Professor of Strategy and Entrepreneurship
London Business School

Are overconfident CEOs born or made?

Most acquisitions fail. That’s not even a point of debate or opinion anymore; the evidence from ample, solid academic research is quite overwhelming: about 70% of acquisitions destroy value, and this has been the case for many, many decades.

The question is, of course, what causes acquisitions to fail, and what causes managers to undertake them in spite of their rather dismal track record? Various complementary explanations have been offered but, remember, “failure” here simply means that the acquiring company does not create sufficient extra value out of the acquisition to recoup its (usually rather hefty) acquisition premium. One prominent explanation is that the average CEO suffers from “hubris”, or “overconfidence”. They think they will be able to create more value through the acquired company than these silly people who are currently running the show, because of “synergies” or simply because they’re much better and smarter than the sorry souls who are currently messing about in that block of bricks they call a firm.

Therefore they’re willing to pay an acquisition premium. Yet, it’s apparent that usually they are overestimating their abilities, because the average CEO/acquisition does not create any surplus value – quite the contrary. Fact is (assuming that managers are well-intended and do expect to create value through their acquisitions; some people even disagree with this assumption), on the whole one can only conclude that most of them are overconfident because in 70% of the cases they don’t manage to pull it off.

But where does their overconfidence come from? Does the average CEO suffer from hubris because that’s the type of person that makes it to the corporate top? That’s one possibility. The other one is that, over the course of their tenure, often top managers gradually become overconfident, rather than that they’re suffering from hubris from the get-go.

Professors Matthew Billett and Yiming Qian from the University of Iowa examined this exact issue, using a sample of 2,487 American CEOs undertaking a combined 3,795 deals over the period 1980-2002, and they found some very compelling evidence that overconfident CEOs are made and not born that way.

They initially uncovered four things. 1) They discovered that CEOs’ first deals, on average, did not destroy value: Their effect on a company’s market value was pretty much zero, 2) those CEOs who had experienced a negative stock market effect in response to their first acquisition usually lost their appetite for doing any more deals, 3) in contrast, those CEOs who – hurrah! – had experienced a positive stock market response to their first take-over got the hots for deal-making; they were very likely to undertake even more acquisitions in the ensuing years, 4) those subsequent deals, however – that is, take-overs by CEOs who had done some before – on average did destroy shareholder value! Hence, the consistent finding in academic research that acquisitions destroy value seems to be caused by CEOs’ later deals only. Matthew and Yiming concluded that first-time, successful deals make CEOs overconfident, which not only stimulates them to do even more deals, but also makes them inclined to pay even heftier take-over premiums for subsequent ones, which they usually are unable to recoup after the acquisition.

Finally, they also examined “insider-trading”; whether CEOs would purchase their own company’s stock in the period preceding the acquisition (confident that they would increase in value as a result of the deal). Most CEOs did, whether they were first time deal makers or experienced acquirers. However, the effect for experienced acquirers (people who had done deals before) was twice as big as for the novices. Apparently, overconfident serial acquirers – who mostly ended up destroying shareholder value – most of the time fell into their own hole; they bought the shares whose value they were about to destroy! Guess there’s a hint of justice in this story after all…

Leave a Comment