How Do Ambiguity and Risk Aversion Affect Price Volatility under Asymmetric Information?

• Date: 01 August 2015
• DOI: http://doi.org/10.1111/ajfs.12101
• Published date: 01 August 2015

How Do Ambiguity and Risk Aversion

Affect Price Volatility under Asymmetric

Information?*

Guangsug Hahn

Division of Humanities and Social Sciences, Postech

Joon Yeop Kwon**

Graduate Program for Technology and Innovation Management, Postech

Received 31 October 2014; Accepted 1 June 2015

Abstract

This paper investigates the effects of ambiguity and risk aversion on asset price volatility

when uninformed traders face ambiguity. We find that the effects of ambiguity on price

volatility depend on the degree of risk aversion. If the degree of risk aversion is sufficiently

low, then ambiguity has little influence on price volatility, even when the degree of ambiguity

is extremely high or almost all traders have ambiguous information. In contrast, if traders

are sufficiently risk-averse, ambiguity effects on price volatility are amplified by the degree of

risk aversion.

Keywords Ambiguity; Ambiguity premium; Asymmetric information; Price volatility; Risk

aversion

JEL Classification: D81, D82, G12

1. Introduction

During financial crises, asset prices may fluctuate drastically over a short period. A

recent example is the United States financial crisis in 2008, which was triggered by

*We would like to thank the editor, an associate editor, and two anonymous referees for their

valuable comments and suggestions, which substantially improved our paper. We are also grate-

ful to Hong Chong Cho and Beum-Jo Park for valuable comments at the Joint Conference with

Allied Korea Finance Associations in 2014. We thank Hyeng Keun Koo and Dong Chul Won

for their helpful comments. Hahn would like to thank Chenghu Ma for his hospitality during

the visit to the School of Management at Fudan University, where a part of the paper was writ-

ten. This work was supported by the National Research Foundation of Korea Grant funded by

the Korean Government (NRF-2008-332-B00139). The usual disclaimer applies.

**Corresponding author: Joon Yeop Kwon, Graduate Program for Technology and Innova-

tion Management, POSTECH, Pohang 790-784, South Korea. Tel: +82-54-279-2692, Fax:

+82-54-279-2694, email: jykwon@postech.ac.kr.

Asia-Pacific Journal of Financial Studies (2015) 44, 616–634 doi:10.1111/ajfs.12101

616 ©2015 Korean Securities Association

the collapse of the subprime mortgage markets, as a result of the housing market

bubble bursting. The Dow Jones Industrial Average fell by more than 50% and the

VIX index (implied volatility) reached 68.01% during that event (Schwert, 2011).

Economists have attempted to explain why stock prices are unreasonably volatile.

The prior literature asserts that risk aversion or ambiguity (ambiguous information)

contributes to increasing asset price volatility. Grossman and Shiller (1981), LeRoy

and LaCivita (1981), and Lansing and LeRoy (2014) show that a higher risk aversion

of traders can cause asset prices to be more volatile. Recent financial crises have

motivated financial economists to pay attention to the role of ambiguity aversion in

explaining the behavior of investors in asset markets. Indeed, recent studies have

begun to attribute high price volatility to ambiguous information in asset markets.

Assuming a representative investor, Epstein and Schneider (2008) and Illeditsch

(2011) show that ambiguous information contained in signals can cause extremely

high price volatility. Dow and Werlang (1992b) and Ozsoylev and Werner (2011)

demonstrate that ambiguity can increase price volatility such that excess volatility

occurs in the sense that price volatility exceeds payoff volatility.

The purpose of this paper is to investigate the interactive effects of risk aversion

and ambiguity on price volatility under asymmetric information. To do this, we

adopt the model of Grossman and Stiglitz (1980) and assume that uninformed tra-

ders have ambiguous information. This setup is similar to Mele and Sangiorgi

(2015), but they attempt to characterize equilibrium asset prices under endogenous

information acquisition. However, we focus on price volatility without information

acquisition. Uninformed traders are assumed to have multiple beliefs about the

mean of a risky asset’s true value but exact information about its variance. This

assumption simplifies the analysis of asset market equilibrium.

1

To analyze the behavior of traders facing ambiguity, which is evidenced by the

Ellsberg (1961) paradox, we can consider models of ambiguity-averse preferences

such as the maxmin expected utility (MEU) model (Gilboa and Schmeidler, 1989),

the smooth ambiguity model (Klibanoff et al., 2005), the multiplier utility model

(Hansen and Sargent, 2001), and the variational utility model (Maccheroni et al.,

2006). Although ambiguity (belief) and the ambiguity aversion (taste) are not sepa-

rated in the MEU model, we adopt this model to reflect the extremely conservative

behavior of traders facing ambiguity in that they consider the worst-case probability

distribution. The experimental study of Ahn et al. (2014) demonstrates that the

MEU model can explain the evidence of ambiguity aversion better than the smooth

ambiguity model.

Finance literature adopting the MEU model has shown that portfolio inertia

arises due to ambiguity.

2

Traders with ambiguity choose not to participate in trad-

1

In general, uninformed traders may have exact information about neither mean nor vari-

ance, as assumed in Easley and O’Hara (2009, 2010) and Ozsoylev and Werner (2011).

2

See Dow and Werlang (1992a), Cao et al. (2005), and Ozsoylev and Werner (2011) among

others.

Ambiguity, Risk Aversion, and Price Volatility

©2015 Korean Securities Association 617