Ownership structure and analysts’ forecast properties: a study of Chinese listed firms

• Pages: 54-78
• Date: 01 February 2016
• Published date: 01 February 2016
• DOI: https://doi.org/10.1108/CG-02-2015-0018
• Author: Sun Liu
• Subject Matter: Strategy,Corporate governance

Ownership structure and analysts’

forecast properties: a study of Chinese

listed firms

Sun Liu

Sun Liu is based at

International Business

School, Xi’an Jiaotong-

Liverpool University,

Suzhou, China.

Abstract

Purpose –The purpose of this paper is to investigate the association between ownership structure and

the properties of analysts’ forecasts in China’s unique corporate setting.

Design/methodology/approach –Multiple regression models were used to examine the influence of

ownership structure mechanisms on analysts’ forecast properties for listed Chinese firms during the

period 2008-2012.

Findings –The paper finds that analysts’ forecast accuracy is higher for listed firms with high levels of

foreign ownership and managerial ownership. However, the complex pyramidal ownership structure

could make corporate information less transparent and then increase the complexity of forecasting;

hence, it results in less precise analysts’ forecasts. Interestingly, the relationship between state

ownership and analysts’ forecast properties appears to be non-linear (an inverted U-shape), and the

inflection point at which the relationship becomes negative occurs at state ownership over 45 per cent.

Originality/value –To the best of the author’s knowledge, this paper is the first to investigate the

influence of ownership structure mechanisms on the properties of analysts’ forecasts in an emerging

market, and the findings provide some insight on how the properties of analysts’ forecast might be

shaped by certain ownership and control features in the context of concentrated state ownership and

complex pyramidal ownership structure.

Keywords Corporate governance, Corporate ownership, Information

Paper type Research paper

1. Introduction

The question of the determinants of analyst forecast properties, such as forecast accuracy

and dispersion, has received a great deal of attention in the accounting and finance

literature. Most prior studies have focused on the factors relating to analysts’ compensation

and expertise/reputation, analysts’ commission income, analysts following, forecast

horizon, earnings volatility, firm size and growth opportunity (Francis and Philbrick, 1993;

Lang and Lundholm, 1996;Brown, 2001;Duru and Reeb, 2002;Gu and Wu, 2003;

Karamanou and Vafeas, 2005;Cowen et al., 2006;Ali et al., 2007;Jacob et al., 2008;

Barniv, 2009,Firth et al., 2013;Xu et al., 2013). However, the association between

ownership structure mechanisms and analysts’ forecast properties, especially in emerging

markets, remains understudied.

China is an interesting case because of its unique institutional environment. For example,

Chinese listed firms typically have a more complex ownership structure than the firms in the

West, and a considerable amount of listed firms’ shares is held by the government. As the

controlling shareholder, the government has appointed the majority of senior managers of

state-owned enterprises (SOEs). Also, unlike Western enterprises widely using equity

incentives to align the interest of managers and shareholders, listed Chinese firms design

the managerial remuneration contracts mainly on the basis of accounting performance

measures, such as sales and/or profits, rather than stock performance (Groves et al., 1994).

Consequently, managers have neither looked to the stock price as an indicator of the firm’s

Received 6 February 2015

Revised 25 September 2015

9 November 2015

Accepted 10 November 2015

PAGE 54 CORPORATE GOVERNANCE VOL. 16 NO. 1 2016, pp. 54-78, © Emerald Group Publishing Limited, ISSN 1472-0701 DOI 10.1108/CG-02-2015-0018

performance nor shown any interests in establishing effective communication channels

with investors. Worse, if the firm’s performance is poor, then managers are tempted to

suppress unfavourable information and to cook the accounting books to secure their

emoluments, but at the expense of shareholders’ interests (Liu and Lu, 2007;Yang et al.,

2012). The China Securities Regulatory Commission (CSRC), therefore, issued the Trail

Measures for the Administration of Equity Incentive Plan of Listed Companies in December

2005 with the aim of increasing the level of managerial ownership in listed firms and then

bringing the interests of managers and shareholders into line. However, reforms in the

economical and political systems, which could lead to decentralised decision-making

authorities and to reduced concentrated state ownership, have largely lagged behind,

because of the government concerns about losing control over large listed firms and the

safety of state assets.

There is much evidence that corporate disclosure is strongly influenced by institutional

environment, including corporate ownership and control characteristics (Ball, 2001;Ball

et al., 2003;Eng and Mak, 2003;Leuz et al., 2003;Dobler, 2008;Cormier et al., 2010;

Zechman, 2010;Heitzman et al., 2010;Hermalin and Weisbach, 2012). Also, corporate

disclosure quality could be reflected in the properties of analysts’ forecasts, because

corporate disclosure is one of the primary information sources for financial analysts who

play an important role in facilitating information transparency in capital markets (Lang and

Lundholm, 1996;Brown et al., 2011). For example, financial analysts act as information

intermediaries to generate valuable information, such as earnings forecasts and stock

recommendations, which are vital for investment decision-making. In addition, analysts can

engage in private information production to uncover any manager’s misuse of firm

resources (Healy and Palepu, 2001). Early developed market studies demonstrate that

analysts’ forecast error and bias are associated with corporate governance features, such

as ownership and control, because corporate governance quality can shape the

informativeness of a firm’s corporate disclosure, and corporate transparency can reduce

uncertainty surrounding the firm’s future performance (Bhat et al., 2006;Byard et al., 2006;

Ali et al., 2007;Haw et al., 2010;Brown et al., 2011;García-Meca and Sánchez-Ballesta,

2011). Given China’s unique corporate setting, this paper extends the above stream of

research by investigating the association between ownership structure mechanisms and

analysts’ forecast properties for Chinese listed firms.

Using a sample of 1,384 firm-year observations during the period 2008-2012, this study

finds that analysts’ forecast accuracy is positively related to the level of foreign ownership,

indicating that the demand for greater corporate information transparency by foreign

investors can reduce forecasting complexity and, hence, result in more precise analysts’

forecasts (Barniv, 2009). Also, there is a positive association between managerial

ownership and analysts’ forecast accuracy; this finding is in line with the incentive

alignment hypothesis (Jensen and Meckling, 1976;Nasir and Abdullah, 2004;Bhat et al.,

2006) which states that managerial ownership can be an effective means to align the

interests of managers and shareholders and, then, mitigate information asymmetry.

However, analysts’ forecast accuracy is lower for firms with more complex pyramidal

ownership structure, supporting the view that with the increase of pyramidal layers,

managers are more likely to act in self-interest and less likely to disclose corporate

information in a timely and objective manner (Aghion and Tirole, 1997;Baker et al., 1999,

Fan and Wong, 2002). Interestingly, this study finds that the association between state

ownership and forecast accuracy is likely to be non-linear (an inverted U-shape). Initially,

the relationship is positive when the level of state ownership is lower than 45 per cent (the

inflection point). However, this relationship becomes negative when the state owns more

than 45 per cent of the listed firms’ equity. Furthermore, it appears that institutional

ownership does not have any significant influence on analysts’ forecast accuracy, and

analysts’ forecast dispersion is not associated with most ownership structure variables,

including pyramidal structure, foreign ownership, managerial ownership and institutional

ownership, at any significance levels.

VOL. 16 NO. 1 2016 CORPORATE GOVERNANCE PAGE 55