Content uploaded by Kee H. Chung

Author content

All content in this area was uploaded by Kee H. Chung on Jan 06, 2021

Content may be subject to copyright.

Hedge Fund Activism and Corporate M&A Decisions#

Szu-Yin (Jennifer) Wua and Kee H. Chunga,b*

a School of Management, University at Buffalo, The State University of New York (SUNY),

Buffalo, NY 14260, USA

b SKK Business School, Sungkyunkwan University, Republic of Korea

# The authors thank Gustavo Manso (Department Editor), an anonymous associate editor, and two

anonymous referees for many valuable comments and suggestions that significantly improved the paper.

The paper also benefitted from the useful suggestions provided by Alon Brav. The authors thank Veljko

Fotak, Sahn-Wook Huh, Jack Jiang, Wei Jiang, Tingting Liu, Dominik Rösch, Cristian Tiu, Jared Wilson,

Brian Wolfe, Shelly E. Webb, seminar participants at the State University of New York at Buffalo,

Duquesne University, Ewha Womans University, National Taiwan University, National Taiwan Normal

University, Sookmyung Women’s University, and Ulsan National Institute of Science and Technology, and

session participants at the 2017 FMA conference, the 2017 SFA conference, and the 2018 MFA conference

for their valuable comments and suggestions. The authors also thank Florian Peters for providing the forced

CEO turnover data, Alon Brav for sharing the hedge fund data, and FactSet for providing technical

assistance with the SharkRepellent database. All of the remaining errors are our own.

* Corresponding author: Kee H. Chung, School of Management, University at Buffalo, The State University

of New York (SUNY), Buffalo, NY 14260, USA; E-mail: keechung@buffalo.edu; Tel: +716-645-3262.

Hedge Fund Activism and Corporate M&A Decisions

Abstract

This paper shows that hedge fund activism is associated with a decrease in mergers and acquisitions (M&A)

and offer premiums and an increase in stock and operating performance. Activist hedge funds improve

target firms’ M&A performance by reducing poor M&A, diversifying M&A, and the M&A of firms with

multiple business segments. Activist hedge funds improve target firms’ M&A decisions by influencing their

governance practices. We show that our results are unlikely driven by selection bias. Overall, activist hedge

funds play an important role in the market for corporate control by increasing the efficiency of target firms’

M&A activities through interventions.

JEL classification: G23, G32, G34

Keywords: Hedge fund activism, M&A performance, Schedule 13D filings, Event study, Abnormal stock

returns, Corporate governance, Selection bias

1

1. Introduction

Prior research shows that corporate mergers and acquisitions (M&A) often destroy shareholder value.

For instance, Moeller et al. (2005) find that shareholders of acquiring firms lost a total of $240 billion from

1998 through 2001 because of value-destroying acquisitions made by some highly valued firms. In this

paper, we examine the role of hedge fund activism in corporate M&A. Prior studies show that hedge fund

activism influences operating performance, dividend policy, corporate governance, and firm productivity.

1

For example, Brav et al. (2008) find that firms targeted by hedge funds have larger dividend payments,

better operating performance, and higher CEO turnovers after activism. Our study contributes to the

literature by exploring whether hedge fund activism affects the extent and quality of corporate M&A.

Opponents of hedge fund activism believe that high dividend payouts and stock repurchases

advocated by hedge funds often force target firms to pass up profitable investment opportunities by reducing

internal capital [see Bebchuk et al. (2015) for a detailed discussion]. In contrast, proponents of hedge fund

activism believe that it creates net benefits for shareholders. Brav et al. (2018) provide supportive evidence

for this position by showing that innovation efficiency increases after hedge fund interventions despite a

reduction in R&D expenditures. Although prior research explores various channels through which hedge

funds influence corporate decisions, there is relatively little evidence regarding the role of hedge fund

activism in M&A decisions and its ramifications for shareholder wealth.

2

We provide such evidence in this

study.

The nature of hedge fund activism as it pertains to M&A activity depends on target firm

characteristics. Hedge funds acquire ownership stakes in small firms to make them attractive takeover

targets and subsequently sell them at a profit. Similarly, they engage in activist risk arbitrage, as analyzed

in Jiang et al. (2018). For large firms, however, hedge fund activism could be quite different, e.g., instead

of making them attractive takeover targets, hedge funds make them better acquirers. Prior research focuses

1

See Kahan and Rock (2007), Becht et al. (2008), Brav et al. (2008), Clifford (2008), Klein and Zur (2009), Boyson

and Mooradian (2011), Brav et al. (2015), and Gantchev et al. (2018).

2

Bethel et al. (1998) show that activist block purchases are followed by a decrease in the frequency of M&A.

However, the authors do not examine the effect of activist block purchases on the quality of M&A.

2

on hedge funds’ ability to make firms attractive takeover targets (e.g., Greenwood and Schor, 2009; Boyson

et al., 2017) or improve the terms of already announced deals after acquiring shares of target firms (e.g.,

Jiang et al., 2018). In contrast, our study focuses on hedge funds’ ability to improve target firms’ acquisition

decisions through activist interventions.

Using a sample of activist campaigns by hedge funds from 1998 to 2012 and M&A announcements

made by US companies from 1993 to 2015, we examine whether hedge fund activism affects the extent and

quality of M&A activities. We employ the difference-in-differences method to isolate the effect of hedge

fund activism on M&A decisions from other effects using the treatment group of firms with hedge fund

activism and the control group of firms with no hedge fund activism. We find that hedge fund activism is

associated with a significant decrease in the number of M&A, the total transaction value of M&A, the

average deal value, and the relative deal size.

3

To assess the effect of hedge fund activism on shareholder

wealth and firm performance, we calculate the acquirer’s cumulative abnormal returns around M&A

announcements, changes in earnings forecasts, and two post-M&A performance measures (i.e., buy-and-

hold abnormal returns and industry-adjusted returns on assets) for both the treatment group and the control

group. We find that hedge fund activism is associated with significant improvements in these performance

metrics.

We explore whether lower M&A activities and more favorable market and analyst reactions are

consequences of improved M&A decisions that result from hedge fund interventions. Analyzing whether

post-intervention acquisitions are of higher quality is essential because it helps demonstrate whether and

how activists mitigate overinvestment and empire-building problems likely to be present in corporate M&A

decisions. This finding also sheds light on the bigger picture of whether activist hedge funds are short-

termists that prevent profitable investments (Bebchuk et al., 2015). Hedge fund activism may improve

M&A performance by reducing bad acquisitions. Prior research shows that diversifying M&A destroy firm

3

These results are in line with the finding of prior research (e.g., Bebchuk et al., 2015; Brav et al., 2018) that firms

targeted by activists typically reduce capital and R&D expenditures.

3

value (see, e.g., Morck et al., 1990; Matsusaka, 1993; Shleifer and Vishny, 1989; Schoar, 2002).

4

We show

that hedge fund activism is associated with the reduced likelihood of an acquisition with a low cumulative

abnormal stock return around the M&A announcement date. We also find that hedge fund activism is

associated with a significant reduction in the frequency and dollar value of diversifying M&A.

Berger and Ofek (1995) show that a firm with diversified business segments is worth less than the

sum of the stand-alone value of each business segment (i.e., the diversification discount). Graham et al.

(2002) argue that the diversification discount could be attributed to the fact that diversifying firms tend to

acquire inefficient assets or companies. Based on these findings, we test whether the effect of hedge fund

activism on M&A activities is stronger for target firms with multiple business segments than for target

firms with a single business segment. Consistent with our expectation, we find that hedge fund activism

plays a significant role in reducing M&A by target firms with multiple business segments.

Prior studies show that entrenched managers or boards are associated with excessive M&A activities

at the expense of shareholders (see, e.g., Masulis et al., 2007, 2009; Harford and Li, 2007; Schmidt, 2015).

Other studies (see Brav et al., 2008; Boyson and Mooradian, 2011; Gantchev et al., 2018) show that activist

hedge funds take governance-related actions to improve managerial and firm performance. These results

suggest that hedge funds’ interventions may have a more substantial impact on target firms with weaker

governance. Consistent with this expectation, we find that target firms with relatively weak governance

before hedge fund interventions exhibit a significant reduction in M&A activities and a considerable

improvement in stock and operating performance afterward. In contrast, target firms with strong governance

do not show such patterns. Overall, these results are consistent with the idea that hedge fund activism

improves target firms’ M&A decisions by helping them to make fewer and better mergers and acquisitions,

especially when their internal governance mechanisms are ineffective.

4

Fan and Goyal (2006) find that vertical mergers are associated with positive wealth effects that are significantly

larger than those for diversifying mergers. Brav et al. (2008) show that activist hedge funds push firms to focus on

their core business. McCahery et al. (2016) show that disagreement with corporate strategies such as diversifying

M&A is one of the most important factors that trigger activist interventions.

4

Activist hedge funds often publicly criticize or demand changes in the target firm’s M&A activities

and strategies and state governance-related objectives in the public campaign. We show that hedge funds’

public and critical comments on the firm’s M&A activities and governance structure are associated with an

improvement in its M&A performance. To explore whether improvements in governance structure are

mechanisms through which hedge fund activism improves M&A performance, we examine whether hedge

fund activism is associated with changes in the board of directors, board independence, and CEO turnover.

5

We show that 28.7% of the activist hedge funds (or their designees) in our study sample obtain board seats

during the post-intervention period and provide evidence that hedge funds’ board appointments help

improve M&A performance. We also find increases in forced CEO turnovers and board independence after

hedge fund interventions and establish links between these increases and M&A performance. Overall, these

results indicate that the superior post-intervention M&A performance may be attributed, at least in part, to

improvements in governance structure associated with hedge fund activism.

Although our empirical findings are consistent with the idea that hedge fund activism improves M&A

decisions, we cannot rule out the possibility that other factors drive the results. For instance, hedge funds

could select firms that already have plans to improve their business strategies or governance practices

voluntarily without their influence or pressure. To explore this possibility, we conduct our analysis using a

sample of hedge fund interventions in which hedge funds and their target firms have confrontational

engagements.

6

We consider this analysis to be an alternative test for the effect of hedge fund activism

because improvements in M&A performance for these firms are less likely due to the voluntary decisions

taken before hedge funds intervene. The majority of firms in this sample ultimately accommodate or adopt

specific changes in governance practices or business strategies proposed by activist hedge funds. Similar to

the results from the full study sample, we find a decrease in M&A activities and an increase in stock returns

for target firms with confrontational interventions.

5

Prior research (e.g., Brav et al., 2008; Boyson and Mooradian, 2011) shows that hedge fund activism involves

changes in these governance variables.

6

We adopt this approach from Brav et al. (2015) and Brav et al. (2018). They use the method to examine the effect of

hedge fund activism on plant productivity and innovation efficiency.

5

To further explore whether improved M&A performance is due to hedge funds’ stock picking skills

or their activism, we also conduct our tests using those hedge funds that switch from passive to active

investors. Following Brav et al. (2015) and Brav et al. (2018), we include a hedge fund-firm pairing in the

switch group if the hedge fund filed at least one Schedule 13G or 13G/A on the firm within one year

preceding the initial Schedule 13D filing. We show that the switch group exhibits lower M&A activities

and better M&A performance than the control sample, adding credence to the idea that performance

improvements are likely due to activism rather than stock-picking skills.

Our study makes several contributions to the literature. First, in contrast to prior research that explores

hedge funds’ ability to convert a company into a takeover target, we investigate whether hedge funds make

their target firms better acquirers. Second, although both our study and Jiang et al. (2019) analyze the effect

of hedge fund activism on acquirers, our study differs in design and extends our understanding of key

mechanisms. That is, our study analyzes whether hedge funds improve the quality and efficiency of a firm’s

acquisition activity through interventions. In contrast, Jiang et al. (2019) focus on whether hedge funds

improve the terms of already announced deals on behalf of the acquiring firm’s shareholders. As a result,

our study sheds further light on the more specific mechanisms through which hedge funds affect target

firms’ M&A decisions. Also, our analysis is based on a sample of 429 firms that are subject to hedge fund

activism, whereas Jiang et al. (2019) use a smaller sample of 58 firms.

Third, our paper provides several results that complement or extend the findings of contemporaneous

work by Gantchev et al. (2019).

7

While Gantchev et al. (2019) focus on cumulative abnormal returns (CARs)

around M&A announcement days, we provide evidence regarding CARs, changes in analysts’ earnings

forecasts, and two post-M&A long-term performance measures. We find evidence that hedge fund activism

is associated with significant improvements in these performance metrics. We also examine the relation

between hedge fund activism and offer premiums and find lower offer premiums after the activist

intervention, suggesting that hedge fund activism reduces overpayments for target assets. Gantchev et al.

7

This paper came to our attention after we had completed the initial draft of our paper.

6

(2019) show that activist hedge funds improve their target firms’ future acquisition decisions by removing

CEOs with poor M&A records, increasing CEOs’ pay-for-performance sensitivity, and changing board

compositions. Our study complements these results by showing that target firms with weak governance

before hedge fund interventions exhibit significant improvements in M&A performance after hedge fund

interventions, whereas target firms with strong governance do not exhibit such improvements. Our study

also shows that hedge funds’ public and critical comments on the target firm’s M&A activities and

governance structure are associated with an improvement in its M&A performance. Finally, we consider

additional governance variables (e.g., the percentage of independent directors) as possible channels through

which activist hedge funds influence corporate M&A decisions.

