That aggregate loss is not evenly spread. The authors find it is driven by a small number of acquisitions with negative synergy gains made by firms carrying extremely high valuations. Strip those deals out and the wealth of acquiring shareholders would have increased. So the honest reading is not that acquisitions destroy value, but that a handful of very large deals done by very highly valued companies destroyed enough value to swamp everything else in the average.
That distinction should change how a private owner receives the statistic. You are not a richly valued public company using inflated paper to buy something enormous, which is the profile the losses concentrate in. The mechanism the paper describes, where a high valuation gives management the room to make a poor acquisition, is a specific condition rather than a universal law. The useful question is whether any of that mechanism applies to you: are you buying because the opportunity is good, or because your own position currently makes it easy?
Read directly against Fuller, Netter and Stegemoller on private targets, and against the McKinsey argument for programmatic acquisition, both on this shelf. Those works describe smaller, repeated, privately negotiated deals; this one describes the opposite end of the market. The shelf holds both because the answer to "do acquisitions work" depends almost entirely on which kind of acquisition is meant.