by Eric Magar (Spanish version)
Not as well as winning the presidential race would suggest, inspection of aggregate data reveals.
First, the PRI failed to win the chamber of deputies’ majority that was generally taken for granted. Many attributed this failure to faulty ballot design. Although the IFE estimated that as many as 1.6 million votes were wasted for this reason, the story of the missing majority is more complex. The PRI has great responsibility for it.
Ease to vote for the PRI without invalidating the congresional vote Common PRI-PVEM congressional candidate Yes No ------------------------------------ Same coali- | (1) easy | (2) hard | tion status Yes | 33,504 | 19,923 | in all the |----------------+------------------ state's | (3) easy | (4) very hard | districts No | 5,673 | 7,429 | ------------------------------------ No. of secciones 39,177 27,349
Combining districts with and without a PRI-Green coalition confused many. To support the coalition’s common candidates for president and Congress, the voter could mark the PRI logo on the ballot, the Green’s or both. This new way of voting was invalid, however, in the 101 districts of 19 states where the parties that have teamed in all federal elections since 2003 nominated separate candidates for federal deputies. Supporters in those districts received contradictory signals about the correct way to vote, especially in the seven states where the PRI and Green nominated common candidates in some districts but not in others (cell 4 in the matrix above).
For each electoral section (sección, a unit of aggregation not much above the voting precinct), figure F:epnVpri reports the vote difference between Enrique Peña Nieto and the PRI’s deputy candidate. Votes cast in favor of the PRI or the PRI and PVEM are considered, but not votes cast for the Green party alone. The differential in the horizontal axis is a simple measure of the PRI candidates’ relative performance. Sections at zero had identical PRI vote tallies for both races. Relative to a normal distribution, pictured for reference, sections reveal a clear and distinct rightward shift. Sections where Peña Nieto underperformed did not offset sections where he overperformed relative to the congressional candidate in the same ticket. On average, the PRI with Peña Nieto on the ballot won 18 votes more per section than the PRI’s congressional ticket. Across more than 66 thousand sections that comprise the country as a whole, the total difference was about 1.2 million votes, or seven percent more votes for the presidential than for the congressional candidates. The figure distinguishes sections with difficulty for the voter from those without. The segment of gray-colored bars visible above black bars correspond to cell 4 sections, the most problematic subset for the voter. Red bars, in contrast, include only sections with common PRI-Green deputy candidate (cells 1 and 3), where the vote could not be invalidated by mistake. The visible portion of black bars includes cell 2 sections only. The shift is most obvious in the gray and black bars with respect to the red, suggestive of many congressional votes for the PRI unwillingly invalidated. But notice how the rightward shift relative to the reference curve is also visible in the red base. And the red includes the bulk of the electoral sections. In other words, it is highly likely that a very large number of voters who supported Peña Nieto deliberately chose not to vote for the PRI.
Second, there is more evidence in aggregate data of voters reluctant to support the PRI. While most of those who voted for Peña Nieto did so marking the PRI logo alone, nearly a one in four preferred to mark the Green party’s as well or instead. Figure F:reacios shows that this occurred across congressional districts. The figure separated districts from the state of Chiapas (hollow points) where the Green party was awarded the joint candidate in the concurrent gubernatorial election, with spillover to the presidential race. In districts with very strong Peña Nieto support, the share contributed by the PRI is about 80 percent, dropping to 70 percent as the Peña Nieto became smaller. To the extent that voting is expressive, that so many included the Green party in their vote is indication of reservations towards Peña Nieto’s party in the electorate.
Finally, PRI deputy candidates received a smaller vote in 2012 than in 2009. Given the high turnout of the general election, the absolute number of votes won by the PRI was in fact higher, but that was far from true in relative terms: 38.9 percent in 2009 vs. 31.6 in 2012, or 39.2 percent to 35.5 if the PRI-Green votes are also added. If the PRI won in 2012 it was not thanks to its strength. The weakness of the PAN and the PRD had a lot to do with it.