Journal of Corporate Finance 73 (2022) 102189

Contents lists available at ScienceDirect

Journal of Corporate Finance

journal homepage: www.elsevier.com/locate/jcorpfin

Corporate divestitures around acquisitions

Nihat Aktas a, *, Aleksandra Baros b, Ettore Croci c
a
WHU Otto Beisheim School of Management, Germany
b
University of Bologna, Italy
c
Università Cattolica del Sacro Cuore, Italy

A R T I C L E I N F O A B S T R A C T

Keywords: Divestitures often accompany acquisitions. Relying on a global sample, we provide support for the
Divestiture efficient restructuring view of acquisition-related divestitures. These divestitures add on average
Acquisition 2% to the total value creation of the acquisition process, translating into a value increase of $149
Asset sales
million. We test various efficiency channels potentially explaining this value contribution. The
Asset restructuring
Value creation
results indicate that the value contribution varies with the synergistic potential of the acquisition
process. We do not find empirical support for efficiency-improvement related to agency correc­
tion, financial constraint relaxation, and regulatory concern anticipation. Examining returns for
divestitures only, we find that those around acquisitions are not transactions with weak bargai­
ning positions.

1. Introduction

Acquisitions constantly redraw firm boundaries (Rhodes-Kropf and Robinson, 2008), generating opportunities to restructure by
selling old assets while buying new ones. Real world cases of acquisition-related divestitures abound. For example, Royal Dutch Shell
sold more than $27 billion in assets since its $54 billion acquisition of BC Group in 2015, Anheuser-Busch InBev sold assets to ease the
merger with SABMiller in 2015, and the Walt Disney Company agreed to sell 21 Fox Regional Sports Networks for almost $10 billion as
part of its $71.3 billion acquisition of Twenty-First Century Fox.
The examples mentioned above share a common trait: a focal deal, i.e. the acquisition, is accompanied by corporate divestitures.1
These acquisition-driven divestitures are instrumental for the completion and the success of the focal transaction (Capron et al., 2001).
Departing from prior literature that has emphasized agency-related correction and refocusing, we investigate the role of asset sales as a
tool to bolster the acquisition synergies.2 Therefore, this paper examines the value effect of asset restructuring plans centered on a focal
acquisition complemented by divestitures, henceforth denoted acquisition-centered restructuring process. Using a global sample, we
study the entire restructuring process and assess the contribution of corporate divestitures to its value creation.
From a theoretical perspective, in a neoclassical model with profit-maximizing firms (see, e.g., Maksimovic and Phillips, 2001,
2002), firms will either divest their assets when they have the opportunity to substitute them with better ones, improving the efficiency

* Corresponding author.
E-mail addresses: nihat.aktas@whu.edu (N. Aktas), aleksandra.baros@unibo.it (A. Baros), ettore.croci@unicatt.it (E. Croci).
1
Throughout the paper, we use the term focal to denote the acquisition around which we identify the divestiture(s).
2
For studies about agency-related correction and refocusing, see, e.g., Kaplan and Weisbach (1992), Comment and Jarrell (1995), John and Ofek
(1995), and Fluck and Lynch (1999). There is also a vast literature that has examined divestitures in isolation without associating them to ac­
quisitions (see, e.g., Lang et al., 1995; Mulherin and Boone, 2000; Schlingemann et al., 2002; and Bates, 2005).

https://doi.org/10.1016/j.jcorpfin.2022.102189
Received 2 June 2021; Received in revised form 26 February 2022; Accepted 20 March 2022
Available online 23 March 2022
0929-1199/© 2022 Elsevier B.V. All rights reserved.
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

of their capital allocation, or sell the parts of the business that command a price above their replacement cost in the market. In both
scenarios, divestitures around acquisitions are expected to contribute positively to the total value creation of the acquisition-centered
restructuring process, leading to an overall efficiency improvement. We identify four mutually non-exclusive efficiency channels that
potentially explain this positive value contribution. The first channel relates to better unlocking operating synergies in the focal deal.
According to this perspective, divestitures might be implemented both ex-ante to prepare the firm for the upcoming transaction (Arcot
et al., 2020) and ex-post to efficiently reconfigure the assets within the combined firm (Capron et al., 2001). By selling assets ex ante, the
bidder might devote more resource to the upcoming acquisition, which can ease synergy extraction. Ex post, divestitures allow the firm
to get rid of redundant resources, resulting in enhanced overall operating efficiency. In other words, divestitures permit to the bidder to
be in a more favorable position to take advantage of the synergies created with the focal deal. It does not necessarily imply that the
divestiture itself adds synergies to those that the bidder can achieve with the focal deal, merely that it facilitates their implementation.
The synergy channel, therefore, suggests that the acquirer is expected to become more efficient not only because of the focal deal but
also thanks to the divestitures around it.
Second, agency correction can also be a possible driver of asset sales. Refocusing divestitures often happen to correct past agency-
driven mergers, resulting in a reduction of agency and coordination costs (see, e.g., Kaplan and Weisbach, 1992; Comment and Jarrell,
1995; John and Ofek, 1995; Fluck and Lynch, 1999), and in an improvement of the allocation of resources (Maksimovic and Phillips,
2001, 2002). The third efficiency channel relates to financial constraint relaxation, as divestitures may ease the focal deal financially
(see, e.g., Arnold et al., 2018; Edmans and Mann, 2019; Mavis et al., 2020) or help the firm to reduce its financial risk ex-post
(Bongaerts and Schlingemann, 2020). Finally, regulatory concern anticipation is another potentially important efficiency channel. If
some divestitures are undertaken ex-ante in anticipation of regulatory costs associated with the focal acquisition (see, e.g., Aktas et al.,
2004; Fidrmuc et al., 2018), firms may avoid the cost associated with selling under urgency.
These four efficiency arguments lead to our main prediction: Under the efficient restructuring hypothesis, divestitures ease value
extraction from the focal acquisition, contributing positively to the total value creation of the acquisition-centered restructuring
process. A corollary to this hypothesis is that acquisition-related divestitures are neither fire sale transactions nor transactions with
weak bargaining power.
The efficient restructuring view is certainly not the only explanation for acquisition-related divestitures. Agency-based consider­
ations, such as pursuing managerial objectives with the sale proceeds, can also be a possible motive behind the decision to sell assets
(see, e.g., Lang et al., 1995). Unexpected difficulties in financing the focal deal and unanticipated regulatory requests may also lead to
inefficient divestitures, increasing the likelihood of a sale at a dislocated price (see, e.g., Shleifer and Vishny, 2011; de Bodt et al.,
2014). In these cases, divestitures have less value-creation potential for the seller. At best, we should observe, ceteris paribus, a non-
positive value effect.
Using a global sample, we provide new evidence about the acquisition-centered restructuring process and the associated value
effects by focusing on relatively large acquisitions of more than $50 million in value and representing at least 5% of the acquirer's
market capitalization. These transactions, which we label focal acquisitions, can trigger a reorganization of the firm's assets. We test
our hypotheses using a sample of 6845 focal acquisitions announced between 1996 and 2016, ensuring that there is no overlapping
acquisition that may contaminate the measurement of the value creation.3 We associate the focal acquisition with all divestitures
taking place between 1 year before the announcement and 1 year after the deal becomes effective.
About 13% of the acquisitions are associated with corporate divestitures (i.e., out of the 6845 focal acquisitions in our sample, 876
focal deals are associated with almost 1400 corporate divestitures). The amount of assets divested is substantial: divestitures represent
on average about 33% of the acquisition value. These statistics suggest that a non-negligible number of firms reorganize their assets
around acquisitions. Divestitures in the post-closing period are slightly more important than divestitures in the pre-announcement
period, and they are more frequent in large deals, acquisition of listed targets, cash deals, and cross-border deals.
To investigate the overall value effect, we adopt a three-step procedure: first, we measure the stock market reaction at the
announcement of the focal acquisition; second, we compute the abnormal returns associated with the related divestment an­
nouncements; finally, we add the abnormal returns at the acquisition announcement to those calculated for divestitures to obtain a
measure of the value creation associated with the overall acquisition-centered restructuring process. This approach allows us to
mitigate the concerns relative to the anticipation effect.4
Focal acquisitions create value for the acquirers, with an average 3-day announcement abnormal return (CAR) of 2.34%. Univariate
analysis shows that the average CAR of divesting acquirers is significantly lower than that of non-divesting acquirers (1.73% vs.
2.44%). However, this underperformance disappears when we account for the value creation of the whole acquisition-centered
restructuring process (3.03% vs. 2.44%). Once we control for deal and firm characteristics, we find evidence that divestiture activ­
ity enhances the total value creation of the focal deal by 2.00%, which translates into a dollar gain of $149 million for the average
sample firm implementing an acquisition-centered restructuring process. Divestitures substantially contribute to the total value
created by the acquisition-centered restructuring process, a result which is consistent with our efficient restructuring hypothesis.
To better understand the positive value contribution of divestitures around acquisitions, we test the four efficiency channels. The
first efficiency channel is related to divestitures helping unlock the focal deal's synergy potential. To this end, we rely on two contextual

3
We examine these more complex programs in Section 4.4 when we propose an alternative way to identify restructuring programs.
4
Given acquisitions and divestitures in our sample are interrelated, their announcement returns are potentially affected by an anticipation effect
(Cai et al., 2011; Wang, 2018). To account for the anticipation effect, we sum the abnormal returns of the acquisition and divestitures included in the
whole process.

