"Striking it Richer," a paper by Emmanuel Saez (an economist at UC Berkeley) looks at the way that the dividends of the slow US "economic recovery" have been distributed. Saez finds that 121% of the economic gains since 2009 have been captured by the richest 1% of Americans -- in other words, despite economic growth, the poorest 99% of Americans actually got poorer through the "recovery."
Read the rest
This confirms a pattern that Matt Stoller highlighted: that income inequality increased more under Obama than under Bush. And the new Saez paper also describes how it came about. In short form, income to the top 1% is significantly influenced by capital gains. Remember, the tax reporting is not clean here: rising equity and bond markets help all those private equity and hedge fund professionals, who are able to get capital gains treatment for what ought to be labor income. But the paper also stresses that the lower orders were hit hard in the aftermath of the global financial crisis than in the dot-bomb era, which also saw a big drop in capital gains. That isn’t as hard to understand. The collapse of the dot-com mania didn’t impair the real economy overmuch because it was not fueled in a meaningful way by borrowings. By contrast, the housing bubble, and more important (in terms of damage to the financial system) the much housing exposure created synthetically by CDOs that consisted entirely or mainly of credit default swaps was highly geared, hence when it collapsed, it took credit providers down with it.
"Computational Complexity and Information Asymmetry in Financial Products," a new paper by Princeton computer scientists and economists Sanjeev Arora, Boaz Barak, Markus Brunnermeier, and Rong Ge suggests that complex financial derivatives are computationally intractable: that is, once you have mixed together a bunch of weird-ass securities and derivatives, you literally can't tell if the resulting security is being tampered with as it pays off (or doesn't). Freedom to Tinker's Andrew Appel likens it to cryptography: you can mix together a bunch of known quantities to get a new number that can't be turned back into the old numbers.
Read the rest
The paper shows the example of a high-volume seller who builds 1000 CDOs from 1000 assert-classes of home mortages. Suppose the seller knows that a few of those asset classes are "lemons" that won't pay off. The seller is supposed to randomly distribute the asset classes into the CDOs; this minimizes the risk for the buyer, because there's only a small chance that any one CDO has more than a few lemons. But the seller can "tamper" with the CDOs by putting most of the lemons in just a few of the CDOs. This has an enormous effect on the senior tranches of those tampered CDOs.
In principle, an alert buyer can detect tampering even if he doesn't know which asset classes are the lemons: he simply examines all 1000 CDOs and looks for a suspicious overrepresentation of some of the asset classes in some of the CDOs. What Arora et al. show is that is an NP-complete problem ("densest subgraph").