Re: People are talking about economics today for some reason

1

I don't think it's quite fair, but it is largely true, with the idea being that investors cannot beat the market and in the long run will do no better or worse than the market. A sort of fairy tale for young economists.

But isn't a proof by contradiction a proof where you start with the opposite of what you want to assume and then argue it to its logical conclusion, which proves that it is impossible and therefore your original assumptium is correct?

So in this example you would argue that if markets weren't efficient, then that means there are firms that are way overvalued compared to their assets and argue from there that black is white.


Posted by: Martin Wisse | Link to this comment | 10-14-08 12:18 AM
horizontal rule
2

It's a "proof by equilibrium". If prices don't reflect the underlying fundamentals, then the market is out of equilibrium, and speculators will act to move it back towards equilibrium. Economists are hypnotized by equilibrium.


Posted by: Walt Someguy | Link to this comment | 10-14-08 12:23 AM
horizontal rule
3
No one would buy an argument to the conclusion that there are no hitchhikers that proceeded by hypothesizing the existence of a hitchhiker, and then asserting that surely a kind soul will give him or her a ride pretty soon.

Is this disproof by... analogy? Oh, Ben.

So is there no way to define a proof by contradiction such that the disproof of the contradictory case is displaced in time? It seems like there must be a way to do this.

That said, the argument as stated seems like it must be a simplification, because I can think of mathematical ways you could approximate the displaced proof as I mention above.

That said, it does seem like this hypothesis, even if true, leaves room for some fairly massive instability in things in the short term, which happily seems to match the reality.


Posted by: Sifu Tweety | Link to this comment | 10-14-08 12:24 AM
horizontal rule
4

Pwned, and so succinctly. That's it, Walt, Lakoff was right.


Posted by: Sifu Tweety | Link to this comment | 10-14-08 12:25 AM
horizontal rule
5

Lakoff is a Komodo dragon, the largest member of the lizard family, and a filthy liar.


Posted by: Walt Someguy | Link to this comment | 10-14-08 12:28 AM
horizontal rule
6

I think that I might have been an economist but for the honest to God truth that I found a hundred dollar bill on the ground on the way to school in the fifth grade (me, not the bill, you pedants). That experience rather soured me on the whole conceit.

The analogy(!) in the OP is inapt. A better one would be: there are no hitchhikers because surely a murderous soul will come across them and drag them to their dungeon lair before anyone else can. After a bit, the supply of willing hitchhikers will dry up.


Posted by: foolishmortal | Link to this comment | 10-14-08 12:28 AM
horizontal rule
7

The efficient market hypothesis is by its nature not susceptible to deductive proof, by contradiction or otherwise. There is a mountain of empirical evidence for the so-called weak form of the efficient market hypothesis, but the so-called strong form is controversial. I delegate further explication of the wherefores to PMP.


Posted by: Knecht Ruprecht | Link to this comment | 10-14-08 12:30 AM
horizontal rule
8

||
New crush: Bethany McLean of Vanity Fair
|>


Posted by: Otto von Bisquick | Link to this comment | 10-14-08 12:36 AM
horizontal rule
9

There is a mountain of evidence for the falseness of the efficient market hypothesis as well. It's all pretty murky.


Posted by: Walt Someguy | Link to this comment | 10-14-08 12:38 AM
horizontal rule
10

There's a space elevator of evidence for the murky market hypothesis.


Posted by: Sifu Tweety | Link to this comment | 10-14-08 12:39 AM
horizontal rule
11

that's why it ain't called the efficient markets theorem.


Posted by: dsquared | Link to this comment | 10-14-08 12:46 AM
horizontal rule
12

So, at what point does it become the "efficient markets bullshit"?


Posted by: foolishmortal | Link to this comment | 10-14-08 12:53 AM
horizontal rule
13

d^2 you famous bastard! Krugman owes you some Budweiser, you're so sweet.


Posted by: Sifu Tweety | Link to this comment | 10-14-08 12:54 AM
horizontal rule
14

12: When the last economist is strangled with the guts of the last libertarian.


Posted by: Walt Someguy | Link to this comment | 10-14-08 12:58 AM
horizontal rule
15

13: I've already given him an award, what more could he ask for?


Posted by: foolishmortal | Link to this comment | 10-14-08 1:02 AM
horizontal rule
16

The prize is 10 million kr, which is presently about $1.414 million. BevMo! is currently out of 24 packs of longnecks (no cans, please, this is a special occasion), but were they available PK would be able to pick them up for $17.99 per case. At that price, his prize would buy him 1.886 million bottles of Bud.

Dude, Krugman could get soooo drunk if he bought all that beer.


Posted by: Otto von Bisquick | Link to this comment | 10-14-08 1:58 AM
horizontal rule
17

There is a mountain of evidence for the falseness of the efficient market hypothesis as well.

In all seriousness, I'm not familiar with the evidence against the weak form of the EMH (basically, that previous movements of an asset price impart no useful information about future movements). Could you point some out to me?

As far as some of the stronger formulations of the EMH, I'll agree that the evidence is ambiguous. My personal view is that, whatever the truth or falsehood of the EMH, most people (i.e. all non-financial professionals and about 3/4 of actual financial professionals*) should act as if the EMH were true. That is, don't try to pick stocks.

*Who are the remaining 25%? Those with access to material non-public information, and those who are willing and able trade on the basis of insights from behavioral finance--a strategy that requires (1) unusual insight into market psychology (think Eddie Murphy in Trading Places); (2) copious liquidity; and (3) gonads of steel.


Posted by: Knecht Ruprecht | Link to this comment | 10-14-08 2:25 AM
horizontal rule
18

The prize is 10 million kr, which is presently about $1.414 million.

Prediction: before the day is out, someone over at The Corner will have calculated written a snarky comment about how much Krugman saved in taxes because of the Bush tax cuts.


Posted by: Knecht Ruprecht | Link to this comment | 10-14-08 2:27 AM
horizontal rule
19

I'm not familiar with the evidence against the weak form of the EMH (basically, that previous movements of an asset price impart no useful information about future movements). Could you point some out to me?

Lo, Campbell and McKinlay's "A Non-Random Walk Down Wall Street" ought to have done for weak-form EMH. Momentum anomalies are now totally well-established. I'm really amazed (although not surprised) that the proposition "weak form EMH is still standing" is still being taught. It's definitely false.

My personal view is that, whatever the truth or falsehood of the EMH, most people (i.e. all non-financial professionals and about 3/4 of actual financial professionals*) should act as if the EMH were true. That is, don't try to pick stocks.

Also not true. In general, you shouldn't do anything unless you're prepared to put some effort into it and ready to accept that it might fail, but I never understand why people say "oooooh you mustn't try to pick stocks" rather than "oooooh, never open a small sandwich shop" (which also usually lose money and go bankrupt). It's ideology.