2
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

factors that correlate with the cost reduction potentials of the focal deal, which are a key driver of merger synergies (see, e.g., Houston
et al., 2001; Devos et al., 2009). The first factor is the relative size of the focal deal, as the larger the relative size of the target the greater
the potential for realizing meaningful synergies (Devos et al., 2009). Moreover, Jansen et al. (2013) argue and document a positive
association between relative size and value creation in synergy-driven transactions (i.e., positive net present value deals). Therefore,
relatively large deals are more likely to trigger value-creating restructuring activities, such as asset sales, to exploit operating synergies
better. The second factor is the tightness of employment protection legislation at the country level, which hinders the acquiring firm's
ability to exploit merger synergies through workforce reductions fully (see, e.g., Alimov, 2015; John et al., 2015; Dessaint et al., 2017).
In particular, in context where downsizing through lay-offs is difficult to achieve (i.e. high employee protection countries), asset sales
might be considered as an alternative, as employees tied to the asset being sold are transferred to the buyer with the transaction.
However, given their low synergy potential in such a context, these assets are likely to command a low price.
We put these ideas to the test. Our results indicate that the positive value effect of acquisition-related divestitures increases with the
relative size of the focal deal, and the presence of tight employee protection laws attenuates their value contribution. Our estimates
imply that the value effect of divestitures increases by 1.07% (i.e., from 2.01% to 3.08%) when we raise the relative size threshold of
the focal deal from 5% to 33%. Moreover, the value contribution of acquisition-related divestitures decreases, depending on the
considered model, by 1.30% and 1.40% for one standard deviation increase in the Employee Protection Law (EPL) index.
The second efficiency channel is agency correction. Divestitures implemented in response to previous agency-driven mergers are
also expected to be associated with a positive value effect. However, the value contribution of acquisition-related divestitures in our
sample does not depend on the firm's past acquisitions (whether agency-driven or not), a result inconsistent with the agency correction
channel. The third efficiency channel is selling asset to reduce the firm's financial constraint. We consider three traits associated with
more severe financial constraints. The first one is whether the focal deal is entirely paid with cash, an important deal characteristic
likely to put pressure on the firm's financial situation. We also consider young and small firms as financially constrained firms (Hadlock
and Pierce, 2010; Denis and Sibilkov, 2010). No matter the proxy used, we find that the value contribution of divestitures does not
depend on the firm's financial constraint status in our sample. The last considered efficiency channel is selling assets in anticipation of
regulatory concern. Regulatory actions are indeed known to be costly for acquirers (see, e.g., Aktas et al., 2004; Fidrmuc et al., 2018).
We examine, therefore, whether the value contribution of divestitures is more pronounced when the focal deal is associated with high
regulatory risk (i.e., these are relatively large deals implemented by market leaders). This appears not to be the case in our sample, a
result inconsistent with the regulatory channel.
We conduct several tests to ensure that our findings are not due to the omitted variable bias. In particular, we add controls for a
variety of firm and industry characteristics that have been found in prior work to affect both divestitures and acquisitions. Our main
finding is also robust to the inclusion of acquirer fixed effects, which allows us to control for any time-invariant firm characteristics and
mitigate the omitted variable bias. Additional robustness checks include using alternative divestiture measures and abnormal return
methods and restricting the sample to US acquirers. To further assess the validity of our findings, we rely on the dormant period
approach (i.e., a period without any acquisition or divestiture) as an alternative method to identify the asset restructuring process (see,
e.g., Aktas et al., 2013). Our main findings are robust to these alterations.
To connect divestitures and acquisitions, we rely on the time dimension in our main analyses. While this empirical choice might
seem reasonable and easy to implement, there is a risk that some unrelated divestitures are tied to the considered focal deal. To
alleviate this issue and to better connect divestitures to acquisitions, in addition to the time dimension, we also consider the industry
and country affiliations of the divested assets. Our results indicate that the value contribution of divestitures around acquisitions is
robust to the different measures adopted.
Another concern is that divesting might be a suboptimal choice during an acquisition process. Comparing abnormal returns for
divestitures embedded in an acquisition-centered restructuring process and those that are not, we do not observe significant differ­
ences, indicating that the existence of a focal acquisition does not weaken the seller's bargaining power in this specific context.
It is also important to note that we do not claim a causal impact from divestitures to value creation in acquisitions. Our results do
not imply that if we pick a random acquirer and force it to divest around the focal deal, this will cause an increase in its value. Likely,
acquisitions and divestitures are jointly determined. In equilibrium, acquirers needing to disinvest to bolster acquisition-related
synergies do it, and the ones not needing to disinvest do not do it. Consistent with this equilibrium argument, we show similar
buy-and-hold abnormal returns (BHAR) for acquirers with and without divestitures from 1 year before the announcement of the focal
deal to 1 year after its completion.
Our paper offers several contributions to the literature. First, our article relates to studies examining divestitures that follow an
acquisition (see, e.g., Kaplan and Weisbach, 1992; Capron et al., 2001; Maksimovic et al., 2011). Kaplan and Weisbach (1992) and
Maksimovic et al. (2011) document that acquirers divest most of the target assets in the years following the acquisition. In an important
departure, we consider all divestitures around the focal deal, not limiting our analysis to the divestiture of target assets in the post-
closing period. Capron et al. (2001) also study divestitures as a mean to dynamically reconfigure the assets within the combined
firm to exploit synergies better. Still, they neither examine its value effects nor consider divestitures occurring before the completion of
the focal deal.
Second, considering the entire acquisition-related restructuring process allows a more precise assessment of the value effect. Other
papers have also analyzed acquisition-related divestitures, but they mostly focus on divestitures in the pre-announcement phase (see,
e.g., Arcot et al., 2020; Mavis et al., 2020). Arcot et al. (2020) document that firms undertake divestitures ex-ante to prepare the firm to
its upcoming acquisitions. Mavis et al. (2020) also examine divestitures in the pre-announcement phase and primarily focus on the
financing motive. In comparison to these two recent studies, we emphasize the synergy channel to explain the value contribution of
divestitures. Additionally, we do not limit our analysis to divestitures in the pre-announcement phase, and we account for the

3
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

anticipation effect by considering the total value creation of the acquisition-centered restructuring process. However, it is worth
emphasizing that our results do not question the financing motive behind asset sales, but what we show is that there is no cross-
sectional difference between financially constrained and unconstrained bidders in the contribution of these asset sales to the total
value creation.
Finally, related to the previous point, we also add to studies that have mainly examined acquisitions and divestitures in isolation
(see, e.g., Mulherin and Boone, 2000, Maksimovic and Phillips, 2001; Schlingemann et al., 2002; Bates, 2005). For example, Mak­
simovic and Phillips (2001) document an active market for corporate assets that generates efficiency gains and improves resource
allocation. We also confirm the contribution of asset sales to allocative efficiency, but in the context of an acquisition-centered
restructuring process.

2. Sample description and variable definitions

2.1. Sample description

Acquisition's data are from Thomson One Banker M&A database and cover the deals announced between 1996 and 2016.5 We
consider only acquisitions announced by publicly listed companies with a transaction value of at least $50 million in which the target is
not owned by the government, a joint venture, or a mutual. Acquirer and target firms should operate in neither the financial industry
(SIC code 6000–6999) nor the utility sector (SIC code 4900–4999). For the acquisition to be included in the sample, the acquirer must
own less than 20% before the deal and more than 90% after completion. Since we are interested in acquisitions triggering a reor­
ganization of the portfolio of assets of the acquiring firm, we require that the deal value be at least 5% of its pre-deal market
capitalization.
To measure the value of the whole acquisition-centered restructuring process, we link divestitures to a particular acquisition
controlling for other conflicting deal announcements. Therefore, we start by identifying all companies that made acquisitions during
our sample period. Since we define the acquisition-centered restructuring period as the period starting 1 year before the announcement
of the focal acquisition and ending 1 year after its completion date, we need to make sure that we drop all acquisitions that overlap.6
This implies that all acquisition-centered restructuring processes we study have exactly one focal acquisition in the period examined.
While this eliminates complex programs composed of a series of acquisitions in a short time, this approach provides two important
advantages. First, it allows us to perform a clean analysis of the whole acquisition-centered restructuring process, facilitating the
association between divestitures and focal acquisitions, and ensuring that no other deals may contaminate the value creation of the
considered focal deal. Second, since we have only one focal acquisition in the process, we can determine the process's beginning and
end. Finally, we apply some final filters. We retain only transactions for which: (i) the length of the period between deal announcement
and completion, i.e., the interim period, is <3 years; (ii) the acquirer has financial data available in Thomson Reuters' Worldscope
database and stock price data in Thomson Reuters' Datastream; (iii) the value of acquirer's assets at the end of the year before the
acquisition is not negative. Our final sample comprises 6845 focal acquisitions made by 5419 different acquirers from 60 different
countries.
For each focal acquisition in our sample, we search for the divestitures carried out from 1 year before the announcement date to 1
year after the completion date. We choose this period to strengthen the association between acquisition and divestitures and to reduce
the risk that divestiture decisions are unrelated to the acquisition one. Maksimovic et al. (2011) use a longer horizon (3 years) for sales
at plant level. However, they look at the decision to sell the assets bought with the acquisition. Since we want to examine the changes in
the asset portfolio triggered by the focal deal but not limited to the assets acquired with the transaction, a 3-year period carries a high
risk of including unrelated divestitures. Because of this, we opt for a more conservative approach and limit the period from 1 year
before the announcement to 1 year after the completion. Using Thomson One Banker M&A database, we consider divestiture deals that
are classified as acquisition of certain assets and acquisition of assets with a non-missing deal value. We consider only completed
divestitures with known transaction value. We require that the divesting firm has financial and price data available in Worldscope.
Overall, we identify 17,806 divestitures that satisfy these requirements, out of which 1397 divestiture events are related to 876 focal
acquisitions implemented by 780 unique firms. Table 1 presents the breakdown of the focal acquisitions and the related divestitures by
year, and Appendix C gives the sample distribution by country.
Panel A of Table 1 describes the global acquisition deals in our final sample. The sample period from 1996 to 2016 covers three
cycles. The number of transactions increases in the late 1990s, then declines in the early 2000s, before picking up again in 2005–2009,
and again towards the end of our sample period. The cyclicality of acquisition activity is also in line with prior literature (see, e.g.,
Harford, 2005; Betton et al., 2008; Maksimovic et al., 2013; Ahern and Harford, 2014). The average acquisition value in the sample is
$913 million, with the median deal amounting to $186 million. Even if smaller on average than acquisitions, the divestitures identified
are sizeable deals, with an average value of about $262 million (Table 1, Panel A). Through time, the divestiture activity and ac­
quisitions follow a similar trend.

5
We stop at 2016 in terms of announcement year because we need to be sure that the acquisition deal was completed ex post, and we need 1 year
after the completion to verify if there was a related divestiture or not.
6
For instance, if a specific company had three acquisitions, out of which the second was announced in the year after the first acquisition was
completed, whereas the third took place 3 years after the second acquisition was completed, we drop the first two acquisition altogether from the
sample, and only keep the third acquisition in the sample.

4
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

Table 1
Acquisition and divestiture activity by year.
Panel A. Acquisition and divestiture value by year

Year Acquisitions Divestitures

Mean Median N Mean Median N

1996 453.03 143.74 205 102.76 17.00 59

1997 438.85 136.63 249 78.55 29.00 77
1998 641.96 140.70 283 153.55 39.02 78
1999 1327.24 169.70 342 116.11 34.00 105
2000 1177.02 178.40 385 180.42 34.29 103
2001 885.69 163.54 243 284.59 30.96 78
2002 724.32 146.67 211 186.49 23.50 58
2003 450.83 141.96 238 80.23 35.52 42
2004 738.96 165.89 279 186.19 54.51 66
2005 928.55 186.00 325 157.55 25.00 93
2006 827.44 188.91 382 174.12 47.50 76
2007 848.66 177.00 475 374.48 35.23 107
2008 960.78 181.59 280 155.42 50.00 48
2009 1225.60 213.95 227 161.34 42.86 40
2010 793.45 210.00 343 261.11 35.25 48
2011 859.98 222.71 355 233.32 70.80 59
2012 718.76 200.78 364 237.92 70.00 36
2013 696.46 210.55 287 123.65 48.50 18
2014 1149.19 215.73 452 1220.80 138.13 75
2015 1296.18 241.20 503 464.95 160.00 72
2016 1188.97 250.00 417 322.21 157.90 59
Total 912.61 186.00 6845 262.38 40.00 1397

Panel B. Summary statistics

Mean Median Min Max N

Divestiture 12.80% 0.00% 0.00% 100.00% 6845

Acquisition Value – without divestiture 681 174 50.00 52,178 5969
Acquisition Value – with divestitures 2492 376 50.00 101,476 876
Divestiture Value 418 59 0.02 18,134 876
Divestiture Value/Acquisition value 32.83% 13.06% 0.00% 136.27% 876
Divestiture Pre 5.83% 0.00% 0.00% 100.00% 6845
Divestiture Interim 2.07% 0.00% 0.00% 100.00% 6845
Divestiture Post 7.54% 0.00% 0.00% 100.00% 6845
Divestiture Value – pre 397 38 0.03 18,134 399
Divestiture Value – interim 414 79 0.03 7426 142
Divestiture Value – post 290 59 0.02 7008 516
Div. Value/Acquisition value – pre 19.95% 8.75% 0.00% 57.97% 399
Div. Value/Acquisition value – interim 5.14% 7.90% 0.00% 7.90% 142
Div. Value/Acquisition value – post 23.40% 11.33% 0.02% 73.10% 516

Panel A reports the mean and median value (in $ million) of focal acquisitions and related divestitures by year. The focal acquisitions included in the
sample are announced over the 1996–2016 period. They have a relative size larger than 5% and are control transactions, with the acquirer owning less
than 20% ownership before the deal and at least 90% after. Panel B reports summary statistics on divestiture (i.e., a binary variable identifying if a
divestiture is associated with a focal deal), acquisition value with and without divestiture, and on the value of the divestitures in the acquisition
process (in $ million and in proportion relative to the value of the focal deal). Pre refers to divestitures taking place before the announcement of the
focal deal. Interim identifies divestitures implemented between the announcement date of the focal deal and its completion date, and Post divestitures
after the closing of the focal deal.

Panel B of Table 1 presents summary statistics on the acquisitions and divestitures in our sample. The average of the divestiture
dummy indicates that divestitures in our sample accompany 12.80% of the focal deals. Acquisitions with corporate divestitures are
larger on average than isolated acquisitions ($2492 million vs. $681 million). We also report summary statistics on the value of the
divestitures at the restructuring program level.7 The average divestiture value at the restructuring program level is $418 million.
Collectively, divestitures are, on average, not negligible and represent almost one-third of the acquisition value. These summary
statistics emphasize the importance of accounting for divestitures in the acquisition-centered restructuring process.
Concerning the timing of the divestitures in the restructuring process, 5.83% of the processes do have a least one divestiture in the
pre-announcement phase (i.e., the year before the announcement), 2.07% in the interim phase ((i.e., the period between

7
If there is more than one divestiture in a given restructuring program, we first sum their value before computing the corresponding summary
statistic.

