There is a new book on election reforms for increasing voter participation that will be available soon, “American Voter Turnout: An Institutional Perspective”, by David Lee Hill (a professor at Valdosa State University). The description of the book from the publisher summarizes the book well:

Using a combination of existing and original research, this new text provides a simple explanation for the low turnout in American elections: rather than creating an environment conducive to participation, the institutional arrangements that govern structure participation, representation, and actual governance in the United States create an environment that discourages widespread participation. To explore this argument, the author examines the origins and development of registration laws, single-member districts, such as the Electoral College, and the separation of powers and the impact these institutions have on turnout levels in American national elections. To this end, the text employs a narrative discussing the impact of institutions on turnout in the United States and across nations, supported with extensive yet accessible data analysis. Hill not only provides students with explanations for the low turnout characteristic of American elections, but also demonstrates the powerful impact of institutions on political life.

I’ve had the opportunity to review an advance uncorrected proof of this new book, and have found it a straightforward and accessible treatment of the available research on the various “institutional” reforms that have either been used or have been discussed in recent years to help increase voter participation in the United States. By “institutional”, political scientists typically mean procedural or regulatory reforms — for example, the types of changes recommended long ago by political scientists Ray Wolfinger and Steven Rosenstone in their seminal book, “Who Votes”: reforms seeking to mitigate or eliminate some of the common hurdles for easy voter registration processes (some of the Wolfinger-Rosenstone ideas were incorporated into NVRA in 1993, and have continued to be important ideas behind many attempts to make the registration process easier for eligible citizens).

After covering the available literature, and providing some analysis of his own, Hill concludes by making two different types of recommended reforms: changes that seek to increase access, and changes that seek to alter the basic nature of the electoral process in the United States. As for improving access to the electoral process, Hill recommends implementation of election day voter registration, voting on weekends or making federal elections federal holidays, expansion of by-mail and early voting, and the examination of Internet voting. When it comes to improving the nature of the electoral process, Hill advocates proportional allocation of electoral votes in presidential elections and public financing of campaigns.

Generally, this is a very approachable book and it provides a straightforward introduction to the research perspective and literature that political scientists have developed in the past few decades to try to better understand why many or most American citizens do not participate in the electoral process. Hill’s book is one that might be useful for undergraduate classes, or for election reformers and election administrators who are interested in learning more about the academic research in this area.

The qualms I have about Hill’s book are beyond the reach of this particular publication. For one, and this is a generic problem for academic researchers facing long delays between the research and writing phases, and the eventual publication of our work, is that Hill has not been able to examine and include some of the very recent research on some of the important reforms he is advocating. This includes much of the research done in the past two years on election day voter registration, on early voting, “supercenter” voting, and Internet voting.

Secondly, research typically focuses exclusively on particular reforms, and then tries to extrapolate how the specific reform would increase voter participation (say how election day voter registration itself would increase voter turnout). But note that Hill has a broader objective here — he recommends a sweeping set of reforms, based on independent examination of each reform’s possible effect on voter turnout. What we don’t know is how these reforms might interact; for example, is it necessarily the case that implementing weekend voting and expanding voting-by-mail nationally would together increase turnout? Would both together increase turnout even higher than either might if it were implemented across the nation by itself? Or might their joint implementation have some unexpected or unanticipated effect, that might even act to reduce participation? This leads to an interesting problem for researchers who study election reform and political participation: what is the effect of multiple reforms on participation, and do some reforms have a positive and interactive effect on voter participation. In other words, if we pair up different reforms, do we find that they produce a positive effect on participation above and beyond the effect each reform might provide if implemented alone?

Third, there is not much discussion in Hill’s book of possible costs associated with these reforms, or the potential problems that might arise were they implemented. Accessibility is obviously an important goal for election reformers, but we must also keep in mind other important goals like privacy, accuracy, integrity and security. As Thad and I have written about in “Point, Click and Vote”, balancing these various goals with respect to the possibility of Internet voting is a complicated proposition (a point that we are returning to in our new book manuscript on the electronic voting debate, which we hope will be published sometime in the latter half of 2006). Trying to achieve all these goals simultaneously is difficult, but we should examine election reforms on all dimensions as we work to understand the broader ramifications of election reforms.

And finally, we are getting into the territory I wrote about recently, regarding the King-Zeng paper on extreme counterfactuals. For the types of reforms that Hall writes about in his book, we don’t have a lot of data on which we can base inferences about the reform’s effect on voter participation. We certainly have even less data if we are going to consider the effects of multiple reforms on turnout. But if someone tries to examine the interactive effects of election reforms, they will need to be careful about the extreme counterfactual problem.