Archegos and marking academic literature to reality

Last week, Bill Hwang’s family office, Archegos, imploded as it was unable to meet the margin calls emanating from steep declines in the prices of stocks that Hwang had bought with huge leverage. Mark to market is a very powerful discipline that spares nobody however rich or powerful. This ruthless discipline makes financial markets self-correcting unlike many other social institutions.

Academic literature in particular is much more insulated from the discipline of mark to reality. Old papers discredited by subsequent developments or even subsequent research continue to be cited and quoted (this is the replication crisis in economics and finance). To borrow accounting terminology, the academic community tends to carry the old literature at historical cost without sufficiently stringent periodic impairment tests.

There is a large stream of finance and accounting literature which is probably badly impaired by last week’s developments. I refer to the literature that uses percentage of institutional shareholding in a company as a proxy for various things including corporate governance. What we are learning now is that Archegos used over the counter derivatives like swaps and contracts for differences to invest in a range of companies with very high leverage. The banks who sold these derivatives to Archegos bought shares in the companies to hedge the derivatives that they had sold. The shareholding pattern of these companies would then show the Archegos counterparties (banks) as the principal shareholders of these companies though in economic terms, the real owner of the shares was Archegos. Media reports suggest that this includes companies which were targeted by short sellers (and presumably had corporate governance concerns).

In the case of these companies with possibly dubious corporate governance, academics and investors might have been reassured on observing that say two-thirds of the shares were owned by institutions without realizing that much of the holding was the family office of a person who had committed insider trading. I think this is another illustration of Goodhart’s law: “Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes.” The lesson that the academic literature must learn from that law is that the longer established a proxy measure is, the more ruthlessly one must apply an impairment test and mark it to reality.

Posted at 9:02 pm IST on Wed, 31 Mar 2021         permanent link

Comments