Price and Probability: Decomposing the Takeover Effects of Anti‐Takeover Provisions

• Author: MARIA GUADALUPE,VICENTE CUÑAT,MIREIA GINÉ
• DOI: http://doi.org/10.1111/jofi.12908
• Date: 01 October 2020
• Published date: 01 October 2020

THE JOURNAL OF FINANCE •VOL. LXXV, NO. 5 •OCTOBER 2020

Price and Probability: Decomposing the

Takeover Effects of Anti-Takeover Provisions

VICENTE CUÑAT, MIREIA GINÉ, and MARIA GUADALUPE∗

ABSTRACT

We study the effects of anti-takeover provisions (ATPs) on the takeover probability,

the takeover premium, and target selection. Voting to remove an ATP increases both

the takeover probability and the takeover premium, that is, there is no evidence of

a trade-off between premiums and takeover probabilities. We provide causal esti-

mates based on shareholder proposals to remove ATPs and address the endogenous

selection of targets through bounding techniques. The positive premium effect in less

protected firms is driven by better bidder-target matching and merger synergies.

ANTI-TAKEOVER PROVISIONS (ATPS)SUCH as staggered boards, dual-class

shares, poison pills, and similar governance mechanisms have been found to

affect firm value (e.g., Gompers, Ishii, and Metrick (2003), Cuñat, Giné, and

Guadalupe (2012)).1In the context of the takeover debate, proponents argue

that ATPs create value by allowing managers to negotiate a higher price in

the event of a hostile bid and encouraging more long-term investment (Stein

(1988), Harris (1990)). However, they may also reduce or delay the possibility

of a takeover (Ryngaert (1988), Pound (1987), Malatesta and Walkling (1988),

Comment and Schwert (1995), Karpoff, Schonlau, and Wehrly (2017)). The

∗Vicente Cuñat is at LSE. Mireia Gine is with WRDS, University of Pennsylvania and IESE

Business School. Maria Guadalupe is with INSEAD. We would like to thank Laurent Bach;

Bernard Black; Marco Becht; Emiliano Catan; François Derrien; Andrey Golubov; Denis Gromb;

Nadia Malenko; Marco Pagano; Miikka Rokkanen; Antoinette Schoar; Rob Shonlau; Yishay Yafeh;

and seminar/conference participants at MIT-Sloan, McCombs UT Austin, Columbia Business

School, LSE, HEC, LBS, Exeter, Carlos III, INSEAD, Duke University, UNC Kenan Flagler,

Tilburg University,Amsterdam University, University of Rotterdam, Tel Aviv University, Hebrew

University, CEPR Adam Smith workshop, SFS Cavalcade, ECGI, CGCG Colloquia, FMA Europe,

and AFA meetings for helpful comments and suggestions. We are also grateful to Amit Seru, an

anonymous Associate Editor, and three anonymous referees for constructive comments and sug-

gestions that have significantly improved the paper. We have read The Journal of Finance disclo-

sure policy and have no conflicts of interest to disclose.

Correspondence: Vicente Cuñat, Department of Finance, The London School of Economics,

Houghton St., London WC2 2AE, UK; e-mail: v.cunat@lse.ac.uk.

This is an open access article under the terms of the Creative Commons Attribution-Non

Commercial-NoDerivs License, which permits use and distribution in any medium, provided the

original work is properly cited, the use is non-commercial and no modifications or adaptations are

made.

1Reducing the threat of a takeover can destroy value by weakening managerial discipline

(Scharfstein (1988), Bertrand and Mullainathan (2003), Gormley and Matsa (2016)).

DOI: 10.1111/jofi.12908

© 2020 The Authors. The Journal of Finance published by Wiley Periodicals LLC on behalf of

American Finance Association

2591

2592 The Journal of Finance®

trade-off between the price (i.e., premium) and the probability of a takeover

has become common wisdom and accepted as fact.

The goal of this paper is to provide causal estimates that allow us to as-

sess the extent of the trade-off between price and probability, and to identify

the types of mergers that ATPs deter versus allow to happen. We also aim to

identify the channels through which firm-level ATPs create or destroy value

for firms and for the economy as a whole. Establishing causal effects is impor-

tant given the evidence of endogeneity of governance structures (Schoar and

Washington (2011), Karpoff, Schonlau, and Wehrly (2017)) and the potential

of mergers and acquisitions to create or destroy value (Morck, Schleifer, and

Vishny (1990), Maksimovic and Phillips (2001), Burkart, Gromb, and Panunzi

(1998), Schoar (2002)).