5
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

announcement and closing), and 7.54% in the post-closing phase (i.e., the year after the deal is closed).8 These proportions indicate
that most of the divestitures take place in the year prior to acquisition announcement and in the post-closing period. The size of these
divestitures in dollar value and relative to the size of the focal deal are almost comparable. The proportion of focal deals with di­
vestitures in the interim period is smaller, indicating that companies rarely sell while closing a major transaction.

2.2. Variable definitions and summary statistics

Following the extant literature on acquisitions and divestitures, we employ a large set of firm, industry, deal, and country char­
acteristics to describe our sample, and as explanatory variables in our multivariate analyses. This subsection provides a succinct
description of the considered variables. We provide detailed variable definitions in Appendix A.
At the firm level, we control for financial performance, debt capacity (or financial flexibility), and investment with the following
variables: ROA, earnings before interest and depreciation divided by total assets; Leverage, total debt divided by total assets; Cash
holding, cash reserves divided by total assets; Dividend Payer, binary variable identifying firms that pay cash dividend; R&D, research
and development expenses divided by total assets; CAPEX, capital expenditures divided by total assets; Tobin's Q, sum of market value
of equity and total debt, divided by total assets; Diversified, binary variable identifying multi-segment firms; Serial Acquirer, binary
variable identifying firms that have implemented other acquisitions in the 3-year period before the announcement of the focal ac­
quisitions. Finally, following Gaspar and Massa (2005) and Peress (2010), we use Excess Price Margin as a proxy for firm's market
power. It represents the ability of the firm to price above marginal cost.
Among the industry-level variables, we consider the Herfindahl Index and the M&A Liquidity variables. We estimate industry
concentration with Herfindahl index as the sum of the squares of the market shares of all firms sharing the same three-digit SIC code, in
which market share is defined as sales of a firm divided by the sum of sales within the industry (Fidrmuc et al., 2018). Following
Schlingemann et al. (2002), we compute the liquidity of the M&A market in the industry of the acquiring firm in a given year as the
total deal value of acquisitions divided by the sum of total assets for each 2-digit SIC industry at the country level.
We also control for deal characteristics that are known to affect announcement abnormal returns (see, e.g., Betton et al., 2008, for a
review). The considered deals characteristics are: Relative Size, the acquisition value divided by the market capitalization of the
acquiring firm from the year prior to the deal announcement; Cross Industry, binary variable identifying cross-industry transactions;
Cross Border, binary variable identifying cross-border transactions; Public Target, indicator variable identifying transactions in which
the target is a listed company; and Stock (Cash), binary variable identifying fully stock (cash) paid transactions.
Finally, at the country level, we account for the general state of the economy with the GDP growth and GDP per capita variables. In
some of our analyses, to better account for the ease of financing at the country level, we rely on broad indexes such as stock market
development and banking development (see, e.g., Levine and Zervos, 1998; Demirgüc-Kunt et al., 2013; Hsu et al., 2014). Stock Market
Development corresponds to the aggregate stock market capitalization as a percentage of the corresponding country GDP, and Banking
Development is the domestic credit to private sector by banks as a percentage of the corresponding country GDP. To measure for the
tightness of labor rights and employment protection at the country level, we use Employment Protection Law (EPL), an index measuring
the strictness of regulations that an employer must follow to dismiss a worker (see, e.g., Dessaint et al., 2017; Ahmad and Lambert,
2019).
Table 2 presents the summary statistics for the considered firm-, industry-, deal-, and country-level variables,9 which indicate the
existence of a systematic difference between acquirers associated with divestitures and the ones without divestitures. Divesting
acquirers are on average larger (both in terms of total assets and market capitalization), more diversified, more leveraged, more
experienced with acquisitions, and more profitable (ROA). Moreover, they have lower growth opportunities (Tobin's Q), hold less cash,
pay more dividends, and invest more. In terms of deal characteristics, full stock-payment is relatively more common in acquisitions
without divestiture. Firms appear to divest relatively more around the focal deal when the target firm is publicly listed, foreign, and the
method of payment is fully in cash. Relative size is larger in acquisitions without divestitures, probably due to the significantly smaller
size of the acquirer in those transactions. The country characteristics indicate that acquirers with divestitures operate in slightly more
stock market-oriented economies than acquirers without divestitures. Finally, the considered proxy for employment protection is not
statistically different across the two subsamples.

3. Value creation and acquisition-centered restructuring process

This section assesses the efficiency of the acquisition-centered restructuring process. First, we explain the adopted approach to
measure the total value creation associated with the restructuring process and provide univariate comparisons. Then, in a multivariate
setting, we examine the determinants of the total value creation, emphasizing the value contribution of divestitures. Finally, we
explore the efficiency channels through which divestitures might contribute to the total value creation.

8
The sum of these proportions is higher than 12.80%, because a given process may have several divestitures occurring at various stages of the
restructuring process.
9
All continuous variables are winsorized at the 1st and 99th percentile.

6
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

Table 2
Summary statistics.
All acquisitions Acquisitions with divestitures Acquisitions without divestitures P-value

Mean Median N Mean Median N Mean Median N Mean Median

Firm characteristics
ROA 10.78% 12.33% 6609 12.61% 13.21% 849 10.51% 12.10% 5760 0.00 0.00
Leverage 21.07% 18.47% 6837 24.17% 23.02% 874 20.61% 17.62% 5963 0.00 0.00
Cash Holding 18.60% 11.41% 6836 13.21% 7.80% 874 19.39% 12.00% 5962 0.00 0.00
Dividend Payer 57.51% 100.00% 6693 67.94% 100.00% 867 55.96% 100.00% 5826 0.00 NA
Total Assets ($m) 2900 614 6842 7140 1726 875 2278 553 5967 0.00 0.00
Market Value ($m) 3103 735 6845 7427 1482 876 2468 675 5969 0.00 0.00
R&D 2.49% 0.00% 6842 2.28% 0.19% 875 2.52% 0.00% 5967 0.13 0.00
Capex 6.12% 4.03% 6792 6.59% 4.57% 871 6.05% 3.92% 5921 0.03 0.00
Tobin's Q 2.36 1.64 6838 1.98 1.54 875 2.42 1.66 5963 0.00 0.00
Diversified 66.09% 100.00% 6845 79.68% 100.00% 876 64.10% 100.00% 5969 0.00 NA
Serial Acquirer 30.40% 0.00% 6845 52.63% 100.00% 876 27.14% 0.00% 5969 0.00 0.00
Excess Price Margin − 0.11 0.00 6303 − 0.07 0.00 842 − 0.11 0.00 5461 0.06 0.48

Industry characteristics
Herfindhal Index 0.30 0.18 6647 0.32 0.20 872 0.30 0.18 5775 0.12 0.07
M&A Liquidity 0.07 0.02 6646 0.08 0.03 871 0.07 0.02 5775 0.20 0.02

Deal characteristics
Stock 19.49% 0.00% 6845 14.04% 0.00% 876 20.29% 0.00% 5969 0.00 0.00
Cash 29.38% 0.00% 6845 32.99% 0.00% 876 28.85% 0.00% 5969 0.01 0.01
Relative Size 0.96 0.30 6845 0.75 0.26 876 0.99 0.30 5969 0.00 0.03
Cross Border 30.85% 0.00% 6845 35.27% 0.00% 876 30.21% 0.00% 5969 0.00 0.00
Cross Industry 60.83% 100.00% 6845 59.59% 100.00% 876 61.02% 100.00% 5969 0.42 NA
Public target 25.61% 0.00% 6845 36.64% 0.00% 876 23.99% 0.00% 5969 0.00 0.00

Country characteristics
Stock Market Dev. 1.12 1.15 6705 1.19 1.25 864 1.11 1.14 5841 0.00 0.00
Banking Dev. 0.84 0.60 6397 0.80 0.57 818 0.84 0.60 5579 0.00 0.01
GDP Growth 3.16 2.86 6799 2.91 2.86 874 3.20 2.86 5925 0.00 0.67
GDP per Capita (ln) 10.39 10.68 6799 10.59 10.70 874 10.36 10.68 5925 0.00 0.32
EPL 0.84 0.09 5627 0.81 0.09 831 0.84 0.09 4796 0.32 0.99

Value creation
Acquisition CAR (− 1,1) 2.34% 1.07% 6845 1.73% 0.74% 876 2.44% 1.12% 5969 0.03 0.12
Divestiture CAR (− 1, 1) 1.07% 0.56% 876
Total CAR (− 1, 1) 2.52% 1.19% 6845 3.03% 1.84% 876 2.44% 1.12% 5969 0.14 0.03

The table presents summary statistics for our sample of focal deals announced over the 1996–2016 period. Variable definitions are in Appendix A. All
variables are winsorized at 1% on both tails. The last two columns report the P-values of the tests for the difference in means and medians. N is the
number of observations. NA denotes cases for which the difference in medians test is not available due to the empirical distribution of the variable.

3.1. Measuring value creation and univariate results

Under the considered efficient restructuring hypothesis, acquisition-centered divestitures are expected to add to the total value
creation of the restructuring process. We examine the abnormal returns associated with the acquisition-centered restructuring process
to test this intuition.
We start by computing the abnormal returns at the announcement of the focal acquisition using the classical market model.10 As
common in M&A literature, the considered event window is the interval (− 1, +1) centered on the announcement day of the focal deal.
Given the relation between acquisitions and divestitures, which are part of the considered restructuring process, an anticipation effect
may affect these abnormal returns (see, e.g., Cai et al., 2011; Wang, 2018). The abnormal returns associated with the divestitures
before the acquisition likely incorporate part of the value effect of the focal acquisition. Similarly, the market could anticipate the
abnormal returns of the post-closing divestitures at the time of the acquisition announcement. Since we are interested in the total value
creation of the acquisition-centered restructuring process, we sum the abnormal returns of the acquisition announcement and the ones
of the divestiture announcement(s) and denote the variable Total CAR.11 By including all the events in the acquisition-centered
restructuring process, our value creation measure mitigates the concerns relative to the anticipation effect.
We report the average and median abnormal returns associated with the focal deal, the associated divestitures, and the whole

10
Results using a market-adjusted model, which does not require an estimation window, are qualitatively similar and presented in Section 4.2.
11
Loderer and Martin (1990) also use the sum of bid announcement effects experienced by a firm in response to acquisitions during a given period.

7
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

acquisition-centered restructuring process in the first three rows of the last panel in Table 2. Focal acquisitions in our sample are on
average value-creating for acquiring firm shareholders, with an average 3-day abnormal return of 2.34%. These positive abnormal
returns are consistent with the existence of positive synergies associated with the focal transaction on average, and the acquirer keeps a
portion of these synergies for its own shareholders. We also observe that the value creation at acquisition announcement differs be­
tween divesting and non-divesting acquirers, being significantly lower for the former ones (1.73% vs 2.44%). The anticipation effect
can partially explain this difference due to the divestitures taking place before the announcement of the focal deal. Thus, to estimate
the value creation of the whole process for divesting acquirers, we also need to consider the value creation of all related divestitures.
We find that the value creation associated with divestitures is on average around 1.07%. When we add the abnormal returns of
divestiture events to those of the acquisition announcement to compute the Total CAR measure, the underperformance for divesting
acquirers disappears. Indeed, the average total value creation is 3.03% for divesting acquirers.12 The univariate analysis shows that the
considered acquisition-centered restructuring processes are on average value-enhancing for shareholders, and therefore, efficiency-
driven to a large extent.

