Managerial control divergence and analysts’ information precision

• Author: Yong‐Chul Shin,Surjit Tinaikar,KoEun Park
• Published date: 01 September 2017
• Date: 01 September 2017
• DOI: http://doi.org/10.1111/corg.12210

EMPIRICAL

Managerial control divergence and analysts’information

precision

KoEun Park |Surjit Tinaikar |Yong‐Chul Shin

University of Massachusetts Boston, Boston,

Massachusetts, USA

Correspondence

Surjit Tinaikar, Assistant Professor, 100

Morrissey Blvd., College of Management,

UMass‐Boston, Boston, MA 02125, USA.

Email: surjit.tinaikar@umb.edu

Abstract

Research Question/Issue: This study addresses how financial analysts’information envi-

ronment varies with the extent to which insiders’voting rights diverge from their cash flow rights

(i.e., control divergence) in the firm.

Research Findings/Insights: This study finds evidence that both analysts’public and private

information precisions are decreasing with insiders’control divergence for a sample of US dual

class share firms. Dual class share firms feature ownership structures that allow us to measure

the control divergence between insiders’voting rights and their cash flow rights. Furthermore,

we document that these effects are more pronounced for firms with lower firm‐level uncertainty

and for periods prior to the introduction of the SOX Act. Our results are robust to omitted vari-

ables, endogenous ownership, and any selection biases associated with dual class equity.

Theoretical/Academic Implications: Analysts’skill and ability to formulate accurate fore-

casts are influenced by the quality of their information environment. We provide insights on

how ownership mechanisms such as control divergence impact this ability by affecting their infor-

mation environment. By using empirical proxies that precisely measure analysts’information envi-

ronment, we are able to better capture the adverse impact of control divergence on the public

and private information that analysts convey in their forecasts.

Practitioner/Policy Implications: Policy makers are often concerned about issues that can

constrain information processing by intermediaries such as analysts. Control divergence mecha-

nisms can negatively impact analysts’information processing by affecting their information envi-

ronment and consequently their ability to effectively forecast earnings and stock prices. These

have larger implications for social welfare and optimal capital allocation in the economy. Policy

makers may therefore need to evolve better regulatory norms to negate some of the pernicious

consequences of such ownership structures on information intermediation.

KEYWORDS

Corporate Governance, control divergence, analysts’information environment, dual class firms,

SOX

1|INTRODUCTION

This study examines how analysts’information environment varies

when managers’control rights within a firm diverge from their cash

flow rights in dual class share firms (viz. control divergence).

1

By ana-

lysts’information environment, we refer to analysts’precision of public

and private information as measured by the empirical proxies designed

by Barron, Kim, Lim and Stevens (1998, hereafter BKLS) to capture the

public and private information that analysts convey in their forecasts.

Prior research suggests that control divergence enables insiders to

extract private rents through self‐dealing, excess compensation and

expropriation without proportionately suffering the consequences of

their actions (Bertrand & Mullainathan, 2003; Faccio & Lang, 2002;

Masulis, Wang, & Xie, 2009). This inclination to divert corporate

resources at the expense of minority shareholders can create incen-

tives to produce noisier and more opaque financial reports. We refer

to this as the “entrenchment effect”.Consistent with this view, empir-

ical findings in the accounting literature show that control divergence

Received: 9 November 2015 Revised: 4 January 2017 Accepted: 3 March 2017

DOI: 10.1111/corg.12210

294 © 2017 John Wiley & Sons Ltd Corp Govern Int Rev. 2017;25:294–311.wileyonlinelibrary.com/journal/corg

is associated with earnings opacity (Gopalan & Jayaraman, 2012; Haw,

Hu, Hwang, & Wu, 2004), low earnings informativeness (Fan & Wong,

2002; Francis, Schipper, & Vincent, 2005), low levels of compensation

disclosure (Tinaikar, 2014), and less timely loss reporting (Khurana,

Raman, & Wang, 2013; Lim & Tan, 2009). In this study, we widen the

scope of these control divergence studies from looking at firm‐level

information environment to investigating the information environment

of intermediaries such as analysts. Analysts’own information environ-

ment is a function of not just the disclosure environment of the firms

they follow, but of many other factors including analysts’own idiosyn-

cratic interpretations of the information provided by these firms, their

information acquisition and processing costs, and their own incentives

to generate private information from sources that go beyond public

reports.

2

The only study that attempts to examine the association of

analysts’information environment with control divergence finds no

measurable impact of control divergence on properties of analysts’

forecasts such as forecast accuracy and forecast dispersion for a sam-

ple of countries from East Asia and Western Europe (Haw, Ho, Hu, &

Wu, 2010). We revisit this research question by using a sample of US

dual class share firms. Our findings indicate that control divergence

in these firms negatively impacts analysts’precision of public and pri-

vate information. Our results are robust to corrections for selection

biases associated with dual class equity and endogenous ownership

structures.

