Gillian Tett on the causes of the financial crisis

[nwhyte]

I took a detour from my usual intellectual pursuits at lunchtime yesterday and wandered over to my former workplace at CEPS, to hear the Financial Times journalist Gillian Tett talk about her book, Fool's Gold: How Unrestrained Greed Corrupted a Dream, Shattered Global Markets and Unleashed a Catastrophe. I was CEPS' researcher on Balkan issues

Comments

[homais]

"I have remarked before that although I work in politics, I find that it is anthropology, rather than political science, which gives me much better insights into what I am doing and more useful ideas about what to do next"

What's sad is that I agree with you, and I'm a professional political scientist (in training, anyway). We're good at some things but highly granular accounts of how systems (or towns, or businesses, or bureaucratic agencies) actually work, are not among those things. Your average political science study, at least on my side of the Atlantic, is really heavy on models and high-level explanations, but the details sort of suck. You wouldn't learn much about how anyone actually does anything by reading most of my colleagues.

Mind you, I'm from a minority faction of the discipline that is more anthropologically-minded, so it's self-serving of me to say what I just said. But there you have it.

[aliceinfinland]

I love her stuff and also heard her speak recently, as part of a panel on press coverage of the financial crisis. The problem is not, as some in that room claimed, that the press didn't warn and inform about the property bubble (exacerbated by new accounting practices) and the derivatives mess. Tett for one wrote about it regularly from at least 2005-2006. Did she or the room come to any conclusion on why the warnings weren't heeded - was it that they didn't come early enough, when the regulations were being made, or the issues weren't popular with politicians or the politicians found themselves powerless against bankers or what?

[coth]

These problems are systemic for a given (international, inter-organisational, inter-industry) scale of system. Every system/organisation solves problems local to itself, leaving any problems arising for the larger system outside of the local scope.

If you have enough perspective on the problem to give the systemic warning you are not in the class of people taking the local decisions. And if you are in the class of people taking the local decisions then the most weight you will give any warning from people outside that class will at best modify a decision taken primarily on local factors.

My experience as a business systems analyst would indicate that the majority of people making business decisions have little systems understanding, and are not generally encouraged to understand how their local systems are embedded in and affect the larger systems. This is especially true for people who work hard for years in very narrow, locally comples disciplines before becoming decision makers.

[scbutler]

"she feels that we should not describe banks as "too big to fail" but should ask if they are in fact "too big to manage"."

A brilliant insight. The arguments against bigness in capitalism are all philosophical, whereas the arguments in favor are all supported by the numbers.

[nwhyte]

Do you mean it that way round?

I summarised her argument rather brutally, though the soundbite was hers. What she was saying was that the difficulty of tracking all the relevant information in a very large bank is excessive, and we might do better in general with smaller banks.

[scbutler]

That's exactly what I mean, though I was also expanding her remark to ecompass capitalism (and I am a capitalist). Capitalism tends to oligopoly, which is just a fancy word for big and concentrated, because that's how you get more efficient and cut costs. But, as any good risk manager knows, concentration is not a good thing in the long run. Which suggests that capitalism, in order to survive in the long run, requires regulation and control.

[wwhyte]

My impression is that coming up to the crisis there was a big problem with bad information, not just between the banks and retail investors, but between the holders of bad debts (such as CDSes) and the (supposedly sophisticated) institutions that bought those bad debts. CDSes and other instruments were systematically thought to be less risky than they actually were

[nwhyte]

As far as I followed the talk (and I'll obviously have to get the book now), she placed the crucial failure point a step farther back: the bankers themselves ought to have damn well known that CDSes couldn't reliably be applied to mortgages, as indeed was the conclusion of the banking tribe she first studied at J.P. Morgan who didn't want anything to do with such a project. Her explanation for why other bankers did not realise that extending the credit market in that way was unsustainable is simple: greed

[nickbarnes]

Ah, 1986. I had a brief and hopeless crush on Gillian, and seeing her name all over the financial media these days always brings a faint smile.