We start by showing that the expected gains from adopting ATPs can ac-

crue to shareholders in three ways. The first is the effect of such provisions

on the probability of being acquired (i.e., the deterrent effect). The second is

the effect on the premium paid conditional on a successful acquisition. This is

based on the price paid for the target at auction, which is determined in turn

by the relative bargaining power of the parties, the degree of competition (i.e.,

number of bidders), and the potential for synergies. Note that the effect is ex

ante ambiguous: for example, if ATPs give managers more bargaining power,

removing them should result in a negative effect on the premium while if they

attract less competition, removing them implies a positive effect (Bulow and

Klemperer (1996)). The third is the selection effect, whereby an ATP changes

the population of firms that become targets. For example, the additional firms

that end up being taken over because they dropped an ATP may be those with

the lowest (highest) potential for value creation, implying negative (positive)

selection. The first two effects have been the focus of existing research, as they

are important determinants of shareholder value. The third, albeit seldom dis-

cussed, is also important, as one cannot infer the takeover premium by compar-

ing firms that are taken over with and without ATPs because the population

of target firms changes when such provisions are in place.2

To provide causal estimates of the components of the expected premium in

this setting requires some form of random assignment in the adoption of ATPs.

We employ two different specifications. The first is a regression discontinuity

design (RDD) for the takeover probability and expected premium. We use data

on all shareholder-sponsored proposals (2,882 proposals in 927 different firms)

to remove an ATP voted on at annual meetings of S&P 1500 firms between

1994 and 2013. We rely on vote outcomes being random in a narrow interval

around the majority threshold, leading to a discrete change in the probabil-

ity of dropping a provision (see Cuñat, Gine, and Guadalupe (2012, 2013)).

The second specification is a matching estimator that is validated using the

2Note that most existing studies focus on effective (conditional) takeover premiums that are

conditional on a takeover offer being made. Since premiums do not exist in the absence of a

takeover bid, changes in these “conditional” premiums are subject to selection bias, as we discuss

below.

Price and Probability 2593

identification strategy proposed by Angrist and Rokkanen (2015).3Since the

two estimation strategies rely on a different set of assumptions, our study

contributes to our understanding by providing consistent results across dif-

ferent techniques that evaluate the effects at different points of the sample.

Even when armed with a source of exogenous variation, we still need to cor-

rect for selection problems inherent in the estimation of the conditional pre-

mium given the co-determination of premiums with the population of firms

taken over. We use the bounding estimation strategy proposed by Lee (2009)

to estimate upper and lower bounds for the effect of ATPs on the takeover

premium. Because we need a distribution of premiums to apply the bounding

technique, which cannot be done at the exact discontinuity, we provide two

sets of results for the conditional premium using Lee bounds. The first is on

the full sample using validated matching (Angrist and Rokkanen (2015)) and

the second is in an interval around the discontinuity.

Across specifications and samples, we find that voting to remove an ATP has

a significant positive effect on both the probability of a firm being taken over

and the future takeover premium. At the majority threshold (classic RDD),

passing a proposal to drop an ATP increases the likelihood of a takeover within

five years by 9.2% (1.8% per year) and increases the expected value of future

takeover premiums by 4.1%. For firms away from the discontinuity, the ef-

fects are smaller but also positive and significant: voting to remove an ATP

increases the probability of a takeover within five years by 4.1% (0.8% per

year) and increases the expected value of future takeover premiums by 2.6%.

These results are intent to treat (ITT) effects that measure the effect of pass-

ing a proposal. In Section V, we discuss possible calculations of the effect of the

provision itself (treatment on the treated). We also show that the results are

similar across different definitions of the share of votes passed (in particular,

the different treatment of abstentions across votes).

Total shareholder gains can be expressed as an unconditional premium that

includes both firms that experience a takeover (and realize a takeover pre-

mium) and those that do not (with a takeover premium of zero). The effect on

the expected unconditional premium is not subject to the inherent selection

problem of the conditional (i.e., realized) takeover premium because the popu-

lations of the treatment and control groups are comparable. However,we would

also like to determine whether a given firm is able to obtain a higher or lower

premium if it drops the ATP and a merger does happen. To do so, we cannot

simply compare the premiums of firms taken over with or without ATPs, as we

can only observe takeover premiums for the firms that are taken over and we

need to account for different selection patterns in the two groups. We therefore

use the bounds methodology developed by Lee (2009) to provide estimates that

3Angrist and Rokkanen (2015) build on the fact that in the regression discontinuity design, we

observe the assignment variable (the vote in our case), which is the only source of heterogeneity.

They propose a matching estimator and use the regression discontinuity approach as a tool for

validating the conditional independence assumption of the model. We explain the method and

intuition further in Section III.