Introduction

Behavioral finance is considered one of the most controversial branches in the New Finance area beside the fractal markets hypothesis (FMH) and the artificial neural networks (ANN) which are gaining wider attention from financial economic community both academically and professionally. Its approach and building blocks have brought the underlying assumptions of Modern Finance theory and evidence into question. Although in its beginning stage, behavioral finance has been explored by many reputable financial economists who conducted extensive research and employed non-economic paradigms from its sister social science disciplines such as individual behavior and decision-making under uncertainty from psychology, group dynamics and decisions from sociology, and collective decision-making from political science. Their attempts have yielded practical and fruitful insights about many puzzles and anomalous phenomena in both the financial markets and the corporate settings, which are congruent with the analytical framework of positive theory suggested by Friedman (1953).

 

Section two contrasts between the assumptions of modern finance and those of the behavioral finance which would enable us to understand the building blocks and the logic behind the area. Section three provides the theoretical foundation for behavioral finance by discussing the "individual primers" that could partially serve as the logical explanations for market reality as opposed to market optimality. In Section four and Section five, the "market and firm domains" are discussed in relation to the previous section on individual primers. Section six concludes the theme of this essay and presents my expectations for its future direction.

 

Modern vs. Behavioral Finance Theory Back to Top

Jegadeesh (1995) asserts that modern finance theory is built on the notion that the actions of all economic agents are guided by the criterion of maximizing expected utility. In the normatively constructed model, all agents take actions based upon rational expectations with the absence of non-economic factors in their utility functions. Fridson (1994) views modern finance along the same line as Jegadeesh does that the assault on the efficient markets hypothesis (EMH) has progressed well beyond the identification of minor anomalies. Based upon the normative assumptions and the EMH, rational and fully informed investors (i.e., representative agents) quickly eliminate any tendency of a class of securities to deviate, on a risk-adjusted basis, from a market rate of return and thereby forcing their prices to resume or converge to their equilibrium level which reflect their true fundamental values. In light of these critical perspectives, we can recapitulate that modern finance is a normative theory based on ex ante assumptions about representative economic agent's rational behavior that individuals are: 1) risk-averse expected utility maximizers; 2) unbiased Bayesian forecasters; and 3) rational-expectations decision-makers.

 

Behavioral finance, as further noted by Jegadeesh (1995), takes the position that not all economic decisions can be described by the equilibrium conditions in the efficient market economy. Thaler (1994) states that sometimes, in order to fine the solution to an empirical puzzle, it is necessary to entertain the possibility that some of the economic agents behave less than fully rationally some of the time. The notion of less-than-fully-rational actions or biased expectations of economic agents is prevalent and easily witnessed in the market reality. In order to characterize actual decision-making behavior which underlies the operations of the actual markets, some financial economists began to incorporate positive paradigms into the process of economic agents' decision-making under uncertainty in hope to find some logical explanations as to why markets do not function as modern finance theory has predicted. Along with this line of argument, we can say that behavioral finance is a positive theory based on ex post realizations of economic decision-makings involving the psychology of individuals that they are: 1) loss-averse expected regret minimizers, 2) biased non-Bayesian forecasters, and 3) naive-expectations decision-makers.

 

The proponents of positive theory such as Kuhn (1970) suggests that behavioral finance should start with the study of market anomalies, i.e., empirical facts for which there is wide agreement that the standard paradigm lacks explanatory power. The joint hypothesis of EMH and capital asset pricing model (CAPM) which is held sacredly by most modern financial economists is now under severe attack by a series of puzzles from which empirical studies have uncovered through the use of advanced statistical techniques and the abundance of market price information. Haugen (1993) comments that actual securities markets do not always exhibit desirable EMH and CAPM characteristics and are pervasively anomalous that: 

1. Stock prices do not respond quickly and accurately to new information.
2. Trading rules and investment strategies do not fail to produce abnormal returns.
3. Professional investors and money/portfolio managers do not fail to produce abnormal returns.
4. Changes in expected returns are not completely captured by the term- and risk-premia.
5. Excess returns driven by factors other than riskfree rate and the beta are not entirely stochastic.