We focus on financial analysts for two reasons. First, corporate

ownership characteristics such as managerial ownership do appear to

influence financial analysts’decision to follow firms and their informa-

tion environment (Baik, Kang, & Morton, 2010; Han, Jin, Kang, & Lobo,

2014; Lang, Lins, & Miller, 2004; Moyer, Chatfield, & Sisneros, 1989;

Sabherwal & Smith, 2008). While our study does not focus on manage-

rial ownership per se, we do believe that related ownership character-

istics such as control divergence constitute important governance

structures in terms of the entrenchment forces they can unleash

(Bebchuk, Kraakman, & Triantis, 2000; Claessens, Djankov, & Lang,

2000; Shleifer & Vishny, 1997). We therefore expect and find them

to have a significant negative impact on analysts’information

environment. Second, financial analysts play an important role in

information intermediation by evaluating information from public and

private sources, in generating earnings forecasts and other estimates,

and in facilitating market efficiency (Brown & Rozeff, 1978; Busse &

Green, 2002; Givoly & Lakonishok, 1979). Thus, any governance

characteristic that has the potential to impact analysts’information

intermediation would be of interest to both academic researchers

and policy makers alike.

Our study is probably most similar to Han et al. (2014) who, using

the BKLS measures, hypothesize that the relationship between mana-

gerial ownership and analysts’information environment could be char-

acterized by either entrenchment or incentive alignment.

3

They find

that managerial ownership is positively associated with the precision

of analysts’public and idiosyncratic information consistent with incen-

tive alignment. Our paper is different from Han et al. (2014)'s paper in

that we examine the impact of control divergence on analysts’infor-

mation environment. In doing so, we address an important lacuna in

the evidence presented in Haw et al. (2010) who find no measurable

impact of control divergence on analysts’forecast properties, and that

in Han et al. (2014) who find a positive incentive alignment effect for

managerial ownership but do not find any entrenchment effects. As

in Han et al. (2014), we employ the BKLS model to measure analysts’

information environment. The BKLS model synthesizes multiple

aspects of analysts’forecast properties (i.e., forecast accuracy, forecast

dispersion, and analyst coverage) to construct information environ-

ment measures.

Investigating control divergence necessitates the use of ownership

structures where both control rights and cash flow rights could be eas-

ily separated. Dual class share firms differ from single class share firms

in that they are characterized by a separation of control rights from

cash flow rights (Claessens et al., 2000; Cronqvist & Nilsson, 2003;

Lins, 2003). Furthermore, unlike in other regions such as East Asia,

Western Europe, and Canada, US dual class share firms are typically

not accompanied by other opaque control enhancing characteristics

such as stock pyramids, cross‐ownership, or interlocking ownership

(Bebchuk et al., 2000; La Porta, Lopez‐de‐Silanes, Shleifer, & Vishny,

1999). This makes the separation of control rights from cash flow

rights relatively transparent and thus makes it easier to isolate the

impact of control divergence on a specific firm‐level attribute. While

control divergence could also be achieved using pyramidal or horizon-

tal ownerships, such ownership mechanisms also have limitations. This

is because the primary motivations for choosing these structures may

have less to do with achieving separation of control rights from cash

flow rights and more to do with reducing taxes on inter‐company div-

idends (Morck, 2003), hiding the identity of the ultimate owner from

the market (Bianchi, Bianco, & Enriques, 2001), or offering certain

financing advantages in poorly developed capital markets (Almeida &

Wolfenzon, 2006). Empirical evidence also indicates that control diver-

gence achieved using pyramids and horizontal structures, being a by‐

product of such structures, tends to be minimal and smaller than that

in dual class share firms (Franks & Mayer, 2001; Lefort & Walker,

2000). Given these limitations of other control divergence structures,

US dual class shares seem to be a natural and feasible choice for exam-

ining our research questions.

In addition to our primary hypotheses, we also explore how exog-

enous variations in firm‐level uncertainty and the enactment of the

Sarbanes‐Oxley Act (SOX) impact this relationship. We focus on firm‐

level uncertainty because analysts have been known to have behav-

ioral biases and misunderstand the implications of financial information

during periods of high uncertainty resulting in less accurate forecasts

(Amiram, Landsman, Owens, & Stubben, 2014; Zhang, 2006). We con-

jecture that analysts may experience similar hurdles in comprehending

the implications of control divergence during uncertain times. Our

focus on SOX is also motivated from the findings in prior literature that

institutional environments that offer better shareholder protection and

mandate higher governance transparency improve analysts’forecast

accuracy (Bhat, Hope, & Kang, 2006; Hope, 2003). We believe that

both firm‐level uncertainty and SOX influence not only the trade‐off

of costs (Dye, 1998; Fishman & Hagerty, 2003) and benefits (Botosan,

1997; Diamond & Verrecchia, 1991) of public disclosure, but also the

private information acquisition and processing costs for the analysts

(Begley, Gao, & Cheng, 2009; Kim & Verrecchia, 1994). Our results

indicate that the relationship between analysts’information precision

and managerial control divergence tends to be weaker in environments

PARK ET AL.295