?

Log in

No account? Create an account

Previous Entry | Next Entry

On dangers of Behavioral Finance

After reading the debate on Behavioral Finance vs. Efficient Markets Theory, I came to a view that the government should ban Behavioral Finance as a destabilizing force. For efficient capital allocation we need markets to behave predictably and according to the fundamentals. The more investors behave rationally, the more predictable markets become. Contrariwise, if the majority of investors subscribe to the Behavioral Finance and bet on stocks that they believe would be revised next quarter by the analysts (as prof. Thaler put it), then the markets destabilize and grow further from fundamentals, with such dire consequences as triggering over/underinvestment and amplifying business cycle. Thus Behavioral Finance becomes self-fulfilling prophecy, and a dangerous one at that. Hence, I propose to denounce Behavioral Finance as a form of financial terrorism.

Comments

( 11 comments — Leave a comment )
oxfv
Nov. 3rd, 2002 02:25 pm (UTC)
Don't know anything about the topic but it sounds such as when things like that Behavioral Finance got banned the market becomes as predictable as the tick-tack-toe game. This would make free market a pathetic institute, also limiting its potential a great deal. However, you're right that if many people subscribe to that "Behavioral Finance" beast the market is going to develop brownian movement (or worse yet, the luddite movement). Thus, I propose the permissive approach: let's accredit a limited number of people to allow to practice this destabilizing technique. How's that?
_mak_
Nov. 4th, 2002 05:19 am (UTC)
Замечательная мысль! А лицензии будем раздавать по блату и за взятки! От Сталина к Брежневу, так сказать.
lnvp
Nov. 3rd, 2002 05:22 pm (UTC)
Замечательно :^)

Рисуется радужная картина: в одну ночь по всем университетским кампусам происходят задержания идеологов вредоносного Бихейвиористического Финансизма (для опознания личности на предмет возможных связей с Аль-Кайдой). Наутро Президент издаёт Декрет (после недолгих дебатов недвусмысленно поддерживаемый сенатом), позволяющий продолжать опознания личностей на предмет возможных связей сколь угодно долго (впрочем, вроде бы такая инструкция уже есть). На местах стихийно возникают первичные ячейки молодёжной организации рационалистов — помогающие выявлять скрытых иррационалов и примкнувших к ним попутчиков...
_mak_
Nov. 4th, 2002 09:13 am (UTC)
Вот-вот. Наконец-то я вижу единомышленника. :-)
mi_b
Nov. 4th, 2002 01:07 am (UTC)
About 99% of practicioners subscribed to behavioral finance long before it was invented ;) I mean very few people actually making important asset allocation decisions believe in EMH, CAPM or all other academic crap. Some of them might use EM-based models as a bechmark but very few really believe in it.

If you look at EMH from inside the market, you will realise that even given full investor rationality, information gathering and processing is a costly process. This process is endogenous for a market participant and exogenous for an observer.
_mak_
Nov. 4th, 2002 05:26 am (UTC)
I am not sure about 99%. May be 99% of small and individual investors?

I thought that large institutional investors always had some flavor of model, and they were the ones to make the weather.

But if it were true that 99% of money in the market make decisions this way, then, in your view, should we abandon the notion that markets reflect something fundamental, or do you see other mechanisms that ensure that?
mi_b
Nov. 4th, 2002 06:01 am (UTC)
I meant 99% of proffesional market participants. It is easy to "prove" this point: if you are a professional money manager believing in EMH, you should, basically, put all the money in an index fund, ensure the tracking error is small, and retire/jump out of the window (I ignore here some fine details like risk preferences asymmetry). Although some money is invested in exact index tracking funds, the number of people managing this money is small (because there are huge economies of scale in running a tracking fund.) Thus, at any given moment, the number of money managers who are alive, not retired and not believing in EMH - i.e. still trying to beat the market - is big.

This does not prove that "markets reflect something fundamental". Of course, they do. My point was that behavioral finance is not something new for the people from the Street. It is rather an attempt to close the gap between practicioner's beliefs and mainstream economic theories.
_mak_
Nov. 4th, 2002 09:07 am (UTC)
I agree, that nobody believes that markets are truly efficient. But the question is, when professionals try to beat the market are they trying to predict herd psychology (which is what behavioral finance seems to be doing) or are they trying to spot deviations between market price and the intrinsic value (whichever model you use for that). My belief (probably naive) is that they actually try to find intrinsic value, and that in their doing so market does move to that intrinsic value. Whereas if they tried to predict where all the idiots would put money tomorrow, they would behave "irrationally" themselves, and if majority of practitioners behaved that way I do not see a way for the market to reflect anything fundamental.