Theoretical Foundation and Individual Primers in Behavioral Finance Back to Top

De Bondt and Thaler (1994) criticize that although modern finance usually makes predictions about market outcomes and firms' behavior, there is an underlying set of assumptions about individual behavior that are used to derive these predictions. However, this set of assumptions can be counter-argued on two fronts, that some assumptions are invalid, e.g., economic agents often violate the substitution axiom of expected utility theory, and that the whole set is incomplete, e.g., the theory has little to say about important aspects of economic behavior such as the role of social norms. Therefore, we have to arrive at a better characterization of individual behavior in order to enhance the currently acceptable finance theory by embarking onto the new domains upon which modern finance has not yet explored. The following behavioral concepts and their contributors can be referred to as the individual psychological primers which seem to be very useful in the area of behavioral finance: 

1. Loss Aversion and Prospect Theory: Markowitz (1952), Kahneman & Tversky (1974, 1979)
2. Framing and Mental Accounting: Tversky & Kahneman (1981) and Thaler (1985)
3. Regret Avoidance, Responsibility Shifting, and Agent's Prudence: Kahneman & Tversky (1982) and Shefrin & Statman (1993)
4. Cognitive Errors and Non-Bayesian Learning & Forecasting: Tversky & Kahneman (1971, 1982)
5. Overconfidence and Biased Expectations: Lichtenstein, Fishchhoff, and Phillips (1977)
6. Over/underoptimism and Naive Expectations: De Long et al. (1990), Lee et al. (1991), Lakonishok et al. (1994), Bauman & Dowen (1988), and La Porta (1995)
7. Fashions, Fads, and Popular Models: Aronson (1991) and Shiller et al. (1990)

Loss Aversion and Prospect Theory 

In his seminal paper on portfolio selection theory, Markowitz (1952) also suggests semi-standard deviation as being a more appropriate measure of risk than the standard deviation. In the field of mathematical psychology, Kahneman and Tversky (1971, 1979) postulate that individuals are loss-averse rather than risk-averse because their pain associated with a given amount of loss is greater than their pleasure derived by an equivalent gain, i.e., losses loom larger than gains, which is the theme of their prospect theory.

 

Framing and Mental Accounting

 

Form of payments and framing of probabilities for identical payoffs matter in the eyes of investors (Tversky and Kahneman, 1981). Loss-averse investors try to minimize their expected loss from making wrong investment decisions by using mental accounting, such as setting aside special accounts to discipline payments or receipts, to mitigate self-control problems (Thaler, 1985)

 

Regret Avoidance, Responsibility Shifting, and Agent's Prudence

 

Regret is the feeling of ex post remorse about a decision that leads to a bad outcome. One strategy for investors to avoid regret is to shift their responsibility for such a decision to the more prudent agents (Shefrin and Statman, 1993). Holding the quality of decisions constant, if the principal-agent relationship reduces regret, the investors' expected utility would increase.

 

Cognitive Errors and Non-Bayesian Learning

 

Kahneman and Tversky (K&T) argue against the normative rational expectations assumption of Bayesian learning and forecasting that, in reality, individuals often commit cognitive errors by 1) underweighing the base-rate information (K&T, 1971) and 2) mismapping the probability (K&T, 1982) when evaluating future stock prices and expected returns given available information, which result in either overreaction or underreaction.

 

Overconfidence and Biased Expectations

 

Lichtenstein, Fishchhoff, and Phillips (1977) find that individuals are overconfident and frequently overestimate the reliability of their knowledge. Professional investors are more confident of their predictions in fields where they have self-declared expertise, holding their predictive ability constant.

 

Overoptimism and Naive Expectations

 

Contrarian strategies like naive extrapolation of past earnings growth (Lakonishok, Shleifer, and Vishny, 1993) and naive reliance on security analysts' forecasts of future earnings growth of well-known stocks despite their inherent overoptimism (Bauman and Dowen, 1988; La Porta, 1995) are examples of uninformed investors' naive expectations and time-varying sentiments.

 

Fashions, Fads, and Popular Models

 

Aronson (1991) points out that individuals are influenced by their social environment and that they often feel pressured to conform. Shiller, Kon-ya, Tsutsui, and Case (1990) attribute speculative markets phenomena like the stock market clash of 1987, the real estate boom of 1988, and the underpricing of IPOs to popular models where naive investors form their expectations based upon aggregate opinions and rules of thumb.

 

Empirical Puzzles in Financial Markets and Behavioral Explanations Back to Top

Based upon the positive theory of individual decision-makings presented and discussed above, we can relate it to the empirical findings about the market anomalies for which those constructs can provide viable explanations. However, behavioral explanations should not be regarded as the only cause of such anomalies, as market and fundamental factors which are not allowed to enter into the traditional assumptions such as transaction costs, agency costs, market microstructure, government's interventions through taxes, subsidies and regulations, and the advancement in information technology can have substantial impacts on the markets. Some of the empirical puzzles and their accompanying behavioral contributors are listed as follows: 

