Low Risk and High Return – Affective Attitudes and Stock Market Expectations

• Author: Alexandra Niessen‐Ruenzi,Alexander Kempf,Christoph Merkle
• Date: 01 November 2014
• DOI: http://doi.org/10.1111/eufm.12001
• Published date: 01 November 2014

European Financial Management, Vol. 20, No. 5, 2014, 995–1030

doi: 10.1111/eufm.12001

Low Risk and High Return – Affective

Attitudes and Stock Market

Expectations

Alexander Kempf

Centre for Financial Research (CFR), University of Cologne, Germany

E-mail: kempf@wiso.uni-koeln.de

Christoph Merkle

University of Mannheim, Germany

E-mail: chmerkle@mail.uni-mannheim.de

Alexandra Niessen-Ruenzi

Department of Finance, University of Mannheim, L9, 1-2, 68131 Mannheim, Germany

E-mail: niessen@bwl.uni- mannheim.de

Abstract

This experimental study investigates the impact of affective attitudes on risk and

return estimates of stocks. Participants rate well-known blue-chip firms on an

affective scale and forecast risk and return of the f irms’ stock.We find that positive

affective attitudes lead to a prediction of high return and low risk, while negative

attitudes lead to a prediction of low return and high risk. This bias increases with

participants’ confidence in their ratings and decreases with financial literacy.

Firm characteristics such as a firm’s marketing expenditures and the strength of

its brand have a positive impact on its affective rating.

Keywords: affective attitudes,risk and return expectations,behavioural finance,

affect heuristic

JEL classification: D80, G02, G11

We thank seminar participants at the FMA 2009 European conference, the IMEBE 2009

conference, the DGF 2009 conference, the University of Mannheim, and the University of

Cologne for valuable comments. We also thank the editor John Doukas and two anonymous

referees for very helpful comments. The authors acknowledge research support from the

Centre for Financial Research at the University of Cologne. All errors are our own.

Correspondence: Alexandra Niessen-Ruenzi.

C

2013 Blackwell Publishing Ltd

© 2013 John Wiley & Sons Ltd

Alexander Kempf, Christoph Merkle and Alexandra Niessen-Ruenzi

996

© 2013 John Wiley & Sons Ltd

1. Introduction

Standard finance theory predicts that rational investors trade off a stock’s risk and

expected return leading to a positive correlation of risk and expected return in equilibrium

(Sharpe, 1964). This prediction is based on rational investors who judge risk and return

of a stock solely based on fundamental information in the context of a market model.

However, several empirical papers suggest that individual investors expect a negative

correlation of a stock’s risk and return (Shefrin, 2001; Kaustia et al., 2009; Amromin

and Sharpe, 2005; Weber et al., 2013). In this paper, we investigate whether affective

attitudes towards a firm can explain these biased expectations.

Weadopt a relatively broad definition of the term affect which includes both, emotions

as affective states and attitudes as affective dispositions (cp. Clore and Gasper, 2000).

Affect thus encompasses immediate experiences of feelings and emotions, the associated

channels of mental processing, and the resulting attitudes towards objects.1The opposite

term to affect is cognition, which comprises the reason-based evaluation of situations

and objects. Psychologists and neuroscientists emphasise the outstanding role of affect

in information processing and decision making (Frijda, 1986; Ekman and Davidson,

1994; Damasio, 1994; LeDoux, 1996). They argue that affect influences all stages of

information processing and decision making, partly without conscious awareness of this

influence.

In a financial setting, where investors are supposed to evaluate the risk and return

prospects of assets, affective attitudes are likelyto get involved, in particular when assets

represent companies towards which affective attitudes are likely to exist. The way affect

operates in this context can be best described by an affect heuristic, which maintains

that people form a global perception of an object based on their affective attitude

and derive judgments and expectations about this object from this perception (Finucane

et al., 2000, Slovic et al., 2004). A positiveaffective attitude towards a companygenerates

an expectation of high return and low risk for its stock, while a negativeaffective attitude

towards a companyleads to a stock market expectation of low return and high risk. Thus,

in contrast to standard finance theory, the affect heuristic predicts a negative correlation

of risk and expected return. Since risk and return expectations are commonly regarded as

the main determinants of asset allocation decisions, the affect heuristic can thus severely

bias investors’ portfolios.

