EX ANTE RISK PREMIA AT THE N Y S E ANALYSIS OF EXPERTS' BEHAVIOUR AT THE INDIVIDUAL LEVEL

Alain Abou

Description

Niveau: Supérieur, Doctorat, Bac+8
1 EX-ANTE RISK PREMIA AT THE N.Y.S.E.: ANALYSIS OF EXPERTS' BEHAVIOUR AT THE INDIVIDUAL LEVEL Alain Abou* and Georges Prat** * Research Associate Professor, CNRS - ** Research Professor, CNRS Corresponding author: MODEM, PARIS-X University 200, avenue de la République 92001- Nanterre Cedex France Abstract – The ex-ante risk premium is defined as the spread between the expected return related to a portfolio of industrial stocks and the riskless rate. The expected return of industrial stocks at the NYSE (S&P400 industrial index) is deduced from surveys carried out by J. Livingston on a panel of experts for one and two semesters time horizons, while the riskless rate is given by zero coupon bonds with maturities corresponding to the forecasts horizons. It then becomes possible to compute about 3000 individual values of ex-ante risk premia over the period 1952-1993. With respect to the ex-post market premium analysed in the literature, these ex-ante premia offer the main advantage to be based on information available at time t of the financial decisions. Three main lessons may be drawn from our study. First, the values of these ex-ante premia seem rather realistic and especially more than those of the ex-post ones considered in the literature.

• industrial stock price


• premium


• p400 industrial


• ante premia


• market premium


• individual ex-ante


• ex-ante risk


• stock returns


• post premium

Sujets

• Analysis
• Livingston
• Premium

Informations

• Publié par: profil-zyak-2012
• Nombre de lectures: 16
• Langue: English

Extrait

EX-ANTE RISK PREMIA AT THE N.Y.S.E.: ANALYSIS OF EXPERTS BEHAVIOUR AT THE INDIVIDUAL LEVEL   Alain Abou* and Georges Prat**   * Research Associate Professor, CNRS - ** Research Professor, CNRS Corresponding author: [e-mail]   MODEM, PARIS-X University 200, avenue de la République 92001- Nanterre Cedex France         Abstract  Theex-ante risk premium is defined as the spread between the expected return related to a portfolio of industrial stocks and the riskless rate. The expected return of industrial stocks at the NYSE (S&P400 industrial index) is deduced from surveys carried out by J. Livingston on a panel of experts for one and two semesters time horizons, while the riskless rate is given by zero coupon bonds with maturities corresponding to the forecasts horizons. It then becomes possible to compute about 3000 individual values ofex-ante risk premia over the period 1952-1993. With respect to theex-post market   premium analysed in the literature, these ex-antepremia offer the main advantage to be based on information available at timetof the financial decisions. Three main lessons may be drawn from our study. First, the values of these ex-antepremia seem rather realistic and especially more than those of theex-postones considered in the literature. Second, the professional affiliation which may determine the information used by forecasters to build their expectations appears to be a significant criterion discriminating the risk premia. Third,ex-antepremia depend both on general factors (macroeconomic variables) and individual factors (i.e. agents forecasts about inflation and industrial production growth rate), the last ones contributing to explain the heterogeneity between experts.   Key words:    ex-ante risk premium; stock market; individual survey data; expectations heterogeneity.   JEL classification : D81 ; D84 ; E44 ; G12 ; G14.                              

 

1

EX-ANTE RISK PREMIA AT THE N.Y.S.E.: ANALYSIS OF EXPERTS BEHAVIOUR AT THE INDIVIDUAL LEVEL  

    1 - INTRODUCTION     The risk premium of a risky equity or a portfolio is an excess return defined as the difference between the expected return of this asset and the riskless rate. Assuming efficient financial markets, so that asset price expectations are rational, the theory offers three frameworks to explain the value of the excess returns required by investors to hold risky assets.   The Capital Asset Pricing Model (CAPM) suggested by Sharpe (1964) defines the first approach. According to this model, only the systematic risk is paid to investors and then the risk premium of an equity or a portfolio is proportional to the whole market risk premium, the coefficient of proportionality being the so-called beta coefficient. The failure of the CAPM to explain risk premia suggests either to relax the partial equilibrium hypothesis on which this model is based, or in a more radical way, to replace the equilibrium of the investor portfolio by the arbitrage principle. This leads to the two others following approaches.   The general intertemporal equilibrium framework (Lucas, 1978), according to which the representative agent maximizes his expected utility, gives rise to the second approach figured by the Consumption-CAPM. Similarly to the CAPM, this model enables to specifya priorithe factors of the risk premium that is, the conditional covariance between stock returns and the marginal utility of consumption. The many debates about the so-calledequity premium puzzle and thevolatility puzzle ends today in the idea that if, at the best, the first two moments of stock premia may be represented, this model still cannot explain the values of premia per date. 1            Substituting the arbitrage paradigm to the equilibrium one, the third approach suggests that common factors may be extended to any macroeconomic variable: this is the Arbitrage Pricing Theory(APT) suggested by   Ross (1976). This model says that when no arbitrage opportunities prevail in the market, the risk premium of an equity or a portfolio depends on a linear form of a set of common factors that might include the market factor. At the opposite of the previous ones, this approach does not limita priori the number of common factors to the single market factor, although only the non-diversifiable risk remains rewarded. Hence, commons factors are not restricted to the market (cf. CAPM) or to consumption (cf. C- CAPM). However, the main drawback of the APT is that neither the nature of each factor nor their number are specifieda prioriby the model. As a result, only empirical analysis allows identifyinga posteriori the factors of risk premia. In fact, macroeconomic variables frequently appear to be relevant factors: inflation rate, production growth rate, spreads of interest rates, rate of increase in money supply, etc 2 .      In this paper, we follow an approach based on the conditional APT, so that the common factors of the risk premia will not be identified by theory, but by empirical analysis. Moreover, empirical tests of the APT require an hypothesis to represent expected stock returns,                                                   1  See the Campbell and Cochrane (1999) approach based on consumption habits, and the one of Cechetti, Lam and Mark (2000) based on distorted expectations. 2  See in particular Roll and Ross (1980); Chen, Roll and Ross (1986); Elton, Gruber and Mei (1994).

 

2