50%+1 < 60%

One of the key grafs from the Atlantic article about Rove is this one, which contains the explanation for why Rove’s ambitions to create a McKinley-like realignment were always doomed (leaving aside Mayhew’s entirely correct point that realignments are longer-term and more complex processes than are usually assumed):

One of the big what-ifs of his presidency is how things might have turned out had he stuck with it (education remains the one element of Rove’s realignment project that was successfully enacted). What did become clear is that Rove’s tendency, like Bush’s, is always to choose the most ambitious option in a list and then pursue it by the most aggressive means possible—an approach that generally works better in campaigns than in governing. Instead of modest bipartisanship, the administration’s preferred style of governing became something much closer to the way Rove runs campaigns: Steamroll the opposition whenever possible, and reach across the aisle only in the rare cases, like No Child Left Behind, when it is absolutely necessary. The large tax cut that Bush pursued and won on an almost party-line vote just afterward is a model of this confrontational style. Its limitations would become apparent.

It should be noted that, for the first term, the “50%+1″strategy was, in fact a very effective governing tool. Bush was very successful at getting his agenda through Congress despite his narrow “victory” precisely because he ignored vacuous invocations of “mandates” and realized that your power in domestic policy is about how many votes you can get in Congress, and simply getting the minimum necessary coalition allowed for the maximum policy gains. But this strategy is entirely incompatible with a long-term realignment, which requires adding allies rather than simply paying off existing ones. Social Security, among some other New Deal policies, worked for FDR precisely because they created the large coalition of supporters (although this meant not getting some things he wanted and making some horrible compromises with Southern Democrats.) And because of this existing constituency, privatizing social Security was never going to be broadly popular or an effective coalition-building device. Seeking the minimum possible winning coalition is never going to be compatible with engineering a major realignment, and Bush’s historically narrow victory as a wartime president with a decent economy makes clear. And even worse for Rove, 50%+1 becomes a lot less effective as the President becomes less popular, and hopeless on domestic policy when you’ve lost Congress.