3.2. Multivariate results

To examine whether divestitures contribute to the total value creation of the acquisition-centered restructuring process, in Table 3,
we estimate OLS regressions controlling for firm, deal and industry characteristics that are known to affect announcement returns. Our
specifications also include country and industry fixed effects to account for time-invariant country and industry unobservable factors,
and year dummies to control for changing economic and financing conditions through time. Columns 2 and 4 also consider acquirer
fixed effects to mitigate further omitted variable biases.
The first two columns report on the acquirer 3-day abnormal returns around the acquisition announcement as the dependent
variable. The coefficient estimate of Divestiture, which identifies whether the focal deal is associated with divestitures, is positive in
both columns, but only statistically significant in the model without firm fixed effects. Concerning the control variables in Column 1,
the value effects are lower for firms with higher total assets and valuation and acquiring other public companies, in line with prior
literature (see, e.g., Moeller et al., 2004). The negative coefficient for Tobin's Q supports the view that acquirers signal their over­
valuation to the market (Dong et al., 2006). Firms more involved in R&D have lower abnormal returns. Firms that are diversified and
with high leverage are associated with higher stock price reaction. Relative size has a positive coefficient estimate, which is consistent
with the existence of high synergy potential in relatively large value-creating deals (see, e.g., Jansen et al., 2013). Differently from
Harford and Uysal (2014), the coefficient estimate of M&A liquidity is positive in a worldwide context, indicating that acquisitions in
industries with more M&A activity have higher abnormal returns. However, in the model with firm fixed effects (Column 2), only the
coefficient estimates of Size and Public Target retain their significance.
In the last two columns of Table 3, we study the determinants of the total value creation associated with the acquisition-centered
restructuring process. The dependent variable is Total CAR, the sum of the abnormal return for the acquisition and those of the eventual
divestitures. Controlling for firm, deal, and industry characteristics, we find that divestitures enhance the value creation of the
restructuring process. In both models, the coefficient estimate of Divestiture is positive and significant, indicating that divestitures
enhance the total value creation of the focal deal by circa 2.00%, which translates into a dollar gain of $149 million for the average firm
implementing an acquisition-centered restructuring process.13 Concerning the remaining explanatory variables, results in the total
CAR models are remarkably similar to those in the acquisition CAR models.
Taken collectively, our results show that firms exploit the acquisition event to restructure their asset portfolio, and this restruc­
turing process is efficiency-driven. To provide further evidence consistent with the efficient restructuring hypothesis, we next explore
four efficiency channels potentially explaining the value contribution of acquisition-related divestitures. These are unlocking synergy
potential, agency correction, financing constraint relaxation, and regulatory concern anticipation.

3.3. Unlocking synergy potential and total value creation

We consider two contextual factors that correlate with the synergistic potential of the restructuring process. The first factor is the
relative size of the focal deal, and the second one is the tightness of employment protection legislation at the country level.
Relative size is an important determinant of acquirer announcement returns (see, e.g., Asquith et al., 1983; Moeller et al., 2004). In
synergy-driven transactions, relatively large deals result in larger positive abnormal returns for acquirers (Jansen et al., 2013).
Relatively large deals are also likely to trigger more asset restructuring activities to facilitate the focal deal's completion and better
exploit operating synergies. The focal deals in our sample are value-creating on average, as documented in Table 2, and the coefficient
estimate of relative size is positive in Table 3, consistent with the existence of more synergy potentials in relatively large transactions.
In Panel A of Table 4, we assess whether the positive value effect of corporate divestitures increases with the relative size of the
focal deal. To this end, we rely on subsamples with increasing relative size threshold for the focal deal to be included in the corre­
sponding sample. Our baseline results reported in Table 3 rely on a sample of deals with a minimum relative size of 5%. The first and

12
The sum of acquisition CAR (1.73%) and divestiture CAR (1.07%) does not equal Total CAR (3.03%) because the variables are winsorized to
limit the influence of outliers. Without winsorization, the acquisition CAR is 1.89%, the divestiture CAR 1.34%, and total CAR 3.23% (=1.89% +
1.34%).
13
In unreported analysis, we obtain a similar value effect (2.09%) when we rely on 5–day CARs instead of 3–day CARs.

8
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

Table 3
Value contribution of divestitures in acquisition-centered restructuring programs.
Acquisition CAR Total CAR

(1) (2) (3) (4)

Divestiture 0.0062* 0.0067 0.0201*** 0.0203**

[0.0033] [0.0082] [0.0042] [0.0096]
ROA 0.0034 0.0400 0.0095 0.0524
[0.0141] [0.0494] [0.0149] [0.0513]
Leverage 0.0190** 0.0070 0.0198** 0.0087
[0.0081] [0.0330] [0.0084] [0.0349]
Cash Holding 0.0012 0.0055 0.0014 0.0045
[0.0107] [0.0454] [0.0111] [0.0492]
Dividend Payer − 0.0017 0.0102 − 0.0028 0.01
[0.0030] [0.0124] [0.0032] [0.0131]
Size − 0.0081*** − 0.0209*** − 0.0082*** − 0.0207***
[0.0011] [0.0070] [0.0011] [0.0075]
Excess Price Margin 0.0032 − 0.0112 0.0017 − 0.0177
[0.0038] [0.0159] [0.0040] [0.0166]
Herfindhal Index 0.0053 − 0.0054 0.0071 − 0.0061
[0.0059] [0.0358] [0.0061] [0.0370]
R&D − 0.0821** − 0.1328 − 0.0651 − 0.0465
[0.0410] [0.1724] [0.0423] [0.1750]
Capex − 0.0141 − 0.0152 − 0.0056 − 0.0394
[0.0235] [0.1122] [0.0246] [0.1149]
Tobin's Q − 0.0023*** − 0.0046 − 0.0024** − 0.0063*
[0.0009] [0.0031] [0.0010] [0.0032]
Diversified 0.0076*** − 0.0044 0.0087*** − 0.0018
[0.0029] [0.0105] [0.0030] [0.0109]
M&A Liquidity 0.0279*** − 0.0055 0.0324*** 0.0105
[0.0103] [0.0328] [0.0111] [0.0362]
Stock 0.0028 0.013 0.0023 0.0105
[0.0042] [0.0122] [0.0044] [0.0126]
Relative Size 0.0040*** 0.0056 0.0040*** 0.0056
[0.0011] [0.0041] [0.0012] [0.0042]
Cross Border 0.0035 0.0064 0.0034 0.0091
[0.0029] [0.0083] [0.0030] [0.0087]
Cross Industry − 0.0012 − 0.0051 − 0.0021 − 0.0048
[0.0025] [0.0077] [0.0026] [0.0083]
Public Target − 0.0274*** − 0.0332*** − 0.0282*** − 0.0322***
[0.0030] [0.0081] [0.0032] [0.0086]
Serial Acquirer 0.0022 0.0008
[0.0028] [0.0029]
Acquirer FE No Yes No Yes
Country FE Yes No Yes No
Industry FE Yes No Yes No
Year FE Yes Yes Yes Yes
Adjusted R2 0.082 0.116 0.077 0.110
Observations 6011 2374 6011 2374

The table presents the coefficient estimates of OLS regressions where the dependent variable is Acquisition CAR (i.e., the 3-day CAR around the
acquisition announcement) in the first two columns, and Total CAR (i.e., acquisition announcement 3-day CAR plus divestiture announcement 3-day
CARs) in the last two columns. The specifications in Columns 1 and 3 control for time, industry, and country-level fixed effects (FE). In Columns 2 and
4, we control for acquirer, and time FE. Variable definitions are in Appendix A. All variables are winsorized at 1% on both tails. Standard errors are
clustered at firm level and reported within brackets. ***, **, and * indicate significance at the 1%, 5%, and 10%, respectively.

second columns in Panel A of Table 4 report the total CAR regressions on the subsamples of focal deals with a relative size higher than
10% and 33%, respectively. The regression models include the same set of control variables as in Column 3 of Table 3, whose coef­
ficient estimates are suppressed for brevity. Regardless of the relative size threshold, the results indicate that divestitures positively
contribute to the total value creation of the acquisition-centered restructuring process. The value contribution clearly increases with
the relative size of the deals.
Our estimates imply that when we increase the relative size threshold from 10% to 33%, the value contribution of divestitures
increases by 82 basis points (i.e., the coefficient estimate of the divestiture dummy increases from 2.26% to 3.08%). The increase is
even more pronounced if we consider our initial sample with a relative size threshold of 5% (i.e., the coefficient estimate of the
divestiture dummy increases by 107 basis points from 2.01% to 3.08%).
Next, exploiting the cross-country dimension of our sample, we examine the tightness of employment protection legislation at the
country level as a factor correlated with synergy potential. Dessaint et al. (2017) document that potential synergies are lower in a high
employee protection environment, because labor market rigidity limits the implementation of cost synergies through workforce
reduction. Divestitures might be considered as an alternative to workforce reduction. However, given their low synergy potential in

9
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

Table 4
Value contribution of divestitures and synergistic potential of the focal deal.
Panel A. Relative size of the focal deal as a proxy for synergistic potential

Relative size >10% Relative size >33%

(1) (2)

Divestiture 0.0226*** 0.0308***

[0.0050] [0.0075]
Controls Yes Yes
Country FE Yes Yes
Industry FE Yes Yes
Year FE Yes Yes
Adjusted R2 0.081 0.092
Observations 4929 2768

Panel B. EPL index as an inverse proxy for synergistic potential

(1) (2)

Divestiture 0.0297*** 0.0300***

[0.0059] [0.0059]
EPL 0.0053*
[0.0028]
EPL × Divestiture − 0.0153*** − 0.0140***
[0.0047] [0.0048]
GDP Growth − 0.0013 0.0001
[0.0013] [0.0017]
GDP per Capita 0.0122** − 0.1358**
[0.0062] [0.0633]
Stock Market Development − 0.0010 0.0038
[0.0060] [0.0115]
Banking Development − 0.0051 0.0190*
[0.0047] [0.0104]
Controls Yes Yes
Country FE No Yes
Industry FE Yes Yes
Year FE Yes Yes
Adjusted R2 0.064 0.062
Observations 4614 4614

The table presents the coefficient estimates of OLS regressions where the dependent variable is Total CAR (i.e.,
acquisition announcement 3-day CAR plus divestiture announcement 3-day CARs). Panel A relies on the relative size
of the focal deal as a proxy for synergistic potential. Columns 1 and 2 present the results for the subsample of focal
deals with relative size larger than 10% and 33%, respectively. Panel B relies on Employee Protection Law (EPL) index
at the country level as an inverse proxy for synergistic potential of the focal deal. Each model includes the same set of
controls as in Column 3 of Table 3, whose coefficients are suppressed for brevity. Variable definitions are in
Appendix A. All variables are winsorized at 1% on both tails. Standard errors are clustered at firm level and reported
within brackets. ***, **, and * indicate significance at the 1%, 5%, and 10%, respectively.

such an environment, these assets are likely to be sold at a low price. Therefore, we expect labor market rigidity to attenuate the value
contribution of acquisition-related divestitures. We rely on the Employee Protection Law (EPL) index of the OECD as a proxy for labor
market rigidity at the country level. Specifically, we employ the summary indicator for individual dismissals of regular workers.
Panel B of Table 4 reports the estimation results of two specifications with Total CAR as the dependent variable. We augment the
baseline specification (Column 3 of Table 3) with EPL and its interaction with Divestiture, and with the following country-level vari­
ables: GDP Growth, GDP per Capita, Stock Market Development, and Banking Development. We suppress the coefficient estimates of the
control variables from the baseline model for brevity. Columns 1 and 2 report on the specification without and with country fixed
effects, respectively. In the latter case, the individual EPL term is omitted, because it shows little within-country variation.14
The coefficient estimate of the interaction term between EPL and Divestiture is negative and statistically significant in both models.
This result indicates that, as expected, tight employment protection at the country level attenuates the positive contribution of di­
vestitures to the total value creation of the restructuring process. In terms of economic impact, a one standard deviation increase in the
EPL index reduces the value contribution of acquisition-related divestitures between 1.30% and 1.40%, depending on the model.15
These results suggest a sizeable economic impact, given that the average of the total value creation associated with the acquisition-

14
We estimate the specification in Column 2 also including the EPL index. As expected, its inclusion does not alter the coefficients of the variables
of interests. Results are available from the authors upon request.
15
The standard deviation of the EPL index is 0.92 in our sample.

10
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

centered restructuring process is 3.03% in our sample.

To account for the timing of divestitures around the focal deal, we next replace the divestiture variable, in the specifications re­
ported in Table 4, with three binary variables identifying whether the divestitures take place in the pre-announcement, interim, or
post-closing periods. Panel A of Table 5 reports the estimation results of three different samples constructed using the 5%, 10%, and
33% relative size thresholds, respectively. The coefficient estimates of all three divestiture variables are positive. However, only the
two binary variables identifying divestitures in the pre-announcement and post-closing periods are statistically significant across the
three samples. The value contribution of divestitures in the pre-announcement and post-closing periods also increases with the relative
size threshold. For example, divestitures in the pre-announcement (post-closing) period are associated with an increase of 2.57%
(2.43%) in Total CAR in focal deals with a relative size larger than 33%. Compared with the results of the sample with a relative size
larger than 5% (Column 1), the value contribution increases by circa 1% both for pre-announcement and post-closing divestitures.
Panel B of Table 5 considers the timing of the divestitures in interaction with the EPL index. We find that divestitures both in the
pre-announcement and post-closing periods are associated with a lower contribution to the total value creation in high EPL envi­
ronment. Collectively, these results further corroborate the view that divestitures are part of a value-increasing asset restructuring
process intending to unlock potential synergies from the focal deal.

