Will Perry’s Extremism Matter?

Taibbi cries out that Perry's social security rhetoric will sink him. Bernstein differs:

[T]here is still plenty of time for Perry to blunt charges of extremism by modifying his positions and, especially, emphasis. Yes, the Obama campaign has the debate footage saved and ready to deploy. And yes, it will make no sense at all when Perry says (if he wins the nomination) that despite still believing what he’s said in the past, he remains committed to delivering every dollar of Social Security benefits to absolutely everyone who has ever paid into the system. It won’t make sense — but if the economy is awful and Obama is at 35% approval, it won’t matter because no one will be listening to the president any more.

I don't agree. Even in a steep recession, an Obama-Perry fight is much likelier to end in an Obama victory. No one wants to return to Bushism. Very few want to abolish social security and Medicare. Gelman looks beyond the horse-race:

[T]he evidence is that political extremism might cost a presidential candidate about 1 or 2 percentage points of the vote. (Political scientist Steven Rosenstone wrote about this in his 1983 book, Forecasting Presidential Elections, where he argued that, contrary to what some disaffected Democrats wanted to believe, Ted Kennedy would’ve done even worse than Jimmy Carter in the 1980 general election.) But my second point is that, yes, it could matter a lot who the Republican nominee is in 2012. The new president can have a big impact on all sorts of policies. In his phrasing, Tucker was making the common slip of assuming that all that matters is which party wins the election and forgetting about policy after the election has been decided.