1. Price Volatility Puzzle: Black (1986), Shiller (1981, 1984) and Summers (1986)
2. Trading Volume Puzzle: Lowenstein (1988) and De Bondt (1992)
3. Contrarian Strategies Puzzle: Basu (1977), Shefrin & Statman (1985), De Bondt & Thaler (1986, 1987), and Jaffe, Keim, Westerfield (1989)
4. Momentum Strategies Puzzle: Chan, Jegadeesh, and Lakonishok (1996)
5. Closed-End Mutual Funds and Initial Public Offerings (IPOs) Puzzles: Lee, Shleifer, and Thaler (1991), Ibbotson, Sindelar, and Ritter (1988), and Ritter (1991)
6. Equity Premium Puzzle: Mehra & Prescott (1985)

Price Volatility Puzzle and Non-fundamental (Noise) Factors

 

Black (1986) relates the volatilities in price and value to the rate of arrival of public information about the firm's earnings and the interpretation of such information by rational and noise traders through their trading activities. Shiller (1981) examines the volatility of stock prices and concludes that it is too high to be justified by fundamental information about the firm's earnings prospects. Summers (1986) criticizes that stock price efficiency and volatility cannot be captured merely by using the standard CAPM, but does not reject the EMH. Noise factors are found to be non-random and create excess price volatility. De Long, Shleifer, Summers, and Waldmann (1990) observe that not only the noise trader risk is non-random, but it also increase the fundamental risk of the rational traders.

 

Trading Volume Puzzle and Overconfidence

 

In an absence of noise traders, heterogeneous beliefs among rational traders will not generate excessive trading if rationality is common knowledge. In reality, the high trading volume produced by well-informed institutional investors through their active portfolio management is much higher than the one produced by individual traders. De Bondt (1992) states that overconfidence of portfolio managers and agency relationship between clients and money managers can be used to explain the trading volume puzzle and the belief that professional investors can outperform the market.

 

Contrarian & Momentum Strategies Puzzle and Cognitive Errors

 

The value-based investment strategies such as buying losers/selling winners (Shefrin and Statman, 1985), buying low P/E ratio stocks (Basu, 1978; Jaffe, Keim, and Westerfield, 1989), buying low M/B ratio and low past return stocks (De Bondt and Thaler, 1985, 1987) tend to generate unusual returns because the prices of losing firms are irrationally depressed by naive investors and/or biasedly undervalued by analysts who commit cognitive errors (overreaction) in favor of the winners. De Bondt and Thaler (1987) theorize that contrarians are compensated for bearing the perceived risk (based on their subjective probability beliefs), rather than the true risk (based on market objective probability measure), in order to explain this puzzle using the standard CAPM. Momentum strategies (Chan, Jegadeesh, and Lakonishok, 1996) benefits from investors' underreaction to earnings reports that good news follows good, and bad news follows bad.

 

Closed-End Mutual Funds & IPO Puzzles and Overoptimism

 

The share prices of the closed-end mutual funds, net of agency costs, should be sold at premium rather than discount. The puzzle is that they usually sell at discount. De Long, Shleifer, Summers, Waldmann, (1990) interpret this anomaly in the context of noise traders risk that when they are optimistic, the discount is low. Since rational traders are facing two types of risk - fundamental and noise trader - they would only buy closed-end funds at discount. In the case of IPOs, they appear to be underpriced initially and overpriced in the long run. Ibbotson, Sindelar, and Ritter (1988) and Ritter (1991) find that the average IPO outperformed the market on its first trading day. This puzzle can also be explained by investors' overoptimism about the growth potential of the IPOs despite their realized disappointment and by attributing to Shiller's popular models.

 

Equity Premium Puzzle and Loss Aversion/Mental Accounting

 

The return differential between stocks and the risk-free interest rate appears to be too large (about 6%) to be consistent with standard CAPM. Mehra and Prescott (1985) examine the representative investor's risk aversion in relation to the historical equity premium and find that the level of risk aversion is excessive. Benartzi and Thaler (1993) use the concept of loss aversion and mental accounting (frequency of and horizons for portfolio evaluation) to explain this puzzle. They find that the shorter the horizon together with the higher the loss aversion, the larger the equity premium would be. If the horizon of the typical investor were 20 years, the equity premium would fall to around 1.5%.

 

Empirical Puzzles in Corporate Finance and Behavioral Explanations Back to Top

Using an economic definition offered by March and Simon (1993), a corporation is a system of coordinated action among individuals and groups whose preferences, information, interests, or knowledge differ. Modern finance theory should also be able to explain why organizations exist and function the way they normally do. The Modigliani-Miller's (1958) or MM propositions for capital structure and dividend policy irrelevancies are held by corporate finance theorists and practitioners as valid under the strict assumptions of frictionless market. That is, in the absence of agency costs, taxes, and other frictions, the assignment of property rights should not affect either the firm's operations or its market value. As a result of the MM propositions, modern theory in corporate finance seeks the various ways in which taxes, information asymmetries, and self-interest in agency relationship change optimal financing and investment decisions as well as the economic forces that push the corporation toward its optimal ownership structure. The decision-making behavior of various stakeholders (i.e., stockholders, bondholders, management team, employees, suppliers, and customers) that make up the firm becomes interrelated and influential in the design of optimal claims structure. Insofar as actual decisions differ from their theoretically normative ideal, corporate finance requires a new analytical framework. The empirical puzzles to be discussed below are examples of how behavioralism can offer plausible explanations and assist corporate financiers to achieve their desirable contractual organizational structure: 

