Via Greg Mankiw, Brian Knight and Nathan Schiff of Brown University find that...voters in early primary states are disproportionately influential!

Two Brown University economists have, for the first time, quantified the substantial effects of winning early in the race for the presidential nomination. In a National Bureau of Economic Research working paper, Brian Knight and graduate student Nathan Schiff demonstrate that voters in early primary states such as Iowa and New Hampshire have up to 20 times the influence of voters in later states in the selection of candidates.

Not being an economist, I'm in no position to question their result, but my first instinct is that this figure seems way too low. In 2004, at least, given that all the major candidates besides John Kerry dropped out after Super Tuesday, didn't all the subsequent states have essentially no influence on the process? Shouldn't the number be a little closer to, you know, infinity?

--Josh Patashnik