Home > News > Would “Democracy Day” Boost Voter Turnout in the U.S.? Some New Evidence
90 views 45 sec 0 Comment

Would “Democracy Day” Boost Voter Turnout in the U.S.? Some New Evidence

- March 20, 2009

What, if anything, should be done about the level of voter turnout in this country? Many different approaces have been suggested, but one that keeps coming back is designating Election Day as a national holiday, i.e., “Democracy Day.”

In a new working paper, Henry Farber of Princeton University analyzes data on voter turnout by state government employees in states that either have an official “Democracy Day” or don’t, with non-state employees serving as a control group. Thirteen states have such holidays.

The highlights:

* Average turnout is statistically significantly higher (p<.10) in election holiday states for all groups of workers except federal government employees and the non-employed in 2006. * "If the election holiday policy was not related to other factors affecting turnout in a state, then the policy should affect only voter turnout for state employees. It is clear that this is not the case. While there is a significant difference between voter turnout among state employees in states with and without the election day holiday, there is also a significant positive relationship of voter turnout with the election day holiday for other classes of individuals. Of course, this cannot be the direct result of the eleciton holiday policy because it applies only to state employees. It may be that states with such a policy for state employees are states where other individuals value higher voter turnout." * "There is no significant effect of the eleciton holiday on the turnout rate in either 2004 or 2006 of state workers relative to those not employed, private sector workers, or local government workers." * Farber goes on to institute statistical controls for an array of other factors, but these follow-ups do little to dislodge the results summarized above. His conclusion? "Having an eleciton holiday, by itself, is not an effective strategy to increase voter turnout." Others who follow research on voter turnout more closely than I do may wish to weigh in on Farber's findings and conclusions. [Hat tip to Bob Goldfarb]

Topics on this page