Posted by: dsquared | Link to this comment | 10-14-08 2:39 AM
horizontal rule
20

oooooh, never open a small sandwich shop

I'm also on the record in the very forum as recommending that no one ever start a business (especially a restaurant) because it will probably fail. The reason is not ideology, but constitutional risk aversion.

Fortunately for the world as a whole, there are a lot of entrepreneurs out there with enough irrational confidence in their own abilities that businesses continue to be started and the capitalist economy continues to survive.

I would also concede that a lot of people seem to get a their jollies out of picking stocks and playing the market, and far be it from me to stand between them and their pleasure. But certainly no one should feel any guilt about letting Vanguard do the dirty work for them, because chances are they will be better off in the long run, even before accounting for the value of their personal time they didn't waste analyzing stocks.


Posted by: Knecht Ruprecht | Link to this comment | 10-14-08 2:53 AM
horizontal rule
21

People are talking about economics today for some reason

Well, you know, there's nothing really going on now, except boredom and ennui. Speaking of which, Ben, when I saw this, I thought of you:

Gentles, and Citizens of America, I put it to you, a query of the curious disposition: Have you not noticed, in instances numbered too many to credit the Phenomenon to merest Chance, that the Members of the Opposite-Sex, viz. WOMEN, do so redden at the Cheeks and swoon with Covetousness, upon the sight of the prolifick and persuasive Pamphleteer, viz. & c. g., ME, Sir THOMAS PAINE?

max
['You're missing your calling, man.']


Posted by: max | Link to this comment | 10-14-08 2:54 AM
horizontal rule
22

so that it ought also to explain (a) how the stock prices deviated from what's justifiable in the first place and (b) why it won't take arbitrarily long for the rational and clever speculators to be summoned from the vasty deep.

The hypothesis implies nigh-perfect information, and if we actually took it seriously, would require instantaneous transmission of information. (That is, in excess of the speed of light.) You could start waving away assumptions (and there are a lot of them if you take it seriously), but then you wind up with something that is almost, but not quite, entirely unlike the EMH. People pimp that sucker anyways, usually when they're arguing that you should do something dumb.

I'm also on the record in the very forum as recommending that no one ever start a business (especially a restaurant) because it will probably fail. The reason is not ideology, but constitutional risk aversion.

A new restaurant probably WILL fail. Any given random person probably WILL make bad stock picks. But I don't think you need to invoke the Efficient Markets Platonic Ideal to get there, or if you do, you probably need to work up some better arguments.

max
['Death to the meme!']


Posted by: max | Link to this comment | 10-14-08 3:06 AM
horizontal rule
23

I know better than to get into a pissing match with D-Squared over finance, but my amateur understanding of Lo et al is that they (1) their attack is aimed at the random walk hypothesis, not EMH itself; (2) they explicitly decouple the two; and (3) they don't necessarily dispute a weakish form of EMH, namely that while it's possible to identify pricing anomalies through targeted research, the returns to this activity are subject to the same normalizing pressures as any other business.

Notwithstanding my careless wording in 17, I'm not sure that there is any real daylight between my views and those of Lo et al.


Posted by: Knecht Ruprecht | Link to this comment | 10-14-08 3:11 AM
horizontal rule
24

This time with formatting corrected

I know better than to get into a pissing match with D-Squared over finance, but my amateur understanding of Lo et al is that they (1) their attack is aimed at the random walk hypothesis, not EMH itself; (2) they explicitly decouple the two; and (3) they don't necessarily dispute a weakish form of EMH, namely that while it's possible to identify pricing anomalies through targeted research, the returns to this activity are subject to the same normalizing pressures as any other business.

Notwithstanding my careless wording in 17, I'm not sure that there is any real daylight between my views and those of Lo et al.


Posted by: Knecht Ruprecht | Link to this comment | 10-14-08 3:13 AM
horizontal rule
25

But I don't think you need to invoke the Efficient Markets Platonic Ideal to get there

I don't. Or at least only in a weakish form. My point is that EMH is close enough to true that you have to have a pretty meaningful competitive advantage in the timely acquisition of information to make stock picking worthwhile, and most people (and a pretty large fraction of professional asset managers) are exceedingly unlikely to have such an advantage on a sustainable basis.

Even some of the more intellectually honest stock pickers (Lynch, Buffett)--guys who see themselves as living refutations of the EMH--are pretty upfront about the fact that finding fundamental mispricings is a tedious, resource-intensive business. Lynch, for example, encourages his readers to keep an eye out for information under the radar, like the fact that a local factory is adding a third shift. Realistically, how often is the average investor likely to come across that kind of gem?


Posted by: Knecht Ruprecht | Link to this comment | 10-14-08 3:20 AM
horizontal rule
26

You're right that they do make that distinction, but the papers in that book (and even more, Lo's papers on technical trading rules) are explicitly about using information about the history of securities prices to forecast securities prices and so in that sense, their results aren't consistent with the weak form of EMH.


Posted by: dsquared | Link to this comment | 10-14-08 3:52 AM
horizontal rule
27

I'm of the opinion that people who start restaurants are idealists and altruists sacrificing themselves for the sake of nothing much. I don't know what, diners I guess, or people who rent storefronts. They take their life savings and run their credit to the limit and mortgage their homes and work 80 hours a week for years just so they can file bankruptcy. It's like a big suttee or potlatch or burnt offering to the God of Capital.

In one case I know, failure is a self-fulfilling prophecy. He did everything right, but because it's a restaurant he got credit on unfavorable terms, and for that reason he isn't paying his debt down at all quickly, so he's more likely to fail if there's any adversity. He did absolutely everything right and is a booming success. (Actually, he did put more money into decor than he perhaps should have, but coffeeshops seem to need a sort of a "look", compared to ethnic restaurants which can be tacky. His place is about halfway between funky and swipple).


Posted by: John Emerson | Link to this comment | 10-14-08 5:33 AM
horizontal rule
28

"Is a booming success" in terms of having lots of business, loyal customers, and a sterling reputation, natch. Not financially.


Posted by: John Emerson | Link to this comment | 10-14-08 5:35 AM
horizontal rule
29

Realistically, how often is the average investor likely to come across that kind of gem?

The one thing I've experienced is that sometimes it's not so much "under the radar" stuff as big stuff that runs counter to preconceived notions, and so gets ignored/undervalued. 3 examples, instances where, had I been in a position to buy stock, I genuinely would have, and would have made a killing: 1. Chrysler's ca. 1993 rollout of new vehicle lines across all its brands; 2. iMac; 3. iPod. Those sound like stupid examples, but if you look back at the relevant stock prices, you'll see that none of them moved the company values in the short term after the announcements/rollouts, but all 3 showed big and lasting increases in the 1-5 year range.

