Spoiled Ballots Say Something

Honduras has just concluded its national primary elections to choose candidates to compete in the national elections for President, Congress, and Mayorships this coming November. While the official results are not expected before the end of the month, there are some disturbing trends in the voting:  spoiled and blank ballots are becoming much more common.  Spoiled ballots are ones that are defaced by political messages, or vote for more than one candidate for office where only one selection is allowed.  Blank ballots are those that contain no discernible vote.

In the low information, highly politicized elections in Latin America, spoiled ballots and blank ballots mean something.  As Driscoll and Nelson (2014) state:

"Scholars interpret blank and spoiled ballots as resulting from some combination of voter incapacity, where citizens lack the requisite skills or information to cast a valid ballot, and political motivations, when voters deliberately signal their malcontent."

Driscoll and Nelson go on to show that it's less about incapacity in their case study of Bolivian elections, and more about political intentions.  In the Bolivian 2011 election, where 60% of the votes cast were spoiled votes, they conclude that both blank and spoiled ballots are the result of political concern, but that blank ballots were more likely to be from  politically more sophisticated voters.

In Honduras, a quick examination of the Tribunal Supremo Electoral's (TSE) archive of the results of past elections shows that the number of spoiled and blank ballots in general elections oscillates between 4% and 8%.  The rate seems to increase when the candidates are more contentious, as in the election between Zelaya and Lobo Sosa in 2005, and drop when the election is less contentious as in the election between Carlos Reina and Oswaldo Ramos in 1993.

Unfortunately, the TSE archive does not preserve complete data on the primary elections before 2012.  In the  2012 primary, the TSE data show the rate of spoiled and blank votes by political party:

National Party          14.15%
Liberal Party            13.73%
Libre                          5.27%

Yet voter dissatisfaction with Party candidates dropped in the general election of 2013 to 4.88% while overall voter participation rose.

While the full and final results are not yet in, a comparison with this year's reported results to date is nonetheless instructive:

National Party          16.73%
Liberal Party            13.43%
Libre                          6.85%

Honduran press reports indicate a campaign to inflate the number of votes cast in the National Party with at least two videos surfacing showing polling place workers filling out left over ballots and stuffing them into the National Party ballot box.  Twitter was full, after the election, of images of National Party ballots marked with messages against  candidate and current President Juan Orlando Hernandez (here, and here for example). In addition, the TSE has failed to purge voter roles of the deceased, so that even the dead could vote in the primary.

All three parties have strong central administrations that control the degree to which different factions have a say in party positions.  All three parties have had the same faction of the party in control of the central administration since the 2009 coup.

Our interpetation of the data suggests that internal party dissatisfaction with the candidates available is increasing in all three parties, but it's especially notable in the National Party. As Driscoll and Nelson note in their conclusions:

"we show that these votes are intentional demonstrations of citizen dissatisfaction, signaling to elite voters disapproval of the electoral process."

To write these voters off as anomalous is to risk loosing the confidence of the voters that brought them to power. A study of where these spoiled and blank votes come from could be used by a smart political party, as a cue to where it should concentrate its efforts to consolidate Party support.