A Canadian on Wall Street

Thinking I had cleverly coined a new term, I did a search on Google for it and discovered this interesting, somewhat academic paper from Brad DeLong and co. I like the concept of “noise trader risk” when trying to fade the noise trader.

I was at a public library sale the other day and picked up a few decent investing related books for a buck a piece — one on commodity futures trading from the 60s, and the other a decent overview of the strategies and issues of Long Term Capital Management, aka LTCM, aka the fund that last broke the financial system until a huge round of bailouts. Basically, like Ecclesiastes says, ain’t nothing new under the sun. I’m paraphrasing.

One of the things that was really clear from the latter book was that LTCM was using massive amounts of leverage to get their returns. They’d borrow money very cheaply, and throw it behind trades that would normally return about 2% or so, but due to the scale of their bets, they’d get a massive return on equity (if a piddling return on assets). It’s only when those “sure bets” started to fail — once again, the problem of induction rearing its ugly head, and in fact, this book went into a fair bit of detail on it for a book from 2000, before the recent success of the Black Swan — that everything went to pot.

This is essentially how hedge funds do their business, and make their spectacular returns, and equally gigantic flameouts. It really makes you wonder if the party line being sold to people, that you just sock away your money in a well-diversified fund and cross your fingers that everything does about the same as it always has (for the last hundred years), is really all that effective. It works until it doesn’t. I just find it asinine how people will recommend this strategy over learning about individual stocks, like it’s getting your teeth cleaned or grocery shopping — I don’t think it’s ever really possible to make it as staid as all that.

In any case, it also reminds me of ideal betting strategy, except in the case of blackjack you tend to have a better sense of the odds (as some people have described it, the known unknown) versus investing (the unknown unknown) and can feel comfortable raising your bet accordingly. Of course, this’ll get you thrown out of the casino if you do it enough times and start winning big, and in investing it’s all but impossible to get that level of comfort, without applying a little judicious ignorance. Sort of like creating a “pseudo-known unknown” by having a myriad of blind spots.

Interesting article about dark pools from Reuters, regarding the rise of off-exchange trading and the difficulties this causes in determining a single price for anything. Although the article indicates the average order size has gone down to 250 shares versus 1500 in the past ten years and seems to believe this is a negative, I wonder if that’s the case? Given the increasing actual share volume, wouldn’t this just mean larger investors can’t move the market as easily with their trades? I suppose it would lead to more so-called efficiency, though, so I guess it’s possible that it’s more difficult to make money. Less liquidity leads to bigger price spreads leads to more potential for profit — this is what I’d call a systemic method of investing or perhaps meta-investing, i.e. taking advantage of the structure or system’s characteristics to make money, a la market-makers. Reminds me a little like John Turturro’s character in Rounders, wanting to make a lot of little wins versus the big score that Matt Damon makes by taking down John Malkovich’s character, the grinder lifestyle. This is particularly interesting:

“If a retail customer puts in an order that is very well priced, and it becomes the best price available, all of a sudden, all these investors in dark pools have to trade off of that price,” said Bernard Donefer, Associate Director of the Subotnick Financial Services Center at Baruch College in New York. “The national best bid offer could be created by small retail investors. It used to be the opposite. It used to be that the institutional investors set the price and the retail investors got that price.”

It just occurred to me today that if you think of stock market fluctuations being in the same vein as weather patterns — varying at the micro level, somewhat gradual and predictable at a short-term level and almost totally unpredictable the further out you go in time — it puts a different perspective on a lot of things. Trying to determine future price movements based on a large aggregation of historical data might be akin to trying to predict future weather by looking at when it last rained, which seems silly as there’s zero precision to this even if there were some kind of general cycling. Daytrading might be a little like trying to figure out whether the temperature would be a degree higher or lower at any given moment, or the precise windspeed or direction. In both, there are elements of feedback at play that can damp down or bubble up at any moment. And there’s the same lack of catastrophe scenario analysis. Look at the way Hurricane Katrina was predicted, its heading well known, and yet people did not evacuate in time; could the same be said for the collapse of the US mortgage market and its subsequence knock-on effects?

Maybe this is all a bit tenuous, but it certainly gives me more of an ability to be sanguine about my investing when I can view short-term market fluctuations in the same way I might view weather, as mostly random.

On a note related more to the last entry — I wonder if there’s any evidence of investors who’ve had no systematic method of stock picking doing well over long periods of time. In one year time frames, darts have often easily outperformed professionals, but does there always have to an analytical, articulated methodology of stock picking? I still wonder if anyone would admit that they used an intuitive or hunch-based method or if, in the process of success, any method like this would be retroactively explained by some convenient, well-fit “system”.

1. I’ve started a paper trading journal, simply to document the reasoning behind any moves I make in my portfolio. It’s a surprisingly good way to have to articulate a position: what’s the strategy, what motivated it. Quite often, it’s initiated by some reading or discussion I’ve done. To be honest, much of my investing is not truly quantitative or based on fundamental analysis, even though I’m steadily hoarding more and more unread “Analysis of Financial Statements” textbooks, in the event that flammable materials become scarce. It’s almost talismanic, my need to pore over these strategies and then not actually apply it to most of investing. And in retrospect, some of my biggest scores have been almost completely random, or motivated by the flimsiest of rationales. I’ve been reading Mauboussin’s book More Than You Know, a collection of some of his old essays, and there’s a good quote about how it’s possible to do well investing even if you don’t adopt a smart methodology, but that the odds are against you over time, so it’s quite possible that if I don’t shape up (hence the trading journal) I could be a big ol’ washout.