Why? Because each time the consensus was that the company was flawed, and therefore the products would be flawed, and therefore they wouldn't make a difference. As a car/computer enthusiast, I recognized that these were game-changing products. I'll add that there haven't been other examples of new products that I thought would change companies' values - it's not like every new Steve Jobs speech makes me think "Buy APL!!!" Also, of course, I don't think it means that I'm some brilliant stock picker. I just think it means that, just as you get stupid elite/insider consensus in every other field, so you can get it among analysts, and I think it's possible for someone who is well-informed in a subject to recognize important information that Wall St undervalues. The opportunities to do so are rare and fleeting, but not necessarily small (and of course worthless if you don't get out in a timely fashion, which can be hard to do if you're patting yourself on the back for buying early).


Posted by: JRoth | Link to this comment | 10-14-08 5:43 AM
horizontal rule
30

When I first read about HIV being transmitted to hospital workers, my first thought was "buy latex, buy gloves", because latex is the barrier protection of choice. I read about it in the newspaper, but if I'd been following medical literature I could have known several weeks earlier. For the next year or more gloves were hard to buy, since everyone already in the business was working at capacity.

I'm not sure that buying latex would have been smart, but buying gloves would have. Usage of latex gloves multiplied many times in the next few months.


Posted by: John Emerson | Link to this comment | 10-14-08 6:10 AM
horizontal rule
31

With obscene operators hidden to protect the innocent

(q   q (1R  ))| (q   q (1R  ))|

which expresses in terms of Keynesian probabilities that tomorrow's price is judged as
likely to exceed as to fall short of today's price plus interest

Q.E.D.


Posted by: bob mcmanus | Link to this comment | 10-14-08 6:11 AM
horizontal rule
32

31:I have stupified the thread.

It was a joke. I have barely begun my first cup of coffee. Maybe I'll be back,

Treatise on Probability is a waycool book.

Another joke.


Posted by: bob mcmanus | Link to this comment | 10-14-08 6:27 AM
horizontal rule
33

[blah blah blah]


Posted by: | Link to this comment | 10-14-08 6:40 AM
horizontal rule
34

I'll add that there haven't been other examples of new products that I thought would change companies' values - it's not like every new Steve Jobs speech makes me think "Buy APL!!!"

As an example I actually took advantage of, Nintendo was vastly underpriced for the first few months of 2006, well after it was apparent that the Wii was a monster hit in the 2005 Christmas season. I suspect the story here was that the business press was focused on the 360 vs. PS3 throwdown, and Nintendo doesn't trade as an ADR and is thus underfollowed by American analysts.

Maybe part of the reason people are more supportive of the weak EMH is that some of the vocal detractors of the strong and semistrong forms buy into the ideology of the weak form? Certainly the people I associate with Fama-bashing are mostly Grahamian value types who don't believe in the usefulness of price data.


Posted by: snarkout | Link to this comment | 10-14-08 6:46 AM
horizontal rule
35

33, soon to disappear

But troll, a million experts predicting the weather makes it rain! It's the magic of markets!


Posted by: bob mcmanus | Link to this comment | 10-14-08 6:46 AM
horizontal rule
36

One follow-on to 29: the iPhone further illustrates my point, in that there wasn't a big opportunity, even though its success has exceeded expectations*, and it had its share of analyst detractors. Why? Because the professionals have now bought into Apple as a company, and its value tends to settle where it should be because there are countervailing arguments. 10 years ago, pre-iMac, the professional consensus was that Apple was doomed, or that their future was in clones, or software, or whatever. The analysts weren't prepared to view the iMac for what it was, and so even the ones who thought it might succeed didn't view it as a game-changer.

* The evidence is that they've already reached the 10 million mark, which was their goal for Dec. 31 2008, and which most analysts said was unrealistic.


Posted by: JRoth | Link to this comment | 10-14-08 7:24 AM
horizontal rule
37

10 years ago, pre-iMac, the professional consensus was that Apple was doomed

And ten years ago, I'd been hearing that for the previous ten years as well.


Posted by: apostropher | Link to this comment | 10-14-08 7:32 AM
horizontal rule
38

And ten years ago, I'd been hearing that for the previous ten years as well.

Well, yeah. But in that ten years, Apple never produced anything to really change things (Newton was close, but they never got it right, on a variety of metrics). When the G3 processor came out, that was a good sign, but it on its own it wasn't enough. But once the G3 went into the iMac, and it was a good price, suddenly it became clear that Apple had a hot product that would sell big, and change that decade-old narrative.


Posted by: JRoth | Link to this comment | 10-14-08 7:44 AM
horizontal rule
39

Newton could have been great. They were only a couple of generations away from something that would have been genuinely revolutionary. But Newton was not Jobs' child so when he returned it had to be strangled and will never be brought back. Jobs is such a dick about other people's ideas that I predict he will resist allowing the iPhone to evolve in the direction of Newton. That's too bad, because there really is a niche out there for a device about the size of a hardback book which can do basic web browsing and serve as an organizer as well as do lightweight office type stuff. Amazon's Kindle will evolve in that direction, I believe.


Posted by: togolosh | Link to this comment | 10-14-08 8:04 AM
horizontal rule
40

Knecht: Dsquared already mentioned momentum -- stock prices have a slight tendency to keep moving in the same direction over periods of a year. Over long time horizons, they tend to move in the opposite direction. Market-wide price-dividend ratios forecast market-wide returns about a year in advance. Returns on large stocks forecast returns on small stocks by about a day.


Posted by: Walt Someguy | Link to this comment | 10-14-08 8:05 AM
horizontal rule
41

Yeah... The efficient market hypothesis is pretty much bullshit, and I'd agree with dsquared that one of the biggest nails in its coffin is evidence of ten billion types of price and earnings momentum. Things like anti-correlated long-term (3-5 year) returns and persistent medium-term returns (1 year) are nearly impossible to explain in terms of risk factors, have been established in the finance literature since about 93-95, and yet keep popping up everywhere else we look for them. I've got a paper at the moment by a few academics/hedgies which establishes similar momentum effects in commodities, currencies, and bonds as well as the already well-known effect in foreign equity markets.

Now, there is also some evidence of persistent alpha out there for some individuals and mutual funds (Oddly enough, Lynch isn't one of those. If you look at his historic returns, they have lots of alpha against a three-factor model, but most of it disappears when momentum is taken into account). There is evidence that U.S. institutions as a whole have alpha, which was identified by looking through the quarterly disclosures that any institution investing more than $100 million must make. However, though the alpha is robust enough to be statistically significant, it is tiny enough that any sort of management fee would wipe it out. Basically, it's just proof that individual investors and foreign investors kind of suck in their stock choices, leaving U.S. institutions with better-quality goods.

