Byrne's Eye View

It occurs to me that the same people who said capitalism was exploitative are the ones who push universities to divest their assets in South Africa (during Apartheid), Israel, and the like. If they really think capitalism is oppressive and unfair, shouldn't they demand that Harvard invest all of its endowment in oppressive countries?

A SocGen trader making less than 100,000 Euros a year dreamed big and bet bigger, and now SocGen has announced a loss of $7 billion on his unauthorized trades. I have a question: has anyone, ever, issued a press release like this: "A rogue trader at BancCo has been fired for exceeding his position size limits. He made the firm $100 million last quarter, and is the only reason they beat their estimates."

It reminds me of a point about sexual harassment: that nobody has ever been prosecuted when it turned out that his advances were wanted, after all. There should be a term for doing something wrong on average that doesn't get you punished if you're lucky enough to accidentally benefit the party you'd usually harm. For now, I'm calling them Victimish Crimes.

Michael Lewis tries to analyze Goldman's giant subprime bet (helpfully, he ignores other companies that made similar bets, so he can write a story about how Goldman was unique rather than uncommon). Really, I don't see what's special about this situation, other than the scale: if everyone agreed that there was a 95% probability that mortgages would pay off, we'd either see 95% of traders make a little money or 5% make a lot of money.

But what's crazy about Lewis is not that he has to be dishonest to tell the story he has in mind: it's that his story is so dull. According to Lewis, the problem with Goldman Sachs is that a company with 30,000 people, most hired for their smarts, don't all agree all the time.

Alan Greenspan has joined Paulson & Co., which made an insane amount of money recently betting on a collapse in mortgage prices. He probably won't end up with the same net worth as Mr. Paulson, though.

It's probably bond trading 101 that if a country's capital disappears into a mushroom cloud, that country's bonds will drop in value. This is not necessarily true: distressed-debt traders are buying North Korean bonds (defaulted in 1972, price at 32 cents on the dollar). This seems to be a case in which the probability of repayment is so low that the only thing they're trading on is volatility: North Korea is not just poor -- it's stable -- so a bull market in their bonds either indicates that they're expected to get closer to South Korea or that they're expected to get closer to the kind of chaos that would allow South Korea (or China, or Russia, or an internal Junta) to swoop in and reform things.

If Amazon.com patents the "One-more-click purchase", will they have a monopoly by induction?

I'm a proponent of prediction markets. It seems obvious to me that if you want to know what smart people know, the way to do it is to pay them in proportion to their ability to make good predictions. We already do this in an ad hoc sense -- people who make the right investments and career choices make more money -- but formalizing it can't hurt. Most of the people who discuss this seem opposed to prediction markets, though. Fortunately, I can usually address their claims on my terms:

1. Prediction markets are too easy to manipulate seems to mean "I could consistently make money betting against suspiciously high-priced 'establishment' candidates on political betting markets. This would bring prices closer to reality, and make me richer, too!"


2. Prediction markets aren't liquid enough to be accurate implies "It would be profitable to be a market-maker in a prediction market, because the bid/ask spread is so high."


3. They give really weird results says "Despite these high spreads, there are arbitrage opportunities."¹


4. Prediction markets are less accurate than polls also means "I can bet that prediction market results will revert to poll results, and automatically make money."


5. Prediction markets are only as accurate as polls, on average means "I don't understand sampling, and think that two measures of the same variable with the same accuracy, when averaged together, are just as accurate as one. I can use this to empirically prove that you can always survey one less person in your polls, and still be just as accurate."


6. Prediction markets will help terrorists make money from assassinations implies that "terrorists would rather make $5,000 dollars on a 90-point jump in an assassination contract than many millions of dollars using derivatives to short the broader market in advance of market-disrupting action." You'd need a multi-billion dollar prediction market in airline-based terrorism for it to make more sense to bet on the prediction market rather than buying puts on airlines before 9/11, for example.

That covers it, I think. On the other hand: Daniel Dennett can explain Christianity as a Darwinian phenomenon involving the cost of acquiring new information versus mimicking people around you. His opponents can explain him as a religious phenomenon involving the necessity of doubt to make faith meaningful, or the temptations of a materialist outlook. In other words, I'm sure there's an anti-PM system of thought in which my examples are nonsense. I just have no idea what it might be: in every example, I've explained how a market skeptic who really believes in market skepticism can make money at the expense of radical, fundamentalist free-market demagogues like me. Is there a counterargument?

[1] The Al Gore arbitrage mentioned in that article is over: whoever shorted the Gore-President and bought the Gore-Nomination made a low-risk profit.

... if they charge passengers per mile but reimburse them per minute.

Drivers get more reckless when they have to wear seatbelts, because they feel safer. Depending on which data you use, the resulting change in total accidents either underwhelms or swamps the safety effects of seatbelts. This works the other way, too: on icy roads, drivers slow down and drive cautiously, drastically reducing fatalities. This leads to an obvious conclusion: the way to make driving safer is not to mandate safe behavior, but to mandate behavior that feels as unsafe as possible compared to how safe it is. If people think that driving with an eyepatch makes them 300% more likely to get in an accident, but it really only makes them 10% more likely, mandating eyepatches makes driving much safer.

Now, consider: people don't judge lawmakers based on how effective laws are, but on how effective they seem. So lawmakers have no reason to pass a law that, as far as their constituents know, makes driving 300% more fatal. In fact, they have exactly the opposite incentive: to pass the laws that appear the safest, but that cause more accidents so there's a further need for safety-generating legislation.

Public elections are perfectly designed to produce what is literally the worst law imaginable. QED.

Today I walked past the Bear Stearns headquarters, and noticed that they have a gaudy Christmas wreath in the lobby. I'm wondering how to interpret this indicator, which is more fine-grained and timely than the Beige Book. My guess is that it's positive: it means they can afford to have a minion put up the wreath without really thinking that this is not the thing to do after losing a few billion dollars on mortgages. Or it's negative: some slightly higher-up Bear Stearns executive has nothing better to do than make lobby-decorating decisions. And since the wreath is still up, both theses are valid if inverted: are they too poor to pay someone to take it down, or too busy to get around to it? Econometrics is a tough subject.

In other news, Barnes & Noble has hired a greeter for their Fifth Avenue location. Is it a bullish indicator that a major player in a highly competitive industry can afford that kind of frivolous expense -- or should I read a recession into the fact that Manhattanites are being treated like Wal-Mart shoppers?