Rational-efficient financial markets
ANOMALIES, BEHAVIORAL FINANCE?
Following an initial period of success in the 1970s and 1980s, the economic hypotheses of rational behavior and efficient markets have gained a bad reputation during the last two decades. So-called anomalies in empirical studies abound and many economists and market participants seem to be infatuated with a new paradigm, called behavioral finance.
Nevertheless, consider this:
* Anomalies are nothing more than empirical observations awaiting a rational, fundamental explanation. Obviously, anomalies do not fit the usual stylized and simplified textbook theories that we all know. But reality is more complicated than theory; that is why we make and use theories (Fama, 1991). Explaining empirical facts to the nth statistical decimal takes much effort and time, but slowly we are getting there.
* So-called anomalies that already seem to have found rational solutions (using market imperfections such as transaction costs, limited information and asymmetric information problems, methodological mistakes in research, etc.) remain alive in the popular literature. This begs the question: Do we really want an explanation? Why do many people find these anomalies and psychological stories so fascinating and won't let them die (Bowman and Buchanan, 1995)?
A brief preview to understanding financial market anomalies
* EFFICIENT MARKETS VERSUS THE RANDOM WALK HYPOTHESIS
Definition: “A market is efficient with respect to information set It if it is impossible to make economic profits by trading on the basis of this information set.” (Jensen, 1978; Levich, 1985; Ross, 1987). I.e. there must be no arbitrage opportunities, after costs, after risk premium, using only ex ante information
Most textbooks and many academic articles continue to suggest that efficient markets imply a random walk behavior in market prices. This is wrong. The random walk hypothesis (Samuelson, 1965) is a special and extreme hypothesis that can be derived from the basic fundamental value model of asset prices only when one is willing to assume:
1* that cash flows (stock dividends) and discount rates (required returns) in the present value calculation (dividend discount model) are constant or can only change permanently (no temporary changes allowed) and immediately (no effects of expected future changes);
2* that the special and extreme hypothesis of perfect-foresight-with-error expectations is a valid description of rational expectations, i.e. that economies are of the stationary stochastic type and that information is completely transparent. (For further details on the misperceptions about rational expectations, see my other page
None of these special assumptions is a necessary component or a correct approximation of reality.
It is easy to show that, for example, temporary changes in dividends or expected future changes in dividends require predictable changes in stock prices (i.e. capital gains/losses) in order to keep required and expected returns for new investors unchanged. The exchange rate overshooting model is another well-known example of non-random but rational asset price behavior.
Thus, understand this quote: “Of course [!!!], as Leroy (1973) and Lucas (1978) have shown, the unforecastability of asset returns is neither a necessary nor a sufficient condition [!!!] of economic equilibrium. And, in view of the empirical evidence in Lo and MacKinlay (1988b), it is also apparent that historical stock market prices do not follow random walks." (Lo and MacKinlay, 2001, p.115, emphasis added)
Or else, LeRoy (1989, p.1613, emphasis added): "This survey should by now have made amply clear that the transition between the intuitive idea of market efficiency and the martingale model is far from direct. Few financial economists, surprisingly, have taken direct issue with the prevailing practice in the finance literature of identifying market efficiency with the validity of a particular specialized model of equilibrium in financial markets," [...] (p.1615) "Market efficiency is a complex joint hypothesis. Some elements of this joint hypothesis are central to economists' way of thinking, like rationality and rational expectations, while others are no more than convenient auxiliary assumptions, like the martingale model."
Note: Also see my pages on the random walk myth and rational expectations.
* THE THEORETICAL ASSUMPTION OF PERFECT ASSET MARKETS
Theories are abstractions of reality. We make simplifying assumptions in order to focus on some interesting phenomenon or question. One of the key simplifying assumptions in financial market analysis is the assumption of "perfect asset markets". Meaning:
1* perfect competition (atomistic market, with infinite numbers of buyers and sellers)
2* zero transaction costs
3* information costless and completely transparent
4* all income claims complete divisible and tradable
5* lending and borrowing at single, uniform interest rate
6* no (government) regulation that influences behavior of market participants (e.g. tax system)
None of these special assumptions is literally correct in reality.
Desiring to understand observed market anomalies requires us to develop more complicated descriptions of behavior in cases where one or more of the previous simplifying assumptions fail. That is sometimes extremely difficult.
Of course, many studies have examined some of the previous assumptions, usually on a case-by-case basis. Sometimes, the conclusion is that although relaxing the specific assumption helps, it doesn't eliminate the anomaly entirely. My suggestion: don't reject this approach to the solution of market anomalies, but add one of the other assumptions. Market imperfections are not mutually exclusive.
* METHODOLOGICAL ERRORS IN RESEARCH
Theories should be tested against empirical observations. Thus, economists derive hypotheses or specific observable implications from economic theory and also devise empirical methodologies to be used. However, sometimes what appears to be logical is nevertheless incorrect. A basic mistake is of course to confuse statistical significance of estimated relationships with causality or economic significance. To confuse absolute returns with abnormal returns adjusted for time-varying interest rates and risk is another. Another mistake is to assume that the normal probability distribution applies to all risk analysis, ignoring skewness and fat tails. Other mistakes are more subtle and often highly technical in nature. Discovering these methodological errors or biases is difficult, but they do exist and unfortunately also persist.