There are piles and piles of evidence that I believe add up to a market that's quite difficult to beat. I'd also add that, in my very limited experience as an equity analyst at a fundamentals-obsessed firm, the vast majority of the time, really are pretty accurate looking. Of course, we're not in such a time anymore, but now it becomes more a matter of how long the market can stay stupid. Unfortunately, the answer can be 10 years or more.


Posted by: Po-Mo Polymath | Link to this comment | 10-14-08 8:07 AM
horizontal rule
42

Isn't most of the money made in the exceptions, regardless of whether it's just dumb luck or whether its smarts?


Posted by: John Emerson | Link to this comment | 10-14-08 8:21 AM
horizontal rule
43

10 years ago, pre-iMac, the professional consensus was that Apple was doomed

The consensus of MBA's perhaps, but certainly not the consensus of people with technical knowledge. And there's the rub, really.

I was in the valley at the time, surrounded by real tech talent, and the day after a discussion with colleagues about the glimmerings of turn-around there, I bought as much AAPL as I could at about $7. It's a shame I didn't have any real money to put into it, as I sold nearly the lot at $185, which sounds great but it wasn't a lot of stock. So you decide if I was lucky or knew what I was doing, but the market signaling on that sort of thing doesn't seem to be that consistently good. It catches up eventually.


Posted by: efficient market? | Link to this comment | 10-14-08 8:28 AM
horizontal rule
44

42: Nah, I'd roughly estimate that most of the money is made in managing the money, which is why publicly-traded fund managers absolutely kicked ass over the past 5-10 years, though people are very bearish on them now for somewhat obvious reasons.


Posted by: Po-Mo Polymath | Link to this comment | 10-14-08 8:31 AM
horizontal rule
45

Nah, I'd roughly estimate that most of the money is made in managing the money

Telling, that.


Posted by: soup biscuit | Link to this comment | 10-14-08 8:34 AM
horizontal rule
46

44, 45: Churn Rules!


Posted by: JP Stormcrow | Link to this comment | 10-14-08 8:45 AM
horizontal rule
47

Oddly enough, Lynch isn't one of those. If you look at his historic returns, they have lots of alpha against a three-factor model, but most of it disappears when momentum is taken into account

I have a long and unstructured rant on this subject, which awaits conversion into something publishable, on the general subject of regression models not dealing at all well with market timing, and the literature not taking this well-known problem (which has an equally well-established solution) into account.

Example - both of us buy a Vanguard index fund and the market goes up 20%. You "buy and hold", but I sell and go to cash, earning 1%, and the market goes down 20%. Then I buy back in and the market goes up 4%.

By construction, my "alpha" is zero - at all times during this thought experiment my portfolio has consisted only of assets where the return is completely explainable as the reward to risk. But on the other hand, despite not having any alpha, I appear to have had lower volatility to you and indisputably have $1.26 for every $1 you have.


Posted by: dsquared | Link to this comment | 10-14-08 8:48 AM
horizontal rule
48

Isn't a presumption of predictability, or an exploitation of the natural human tendency to seek security in predictability, at the core of the Neo-Classical program of Friedman/Mundell/Lucas?

Rational expectations, lifetime earnings, monetarism, inflation expectations, NAIRU as a formula, emphasis on controlling inflation over full employment...all based on this weird epistemology that says the future, while not controllable in the aggregate, is foreseeable?

Strange kind of conservatism.


Posted by: bob mcmanus | Link to this comment | 10-14-08 8:50 AM
horizontal rule
49



Posted by: | Link to this comment | 10-14-08 8:56 AM
horizontal rule
50

This kind of "no-arbitrage" condition is very frequently used to "prove" stuff in economics. It follows tautologically from the assumption that a large number of people are effective, opportunistic, rational, profit maximizers, constantly scouring the markets for profit opportunities. And in truth, under capitalism social institutions do create a large number of such types and make it easy for them to do their thing.

The hitchhiker analogy would be a social theory that posited that people were relentlessly benevolent, constantly scouring their surroundings for opportunities to do good and provide aid to others. Under such circumstances, hitchhikers would be as rare as $20 bills lying on the sidewalk.


Posted by: PGD | Link to this comment | 10-14-08 9:03 AM
horizontal rule
51

My point is that EMH is close enough to true that you have to have a pretty meaningful competitive advantage in the timely acquisition of information to make stock picking worthwhile, and most people (and a pretty large fraction of professional asset managers) are exceedingly unlikely to have such an advantage on a sustainable basis.

... such that they beat the market? I agree, actually. But that doesn't quite make it even to weak EMH, unless you take the long view.

[Considers it, backs up, tries again.]

Roughly, I don't think in terms of EMH, since I think it's silly for the purposes it is used for. Like this guy:

On Friday, Mr. Shiller told me of a conversation he had with an economist friend of his. The man had spent his entire career advocating the efficient market hypothesis, which posits that all known information about a stock is already priced into it. But with the market in collapse, the economist sold all his stocks. "I feel like a Christian Scientist who has come down with appendicitis," he told Mr. Shiller.
(vulgar?) EMH most oftens seems to be used to argue that prices will go up, so pony up cash now. It strikes me as a just so story - lose money, the markets are efficient. Make money, well, you're a genius and the markets are efficient. (Cut taxes because the markets are efficient!)

What you're saying is more along the lines of 'I can't go head-to-head with those sods over there'; that's probably mostly true (or true most of the time), but it doesn't mean the market is efficient, in terms of the hypothesis platonic ideal. It's merely that others have more resources than you; potentially enough to make the market inefficient enough for you to lose money.

max
['Love those stock pools!']


Posted by: max | Link to this comment | 10-14-08 9:09 AM
horizontal rule
52

Also, I've always wondered about the fact that it is institutionally harder to go short than to go long (margin calls, etc.). Doesn't that make it somewhat more difficult to arbitrage opportunities that rely on a stock dropping?


Posted by: PGD | Link to this comment | 10-14-08 9:12 AM
horizontal rule
53

||

I have little faith in the voter registration process. After moving to a new state, I tried to register by mail due to the remarkable ineptitude of the DMV, waited weeks without hearing anything, and found some "are you registered?" website that had no record of me. Today I called to find out what's going on, and talked to someone who says that I am registered, and that she remembered talking to me about it before. The latter is false, assuming I didn't call in my sleep. Hopefully the former isn't?

|>


Posted by: essear | Link to this comment | 10-14-08 9:13 AM
horizontal rule
54

47: Well... you would technically be buying zero-alpha assets the entire time, but a regression of your returns over the entire period would show a beta considerably below one and a good-sized alpha, since regression models assume constant betas to all factors.

Now, that regression would not reveal what precisely caused your alpha, but it would show that it existed.