The paper proceeds as follows. Section 2 describes data sources and variable measurement methods

and presents descriptive statistics. Sections 3 through 5 present our main empirical findings on the relation

between hedge fund activism and the extent and quality of M&A. Section 6 explores various governance

mechanisms through which hedge fund activism affects M&A performance. Section 7 explores other

possible explanations for our empirical results. Section 8 summarizes our main findings and concludes the

paper.

2. Data sources, variable measurement methods, and descriptive statistics

2.1. Data sources and sample construction

We collect data on activist campaigns from the SharkRepellent database that covers corporate

activism campaigns from various sources, including activist press releases, Schedule 13D filings with the

Securities and Exchange Commission (SEC), company websites, and financial trade publications. The

initial sample includes 10,040 activist events in over 30 countries during the 1984-2019 period. The number

of campaigns reduces to 8,158 once we exclude non-US campaigns. We identify 4,286 activist hedge fund

campaigns by matching the activist names in the SharkRepellent database with those in the list of 661

activist hedge funds provided by Alon Brav using the information from corporate websites, news articles

7

that discuss the activists, and SEC filings.

8

We use the announcement date of the activist campaign as the

event date.

9

We match each firm targeted by activist hedge funds with a firm with the same name in the

CRSP/Compustat database. This process reduces the number of campaigns to 3,125 with valid GVKEY

and PERMNO identifiers. Of these, 1,857 have campaign announcement dates between 1998 and 2012.

After excluding 339 campaigns with missing financial/accounting data in the announcement year, our study

sample comprises 1,518 hedge fund campaigns and 1,401 firms with hedge fund activism.

10

We retrieve all mergers and acquisitions for US companies from the Thomson Reuters Securities

Data Company (SDC) database with (i) announcement dates between 1993 and 2015, (ii) disclosed

transaction values greater than $1 million, (iii) the acquirer owns less than 50% of the target stock before

the acquisition and owns more than 50% after the purchase, and (iv) economically significant deals, i.e.,

deals with a relative deal size (i.e., the ratio of the transaction value to the market value of the acquirer)

larger than 1% (Moeller et al., 2005; Hsieh and Walking 2005). As in Gantchev et al. (2019), we include in

the study sample only those firms that undertook at least one qualified M&A during a five-year window

before hedge fund activism.

11

Our final sample comprises 465 activist hedge fund campaigns and 429

unique firm-year observations with prior M&A experience over a five-year window leading up to the

activism announcement year.

8

We manually match activist names and consolidate changes in fund names before merging with Alon Brav’s data.

For instance, we consolidate “Ramius LLC” and “Ramius Capital Group LLC” in the SharkRepellent database as

“RCG HOLDINGS LLC.” Similarly, we consolidate “Riley Investment Management LLC,” “B. Riley and Co., Inc.,”

and “B. Riley & Co. LLC” as “SACC PARTNERS LLC” because Bryant R. Riley owns Riley Investment

Management (RIM) and RIM is the investment adviser to and general partner of SACC Partners. For a campaign with

multiple activists in the SharkRepellent database, we consider it a hedge fund campaign if there is at least one activist

name matched with the list.

9

This is the date of an activist press release or the date of a Schedule 13D or other SEC filing that indicates the

initiation of hedge fund activism.

10

Of the 1,401 firms with hedge fund activism, 1,295 firms had one hedge fund campaign, 95 had two hedge fund

campaigns, and 11 had three hedge fund campaigns. As a result, the total number of hedge fund campaigns during the

study period is 1,518 (= 1,295 + 2 x 95 + 3 x 11).

11

Gantchev et al. (2019) find evidence that hedge funds systematically target firms that had poor M&A in the past.

An implicit assumption in both our paper and Gantchev et al. (2019) is that firms with prior M&A activity are better

suited for the test of overinvestment and empire-building problems than firms without prior M&A activity. It should

be noted, however, that prior M&A activity is only a proxy for overinvestment and empire-building problems because

firms with an overinvestment problem can still have bad acquisitions in the future even if they did not have bad

acquisitions in the past (but overinvested in different ways) and hedge fund activism can mitigate this future behavior.

8

We obtain stock market and accounting data from the Center for Research in Security Prices (CRSP)

and Compustat, stock ownership from the Thomson Reuters Stock Ownership database, and analysts-

related data (e.g., earnings forecasts and the number of analysts following a firm) from the Institutional

Brokers’ Estimate System (I/B/E/S).

2.2. Sample characteristics

Column 1 in Table 1 shows the distribution of 1,401 firms with hedge fund activism from 1998

through 2012. For convenience, we use the acronym FHFA to denote “firms with hedge fund activism.” Of

the 1,401 FHFA, 429 firms had at least one M&A during the five years before hedge fund activism, while

972 firms did not have any M&A during the same five-year period. Column 2 shows the distribution of the

429 firms with prior M&A, and column 3 shows the distribution of the 972 firms with no prior M&A.

Column 4 shows the distribution of the 1,518 activist hedge funds. Of the 1,518 activist hedge funds, 465

are associated with firms with at least one M&A during the five years before hedge fund activism, while

1,053 are associated with firms with no M&A deal during the same five-year period. Column 5 shows the

distribution of the 465 activist hedge funds, and column 6 shows the distribution of the 1,053 activist hedge

funds. Similar to the result reported in Boyson et al. (2017), we find a higher level of hedge fund activism

during the 2005-2008 period.

2.3. An example of hedge fund activism

On March 18, 2008, HealthCor Management filed a Schedule 13D with the SEC indicating that it

owned 6.83% of Magellan Health Services. HealthCor Management had been a passive blockholder since

its initial Schedule 13G filing on August 6, 2007. The Schedule 13D filing included a letter sent to Rene

Lerer, the President and CEO, and the board of directors, which praised Magellan’s growth and large cash

flows but noted that its previously announced plan to make diversifying acquisitions using its large cash

balance as problematic. Magellan had previously highlighted acquisitions as a means for diversification

and growth. For instance, before public pressure from HealthCor Management, Magellan had undertaken

9

two large acquisitions of more than $400 million in value: (1) National Imaging Associates Inc, a provider

of radiology benefits management services, for an estimated $122 million in 2005 and (2) Icore Healthcare,

a provider of specialty pharmaceutical management services, for $285 million in 2006. Both target firms’

three-digit SIC codes are different from that of Magellan Health Services. HealthCor Management noted

that:

“...the underperformance of ICORE since its acquisition makes us cautious about future acquisitions

that are consummated for the sake of diversification; … We see no reason to waste more capital for

the sake of GRANDEUR. …‘empire building acquisitions outside of a company's core competency

are not being viewed favorably by investors.”

Many believed that Magellan was expanding beyond its core behavioral health care business because

of these acquisitions. Although Magellan had previously highlighted acquisitions as a means for growth, it

had only one significant acquisition of $110 million in value during HealthCor Management’s active

monitoring period. HealthCor Management increased its stakes in Magellan in 2008 and made a smooth

exit in 2012 after becoming a passive blockholder in May 2009.

Table IA.1 in the Internet Appendix provides a summary of additional examples from Schedule 13D

filings, other SEC filings, and press releases for our study sample of 429 FHFA in which activist hedge

funds explicitly criticized, condemned, or demanded changes in firms’ M&A activities and strategies.

2.4. Construction of the control sample and descriptive statistics

Following Abadie and Imbens (2006, 2011), we construct a control sample of firms using the nearest-

neighbor matching estimator.

12

The initial set of potential control firms includes all firms in Compustat with

12

This method has the advantage of guaranteeing that the covariates of the two samples, treatment and control, are

similar on average and the two groups have similar values for each characteristic. See Cremers, Giambonda, Sepe,

and Wang (2018) for a comparison of the nearest-neighbor matching estimator and the propensity score

matching (PSM) method.

10

at least one M&A deal during the five years before the activism event year but was not targeted by any

activist hedge funds during the same five years. We first estimate the likelihood of becoming a target firm

using the logistic regression model. Following prior research, we include in the model the market value of

equity (MVE), Tobin’s Q, leverage ratio, sales growth rate, return on assets (ROA), R&D ratio, dividend

yield, institutional ownership, and the number of analysts.

13

We report the results of the logistic regression

in Table IA.2 in the Internet Appendix. We then consider all the explanatory variables that are statistically

significant in the logistic regression

14

to obtain 429 matching pairs of treatment and control firms with the

same two-digit SIC code and fiscal year using the nearest-neighbor matching estimator. We construct the

final control sample using the following matching variables: MVE, Tobin’s Q, leverage ratio, sales growth

rate, dividend yield, institutional ownership, and the number of analysts. The appendix provides definitions

of these variables. We winsorize all accounting variables at the 1st and 99th percentile values.

Table 2 reports the summary statistics in the event year for the 429 matching pairs of the treatment

and control firms. The last three columns show the difference, the t-statistic, and the probability of the

equality of the mean value between the treatment and control groups. Panel A shows that the treatment and

control groups are similar in firm characteristics. Although M&A activity is not one of the matching

variables, the treatment and control firms have similar acquisition expense ratios: the treatment (control)

firms spend an equivalent of 3.9% (4.9%) of their total assets in M&A during the event year. The treatment

and control firms also have similar institutional ownership and analyst following. Panel B shows that, on

average, the treatment (control) firms have 0.438 (0.478) M&A with a total deal value of $275 ($154)

million in the intervention year, and have 1.403 (1.347) M&A with a total deal value of $1,403 ($1,662)

million during the three years before hedge fund interventions. Panel C shows that the treatment and control

firms are also similar in various corporate governance dimensions, including CEO ownership, director

ownership, antitakeover defenses, and the E-index. Panel D provides the summary statistics of M&A deals

13

See Bethel et al. (1998), Brav et al. (2008), Boyson and Mooradian (2011), Brav et al. (2018), and Appel et al.

(2019).

14

All variables with statistical significance have the same sign as those reported in Brav et al. (2008).

11

included in the performance analysis. The average deal transaction value is $379 ($461) million, and the

average offer premium is 0.621 (0.558) for the treatment (control) firms.

15

3. Hedge fund activism, M&A activities, and stock/operating performance

3.1. M&A activities after hedge fund interventions

In this section, we examine the relation between hedge fund activism and M&A activities using the

429 matching pairs of the treatment and control firms with at least one M&A deal during the five years

before the activism event year. The panel data include the observations from three years before an activism

event to three years after an activism event, (t – 3, t + 3), where t denotes the year of the activism event.

We measure M&A activities by the number of M&A, the total transaction value of M&A, the average deal

value, and the average relative deal size in the pre-intervention period (t – 3, t – 1) and the post-intervention

period (t + 1, t + 3), respectively.

16

We use the following regression model to examine the relation between

hedge fund activism and M&A activities:

where is one of the four M&A activity variables described above, is equal to one for FHFA

(i.e., firms with hedge fund activism) and zero for their matching control firms, is equal to one for

the post-intervention period (t + 1, t + 3) and zero otherwise, and denotes a set of control variables.

controls for trends in common to both the treatment and control groups. We use the log of

(1 + Number of M&A), (1 + Total value of M&A), and (1 + Average deal value) in the regressions. Because

we use the log-linear regression, we interpret regression coefficients as semi-elasticities if they are small

numbers (i.e., the percentage change in the dependent variable for a one-unit change in each explanatory

15

We measure the offer premium by (transaction value/the target’s market value of equity) – 1.

16

This approach mitigates the potential bias in standard errors in a pooled-regression model controlling for both the

firm and year fixed effects (Bertand, Fuflo, and Mullainathan, 2004).

12

variable).

17

We also include the year and firm fixed effects ( and ) in the model. Hence, our analysis

focuses on within-firm variations in variables that are both firm-specific and unrelated to common factors.

Following prior research, we use the market value of equity (MVE), firm age, and Tobin’s Q as control

variables (see Bebchuk et al., 2015; Brav et al., 2018).

The results (see the odd-numbered columns in Table 3) show that the estimates are all negative

and significant, indicating that hedge fund activism is associated with a decrease in the number of M&A

transactions, the total transaction value of M&A, the average deal value, and the average relative deal size.

The coefficients in columns 1 and 3 suggest that FHFA had, on average, 20.1% fewer [= 100(e-0.225 – 1)]

M&A and 63.2% lesser [= 100(e-1.001 – 1)] M&A (in dollar value) than the control group in the post-

intervention period. The average deal value of FHFA is 58.4% smaller [= 100(e-0.876 – 1)] than that of the

control group. Considering that the average duration of hedge fund activism is around two years (Boyson

and Mooradian 2011; Brav et al., 2008), we also estimate regression model (1) using a shorter observation

window of (t – 2, t + 2) and find qualitatively similar results (see the even-numbered columns in Table 3).

To assess the robustness of the results, we estimate regression model (1) using the value of each

variable in each year and provide the results in the odd-numbered columns of Table IA.3 in the Internet

Appendix. The coefficients on DHFA × POST are negative and significant, indicating that FHFA had fewer,

lesser, and smaller M&A than the control group in the post-intervention period. The even-numbered

columns in Table IA.3 show that the shorter window (t – 2, t + 2) results are qualitatively similar. We also

estimate regression model (1) using an alternative definition of the dependent variable. Instead of the

number of M&A in each year, we use a binary variable which equals one if there is at least one M&A in

each year and zero otherwise. Columns 1 and 2 of Table IA.4 in the Internet Appendix show the results of

the linear probability model. Columns 3 and 4 report the marginal effect of each explanatory variable at the

17

In the log-linear regression model log(Yi) = α + β Xi + εi , the literal interpretation of β is that a one-unit increase in

X is associated with an expected increase in log(Y) of β units. In terms of Y itself, this means that the expected value

of Y is multiplied by eβ, or alternatively, the expected percentage change in Y (%Y) is equal to 100 (eβ – 1). For

small values of β, eβ ≈ 1+ β. Hence, %Y = 100 (1+ β – 1) = 100β. For example, if β = 0.1, the expected percentage

change in Y is approximately 10%. If β = –1.5, a one unit increase in X is associated with the expected percentage

change in Y of 100 (e-1.5 – 1) = –78%.

