Voters have up to 5 times more influence in early primaries

Voters in states with early primary races such as Iowa and New Hampshire have up to five times the influence of voters in later states in selecting presidential candidates, according to research by Brown University economist Brian Knight. The paper, the first to quantify the effects of early victories in the race for the presidential nomination, is co-authored by Nathan Schiff and published in The Journal of Political Economy.

Knight and Schiff developed a statistical model that examines how daily polling data responds to returns from presidential primaries. In the model, candidates can benefit from momentum effects when their performance in early states exceeds expectations. For example, Knight and Schiff found that in 2004, John Kerry benefited from surprising wins in early states and took votes away from Howard Dean, who held a strong lead prior to the beginning of the primary season. According to their research, Schiff and Knight predict that if states other than Iowa and New Hampshire had voted first in 2004, the Democratic nominee may have been John Edwards, rather than John Kerry.

"Clearly, the primary calendar plays a key role in the selection of the nominee," said Knight, associate professor of economics. "Evidence that early voters have a disproportionate influence over the selection of candidates violates 'one person-one vote' — a democratic ideal on which our nation is based."

Knight and Schiff also simulate the 2004 primary as a simultaneous national primary, which they predict would have been much tighter than Kerry's landslide victory, due to the absence of momentum effects. Additionally, the research demonstrates how this disproportionate influence of early voters affects candidates' allocation of campaign resources, as measured by advertising expenditures. They found that candidates spent a disproportionately high amount in states with early primaries. The economists conclude, "While these results are specific to the 2004 primary, we feel that they are informative more generally in the debate over the design of electoral systems in the United States and elsewhere..."

Knight's current work addresses the policy implications of this research, exploring which system is the best in terms of selecting the best candidates. The work considers whether there should be a national primary in which every state votes on the same day, the current sequential system, or possibly a hybrid system with a rotating regional primary.