in the New York Times yesterday and today makes manifest the failure (because it's simply incapable) of quant. From yesterday we read that S&P fiddled, and intentionally fiddled the models, while today that J.P. Morgan fiddled. While quant may have intended to do the right thing (whatever that is), policy always trumps data, given the chance.
One S&P staffer, at least, tried to fight the good fight:
The idea of asking bankers what they thought about a change in the firm's methods shocked some S.& P. analysts and executives, including one who fired back, "What does 'rating implication' have to do with the search for truth? Are you implying that we might actually reject or stifle 'superior analytics' for market considerations?"Ummm, yes, yes you did.
For those who've been attempting to foist responsibility off on the GSEs, you're full of shit and always have been. This was a private capital raid on the public, and thence the taxpayer once their sandcastles collapsed. The entire mess was driven by the lust for profit, the public and even shareholders be damned. It wasn't some government agency or GSE ordering the financial services industry to inflate house prices. Yes, Dubya and Greenspan started the snowball rolling by cratering interest rates, but without the rating agencies dousing these derivatives with their holy water, the process couldn't have continued.
Accounts of the financial crisis have shown that credit analysts placed too much trust in their models. They routinely relied on historical projections that failed in an environment where home prices were falling fast.Once again, was it just the Suits, or the quants, too who fiddled while the US burned? In the words of Murphy, "admit nothing, deny everything, demand proof":
The Justice Department contends it was more than just naïveté. The suit claims that S.& P. employees deliberately used models to produce inflated, fraudulent ratings.
Rating agencies are paid to evaluate securities by the issuers. An issuer might decide not to use a rating if it were lower than those available from rivals. In that case, the credit rating agency might lose valuable fees.It was a marketing reaction, just as the non co-opted have said from the beginning:
According to the Justice Department, S.& P. started to tinker with its models relatively early in the housing boom.
S.& P. used a model called E3 to help rate certain types of C.D.O.'s. The problem, according to the Justice Department, is that the model could turn out ratings that did not meet the issuers' expectations."Guilty, guilty, guilty!" Who's more guilty, the quants or the Suits? You decide.
In such cases S.& P. had a solution, the suit said. The credit rating firm switched to a different model -- a more forgiving version called E3 Low -- that could provide more attractive ratings on certain C.D.O.'s.
The suit says S.& P. gave instructions on when to use E3 Low. "If the transaction passes E3.0, GREAT!!," the instructions said, according to the government. But if it does not, "then use E3 Low."