13

mean value estimated from the logistic regression model. Columns 1 and 3 show the results using the

window (t – 3, t + 3), and columns 2 and 4 show the results using the window (t – 2, t + 2). The results

show that the coefficients on the interaction term are all negative and significant, indicating that FHFA had

fewer M&A than the control sample in the post-intervention period.

3.2. Stock and operating performance

Following prior studies (e.g., Chen et al., 2007; Masulis et al., 2007; Bouwman et al., 2009), we use

four M&A performance measures to evaluate the effects of hedge fund activism. The first measure is the

cumulative abnormal return around the M&A announcement date. We estimate the market model for each

M&A using daily stock returns and the CRSP value-weighted market returns during a 200-day estimation

window (t – 211, t – 11), where t denotes the announcement date. We then use the estimated parameters to

calculate the cumulative abnormal return (CAR) during the 11-day (t – 5, t +5) event window centered on

the announcement date.

18

The second measure is the change in analysts’ earnings forecasts between the pre- and post-M&A

periods. We measure the change in the acquiring firm’s earnings per share (∆EPS) by the difference

between the first median analyst forecast in the three months after the M&A completion date and the last

median analyst forecast in the three months before the M&A announcement date divided by the absolute

value of the latter. We obtain analyst forecasts from the I/B/E/S database. The third measure is the post-

acquisition long-term operating performance. We measure operating performance by return on asset (ROA),

i.e., the ratio of earnings before tax to total assets. To account for industry-wide factors, we use industry-

adjusted ROA (IROA), i.e., the difference between the acquiring firm’s ROA and the median ROA for all

firms in the same three-digit SIC code as the acquiring firm. We measure the change in IROA (IROA) by

18

We also calculate CAR using the market model parameters estimated over the (-210, -60) period to assess the

robustness of our results. We find that these CAR values are highly correlated (correlation coefficient = 0.98) with the

CAR values calculated from the market model parameters estimated over the (-211, -11) period. Regression results

using these new CAR values are almost identical to those reported in the paper.

14

the difference in IROA between the year of the deal announcement and three years after the deal

announcement.

The fourth measure is the post-M&A long-term stock performance. Following Lyon et al. (1999), we

measure the long-term stock performance by the buy-and-hold abnormal return (BHAR) during the 36

months after the deal announcement (i.e., the difference in buy-and-hold returns between the acquirer and

a reference portfolio, where the reference portfolio is composed of firms that are similar in size and book-

to-market ratio to the acquirer).

19

To obtain reference portfolios, we first sort all NYSE firms into ten

portfolios (size deciles) according to their market capitalizations at the end of June in each year. We then

sort firms in each size decile into quintiles based on their book-to-market ratios in the preceding year. We

place Nasdaq and AMEX firms in the appropriate size and book-to-market portfolios based on their market

capitalizations at the end of June in each year and their book-to-market ratios in the preceding year. Panel

D in Table 2 shows descriptive statistics for the four performance measures.

We employ the following regression model to measure the effect of hedge fund activism on stock

and operating performance:

where is one of the four performance measures described above. are

the same as defined in regression model (1). represents a vector of control variables, including

firm size, firm age, Tobin’s Q, ROA, leverage ratio, relative deal size, method of payment, the status of the

target firm, tender offer, and diversifying deal (Chen et al., 2007; Moeller et al., 2004, 2005; Fuller et al.,

2002; Arikan and Stulz, 2016). We also include the year and industry fixed effects ( and ) in the

regression model.

19

See Fama and French (1993).

15

The results (see the odd-numbered columns in Table 4) show that the coefficients ( on the

interaction term are positive and significant for three of the four performance measures, indicating that

FHFA had better M&A performance than the control firms in the post-intervention period. For instance,

hedge fund activism is associated with a two percentage points improvement in CAR. The effect is

statistically significant and economically meaningful, translating to an increase of $92.67 million in

shareholder value for an average treatment firm.

20

The results also show that hedge fund activism is

associated with a 3.8 percentage points improvement inIROA and a 20.9 percentage points improvement

in ∆EPS. The effect of hedge fund activism on BHAR is positive (0.209) but statistically insignificant.

Overall, these results are inconsistent with the claim that hedge fund activism produces short-term

improvements in performance at the expense of long-term performance. We report the results from the

shorter window of (t – 2, t + 2) in the even-numbered columns. The coefficients are comparable to those

reported in the odd-numbered columns. For instance, hedge fund activism is associated with a 2.1

percentage points increase in CAR and a 17.8 percentage points increase in BHAR (which is equivalent to

5.6 percentage points per annum). One possible interpretation of the smaller increase in CAR relative to

BHAR may be that the stock return during the 11-day event window surrounding the announcement date

does not fully capture the value of hedge fund activism. The coefficients on are all negative and

significant in most regressions, indicating that the M&A performance of firms in the treatment group is

lower than that of the control group before the activism.

21

To shed further light on the economic significance, we construct a variable reflecting the annual

M&A reaction value at the firm level. Following Brav et al. (2018), we first estimate the M&A reaction

20

We obtain this figure from 0.02 x $4,633 million = $92.67 million, where 0.02 is the coefficient on the interaction

term in column 1 of Table 4 and $4,633 is the average market capitalization of treatment firms (see Table 2).

21

Note that the difference in BHAR between the treatment and control group during the post-intervention period is

measured by According to the results in column 7, 0.209 + (–0.351) = –0.142, which is

equivalent to –4.5 percentage points per annum. According to the results in column 8, = 0.178 + (–0.278) =

–0.10, which is equivalent to –3.2 percentage points per annum. Hence, despite large improvements in BHAR

associated with hedge fund activism, the BHAR of the treatment group is still smaller than the BHAR of the control

group because the BHAR of the treatment group was much lower (35.1 and 27.8 percentage points according to the

coefficients on DHFA in columns 7 and 8) than the BHAR of the control group before hedge fund activism.

16

value of each deal by the product of CAR and the market value of equity. We use the market value of equity

on the day before the deal announcement. As a robustness check, we also use the market value of equity at

the end of the last fiscal year. For each acquiring firm, we estimate the annual M&A reaction value by the

summation of the market reaction value across all M&A deals announced in a given year. We then estimate

regression model (2) using the yearly M&A reaction value as the dependent variable and provide the results

in Table IA.5 in the Internet Appendix. The results show that hedge fund activism is associated with an

increase of $88 million or $78 million in market capitalization, depending on whether we use the market

value of equity before the deal announcement or at the end of the last fiscal year. These figures are

comparable to (albeit smaller) to the figure ($92.67 million) we obtain using the coefficient (0.02) on the

interaction term in column 1 of Table 4 and the average market capitalization ($4,633 million) of treatment

firms (see footnote 20).

4. Hedge fund activism and the quality of M&A

We showed in Section 3 that hedge fund activism is associated with a decrease in M&A activities

and an increase in M&A performance measures. In this section, we explore whether FHFA decrease M&A

activities and improve M&A performance by making better M&A decisions (e.g., reducing poor M&A).

4.1. Test of whether hedge fund activism reduces poor acquisitions

Prior research shows that active institutional monitoring improves a firm’s M&A decisions (see, e.g.,

Chen et al., 2007; Roosenboom et al., 2014).

22

In a similar vein, we conjecture that shareholder activism

provided by hedge funds reduces poor acquisitions. To test this conjecture, we modify regression model (1)

by replacing the dependent variable with a dummy variable indicating poor M&A. We first calculate CAR

for each M&A deal and consider an M&A deal poor if it belongs to the bottom tercile of the CAR

22

Roosenboom et al. (2014) suggest that increased interventions by institutions are likely to increase pressure on

managers to withdraw deals with negative announcement returns. However, withdrawal of deals with poor

announcement returns has been viewed as evidence that managers learn from and react to the market, even without

activist interventions (Luo, 2005).

17

distribution in a given year. The dummy variable for poor M&A is equal to one if the firm has at least one

poor M&A in a given year and zero otherwise. Column 1 in Table 5 shows the results of the linear

probability model, and columns 2 and 3 show the results of the logistic regression model with different

fixed effects.

23

The results show that the coefficients on the interaction term are all negative and significant,

indicating that FHFA had fewer poor M&A deals in the post-intervention period than the control firms.

These results are consistent with the idea that hedge fund activism improves M&A performance (i.e., lower

M&A activities, together with more favorable market and analyst reactions to M&A activities) by reducing

the likelihood of a poor acquisition.

4.2. Test of whether hedge fund activism reduces diversifying M&A

Brav et al. (2008) show that activist hedge funds pressure target firms to focus on their core business

and oppose acquisitions outside the scope of their core competency. McCahery et al. (2016) show that

disagreements about corporate strategies such as diversifying M&A often trigger activist interventions. In

a similar vein, we expect hedge fund activism to reduce diversifying M&A. To test this conjecture, we

estimate regression model (1) for diversifying M&A and non-diversifying M&A separately. Following

prior research, we assume that a takeover is a diversifying acquisition if the acquiring and acquired firms

have different three-digit SIC codes.

The first four columns in Table 6 show the results for diversifying M&A and the next four columns

show the results for non-diversifying M&A. We find that the coefficients on the interaction term are all

negative and significant for diversifying M&A. The results indicate that FHFA made 14.4% fewer [= 100(e-

0.155 – 1)] and 48.4% lesser [= 100(e-0.662 – 1)] (in total dollar value) diversifying M&A deals in the post-

intervention period than the control firms. In contrast, the corresponding coefficients for non-diversifying

M&A are not statistically different from zero. As in Table IA.3 in the Internet Appendix, we also repeat

regressions using the number of M&A deals and the total transaction value of M&A deals in each year and

23

We exclude the dummy variable DHFAi when the firm fixed effects are included in the model.

18

show the results in Table IA.6 in the Internet Appendix. The results are qualitatively similar to those in

Table 6, i.e., hedge fund activism is associated with a significant decrease in diversifying M&A but not in

non-diversifying M&A. As in Table IA.4 in the Internet Appendix, we repeat regressions using the two

binary dependent variables and show the results in Table IA.7. Again, we find that the coefficients on the

interaction term are negative and significant only for diversifying M&A. Overall, these results suggest that

hedge fund activism improves M&A performance by reducing diversifying M&A.

4.3. Test of whether the effect of hedge fund activism on M&A activities is stronger for target firms with

multiple business segments

Berger and Ofek (1995) show that the market value of a firm with diversified business segments is

smaller than the sum of the stand-alone value of each business segment. Graham et al. (2002) argue that

diversifying firms tend to acquire inefficient assets or firms. Based on these findings, we conjecture that

the effect of hedge fund activism on M&A activities is stronger for target firms with multiple business

segments than for target firms with a single business segment. To test this conjecture, we employ the

following regression model:

where () is equal to one for firms with multiple segments (single

segment) and zero otherwise, and all other variables are the same as defined in regression model (1). We

obtain the business segment information from the Compustat Industrial Segment (CIS) database. Following

Berger and Ofek (1995), we exclude financial service firms and firms with financial service segments (SIC

19

codes between 6000 and 6999).

24

We consider a firm to have multiple segments if it reports more than one

business or operating segment with different SIC codes one year before the activism event year.

25

Panel A

of Table 2 shows that the mean number of business segments (1.84) for FHFA is similar to that (1.79) for

the control group.

Table 7 shows the regression results using Log(1 + Number of M&A) and Log(1 + Total value of

M&A). For ease of comparison, we report the estimates of for firms with multiple business

segments in the odd-numbered columns and the corresponding estimates ( for firms with a single

business segment in the even-numbered columns. The coefficients on the interaction term are negative and

significant only for firms with multiple business segments and diversifying M&A. These results are

consistent with our expectation that hedge fund activism plays an important role in reducing poor M&A,

mainly for firms with multiple business segments. To assess the robustness of the results, we also estimate

regression model (3) using Log(1 + Number of M&A) and Log(1 + Total value of M&A) in each year and

provide the results in Table IA.8 in the Internet Appendix. We find that the results are qualitatively similar.

4.4. Test of whether hedge fund activism reduces offer premiums

Prior research shows that entrenched managers tend to make overpayments in acquisition deals (see

Harford et al., 2012). We conjecture that hedge fund activism reduces overpayments for target assets. To

test this conjecture, we use a subsample of M&A deals used in Section 3.2 to calculate the offer premium.

Following Moeller et al. (2004, 2005) and Officer (2003), we measure the offer premium by (transaction

value/the target’s market value of equity) – 1 and truncate the value at 0 and 2.

26

We then estimate

regression model (2) using the offer premium as the dependent variable.

Columns 1 and 2 in Table 8 show the results when we include both acquiring firm characteristics and

deal characteristics as control variables. Columns 3 and 4 show the results when we further include target

24

Our results are robust to the inclusion of financial service firms and firms with financial service segments.

25

Tong (2011) also uses the number of segments as a measure of firm diversification.

26

We use the target firm’s market value of equity 30 days before the deal announcement.

