The most important political change of the past half century is the Democrats’ and Republicans’ transformation from loose ideological coalitions to sharply distinct parties of the left and right. In Washington, the parties are now too far apart ideologically for either to count on winning support from the other side.
However, the country’s biggest problems are too large for one party to handle, at least in any consistent way. The Democrats did pass health reform on a party-line basis, a remarkable accomplishment, but they did it by the skin of their teeth and with a Senate supermajority which has evaporated. That is not a trick they can keep performing.
Under those conditions, the only way to achieve sustainable bipartisanship is to divide control of the government, forcing the parties to negotiate in order to get anything done. That pulls policy toward the center, which encourages reasonableness.
This is an argument I expect to hear a lot from centrist pundits in the coming months, but it's highly oversimplified.
Imagine that Republicans take back control of the House in November. It is true that this arrangement would force more compromise on must-pass budget and appropriations bills. However, many of the "remaining challenges" that Rauch describes as "daunting" -- "the economy (especially employment); financial reform; energy and the environment; above all, an impending fiscal train wreck" -- are likely to be more, not less, difficult to address with a Republican House.
Here's why. Under the gridlock zone model of Congress, legislative action on any issue is currently impossible when the status quo lies between the filibuster pivot in the Senate (i.e. the 41st most conservative senator) and the veto pivot (the most liberal Democrat whose vote would be needed to override an Obama veto) -- a filibuster blocks any move to the left, and an Obama veto blocks any move to the right. Here's a figure illustrating the idea from Keith Krehbiel's Pivotal Politics:
If the new median voter in the House or the new filibuster pivot in the Senate is more conservative than the current filibuster pivot, then the "gridlock zone" expands to the right, blocking action on more issues even if those proposals would move the policy status quo toward the center. The relevant change in policy is likely to be more gridlock, not more policy compromise on important issues.
Rauch also advances the silly claim that Obama would benefit politically from divided government:
In the face of those challenges, here is a two-word prescription for a successful Obama presidency: Speaker Boehner.
...To regard the prospect of a House turnover this fall as a calamity for Democrats is understandable but short-sighted. Speaker Gingrich made it possible for Bill Clinton to leave office with glowing approval ratings by allowing him to govern from the center of the country, instead of the center of his party. Speaker Boehner would do the same for Barack Obama.
But I've argued, it wasn't Newt Gingrich who "made it possible for Bill Clinton to leave office with glowing approval ratings"; it was the booming economy. Clinton's move toward the center after 1994 may have helped increase his 1996 vote totals and approval ratings, but the long economic expansion that took place during his time in office was the driving force behind his political success.
Likewise, while Obama might realize some benefit from moving toward the center, divided government isn't necessary for such a move, and it would drastically limit his ability to enact his policy agenda in Congress. It's hard to see how a GOP takeover of the House or Senate would be a net win for the President.
[Update: Added last paragraph for clarity.]