Residual Votes and Abstentions in the 2016 Election

Abstract

This paper provides a preliminary analysis of the increase in the residual vote rate from 2012 to 2016, when it increased from 0.99% to 1.87% nationwide. It is reasonable to assume that this spike in the residual vote rate is due to a rise in abstentions. However, there are currently other trends in election administration, such as an increasing reliance on vote-by-mail, that could also be driving up the residual vote rate. And, even if the recent up-tick in the residual vote rate is primarily due to an increase in abstentions in 2016, it is not a priori obvious that the source of new abstentions was equally distributed among disaffected Democrats and Republicans. The analysis in this paper relies on a combination of public opinion data and election returns to address these issues. We find, first, that the increase in abstentions in 2016 was most likely due to disaffected Republicans, rather than an across-the-board phenomenon. We also confirm that the increase in the 2016 residual vote rate was not due to changes in voting technologies between 2012 and 2016. We address three issues in the conclusion that this analysis raises: (1) the potential for the growth of protest voting in the U.S., (2) the likelihood that there is a significant under-reporting of voter abstentions in public opinion surveys, leaving a role for aggregate analysis to study this phenomenon, and (3) cautions about the use of the residual vote rate as a metric to gauge the accuracy of voting technologies.

Keywords: residual vote rate, elections, 2016 election, lost votes, nevada