The main contribution of this paper is to investigate,whether affective attitudes towards

a company have an impact on how a person estimates risk and return of the company’s

stock. Furthermore, we analyse whether the impact of affective attitudes on risk and

return expectations depends on financial literacy of investors and their confidence in

evaluating a firm. Finally, we explore whether the management of a fir m can actively

influence which affective attitudes investors have for this firm.

We analyse the impact of affective attitudes on risk and return estimates in an

experimental setting. Participants in the experiment provide their affective evaluation

of companies on a semantic differential scale (Osgood et al., 1957), a standard tool in

psychology to elicit affectiveattitudes. Participants are further asked to predict future risk

and return of these company’s stock. They also get access to fundamental information

about the companies to prevent that an affective response in the estimation of returns is

merely due to a lack of information.

1In our experimental setting we will mostly refer to ‘affective attitudes towards companies’.

Low Risk and High Return 997

© 2013 John Wiley & Sons Ltd

Our results show that affective attitudes have a strong impact on participants’ risk

and return estimates. Positive ratings lead to an estimation of high expected return and

at the same time low risk. In contrast, negative ratings produce the opposite prediction

of low expected return and high risk. The estimates provided in the experiment clearly

violate the predictions of standard finance theory and suggest the presence of a bias

in expectation formation. The results hold no matter whether the affective ratings are

elicited before or after the risk and return estimation, i.e. we find that an immediate

activation of affective reactions before the risk and return estimations is not necessary

for affect to have an impact on these estimations.

We also find that our results are stronger for participants with lower financial literacy.

This suggests that participants with higher financial literacy are at least partly able

to cognitively correct their affective attitude when confronted with the estimation

task. Furthermore, the results are stronger for participants who indicate that they feel

confident in evaluating the firm. Conf idence thus contributes to a more affectivelydriven

evaluation. Finally, we explore which firm characteristics determine its affective rating

among participants. We find that views on company image, brand and products carry

over to investment expectations. This result supports Fehle et al. (2005), who argue that

firms can use advertising to impact investor behaviour.

There are two important implications arising from our findings. First, our results imply

that investors who are subject to an affect heuristic potentially overweight stocks which

they evaluate positively because they believe these stocks to be more profitable and

less risky than the average investment. At the same time, these investors might ignore

negatively evaluated stocks, which in combination leads to insufficient diversif ication

of their portfolio holdings. Underdiversification of individual investors’ portfolios has

been shown frequently (Grinblatt and Keloharju, 2001; Anderson, 2007; Kumar and

Goetzmann, 2008), and might at least partly be caused by affective attitudes towards

stocks. As firms are rated similarly across participants, our results might also have

an impact on aggregate stock returns. Specifically, the literature on investor sentiment

suggests that stocks with positive affective attitudes should have higher returns in the

short term and then mean revert afterwards (Baker and Wurgler, 2006). Our results

confirm that this might indeed be the case. Second, our results imply that firms with a

positive affective rating might attract a different shareholder clientele than f irms with a

negative affective rating. We show that financially illiterate participants are particularly

prone to the affect heuristic, which is in line with the general finding that especially

individual investors are often subject to behavioural biases (Barberis and Thaler, 2003).

This might have important consequences for firms since a fir m’s shareholder structure

might determine which policies it can adopt (Hartzell and Starks, 2003; Graham and

Kumar, 2006).

Our study builds on the growing literature on the importance of emotions and affective

attitudes for economic decision making (Elster, 1998; Loewenstein, 2000; Loewenstein

et al., 2001) and particularly financial decision making (Goetzmann and Zhu, 2005;

Subrahmanyam, 2008). Forinstance Loewenstein et al. (2001) and Hirshleifer (2001) lay

down a conceptual foundation of the impact of emotions on financial decision making.

Our results empirically support the views expressed in this literature by showing that

affective attitudes play an important role in the estimation of a stock’s risk and return.

Further we extend the literature on the impact of the affect heuristic on information

processing. MacGregor et al. (2000) introduced the affect heuristic into finance and show

that affect is important for evaluating classes of securities, especially when fundamental

information is scarce. Our paper illustrates that affect is important for the evaluation of