3.4. Agency correction, financial constraint relaxation, and regulatory concern anticipation

The next efficiency channel that we test is agency correction. Prior literature abounds with evidence about agency-related
correction and refocusing following value-destroying mergers (see, e.g., Kaplan and Weisbach, 1992; Comment and Jarrell, 1995;
John and Ofek, 1995; Fluck and Lynch, 1999). Divestitures motivated by the correction of previous agency-driven mergers are likely
associated with a positive value effect. To test for the existence of the agency correction channel in our sample, we consider the
historical acquisitions undertaken by our sample firms. We use a binary variable, Past Acquisition, to identify focal deals in which the
acquirer has made other (or value-destroying) acquisitions in the previous 3 years. We then augment our baseline specification with
Past Acquisition and its interaction with Divestiture and report the estimation results in Panel A of Table 6.
In Column 1, Past Acquisition takes the value of 1 if acquisitions precede the focal deal in the previous 3 years, and 0 otherwise.16 In
Column 2, to account for the existence of agency-driven acquisitions, Past Acquisition is a binary variable identifying whether at least
one value-destroying acquisition (i.e., negative acquirer CAR) precedes the focal deal in the previous 3 years, and 0 otherwise. The
coefficient estimate of Divestiture is positive and statistically significant at the 1% level in both models, indicating that divestitures
contribute positively to the total value creation, even in the absence of previous acquisitions. The impact of Past Acquisition on the total
value creation is negative, but only marginally significant in Column 2. The total value creation is associated with a decrease of 0.61%
when the firm has implemented at least a value-destroying deal in the previous 3 years. Most importantly, the interaction term between
Past Acquisition and Divestiture is negative but statistically insignificant in both specifications, indicating that the value contribution of
divestitures to the total value creation of the restructuring process in our sample does not depend on the correction of mistakes from the
firm's past acquisitions. These results fail to support the agency correction channel.
We consider another important efficiency channel: the firm's financial constraint relaxation thanks to the divestiture proceeds. The
financing motive is an important driver of the disinvestment decision before acquisitions emphasized in prior literature (see, e.g.,
Arnold et al., 2018; Edmans and Mann, 2019; Mavis et al., 2020). Divestitures following acquisitions might also be triggered by the
firm's willingness to alleviate its financial constraints (Bongaerts and Schlingemann, 2020). If the considered divestitures allow the
acquirer to mitigate its financial constraint, then these divestitures are also expected to enhance the total value creation associated
with the restructuring process.
We consider three characteristics related to the firm's financial constraint status. The first is whether the focal deal is entirely paid
with cash, as the financial constraint argument is more relevant for cash-financed transactions. We also consider age and size as two
additional proxies for financial constraints, with small and young firms being relatively more financially constrained (Hadlock and
Pierce, 2010; Denis and Sibilkov, 2010). In Panel B of Table 6, we add to the baseline specification the considered financial constraint
proxy and its interaction with Divestiture. We expect a positive interaction term under the financing motive as an efficiency improving
channel. Each column is devoted to a specific financial constraint proxy. None of the estimated interaction terms are positive and
statistically significant at conventional levels. The interaction term in Column 1 is negative and statistically significant, indicating that
divestitures around cash acquisitions contribute less to the total value creation. Columns 2 and 3 show that the value contribution of
divestitures implemented by small and young firms does not differ from our sample's remaining firms. These results indicate that the
value contribution of acquisition-related divestitures is not due to the relaxation of the firm's financial constraint.
The last considered efficiency channel is selling assets in anticipation of regulatory concerns. Regulatory actions are known to be
costly for firms as they may delay the closing of the focal deal and/or be associated with forced disinvestment for the focal deal to go
through (see, e.g., Aktas et al., 2004; Fidrmuc et al., 2018). In anticipation of regulatory challenges, divestitures may help the firm
extract more value in these transactions as a seller. This will avoid negotiating under urgency and selling the divested asset at a
discount. To test this intuition, we first identify focal deals likely to be associated with potentially high regulatory risk and then
examine whether the value contribution of these divestitures is relatively larger or not.
Large transactions from market leaders are likely to attract relatively more scrutiny from regulators. Therefore, we augment our

16
We have also used a 5-year window prior to the focal deal to identify historical acquisitions. The results are similar to the 3-year window
(unreported).

11
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

Table 5
Timing of divestitures around the focal deal.
Panel A. Replication of Tables 3 and 4 – Panel A

Relative size >5% Relative size >10% Relative size >33%

(1) (2) (3)

Divestiture Pre 0.0171*** 0.0180** 0.0257**

[0.0061] [0.0074] [0.0120]
Divestiture Interim 0.0116 0.0160 0.0173
[0.0091] [0.0109] [0.0169]
Divestiture Post 0.0142*** 0.0160*** 0.0228**
[0.0052] [0.0062] [0.0090]
Controls Yes Yes Yes
Country FE Yes Yes Yes
Industry FE Yes Yes Yes
Year FE Yes Yes Yes
Adjusted R2 0.133 0.144 0.176
Observations 6011 4929 2768

Panel B. Replication of Table 4 – Panel B

(1) (2)

Divestiture Pre 0.0272*** 0.0274***

[0.0087] [0.0087]
Divestiture Interim 0.0097 0.0096
[0.0124] [0.0124]
Divestiture Post 0.0240*** 0.0245***
[0.0073] [0.0073]
EPL 0.0049*
[0.0027]
EPL × Divestiture Pre − 0.0166** − 0.0156**
[0.0071] [0.0071]
EPL × Divestiture Interim 0.0062 0.0068
[0.0118] [0.0118]
EPL × Divestiture Post − 0.0136** − 0.0128**
[0.0059] [0.0060]
Controls Yes Yes
Country FE No Yes
Industry FE Yes Yes
Year Yes Yes
Adjusted R2 0.107 0.104
Observations 4614 4614

Replication of the baseline results in Tables 3 and 4 with the independent variable of interest, Divestiture, by three binary variables identifying di­
vestitures in the pre, interim, and post periods. In all models, the dependent variable is Total CAR (i.e., acquisition announcement 3-day CAR plus
divestiture announcement 3-day CARs). Column 1 of Panel A replicates Column 3 of Table 3, while Columns 2 and 3 replicate the two models in Panel
A of Table 4. Panel B replicates the two models in Panel B of Table 4. Each model includes the same set of controls and fixed effects (FE) as in the
original specification, whose coefficients are suppressed for brevity. Variable definitions are in Appendix A. All variables are winsorized at 1% on both
tails. Standard errors are clustered at firm level and reported within brackets. ***, **, and * indicate significance at the 1%, 5%, and 10%,
respectively.

baseline specification with High Regulatory Risk, a binary variable identifying large deals implemented by market leaders, and its
interaction term with Divesture. We define market leaders based on the firm's market share in the corresponding country and industry.
Firms in the top tercile are considered as market leaders. We use two relative size thresholds to identify large focal deals, respectively
10% and 20%. Panel C of Table 6 reports the result. In both columns, the coefficient estimate of the interaction term is statistically
insignificant. The anticipation of regulatory concerns does not seem to explain the value contribution of divestitures in our sample.

4. Additional results and robustness checks

This section reports additional analyses and robustness checks. We first consider alternative divestiture definitions. We then assess
the robustness of our results to controlling for external sources of financing (such as equity and debt issues), restricting the sample to
US acquirers only, adopting a different return generating process, and a change in the considered employee protection index. Next, we
examine the impact of acquisition-related divestitures on buy-and-hold abnormal returns. A further test uses the dormant period
approach as an alternative method to identify asset restructuring programs. Finally, we compare the wealth effect of acquisition-
related divestitures and those implemented in isolation.

12
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

Table 6
Agency corrections, financial constraints, and anticipation of regulatory concerns.
Panel A. Agency correction

All acquisitions Value destroying acquisitions

(1) (2)

Divestiture 0.0250*** 0.0243***

[0.0059] [0.0051]
Past Acquisition − 0.0036 − 0.0061*
[0.0028] [0.0031]
Past Acquisition × Divestiture − 0.0099 − 0.0132
[0.0078] [0.0081]
Controls Yes Yes
Country, Industry, and Year FE Yes Yes
Adjusted R2 0.077 0.078
Observations 6011 6011

Panel B. Financial constraint relaxation

Cash Payment Small Firm Young Firm

(1) (2) (3)

Divestiture 0.0246*** 0.0196*** 0.0217***

[0.0054] [0.0039] [0.0042]
Financial Constraint 0.0084*** − 0.0003 0.0018
[0.0026] [0.0041] [0.0032]
Financial Constraint × Divestiture − 0.0128* 0.0029 − 0.0076
[0.0075] [0.0139] [0.0120]
Controls Yes Yes Yes
Country, Industry, and Year FE Yes Yes Yes
Adjusted R2 0.078 0.077 0.077
Observations 6011 6011 6011

Panel C. Regulatory concern anticipation

Relative size >10% Relative size >20%

(1) (2)

Divestiture 0.0236*** 0.0200***

[0.0056] [0.0047]
High Regulatory Risk 0.0015 0.0034
[0.0027] [0.0030]
High Regulatory Risk × Divestiture − 0.0067 0.0003
[0.0078] [0.0085]
Controls Yes Yes
Country, Industry, and Year FE Yes Yes
Adjusted R2 0.077 0.077
Observations 6011 6011

The table presents the coefficient estimates of OLS regressions where the dependent variable is Total CAR (i.e., acquisition
announcement 3-day CAR plus divestiture announcement 3-day CARs). Panel A considers agency correction as a potential
factor explaining the value contribution of divestitures. In Column 1, Past Acquisition takes the value of 1 if the focal acquisition
is preceded by acquisitions in the previous 3 years, and 0 otherwise. In Column 2, Past Acquisition takes the value of 1 if the focal
deal is preceded by at least one value destroying acquisition (i.e., negative acquirer CAR) in the previous 3 years, and
0 otherwise. Panel B considers financial constraint relaxation as a potential factor explaining the value contribution of di­
vestitures. We rely on three different binary variables to distinguish financially constrained firms. The column heading
identifies the considered proxy. In Column 1, the financial constraint proxy is Cash Payment, a binary variable identifying focal
deals that are fully paid in cash. In Column 2, the financial constraint proxy is Small Firm, a binary variable identifying firms
that are in the bottom tercile based on total assets. In Column 3, the financial constraint proxy is Young Firm, a binary variable
identifying firms that are in the bottom tercile based on firm age. Panel C considers the regulatory concern anticipation as a
potential factor explaining the value contribution of divestitures. High Regulatory Risk is a binary variable identifying deals
more likely to generate regulatory concerns (i.e., relatively large deals undertaken by market leaders). We define market
leaders based on the firm's market share in the corresponding country and industry. Firms in the top tercile are considered as
market leaders. We use two relative size thresholds to identify large focal deals, respectively 10% and 20% in Columns 1 and 2.
Each model includes the same set of controls and fixed effects (FE) as in Column 3 of Table 3, whose coefficients are suppressed
for brevity. Variable definitions are in Appendix A. All variables are winsorized at 1% on both tails. Standard errors are
clustered at firm level and reported within brackets. ***, **, and * indicate significance at the 1%, 5%, and 10%, respectively.

13
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

4.1. Alternative divestiture definitions

Our main analysis employs binary variables to identify whether acquirers carry out divestitures during the period going from 1 year
before the acquisition announcement to 1 year after its completion. One drawback of these variables is that they do not account for the
importance of the divestiture in terms of size relative to the acquisition. While Table 1 shows that on average divestitures are
important, there is still variation in the size of the assets sold around acquisitions. To alleviate the concern that our results are driven by
divestitures of negligible value, Panel A of Table 7 replicates our main analysis with the following alterations. In Column 1, we replace
the divestiture variable in the baseline specification with Divestiture Intensity, defined as the ratio between the dollar value of the
divestitures and the value of the acquisition. In Columns 2 and 3, we ignore small divestitures, and identify divestitures with a relative
size of at least 1% and 10% relative to the focal acquisition, respectively. The estimates shown in Panel A corroborate our previous
results, showing that divestitures positively contribute to the value creation of the acquisition-centered restructuring process.
To better connect divestitures to acquisitions, in addition to the time dimension, we also consider the industry and country af­
filiations of the divested assets. Panel B of Table 7 reports these tests. The heading of the columns identifies the considered industry/
country affiliation of the divested assets. In Columns 1 and 2, Divestiture identifies focal deals associated with divestitures from the
same industry as the acquirer and the target, respectively. In Columns 3 and 4, Divestiture identifies focal deals associated with di­
vestitures from the same country as the acquirer and the target, respectively. In all the regressions, the sign and magnitude of the
coefficient estimate of Divestiture is comparable to our baseline estimate.