20

firm characteristics in the regression model. The results show that FHFA paid about 26-29 percentage points

lower premiums than the control group in the post-intervention period. The effect is statistically significant

and economically sizable, translating to a decrease of around $250 million on average.

27

We note that the

offer premium itself is not a clean measure of the overpayment because larger offer premiums could reflect

larger synergistic values of M&A. Taken together with the results in Section 3.2, however, the market reacts

more favorably to acquisition announcements made by FHFA in the post-intervention period. The synergy

hypothesis is less plausible because it predicts a positive coefficient on the interaction term. Overall, our

evidence indicates that hedge fund activism improves M&A performance by reducing the offer premium.

5. Test of whether the effects of hedge fund activism on M&A activities and stock/operating

performance vary with corporate governance

Numerous studies have examined the relation between corporate governance and M&A decisions.

For example, Masulis et al. (2007, 2009), Harford and Li (2007), and Schmidt (2015) show that entrenched

managers or boards are associated with a high propensity to make acquisitions at the expense of

shareholders. Prior research (see, e.g., Brav et al., 2008; Boyson and Mooradian, 2011; Gantchev et al.,

2018) also shows that activist hedge funds take governance-related actions to rescind antitakeover defenses,

enhance board independence, seek fair board representation, and oust a CEO or chairman. These results

suggest that hedge funds play a more substantial role in target firms with weaker governance through

interventions.

We use the E-index, CEO ownership, and director ownership as our empirical proxies for the quality

of corporate governance. We obtain CEO ownership from the ExecuComp database and director ownership

and antitakeover provisions from the Investor Responsibility Research Center (IRRC) database. Table A1

provides definitions of these variables. Of the 429 treatment firms, we obtain CEO ownership for 210 firms

27

We obtain this figure from 0.28 x $890 million = $250 million, where 0.28 is the coefficient on the interaction term

in column 1 of Table 8, and $890 million is the average market capitalization of target firms 30 days before the deal

announcement (see Table 2).

21

from the ExecuComp database, aggregate director ownership for 185 firms from the IRRC database, and

the E-index for 186 firms from the IRRC database in the activism announcement year. We consider a firm

to have weak (strong) governance if its E-index is above (below) the median value of the sample in the

activism event year, or if its CEO ownership or director ownership is below (above) the median value of

the sample. We then estimate the following regression models:

where WEAK_CG (STRONG_CG) equals one for firms with weak (strong) corporate governance and zero

otherwise, and all other variables are the same as defined in regression models (1) and (2). Panel A1 to

Panel A3 in Table 9 show the results for M&A activities using each of the three governance proxies, and

Panel B1 to Panel B3 show the results for stock and operating performance. We do not report the results of

the control variables for brevity. For ease of comparison, we provide the estimates of and for firms

with weak governance in the odd-numbered columns and the corresponding values and for firms

with strong governance in the even-numbered columns.

Panel A1 to Panel A3 show that the coefficients () on for firms with weak

governance are negative and significant in all regressions, but only one coefficient () is significant for

22

firms with strong governance.

28

Panel B1 to Panel B3 show that the coefficients () on the interaction term

for firms with weak governance are positive and significant in 10 of 12 regressions, but none of the

coefficients () is significant for firms with strong governance. The F-test statistics for the equality of the

coefficients on the interaction term ( are significant in most regressions. These results suggest that

shareholders benefit more from hedge fund activism when firms have weaker governance at the time of an

activism campaign because hedge fund activism improves their governance practices and, as a result,

improves the quality of M&A. In the next section, we investigate whether hedge fund activism is associated

with changes in governance practices and whether such changes can explain changes in M&A performance.

6. Mechanisms through which hedge fund activism affects M&A performance

6.1. Hedge fund activism on the firm’s M&A

Activist hedge funds often criticize or express concerns about the firm’s M&A activities or

acquisition strategies in their campaign. From 13D and 13D/A filings and press releases, we identify 39

cases in our sample in which activist hedge funds publicly and explicitly criticize, condemn, or demand

changes in the target firm’s M&A activities and strategies, or question or oppose a pending acquisition

proposal. We provide an excerpt from each of these 39 cases in Table IA.1 in the Internet Appendix. These

excerpts suggest the critical role of activist hedge funds in shaping target firms’ M&A decisions. To

examine whether these public comments have a positive effect on M&A performance, we estimate the

following regression model:

28

Table IA.9 in the Internet Appendix provides the results when we use yearly values of M&A activities. We find that

the results are qualitatively similar.

23

where is equal to one for the 39 firms and zero for their matching control firms. All other

variables are the same as defined in regression model (1). For brevity, we report only the estimates of ,

and Columns 1, 2, and 3 of Panel A in Table 10 show that the coefficients on the interaction term

are positive and significant when we measure M&A performance by CAR, EPS, or IROA, indicating

that hedge funds’ public and critical comments on a firm’s M&A activities are associated with an

improvement in its M&A performance. As in column 7 of Table 4, we find that the coefficient on the

interaction term is not significant when we measure M&A performance by BHAR.

6.2. Hedge fund activism on corporate governance

Activist hedge funds often propose or demand changes in a target firm’s governance structure (Brav

et al., 2008). Using shareholder activism campaigns reported in the SharkRepellent database, we identify

activist hedge fund campaigns and stated objectives related to the target firm’s governance issues, including

curtailing compensation, promoting board independence, rescinding takeover defenses, removing directors

and officers, seeking a board representative for the activist group, and enhancing other governance-related

metrics. Activist hedge funds explicitly state at least one governance-related objective in the public

campaign for 195 (45%) target firms in our study sample. These results suggest that the improvement of

governance practices is a possible mechanism through which hedge funds influence target firms’ M&A

decisions. To examine whether the M&A performance of these 195 firms is different from that of the

matching control firms, we estimate regression model (6) after replacing PUBLIC with GOVERNANCE,

which is equal to one for the 195 firms and zero for the matching control firms. Columns 5 through 8 of

Panel A in Table 10 show that the coefficients on the interaction term are positive and significant when we

measure M&A performance by CAR, EPS, or IROA, suggesting that hedge funds’ public and critical

comments on targets’ governance-related issues are associated with an improvement in these firms’ M&A

performance.

24

6.3. Effects of specific changes in governance structure on M&A performance

Prior research shows that activist hedge funds take various governance-related actions to improve

managerial and firm performance. In this section, we explore four governance-related variables, including

the three governance variables used in Brav et al. (2008) and Boyson and Mooradian (2011), as possible

channels through which activist hedge funds influence corporate M&A decisions.

6.3.1. Obtaining seats on the board of directors

Prior studies show that shareholder activism is associated with changes in board behavior and board

composition (see Gow et al., 2016; Keusch, 2018; Bebchuk et al., 2020). In our study sample of the 429

FHFA, 122 (28.4%) activist hedge funds obtain at least one board seat in the campaign. Of the 122 activist

hedge funds, 64 obtain one board seat, 30 obtain two board seats, 21 obtain three board seats, and 7 obtain

four or more board seats. These results suggest that board appointment is a possible mechanism through

which hedge funds influence a firm’s M&A decisions. To examine the relation between M&A performance

and hedge funds’ board appointment, we estimate regression model (6) after we replace PUBLIC with

BOARD, which is equal to one for the 122 target firms and zero for the matching control firms. POSTi,t is

equal to one for the post-board-appointment period and zero otherwise. Columns 9 through 12 of Panel A

in Table 10 show that the coefficients on the interaction term are positive and significant when we measure

M&A performance by CAR, EPS, or IROA, suggesting that hedge funds’ board appointment is

associated with an improvement in M&A performance.

6.3.2. CEO-chairman joint appointment

We examine whether hedge fund activism is associated with board independence. Board

independence is likely to be lower when a CEO also serves as the chairman of the board. The joint

appointment gives the CEO concentrated power, allowing for more managerial discretion. The joint

appointment also allows the CEO to effectively control information sharing with other board members,

impeding effective monitoring and independent oversight. We use the data in ExecuComp and manually

25

check and cross-reference the CEO titles in DEF 14A and the 10-K forms. We find that 40.3% (39.9%) of

the treatment (control) group had a CEO who is also the chairman of the board in the pre-intervention period.

The difference in the proportion between the treatment and control groups is statistically insignificant (z-

statistic = 0.12). By the end of the third year from the intervention, 30.6% (38.6%) of the treatment

(control) group had a CEO who is the chairman of the board. The proportion of firms with a CEO-chairman

joint appointment is significantly lower for the treatment group with a p-value of 0.023 (z-statistic = 2.27).

The difference in the proportion between the pre- and post-intervention periods is statistically significant

for the treatment group with a p-value of 0.005 (z-statistic = 2.79) but insignificant for the control group

(z-statistic = 0.35), suggesting that hedge fund activism tends to increase board independence by reducing

joint appointments.

To examine whether M&A performance is related to a CEO-chairman joint appointment, we

estimate the following regression model using the M&A deals of the treatment and control firms in the

post-intervention period:

where JOINTi,t is equal to one for firms that eliminated the CEO-chairman joint appointment and zero

otherwise, and all other variables are the same as defined earlier. Columns 1 through 4 of Panel B in Table

10 show that the coefficients on the interaction term are positive and significant when we measure M&A

performance by CAR or EPS, but insignificant when we measure M&A performance by IROA or BHAR.

Hence, the results partially support the idea that the elimination of the CEO-chairman joint appointment

improves M&A performance.

6.3.3. Board independence measured by the percentage of independent directors

26

An independent director is a board member who does not have a material relationship with the

company. Using the corporate governance data from the Investor Responsibility Research Center (IRRC)

and the Institutional Shareholder Services (ISS), we obtain the percentage of independent directors for each

treated and control firm during the pre- and post-intervention periods. During the pre-intervention period (t

– 3, t – 2, and t – 1), the yearly mean values of the percentage are 70.5%, 71.1%, and 72.7% for the treatment

group, and 68.5%, 71.2%, and 72.3% for the control group. The differences in the mean between the two

groups (2.0, –0.1, and 0.4 percentage points) are not statistically significant. During the post-intervention

period (t + 1, t + 2, and t + 3), the yearly mean values of the percentage are 77.2%, 79.8%, and 80.6% for

the treatment group, and 75.1%, 75.8%, and 76.6% for the control group. The difference in the mean

between the two groups in year t + 2 and t + 3 is four percentage points (significant at the 1% level),

indicating that hedge fund activism is associated with an increase in board independence.

To examine whether M&A performance is related to board independence, we estimate regression

model (7) after we replace JOINTi,t with INDEPi,t, where INDEPi,t is equal to one for firms that increase the

percentage of independent directors after the hedge fund intervention and zero otherwise. Columns 5

through 8 of Panel B in Table 10 show that the coefficients on the interaction term are positive and

significant when we measure M&A performance by CAR, EPS, or IROA, but insignificant when we

measure M&A performance by BHAR. Hence, the results partially support the idea that board

independence improves M&A performance.

6.3.4. CEO turnover

We consider CEO turnover as a possible mechanism through which hedge fund activism affects

M&A performance. We merge the CEO turnover data with our study sample to identify forced CEO

turnovers after hedge fund interventions.

29

The results show that 54 CEOs (12.6%) are forced to leave in

the treatment group, while 17 CEOs (4%) are forced to leave in the control group during the post-

29

We thank Florian Peters for sharing forced CEO turnover data. Classification of CEO turnover as forced or voluntary

follows the method in Peters and Wagner (2014) and Jenter and Kanaan (2015).

27

intervention period. The z-statistic for testing the equality of proportion between the two groups is 4.50 (p

= 0.00), indicating that hedge fund activism increases CEO turnovers. To investigate whether there is a link

between forced CEO turnovers and M&A performance, we estimate regression model (7) after we replace

JOINT with TURNOVER, where TURNOVER is equal to one for firms with forced CEO turnovers and

zero otherwise. Columns 9 through 12 of Panel B in Table 10 show that the coefficients on the interaction

term are positive and significant when we measure M&A performance by CAR, EPS, or IROA, but

insignificant when we measure M&A performance by BHAR. Hence, the results partially support the idea

that CEO turnover improves M&A performance, complementing the finding of Gantchev et al. (2019) that

activist hedge funds improve their target firms’ acquisition decisions by removing CEOs with poor records.

On the whole, these results collectively suggest that hedge fund interventions are associated with

improvements in corporate governance, and these improvements are channels through which hedge fund

activism improves M&A performance.

7. Alternative explanations

We have thus far shown that hedge fund activism is associated with improvements in M&A

performance. However, the hedge fund activism targets in our study sample are not randomly selected, and

thus our results are subject to selection bias. In this section, we explore whether selection bias can explain

our findings.

7.1. Test of selection bias using confrontational hedge fund interventions

An alternative explanation for our findings is that hedge funds select those firms that would improve

M&A performance even in the absence of their intervention. However, this explanation is less applicable

to cases in which hedge funds and target firms have confrontational engagements. In these cases, it would

be less likely that the improved M&A efficiency is due to target firms’ voluntary decisions rather than

28

hedge fund interventions. Based on these considerations, we conduct our analysis using a sample of

confrontational hedge fund interventions.

30

We consider a hedge fund’s activism to be confrontational when the hedge fund is involved in the

following activities specified on Item 4 (Purpose of Transaction) of Schedule 13D or 13D/A (Brav et al.,

2015; Brav et al., 2018): (i) it intends to take over the target or makes a takeover bid; (ii) it sues the target

firm or files a complaint with a court; (iii) it threatens to launch or actually launches a proxy contest; and

(iv) it makes shareholder proposals, denounces the management team, or demands management changes.