But yes, I agree that regression models have some major faults. If dealing with any portfolios that disclose holdings, I prefer some of the approaches based on characteristic-portfolios that allow one to not only accomodate changing exposures to risk factors, but also to attribute alpha (if any) to stock selection and risk factor timing.

I'm not yet aware of any academic models that do a good job of analyzing market timing without relying on disclosed portfolios, but do you guys know of any, dsquared and Walt?


Posted by: Po-Mo Polymath | Link to this comment | 10-14-08 9:20 AM
horizontal rule
55
Posted by: | Link to this comment | 10-14-08 9:21 AM
horizontal rule
56

kalashnikov
i recalled how we in hs used to assemble and dissemble it, i was pretty good at it


Posted by: read | Link to this comment | 10-14-08 9:23 AM
horizontal rule
57

There is surely a more cogent way to phrase this, but it always strikes me that the role of time delays are never given enough attention in these theories. Even for relatively simple and well-understood systems such as motor controls, steering, a million manufacturing processes, the whole "game" is in managing the second-order effects arising from delays. These models seem to either assume this all away, or that the "smart" guys like Soros act as the feedback/feedforward controllers (but even then their "opportunity" arises when they measure the gap and act, both of which take time). I dunno, am interested to learn if there has ever been some exploration of injecting some of the prinicples of Control Theory into these models. (Maybe it is already baked in?)


Posted by: JP Stormcrow | Link to this comment | 10-14-08 9:24 AM
horizontal rule
58

56: read for the win!


Posted by: JP Stormcrow | Link to this comment | 10-14-08 9:25 AM
horizontal rule
59

This article seems relevant, they make reference to D2s point about market timing being difficult to model.


Posted by: PGD | Link to this comment | 10-14-08 9:27 AM
horizontal rule
60

52: Absolutely. An entire branch of behavioral finance research looked into limits to arbitrage that arise due to the difficulty of shorting stocks in some markets and the problems with arbitrage trades that rely on other people's money. The latter argument is especially persuasive to me.

The argument in short is: Since arbitrage trades are often technical and difficult to explain and/or find, any arbitrage fund or trader will be assessed mostly on performance/reputation because few managers or investors could assess the positions on their fundamental merit. So if a great new arbitrage trade has been found, but it will take a while to resolve itself, there's a possibility of that trade going against the arbitrage fund in the meantime. Should that happen, investors or managers who only see the poor performance may demand money be pulled out of the trades, which ultimately makes the relative values of the assets go even further out of whack, potentially causing more forced liquidation, etc. So any noise in relative asset prices can drastically reduce the effectiveness of arbitrageurs, since they're almost always investing someone else's money.


Posted by: Po-Mo Polymath | Link to this comment | 10-14-08 9:31 AM
horizontal rule
61

It's merely that others have more resources than you

The only smart thing I remember McArdle ever saying was that the EMH in theory depends on a majority of investors suckers picking stocks badly in order to create the information useful to successful investors. The relevant thought-experiment involves everyone buying index funds.

Saying risk & return have anything to do with stock prices is like saying it is utility that determines the price of hulahoops.


Posted by: bob mcmanus | Link to this comment | 10-14-08 9:32 AM
horizontal rule
62

(but even then their "opportunity" arises when they measure the gap and act, both of which take time).

Between the idea
And the reality
Between the motion
And the act
Falls the arbitrageur.


Posted by: ben w-lfs-n | Link to this comment | 10-14-08 9:37 AM
horizontal rule
63

The only smart thing I remember McArdle ever saying was that the EMH in theory depends on a majority of investors suckers picking stocks badly in order to create the information useful to successful investors.

It's not inconsistent that there simply be no clever speculators, is it?


Posted by: ben w-lfs-n | Link to this comment | 10-14-08 9:39 AM
horizontal rule
64

54 - There was a paper two years that got some media attention that had some way of matching dynamic strategies to measure hedge fund performance. The authors claimed that once you used it, there was no evidence of significant alpha.


Posted by: Walt Someguy | Link to this comment | 10-14-08 9:45 AM
horizontal rule
65

And of course stock picking is one area where anecdata is fraught with self-delusion. I kick myself to this day for not acting on a clear sense that AOL would "win"* against other online competitors way back in the day, but I've surely forgotten half a dozen mediocre and two disastrous decisions I would have made around the same time.

Of course this phenomenon is most clearly seen in folks who gamble in casinos. How many, "Oh, but I/my uncle/my friend wins", stories have you heard. Habitual casino gamblers are people who lose money gambling in casinos (don't get me wrong it can be fun), if they are comping you big time you are a big loser. For instance, a particularly lame attempted attack on McCain I saw was that he had never declared any gambling winnings on his taxes, despite his frequent gambling.

*Part of why I did not act, was because of my limited imagination with regard to the potential financial magnitude of what a "win" in online services represented.


Posted by: JP Stormcrow | Link to this comment | 10-14-08 9:48 AM
horizontal rule
66

Tangentially related, and certainly a methodology open to criticism, but that won't stop me from recommending it as a political education piece for our undecided members.


Posted by: Sir Kraab | Link to this comment | 10-14-08 9:55 AM
horizontal rule
67

The efficient markets hypothesis is a simplified model of reality. Like other simplified models it doesn't have to match reality exactly to be useful. So just pointing out cases in which it doesn't match reality exactly is kind of pointless.


Posted by: James B. Shearer | Link to this comment | 10-14-08 9:58 AM
horizontal rule
68

66 - even if you exclude Clinton's extreme from the Democrats they still thump the hell out of the republicans. That's impressive.


Posted by: togolosh | Link to this comment | 10-14-08 10:06 AM
horizontal rule
69

More responses to trolls should involve claims of knowing how to use a Kalashnikov. That's great.


Posted by: essear | Link to this comment | 10-14-08 10:08 AM
horizontal rule
70

VII
O thin men of Haddam,
Why do you imagine golden parachutes?
Do you not see how the arbitrageur
Walks around the feet
Of the women about you?


Posted by: essear | Link to this comment | 10-14-08 10:25 AM
horizontal rule
71

66. Be very careful what you wish for. You may have an "excludes Obama" figure in eight years. Past performance being no guide to future performance, etc.


Posted by: Tassled Loafered Leech | Link to this comment | 10-14-08 10:30 AM
horizontal rule
72

I was of three minds,
Like a tree
In which there are three arbitrageurs.


Posted by: PGD | Link to this comment | 10-14-08 10:34 AM
horizontal rule
73

71: Past performance being no guide to future performance, etc.

Totally acknowledged. (Though I'm not sure what you mean by "be careful what you wish for.") I'm certainly not suggesting that the ups and downs were solely the result of a Democratic or Republican administration, but I do think it's a helpful thought experiment for voters who automatically associate Democrats with "tax and spend" policies that don't benefit them.