To your another point, I was joking when proposing to crack down on behavioral finance as a field of study. I see value in exploring almost anything that lends itself to scientific method.

And on yet another subject - are you professionally/seriously interested in this, or you are an occasional reader like I am? Because if you are serious I would love to have an opportunity to ask you some questions sometimes...
mi_b
Nov. 4th, 2002 09:44 am (UTC)
You might take a look at one of my last LJ postings - the one about finance and biology and at some earlier discussions about bubbles and rationality. Actually, it is pretty hard to give any exact meaning to the "intrinsic value" (I guess, it is usually called "fundamental value"). Is there such a thing?

This does not imply that ALL models are crap - many are quite useful, especially for derivatives. I would rather say that people's preferences, present and future, are one of fundamental aspects of the market.


Whereas if they tried to predict where all the idiots would put money tomorrow, they would behave "irrationally" themselves, and if majority of practitioners behaved that way I do not see a way for the market to reflect anything fundamental.


Behaving rationally means (vaguely) optimally using available information to optimize future well-being. So, if you are good at predicting the herd behaviour effect on stock prices, the rational thing for you is to forget about models and fundamentals and to forerun the herd.

The stuff is rather close to my professional activities - I am a practicioner with "hard science" background.
_mak_
Nov. 4th, 2002 07:04 pm (UTC)
>>Actually, it is pretty hard to give any exact meaning to the "intrinsic value" (I guess, it is usually called "fundamental value"). Is there such a thing?

Well the value exists even if we do not know it and cannot easily find. In the end everything comes to the discounted value of free cash flows/earnings/whatever, even "people's preferences, present and future". The fact that those future streams of payments are highly uncertain is true, but the same uncertainty does not stop us from valuations of options and futures. It is labor-intensive too, but that has an advantage – if not everybody can do it there is an opportunity for those who can.

Now, if we talk specifically about the subjective preferences, I agree that they are fundamental aspects of the market. (I am sort of replying to both this comment and your other post). Moreover, we can easily quantify some of them, such as premium on US-owned company vs. developing-world company, given equal fundamentals. You can argue that a developing-world company carries more risk than an American company, but even if you specifically take companies with the same risk, American company will still be valued higher. That’s a quantifiable value of a subjective preference. Another one is a premium a company gets when it hires a star CEO. Latest accounting/technology/marketing fad may carry a premium without changing fundamentals.

But the only way to get a premium for the equal good is some sort of marketing, and that’s what all those examples are – obvious and commonly known marketing ploys. They are so common that are immediately recognized and therefore give no advantage to anybody. The goal is to find something before it becomes obvious. So we may want to study all sorts of marketing tactics and if we spot one that we believe may lead to increase of subjective value we may want to bet on its success.

The problem is that even in the normal marketing of goods we rarely know whether it will be successful, let alone the marketing of the stock. Then there is a question of the precise valuation, i.e. when to sell.

The point is, if we find it hard to value the fundamentals, isn’t it ten times harder to value subjective preferences?
mi_b
Nov. 5th, 2002 05:46 am (UTC)
Well the value exists even if we do not know it and cannot easily find. In the end everything comes to the discounted value of free cash flows/earnings/whatever, even "people's preferences, present and future".

How can you tell that something exists if there is no way to observe or measure it? Discounting has to be done at acertain rate, but at which rate? Discounting depends on temporal and risk preferences of each individual To aggregate this into a single "representative" discount you have to assume, among other things, that there is an equilibrum, which may be questionable. By the way, I met very few people who would know what they risk aversion is. The standard theory here starts with several assumptions that are plainly wrong.

the same uncertainty does not stop us from valuations of options and futures It is no the same. Futures and, to a smaller extent, options, are valued from a "no-arbitrage" argument. There is no way to apply anything like that to value equities. There is no riskless arbitrage in case of a mispricing.

By subjective preferences I meant rather risk preferences and intertemporal preferences depending on market state.

I am not sure how good is your marketing analogy. Selling stock is not like selling other goods - except for IPO and new rights issues, the sellers do not have much control over company policy; the sellers are not usually engaged in any kind of advertisement campaign. Even during IPOs the selling strategy is quite special and the underwriter is not expected to maximize the IPO price.

The point is, if we find it hard to value the fundamentals, isn’t it ten times harder to value subjective preferences?

It is not hard to value past risk preferences - just make a search for something like "equity premium puzzle". The problem is that we need to know future risk preferences. Of course, it is not easy to predict them.
( 11 comments — Leave a comment )