CEO's Total Wealth Characteristics and Implications on Firm Risk

• Date: 01 March 2018
• Published date: 01 March 2018
• Author: Timo Korkeamäki,Daniel Pasternack,Eva Liljeblom
• DOI: http://doi.org/10.1111/irfi.12139

CEO’s Total Wealth Characteristics

and Implications on Firm Risk*

TIMO KORKEAMÄKI

†

,EVA LILJEBLOM

†,‡

AND DANIEL PASTERNACK

†,§

†

Hanken School of Economics, Helsinki, Finland,

‡

Lund University School of Economics and Management, Lund, Sweden and

§

Elite Asset Management Plc., Helsinki, Finland

ABSTRACT

We study the connections between firm risk and the CEO’s personal wealth

characteristics, using a unique dataset on CEO wealth and its components.

Consistent with decreasing absolute risk aversion, we find that wealthier CEOs

are associated with higher risk firms. Riskier firms tend to have CEOs whose

wealth is more independent of the firm. We also find that CEOs with high

personal portfolio betas run firms with higher betas. CEO’s tenure is negatively

associated with firm risk measured either as beta, idiosynchratic risk, or

volatility of accounting profitability. A possible interpretation is that risk-

averse managers are better able to imprint their risk preferences on the firm

over time. Stronger corporate governance weakens the connection between

CEO wealth characteristics and firm risk.

JEL Codes: G31; G32; J33

Accepted: 15 May 2017

I. INTRODUCTION

Risks that a manager takes as part of her job are likely to be influenced not only

by personal characteristics such as age but also by the size and the decomposition

of the manager’s wealth. Personal overall attitudes toward risk tend to also affect

managers’decision making (Cain and McKeon 2016). Empirical studies on the

connection between managerial wealth and firm risk-taking are typically based

on managerial ownership of the firm, either through shares or executive options

(see e.g. Guay 1999; Rajgopal and Shevlin 2002; Coles et al. 2006; Low 2009).

1

However, endogeneity issues cloud the interpretation of the relation between

equity-based managerial wealth and firm risk because it can be assumed that

* We thank Lee Biggerstaff, Mara Faccio, Martin Holmén, Karin Thorburn, an anonymous referee, and

the seminar participants at Aalto University Business School, Hanken School of Economics, University

of Edinburgh and the 2014 FMA Annual meetings in Nashville, TN for their helpful comments.

1 Share ownership is typically perceived as providing increased incentives for risk taking. The

relationship between options and risk seeking is more complex, see, for example, Guay

(1999), Carpenter 2000, and Ross (2004).

© 2017 International Review of Finance Ltd. 2017

International Review of Finance, 2017

DOI: 10.1111/irfi.12139

International Review of Finance, 18:1, 2018: pp. 35–58

DOI:10.1111/irfi .12139

© 2017 International Review of Finance Ltd. 2017

firms often design executive compensation schemes to motivate managers to

optimal risk taking. Recently, Wei and Yermack (2011) study CEOs’inside debt

positions. Their results indicate that publication of large CEO inside debt

holdings leads to reductions in firm risk, to wealth transfers from equity to debt,

and to reductions in enterprise value. Moreover, Cronqvist et al. (2012) study

CEOs’personal leverage by using data on their most recent home purchases

and find support for endogenous matching of between CEO and firm leverage.

While all these recent studies shed light on the connection between managerial

wealth and corporate risk-taking, they focus only on parts of CEO wealth, as data

restrictions have not allowed studying the effects of CEOs’total wealth. The rare

exceptions to consider CEOs’total wealth include Becker (2006), who studies the

effects of total wealth on performance-based compensation among Swedish

CEOs, and Elsilä et al. (2013), who observe the connection between total wealth

of Swedish CEOs and their firms’subsequent accounting performance.

We contribute to the literature by studying the connections between CEOs’

total wealth and corporate behavior. By using a unique dataset on total CEO

wealth, we at least partly circumvent the endogeneity problems that are inherent

with equity compensation packages designed by the firm. Our detailed dataset

also allows us to study the effects of decomposition of CEOs’assets in great detail.

Our dataset includes the CEOs of all Finnish publicly listed firms, and in addition

to holdings in the CEOs’own firms, it also contains parts of the CEOs’wealth

(and indebtedness) that are not directly attributable to the firm and its

compensation structure. The data allow us to see all individual holdings such

as the CEOs’executive stock options and private investments in different

individual shares of listed and unlisted firms (item by item), mutual funds, real

estate and housing assets, land, and forest.

We find that wealthier CEOs are associated with riskier firms, a result in line

with decreasing absolute risk aversion. We also find support for the hypothesis

that managers whose wealth is more independent of the firm are associated with

riskier firms, a result in line with decreasing relative risk aversion. We further find

that the CEO’s tenure is negatively associated with firm risk measured either as

beta, idiosynchratic risk, or volatility of accounting profitability. A possible

interpretation is that risk-averse managers are better able to imprint their risk

preferences on the firm over time. Our findings are also consistent with Berger

et al.’s (1997) suggestion that managers with longer tenure become more

entrenched, which increases their risk aversion. Prior studies suggest that risk

aversion increases with age,

2

but our results are robust to controlling for age,

and they suggest tenure with the firm, instead of age, as the primary driver of

managers’risk attitude. We also consider the matching relationship between firm

risk and manager’s portfolio risk and find that firms with higher betas have CEOs

with high beta personal stock portfolios.

The rest of the paper is organized as follows. In section 2, we provide a

literature review and discuss our hypotheses regarding the relation between

2 See, for example, Halek and Eisenhauer (2001) and Serfling (2014).

International Review of Finance

© 2017 International Review of Finance Ltd. 20172

International Review of Finance

36 © 2017 International Review of Finance Ltd. 2017