31

We consider a hedge fund intervention to be confrontational if it entails any of the actions listed above.

32

We find that 95 interventions in our sample are confrontational, which account for 22.1% of the sample.

33

The majority of our sample firms with confrontational interventions ultimately accommodate or adopt

specific changes in their governance structure, business strategies, or capital structures.

Table IA.10 in the Internet Appendix shows the regression results using the sample of the

confrontational interventions, together with the nearest-neighbor matched estimator control sample.

Dependent variables in columns 1 to 4 of Panel A are the four M&A activity measures. Similar to the results

from a larger sample shown in Table 3, we find a decrease in M&A activities for target firms with

confrontational interventions: a 24.4% reduction [= 100(e-0.280 – 1)] in the number of M&A, a 77.3%

decrease [= 100(e-1.483 – 1)] in the total deal value, a 73.8% decrease [= 100(e-1.341 – 1)] in the average deal

value, and a 2.5% reduction [= 100(e-0.025 – 1)] in the average relative deal size. The coefficients on the

interaction term are larger (in absolute values) than the corresponding figures in Table 3 for the first three

measures of M&A activity, suggesting that confrontational interventions have stronger effects on target

firms’ M&A. Columns 1 and 2 in Panel B show that the effects of confrontational interventions on CAR

30

Brav et al. (2015) and Brav et al. (2018) use samples of confrontational hedge fund activism to investigate whether

hedge fund activism improves plant productivity and innovation efficiency.

31

Our classification of confrontational events may be incomplete because it is only based on the information contained

in 13D and 13D/A filings.

32

In the case of multiple interventions in the same year, we define this firm-year as a confrontational event whenever

there is at least one confrontational hedge fund intervention.

33

Our fraction of confrontational events is comparable to those reported in Brav et al. (2015) and Brav et al. (2018).

29

and ∆EPS are positive and significant. In contrast, columns 3 and 4 show that the effects on ∆IROA and

BHAR are insignificant. These results partially support the idea that the relation between hedge fund

activism and M&A activities/performance is unlikely to be explained by selection bias.

More than half of the confrontational interventions are followed by adoptions of business plans

advocated by hedge funds (e.g., split into multiple stand-alone entities or announce stock repurchases) or

changes in governance practices (e.g., an increase in the number of independent board members or a

replacement of board members) and the composition of the board (e.g., appointment of hedge-fund-

designated individuals on boards or committees).

34

These adoptions and changes prompted by

confrontational interventions might have resulted in improvements in managerial monitoring and

reductions in agency problems, which ultimately improve the efficiency of M&A activities.

7.2. Test of selection bias using hedge funds that switch from Schedule 13G to 13D filings

We conduct an alternative test of the selection bias discussed above based on the SEC blockholder

ownership reporting rules. Under Exchange Act Section 13(G) and Regulation 13D-G, investors who hold

beneficial ownership between 5% and 20% with no intent of shareholder activism can file a shorter form

Schedule 13G. If a hedge fund were to change from a passive to an active investor, it would need to file a

Schedule 13D. Following Brav et al. (2015) and Brav et al. (2018), we conduct our analysis using only

those hedge funds that switch from passive to active investors to separate the activism effect from stock-

picking skill.

35

First, we search for the Schedule 13G and 13G/A filings submitted by hedge funds in our study

sample of 429 FHFA. A hedge fund-firm pairing is defined as ‘switch’ if the hedge fund filed at least one

13G or 13G/A on the firm within one year before the initial 13D filing (i.e., the hedge fund switched from

13G to 13D). Second, we define passive firms as those firms with (i) a 13G or 13G/A filing reported by

34

The actual number of confrontational hedge fund interventions could be larger than this figure because it is only

based on the information contained in 13D and 13D/A filings.

35

Hedge funds are considered to be sophisticated investors with superior ability to pick stocks (Griffin and Xu, 2009).

30

any of the activist hedge funds defined as switch above, (ii) no 13D filing reported by the activist hedge

funds defined as switch above, and (iii) at least one M&A during the five years before the activism event

year. There are two advantages to this sample construction. First, both switch and passive firms are held by

the same set of hedge funds. Second, both switch and passive firms have prior M&A. The final sample

includes observations from three years before and three years after the 13D filing for switch firms (76

interventions) and initial 13G filing for passive firms (463 firms).

36

We use the following regression models to explore whether our results can be explained by stock-

picking skills rather than by hedge fund activism:

where is equal to one if the hedge fund-firm pairing is a switch from 13G to 13D and zero

otherwise and is equal to one if the firm-year observation is within (t + 1, t + 3) years after the year

of the switch to a Schedule 13D for the sample of the switch, and after the year of the Schedule 13G filing

for the passive sample, and zero otherwise. ,, represent the year, firm, hedge fund, and

industry fixed effects. We use the same measures of M&A activities and M&A performance that we used

in Section 3.

Panel A of Table IA.10 in the Internet Appendix reports the results of regression model (8). The

results show that the coefficients on are negative and significant in all regressions,

indicating that the switch group exhibits lower M&A activities than the passive group in the post-

intervention period. Panel B reports the results of regression model (9). The coefficients on

36

The fraction of switch samples (14.1%) is almost identical to the figure (14%) reported in Brav et al. (2018).

31

are positive and significant in all regressions, indicating that the switch group has better

M&A deals than the passive group in the post-intervention period. Overall, these results suggest that the

reduced M&A activities and improved M&A performance can be attributed at least partially to hedge fund

activism rather than stock-picking skills.

8. Summary and concluding remarks

This paper analyzes how and to what extent hedge fund activism influences corporate M&A decisions.

We show that target firms with prior M&A activity generally reduce the frequency and size of M&A after

hedge fund interventions, and the stock market responds more favorably to M&A announcements after

hedge fund interventions. These results are consistent with the idea that hedge fund activism improves firms’

M&A decision by helping them to make slimmer and better merger and acquisition decisions. We show

that target firms with prior M&A activity generally reduce the frequency and size of M&A after hedge fund

interventions, suggesting that hedge fund activism reduces the overinvestment problem. The stock market

responds more favorably to M&A announcements after hedge fund interventions. Hedge fund activism

plays a significant role in reducing poor M&A, diversifying M&A, and M&A of target firms with multiple

business segments, suggesting that hedge funds pressure target firms to focus on their core business. Hedge

fund activism exerts a significant impact on M&A activities and stock returns for target firms with weak

governance, but no significant impact for target firms with strong governance. We conjecture that activist

hedge funds improve their target firms’ M&A performance through their influence on corporate governance

and find supportive evidence by analyzing several specific governance channels.

To assess the effect of selection bias on our results, we analyze a sample of confrontational hedge

fund interventions. We find that the results are qualitatively similar. We also find evidence that the effect

of hedge fund activism on target firms’ M&A decisions is stronger when there is a clear indication of

activism as revealed by hedge funds’ deliberate move to Schedule 13D investor status from Schedule 13G

investor status. This finding suggests that the relations between hedge fund activism and target firms’ M&A

decisions documented in this study are unlikely explained by selection bias.

32

Opponents of hedge fund activism argue that it forces target firms to sacrifice profitable investment

opportunities to satisfy the short-term interests of shareholders. Our empirical results suggest that hedge

fund activism reduces target firms’ M&A activities, low quality ones in particular, by pressuring target

firms to make fewer but better M&A. Overall, our results suggest that hedge funds play an important role

in the market for corporate control by increasing the efficiency of target firms’ M&A activity through their

activist interventions.

33

References

Abadie, A., Imbens, G., 2006. Large sample properties of matching estimators for average treatment effects.

Econometrica 74, 235-267.

Abadie, A., Imbens, G., 2011. Bias-corrected matching estimators for average treatment effects. Journal of

Business & Economic Statistics 29, 1-11.

Appel, I., Gormley, T. A., Keim, D. B., 2019. Standing on the shoulder of giants: The effect of passive

investors on activism. Review of Financial Studies 32, 2720-2774.

Arikan, A. M., Stulz, R. M., 2016. Corporate acquisitions, diversification, and the firm's life cycle. Journal

of Finance 71, 139-194.

Bebchuk, L., Brav, A., Jiang, W., 2015. The long-term effects of hedge fund activism. Columbia Law

Review 115, 1085-1155.

Bebchuk, L., Brav, A., Jiang, W., Keusch, T., 2020. Dancing with activists. Journal of Financial Economics,

forthcoming.

Bebchuk, L., Cohen, A., Ferrell, A., 2009. What matters in corporate governance? Review of Financial

Studies 22, 783-827.

Becht, M., Franks, J., Mayer, C., Rossi, S., 2008. Returns to shareholder activism: Evidence from a clinical

study of the Hermes UK Focus Fund. Review of Financial Studies 22, 3093-3129.

Berger, P., Ofek, E., 1995. Diversification’s effect on firm value. Journal of Financial Economics 37, 39-

65.

Bertrand, M., Duflo, E., Mullainathan, S., 2004. How much should we trust differences-in-differences

estimates? Quarterly Journal of Economics 119, 249-275.

Bethel, J. E., Liebeskind, J. P., Opler,T., 1998. Block share purchases and corporate performance. Journal

of Finance 53, 605-634.

Boyson, N., Mooradian, R., 2011. Corporate governance and hedge fund activism. Review of Derivatives

Research 14, 169-204.

Boyson, N., Gantchev, N., Shivdasani, A., 2017. Activism mergers. Journal of Financial Economics 126,

54-73.

Brav, A., Jiang, W., Kim, H., 2015. The real effects of hedge fund activism: Productivity, asset allocation,

and labor outcomes. Review of Financial Studies 10, 2723-2769.

Brav, A., Jiang, W., Ma, S., Tian, X., 2018. How does hedge fund activism reshape corporate innovation?

Journal of Financial Economics 130, 237-264.

Brav, A., Jiang, W., Partnoy, F., Thomas, R., 2008. Hedge fund activism, corporate governance, and firm

performance. Journal of Finance 63, 1729-1775.

34

Bouwman, C. H. S., Fuller, K., Nain, A. S., 2009. Market valuation and acquisition quality: Empirical

evidence. Review of Financial Studies 22, 633-679.

Chen, X., Harford, J., Li, K., 2007. Monitoring: Which institutions matter? Journal of Financial Economics

86, 279-305.

Clifford, C. P., 2008. Value creation or destruction? Hedge fund as shareholder activists. Journal of

Corporate Finance 14, 323-336.

Cremers, M., Giambona, E., Sepe, S. M., Wang, Y., 2018. Hedge fund activism, firm valuation and stock

returns, Available at SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2693231.

Fama, E. F., French K. R., 1993. Common risk factors in the returns on stocks and bonds. Journal of

Financial Economics 33, 3-56.

Fan, J., Goyal, V., 2006. On the patterns and wealth effects of vertical mergers. Journal of Business 79,

877-902.

Fuller, K., Netter, J., Stegemoller, M., 2002. What do returns to acquiring firms tell us? Evidence from

firms that make many acquisitions. Journal of Finance 57, 1763-1794.

Gantchev, N., Gredil, O., Jotikasthira, C., 2018. Governance under the gun: Spillover effects of hedge fund

activism. Available at SSRN: http://papers.ssrn.com/sol3/papers.cfm?abstract_id= 2356544.

Gantchev, N., Sevilir, M., Shivdasani, A., 2019. Activism and empire building. Journal of Financial

Economics, forthcoming.

Gow, I. D., Shin, S. S., Srinivasan, S., 2016. Consequences to directors of shareholder activism. Harvard

Business School Working Paper, No. 14-071.

Graham, J., Lemmon, M., Wolf, J., 2002. Does corporate diversification destroy value? Journal of Finance

57, 2379-2403.

Greenwood, R., Shor, M., 2009. Investor activism and takeovers. Journal of Financial Economics 92, 362-

375.

Griffin J. M., Xu, J., 2009. How smart are the smart guys? A unique view from hedge fund stock holdings.

Review of Financial Studies 22, 2531-3570.

Harford, J., Humphery-Jenner, M., Powell, R., 2012. Sources of value destruction in acquisitions by

entrenched managers. Journal of Financial Economics 106, 247-261.

Harford, J., Li, K., 2007. Decoupling CEO wealth and firm performance. Journal of Finance 62, 917-949.

Hsieh, J., Walkling, R. A., 2005. Determinants and implications of arbitrage holdings in acquisitions.

Journal of Financial Economics 77, 605-648.

Jenter, D., Kanaan, F., 2015. CEO turnover and relative performance evaluation. Journal of Finance 70,

2155-2184.

35

Jiang, W., Li, T., Mei, D., 2018. Influencing control: Jawboning in risk arbitrage. Journal of Finance 73,

2635-2675.

Jiang, W., Li, T., Mei, D., 2019. Activist M&A arbitrage in acquirers. Finance Research Letters 29, 156-

161.

Kahan, M., Rock, E. B., 2007. Hedge funds in corporate governance and corporate control. University of

Pennsylvania Law Review 155, 1021-1093.

Keusch, T., 2018. Board monitoring, CEO incentives, and shareholder activism. Available at SSRN:

https://ssrn.com/abstract=2746135 or http://dx.doi.org/10.2139/ssrn.2746135.

Klein, A., Zur, E., 2009. Entrepreneurial shareholder activism: Hedge funds and other private investors.

Journal of Finance 64, 187-229.

Luo, Y., 2005. Do insiders learn from outsiders? Evidence from mergers and acquisitions. Journal of

Finance 60, 1951-1982.