4.2. Further robustness checks

Debt and equity issues are alternative sources of financing a firm can use to fund its future investments (Hovakimian and Titman,
2006). While we already control for the financial characteristics of the firm (such as leverage and cash holding) in Table 3, and also for
the ease of financing at the country level in Panel B of Table 4, we repeat the analyses including variables that control for equity and
debt issues to better account for alternative modes of financing. For each focal acquisition, we record equity and debt (both bond and

Table 7
Alternative divestiture measures.
Panel A. Relative size of the divested assets

Divestiture intensity Divestiture relative size >1% Divestiture relative size > 10%

(1) (2) (3)

Divestiture 0.0347*** 0.0224*** 0.0202***

[0.0080] [0.0044] [0.0052]
Controls Yes Yes Yes
Country FE Yes Yes Yes
Industry FE Yes Yes Yes
Year FE Yes Yes Yes
Adjusted R2 0.077 0.078 0.075
Observations 6011 6011 6011

Panel B. Industry and country affiliations of the divested assets

Acquirer industry Target industry Acquirer country Target country

(1) (2) (3) (4)

Divestiture 0.0211*** 0.0218*** 0.0199*** 0.0228***

[0.0050] [0.0054] [0.0041] [0.0051]
Controls Yes Yes Yes Yes
Country FE Yes No Yes Yes
Industry FE Yes Yes Yes Yes
Year FE Yes Yes Yes Yes
Adjusted R2 0.076 0.078 0.076 0.079
Observations 6011 2661 6011 6011

This table provides various additional analyses adopting alternative divestiture measures. In all models, the dependent variable is Total CAR (i.e.,
acquisition announcement 3-day CAR plus divestiture announcement 3-day CARs). Panel A accounts for the intensity of the divestiture relative to the
focal deal. In Column 1, we replace the divestiture binary variable in the baseline specification with Divestiture Intensity, defined as the ratio between
the dollar value of the divestitures and the value of the acquisition. In Column 2, Divestiture is a binary variable identifying focal deals with di­
vestitures whose total value is at least 1% of the focal deal value. In Column 3, Divestiture is a binary variable identifying focal deals with divestitures
whose total value is at least 10% of the focal deal value. Panel B accounts for the industry and country affiliations of the divested assets to better
connect divestitures to focal acquisitions. In Columns 1 and 2, Divestiture is a binary variable identifying focal deals whose divestitures are from the
same industry as the acquirer and the target, respectively. In Columns 3 and 4, Divestiture is a binary variable identifying focal deals whose divestitures
are from the same country as the acquirer and the target, respectively. Each model includes the same set of controls and fixed effects (FE) as in Column
3 of Table 3, whose coefficients are suppressed for brevity. Variable definitions are in Appendix A. All variables are winsorized at 1% on both tails.
Standard errors are clustered at firm level and reported within brackets. ***, **, and * indicate significance at the 1%, 5%, and 10%, respectively.

14
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

loan) issues of the acquiring firms that have occurred from 1 year before the acquisition announcement to 1 year following its
completion. Appendix B reports the descriptive statistics of these issues. In Column 1 of Table 8, we estimate the Total CAR regression
after including the dummies for external financing into the specification. The magnitude of the coefficient estimate of our variable of
interest, Divestiture, is similar to the ones we obtained in our baseline models in Table 3.
Another robustness check is to restrict the sample to US acquirers. Column 2 of Table 8 reports the result. The coefficient estimate of
Divestiture is 3.21%, and it is statistically significant at the 1% level. The value contribution of divestitures for US acquirers appears to
be relatively larger than the results obtained with the initial global sample (3.21% versus 2.01%). This is not that surprising as the US is
a relatively more stock market-oriented economy with the M&A market playing an important role in resource allocation. The global
sample allows us to exploit differences in employee protection laws and better isolate the synergy channel of our efficient restructuring
hypothesis.
Our main analyses employ the market model to estimate the abnormal returns around acquisition and divestiture announcements.
In some cases, the 200-day window used to estimate the parameters of the market model may overlap with the event window of either
acquisitions or divestitures, potentially introducing a bias in the measurement of the abnormal return. While we expect this bias to be
relatively negligible, we rerun the main analysis using abnormal returns computed with the market-adjusted approach, which does not
require an estimation window. Appendix B reports the univariate statistics. While the values of the abnormal returns are slightly larger
on average with the market-adjusted approach than those with the market-model approach reported in Table 2, the picture that
emerges is similar. Column 3 of Table 8 shows the regression result, which closely mirrors those in Table 3, confirming the view that
our results are robust to the choice of the normal return generating process.
As further robustness checks, we perform a battery of unreported analyses related to employee protection. First, we consider
Bargaining Coverage (i.e., total number of employees covered by collective bargaining agreements divided by all wage and salary
earners with the right to bargain in employment) as an alternative proxy for employee protection. The results are comparable to the
models with the EPL variable. Second, we assess whether the negative impact of EPL on the value contribution of divestitures is
concentrated in industries with high labor dependence. Following Levine et al. (2020), we use high labor volatility at the industry level
as a proxy for labor dependence. Our unreported results indicate that the negative joint effect of EPL and divestitures on total value
creation is driven mainly by acquiring firms in industries with high labor dependence. Third, given that 31% of the focal acquisitions
are cross-border in our sample, which implies that there may be a difference in the tightness of the employment protection in the
countries of the target and acquiring firms, we examine whether our main findings are robust to the use of the EPL index of the target
company in the focal deal in lieu of the one of the acquirers. Our findings are not sensitive to this change. Finally, the EPL measure used
in our analysis is the summary indicator for individual dismissals of regular workers. As a last check, we therefore re-run our models
employing the summary indicator for individual and collective dismissals of regular workers. Given that two measures have an
extremely high correlation (0.98), it is not surprising that our results are unaffected by the choice of the EPL measure.

4.3. Buy-and-hold abnormal returns

Acquisitions and divestitures are likely to be jointly determined through an optimization process, and in equilibrium, firms that
need to disinvest to bolster acquisition-related synergies do it, and the ones that do not need to do disinvestment do not do it. To

Table 8
Additional results and robustness checks.
Controlling for equity and debt issues US acquirers Market-adjusted total CAR BHAR

(1) (2) (3) (4)

Divestiture 0.0200*** 0.0321*** 0.0229*** 0.0098

[0.0042] [0.0060] [0.0043] [0.0098]
Equity Issue 0.0006
[0.0030]
Debt Issue 0.0058**
[0.0028]
Controls Yes Yes Yes Yes
Country FE Yes No Yes Yes
Industry FE Yes Yes Yes Yes
Year FE Yes Yes Yes Yes
Adjusted R2 0.077 0.077 0.082 0.098
Observations 6011 2661 6011 6011

This table provides various additional analyses and robustness checks. In Column 1, we augment the baseline specification with two additional binary
variables identifying equity and debt issues around the focal deal (i.e., these are issues that have occurred within the acquisition period of interest,
from 1 year prior to the announcement, to 1 year after completion of the focal acquisition). In Column 2, we restrict the sample of observations to US
acquirers. In the first two columns, the dependent variable is Total CAR (i.e., acquisition announcement 3-day CAR plus divestiture announcement 3-
day CARs). In Column 3, the dependent variable is Total CAR measured using the market-adjusted approach instead of the market model. In Column 4,
the dependent variable is the buy-and-hold abnormal returns (BHAR) for our sample firms over the period going from 1 year prior to the
announcement of the focal deal till 1 year after its completion. Each model includes the same set of controls and fixed effects (FE) as in Column 3 of
Table 3, whose coefficients are suppressed for brevity. Variable definitions are in Appendix A. All variables are winsorized at 1% on both tails.
Standard errors are clustered at firm level and reported within brackets. ***, **, and * indicate significance at the 1%, 5%, and 10%, respectively.

15
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

examine whether this equilibrium argument is at play in our sample, we assess the value effect of the asset-related restructuring process
over a longer window by computing the buy-and-hold abnormal returns (BHAR) over the period from 1 year before the announcement
of the focal deal, till 1 year after its completion. The BHAR of fim i is given by the following equation:
( ) ( )
BHARi = Π 1 + Ri,t –Π 1 + RM,t , (1)

where Ri,t is the stock return of firm i on day t, and RM,t is the return of the market index on day t. Appendix B reports the summary
statistics and Column 4 of Table 8 the regression result with BHAR as the dependent variable. The average BHAR around the focal
acquisition is 4.68% in our sample (see Appendix B), and it is not statistically different between the two acquirer groups (i.e., without
and with divestitures). This univariate comparison is confirmed with the multivariate regression result reported in Column 4 of
Table 8. The divestiture dummy is statistically insignificant. These results are consistent with the equilibrium argument.

4.4. Alternative approach to identify asset restructuring programs

To further assess to robustness of our findings, we rely on the dormant period approach as an alternative method to identify the
asset restructuring process (see, e.g., Aktas et al. (2013) for a similar approach). To identify the start (and the end) of a new asset
restructuring process, we impose a certain number of years without any asset-related restructuring activity (i.e., a dormant period with
neither acquisition nor divestiture). We consider both 1-year and 3-year as dormant period.
Panel A of Table 9 reports the summary statistics on the asset restructuring programs identified with the dormant period approach.
The 1-year dormant period approach leads to the identification of 23,191 asset restructuring programs. The average number of deals (i.
e., a deal being either an acquisition or a divestiture) is 1.48, and the corresponding median is 1.00. Out of these 23,191 programs,
11,537 of them include at least a divestiture, 12,666 of them include at least an acquisition, and the number of programs including
both divestitures and acquisitions is 1012. The 3-year dormant period approach, being more restrictive, leads to the identification of
less restructuring programs as expected (i.e., 14,859), but with a larger average number of deals included in the program (i.e., 1.82
deals per program on average).
Panel B of Table 9 reports the regression results. Columns 1 and 2 consider the programs with at least one deal (either an acquisition
or a divestiture), while Columns 3 and 4 restrict the sample to programs that include at least one acquisition. The dependent variable is
Total CAR, which sums the announcement 3-day CAR of the deals included in the program. The results are largely consistent with our
initial findings, with divestitures having a positive contribution to the total value created in the asset restructuring program. These
results further support the efficient restructuring hypothesis.

Table 9
Dormant period approach to identify asset restructuring programs.
Panel A. Summary statistics on restructuring programs

1-year hiatus programs 3-year hiatus programs

Number of restructuring programs 23,191 14,859

Average number of deals per program 1.48 1.82
Median number of deals per program 1.00 1.00
Number of programs with at least one divestiture 11,537 7798
Number of programs with at least one acquisition 12,666 8204
Number of programs with divestitures and acquisitions 1012 1143

Panel B. Total CAR regressions

Programs with at least one deal Programs with at least one acquisition

1-year hiatus 3-year hiatus 1-year hiatus 3-year hiatus

(1) (2) (3) (4)

Divestiture 0.0048** 0.0087*** 0.0235*** 0.0355***

[0.0020] [0.0033] [0.0047] [0.0064]
Controls Yes Yes Yes Yes
Country FE Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes
Year FE Yes Yes Yes Yes
Adjusted R2 0.030 0.029 0.054 0.055
Observations 16,455 10,016 9385 5773

The table presents summary statistics and regression results for determinants of wealth creation for asset restructuring programs preceded and
followed by 1- or 3-year dormant period. Panel A reports summary statistics for 1- and 3-year hiatus restructuring programs. Panel B presents the
coefficient estimates of OLS regressions where the dependent variable is Total CAR (i.e., sum of all deal 3-day CARs that occurred during the program)
associated with the 1-year and 3-year hiatus programs, respectively. Each model includes the same set of controls and fixed effects (FE) as in Column 3
of Table 3, whose coefficients are suppressed for brevity. Variable definitions are in Appendix A. All variables are winsorized at 1% on both tails.
Standard errors are clustered at firm level and reported within brackets. ***, **, and * indicate significance at the 1%, 5%, and 10%, respectively.

