Stifel Nicolaus analyst Blair Levin has broken down the odds that Washington approves one of multiple tie-ups in play right now: Microsoft (MSFT) buys all of Yahoo (YHOO) ; Microsoft buys Yahoo search; Yahoo hands over some of search to Google (GOOG).
Blair’s bio makes him worth listening to: He was chief of staff to former FCC chair Reed Hundt for a good stretch of the Clinton administration. In a nutshell, Blair thinks:
- Microsoft buying Yahoo is not a slam dunk, but doable. It gets much harder if it’s reviewed by an Obama administration, or even a McCain administration.
- Microsoft grabbing Yahoo search is easier than an outright acquisition, but not by much; contrary to reports that the concentration of e-mail and other services would pose a problem, the real issue would be the reduction of players in search.
- Yahoo handing over some of search to Google is by far the hardest deal to through DOJ. And the more that Google and Yahoo try to construct a partnership that will pass regulatory muster, the less attractive the deal will look to Yahoo shareholders.
Blair’s logic: Google and Yahoo will construct a deal that will be non-exclusive, and only give Google a portion of Yahoo’s search business. So they’ll argue that the deal still gives Yahoo plenty of incentive to keep working on its own search product and improving it, etc — or at least selling the rest of it to other players, like Microsoft. But regulators are going to have a hard time believing that:
Rather, we believe the government would look to the likely impact of the outsourcing on pricing control and Yahoo search innovation. That is, even if the deal is nonexclusive, we believe the
government would try to discern whether the economics are such that Yahoo would end up, in fact, using Google on an exclusive basis.
And the more that Yahoo and Google try to limit the scope of the deal, Blair argues, the less sense it makes: If Yahoo is only going to hand over a small chunk of its search business and keep pouring resources into its own R&D, what’s the point?