Lyon, J. D., Barber, B. M., Tsai, C., 1999. Improved methods for tests of long-run abnormal stock returns.

Journal of Finance 54,165-201.

Masulis, R., Wang, C., Xie, F., 2007. Corporate governance and acquirer returns. Journal of Finance 62,

1851-1889.

Masulis, R., Wang, C., Xie, F., 2009. Agency costs at dual-class companies. Journal of Finance 64, 1697-

1727.

Matsusaka, J. G., 1993. Takeover motives during the conglomerate merger wave. Rand Journal of

Economics 24, 357-379.

McCahery, J., Sautner, Z., Starks, L., 2016. Behind the scenes: The corporate governance preferences of

institutional investors. Journal of Finance 71, 2905-2932.

Moeller, S. B., Schlingemann, F. P., Stulz, R. M., 2004. Firm size and the gains from acquisitions. Journal

of Financial Economics 73, 201-228.

Moeller, S. B., Schlingemann, F. P., Stulz, R. M., 2005. Wealth destruction on a massive scale? A study of

acquiring-firm returns in the recent merger wave. Journal of Finance 60, 757-782.

Morck, R., Shleifer, A., Vishny, R., 1990. Do managerial incentives drive bad acquisitions? Journal of

Finance 45, 31-48.

Officer, M., 2003. Termination fees in mergers and acquisitions. Journal of Financial Economics 69, 431-

467.

Peters, F., Wagner, A., 2014. The executive turnover risk premium. Journal of Finance 69, 1529-1563.

Roosenboom, P., Schlingemann, F. P., Vasconcelos, M., 2014. Does stock liquidity affect incentives to

monitor? Evidence from corporate takeover. Review of Financial Studies 27, 2392-2233.

Schoar, A., 2002. Effects of corporate diversification on productivity. Journal of Finance 57, 2379-2403.

36

Shleifer, A., Vishny, R., 1989. Managerial entrenchment: The case of manager-specific investment. Journal

of Financial Economics 25, 123-140.

Schmidt, B., 2015. Costs and benefits of friendly boards during mergers and acquisitions. Journal of

Financial Economics 117, 424-447.

Tong, Z., 2011. Firm diversification and the value of corporate cash holdings. Journal of Corporate Finance

17, 741-758.

37

Table 1

The number of firms with hedge fund activism and the number of activist hedge funds during 1998-2012

Column 1 shows the distribution of 1,401 firms with hedge fund activism from 1998 through 2012. For

convenience, we use the acronym FHFA to denote “firms with hedge fund activism.” Of the 1,401 FHFA,

429 firms had at least one M&A during the five years before hedge fund activism, while 972 firms did not

have any M&A during the same five-year period. Column 2 shows the distribution of the 429 firms with

prior M&A, and column 3 shows the distribution of the 972 firms with no prior M&A. Column 4 shows

the distribution of the 1,518 activist hedge funds. Of the 1,518 activist hedge funds, 465 are associated with

firms with at least one M&A during the five years before hedge fund activism, while 1,053 are associated

with firms with no M&A deal during the same five-year period. Column 5 shows the distribution of the 465

activist hedge funds, and column 6 shows the distribution of the 1,053 activist hedge funds.

Year

Number of

firms with

hedge fund

activism

With

prior M&A

Without

prior M&A

Number of

activist

hedge funds

With

prior M&A

Without

prior M&A

(1)

(2)

(3)

(4)

(5)

(6)

1998

39

8

31

40

8

32

1999

29

8

21

29

8

21

2000

37

10

27

38

10

28

2001

47

12

35

50

13

37

2002

42

10

32

44

10

34

2003

51

14

37

54

15

39

2004

60

16

44

64

17

47

2005

125

35

90

135

39

96

2006

175

55

120

195

57

138

2007

215

70

145

233

74

159

2008

158

47

111

179

58

121

2009

74

26

48

78

27

51

2010

99

32

67

112

37

75

2011

129

38

91

133

39

94

2012

121

48

73

134

53

81

Total

1,401

429

972

1,518

465

1,053

38

Table 2

Descriptive statistics

This table provides the summary statistics in the event year for the 429 matching pairs of the treatment and control firms. The last three columns show the difference,

the t-statistic, and the probability of the equality of the mean value between the treatment and control groups. Panel A shows that the treatment and control groups

are similar in firm characteristics. Panel B shows that, on average, the treatment (control) firms have 0.438 (0.478) M&A with a total deal value of $275 ($154)

million in the intervention year, and have 1.403 (1.347) M&A with a total deal value of $1,403 ($1,662) million during the three years before hedge fund

interventions. Panel C shows that the treatment and control firms are also similar in various corporate governance dimensions, including CEO ownership, director

ownership, antitakeover defenses, and the E-index. Panel D provides the summary statistics of M&A deals included in the performance analysis. The appendix

provides definitions of these variables. We winsorize all accounting variables at the 1st and 99th percentile values.

Treatment group

Control group

Difference

TreatmentControl

Mean

S.D.

Median

Mean

S.D.

Median

Difference

t-statistic

p-value

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

Panel A: Firm characteristics

Market value of equity ($ million)

4,633

21,892

373

4,224

16,256

516

410

0.31

0.76

Total assets ($ million)

5,241

20,333

569

5,686

26,249

621

-445

-0.28

0.78

Tobin’s Q

1.208

0.739

1.029

1.245

0.798

1.069

-0.037

-0.70

0.48

Book-to-market ratio

0.681

0.888

0.570

0.658

0.563

0.538

0.023

0.46

0.65

Leverage ratio

0.217

0.215

0.177

0.211

0.190

0.188

0.006

0.40

0.69

Sales growth rate

0.144

0.532

0.078

0.205

0.543

0.124

-0.061

-1.66

0.10

Dividend yield

0.028

0.179

0.000

0.021

0.066

0.000

0.008

0.81

0.42

ROA

0.087

0.118

0.099

0.092

0.097

0.099

-0.005

-0.74

0.46

Capital expenditure ratio

0.045

0.056

0.028

0.043

0.058

0.025

0.001

0.38

0.71

R&D ratio

0.094

0.511

0.000

0.103

0.792

0.000

-0.009

-0.19

0.85

Firm age

19.795

14.216

15.000

19.301

13.624

15.000

0.494

0.52

0.60

Acquisition expense ratio

0.039

0.081

0.001

0.049

0.100

0.003

-0.010

-1.59

0.11

Number of segments

1.835

1.259

1.000

1.793

1.190

1.000

0.043

0.50

0.62

Institutional ownership

0.670

0.290

0.713

0.645

0.299

0.716

0.025

1.25

0.21

Number of analysts

6.938

6.820

5.000

6.897

6.665

5.000

0.041

0.09

0.93

Panel B: M&A activity measures

Number of M&A

0.438

0.886

0.000

0.478

0.882

0.000

-0.040

-0.66

0.51

Value of M&A ($ million)

274

3,488

0.000

154

939

0.000

121

0.69

0.49

Average deal value ($ million)

255

3,479

0.000

100

665

0.000

155

0.91

0.37

Average relative deal size

0.063

0.243

0.000

0.071

0.235

0.000

-0.009

-0.52

0.60

Total number of M&A (past 3 years)

1.403

1.384

1.000

1.347

1.736

1.000

0.056

0.52

0.60

Total value of M&A (past 3 years)

496

3,695

41.09

418

1,662

25.00

78.29

0.40

0.69

39

Treatment group

Control group

Difference

TreatmentControl

Mean

S.D.

Median

Mean

S.D.

Median

Difference

t-statistic

p-value

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

Panel C: Governance variables

CEO ownership

0.015

0.040

0.002

0.021

0.051

0.004

-0.006

-1.33

0.18

Director ownership

0.069

0.115

0.028

0.084

0.134

0.038

-0.014

-1.08

0.28

E-index

2.710

1.403

3.000

2.614

1.300

3.000

0.096

0.68

0.50

Classified board

0.559

0.498

1.000

0.574

0.496

1.000

-0.015

-0.28

0.78

Dual class

0.027

0.162

0.000

0.063

0.243

0.000

-0.036

-1.63

0.10

Golden parachute

0.731

0.445

1.000

0.722

0.449

1.000

0.010

0.20

0.84

Poison pill

0.554

0.498

1.000

0.511

0.501

1.000

0.042

0.81

0.42

Limited ability to amend bylaws

0.376

0.486

0.000

0.392

0.490

0.000

-0.016

-0.31

0.76

Limited ability to amend charters

0.500

0.501

0.500

0.438

0.497

0.000

0.063

1.19

0.23

Supermajority voting

0.247

0.433

0.000

0.165

0.372

0.000

0.083

1.95

0.05

Panel D: M&A deal characteristics

Deal value ($ million)

379

2,765

43

460

1,752

57

-90

-0.86

0.22

Relative deal size

0.192

0.509

0.063

0.182

0.364

0.078

0.010

-0.49

0.71

Cash payment

0.759

0.428

1.000

0.751

0.433

1.000

0.009

-0.45

0.66

Subsidiary target

0.316

0.465

0.000

0.325

0.469

0.000

-0.011

-0.53

0.33

Tender offer

0.031

0.172

0.000

0.019

0.138

0.000

0.011

-1.58

0.94

Diversifying deals

0.379

0.485

0.000

0.345

0.476

0.000

0.032

-1.48

0.95

Target 52-week high

0.868

1.318

0.439

1.245

4.380

0.255

-0.377

-0.88

0.19

Target firm size ($ million)

890

1,801

183

1,056

1,813

292

-166

-0.69

0.25

Offer premium

0.621

0.548

0.456

0.558

0.459

0.444

0.063

-0.93

0.82

CAR

0.001

0.090

0.005

0.006

0.094

0.006

0.005

1.28

0.10

BHAR

-0.366

1.102

-0.348

-0.081

0.875

-0.164

0.285

5.28

0.00

∆EPS

0.003

1.553

0.057

0.103

0.892

0.071

-0.100

-1.62

0.11

∆IROA

-0.024

0.122

-0.013

-0.026

0.115

-0.013

0.002

0.21

0.83

40

Table 3

Hedge fund activism and M&A activities

This table shows the results of the following regression model:

where is one of the four M&A activity variables, is equal to one for FHFA (i.e., firms with hedge fund activism) and zero for their matching control

firms, is equal to one for the post-intervention period (t + 1, t + 3) and zero otherwise, and denotes a set of control variables. We estimate the

model using the 429 matching pairs of the treatment and control firms with at least one M&A deal during the five years before the activism event year. We measure

M&A activities by the number of M&A, the total transaction value of M&A, the average deal value, and the average relative deal size in the pre-intervention period

(t – 3, t – 1) and the post-intervention period (t + 1, t + 3), respectively. We use the average values of the control variables during the pre- and post-intervention

periods. The odd-numbered columns show the results using the observations from three years before an activism event to three years after an activism event, (t –

3, t + 3), where t denotes the activism event year. The even-numbered columns show the results using the observations from two years before an activism event to

two years after an activism event, (t – 2, t + 2). Figures in parentheses are standard errors clustered at the match level. ***, **, and * indicates significance at the 1%,

5%, and 10% level, respectively.

Log(1 + Number of M&A)

Log(1 + Total value of M&A)

Log(1 + Average deal value)

Average relative deal size

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

DHFA POST

-0.225***

-0.170***

-1.001***

-0.720**

-0.876***

-0.638**

-0.029**

-0.029*

(0.066)

(0.064)

(0.286)

(0.297)

(0.268)

(0.280)

(0.014)

(0.017)

POST

-0.410***

-0.279***

-1.527***

-1.107***

-1.264***

-0.928***

-0.005

-0.004

(0.073)

(0.072)

(0.326)

(0.342)

(0.306)

(0.319)

(0.013)

(0.018)

Log(

)

0.138**

0.071

0.746***

0.581**

0.722***

0.608**

-0.015

-0.015

(0.059)

(0.060)

(0.233)

(0.293)

(0.210)

(0.278)

(0.017)

(0.019)

Log(

)

-0.273

-0.706**

-0.278

-2.241

-0.029

-1.784

-0.034

-0.069

(0.246)

(0.314)

(1.039)

(1.436)

(0.988)

(1.350)

(0.061)

(0.105)

0.085

0.074

0.026

-0.036

-0.099

-0.168

0.005

0.001

(0.058)

(0.059)

(0.218)

(0.250)

(0.200)

(0.240)

(0.014)

(0.016)

Constant

0.892

2.081**

0.134

4.130

-0.988

2.551

0.236

0.331

(0.767)

(0.977)

(3.127)

(4.237)

(2.920)

(3.982)

(0.213)

(0.375)

Year fixed effects

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Firm fixed effects

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Sample size

1,575

1,575

1,575

1,575

1,575

1,575

1,575

1,575

Adjusted R2

0.591

0.541

0.548

0.465

0.488

0.407

0.297

0.282

41

Table 4

Hedge fund activism and M&A performance

We use the following regression model to measure the effect of hedge fund activism on stock and operating performance:

where is one of the four measures of stock and operating performance, is equal to one for FHFA (i.e., firms with hedge fund activism) and

zero for their matching control firms, is equal to one for the post-intervention period and zero otherwise, and represents a vector of control

variables. The four performance measures are (1) the cumulative abnormal return (CAR) during the 11-day (t – 5, t +5) event window centered on the announcement

date, (2) the change in analysts’ earnings forecasts (∆EPS) between the pre- and post-M&A periods (i.e., the difference between the first median analyst forecast

in the three months after the M&A completion date and the last median analyst forecast in the three months before the M&A announcement date divided by the

absolute value of the latter), (3) the change in industry-adjusted return on assets (IROA) between the year of the deal announcement and three years after the deal

announcement, and (4) the buy-and-hold abnormal return (BHAR) during the 36 months after the deal announcement. The odd-numbered columns show the results

using observations from three years prior to an activism event to three years after an activism event (i.e., (t – 3, t + 3)) and even columns report the results using

observations from two years before an activism event to two years after an activism event (i.e., (t – 2, t + 2)). Figures in parentheses are standard errors clustered

at the match level. ***, **, and * indicates significance at the 1%, 5%, and 10% level, respectively.