2. Not really related to investing, but sort of on the “I’m a cheapskate” tip, someone really needs to start a “cheap wine blog” based out of British Columbia. I’m not that someone, but I’m cheering you on, nameless potential wineblogger! I’m enjoying stuff like this blog, but frankly when you’re sticking to the BC Liquor Store, you’ve got a limited selection to work with.

More tidbits from Wall Street:

The point of this little review is not just to embarrass official wisdom, though certainly that is always fun, but to undermine confidence in the entire enterprise of conventional mathematized economics. And few subfields are as math-dense as finance. A lot of neat theories grew up in the 1950s, 1960s, and 1970s, only to be challenged by some neater theories in the 1980s and 1990s, but the entire project of clever, influential, and largely empty theorizing about capital markets and the invisible hand has yet to be severely questioned. Even the extensive empirical work by a number of financial economists, often based on thousands, even millions, of data points, fails to provide any significant enlightenment, because it asks such self-contained, even puerile, questions.

As Leontief argued — in the presidential address to the AEA given 20 years before Debreu’s — that self-contained quality is a significant reason for the failings of econometric analysis.

The same well-known sets of figures are used again and again in all possible combinations to pit different theoretical models against each other in formal statistical combat. For obvious reasons a decision is reached in most cases not by a knock-out, but by a few points. The orderly and systematic nature of the entire procedure generates a feeling of comfortable self-sufficiency.

Real progress, he continued, would require crossing disciplinary boundaries — towards engineering (or, he might have added, industrial organization) to understand the process of production, or anthropology and demographics (or psychology) to understand consumption. But in the hermetic world of conventional analysis, prices are explained with regard to other prices, output with regard to other outputs, and so on — a circular, almost onanistic process of analysis. The interesting stuff — OPEC’s rise in the 1970s, for example, and its fall in the 1980s — is typically relegated to the realm of “exogenous shocks.” Some theories of the business cycle explain recessions as purely exogenous phenomena — an intellectual convenience, since it allows the theorist to avoid the labor of explaining booms and busts, and an ideological one as well, since it exonerates market processes themselves as the source of instability.

Studying economics also seems to make you a nastier person. Psychological studies have shown that economics graduate students are more likely to “free ride” — shirk contributions to an experimental “public goods” account in the pursuit of higher private returns — than the general public. Economists also are less generous than other academics in charitable giving. Undergraduate economics majors are more likely to defect in the classic prisoner’s dilemma game than are other majors. And on other tests, students grow less honest — expressing less of a tendency, for example, to return found money — after studying economics, but not after studying a control subject like astronomy (Frank, Gilovich, and Regan 1993).

I’m reading an out-of-print book recommended by Brad DeLong called Wall Street by Doug Henwood, published in 1997. Some parts seem eerily prescient, or maybe it’s just that same old ish keeps happening again and again:

But it’s wrong to blame only the government, despite the American habit of doing so. Virtually every high-end profession around was involved (a point made well by Martin Mayer [1990]). Auditors repeatedly certified fictitious financial statements, lawyers argued on behalf of con artists and incompetents, investment banks bilked naïve S&L managers, and consultants testified as character witnesses for felons. One of these character witnesses was Alan Greenspan, then an undistinguished economist from whom “you could order the opinion you needed” (Mayer 1990). Greenspan praised thrift-killer Charles Keating’s “seasoned and expert” management team for rescuing a “badly burdened” thrift through “sound and profitable” investments. Every word of this was untrue. Greenspan’s reputation, however, survived intact (just as it did his earlier demented jottings for Ayn Rand’s Objectivist newsletter).

In its infinite generosity, Washington came to the rescue. Of course it had no choice; no modern government would dare let a financial crisis turn into a general collapse. Yet the situation is rich with irony. In the early 1990s, Greenspan would craft the Federal Reserve’s bailout of the 1980s mania. And the braindead caretaker administration of George Bush crafted the greatest socialization of private loss in history, the S&L bailout. And, remarkably, almost nobody has suffered serious criminal penalties or political disgrace for this rampant abuse of trust. Huge quantities of public money — some $200 billion, though definitive accountings are hard to come by — were spent with little discussion or analysis, and the affair is now largely forgotten. The chance to use the industry’s partial liquidation as an opportunity to develop new public and cooperative financial institutions was blown. Within a couple of years of the crisis’ passing, no one paid it any mind any longer. It’s as if it never happened.

…

Sluttish accountancy is a long-standing practice. During the 1960s, John Brooks (1973, pp. 160–162) reported, accountants painted a glowing picture in numbers that merger promoters wanted. Investors, many of them naïfs, eagerly bought paper that only a few years later would turn out to be near worthless. “By following conservative practices and their consciences, accountants could have prevented this jiggery-pokery; they did not.” You could say exactly the same thing about the financial disasters of the last 15 years. Had Wall Street analysts, who are presumably competent to do so, given the accountants’ numbers a serious second look, much recent jiggery-pokery might have been detected.