1. Shareholders' Preference for Dividends: Shefrin & Statman (1984)
2. Investors' Perceptions of Earnings Reports: Brealey & Myers (1984), Schipper (1989), and Hand (1989)
3. Managements' Decisions to Invest, Divest, and Reorganize: Staw (1976), Roll (1986), Samuelson & Zeckhauser (1988), and Jensen (1993)

Dividends Preference and Shareholders' Mental Accounting/Self-Control

 

With the higher tax rate on dividends than on capital gains, shareholders should prefer stock repurchase to cash dividends. In fact, the reverse is preferred. Shefrin and Statman (1984) explain this anomaly based on mental accounting and self-control behavioral concepts that investors resist dipping into their stocks but are more comfortable consuming from the flows. Dividends are savored as a separate gain when the stock price rises and consoled as a silver lining when the price falls. Financing consumption out of dividends further avoids the anticipated regret of selling a stock that rises in value. They also suggest the clientele effects to differentiate the preference for high dividends payout of older investors from the preference for low dividends payout of younger ones.

 

Earnings Report Perception and Investors' Overoptimism

 

Investors and managements alike seem to prefer a steady upward trend in earnings with clear future targets. Brealey and Myers (1984) find that investors suffer from financial illusions by accounting manipulation which leads them to be too optimistic. Schipper (1989) contends that earning management is beneficial in providing a means for managements to reveal their private information. Hand (1989) finds that many firms report gains on debt-equity swaps in order to smooth a transitory fall in earnings.

 

Entry, Exit, Reorganization and Managements' Overconfidence

 

Hite, Owers, and Rogers (1987) show that market often reacts positively to sell-offs and corporate retrenchment (exit) than takeovers and corporate expansion (entry). Yet, entry decisions are less difficult to make than exit decisions. Jensen (1993) attributes the difficulty of exit decisions to information asymmetry, agency cost, and the mind set of managers. Managements of firms with large cash flow often invest in even more money-losing projects. Entrapment (Staw, 1976), hubris hypothesis (Roll, 1986), and status-quo bias (Samuelson and Zeckhauser, 1988) are used as behavioral explanations for irrational entry commitments and mergers and acquisitions. Psychologically, managements of bidding firms are convinced that they can run the target firms better than current managements. As a result, they systematically overestimate the benefits of M&A synergy even through they will confront with the winner's curse where the M&A costs are highly overpriced.

 

Conclusion Back to Top

The distinction between modern and new finance is more clear-cut than ever due to their different theoretical approaches -- normative and positive. Behavioral finance takes its firm stance in the new finance school of thought by attacking the underlying assumptions of modern finance theory which follow the optimality principle and frictionless market and using the more mundane empirical observations and the experimental results of extensive studies in other social science disciplines such as psychology, sociology, and political science.

 

On the one hand, excessive reliance on normative assumptions of modern finance can incur substantial costs. First, it focuses more on the outcomes of the ideal individual decision makers and less on the actual decision-making process itself. Second, the normative assumptions sometimes lead to unrealistic characterization of individuals who directly affect the market operations which result in absurd rationalization. And third, the current theories tend to be more immuned from falsifiability without the confirmation or disconfirmation from empirical findings and alternative paradigms, thereby limiting the progress in theoretical development. On the other hand, we also cannot fully rely on the positive approach in behavioral finance alone for us to be called naive believers without the well-structured theoretical guidance from the normative approach. No matter what the circumstance turns out to be, we have to be versatile and critical enough in adopting and criticizing both approaches while opening ourselves for new and other unorthodox dimensions.

 

We can benefit a lot from the vast amount of psychological, sociological, and even political science research done to date by making financial economics more interdisciplinary and receptive to the changes within. Other contacts with physical and computer science disciplines have been progressively made through the fractal and chaos nonlinear dynamic models as well as the artificial neural networks. Financial economics in the next century should be no less interesting than in the past half century - from Markowitz's Portfolio Theory to Fama's EMH and Sharpe-Lintner-Mossin's CAPM to Black-Scholes-Merton's Option Pricing Models. Behavioral-, fractal-, and neural-based asset and derivative pricing models are all the challenges for New Financial theorists and empiricists in the years to come.