CAR

EPS

IROA

BHAR

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

DHFA POST

0.020**

0.021**

0.209*

0.169

0.038**

0.051***

0.209

0.178*

(0.009)

(0.010)

(0.113)

(0.115)

(0.017)

(0.017)

(0.130)

(0.100)

POST

-0.009

-0.008

-0.065

-0.069

-0.016

-0.024**

0.076

0.012

(0.006)

(0.007)

(0.080)

(0.088)

(0.012)

(0.012)

(0.100)

(0.089)

DHFA

-0.021***

-0.021***

-0.168*

-0.134

-0.032**

-0.036**

-0.351***

-0.278***

(0.006)

(0.008)

(0.094)

(0.089)

(0.016)

(0.016)

(0.117)

(0.063)

Log(MVE)

-0.006***

-0.007***

0.021

0.020

0.001

0.003

0.030

0.024

(0.002)

(0.002)

(0.024)

(0.022)

(0.003)

(0.004)

(0.020)

(0.018)

Log(Age)

0.000

0.002

-0.008

-0.008

0.022***

0.026***

-0.034

0.059

(0.005)

(0.006)

(0.059)

(0.071)

(0.008)

(0.010)

(0.089)

(0.056)

Tobin's Q

0.008**

0.012***

-0.015

0.168

0.017**

0.010

0.079

0.195***

(0.003)

(0.004)

(0.192)

(0.168)

(0.007)

(0.009)

(0.090)

(0.054)

ROA

0.077**

0.076**

-0.450

0.422

-0.372***

-0.298***

0.960**

0.888***

(0.033)

(0.035)

(0.670)

(0.499)

(0.088)

(0.102)

(0.391)

(0.241)

42

Leverage ratio

0.003

-0.005

-0.439

-0.090

-0.072

-0.054

0.170

-0.006

(0.012)

(0.014)

(0.320)

(0.253)

(0.067)

(0.056)

(0.203)

(0.149)

Relative deal size

-0.006

-0.010

-0.191

-0.132

0.027

0.057

0.125

0.235

(0.005)

(0.006)

(0.138)

(0.150)

(0.033)

(0.041)

(0.264)

(0.194)

Tender offer

0.002

0.002

0.052

0.038

0.004

-0.010

0.105

-0.038

(0.012)

(0.014)

(0.092)

(0.077)

(0.012)

(0.017)

(0.171)

(0.148)

Cash payment

0.002

0.001

0.161

0.277

-0.050

-0.004

0.218

0.234

(0.006)

(0.008)

(0.334)

(0.360)

(0.035)

(0.036)

(0.250)

(0.194)

Subsidiary target

0.016***

0.017***

-0.041

-0.046

0.006

0.004

-0.043

-0.084*

(0.005)

(0.005)

(0.071)

(0.067)

(0.007)

(0.010)

(0.049)

(0.048)

Diversifying deal

-0.008**

-0.005

-0.101

-0.151**

-0.011

-0.013

-0.134*

-0.085*

(0.004)

(0.005)

(0.068)

(0.068)

(0.007)

(0.009)

(0.072)

(0.049)

Constant

-0.010

-0.098

0.246

0.004

-0.007

0.020

-0.094

-0.822**

(0.045)

(0.085)

(0.239)

(0.238)

(0.040)

(0.041)

(0.350)

(0.377)

Year fixed effects

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Industry fixed effects

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Sample size

2,027

1,537

1,555

1,140

1,140

930

1,375

1,168

Adjusted R2

0.072

0.080

0.025

0.020

0.271

0.387

0.153

0.183

43

Table 5

Hedge fund activism and the avoidance of poor M&A

This table shows the results of the following regression models:

where is equal to one if the firm has at least one poor M&A in a given year and zero otherwise,

is equal to one for FHFA (i.e., firms with hedge fund activism) and zero for their matching control

firms, is equal to one for the post-intervention period and zero otherwise, and represents

a vector of control variables. We first calculate CAR for each M&A deal and consider an M&A deal poor

if it belongs to the bottom tercile of the CAR distribution in a given year. Column 1 shows the results of

the linear probability model, and columns 2 and 3 show the results of the logistic regression model with

different fixed effects. The coefficients in columns 2 and 3 measure the marginal effects of each explanatory

variable at the mean values. Figures in parentheses are standard errors clustered at the firm level. ***, **,

and * indicates significance at the 1%, 5%, and 10% level, respectively.

Linear probability model

Logistic model

Logistic model

Variables

(1)

(2)

(3)

DHFAPOST

-0.051*

-0.055***

-0.054***

(0.026)

(0.013)

(0.012)

POST

0.028

-0.006

-0.006

(0.023)

(0.014)

(0.013)

Log(MVE)

0.064***

0.019***

0.019***

(0.016)

(0.003)

(0.003)

Log(Age)

-0.114*

-0.031***

-0.021***

(0.065)

(0.007)

(0.008)

Tobin's Q

-0.017

-0.001

-0.002

(0.011)

(0.004)

(0.005)

Constant

0.172

(0.191)

Year fixed effects

Yes

Yes

Yes

Industry fixed effects

No

No

Yes

Firm fixed effects

Yes

No

No

Sample size

5,471

5,471

5,425

Adjusted R2

0.081

Pseudo R2

0.045

0.062

44

Table 6

Hedge fund activism and diversifying and non-diversifying M&A activities

This table shows the results of the following regression model:

where is one of the four M&A activity variables, is equal to one for FHFA (i.e., firms with hedge fund activism) and zero for their matching control

firms, is equal to one for the post-intervention period and zero otherwise, and denotes a set of control variables. We estimate the model using the

429 matching pairs of the treatment and control firms with at least one M&A deal during the five years before the activism event year. We assume that a takeover

is a diversifying acquisition if the acquiring and acquired firms have different three-digit SIC codes. We measure the extent of M&A activities by the number of

M&A, the total transaction value of M&A, the average deal value, and the average relative deal size in the pre-intervention period and the post-intervention period,

respectively. We use the average values of the control variables during the pre- and post-intervention periods. Columns 1, 2, and 3 show the results for diversifying

M&A. Columns 4, 5, and 6 show the results for non-diversifying M&A. We show the results using the observations from three years before an activism event to

three years after an activism event. Figures in parentheses are standard errors clustered at the match level. ***, **, and * indicates significance at the 1%, 5%, and

10% level, respectively.

Diversifying M&A

Non-diversifying M&A

Log(1 +

Number of

M&A)

Log(1 + Total

value of

M&A)

Log(1 +

Average deal

value)

Average

relative deal

size

Log(1 +

Number of

M&A)

Log(1 + Total

value of

M&A)

Log(1 +

Average deal

value)

Average

relative deal

size

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

DHFA POST

-0.155***

-0.662**

-0.602**

-0.059*

-0.094

-0.429

-0.340

-0.011

(0.059)

(0.264)

(0.244)

(0.032)

(0.060)

(0.276)

(0.256)

(0.028)

POST

-0.268***

-1.058***

-0.918***

-0.008

-0.225***

-0.880***

-0.757***

-0.033

(0.067)

(0.311)

(0.291)

(0.036)

(0.062)

(0.304)

(0.284)

(0.034)

Log(

)

0.086*

0.410**

0.385**

-0.015

0.086

0.559**

0.553**

-0.036

(0.051)

(0.207)

(0.190)

(0.035)

(0.056)

(0.258)

(0.239)

(0.041)

Log(

)

0.161

0.900

0.854

-0.079

-0.312

-0.993

-0.829

0.060

(0.217)

(1.025)

(0.974)

(0.122)

(0.236)

(0.984)

(0.893)

(0.185)

0.041

0.074

0.043

0.007

0.071

0.085

-0.003

0.015

(0.050)

(0.222)

(0.210)

(0.037)

(0.057)

(0.220)

(0.200)

(0.022)

Constant

-0.441

-2.276

-1.992

0.433

0.400

-0.337

-0.756

0.096

(0.747)

(3.420)

(3.226)

(0.476)

(0.727)

(3.126)

(2.860)

(0.588)

Year fixed effects

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Firm fixed effects

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Sample size

1,575

1,575

1,575

1,575

1,575

1,575

1,575

1,575

Adjusted R2

0.433

0.463

0.446

0.355

0.593

0.546

0.499

0.146

45

Table 7

Hedge fund activism and diversifying and non-diversifying M&A activities for firms with single and multiple business segments

This table shows the results of the following regression model:

where () is equal to one for firms with multiple segments (single segment) and zero otherwise, is equal to one for

FHFA (i.e., firms with hedge fund activism) and zero for their matching control firms, is equal to one for the post-intervention period and zero otherwise,

and denotes a set of control variables. We obtain the business segment information from the Compustat Industrial Segment (CIS) database. We consider

a firm to have multiple segments if it reports more than one business or operating segment with different SIC codes one year before the activism event year. For

ease of comparison, we report the estimates of for firms with multiple business segments in the odd-numbered columns and the corresponding estimates

( for firms with a single business segment in the even-numbered columns. Figures in parentheses are standard errors clustered at the match level. ***, **,

and * indicates significance at the 1%, 5%, and 10% level, respectively.

Diversifying M&A

Non-diversifying M&A

Log(1 + Number of M&A)

Log(1 + Total value of M&A)

Log(1 + Number of M&A)

Log(1 + Total value of M&A)

MULTI_

SEGMENT

SINGLE_

SEGMENT

MULTI_

SEGMENT

SINGLE_

SEGMENT

MULTI_

SEGMENT

SINGLE_

SEGMENT

MULTI_

SEGMENT

SINGLE_

SEGMENT

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

DHFAPOST

-0.195**

-0.066

-0.675*

-0.258

0.090

-0.130

0.050

-0.562

(0.085)

(0.068)

(0.385)

(0.314)

(0.078)

(0.087)

(0.362)

(0.374)

POST

-0.250***

-0.247***

-1.108***

-0.930***

-0.196***

-0.052

-0.823**

-0.337

(0.073)

(0.068)

(0.351)

(0.329)

(0.068)

(0.084)

(0.323)

(0.341)

Log(

)

0.047

0.248

0.053

0.275

(0.048)

(0.212)

(0.053)

(0.239)

Log(

)

0.307

1.343

-0.620**

-1.531

(0.194)

(0.996)

(0.254)

(0.980)

0.008

0.006

0.041

0.070

(0.044)

(0.194)

(0.048)

(0.196)

Constant

-0.730

-2.890

1.365*

3.697

(0.692)

(3.466)

(0.812)

(3.121)

Year fixed effects

Yes

Yes

Yes

Yes

Firm fixed effects

Yes

Yes

Yes

Yes

Sample size

1,409

1,409

1,409

1,409

Adjusted R2

0.687

0.715

0.792

0.776

F-test (MULTI_SEGMENT =

SINGLE_SEGMENT)

1.49

0.72

3.73

1.48

Prob > F

0.222

0.398

0.054

0.224

46

Table 8

Hedge fund activism and the offer premium

This table shows the results of the following regression model:

where Premiumi,t is equal to Transaction valueTarget firm’s market value of equity is equal to one for

FHFA (i.e., firms with hedge fund activism) and zero for their matching control firms, is equal to one for the

post-intervention period and zero otherwise, and denotes a set of control variables. We include acquiring

firm characteristics and deal characteristics in columns 1 and 2, and add target firm characteristics in columns 3 and

4. Figures in parentheses are standard errors clustered at the match level. ***, **, and * indicates significance at the 1%,

5%, and 10% level, respectively.

