Print
E-MailWenyu Wang is Assistant Professor of Finance at Indiana University Kelley School of Business. This post is based on his recent article, published in the Journal of Financial Economics.
A large body of research documents that market reactions to takeover announcements are, on average, neutral or even slightly negative for acquirers. This evidence presents a value-creation puzzle: Why do acquirers pursue takeovers if they do not overtly benefit from the deals? The puzzle seems to contradict both the common assumption that acquirers are value-maximizers and the neoclassical theory of mergers and acquisitions (M&As).
The main goal of my article, Bid Anticipation, Information Revelation, and Merger Gains, recently published in the Journal of Financial Economics, is to reconcile the value-creation puzzle with the neoclassical theory of M&As. I evaluate how the market’s anticipation of future takeovers (i.e., the anticipation effect) and the new information revealed upon bid announcements (i.e., the revelation effect) confound traditional estimates of merger gains and lead to misinterpretation of announcement returns.
My empirical methodology is structural estimation, which is an attempt to fit an economic model to the data, assess how well the model fits the data, and measure fundamental economic parameters. This methodology is common in certain fields of economics, such as industrial organization, and it is becoming more popular in corporate finance.
My structural model formalizes the anticipation and revelation effects as the market’s reaction to firms’ takeover decisions. In the model, firms produce using a pair of complementary assets (e.g., a firm’s productivity and its growth option) and these assets can be reallocated through M&As. For example, a firm with high productivity but low growth option may want to acquire another firm with high growth option but low productivity. When the market evaluates each firm, it takes into account the probability of the firm acquiring or merging with another in the future, and therefore the firm’s pre-merger market value captures the anticipation effect.
The revelation effect emerges as the market’s reaction to a self-selection problem induced by firms’ takeover decisions. Firms decide to acquire, be acquired, or remain independent in response to shocks to their complementary assets, which are not observed by the market. As a result, when a firm announces that it is acquiring or merging with another firm, the market learns something new about the firm’s fundamentals and reassesses its stand-alone value. A significantly positive shock to a firm’s productivity, for example, may induce the firm to expand quickly through acquisitions in order to accommodate growth. A significantly negative shock to a firm’s growth prospect may compel the firm to restore its competitiveness through acquisitions in order to keep up with its peers. Since the realizations of such shocks are unobservable to the market, the market conjectures the underlying takeover motives based on its own information set. A positive (or negative) revelation effect occurs when the market perceives that a positive (negative) shock to productivity (growth prospect) is more likely to have occurred for a given acquirer.
I estimate the model using U.S. M&A transaction data from 1980 to 2012. The model closely matches the observed patterns in the data, including the persistence and variance of firm profitability, the probability of bid incidence, characteristics of acquirers and targets, the average and dispersion of cross-sectional announcement returns, market reactions to bid terminations, and the low predictability of acquirers and targets.
The estimates indicate five main findings. First, M&As on average create significant value for both acquirers and targets. Second, the revelation effect is economically sizable and clouds the traditional interpretation of acquirers’ announcement returns. Specifically, I estimate the revelation effect to average about -5%. Though acquirers are estimated to gain 4% from a typical merger, the revelation effect brings down their average announcement returns to -1%, creating the misleading impression that acquirers gain little from M&As. Third, the merger gains estimated from acquirers’ announcement returns do not capture the total value of an active merger market. Using a model simulation to perform a counterfactual analysis in which no M&As are allowed, I estimate that the value of an active merger market is about 13% for acquirers, part of which is capitalized in their pre-merger market value because of the anticipation effect. Fourth, a variance decomposition of acquirer announcement returns shows that only 30% of the variation is attributable to the uncertainty in acquirers’ merger gains and that the revelation effect explains most of the cross sectional variation in acquirer announcement returns. Fifth, the observed correlation between announcement returns and firm characteristics does not necessarily imply that certain types of firms are more likely to lose in takeovers. Firms with certain characteristics (e.g., large firms) are simply more likely to pursue takeovers associated with negative revelation effects.
Overall, the structural estimation results suggest that a model with value-maximizing firms can explain, both in magnitude and in patterns, the observed empirical facts that were previously perceived to support the value-creation puzzle. The anticipation effect and the revelation effect are large enough to account for the significant discrepancy between a sizable merger gain and a negative acquirer announcement return. The unified model also consistently explains other stylized facts that are prevalent in the data. The main results are robust to including agency costs and hubris motives of M&As in the model.
The full article is available for download here.
