Private communication and management forecasts: Evidence from corporate site visits

• Published date: 01 July 2022
• Author: Xiaoqi Chen,C. S. Agnes Cheng,Jing Xie,Haoyi Yang
• Date: 01 July 2022
• DOI: http://doi.org/10.1111/corg.12419

ORIGINAL ARTICLE

Private communication and management forecasts: Evidence

from corporate site visits

Xiaoqi Chen

1

| C. S. Agnes Cheng

2

| Jing Xie

3

| Haoyi Yang

4

1

Institute for Financial and Accounting Studies,

Xiamen University, Xiamen, China

2

Steed School of Accounting, University of

Oklahoma, Norman, Oklahoma, USA

3

School of Accounting and Finance, Hong

Kong Polytechnic University, Hong Kong, SAR,

China

4

School of Business, Nanjing University,

Nanjing, China

Correspondence

Haoyi Yang, Department of Accounting,

Nanjing University, 16 Jinyin St, Nanjing

210000, China.

Email: hyang@nju.edu.cn

Abstract

Research Question/Issue: This paper investigates whether the communication

between managers and corporate site visitors facilitates managerial learning in

improving management forecast accuracy.

Research Findings/Insights: Using corporate site visit (CSV) data from 2009 to 2016,

we find that the frequency of corporate site visits is positively related to manage-

ment forecast accuracy. The positive relationship between corporate site visits and

forecast accuracy is stronger when visitors have greater expertise and advanced

industry knowledge and when firms are visited during a period of high business

uncertainty and are subject to high sensitivity to industry shocks. These findings are

consistent with the managerial learning hypothesis. Our results are robust to endo-

geneity concerns and alternative CSV measures.

Theoretical/Academic Implications: Previous literature on voluntary disclosure

assumes that managers are endowed with a private set of information and decide

how much of that to reveal. In this paper, we find that corporate site visits serve as a

communication channel through which managers learn from investors. Overall, we

provide evidence that private communication is a two-way channel through which

visitors learn useful information from managers and vice versa.

Practitioner/Policy Implications: We provide evidence that the private information

acquisition process is also a channel of information sharing that increases manage-

ment forecast accuracy. Our results could be useful to policymakers who may care

about the information acquisition cost of investors when evaluating the benefit of

corporate site visits.

KEYWORDS

corporate site visits, management forecasts, managerial learning, forecast accuracy

1|INTRODUCTION

In this paper, we examine whether managers can learn useful informa-

tion from corporate site visitors to improve their management fore-

cast accuracy. Management forecasts are an important way through

which managers disclose information to investors and reduce informa-

tion asymmetry between firms and investors (Beyer et al., 2010;

Cheng et al., 2013). Previous literature on management forecasts is

based on the voluntary disclosure theory, which assumes that

managers are endowed with a private set of information, and that

managers decide how much information to reveal (Grossman &

Hart, 1980). In this paper, we investigate an interesting and important

question: Do corporate site visitors affect management forecast

accuracy?

1

Received: 2 November 2020 Revised: 16 November 2021 Accepted: 17 November 2021

DOI: 10.1111/corg.12419

482 © 2021 John Wiley & Sons Ltd Corp Govern Int Rev. 2022;30:482–497.wileyonlinelibrary.com/journal/corg

Corporate site visits (hereinafter CSVs) become an increasingly

popular way for external shareholders to communicate with managers

(Brown et al., 2015). Unlike the traditional communication channels

such as conference calls, non-deal roadshows, or investors' days, CSVs

allow visitors to observe firms' operating activities and facilities and

communicate with general managers, functional managers, and front-

line managers. During face-to-face meetings, visitors may meet with

CEOs, CFOs, other top executives, board members, and/or other

managers (Cheng et al., 2016,2018). Thus, CSVs provide a two-way

channel between site visitors and managers to closely communicate.

During the communication, site visitors who hold large shares in the

firm have incentives and are willing to inform managers about the

information they have. Also, if visitors, especially institutional inves-

tors and analysts, visit target firms' peers or competitors, they may

have a clearer picture about the industry, and some non-sensitive

information they choose to share can be useful for managers making

forecasts in the future. Therefore, we expect that management

forecast accuracy will improve if managers can get new and useful

information in addition to their private information set from corporate

site visitors.

