And Some Random Thoughts
Regarding “Basel: Faulty,” what further empirical evidence would cause you to question your conclusions? What might make you more confident?
Answers and thoughts: Most importantly, the balance sheet averages I employ for European banks are data I assembled myself from publicly available data. The original source was mostly Bloomberg. I am confident that they mirror reality reasonably well, but it’s possible that they do not precisely reflect the actual aggregates. The BIS has a database containing data that could corroborate my arguments. Unfortunately, I found this database impenetrable. Partly, this was an issue of terminology. I would be most grateful to anyone who could help me access the BIS data or supply the relevant data.
As I point out in the post, much of the asset growth for European banks from 2000 to 2007 was attributable to derivatives. Unfortunately, derivatives accounting was, then at least, a fraught discipline. For instance, when we see UBS’ derivatives book grow from E26 billion in 2000 to E450 billion in 2007, it is not necessarily the case that there has been a commensurate increase in risk to the UBS balance sheet. But it is also possible that the underlying risk was greater, perhaps much greater, than the balance sheet reflects. Valuations were hazy and largely a function of management discretion. Given the euphoria that characterized the mid 2000’s and the license that Basel II provided in valuing derivatives, I think it is unlikely that the actual risk was less than that presented on the balance sheet.
One characteristic of most derivatives is that if one is short (as most banks tend to be) and unhedged, one’s exposure will snowball as the market moves against you. This effect will be compounded if doubts arise about counter party solvency and even more so if potential buyers decide they don’t really understand the instruments. Speculators may also “pile on” the trade if they believe an institution is in trouble. So whatever the exposures might have been on paper at a given balance sheet date, those exposures were certainly far larger in the depths of the crisis.
Remember that by the time the crisis hit, quants and traders were — effectively — running the banks: the inmates were running the asylum. Everyone deferred to them because anyone with the temerity to argue with their math got laughed out of the room. There are innumerable examples of risk managers being fired for questioning the growing exposures. Management, too, was intimidated and, in any case, the quants were making so much money for everyone that management could not afford to reign them in. No one could afford to kill the golden goose. That doesn’t absolve senior management — quite the contrary — but it does help explain how it all came about.
There is a fascinating post-mortem authored by the Swiss bank regulator concerning the losses at UBS. It goes into some depth describing the groupthink that contributed to management’s capitulation to the traders.
Ironically, but not at all surprisingly, the elaborate mathematical models spewed out by the quants during the good times turned out to be as bogus as the BIS “risk ratings” that caused the whole mess. I remember attending a bank analyst conference in, like, November, 2007 when one of the presenters tried to explain CDO’s and CDO squared. Everyone’s jaws just dropped. No one in the audience had ever heard about them and no one could imagine anything so stupid. Also, I remember being astounded to discover that the principal risk model employed by many banks was “Value at Risk.’ I had thought that VAR had been consigned to the dustbin of history after the collapse of Long Term Capital in 1998.
Partly, the ascendancy of quants and their models was due to the length of the “great moderation” that preceded the crash. Everyone was lulled into a sense of false security by the stability of the markets, most notoriously Alan Greenspan (I believe that it was the continued leveraging dictated by Basel I that produced this stability.) Of course, if everyone had read their Minsky, they would have known that an extended period of stability is just a rubber band waiting to snap. (See Bob Barbera’s excellent book “The Cost of Capitalism.”)
Things might not have gotten so out of hand absent the then prevailing devotion to free market Gospel. In his superb book “How Markets Fail,” John Cassidy details how “efficient market theory” (CAPM) came to dominate finance. CAPM reinforced the myth that no asset –derivatives included — could possibly be more risky than it was priced to be. Anyone who disputed this orthodoxy was dismissed as a crank ( e.g. Hyman Minsky and Benoit Mandelbrot. )
Remember too, that this didn’t all happen suddenly. It’s not as if one day all these managements decided to bet their banks on sub prime real estate derivatives. It was an incremental process. Quarter by quarter, management was under pressure to produce the revenue and profits to beat prior quarters. They took on more leverage and increasingly esoteric derivatives to achieve this because they were just repeating what they had successfully done in the past. The incremental risk was pretty much invisible to them, and, anyway, it had all been blessed by the regulators.
To convince me that my “Basel” hypothesis is false, someone needs to identify an alternative source of the cash that flooded into the mortgage market and other credit markets from 2000 to 2007. One possibility is foreign countries. Ben Bernanke once remarked that losses from the sub-prime “bubble” by themselves could not explain the severity of the 2008 crisis. He invoked the “savings glut’ as an explanation. Ben may have almost single-handedly saved western civilization, but I think that the “savings glut” thesis is unlikely.
“Savings glut” is simply another expression for the immense US balance of payments deficit that built up from in the early 2000’s. By 2007, our cumulative trade-related debt totaled nearly $8 trillion. This helps explain why most of the speculation took place in dollar based assets. However, a payments imbalance such as this is essentially a zero sum proposition that creates no NET worldwide demand for credit. Lenders of dollars are borrowing other currencies, usually their own. I have seen no evidence that the largest debt holders – China, Japan, Canada – and their banks were significant buyers of subprime instruments. Germany was to a small extent.
However, it just seems logical to me that the immense US trade deficit and commensurate payments imbalance must have played some role in the crisis. I believe that the connection is that surplus countries were a source of demand for the short term low rate dollar instruments (repo and ABCP) that provided the funding for European bank and shadow bank activity.
The advent of the Euro in 2000 also contributed to the credit bubble, especially within the Eurozone. Prior to 2000, a German bank that wanted to lend to a Greek importer not only had to consider the credit of the borrower, but had to consider how easily that borrower could obtain Deutsche Marks in the future in exchange for his Drachma, a much less liquid currency. The Euro eliminated this constraint, opening the floodgates to credit to the periphery.
It is not clear how significant an influence Basel II had on bank risk profiles and banker behavior, but it certainly heightened risk-taking. Basel II was not officially promulgated until 2004 and was not adopted by many European banks until 2008. It was never officially adopted by US banks. However, the broad outlines of Basel II had been apparent since at least 1999. So banks had been adjusting their business models in anticipation of Basel II for at least 5 years. Driven by “efficient market” dogma, the centerpiece of this regulation was the A-IRB (Advanced Internal Ratings Based) approach, which relied on the banks themselves to provide risk models for their own portfolios. As a quid pro quo for bringing banks on board, it was widely recognized that bank capital ratios would be allowed to drop (see Dan Tarullo’s 2008 book Banking on Basel.)