Corporate Scandals and Household Stock Market Participation

• Author: MARIASSUNTA GIANNETTI,TRACY YUE WANG
• DOI: http://doi.org/10.1111/jofi.12399
• Published date: 01 December 2016
• Date: 01 December 2016

THE JOURNAL OF FINANCE •VOL. LXXI, NO. 6 •DECEMBER 2016

Corporate Scandals and Household Stock Market

Participation

MARIASSUNTA GIANNETTI and TRACY YUE WANG∗

ABSTRACT

We show that, after the revelation of corporate fraud in a state, household stock mar-

ket participation in that state decreases. Households decrease holdings in fraudulent

as well as nonfraudulent firms, even if they do not hold stocks in fraudulent firms.

Within a state, households with more lifetime experience of corporate fraud hold less

equity. Following the exogenous increase in fraud revelation due to Arthur Ander-

sen’s demise, states with more Arthur Andersen clients experience a larger decrease

in stock market participation. We provide evidence that the documented effect is

likely to reflect a loss of trust in the stock market.

THE ABILITY TO REAP EQUITY RETURNS IS OF CENTRAL importance for household

welfare and hence is the subject of a growing literature in finance and economics

(Campbell (2006)). A number of influential papers rely on fixed participation

costs and nonstandard preferences and expectations to explain why households

do not participate in the stock market to the extent that standard portfolio

models would predict (e.g., Mankiw and Zeldes (1991), Poterba and Samwick

(1995), Vissing-Jørgensen (2002)).

Somewhat surprisingly, however, this literature does not examine the extent

to which households’ limited stock market participation is affected by securities

∗Giannetti is from the Department of Finance of the Stockholm School of Economics. Wang

is from the Department of Finance of the Carlson School of Management University of Min-

nesota. We are grateful to two anonymous referees, an Associate Editor, Kenneth Singleton (the

Editor), Anup Agrawal, Allen Ferrell, Diego Garcia, Will Gerken, Cam Harvey,Gur Huberman, Zo-

ran Ivkovic, Jonathan Karpoff, George Korniotis, VojislavMaksimovic, Adair Morse, Stefan Nagel,

Chester Spatt, Michael Weisbach, and seminar participants at Western Finance Association, the

Miami Behavioral Finance Conference, the Fraud and Misconduct Conference at the University of

California, Berkeley, the Law and Economics NBER Meeting, the Harvard Business School Con-

ference on Household Behavior in Risky Asset Markets, the Society for Financial Studies Finance

Cavalcade, the Financial Intermediation Research Society Meetings, the European Finance As-

sociation, the Minnesota Corporate Finance Conference, Boston College, the U.S. Securities and

Exchange Commission, the University of Minnesota, the University of Lausanne, the University of

Zurich, the National University of Singapore, the University of New South Wales, De Paul Univer-

sity,Koc University, and the Southwestern University of Finance and Economics for comments. We

also thank Diego Garcia and Angie Lo for sharing some of the data we use in the paper. Giannetti

acknowledges financial support from the NASDAQ/OMX Nordic Foundation, the Jan Wallander

and Tom Hedelius Foundation, and the Bank of SwedenTercentenary Foundation. The authors do

not have any conflicts of interest, as identified in the Disclosure Policy.

DOI: 10.1111/jofi.12399

2591

2592 The Journal of Finance R

market regulation and corporate governance failures. Episodes of corporate fi-

nancial misconduct, for instance, have been known to wipe out between 20%

and 40% of the valuation of firms that are found to be fraudulent (Karpoff,

Lee, and Martin (2008a), Dyck, Morse, and Zingales (2013)). But these direct

economic costs may be just a fraction of the negative economic consequences of

corporate securities fraud. By undermining trust in financial markets, corpo-

rate financial misconduct may decrease stock market participation and in turn

increase the cost of capital for all firms.