16
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

4.5. Are divestitures around acquisitions associated with weak bargaining power?

Selling assets around acquisitions might be a suboptimal decision in comparison to a sale process organized in isolation. Thus, to
confirm that divestitures are truly part of an efficient value-maximizing strategy, we need to show that firms are, at least, not worse off
by selling assets around acquisitions in comparison to the ones taking place in isolation.
To this end, we rely on a large sample of divestitures announced over the 1996–2016 period and compare the 3-day abnormal
returns associated with divestitures embedded in acquisition processes and the ones that are implemented in isolation. Panel A of
Table 10 shows the univariate analysis. Overall, in line with prior literature (see, e.g., Bates, 2005), we find that divestitures create
value, with an average abnormal return of 1.56%. The univariate analysis indicates that acquisition-related divestitures create on
average lower value in comparison to divestitures implemented in isolation (1.09% vs. 1.60%). However, after controlling for deal and
firm characteristics in Panel B, the picture changes rather dramatically. In this multivariate analysis, we regress the 3-day divestiture
CAR on a binary variable identifying whether the divestiture is related to an acquisition (i.e., the binary variable is denoted Acquisition)
and the same set of control variables as in Column 3 of Table 3. The acquisition dummy is statistically insignificant (see Column 1 in
Panel B of Table 10). Moreover, the timing of the acquisition-related divestiture relative to the focal acquisition does not affect this
result (Column 2). Thus, we do not find that divestitures embedded into an acquisition process are associated with a weak bargaining
power for the seller or that these divestitures are fire asset sales. Therefore, these findings further support the profit-maximizing view.

5. Conclusion

Focusing on a global sample of relatively large acquisitions, we study the entire acquisition-centered restructuring process and
examine all divestitures taking place before and after the announcement of the focal deal. We argue that these divestitures associated
with focal acquisitions are not just used to raise financing but that they are part of a profit-maximizing asset restructuring process to

Table 10
Wealth creation around divestiture announcements.
Panel A. Summary statistics

Period All divestitures (N = 17,806) Divestitures within an M&A process (N = 1397) Divestitures not related to an M&A process (N = 16,409)

Mean Median Mean Median N Mean Median

All 1.56% 0.36% 1.09% 0.30% 1.60% 0.36%

(0.00) (0.62)
Pre 1.27% 0.40% 511
Interim 0.65% 0.43% 188
Post 1.09% 0.19% 698

Panel B. Divestiture CAR regressions

(1) (2)

Acquisition 0.0012
[0.0021]
Acquisition Pre 0.0036
[0.0034]
Acquisition Interim 0.0017
[0.0044]
Acquisition Post − 0.0001
[0.0027]
Controls Yes Yes
Country FE Yes Yes
Industry FE Yes Yes
Year FE Yes Yes
Adjusted R2 0.049 0.049
Observations 13,059 13,059

Panel A presents summary statistics on 3-day cumulative abnormal returns around the divestiture announcement. The first two columns report on all
divestitures, the middle three columns on divestitures that are within an acquisition-centered restructuring process, and the last three columns on the
ones that are not part in an acquisition-centered restructuring process. The P-values of the difference in mean and median tests between the two
subsamples are reported within parentheses below the corresponding coefficient in the last two columns. Panel B presents the coefficient estimates of
OLS regression where the dependent variable is the 3-day cumulative abnormal return around the divestiture announcements. The specifications
include year, industry, and country fixed effects. Acquisition is a binary variable identifying divestitures that are related to an acquisition process.
Acquisition Pre (Post) identifies divestitures that are related to an acquisition process and implemented in the pre-announcement (post-closing) period.
Acquisition Interim identifies divestitures implemented in the interim period of an acquisition process. Each model includes the same set of controls
and fixed effects (FE) as in Column 3 of Table 3, whose coefficients are suppressed for brevity. Variable definitions are in Appendix A. All variables are
winsorized at 1% on both tails. Standard errors are clustered at firm level and reported within brackets. ***, **, and * indicate significance at the 1%,
5%, and 10%, respectively.

17
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

redraw the firm's boundaries.

In support of this efficient restructuring, we document that acquirers divest more assets when they are larger, more diversified,
acquisitive, and have lower market valuation. Furthermore, more than half of these divestitures take place after the completion of the
focal acquisition, a period that is relatively less sensitive to financing and regulatory needs. Examining the efficiency of the acquisition-
driven asset reorganizations, we document that divestitures enhance the value creation associated with acquisitions. We also show that
the value contribution of these divestitures is related to the synergistic potential of the focal deal. We do not find support for alternative
efficiency channels, such as agency correction, financial constraint relaxation, and regulatory concern anticipation.
We also carry out additional tests to examine whether divesting during an acquisition process weakens the seller's bargaining
power, and document that this is not the case. Our results indicate that firms do not make sub-optimal decisions when they sell assets
around acquisitions. Overall, we find evidence that acquisitions are catalysts for value-creating restructuring processes. Large com­
panies seem to adopt a dynamic perspective to restructure their assets, with both the buy- and sell-side being essential activities that
contribute to firm performance.

Acknowledgments

We are grateful to Emanuele Bajo, Massimiliano Barbi, Eric de Bodt, Jean-Gabriel Cousin, Ding Ding (discussant), Andrey Golubov,
Mussa Hussaini (discussant), Peter Limbach, Ellie Luu (discussant), Dimitris Petmezas, Markus Schmid, Merih Sevilir, Fabian Silbereis
(discussant), Linus Siming, Stefano Rossi, and Paul Voss (discussant) for helpful comments and suggestions. We also thank participants
at the FMA Annual Meeting (2020), the Paris December Finance Meeting (2020), the PhD Nordic Finance Workshop (2021), the AFFI
Annual Meeting (2021), and the DGF Annual Meeting (2021) for helpful comments and suggestions. This research did not receive any
specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Appendix A. Variable definitions

Unless explicitly mentioned otherwise, Thomson Reuters' Worldscope is the data source for financial- and accounting-related
variables, Thomson Reuters' Datastream for market-related variables, and Thomson One Banker for transaction-related variables
and equity/debt issues.

A.1. Dependent variables

Acquisition (Divestiture) CAR: Cumulative abnormal return for the acquiring (divesting) firm over the 3-day event window (− 1, +1)
around the announcement day of the focal deal (divestiture). The abnormal return is computed using a market model with parameters
estimated over the estimation period (− 240, − 41) with respect to the announcement day. As a robustness check, we also use abnormal
returns computed with the market-adjusted approach, which does not require an estimation window. We employ the local index
(datatype LI) as market index (Source: Datastream).
Total CAR: It corresponds to the acquisition CAR plus the sum of the divestiture CARs in the acquisition-centered restructuring
process.
BHAR: Buy-and-hold abnormal returns (BHAR) of the acquiring firm over the period from 1 year before the announcement of the
focal deal, till 1 year after its completion (see Eq. (1)).

A.2. Independent variables of interest

Divestiture: Binary variable that takes the value of 1 if the acquirer has done at least one divestiture during the acquisition process (i.
e., from 1 year before the announcement to 1 year after the completion of the focal deal), 0 otherwise.
Divestiture Pre: Binary variable that takes the value of 1 if the acquirer has done at least one divestiture in the year before the
announcement of the acquisition, 0 otherwise.
Divestiture Interim: Binary variable that takes the value of 1 if the acquirer has done at least one divestiture in the period between the
announcement and completion of the focal deal, 0 otherwise.
Divestiture Post: Binary variable that takes the value of 1 if the acquirer has done at least one divestiture in the year following the
completion of the focal deal, 0 otherwise.

A.3. Firm variables

Capex: Capital expenditures divided by total assets.

Cash holding: Cash reserves divided by total assets.
Debt Issue: Binary variable that takes the value of 1 if the firm has issued debt in the form of loan or bond during the acquisition
process (i.e., from 1 year prior to the announcement to 1 year after the completion of the focal deal), 0 otherwise.
Diversified: Binary variable that takes the value of 1 if the firm is active in more than one business segment, 0 otherwise.
Dividend Payer: Binary variable that takes the value of 1 if the firm paid cash dividends in the year before the deal, zero otherwise.
Equity Issue: Binary variable that takes the value of 1 if the firm has issued equity during the acquisition process (i.e., from 1 year
prior to the announcement to 1 year after the completion of the focal deal), 0 otherwise.

18
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

Excess Price Margin: Following Gaspar and Massa (2005) and Peress (2010), the price-cost margin (PCM) is defined as operating
profits (before depreciation, interest, special items, and taxes) over sales (if data are missing, we use operating income). Excess Price
Margin corresponds to the difference between the firm's PCM and the PCM of its industry. The industry PCM is the value-weighted
average PCM across firms in the industry where the weights are based on market share and industries are defined using two-digit
SIC code.
Leverage: Total debt divided by total assets.
Market Value: Market capitalization of the equity of the firm.
Past acquisition: Binary variable that takes the value of 1 if the firm has implemented other deals (without any value and relative size
restrictions) in the previous 3 years, and 0 otherwise. We also consider a signed version, in which Past Acquisition takes the value of 1 if
the focal deal is preceded with at least one value destroying acquisition (i.e., negative acquirer CAR) in the past 3 years, and
0 otherwise.
R&D: Research and development expenses divided by total assets.
ROA: EBITDA divided by total assets.
Serial Acquirer: Binary variable that takes the value of 1 if the firm has made other acquisitions in either the 1-year period before the
announcement of the focal deal or the 1-year period after the closing of the focal deal, 0 otherwise.
Size: Log(total assets in $ million).
Small firm: Binary variable that takes the value of 1 if the firm is in the bottom tercile based on total assets, and 0 otherwise.
Tobin's Q: Market value of the equity plus total debt divided by total assets.
Young firm: Binary variable that takes the value of 1 if the firm is in the bottom tercile based on firm age, and 0 otherwise.

A.4. Industry variables

Herfindhal Index: Sum of squares of the market shares of all firms in a given year, country and two-digit SIC code industry, where
market share is defined as sales of the firm divided by the sum of the sales in the industry.
M&A Liquidity: It measures the liquidity of the M&A market in the industry of the acquiring firm in a given year. It is computed as
the sum of acquisitions in a given year, country, and two-digit SIC code, divided by the sum of total assets of all firms in the same
country and two-digit SIC code.

A.5. Deal characteristics

Cross Border: Binary variable that takes value of 1 if the target and the acquirer are from different countries, 0 otherwise.
Cross Industry: Binary variable that takes value of 1 if the target and the acquirer are from different industries, 0 otherwise Industries
follows the Fama-French 49-industry classification.
High Regulatory Risk: Binary variable that takes the value of 1 if the acquirer is a market leader in its country and industry and if the
relative size of the focal deal is greater than 10% (or 20%), and 0 otherwise. We define market leaders based on the firm's market share
in the corresponding country and industry. Firms in top tercile are considered as market leaders.
Public Target: Binary variable that takes value of 1 if the target of the focal acquisition is a publicly listed firm, 0 otherwise.
Relative Size: Value of the focal deal divided by the market value of the acquiring firm from the last fiscal year before the deal
announcement.
Stock (Cash): Binary variable that takes value of 1 if the method of payment in the focal deal is fully stock (fully cash), 0 otherwise.

A.6. Country variables

Banking Development: Domestic credit to private sector by banks as percentage of the corresponding GDP (Source: World Bank WDI).
EPL: It corresponds to the Employee Protection Law Index, which measures the strictness of regulations that an employer must
follow in order to dismiss an individual worker with a regular contract; it ranges from 0 to 6 and is time-varying. (Source: OECD).
Stock Market Development: Market capitalization of listed domestic companies as a percentage of the corresponding country GDP
(Source: World Bank WDI).
GDP Growth: GDP at purchaser's prices is the sum of gross value added by all resident producers in the economy plus any product
taxes and minus any subsidies not included in the value of the products. It is calculated without making deductions for depreciation of
fabricated assets or for depletion and degradation of natural resources (Source: World Bank WDI).
GDP Per Capita: Gross domestic product divided by midyear population. GDP is the sum of gross value added by all resident
producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. It is calculated
without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources. We use the log
transform of the variable (Source: World Bank WDI).