Posted by: Sir Kraab | Link to this comment | 10-14-08 10:49 AM
horizontal rule
74
Posted by: | Link to this comment | 10-14-08 10:56 AM
horizontal rule
75

What I mean Krabby, is that it is more than likely that the next several years will not be kind to the markets, no matter what President Obama does. I am making the assumption that it is your wish that he be President, regardless.


Posted by: Tassled Loafered Leech | Link to this comment | 10-14-08 11:05 AM
horizontal rule
76

So apparently the next level of horribleness from the wingnuts is the suggestion that Obama had a consensual homosexual affair with a pedophile when he was ten years old. It's beyond caricature.


Posted by: togolosh | Link to this comment | 10-14-08 11:07 AM
horizontal rule
77

73: but I do think it's a helpful thought experiment for voters who automatically associate Democrats with "tax and spend" policies that don't benefit them.

Or for fucking fuckheads like David Broder* whose most recent column made a brief acknowledgment that Bush had totally fucked things up and then went on to preemptively admonish Obama for the short honeymoon he would have and his need to choose between a progressive agenda and fixing things up.

*I would like all the Unfoggedtariat to take a solemn vow that they and their children and their children's children will join me in laying waste to the name and reputation of David Broder. I'm talking old school here, scorched earth. I want children in 2150 to fear the name of "Broder", I want there to be the equivalent of Godwin's Law about bringing up his name in Internet discussions, I want a million naive discussions of moral philosophy to include "but how would that prevent a David Broder?".

... oh, and Cokie Roberts, too.


Posted by: JP Stormcrow | Link to this comment | 10-14-08 11:12 AM
horizontal rule
78

Yeah, David Broder is already sharpening the knives for Obama, predicting that because of Bush's failure, Obama will have a short honeymoon. The indispensable Broder is also pushing a balanced-budget / spending-cuts program which he must have learned from his demented grandmother on her deathbed. His grandmother also warned him that Social Security would never work.

I'm sure that we can count on our friend TLL to explain to everyone if Obama does well by the ruined country Bush left him. After all, TLL is not a Republican and is totally wise to Bush's tricks. He would never join in on the insane jabbering, led by Broder, that will begin on Jan. 21.


Posted by: John Emerson | Link to this comment | 10-14-08 11:18 AM
horizontal rule
79

Don't try your feeble Jedi minds tricks on me, Emerson-Kenobi.


Posted by: Tassled Loafered Leech | Link to this comment | 10-14-08 11:23 AM
horizontal rule
80

predicting that because of Bush's failure, Obama will have a short honeymoon.

"You have a really nice family, really nice. It'd be a shame if anything were to happen to them."


Posted by: Jackmormon | Link to this comment | 10-14-08 11:29 AM
horizontal rule
81

75: What I mean Krabby, is that it is more than likely that the next several years will not be kind to the markets, no matter what President Obama does.

True also for President McCain (*shudder*). Whether or not the economy tanks, middle- and working-class people will be better off with President Obama. (Barring unforeseen alien invasions & etc.)


Posted by: Sir Kraab | Link to this comment | 10-14-08 11:32 AM
horizontal rule
82

77: Count me in.


Posted by: Sir Kraab | Link to this comment | 10-14-08 11:34 AM
horizontal rule
83

Incidentally, I've figured out what the idea that Republicans are fiscally responsible is all about. It's Eisenhower nostalgia from 48 years ago. Eisenhower really cared about that stuff and probably did a fair amount of harm that way. Nixon, Reagan, Bush, and Bush were all big deficit spenders. NOT THAT THERE WAS ANYTHING WRONG WITH THAT. For a lot of people "fiscally conservative" is part of the name of the Republican species -- blackbirds are blackbirds even f they're white.


Posted by: John Emerson | Link to this comment | 10-14-08 11:35 AM
horizontal rule
84

Barring unforeseen alien invasions

The aliens are going back home, which is one reason for the housing collapse in some markets.
http://isteve.blogspot.com/2008/10/my-new-vdare-column-big-enchilada.html

It's Eisenhower nostalgia from 48 years ago.

Some memes will never die, no matter how much the evidence is against them.


Posted by: Tassled Loafered Leech | Link to this comment | 10-14-08 11:45 AM
horizontal rule
85

OK, folks, let's visualize.

Broder's on TV droning on and on, and then he slows down, and stalls, and starts jerkily up again -- and then suddenly slabs of putrescent Broder start shelving off like slabs of glacier, splatting on the ground, faster and faster until there's nothing left of him but an aromatic, quivering, gelatinous heap.

We'll all have to do it at the same time when Broder's on the TV. Ogged's thunb thingie might help. Also ritual swipple objects, yoga, free-range sausages, saffron, artisanal vinegar, heritage tomatoes, and so on.


Posted by: John Emerson | Link to this comment | 10-14-08 11:48 AM
horizontal rule
86

85: Not so fast, it might turn out that Broder's seemingly decrepit human mind and body are actually disguising a super-strong and malevolent inner being whom we do not want to see released, like in Cronenberg's "The Fly".


Posted by: CN | Link to this comment | 10-14-08 11:51 AM
horizontal rule
87

In the end we whipped the Stapuft Marshmallow Man, and we'll whip the Broder cyborg too!


Posted by: John Emerson | Link to this comment | 10-14-08 11:53 AM
horizontal rule
88

That image is from Poe's "The Facts in the Case of M. Valdemar/Edgar Allan Poe "THE FACTS IN THE CASE OF M. VALDEMAR", an underestimated story.


Posted by: John Emerson | Link to this comment | 10-14-08 11:55 AM
horizontal rule
89

Here! Short! Read it!


Posted by: John Emerson | Link to this comment | 10-14-08 12:03 PM
horizontal rule
90

Good read, bad background. Victorians are more fun, with the Gothic undertones.


Posted by: Tassled Loafered Leech | Link to this comment | 10-14-08 12:09 PM
horizontal rule
91

... oh, and Cokie Roberts, too.

I mentioned here the other day that I shocked my good, NPR-listening MIL by telling her that Cokie is worthless or worse. Key evidence? That she was pro-impeachment.

I suppose that sullying her name for the future does not start with turning the elders against her. But it can't hurt.


Posted by: JRoth | Link to this comment | 10-14-08 12:16 PM
horizontal rule
92
Posted by: | Link to this comment | 10-14-08 12:30 PM
horizontal rule
93

Is 92 ToS turning making halting steps toward value-added commenting? Keep it up, ToS! Remember that read, too, was once an unwelcome interjector of nonsense, and look how far she's come!


Posted by: Knecht Ruprecht | Link to this comment | 10-14-08 2:23 PM
horizontal rule
94

60 summarizes the best theoretical argument why economists shouldn't expect the efficient markets hypothesis to be true.

