No, not Heidi Klum models, but economic models and what they tell us, or more importantly, don’t tell us about economic reality.
There’s an old saying that goes like this:
In theory there is no difference between theory and practice. In practice there is.
It’s been variously attributed to Albert Einstein, computer scientist Jan L. A. van de Snepscheut and, are you ready for this? Yogi Bera! Personally, I’m pulling for Yogi myself. It’s totally something he would say. Whomever is responsible for this clever observation, it’s a delightful turn of phrase that makes explicit the risk of presuming your models for the way things work are actually the way things work.
It sure would have been nice if the investment bankers back in 2005 understood this when they started carving up mortgages into smaller and smaller chunks to collateralize them. Yves Smith has an entire book, ECONned, dedicated to the catastrophic failures in the modeling skills of financial economists.
It was a theory posited by a over-smart “quant” names David Li. In 2000, he published an article in the Journal of Fixed Income where he identified a new formula to model correlations, that is the impact the movement of one variable has on another variable. This was a “major breakthrough” and supplanted older, more conservative formulas.
But the entire premise was a fraud. Any approach that tries to model correlation has to assume that the relationships will behave in a predictable way. It is one thing to use past patterns as a heuristic (“the dollar and gold tend to move in opposite directions”) and be aware that the relationship is far from foolproof, and quite another to build elaborate pricing formulas and products on a rotten foundation. Yet the financial services industry did just that, even though Li himself cautioned against undue reliance on his new formula.
Li’s model played a central role in the crisis. It appeared to offer a way to estimate the odds of default in very large groups of securities. This is a crucial input in assessing the riskiness of bonds that are based on pools of loans, such as mortgages and credit cards. And the trick with this model was that rather than relying on historical data, it used credit default swaps (CDS) to estimate the risk that borrowers will fail to pay….
The formula was wildly popular because it appeard to offer a way to price default risk in markets with little historical data, like subprime mortgages. But the logic was hopelessly flawed.
Remember CDOs (Collateralized Debt Obligations)? Remember CDSs (Credit Default Swaps)? AIG? Lehman Brothers? Bear Sterns? Those guys? The smartest guys in the room turned out to be the ones blinded by their own belief in their collective brilliance. They assumed that their models of reality were reality. And because of that, they burned the global economy to the ground without any apparent consequences to themselves.
So, gentle reader, when you have an unregulated market like CDOs and CDSs that goes belly-up, is the right answer less regulation or more regulation? If you ask the GOPers, the answer is always less. And why? Because that’s what the industry wants them to say. The 99%, though, they’re onto this game. They know the answer is, and always will be, more regulation of the financial industry. And let’s be clear. These guys are not job creators.
As economist John Kay opined today,
Models are often useful in illuminating complex problems and quantification is an essential part of decision making. But good models are simplifications, not black boxes whose workings are incomprehensible even to their operators. The relevant model is always specific to the task at hand and there is no objective method of determining the right tool to employ in any particular case. If you do not know the answer to a question, the right response is not to make a number up, but to rethink and frame an alternative question that is capable of being answered.
We do great damage by claiming to know things that are not known, by asserting certainty in the face of uncertainty and ambiguity, and by attaching a veneer of rationality to decisions that have in fact been made on other, rarely articulated, grounds. The paradoxical result is all too obvious. The public sector and large bureaucratic organisations appear as paragons of good decision making process and exemplars of bad decisions. (emphasis added)
But back to the question of models of reality and the reality of models. Let’s face it. Like the old King of Swamp Castle in Monty Python and the Holy Grail, if you build your castle in the swamp, don’t be surprise when it burns down, falls over and then sinks into the swamp. Economic models that are built on the same swamp will eventually do the same thing. Expecting otherwise is insane. Unfortunately, when that castle sinks, we all sink with it.
And that, my friends, is reality.