Offer premium

Variables

(1)

(2)

(3)

(4)

DHFAPOST

-0.281*

-0.283*

-0.289*

-0.262*

(0.156)

(0.154)

(0.156)

(0.155)

POST

0.151*

0.107

0.104

0.091

(0.088)

(0.093)

(0.096)

(0.082)

DHFA

0.107

0.086

0.083

0.077

(0.075)

(0.077)

(0.076)

(0.073)

Log(MVE)-Acquirer

0.025

0.018

0.054**

0.063***

(0.019)

(0.020)

(0.022)

(0.023)

Log(Age)-Acquirer

0.117**

0.101

0.088

0.073

(0.059)

(0.068)

(0.064)

(0.067)

Tobin's Q-Acquirer

0.046

0.056

0.037

-0.003

(0.050)

(0.067)

(0.067)

(0.056)

Relative deal size

0.057

0.024

0.082

0.080

(0.050)

(0.050)

(0.060)

(0.059)

Tender offer

0.147

0.140

0.126

0.144

(0.094)

(0.103)

(0.099)

(0.098)

Equity payment

-0.256***

-0.257***

-0.252***

-0.221***

(0.087)

(0.088)

(0.082)

(0.077)

Diversifying deal

-0.009

0.002

0.000

0.023

(0.063)

(0.068)

(0.067)

(0.066)

Target 52-week high

0.055***

0.056***

0.049***

0.048***

(0.016)

(0.017)

(0.016)

(0.018)

Log(MVE)-Target

-0.060**

-0.065**

(0.025)

(0.026)

Tobin's Q-Target

-0.001

-0.060

(0.031)

(0.134)

Constant

0.118

0.298

0.346

0.412

(0.214)

(0.274)

(0.256)

(0.279)

Year fixed effects

Yes

Yes

Yes

Yes

Industry fixed effects-Acquirer

No

Yes

Yes

No

Industry fixed effects-Target

No

No

No

Yes

Sample size

223

223

223

223

Adjusted R2

0.292

0.282

0.296

0.306

47

Table 9

The effects of hedge fund activism on M&A activities and stock/operating performance for firms with weak and strong corporate governance

This table shows the results of the following regression models:

where WEAK_CGi (STRONG_CGi) equals one for firms with weak (strong) corporate governance and zero otherwise, is equal to one for FHFA (i.e., firms with

hedge fund activism) and zero for their matching control firms, is equal to one for the post-intervention period and zero otherwise, and denotes a set of

control variables. We use the E-index, CEO ownership, and director ownership as our empirical proxies for the quality of corporate governance. We consider a firm to

have weak (strong) governance if its E-index is above (below) the median value of the sample in the activism event year, or if its CEO ownership or director ownership

is below (above) the median value of the sample. Panel A1 to Panel A3 show the regression results for M&A activities using each of the three governance proxies,

respectively. Dependent variables are the number of M&A, the total transaction value of M&A, the average deal value, and the average relative deal size in the pre-

intervention period and the post-intervention period, respectively. Panel B1 to Panel B3 show the regression results for M&A performance measures using each of the

three governance proxies, respectively. As in Table 4, we show the results for four performance measures (i.e., CAR, ∆EPS, IROA, and BHAR). Figures in parentheses

are standard errors clustered at the match level. ***, **, and * indicates significance at the 1%, 5%, and 10% level, respectively.

Panel A: Regression results for M&A activities

Panel A-1: E-index as governance proxy

Log(1 + Number of M&A)

Log(1 + Total value of M&A)

Log(1 + Average deal value)

Average relative deal size

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

DHFAPOST

-0.267*

-0.213

-1.374**

-1.257**

-1.257**

-0.953

-0.034*

-0.035

(0.147)

(0.167)

(0.642)

(0.846)

(0.581)

(0.800)

(0.019)

(0.029)

POST

-0.365**

-0.337**

-1.592**

-1.365**

-1.365**

-1.016

0.009

-0.012

(0.159)

(0.156)

(0.659)

(0.702)

(0.593)

(0.665)

(0.020)

(0.030)

Constant

0.944

-1.571

-3.595

0.106

(1.667)

(6.624)

(6.075)

(0.397)

Firm characteristics

Yes

Yes

Yes

Yes

Year fixed effects

Yes

Yes

Yes

Yes

Firm fixed effects

Yes

Yes

Yes

Yes

Sample size

662

662

662

662

Adjusted R2

0.786

0.752

0.726

0.649

F-test (WEAK_CG =

STRONG_CG)

0.064

0.091

0.096

0.000

Prob > F

0.800

0.763

0.757

0.985

48

Table 9 (continued)

Panel A-2: CEO ownership as governance proxy

Log(1 + Number of M&A)

Log(1 + Total value of M&A)

Log(1 + Average deal value)

Average relative deal size

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

DHFAPOST

-0.348***

-0.119

-1.882***

-0.734

-1.711***

-0.701

-0.035**

-0.029

(0.121)

(0.127)

(0.636)

(0.565)

(0.599)

(0.524)

(0.017)

(0.021)

POST

-0.354***

-0.553***

-1.481***

-2.222***

-1.224**

-1.847***

-0.007

-0.002

(0.111)

(0.128)

(0.540)

(0.597)

(0.514)

(0.554)

(0.017)

(0.020)

Constant

-0.282

-5.377

-5.668

0.242

(1.250)

(6.080)

(5.830)

(0.301)

Firm characteristics

Yes

Yes

Yes

Yes

Year fixed effects

Yes

Yes

Yes

Yes

Firm fixed effects

Yes

Yes

Yes

Yes

Sample size

862

862

862

862

Adjusted R2

0.789

0.738

0.703

0.623

F-test (WEAK_CG =

STRONG_CG)

1.831

1.853

1.640

0.054

Prob > F

0.177

0.175

0.201

0.817

Panel A-3: Director ownership as governance proxy

Log(1 + Number of M&A)

Log(1 + Total value of M&A)

Log(1 + Average deal value)

Average relative deal size

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

DHFAPOST

-0.252*

-0.260

-1.527**

-1.112

-1.457**

-0.947

-0.030*

-0.014

(0.149)

(0.166)

(0.750)

(0.687)

(0.695)

(0.625)

(0.016)

(0.026)

POST

-0.316**

-0.345**

-1.343**

-1.513**

-1.128*

-1.332**

0.009

-0.012

(0.139)

(0.163)

(0.679)

(0.683)

(0.636)

(0.633)

(0.016)

(0.030)

Constant

0.546

-4.899

-6.618

0.469*

(1.739)

(6.986)

(6.473)

(0.242)

Firm characteristics

Yes

Yes

Yes

Yes

Year fixed effects

Yes

Yes

Yes

Yes

Firm fixed effects

Yes

Yes

Yes

Yes

Sample size

638

638

638

638

Adjusted R2

0.791

0.761

0.732

0.607

F-test (WEAK_CG =

STRONG_CG)

0.001

0.169

0.307

0.222

Prob > F

0.973

0.681

0.580

0.638

49

Table 9 (continued)

Panel B: Regression results for M&A performance

Panel B-1: E-Index as governance proxy

CAR

EPS

IROA

BHAR

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

DHFAPOST

0.034*

-0.009

0.430

0.004

0.022**

-0.001

0.495***

0.139

(0.019)

(0.014)

(0.318)

(0.108)

(0.010)

(0.012)

(0.139)

(0.202)

POST

-0.009

0.005

-0.071

0.152*

-0.003

0.005

-0.162

-0.019

(0.013)

(0.009)

(0.158)

(0.090)

(0.006)

(0.007)

(0.108)

(0.184)

Constant

0.021

-0.036

0.003

0.080

(0.023)

(0.199)

(0.034)

(0.302)

Firm characteristics

Yes

Yes

Yes

Yes

Year fixed effects

Yes

Yes

Yes

Yes

Industry fixed effects

Yes

Yes

Yes

Yes

Sample size

903

783

824

750

Adjusted R2

0.151

0.339

0.643

0.273

F-test (WEAK_CG =

STRONG_CG)

3.977

2.862

2.183

2.301

Prob > F

0.047

0.092

0.141

0.131

Panel B-2: CEO Ownership as governance proxy

CAR

EPS

IROA

BHAR

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

DHFAPOST

0.022*

-0.024

0.482*

0.095

0.018*

-0.006

0.565***

0.214

(0.013)

(0.017)

(0.251)

(0.219)

(0.011)

(0.011)

(0.190)

(0.167)

POST

-0.008

0.018*

-0.172

0.131

-0.003

0.001

-0.091

-0.018

(0.009)

(0.009)

(0.170)

(0.165)

(0.006)

(0.007)

(0.115)

(0.150)

Constant

0.017

-0.307

0.034

0.603*

(0.025)

(0.336)

(0.028)

(0.326)

Firm characteristics

Yes

Yes

Yes

Yes

Year fixed effects

Yes

Yes

Yes

Yes

Industry fixed effects

Yes

Yes

Yes

Yes

Sample size

1,093

785

824

794

Adjusted R2

0.166

0.310

0.319

0.261

F-test (WEAK_CG =

STRONG_CG)

5.169

3.166

2.917

1.866

Prob > F

0.024

0.077

0.089

0.173

50

Panel B-3: Director ownership as governance proxy

CAR

EPS

IROA

BHAR

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

WEAK_CG

STRONG_CG

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

DHFAPOST

0.028*

-0.005

0.382

-0.037

0.046*

-0.034

0.528**

0.200

(0.016)

(0.016)

(0.236)

(0.195)

(0.027)

(0.033)

(0.204)

(0.156)

POST

-0.018*

0.002

-0.162

-0.002

-0.015

0.016

-0.124

-0.157

(0.011)

(0.010)

(0.146)

(0.188)

(0.017)

(0.019)

(0.111)

(0.143)

Constant

0.025

-0.512

-0.052

0.296

(0.028)

(0.416)

(0.047)

(0.374)

Firm characteristics

Yes

Yes

Yes

Yes

Year fixed effects

Yes

Yes

Yes

Yes

Industry fixed effects

Yes

Yes

Yes

Yes

Sample size

889

774

726

718

Adjusted R2

0.249

0.278

0.443

0.228

F-test (WEAK_CG =

STRONG_CG)

3.300

2.615

4.126

1.650

Prob > F

0.071

0.107

0.044

0.200

51

Table 10

Mechanisms through which hedge fund activism affects M&A performance

Panel A shows the results of the following regression model using the event sample and the matched control sample:

where = PUBLICi,t is equal to one for the 39 (9%) cases in our sample in which activist hedge funds publicly and explicitly criticize, condemn, or demand

changes in the target firm’s M&A activities and strategies, or question or oppose a pending acquisition proposal, and zero for their matching control firms; EVENTi,t

= GOVERNANCEi,t is equal to one for the 195 (45%) cases in our sample in which activist hedge funds explicitly state at least one governance-related objective in

the public campaign, and zero for their matching control firms; and EVENTi,t = BOARDi,t is equal to one for the 122 (28%) cases in our sample in which activist hedge

funds obtain at least one board seat during the post-intervention period, and zero for their matching control firms. For = PUBLICi,t or GOVERNANCEi,t,

POSTi,t is equal to one for the post-intervention period and zero otherwise. For EVENTi,t = BOARDi, POSTi,t is equal to one for the post-board-appointment period

and zero otherwise. As in Table 4, we show the results for four performance measures (i.e., CAR, ∆EPS, IROA, and BHAR). Columns 1 through 4 show the results

for EVENTi,t = PUBLICi,t. Columns 5 through 8 show the results for EVENTi,t = GOVERNANCEi,t. Columns 9 through 12 show the results for EVENTi,t = BOARDi,t.

Panel B shows the results of the following regression model that use the M&A deals of the treatment and control firms in the post-intervention period:

where EVENTi,t = JOINTi,t is equal to one for firms after they eliminate the CEO-Chairman join appointment during the post-intervention period and zero otherwise;

EVENTi,t = INDEPi,t is equal to one for firms after they increase the percentage of independent directors between the activism event year and the post-intervention

period and zero otherwise; and EVENTi,t = TURNOVERi,t is equal to one for firms after they forced CEO turnovers during the post-intervention period and zero

otherwise. DHFAi is equal to one for the treatment sample (firms with hedge fund activism), and zero for their matching control firms. Columns 1 through 4 show the

results for EVENTi,t = JOINTi,t. Columns 5 through 8 show the results for EVENTi,t = INDEPi,t. Columns 9 through 12 show the results for EVENTi,t = TURNOVERi,t.

For brevity, we report only estimates of , , and . Figures in parentheses are standard errors clustered at the match level. ***, **, and * indicates significance at the

1%, 5%, and 10% level, respectively.

Panel A: Relation between M&A performance and hedge fund activism

EVENT = PUBLIC

EVENT = GOVERNANCE

EVENT = BOARD

CAR

EPS

IROA

BHAR

CAR

EPS

IROA

BHAR

CAR

EPS

IROA

BHAR

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

PUBLIC POST

0.025*

0.308**

0.038**

0.345

(0.013)

(0.151)

(0.016)

(0.226)

PUBLIC

-0.017

-0.006

-0.012

-0.219

(0.011)

(0.117)

(0.010)

(0.227)

GOVERNANCE

0.027**

0.157*

0.025*

0.152

POST

(0.013)

(0.092)

(0.013)

(0.131)

GOVERNANCE

-0.026***

-0.101*

-0.012

-0.346***

(0.009)

(0.055)

(0.009)

(0.088)

52

BOARD POST

0.041**

0.312**

0.043**

0.302

(0.019)

(0.138)

(0.020)

(0.291)

BOARD

-0.003

-0.077

-0.007

-0.298***

(0.009)

(0.064)

(0.010)

(0.084)

POST

-0.012

-0.012

-0.035

-0.361***

-0.019*

-0.090

-0.004

-0.001

-0.013

0.001

-0.094

-0.094

(0.010)

(0.113)

(0.025)

(0.130)

(0.009)

(0.075)

(0.010)

(0.104)

(0.009)

(0.066)

(0.018)

(0.128)

Year FE

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Industry FE

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Sample size

222

195

147

138

906

832

456

528

525

477

339

304

Adjusted R2

0.049

0.213

0.023

0.147

0.075

0.121

0.067

0.327

0.076

0.042

0.152

0.290

Panel B: Relation between M&A performance, hedge fund activism, and board independence and forced CEO turnover

EVENT = JOINT

EVEN = INDEP

EVENT = TURNOVER

CAR

EPS

IROA

BHAR

CAR

EPS

IROA

BHAR

CAR

EPS

IROA

BHAR

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

DHFA JOINT

0.067*

0.552*

-0.024

0.385

(0.039)

(0.332)

(0.022)

(0.355)

JOINT

-0.052

-0.294

0.043**

0.251

(0.032)

(0.321)

(0.017)

(0.158)

DHFAINDEP

0.028*

0.192*

0.071**

-0.275