Data errors or measurement errors are also a likely candidate for incorrect empirical conclusions. Too many researchers appear to not invest in the institutional knowledge necessary to correctly interpret the data available to them.
* THE MARKET FOR ANOMALIES
Many more researchers seem to be interested in perpetuating or expanding on anomalies than solving or understanding the associated problems. One reason may be that there exists a profitable market for anomalies. Do you want to sell your financial market experience on anomalies? Do you want to quickly publish a few more research articles? Try this: Take the most stylized version of an economic theory, almost to the extent of it being a caricature. [Never mind that developers of theories usually simplify their exposition only for them to be able to focus on what they consider to be the key arguments, and not normally intend these simplified theories to become the "true version" of their theory. Never mind that economic theories usually include the ceteris paribus condition that should not be ignored in practice. Etc.] Show that an economic theory doesn't capture all aspects of reality and consequently reject all its implications. Provide your own solution to the puzzle and preferably give the anomaly a new name to distinguish yourself. Start your own investment fund and offer to exploit the 'free lunch' in investment returns.
Subpages:
BIASES AND ERRORS IN EMPIRICAL STUDIES
Over time, discussions about empirical studies have turned up many possible biases and errors in databases and empirical methods. A survey.
With regard to the efficient market hypothesis, most stock market anomalies do not survive the basic hurdle of trading costs. What are these trading costs?
A review of some high profile stock market anomalies. What do we really know? Are they relevant?
Selected general literature:
* Berk, J.B., A critique of the efficient market hypothesis, working paper, preliminary, February 2008. http://www.vgsf.ac.at/activities/JBerk1.pdf
* Rubinstein, M. (2001), Rational markets: yes or no? The affirmative case, Financial Analysts Journal, vol.57 (3) May/June: 15-29 http://www.haas.berkeley.edu/groups/finance/WP/rpf294.pdf
* Bowman, R.G. and J. Buchanan (1995), The efficient market hypothesis - A discussion of institutional, agency and behavioural issues, Australian Journal of Management, vol.20 (2) December: 155-66. http://www.agsm.edu.au/~eajm/9512/pdf/bowman.pdf
* Falkenstein, E. (2001), A review of Shleifer's Inefficient Markets: An Introduction to Behavioral Finance. http://www.efalken.com/papers/inefficient_markets.htm
* Stout, L.A. (2005), Inefficient markets and the new finance, Journal of Financial Transformation, vol.14 August: 95-105 http://www.law.uiuc.edu/content/conferences/capitalmarkets/pdf/stout_inefficient_markets.pdf
* Schwert, G.W. (2003), Anomalies and market efficiency, G.M. Constantinides, M. Harris, R. Stulz (eds.) Handbook of the Economics of Finance, Elsevier: 937-72 http://schwert.ssb.rochester.edu/hbfech15.pdf
* Malkiel, B.G. (2003), The efficient market hypothesis and its critics, Journal of Economic Perspectives, vol.17 (1) Winter: 59-82 http://www.princeton.edu/~ceps/workingpapers/91malkiel.pdf
* Davis, J.L. (2006), The informational efficiency of stock prices: A review. http://www.dfaus.com/library/articles/informational_efficiency_of_stock_prices/
* Fama, E.F. (1998), Market efficiency, long-term returns, and behavioral finance, Journal of Financial Economics, vol.49 ( ) : 283-306. http://introduction.behaviouralfinance.net/Fama98.pdf
* Fama, E.F. (1991), Efficient capital markets II, Journal of Finance, vol.46 (5) December; 1575-1617 http://student.bus.olemiss.edu/files/Riley/FIN%20633/Market%20Efficiency/More%20Articles/Fama%201991%20JF.pdf
* An interview with Eugene Fama (1997) http://www.dfaus.com/library/reprints/interview_fama_tanous/
High returns as compensation for risk? What risk? Surely, we all know beta is dead !?
* Failed wizards of Wall Street (1998), http://www.businessweek.com/1998/38/b3596001.htm
* For Wall Street's math brains, miscalculations (2007), http://www.washingtonpost.com/wp-dyn/content/article/2007/08/20/AR2007082001846.html?hpid=topnews
* Foster, D.P. and H.P. Young (2008), Hedge fund wizards, Economists' Voice, February http://www.bepress.com/ev/vol5/iss2/art1/ discussed in many newspapers, for example M. Wolff, Why today's hedge fund industry may not survive, Financial Times March 18 2008 http://us.ft.com/ftgateway/superpage.ft?news_id=fto031820081426064497&page=1
* Ross, S. (2006), A neoclassical look at behavioral finance: A tale of two anomalies, Nomura Lecture Oxford Univ. www2.maths.ox.ac.uk/mcfg/ncmf/Two Anomalies Talk - Oxford.ppt