The Shareholder Base Hypothesis of Stock Return Volatility: Empirical Evidence

• Published date: 01 March 2018
• Date: 01 March 2018
• DOI: http://doi.org/10.1111/fima.12184

The Shareholder Base Hypothesis

of Stock Return Volatility: Empirical

Evidence

H˚

akan Jankensg˚

ard and Anders Vilhelmsson∗

Weuse Swedish ownership data to explore whether a large and diversified shareholder base leads

to lower volatility by improving the information content of stock prices. We find that volatility

increases in the number of shareholders with respect to both the number of relatively large

shareholders and the fraction of shares held by investors with stakes below 0.1%. Volatility is

also positively related to the number of institutional owners but negativelyrelated to the number

of large and underdiversified institutional owners. Foreign investorshave no impact. Our results

suggest that a large shareholder base does not lowervolatility.

There is great variation in the structure of corporate ownership. Whereas some firms have a

small number of domestic owners, others have a more diversified structure with institutional in-

vestors, privateequity, business spheres, and foreign owners as a significant part of the ownership.

Some firms have a large fraction of small investors, whereas others havea small number of large

ownership blocks. The questions we address in this article are 1) whether stock return volatility

is determined by what kind of investors a firm has and 2) how many investors there are, which

is hereafter referred to as the shareholder base. The literature has noted the importance of stock

return volatility for financial theory, as well as for practitioners in the investor community (e.g.,

Campbell et al., 2001; Zhang, 2010). We contribute to this literature by examining shareholder

base determinants of volatility.

Our inquiry is motivated by an intriguing conjecture in the literature that suggests volatility

decreases with a larger shareholder base. According to Wang (2007), the shareholder base-

broadening effect occurs because each individual has only partial information about the firm. As

the number of investorsg rows,the accuracy of the information available about the stock increases,

which in turn lowers the variance of stock returns. This follows from an extension of Merton’s

(1987) analysis of investor recognition, according to which a small shareholder base leads to

higher expected returns because risks are insufficiently shared among investors. Although the

importance of the shareholder base is hinted at in several studies, it has not, to our knowledge,

The authors thank an anonymous referee, Marc Lipson (Editor), Adri de Ridder, Milda Norkute, Joakim Westerlund,

Jens Forssbaeck, Martin Holm´

en, TomAabo, Martin Strieborny, Ettore Croci, Bj¨

orn Hagstr¨

omer,and Frederik Lundtofte

for useful suggestions. We also thank seminar participants at the Knut Wicksell Centre for Financial Studies, Lund

University; School of Business Economics and Law, Gothenburg University; and School of Business and Social Sci-

ences, Aarhus University. We are grateful to Sven-Ivan Sundqvist at SIS ¨

Agarservice for supplying the ownership data.

Jankensg˚

ard gratefully acknowledgesthe f inancial support of the JanWallander and Tom Hedelius Foundation and the

Tore Browaldh Foundation. Vilhelmsson gratefully acknowledges the financial support of the Marianne and Marcus

WallenbergFoundation.

∗H˚

akan Jankensg˚

ard is an Associate Professorin the Department of Business Administration at Lund University in Lund,

Sweden. Anders Vilhelmsson is an Associate Professor in the Department of Economics at Lund University in Lund,

Sweden.

Financial Management •Spring 2018 •pages 55 – 79

56 Financial Management rSpring 2018

been comprehensively investigated empirically (Merton, 1987; West, 1988; Wang, 2007; Rubin

and Smith, 2009; Li et al., 2011). The current study fills this gap.

Stock return volatility may also decrease if the firm diversifies its shareholder base by attracting

specific groups of shareholders who are better informed compared with individual domestic

investors. The literature has identified two categories of shareholders that are likely to improve

the information content of a firm’s stock price: institutional and foreign investors. According to

Rubin and Smith (2009), institutional investors are more financially sophisticated and therefore

contribute to more efficient gathering and processing of information. Using a sample of US

firms, Rubin and Smith (2009) f ind that institutional ownershipleads to lower volatility for firms

that do not pay a dividend because the information environment is weaker for these firms. Li

et al. (2011) show that trading by foreign investors has a stabilizing influence on volatility in the

context of emerging markets following financial liberalization.

In this study,we investigate whether the effectsfrom increasing and diversifying the shareholder

base persist in an economy that steers a middle course between the Anglo-American governance

model and emerging markets where concentrated ownership is predominant. Sweden has long

been a developed, open economy with liquid financial markets that mixes 1) a traditional gover-

nance model involving strong controlling owners and various mechanisms for corporate control

(e.g., differential voting rights) with 2) a large influx of institutional and foreign owners, whose

influence has grown substantially in recent decades. This variation makes for a useful setting to

test the relation between the size and heterogeneity of the shareholder base, on the one hand, and

stock return volatility, on the other.

We furthermore benefit from access to databases on corporate ownership that offer several

advantages for identifying the shareholder base. SIS ¨

Agarservice collects detailed ownership data

on publicly listed firms in Sweden and provides the actual ownership lists of these firms. This

allows us to characterize firms’ shareholder base in terms of the number of investors, their type

and nationality, cash flow and voting rights, association with a business sphere, and so on for

a broad cross-section of firms, from the very smallest listed f irms to large-cap multinationals.

Furthermore, access to data from the Visby Research in Stock Ownership (VIRSO) files allows

us to identify the size of the shareholder base on various dimensions, such as the number of

individual, institutional, and foreign investors, as well as other categories of owners such as

trusts, operating companies, and churches. As explained in detail later in this article, several

features of these databases are unique and provide a better identification of the shareholder base

than looking at aggregate ownership stakes. We construct a sample of ownership data spanning

from 2000 to 2013, yielding around 1,900 firm-year observations. We use a rich set of control

variables from theories related to earnings uncertainty, information asymmetry, leverage, and

portfolio concentration.

The data do not bear out the hypothesis that a large shareholder base is conducive to lower

volatility. In fact, the evidence is overwhelmingly to the contrary. We first show that volatility

increases with the total number of shareholders. This effect persists when we break down the

shareholder base into groups based on the relative size of their ownership stakes, which captures

differing incentives to produce information about the stock. Volatility increases with the number

of relatively large shareholders (stakes >0.1%) as well as with the size of the “microfloat”

(i.e., the fraction of shares held by small investors with stakes below 0.1%). These results are

generally significant at the 1% level and are robust to various econometric approaches. We

implement an instrumental-variable approach (two-stage least squares [2SLS]) to deal with the

potential problem of small investors self-selecting into high-volatility stocks, which could also

explain the association. The relation remains statisticallysignif icant after taking self-selection into

account.