Appendix B. Additional summary statistics

This appendix provides the summary statistics of the new variables used in Table 7. The first panel reports on equity and debt issues
around focal acquisitions (include issues that have occurred within the acquisition period of interest, from 1 year prior to the
announcement, to 1 year after completion of the focal acquisition). Debt issue is constructed by summing loan and bond issues. Equity

19
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

and debt proceeds are in $ million. Variable definitions are in Appendix A. The second panel reports on the 3-day market-adjusted
CARs around the acquisition announcement and the divestiture announcement. The last row is devoted to Total CAR (i.e., acquisi­
tion announcement 3-day market-adjusted CAR plus divestiture announcement 3-day market-adjusted CARs). The last panel reports on
the buy-and-hold abnormal returns (BHAR) for our sample firms over the period going from 1 year prior to the announcement of the
focal deal till 1 year after its completion. The last two columns report the P-values of the difference in mean and median tests between
the two considered subsamples.

Acquisitions with divestitures Acquisitions without divestitures P-value

Mean Median N Mean Median N Mean Median

B.1. Equity and debt issues

Equity Issues Dummy 26.71% 0.00% 876 32.57% 0.00% 5969 0.00 0.00
Debt Issues Dummy 63.70% 100.00% 876 46.42% 0.00% 5969 0.00 0.00
Equity Proceeds 512.35 178 234 292 115 1944 0.00 0.00
Debt Proceeds 4108 975 558 1536 453 2771 0.00 0.00

B.2. Market-adjusted CARs

Acquisition CAR 2.00% 0.88% 876 2.70% 1.44% 5969 0.03 0.06
Divestiture CAR 1.27% 0.71% 876 5969
Total CAR 3.54% 2.08% 876 2.70% 1.44% 5969 0.05 0.06

B.3. Buy-and-hold returns

BHAR 3.74% 1.46% 876 4.82% 1.13% 5969 0.27 0.64

Appendix C. Descriptive statistics by country

The table presents descriptive statistics at country level for our sample of focal deals announced over the 1996–2016 period. We
report the number of focal deals by country (Column 1); the number and percentage of focal deals with divestitures (Columns 2 and 3);
the number of total divestitures per country and the average number of divestitures per focal deal in a country (Columns 4 and 5). The
category Other countries includes observations from 22 countries with less than 10 focal deals (Argentina, Bahrain, Chile, Colombia,
Croatia, Egypt, Hungary, Iceland, Jordan, Kuwait, Lithuania, Luxembourg, Nigeria, Oman, Peru, Qatar, Saudi Arabia, Slovenia,
Turkey, Ukraine, Utd. Arab Emirates, and Vietnam). N is the number of observations.

Country Focal deals Divestitures

(1) (2) (3) (4) (5)

N With divestiture % with div. N N. per focal deal

Australia 242 46 19.01% 81 1.76

Austria 18 0 0.00% 0 0.00
Belgium 28 5 17.86% 7 1.40
Bermuda 13 0 0.00% 0 0.00
Brazil 53 3 5.66% 4 1.33
Canada 492 64 13.01% 102 1.59
China 583 22 3.77% 23 1.05
Denmark 34 5 14.71% 5 1.00
Finland 63 11 17.46% 18 1.64
France 169 34 20.12% 65 1.91
Germany 99 9 9.09% 23 2.56
Greece 13 0 0.00% 0 0.00
Hong Kong 85 3 3.53% 3 1.00
India 71 2 2.82% 2 1.00
Indonesia 15 0 0.00% 0 0.00
Ireland 30 2 6.67% 3 1.50
Israel 28 2 7.14% 2 1.00
Italy 67 5 7.46% 7 1.40
Japan 351 18 5.13% 21 1.17
Malaysia 31 0 0.00% 0 0.00
Mexico 35 1 2.86% 4 4.00
Netherlands 78 8 10.26% 25 3.13
New Zealand 24 2 8.33% 2 1.00
Norway 48 6 12.50% 11 1.83
(continued on next page)

20
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

(continued )
Country Focal deals Divestitures

(1) (2) (3) (4) (5)

N With divestiture % with div. N N. per focal deal

Philippines 11 1 9.09% 1 1.00

Poland 22 0 0.00% 0 0.00
Portugal 11 1 9.09% 1 1.00
Russian Fed 20 0 0.00% 0 0.00
Singapore 38 0 0.00% 0 0.00
South Africa 53 5 9.43% 8 1.60
South Korea 115 6 5.22% 8 1.33
Spain 51 11 21.57% 22 2.00
Sweden 110 13 11.82% 17 1.31
Switzerland 84 9 10.71% 25 2.78
Taiwan 46 2 4.35% 2 1.00
Thailand 21 1 4.76% 1 1.00
United Kingdom 576 144 25.00% 250 1.74
United States 2942 435 14.79% 654 1.50
Other Countries 75 0 0.00% 0 0.00
Total 6845 876 1397 1.59

References

Ahern, K.R., Harford, J., 2014. The importance of industry links in merger waves. J. Financ. 69, 527–576.
Ahmad, M.F., Lambert, T., 2019. Collective bargaining and mergers and acquisitions activity around the world. J. Bank. Financ. 99, 21–44.
Aktas, N., de Bodt, E., Roll, R., 2004. Market response to European regulation of business combinations. J. Financ. Quant. Anal. 39, 731–757.
Aktas, N., de Bodt, E., Roll, R., 2013. Learning from repetitive acquisitions: evidence from the time between deals. J. Financ. Econ. 108, 99–117.
Alimov, A., 2015. Labor market regulations and cross-border mergers and acquisitions. J. Int. Bus. Stud. 46, 984–1009.
Arcot, S., Gantchev, N., Sevilir, M., 2020. Divest to Acquire. Working paper, Kelley School of Business Research Paper no. 2020-63.
Arnold, M., Hackbarth, D., Puhan, T.X., 2018. Financing asset sales and business cycles. Rev. Financ. 22, 243–277.
Asquith, P., Bruner, R., Mullins, D., 1983. The gains to bidding firms from merger. J. Financ. Econ. 11, 121–139.
Bates, T.W., 2005. Asset sales investment opportunities, and the use of proceeds. J. Financ. 60, 105–135.
Betton, S., Eckbo, B.E., Thorburn, K.S., 2008. Corporate takeovers. In: Eckbo, B.E. (Ed.), Handbook of Corporate Finance, Empirical Corporate Finance, Vol. 2.
Elsevier, North Holland, Amsterdam, pp. 291–429.
Bongaerts, D., Schlingemann, F.P., 2020. The Real Effects of Credit Ratings: Evidence from Corporate Asset Sales. Working Paper. Erasmus University and University
of Pittsburgh.
Cai, J., Song, M.H., Walkling, R.A., 2011. Anticipation, acquisitions, and bidder returns: industry shocks and the transfer of information across rivals. Rev. Financ.
Stud. 24, 2242–2285.
Capron, L., Mitchell, W., Swaminathan, A., 2001. Asset divestiture following horizontal acquisitions: a dynamic view. Strateg. Manag. J. 22, 817–844.
Comment, R., Jarrell, G.A., 1995. Corporate focus and stock returns. J. Financ. Econ. 37, 67–87.
de Bodt, E., Cousin, J.G., Demidova, I., 2014. M&A outcomes and willingness to sell. Finance 35, 7–49.
Demirgüc-Kunt, A., Feyen, F., Levine, R., 2013. The evolving importance of banks and securities markets. World Bank Econ. Rev. 273, 476–490.
Denis, D.J., Sibilkov, V., 2010. Financial constraints, investment, and the value of cash holdings. Rev. Financ. Stud. 23, 247–269.
Dessaint, O., Golubov, A., Volpin, P., 2017. Employment protection and takeovers. J. Financ. Econ. 125, 369–388.
Devos, E., Kadapakkam, P.R., Krishnamurthy, S., 2009. How do mergers create value? A comparison of taxes, market power, and efficiency improvements as
explanations for synergies. Rev. Financ. Stud. 22, 1179–1211.
Dong, M., Hirshleifer, D., Richardson, S., Teoh, S.H., 2006. Does investor misevaluation drive the takeover market? J. Financ. 61, 725–762.
Edmans, A., Mann, W., 2019. Financing through asset sales. Manag. Sci. 65, 2947–3448.
Fidrmuc, J.P., Roosenboom, P., Zhang, E.Q., 2018. Antitrust merger review costs and acquirer lobbying. J. Corp. Finan. 51, 72–97.
Fluck, Z., Lynch, A.W., 1999. Why do firms merge and then divest? A theory of financial synergy. J. Bus. 72, 319–346.
Gaspar, J.M., Massa, M., 2005. Idiosyncratic volatility and product market competition. J. Bus. 79, 3125–3152.
Hadlock, C.J., Pierce, J.R., 2010. New evidence on measuring financial constraints: moving beyond the KZ index. Rev. Financ. Stud. 23, 1909–1940.
Harford, J., 2005. What drives merger waves? J. Financ. Econ. 77, 529–560.
Harford, J., Uysal, V., 2014. Bond market access and investment. J. Financ. Econ. 112, 147–163.
Houston, J.F., James, C.M., Ryngaert, M.D., 2001. Where do merger gains come from? Bank mergers from the perspective of insiders and outsiders. J. Financ. Econ.
60, 285–331.
Hovakimian, G., Titman, S., 2006. Corporate investment with financial constraints: sensitivity of investment to funds from voluntary asset sales. J. Money Credit Bank.
38, 357–374.
Hsu, P., Tian, X., Xu, Y., 2014. Financial development and innovation: cross–country evidence. J. Financ. Econ. 112, 116–135.
Jansen, I., Sanning, L., Stuart, N., 2013. On the relation between the relative size of acquisitions and the wealth of acquiring firms. Appl. Econ. Lett. 20, 534–539.
John, K., Ofek, E., 1995. Asset sales and increase in focus. J. Financ. Econ. 31, 105–126.
John, K., Knyazeva, A., Knyazeva, D., 2015. Employee rights and acquisitions. J. Financ. Econ. 118, 49–69.
Kaplan, S.N., Weisbach, M.S., 1992. The success of acquisitions: evidence from divestitures. J. Financ. 47, 107–138.
Lang, L., Poulsen, A., Stulz, R., 1995. Asset sales, firm performance, and the agency costs of managerial discretion. J. Financ. Econ. 37, 3–37.
Levine, R., Zervos, S., 1998. Stock markets, banks, and economic growth. Am. Econ. Rev. 88, 537–558.
Levine, R., Lin, C., Shen, B., 2020. Cross-border acquisitions: do labor regulations affect acquirer returns? J. Int. Bus. Stud. 51, 194–217.
Loderer, C., Martin, K., 1990. Corporate acquisitions by listed firms: the experience of comprehensive sample. Financ. Manag. 19, 17–33.
Maksimovic, V., Phillips, G., 2001. The market for corporate assets: who engages in mergers and asset sales and are there gains? J. Financ. 56, 2019–2065.
Maksimovic, V., Phillips, G., 2002. Do conglomerate firms allocate resources inefficiently across industries? Theory and evidence. J. Financ. 57, 721–767.
Maksimovic, V., Phillips, G., Prabhala, N.R., 2011. Post-merger restructuring and the boundaries of the firm. J. Financ. Econ. 102, 317–343.
Maksimovic, V., Phillips, G., Yang, L., 2013. Private and public merger waves. J. Financ. 68, 2177–2217.

21
N. Aktas et al. Journal of Corporate Finance 73 (2022) 102189

Mavis, C.P., McNamee, N., Petmezas, D., Travlos, N.G., 2020. Selling to buy: asset sales and acquisitions. J. Corp. Finan. 62, 101587.
Moeller, S.B., Schlingemann, F.P., Stulz, R.M., 2004. Firm size and the gains from acquisitions. J. Financ. Econ. 73, 201–228.
Mulherin, J.H., Boone, A., 2000. Comparing acquisitions and divestitures. J. Corp. Finan. 6, 117–139.
Peress, J., 2010. Product market competition, insider trading, and stock market efficiency. J. Financ. 65, 1–43.
Rhodes-Kropf, M., Robinson, D.T., 2008. The market for mergers and the boundaries of the firm. J. Financ. 63, 1169–1211.
Schlingemann, F.P., Stulz, R.M., Walkling, R.A., 2002. Divestitures and the liquidity of the market for corporate assets. J. Financ. Econ. 64, 117–144.
Shleifer, A., Vishny, R.W., 2011. Fire sales in finance and macroeconomics. J. Econ. Perspect. 25, 29–48.
Wang, W., 2018. Bid anticipation, information revelation, and merger gains. J. Financ. Econ. 128, 320–343.

22