Arbitrageurs are the rational investors the EMH counts on to make sure prices are in line with fundamentals. But there are limits to arbitrage: it's carried out by professionals investing other people's money, so there's always a risk that the mispricing will get much worse before it gets better at some unspecified date in the future. If the stock market is in a bubble and you bet it will go down, and it irrationally goes up 30% in a year, you have a lot to explain to your investors, assuming you have any left.

On the empirical evidence side, stock prices are simply too volatile to be explained purely by new information. Moreover, there are good reasons to believe that stock prices move together far more than you would expect if fundamentals were all that mattered; indeed, it looks like investor sentiment plays a very large role in determining the overall movement of markets, at least in the short run.

Finally, Soros himself has said that he's made a lot of money from predicting how the market will move in the future, irrespective of how it should move. That is, he's often not paying that much attention to the fundamentals, contrary to what the passage quoted in the original post says.

None of this means that you have any real chance in beating the market, though.


Posted by: Barbar | Link to this comment | 10-14-08 2:37 PM
horizontal rule
95

That is, he's often not paying that much attention to the fundamentals, contrary to what the passage quoted in the original post says.

And indeed, if you read the article linked in the post, you will understand that the quoted paragraph is not meant to represent Soros' own thinking.


Posted by: ben w-lfs-n | Link to this comment | 10-14-08 3:06 PM
horizontal rule
96

The main reason I can't get around to believing in an EMH is that there are so many idiots out there. PMP, Dsq, others in the industry, haven't you seen a hundred examples of industry "professionals" advising their clients to do stupid things, or clients insisting against the advice of their advisers on doing things that are twice as stupid? It's not just churning either, it's pointless short sales, poorly conceived margin loans, utterly inappropriate investments for clients who are too old or too young or too inexperienced to know what is going on.

I could probably come up with at least one example for every day I've worked in the industry of people with serious bankrolls making incredibly foolish decisions -- not just "irrational" but downright idiotic. Some days I don't know how to talk to these people without just blowing up at them.

It strains credulity to think that the upper echelons are so much more clever than the retail sales force. Look at our current mess -- a conspiracy of the high, middle and low to totally fuck things up.

Christ, you might as well take the money out into the parking lot and burn it, for all the good most folx get out of trading.

Don't even get me started on all the pump-and-dump boiler-room operations that skim across the surface of illegality like so many pond insects.


Posted by: minneapolitan | Link to this comment | 10-14-08 8:21 PM
horizontal rule
97

Mineapolitan, this guy thinks that the big bankers were just as bad:

For me, all this is very troubling. It suggests that what we economists have to learn from Taleb has nothing to do with the nature of risk - we've all known that - but about others' rationality. We should ditch the assumption - which in a sense is mere courtesy - not only that others are rational but even the weaker assumption that they are nearly so. Perhaps we should indeed regard them merely as "empty suits."

But this is a vastly greater departure from standard practice than anything Taleb has suggested about the nature of risk.
All of which leaves me genuinely puzzled. Were banks risk managers really that bad? Are bosses an order of magnitude stupider than even I had thought? Or is something else happening? Help me.

My own theory is that a lot of bankers took Econ 101 too seriously, believed in equilibrium and Gauss curves, and thought that all the new sophisticated stuff was "just theory" as they say in these United States. ("Say, does this guy Einstein have anything on the ball, or is it all just theory". Apocrypha).

People keep telling me "econ isn't really like that", but as it functions in life, sometimes I think it is. Regardless of what the top guys were doing.


Posted by: John Emerson | Link to this comment | 10-14-08 8:35 PM
horizontal rule
98

The link in 84 got ignored, but deserves contempt. I'm going to remember it next time I'm trying to be fair and give you the benefit of the doubt, TLL.


Posted by: Witt | Link to this comment | 10-14-08 8:44 PM
horizontal rule
99

Yikes, 98 gets it exactly right. If that was supposed to be humor, it didn't work.


Posted by: Not Prince Hamlet | Link to this comment | 10-14-08 8:49 PM
horizontal rule
100

Go ahead, Witt. I don't know what was offensive. Some of the loans were crazy, and it had nothing to do with the ethnicity of the borrower. But if the migrant population that was employed in the buildng trade has moved on, or back, then the areas where they tended to live are now overbuilt. It is only contraversial if you want to assign motive to it.


Posted by: Tassled Loafered Leech | Link to this comment | 10-14-08 8:51 PM
horizontal rule
101

I don't know what was offensive.

Let's start with the fact that it was by Steve Sailer.

Let's end there too. I'm voting with my feet.

It is only controversial if you want to assign motive to it.

I'm not sure what you were referring to here, but I'll say a hearty Amen. I'm quite happy assigning motive to people who want to discuss economic meltdown in ethnic terms.


Posted by: Witt | Link to this comment | 10-14-08 8:58 PM
horizontal rule
102

Wow, Steve Sailer? Good job there, TLL.

"You see, the economic meltdown was predicted in The Bell Curve."


Posted by: Sifu Tweety | Link to this comment | 10-14-08 9:03 PM
horizontal rule
103

TLL, Steve Sailer is not a reliable source. He seems like a smart guy, and he's a persuasive writer, but he's a fanatic. Everything he writes that is in any way tangentially related to race is completely distorted by his obsession with the inferiority of blacks and Mexicans. Given that he's a persuasive writer, he's one of the few people you can read and actually be worse informed than before you read it.


Posted by: Walt Someguy | Link to this comment | 10-14-08 9:09 PM
horizontal rule
104

Shorter 103.last: Sailer:race::McMegan:economics


Posted by: Not Prince Hamlet | Link to this comment | 10-14-08 9:41 PM
horizontal rule
105

For fuck's sake TLL, the housing boom went bust and so all the building jobs are gone. You think the real estate market stopped going up 20% every year because some illegal immigrants went home?

The financial crisis is a complex story involving rising home prices, poor underwriting standards, bad loans being made, bad securities being made, bad bets being made on these securities, and on and on and on and on. And then there's the rating agencies, Alan Greenspan, regulators, yadda yadda. As we can see now, it ended up involving the entire goddamn global financial system, it may bankrupt entire countries, massive banks and insurance companies have already collapsed, others have been taken over by the government, etc etc.

So what kind of stupid fuckwit looks at this picture and thinks, "You know who hasn't gotten enough blame for this? Illegal Mexicans making $6/hour. Plus, poor black people. They're flying under the radar."

A fucking racist shithead, that's who.

And what kind of person reads an article by such a shithead and thinks, "Wow, this is really insightful."


Posted by: Barbar | Link to this comment | 10-14-08 10:31 PM
horizontal rule
106

96: the case for the EMH does not really depend on rationality of all or even most investors, though.

First off, if individual deviations from rationality are random, then the market will still look rational.

