Short Selling Pressure, Stock Price Behavior, and Management Forecast Precision: Evidence from a Natural Experiment

• Author: LIANDONG ZHANG,YINGHUA LI
• Date: 01 March 2015
• Published date: 01 March 2015
• DOI: http://doi.org/10.1111/1475-679X.12068

DOI: 10.1111/1475-679X.12068

Journal of Accounting Research

Vol. 53 No. 1 March 2015

Printed in U.S.A.

Short Selling Pressure, Stock Price

Behavior, and Management Forecast

Precision: Evidence from a Natural

Experiment

YINGHUA LI

∗AND LIANDONG ZHANG

†

Received 26 September 2013; accepted 3 November 2014

ABSTRACT

Using a natural experiment (Regulation SHO), we show that short selling

pressure and consequent stock price behavior have a causal effect on man-

agers’ voluntary disclosure choices. Specifically, we find that managers re-

spond to a positive exogenous shock to short selling pressure and price

sensitivity to bad news by reducing the precision of bad news forecasts. This

finding on management forecasts appears to be generalizable to other corpo-

rate disclosures. In particular, we find that, in response to increased short sell-

ing pressure, managers also reduce the readability (or increase the fuzziness)

of bad news annual reports. Overall, our results suggest that maintaining the

current level of stock prices is an important consideration in managers’ strate-

gic disclosure decisions.

JEL codes: D82; G14; G18; M41

Keywords: Regulation SHO; short selling; corporate disclosure; forecast

precision; annual report readability; managerial incentives

∗Arizona State University; †City University of Hong Kong.

Accepted by Christian Leuz. We appreciate helpful comments from two anonymous

reviewers, Jong-Hag Choi, Kalin Kolev, Bin Miao, and Gord Richardson. We thank Ke

Wang for research assistance. An online appendix to this paper can be downloaded at

http://research.chicagobooth.edu/arc/journal/onlineappendices.aspx.

79

Copyright C, University of Chicago on behalf of the Accounting Research Center,2014

80 Y.LI AND L.ZHANG

1. Introduction

Corporate executives pay considerable attention to secondary market

prices and they have strong incentives to maintain or increase their firms’

stock prices. The accounting literature argues that managers can make

strategic financial reporting or disclosure choices to influence stock prices

(e.g., Healy and Palepu [2001]). A large body of empirical research ex-

amines whether and how corporate disclosures affect stock prices. The

literature, however, provides little directional evidence on whether the

behavior of stock prices has a causal effect on managerial strategic disclo-

sure decisions. The difficulty in establishing causality stems largely from the

endogenous nature of stock prices. In this paper, we use a natural experi-

ment to examine the causal effect of stock price behavior on managers’

voluntary disclosure choices. Specifically, we examine the effect of an ex-

ogenous shock to short selling pressure and consequent price sensitivity to

bad news on management forecast precision, where precision refers to the

specificity of forecasts.1

We focus on management forecasts and their precision for the follow-

ing reasons. First, management forecasts are an important source of corpo-

rate financial information for investors. For example, Beyer et al. [2010]

show that management forecasts provide, on average, approximately 55%

of accounting-based information to the stock market over the 1994 to 2007

period. Thus, it is important to understand factors that affect management

forecast choices. Second, there is a large degree of variation in forecast

precision, and managers have a great deal of control and discretion over

forecast precision (Hirst, Koonce, and Venkataraman [2008]). Once the

decision to issue a forecast is made, managers can issue either qualitative or

quantitative forecasts. Quantitative forecasts can be made as point, range,

minimum, or maximum estimates. For range forecasts, managers can fur-

ther choose their width. Due to litigation or reputation concerns, managers

may have even greater discretion over forecast precision than over the de-

cision of whether to issue a forecast (Cheng, Luo, and Yue [2013], Skin-

ner [1994]). Thus, forecast precision is an ideal setting in which to test

managers’ strategic disclosure choices. Finally, prior research shows that

forecast precision has a significant effect on the sensitivity of market prices

to forecast news (e.g., Baginski, Conrad, and Hassell [1993]). This effect

makes the choice of forecast precision a natural fit in our controlled exper-

iment, as discussed below.

Our experiment is based on Securities and Exchange Commission (SEC)

Regulation SHO (Reg SHO), adopted in 2005. On September 7, 2004, the

SEC passed Reg SHO, which mandated temporary suspension of short-sale

price tests for a set of randomly selected pilot stocks during the period May

2, 2005, to August 6, 2007. The pilot stocks comprise every third stock of the

1Our study focuses on one specific aspect of stock price behavior, price sensitivity to bad

news, as dictated by our experimental setting.

SHORT SELLING AND MANAGEMENT FORECAST PRECISION 81

Russell 3000 Index ranked by average daily trading volume. The suspension

of short-sale price tests (i.e., the uptick test for the NYSE and the bid test for

the NASDAQ) represents an exogenous decrease in short-sale constraints,

leading to an increase in short selling activities for the pilot stocks (e.g.,

SEC [2007], Diether, Lee, and Werner [2009]). Increased trading activities

of pessimistic investors make prices of the pilot stocks more sensitive to

negative news (e.g., Goldstein and Guembel [2008], Grullon, Michenaud,

and Weston [2012]).

This study predicts that managers of pilot firms, in response to the posi-

tive shock to price sensitivity to bad news, reduce the precision of bad news

forecasts to maintain the current level of stock prices.2This prediction is

based on the theory and evidence that the magnitude of price reaction to

a disclosed signal is positively related to its precision (e.g., Kim and Verrec-

chia [1991], Baginski, Conrad, and Hassell [1993], Baginski, Hassell and

Wieland [2007]). In fact, our conjecture requires only that managers be-

lieve that reducing the precision of bad news can minimize its impact on

stock prices. This assumption appears to be true given the survey evidence

that some CFOs admit that they tend to make fuzzy disclosures if the news

is bad (Graham, Harvey, and Rajgopal [2005]).

To conduct our empirical tests, we first obtain a list of the 2004 and 2005

Russell 3000 Index constituents from Russell Investments. The pilot stocks

chosen by the SEC are naturally our treatment stocks. We then assign the

remaining stocks of the Russell 3000 Index to the control group. The sam-

ple period covers eight quarters before and eight quarters after the adop-

tion date of Reg SHO (i.e., May 2, 2005). Following Cheng, Luo, and Yue

[2013], we focus on point and range forecasts from the First Call Database,

because the calculation and classification of forecast news is more straight-

forward for these forecasts. Before conducting our main tests, we first

examine whether Reg SHO affects stock market reaction to management

forecast news. Consistent with the idea that short-sale constraints hinder

incorporation of bad news into prices, we find that stock prices of the pi-

lot firms become significantly more sensitive to negative forecast news after

Reg SHO. In contrast, we find no meaningful or much smaller changes in

price reactions to negative news for stocks in the control group. Moreover,

stock price sensitivity to good news forecasts changes for neither the pilot

nor the control stocks. These results suggest that Reg SHO is an exogenous

shock to price sensitivity to bad news forecasts.

Taking Reg SHO as an exogenous shock to stock price behavior, we

examine whether stock price behavior has a causal effect on managerial

2According to a 2008 NYSE survey, managers appear to be aware of and sensitive to the

impact of eliminating price tests on the amount of short-selling in their firms (Grullon,

Michenaud, and Weston [2012], Fang, Huang, and Karpoff [2013]). They also should have

known whether their firms were selected for the experiment given the public disclosure of

the list of pilot securities on the SEC’s Website s everalmonths before implementation of Reg

SHO and the extensive financial media coverage of the experiment.