Albeit its importance, the evidence related to how CSVs affect

visiting investors and the visited firm is scarce because, as pointed out

by Cheng et al. (2016), firms in the Unites States and Europe either do

not maintain archival records of site visits or prohibit distribution of

such information. Recently, Chinese-listed firms in the Shenzhen

Stock Exchange

2

(hereinafter SZSE) are mandated to disclosure site

visit information. Using CSVs data in China, current literature finds

that analysts improve their forecast accuracy after CSVs (Cheng

et al., 2016). CSVs improve corporation innovation (Jiang &

Yuan, 2018), discipline managers and curb excess corporate invest-

ments (Cao et al., 2017), and exacerbate managers' incentive to

withhold bad news, hence leading to crash risk (Gao et al., 2017).

Since 2009, listed firms on SZSE have been required to disclose

detailed information about each corporate site visit initiated by inves-

tors in their annual reports, including the time, location, communica-

tion method and key communication content, and visitors' names and

institutional backgrounds. We obtain corporate site visit data from

Chinese Research Data Services (CNRDS). The sample period is from

2009 to 2016. Following Cao et al. (2017) and Cheng et al. (2016,

2018), we measure the frequency of CSVs by summing the total num-

ber of site visits for each firm in the second half of the year.

3

Our baseline results show that the frequency of corporate site

visits significantly increases management forecast accuracy. Economi-

cally, an increase of one standard deviation in CSVs increases 17%

the standard deviation of management forecast accuracy. Our

baseline result is robust to various robustness checks, including alter-

native measures of forecast accuracy and CSVs, propensity score

matching (PSM), and alternative time windows to count the frequency

of CSVs.

Our results are not immune to endogeneity problems. One con-

cern is that the potential endogeneity stemming from unobserved

omitted variables may bias our results. We apply several approaches

to address this issue. First, we add firm and year fixed effects in all

regressions to control for firm and year invariant factors. Second, we

employ the Heckman Selection approach. Following Jiang and

Yuan (2018), we use the number of listed firms in the city in which

the firm's headquarter is located as the choice of exclusion restriction.

We find that our results still hold. Finally, we implement the

difference-in-differences (DiD) approach by using first-time high-

speed train connection as an exogenous shock, and we find that after

the connection of high-speed train, treatment firms significantly

increase CSV activities and also experience a significant increase in

management forecast accuracy.

We further explore the managerial learning channel by investigat-

ing how the relation between CSVs and forecast accuracy varies in

circumstances where outside information is more valuable to man-

agers. We find that most of the explanatory power of CSVs can be

attributed to professional investors and analysts who come from

insurance companies, mutual funds, foreign institutions, private funds,

securities agencies, and asset management companies. And we also

find that the effect of CSVs on forecast accuracy is more pronounced

when visitors have advanced industry-level knowledge, and when the

firms are visited during a period of high business uncertainty and are

subject to high sensitivity to industry shocks. All these results are

consistent with the managerial learning hypothesis.

The contribution of our paper is threefold. First, we contribute to

the literature on management forecasts by documenting a positive

effect of private interaction between insiders and outsiders on fore-

cast accuracy. Prior literature has documented CEO ability (Baik

et al., 2011), auditor quality (Clarkson, 2000), internal and external

governance (Ajinkya et al., 2005; Karamanou & Vafeas, 2005), analyst

coverage (Chapman & Green, 2018), and firm size (Baginski &

Hassel, 1997) as determinants of forecast accuracy. Our study differs

from the previous literature as we take a further step to show the

channels through which management forecast accuracy is affected.

We extend the management forecast literature by showing that

managers' disclosure is affected by insiders as well as outsiders, more

specifically, corporate site visitors.

Second, our study adds to the literature on information acquisi-

tion during private interactions between managers and investors. Prior

literature has documented that outsiders, such as analysts, have

information advantages after private interactions with managers via

conference calls, private phone calls, non-deal roadshows, and face-

to-face meetings (Bushee et al., 2011,2017; Green et al., 2014;

Solomon & Soltes, 2015). Managers also benefit from private interac-

tions by utilizing the information they have acquired from investors

(Brockman et al., 2017; Bowen et al., 2018; Chapman & Green, 2017).

Consistent with the previous literature showing that investors have

private information (Zuo, 2016), our results add to this strand of litera-

ture by documenting that managers can acquire useful information

from site visitors to improve forecast accuracy.

We also extend the current literature on corporate site visits.

CSVs have been shown to affect analysts' forecast accuracy (Cheng

et al., 2016), stock prices around the visit date (Cheng et al., 2018),

the trading behavior of mutual fund managers (Liu et al., 2017), corpo-

rate investment decisions (Cao et al., 2017), and corporate innovation

CHEN ET AL.483