Wednesday, January 28, 2009

IBM podcast: building a smarter financial system

I'm interviewed on this podcast about the financial crisis and financial systems (click through to get to an embedded player). If you listen carefully, I say "regularization" instead of "regulation" at one point ;-)

Download the Building a Smarter Financial System Podcast (mp3)

I haven't written much on the financial crisis lately, because I don't feel I have anything particularly interesting to say about how we're going to solve the problem. It's a mess, and getting things back to normal is as much a psychological problem as anything else. We have to make projections about future psychological states of other people -- Keynes' beauty pageant problem -- which makes things particularly tricky. There are plenty of "experts" (especially from a particular profession) speaking and writing authoritatively about possible solutions, as if they knew with high confidence the consequences of a particular policy or plan. See this earlier post on intellectual honesty for what I think about them. See here for an excellent discussion by two honest economists, Russ Roberts and Robin Hanson, about the general dilemma of extracting "truth" from complex systems. (Comments there are worth reading as well.)

Building a Smarter Planet blog:

It can't be a good sign that complex financial topics have begun to dominate dinner-table conversations. But the dire situation in which we find our economies extend far beyond the inner circles of the finance elite.

In this episode of the Building a Smarter Planet podcast series, we focus on the financial services industry and interview Stephen Hsu, professor of physics at the University of Oregon, Jeanne Capachin, an analyst at Financial Insights, Carl Abrams, financial services business manager within IBM Research, and Keith Saxton, global director in IBM's financial markets industry. ...


Seth said...

Thanks for both podcast links. They made for good listening on my round of errands.

Your interview was very good, both the program as a whole and your contributions in particular. I hope this is the start of a long series of Steve Hsu interview appearances.

I was glad you offered phase transitions as the key missing physico-mathematical intuition that needed to be brought into the picture. That's probably the single most mathematically tractable topic economists ought to be picking up on more. The math is there, it's not hopelessly inaccessible to mathematical economists and there's at least one Nobel prize waiting for somebody who brands him or herself as "Prof. Economic Phase Transitions".

Russ Roberts' introspective look at the Problem of Method in economics (though he doesn't use that term) is 'must listen' material for anybody anywhere near economic 'research'. It was very refreshing to hear an economist waxing philosophical about what if anything they know about human economic behavior.

Anonymous said...

"have begun to dominate dinner-table conversations"

Whose dinner tables?

"But the dire situation in which we find our economies"

Who is "we"?

Economics is a pseudoscience. Astrolgers are just as useful for econmic advise.

Sabine Hossenfelder said...

I know how to solve the problem ;-)

It's very easy: Write a paper that says you have analyzed the whole issue and there is no problem, so folks, just go on with business as usual. And please stop throwing public money into the arms of those companies who are complaining the loudest, coz that's only going to make matters worse. Okay, I'm not really being serious, but I am pretty sure if there would be a convincing analysis (note that convincing doesn't mean it has to be "correct") which is able to remove the confidence problem, it would unplug the pipes. Call that a self-fulfilling prophecy.

I actually find it very scary that we are letting our lives be organized by a system whose dynamics we evidently don't understand and whose crisis we have to face helplessly. It is a mystery to me why people accept that.

For more serious suggestion of solution, see the Lightcone Institute.


Carson C. Chow said...

Hey Steve,

You're voice is starting to sound a little like E.J. Dionne's :).

What did you mean by de-regularize as opposed to de-regulate? The former seems to imply it was self-administered.

Anonymous said...

Carson, I think he was having a senior moment (as he hinted).

Anonymous said...

Frankly, I think we understand the dynamics of the economy about as well as we need to. If you play the game by certain rules, you'll do okay—maybe not as well as you'd like, but well enough.

However, if you strike out for the summit of Everest in the face of an impending storm, determined to get there no matter what, bad things will happen. That is essentially what happened here. The role of regulators is to abort the expedition, and get the team back down to a safe base camp. A lot of the team's members might be bummed out and pissed off, but no matter; they're still alive, and more or less intact.

Instead the regulators stood by and led cheers, or muttered to themselves, "let the bastards get what's coming to them," and now we've all been dragged into the muck.

(Please pardon the mixed metaphors...)

Carson C. Chow said...

OK, I thought he hinted that he meant to say it to signify something deep, like a connection to renormalization and criticality:).

Steve Hsu said...

Carson: If you find a deep connection, let me know! It just came out that way...

STS: re: phase transitions in economics, Hayek does talk about "spontaneous order" and some of the more sophisticated agent-based simulations can exhibit phase transitions. The Nobel opportunity you identified probably exists in the middle -- the realm of simple analytical models with phase transitions.

I really liked the Roberts/Hanson podcast, although I think Hanson let Russ off the hook a bit too easily -- was a bit too comforting about the status of econ.

Anonymous said...

Stephen, thanks again for participating! And sorry about keeping that regularization part in... the stream of thinking was just so good! And in general, having a fresh perspective really helped all of the thinking. Thanks!

CW, I love the expedition analogy. And from the best I can tell, you are absolutely right on the point about taking tremendous risk being a fundamental part of the problem. Having a system where people can take risks is great. But the role of regulators, it seems, should be to ensure that those huge risks are contained within the context of those who take them. In other words, I'm fine if people want to bet huge on parts of the system. But when their bets have massive implications on people or parts of the economy not culpable, regulators need to be able to manage down that risk and contain it to the individuals/institutions that take it.

Anonymous said...

I appreciate you putting this out in the open for review but the reality is so absolutely simple and we have all of these people running down a path of ridiculous notions. Scientists have some desire to "complexitize" everything. The answer is really quite simple. Consumer credit increased by 70,000% in the last thirty five years. A fractional reserve system allows us to discount future money for investment today. Not for buying LCD TVs, speculating in financial markets and building up Ponzi schemes in the housing markets. Return to lending only for business investment as was regulated before the banksters criminally lobbied our government for changes thirty years ago and we wouldn't be in this mess. Discounting future wealth to live like drunken sailors is the problem. Remove this and all of the derivatives, financial instruments and other shenanigans you and others are trying to explain are no longer in the system. It's complexity is no longer necessary. Instead we hear you and others trying to rationalize the absolutely useless complexity.

It's the forest through the trees.

Anonymous said...

reply to Albert

Which specific changes in the law did bankers "criminally" lobby for and got their wish 30 years ago?

Seth said...


You have a point that the simplest root cause is just that: Financial. Leverage. Is. Dangerous.

What part of that do we (collectively) not understand?

The potential value of a mathemtical model would be to take the conversation about how we regulate the supply of money-credit (like mass-energy, they are inseparable concepts at a deeper level) out of the realm of shouting matches and into something objective. (Now go back and listen to Russ Robert's identity crisis again, while reading Popper's "The Poverty of Historicism")

The United States went through the foundational debates about banking and money-credit back in the 1830's -- the fight about the Second Bank of the United States (see Temin's The Jacksonian Economy) is one example among several.

Why can't we collectively learn from these experiences? It was a major error of intellectual strategy for economics to lose track of the role of *power* in setting boundary conditions for economic "equilibrium".

Anonymous said...

Listened to the George Mason professors. They were VERY UNimpressive.

It is sickening to hear economists speak of low skill and high skill workers. Economists have no skill.

Anonymous said...

I think Hulk Hogan beat Andre the Giant with vaseline.

Blog Archive