M&A Failure Rates

There are a lot of statistics out there about how often mergers (and/or acquisitions) "fail." Failure is typically defined as something like "not meeting financial goals" or "not increasing shareholder value." I've seen failure rates everywhere from about 55%-70%. This would mean that, basically, 2 out of every 3 mergers "fails."

The conventional wisdom *used* to be that they failed because of financial, legal or other reasons. Basically, there were agency issues and complexities that weren't teased out in the due diligence process, and those things caused the endeavor to fail.

In the last few years, the conventional wisdom has changed. The new convention is that mergers fail due to a lack of cultural compatibility or a failure to adapt the cultures to mesh together. Yet, the story goes, they happen anyway because of the egos of executives and because of the interminable pressure to grow. I'm sure that these reason *do* explain many M&A failures, but...

I would like to interject. Specifically, I would like to point out that there are two major flaws with the simple statement: "Most M&A activities fail."

1) Most startups fail, but that doesn't mean startups are a bad idea or are fundamentally done incorrectly. Maybe only 1 in 30 startups really makes it, but when they do make it, if they return 40X their investment, the system, as a whole, works. In fact, it works BETTER than NOT having 30 startups, lumping all that activitiy together into a single larger company, and hoping that *it* works (this is basically how Japan's economy works). The startup system allows for more flexibility, creativity, and specialization of labor.

So, if 2/3 of mergers fail, but if those that succeed garner huge profits, it might be perfectly logical that they continue to occur. Let's say Coke and Pepsi merged (which would never be allowed, but pretend it were). Maybe, if they merged, the simple economies of scale in production and distribution would allow them to eke out a 5% cost savings. Maybe this would pay for the merger efforts, or maybe it wouldn't. However, what if the reduced competition ALSO allowed them to raise prices by 20%? Then, the boon to the shareholders might be immense. Because of the enormous upside, companies might risk doing something that fails more often than it succeeds.

2) There is another systematic, failure-inducing tendency. M&A activity success is often measured by increases in returns to shareholders. But, this is wrong for 2 reasons.

The first is that comparing to the past is not necessarily relevant. What matters is how the merged entity's stock does compared to how the separate companies' stocks would have done over the same time period. Of course, we can't know this, but perhaps companies decide to merge when their industries are contracting (to achieve economies of scale) or when growth is slowing (in fact, I think this is true). If this is the case, you would see a systematic biasing of the results when comparing to rates of return before the merger activity takes place, when the entire industry/market was different. Of course, there's also the possibility that the company is acquired for "non-financial," strategic reasons, but I think mostly this is not relevant. Ultimately, all the strategic factors should affect the returns of the company's stock eventually. Again, the question is more fundamental: "What would have happened if we didn't do the merger?" NOT "How did we do after the merger compared to before?"

Second, since so many acquisitions (or mergers) are paid for with stock, another related possibility is that companies tend to pursue mergers when they know their stock is overvalued. There is no "insider trading" rule against making acquisitions when you know your stock is "high." When AOL merged with Time Warner, it *knew* its stock price was probably unjustified. So, if companies tend to buy when their stock is high, of course they are going to tend to underperform, stock return-wise, after the merger. A counter-argument to this is that the acquiree might be more likely to sell when their stock is undervalued, and that should offset this fact, but I suspect that's not true because: a) the acquiree makes up less of the total, anyway, so it wouldn't fullly offset, and b) the acquirer has "insider" knowledge of its own stock, but not of the target's, and it (the acquirer) makes the decisions (sometimes the acquiree doesn't even want to be acquired at all).

A little more research and regression analysis could probably separate out some of these factors. I hope someone, somewhere, in some grad school (other than Stanford, because poor old Jim Collins suffers from the correlation=causation disease BIG TIME), who has taken statistics (see previous note on how Stanford sucks), will do this research.