I was reading about the DOJ push back on the proposed American-US Airways merger. It strikes me that you never want to be the last merger in an industry consolidation. When the consolidation begins, say with 8 players, a merger -- even if it results in a very big company -- reduces the number of competitors from 8 to 7. After a while, though, the later mergers are proposing to reduce the players from, say, 4 to 3. This will look worse to the DOJ, who by this point in a consolidation may be feeling remorseful, in retrospect, that it let some of the earlier deals go unchallenged. So the last deal gets to catch up / payback from the earlier deals.
I think this is in part what is happening with the American merger. I don't have the data, but my sense is that earlier mergers (e.g. United and Continental) were far more problematic from an anti-trust standpoint.
Disclosure: living in Phoenix, whose US Airways hub will likely get downsized or eliminated in the merger, my life will be worse likely if the merger is approved. Executives swear Phoenix will remain a major hub but most residents here consider this a "If you like your hub you can keep it" type promise.