This paper takes up the challenge of evaluating whether corporate scandals

decrease households’ willingness to participate in the stock market directly or

indirectly and whether through this channel they generate a negative external-

ity in financial markets. To generate cross-sectional and time-series variation

in households’ exposure to corporate financial misconduct, we assume that

households face greater exposure to frauds committed by firms headquartered

in the state in which they live. This assumption is realistic not only because

households are more likely to be aware of local firms, but also because lo-

cal news coverage and personal interactions increase their exposure to frauds

committed by these firms.1

We ask whether corporate scandals in a state reduce the stock market partic-

ipation of households in that state, controlling for nationwide macroeconomic

conditions and asynchronous local shocks using a host of household- and state-

level controls. We find unambiguous evidence that household stock market

participation decreases following corporate scandals in the state in which the

household resides. Households decrease their stock holdings in fraudulent as

well as nonfraudulent firms, and even households that do not hold the stocks of

fraudulent firms decrease their equity holdings. Thus, the decrease in house-

hold stock market participation is not driven by financial losses associated with

holdings in fraudulent stocks.

One may wonder whether our findings are driven by state-level economic

conditions that are associated with both the revelation of corporate fraud and

household stock market participation. For instance, the revelation of corporate

fraud generally occurs at the beginning of economic downturns that may inde-

pendently drive households’ decisions to reduce their equity holdings (Povel,

Singh, and Winton (2007), Wang, Winton, and Yu (2010)). To rule out these al-

ternative interpretations, we perform two sets of tests that exploit orthogonal

sources of variation in fraud revelation.

The first set of tests uses an arguably exogenous shock to fraud revelation

due to the sudden demise of auditing firm Arthur Andersen in 2002. All of

Arthur Andersen’s clients were forced to change auditors. Since new auditors

have incentives to “clean house,” firms that switched auditor due to Arthur

Andersen’s demise had a higher probability of being identified as having com-

mitted fraud (Dyck, Morse, and Zingales (2013)). This led to an exogenous

short-term increase in the probability of fraud revelation that differs across

1See Grinblatt and Keloharju (2001), Ivkovi´

c and Weisbenner (2005), and Seasholes and Zhu

(2010).

Corporate Scandals and Household Stock Market Participation 2593

states, depending on the fraction of firms in a state that were clients of Arthur

Andersen before its demise. We thus use the fraction of firms in a state that

were clients of Arthur Andersen right before its demise as an instrument for

fraud revelation in that state in the period following the shock. We find that

a one-standard-deviation increase in fraud revelation due to the presence of

Arthur Andersen clients increases the probability that a household exits the

stock market by seven percentage points. However, we do not find a consistently

significant drop in household equity-wealth ratio across specifications, possibly

suggesting that, while the Arthur Andersen shock caused some households to

exit the stock market, other households were unaffected.

The second set of tests exploits within-state variation in households’ lifetime

experience of corporate scandals. Households living in the same state at a par-

ticular point in time can have different corporate fraud experiences depending

on their age and whether they have moved across states. In these specifica-

tions, we absorb any state-level shocks by including interactions of state and

year fixed effects. We find that a one-standard-deviation increase in a house-

hold’s lifetime exposure to local fraud decreases the household’s probability of

holding stocks by 4% and its equity-wealth ratio by almost 10%, compared to

the sample average.

The second set of tests exploiting within-state heterogeneity also implies

that our findings are not driven by fraud revelation causing a deterioration

in state economic conditions, as changes in state conditions should affect all

individuals in the state, independent of their past experiences. Consistent with

this finding, households appear to also sell the stocks of out-of-state firms and

firms in unrelated industries, indicating that local fraud is unlikely to matter

because it leads to rational updating of the probability that other (similar) firms

may have committed fraud. Furthermore, the effect of fraud on stock market

participation appears to be unrelated to risk aversion, as it does not vary with

households’ risk tolerance.

We provide evidence consistent with the view that fraud revelation leads

households to reduce their stock market participation by undermining their

trust in the stock market. In states with high fraud revelation, there is a

decrease in the proportion of individuals that report high confidence in financial

markets, as captured by changes in confidence in big businesses and banks.

Moreover, the stock market participation of high-status individuals, who in

experiments have been found to trust less because they have higher costs

of betrayal (Bohnet and Zechauser (2004) and Bohnet et al. (2008)), is more

negatively affected by fraud revelation.

Besides contributing to the literatures on household stock market participa-

tion and the consequences of financial fraud, this paper adds to the literature

exploring whether trust has an effect on economic transactions. In an influ-

ential paper, Guiso, Sapienza, and Zingales (2008) show that general trust in

others helps explain the decision to participate in the stock market in a cross-

section of households. However, in his presidential address to the European

Economic Association, Fehr (2009) argues that, because general trust may just

be an effect of omitted environmental factors, the most convincing strategy to