But of course there are quite systematic errors that people make. Nevertheless, as long as you have a few smart arbitrageurs who can spot the errors, they can simply get very rich cleaning up the mess and pushing the markets towards rationality.

You don't get theoretical problems with the EMH until you face limits to arbitrage, which is basically the fact that in the real world arbitrage is risky.

97: no, bankers do not take Econ 101 seriously. NOBODY makes decisions simply because "a PhD built a computer model that says to do this." People rely on models when it's politically convenient to do so -- for example, when the models have a track record of generating profits, or when all your competitors are all making a killing by using them.

And then someone comes along and says, "You know what, I know we're making a killing right now but if X Y Z happens then I think we are in danger of losing a lot of money." If things are going well and money is being made, such a person is laughed at and marginalized. It has nothing to do with being blinded by equilibrium and Gaussian distributions; it's about the fact that no one wants the party to end.


Posted by: Barbar | Link to this comment | 10-14-08 10:45 PM
horizontal rule
107

One more thought for TLL -- it should be remembered that racists always had a very hard time interpreting California's real estate bubble while it was happening. As you may recall, California was turning into an intolerable hellhole because of immigration, but it also cost ridiculous amounts of money to buy a home in the state, suggesting it was a desirable place to live. How to reconcile? Very tricky!

So for example Sailer wrote in a comment in this 2006 thread:

Here is a way to quantify roughly the impact of illegal immigrants on real estate prices in LA, offered by Sandra Tsing Loh, who writes regularly about schools for both the Atlantic Monthly and the LA Times.

In her "Scandalously Informal Guide to Los Angeles Schools," she writes:
1 API point = $1000 worth of real estate
"So in Los Angeles, a three-bedroom home whose local school has a ("baseline acceptable") 800 API will have a price tag of about $800,000, which is why some wags call the API the "Affluent Parent Index." So if you live in anything less than an $800,000 home, your local school's API is likely to be, well, Guavatorina-like (aka: like a sad bruised melon)."
API scores range by about 500 points in LA among schools, with a very strong correlation for neighborhood schools between ethnic makeup and API scores, with the lowest scores found in schools made up primarily of the children of illegal immigrants.
So, that suggests a negative marginal impact of some hundreds of thousands of dollars on home prices in LA neighborhoos where illegal immigrants choose to live.

In other words, a very thoughtful analysis showing why illegal immigrants hold California real estate values DOWN.

In the very same thread he also thoughtfully argues that illegal immigrants drive up real estate prices up. The only consistent theme I see is that Hispanics are bad.


Posted by: Barbar | Link to this comment | 10-14-08 11:49 PM
horizontal rule
108

Wait, he quoted Sandra Tsing Loh as his source for statistics? She's a comedian! Oh, he's a genius.


Posted by: Sifu Tweety | Link to this comment | 10-14-08 11:52 PM
horizontal rule
109

Thanks for the info on Sailer. It did confirm some anecdotal evidence, but I was unaware of his previous work. Look, I know that the housing bust was not caused by the borrowers, but somebody was signing these loans and in my experience the most egregioous cases were people who did not speak English being gamed by a bilingual mortgage broker. Last year LAUSD noticed a drop in enrollment, especially among immigrants (who didn't show up as anticipated). There is not as much opportunity, so some are moving. By itself, that would not deflate demand enough to shrink prices. But the prices were artificially inflated, for alot of reasons, so the bust is amplified.


Posted by: Tassled Loafered Leech | Link to this comment | 10-15-08 1:14 AM
horizontal rule
110

109: but in this case, given the scale and the heterogeneoty, we must be so, so careful how we cast our anecdotes; so, so easy not to gain real information.


Posted by: Sifu Tweety | Link to this comment | 10-15-08 1:22 AM
horizontal rule
111

Oh hell, Sifu, information just clogs my brain. I purge to make room for the porn.


Posted by: Tassled Loafered Leech | Link to this comment | 10-15-08 1:31 AM
horizontal rule
112

I feel you, man.


Posted by: Sifu Tweety | Link to this comment | 10-15-08 1:33 AM
horizontal rule
113

106: I wasn't saying that Economists take the latest computer model seriously. I was saying that at some deep level, they really believe the truisms and assumptions of Econ 101. They didn't do anything "simply because" of Econ 101 -- they had a motive, and magical thinking was part of it.

The argument was at the link.


Posted by: John Emerson | Link to this comment | 10-15-08 5:53 AM
horizontal rule
114

113: link where?

Googling your quote I found something on Economist's View.

Bankers aren't economists, and economists aren't bankers. Bankers believe in the magic of making shitloads of money, that's all.

Economists tend to believe in rationality and so in the absence of evidence to the contrary, they tend to think that professionals know what they're doing, and that everything's for the best. That's what the Taleb v Economists guy is talking about.

This view that we live in the best of all possible worlds is indeed quite popular in economics. On the other hand, many smart economists have also noticed that, for example, failing isn't that bad psychologically if everyone else does exactly the same thing (herd effect), and loads of other basic facts about human psychology. So for example, a couple of psychologists won the Nobel Prize in Economics a few years ago for their research on consistent human deviations from rationality. See also big-time economists like Richard Thaler, Robert Schiller, and Andrei Shleifer.

Taleb is largely a crank who overstates the novelty of his ideas.


Posted by: Barbar | Link to this comment | 10-15-08 10:43 AM
horizontal rule
115

What the guy at the link said is that Taleb's ideas weren't new, but that he could have taught the economists that bankers are idiots.

Explaining losing shitloads of money by the belief in making shitloads of money leaves to be desired.


Posted by: John Emerson | Link to this comment | 10-15-08 11:07 AM
horizontal rule
116

Explaining losing shitloads of money by the belief in making shitloads of money leaves to be desired.

Does it help to notice how profitable the banking industry has been for the last N years?


Posted by: Barbar | Link to this comment | 10-15-08 11:18 AM
horizontal rule
117

106: the case for the EMH does not really depend on rationality of all or even most investors, though.

Sure, but what I'm proposing is that we view investor behavior not on a continuum bounded by "irrational" on one end and "rational" on the other, but rather one that runs from "dead-set-on-doing-things-that-are-against-one's-own-interests" to "completely clueless" to "irrational" to "rational".

And I see behavior all along that continuum every day, so it must exist. Of course it's irrational to sell off your whole portfolio on a day like today, when you've got 20 years till retirement. But that level of bad decision-making is qualitatively different than some of the shenanigans I see people pulling. And I don't even see the worst of the worst, by any means.

It's a fallacy to assume any nadir for human behavior, because people can always be led lower.

Apropos of nothing, what's up with a 29 year-old media savvy American who's never heard of Gilda Radner? We live in a debased age for sure.


Posted by: minneapolitan | Link to this comment | 10-15-08 5:00 PM
horizontal rule