Mergers fail more often than marriages.” So ran a headline on CNN’s website a few years back, accompanying a story referring to a 2004 study by Bain & Company, which found that 70 percent of mergers failed to increase shareholder value.
This well-worn misconception is widespread among the financial media, fueled by high-profile incomplete deals such as MCI WorldCom–Sprint, Comcast–Walt Disney and GE–Honeywell. Late last year, John Cassidy wrote in the New Yorker about Dow Chemical and DuPont’s proposed merger, citing a study I did of mergers in the 1980s, which failed to find consistent evidence of improved performance or productivity gains.
But the notion that most merger deals are doomed to failure is a canard. When it comes to creating value, most mergers succeed.
One reason for the persistence of the popular view that mergers fail is that there is no consensus on how to measure failure or success. When a 2014 McKinsey & Company article suggested that big mergers of pharmaceutical companies were value generators, John LaMattina, the former head of R&D at Pfizer, retorted that such mergers promote short-termism in an industry that requires long-term strategic thinking, and divert employees’ attention from science to office politics. If you have experienced a merger from the inside, you may sympathize.
The cleanest method of settling the debate is by using data on stock returns. Changes in operating income, return on assets, or similar gauges of operating performance are very hard to measure, since it is nigh impossible to say with any certainty at all what would have happened had the merger not gone ahead. It is little surprise, then, that the results of studies using those measures are equivocal, and there is no clear-cut evidence that acquisitions tend to lead to accounting-based or productivity-based improvements.
A second reason the prevailing narrative about merger failure has survived is that financial reporters tend to take a one-dimensional view of value, looking only at the effect of mergers on the stock value of the bidding company, and overlooking the effect on the acquired company. Often, the buying company’s share price falls on the announcement of a merger deal, while that of the target company rises, suggesting that the purchaser has overpaid, a development often popularly interpreted as meaning that the merger is a bad deal.
To be fair, much depends on the question being asked. If we are interested in whether the management of the acquiring company should have done the deal, our main concern will be what happens to the bidder. If, however, we care more about whether the merger is good for the economy or for society at large, the correct focus should be the combined value the merger creates or destroys.
When we look at the overall picture in the data, a quite different story emerges from the popular merger myth.
A highly simplified example illustrates why the analysis of mergers is often misdirected. Two companies, A and B, are worth $10 billion each. If A buys B, A will be able to cut $2 billion of costs out of the combined company. A goes ahead and buys B, agreeing to pay $15 billion. When the deal is announced, B’s value will increase by $5 billion, from $10 billion to $15 billion, while A’s value will decline by $3 billion from $10 billion to $7 billion. Why the $3 billion decline? Because A is paying $15 billion for assets that will be worth $12 billion ($10 billion plus $2 billion in cost cuts). From the perspective of A’s shareholders, executives, and consultants, A has made a bad acquisition, destroying $3 billion. However, from the perspective of all shareholders, this is a very good acquisition. The combined value of A and B has increased from $20 billion ($10 billion + $10 billion) to $22 billion ($7 billion + $15 billion).
When we look at the overall picture in the data, accounting for the combined returns to both bidders and sellers, a quite different story emerges from the popular merger myth. At the times when merger deals are announced, the combined returns are usually positive both statistically and economically. On average, the overall value of both acquirer and acquired increases, which indicates that the market believes the announced deals will create value. This has been the case for the average acquisition going back 30 years, and it remains the case today. If combined returns are positive, mergers certainly create value for the overall market, and, therefore, for investors in index funds.
From there on, however, the story gets more complex. In the longer run (in which the acquired company disappears as the merger is completed), the value of acquiring companies tends to go up in all-cash deals. Add that to the value created at the announcement of a merger, and there is significant value created.
But the value of the acquirer tends to go down in the longer run for stock deals, as was the case, for example, when AOL announced a $164 billion all-stock takeover of Time Warner in January 2000, and in a number of other large stock deals around the dot-com boom.
Stock deals are complicated by the fact that they supply investors with two important pieces of information. The first is straightforward: about the merger itself, and whether the acquisition makes sense for the buyer. The second concerns what the executives of the acquiring company think about the value of the buying company’s stock. If the acquirer is prepared to give the owners of the target company a lot of stock to complete the deal, it is effectively telling the world that it believes its stock to be overvalued—or, at the very least, that it does not think its stock is undervalued, since if it were undervalued, it would not be so willing to give it away.
Nevertheless, the average all-stock deal still creates value overall at the time of the announcement. Thereafter, however, there tends to be a downward drift. Because the value of an all-stock deal is a combination of how the market values the takeover itself and how it values the acquiring company, it is impossible to say whether this loss in value is because of the deal itself or because the buyer was thought to be overvalued in the first place.
In general, then, cash mergers tend to create value in both the short and longer run, while stock deals only do so in the short run. (That may reflect the dynamics of the early 2000s, when this research was conducted. Since then, it is possible that bidding companies have taken the lessons from prior mergers that failed to create value, and learned to do better deals.)
The stock-cash dichotomy may also help explain research that suggests that “megamergers” tend to reduce value while small-company mergers create it, as was found in a 2003 paper by University of Pittsburgh’s Sara B. Moeller and Frederik P. Schlingemann and Ohio State University’s René M. Stulz. It is often hard to secure loans to pay for very large mergers, so such deals tend to be either all-stock or combined cash-and-stock agreements.
So what kinds of deals do investors think create value? In general, the data suggest that mergers motivated by efficiency are more favored by the market than those that attempt to build a company’s market power. Strategically speaking, investors prefer takeovers that are likely to lead to cost cutting to those that entail putting different businesses together and trying to increase revenues. Dow-DuPont is an example of a cost-cutting deal, and one might reasonably expect that, if completed, it will be shown to have created value. However, the deal is so big and complicated that there are many other factors at play, and it is tricky to make any definite prediction about what will happen.
There is some evidence that the market reaction to mergers is a good indicator of the soundness of the deals. Combined acquisition announcement returns are significantly positively related to subsequent success, as outlined in a 1992 paper I coauthored with Michael S. Weisbach, then at the University of Rochester, and now at Ohio State University. We found that acquisitions later divested at a loss had substantially lower announcement returns than acquisitions divested at a gain and acquisitions that were not divested. Acquirers in deals with negative returns are subsequently significantly more likely to receive hostile takeover offers, according to research by Mark L. Mitchell of AQR Funds and University of Pittsburgh’s Kenneth Lehn, while CEOs of acquiring companies in acquisitions with negative returns are significantly more likely to lose their jobs subsequent to the acquisitions, according to work by Lehn and University of Alberta’s Mengxin Zhao.
Aside from a narrow focus on bidders and the complications of stock deals, there is a third reason the popular narrative about mergers is one of failure: stories of big deals going sour are more newsworthy than mergers being successfully completed. The largest deals get the most attention, but as Moeller, Schlingemann, and Stulz show, these are also the most likely to run into problems. The most successful deals are often midsize takeovers that add 10–20 percent to the size of a company, rather than the headline-grabbing megamergers of equals. Moreover, the media often focus on the cultural problems inherent in mergers, which are undoubtedly real and challenging, but, as I discussed earlier, are difficult to measure.
The common view of corporate mergers as an exercise in repetitive failure, then, does not stand up to the scrutiny of the data. The average merger creates value. The media may believe that most mergers fail. Investors, however, do not.
Steve Kaplan is Neubauer Family Distinguished Service Professor of Entrepreneurship and Finance at Chicago Booth.