How Krugman Works

From Krugman, the gist of an article, titled, How I Work:

In the course of describing my formative moment in 1978, I have already implicitly given my four basic rules for research. Let me now state them explicitly, then explain. Here are the rules:

1. Listen to the Gentiles
   
2. Question the question
   
3. Dare to be silly
   
4. Simplify, simplify

I think this is why I appreciate his insights, that they are not dogmatic, #1 and #2, and that they are clear, #4.

Analysis of Forecast Accuracy in the Political Media

Krugman, in a recent post, mentions a recent study comparing the accuracy of various pundits.

Conclusion:

Our findings seem to agree with Tetlock’s research. Some studies suggest that conservatives have more rigid ideologies (Jost et al., 2003) In other words, they would be considered “hedgehogs.” Similarly, lawyers are taught to argue one side with a single analytical method; they, too, would be “hedgehogs” under Tetlock’s model. While not all liberals are foxes and not all conservatives are hedgehogs, these trends may be informative in explaining why our results are as they are. It may be that conservatives are inherently disadvantaged as prognosticators due to their ideological “hedgehoginess.”

Perhaps most importantly, being a good prognosticator seems to be a product of choices, not birth. Anyone can be good; all they need to do is avoid law school and buy into liberalism as an overarching philosophy. There is no inferior ability associated with being born, say, black or female.

Mind Changing Events, ala Paul Krugman

..thinking like a fox.

If one wants to be correct in one's predictions and general models, one needs to be willing to change one's opinions. All around in American culture, we see the veneration of single-minded 'hedgehogs', that are the most strident, and no matter how wrong, that explain away their mistaken judgments, but never consider their model wrong.

I've sometime worried about you, Mr. Krugman, becoming corrupted by media exposure, since the more exposure the more incorrect theorists become, having to spend so much time defending and selling their ideas to an America that likes absolute opinions delivered with absolutely certainty.

Cheers to you, and hoping you stay smart, like a fox.

Financial Reform

This morning, I read Volcker's editorial in the NY Times, and with the reforms he thinks of putting in place - he advocates more control over banks than most of those with any say - the banks "would be free to to innovate, to trade, to speculate, to manage private pools of capital...". My sense is that trading can't go backwards, and that the use derivatives and securitization will continue unabated, albeit correlated with ups and downs in market activity.

Near-Term Changes in Finance

- Even if an equivalent of Glass-Steagall was put in place, little would change, since the money would simply move to hedge funds and private equity firms. A driver for many of the changes was an increasing unequal society, and there is little chance of it becoming more equitable any times soon.

- I assumed that Obama's policies would be 'Better-Same' than Republicans, meaning generally better for people, but with the same entrenched interests. His presidency will not change very much in the US political landscape, and the same corporate interests will be served, for the most part unabated, including finance.

- Being in banking technology, the only significant changes in banking I've seen, and these are not transformative, are algorithmic trading and an intensification of risk systems. One of the big change drivers in banking/finance is technology, such that more and more activities are handled electronically and by technology.

- The middle office, notably a morass of failed trades that need to be settled, and there are large numbers of them, is being moved offsite and overseas, as well as automated. As previously mentioned, the CTB/RTB division had as a component Six Sigma, looking for ways to prevent failed trades, largely by designing better systems. An obvious change likely to occur is the outsourcing of more jobs to Asia, and not just India, as well as to locations within the US, outside of New York.

- Geitner is from the same entrenched groups, and will possibly go to Goldman's after his tenure, and there is talk of him leaving soon.

A Response to Twilight of the Polymaths (NY Times)

Since you are aware of the fox and hedgehog dichotomy, you might be aware of Tetlock’s book and long-term study showing how foxes best hedgehogs in political prediction, Expert Political Judgment: How Good Is It? How Can We Know?

Although depth in specialties has its merits, I can see how that depth also leads to a kind of blindness, a ‘man with a hammer’ problem, while polymaths, or people with broad awareness can sometimes pull together diverse facts to form insights that are not bound by one’s expertise.

Along the same lines, many insights in technology don’t come so much from depth, but by ‘cross-pollination’ of ideas, of taking the basic concepts of one specialty and applying them to a peripherally-related one.

The Illusion of Certainty

This article goes to the point that the quants were part of the problem, which I don't disagree with, it's just how much they can be accountable for:

Wall Street’s Math Wizards Forgot a Few Variables

During the internet bubble, I read a book by a Yale professor, Schiller - he's had a company for the past few years that tracks the RE market - that was titled Irrational Exuberance. Schiller was credited with saying the markets had bubbled and would implode, and his academic analysis was based on market mania, behavioral economics. It was obviously true at the time, but I remember that a finance professor at Fordham didn't get around to reading it until 2 years later.

The problem with numbers, with quant analysis, I think, is the illusion of certainty that comes from mathematical analysis. It often gives a definite number to things that are somewhat fluid, like probability.