When Does a Merger Create Value? Using Option Prices to Elicit Market Beliefs

• Date: 01 June 2014
• Published date: 01 June 2014
• DOI: http://doi.org/10.1111/fima.12026
• Author: Paul A. Borochin

When Does a Merger Create Value?

Using Option Prices to Elicit Market

Beliefs

Paul A. Borochin∗

I introduceand test a method to identify market expectations about value creation in mergers. Post-

announcement market prices reflectbeliefs about both merged and standalone firm values, and the

likelihood of either outcome. Stock prices alone do not contain sufficient information to identify

these latent beliefs. By adding exchange-tradedstock option data, I deliver a clear decomposition of

observed value change into two parts: 1) value creation and 2) newinformation about standalone

value. Previous research has struggled to disentangle the two. This decomposition provides a

strong and practical measureof the market’s expectations about value creation in a merger.

This paper introduces a method to estimate the market’s expectations about a firm’s value in

two possible outcomes for an ongoing merger negotiation: 1) as a merged entity and 2) as a

standalone entity. It provides an answer to the most fundamental question for the shareholders

of a firm involved in such a negotiation. Would they be better off if the merger did or did not

go through? Answering this question requires knowledge of both possible outcomes. Since they

observe only one outcome and not the other, even after the negotiation is finished, shareholders

cannot obtain a direct answer.As such, a strong test of value creation in a merger has thus far been

elusive. However, byidentifying the market’s latent beliefs about a firm’svalue in both the merged

and standalone states, I provide the answer by comparing the two. Furthermore, since these are

expectations about the future, this test is available before the merger is consummated, when

this information is the most valuable. The identification of state-contingent beliefs about firm

value also enables a disentanglement of the effects of perceived synergies from newly revealed

information about the individual firms.

In responding to a newly announced merger, the market forms beliefs about the economic

variables of interest in the proposed transaction. It estimates the probability of a successful

takeover and simultaneously updates its beliefs about the standalone values of the bidder and

target companies. It also develops beliefs about the value of the combined firm that would

be created if the merger proposal succeeds. These beliefs determine stock price reactions to a

merger announcement, but the stock prices alone do not contain sufficient information to identify

the beliefs underlying them. Thus, observed stock price reactions to merger announcements

can reflect a variety of possible expectations about synergies, standalone valuations, and the

likelihood of acquisition. Thus, attempting to pin down the true set of the market’s expectations

from stock prices alone is analogous to solving an underidentified system. There are only two

I am very grateful to David Robinson, Alon Brav, Ron Gallant, Paige Ouimet, Joe Golec, Carmelo Giaccotto, Assaf

Eisdorfer,an anonymous referee, RaghavendraRau (Editor), and seminar participants at Duke University, the University

of Connecticut, and the 2012 FMA Annual Meeting for helpful comments.

∗Paul A. Borochinis an Assistant Professor at the University of Connecticut Business School in Storrs, CT.

Financial Management •Summer 2014 •pages 445 - 466

446 Financial Management rSummer 2014

ways to proceed: 1) reduce the number of variables estimated or 2) add more restrictions through

additional data points.

Previous research has, for the most part, advanced through the former approach of variable

reduction. Brown and Raymond (1986), Barone-Adesi, Brown, and Harlow (1994), Hietala,

Kaplan, and Robinson (2003), Subramanian (2004), Bhagat et al. (2005), and Bester, Martinez,

and Rosu (2011) developed methods to identify some subsets of the full set of beliefs about

mergers. Their results are applicable in some situations where it is possible to identify some of

the variables describing the market’s beliefs about a proposed merger. These past approaches,

however, fall short of a generally applicable method. I propose such a method and demonstrate

its accuracy, robustness, and added value.

This paper solves the identification problem above by adding additional data from option

prices. This allows the separate identification of the full set of market beliefs arising after merger

announcements, regardless of the offer terms. Since contemporary options markets for large

firms are reasonably liquid, they contain sufficient information to analyze all types of merger

negotiations regardless of the offer terms, outcomes, or number of participants.

A similar approach is taken in the parallel research of Barraclough, Robinson, Smith, and

Whaley (2013). This paper complements the Barraclough et al. (2013) large sample application

of a similar method, adding to the literature regarding the use of option data to estimate merger

synergies in three novel ways.

First, I conduct a series of robustness checks for employing options to estimate expected merger

outcomes. I provide evidence that the use of option prices produces reliable estimates of merger

synergies by examining plausible, but counterfactual acquisition scenarios. My method correctly

rules out the counterfactual acquirers, assigning an insignificant probability to their success and

standalone future values close to current market values. I also examine the robustness of my

method to uncertainty about announced offer terms. I find that the expected total offer value is

virtually invariant to the alternative,more generous, specif ications of offer terms. This sensitivity

analysis provides a direct test of whether the market expects the announced offer terms to be

negotiable.

Additionally,I demonstrate the added value of using options data relative to the earlier, simpler

measures of takeover synergy from stock prices alone. I repeat the estimation with a reduced

number of variables obtained by identifying assumptions about firm standalone values and the

probability of success, similar to the methods employed in earlier research. I determine that

the results obtained are highly sensitive to assumptions that have been previously used to reduce

the set of estimated variables, such as previsible firm fallback values. When stock data were the

only source of information regarding takeover expectations, these assumptions were necessary,

but they can now be done away with by including option data.

Finally, although the Bayesian estimations in this paper are conducted with uninformed priors,

this method leaves open the option of including private information about takeover negotia-

tions through alternative prior specifications. This option may be valuable for practitioners and

regulators involved in ongoing merger negotiations.

There has been a substantial stream of research devoted to uncoveringinformation from market

prices during takeover negotiations. To distinguish my approach from the earlier works,I include

a brief summary of them. Brown and Raymond (1986) first propose a no arbitrage argument for

estimating the probability of merger success for cash offers. Their method relies on the identifying

assumption that the target’s standalone value would revert to its premerger state. Since a cash

offer made the target’s merged value predetermined, they are able to solve for the probability of

merger success as a function of the current price. They find this measure has statistical power

in predicting successful takeovers, but cannot make claims about value creation. Barone-Adesi