Brazil, 2018

Brazil has voted today in presidential, congressional, and state elections (governors and assemblies). The far-right candidate, Jair Bolsonaro, has obtained 46.8% of the presidential vote. The runner-up, Fernando Haddad of the Worker’s Party (PT), is on 28.3%. Given that the leading candidate did not win a majority, there must be a runoff. However, as we know, it is very rare for a first-round candidate with over 45% of the votes to lose the second round, and less likely still when the opponent is so far back.

As results come in for Chamber of Deputies, Senate, and state contests, I hope readers will add detail in the comments.

And if anyone has serious basis for hope that Bolsonaro can be defeated in the runoff, please tell. Because the idea of his being President of Brazil is just too depressing to contemplate. Then again, it seems to be both the hemisphere and the era of too-depressing-to-contemplate presidents.

25 thoughts on “Brazil, 2018

  1. The first thing that came to my mind when coming across the Brazilian presidential election results was Portugal’s 1986 presidential election, in which the staunchly conservative Diogo Freitas do Amaral seemed set to prevail in a runoff election after having emerged from the first round of voting with 46% of the vote, well ahead of Socialist Party leader Mario Soares, who arrived a distant second with just 25%. To further complicate matters, the third placed candidate, Francisco Salgado Zenha, who polled 21%, had been a close ally of Soares prior to having a nasty falling out with him earlier that decade; given the ongoing feud between the two, it seemed very unlikely he’d back Soares in the runoff. Soares, however, remained undaunted, and secured the support of fourth-placed candidate and former Prime Minister Maria de Lourdes Pintasilgo (who had won seven percent of the vote in the first round). Then – and contrary to earlier expectations – he won over Salgado Zenha’s left-wing supporters, who in the hour of truth found Freitas do Amaral’s right-wing postures too hard to stomach. With their backing, Soares went on to achieve the seemingly impossible and win the runoff vote over Freitas do Amaral by a narrow 51%-to-49%.

    Will Fernando Haddad be able pull a similar feat in three weeks? I honestly don’t know, but I think it’s too early too rule out that possibility.

    P.S. the latest results with 99.8% of the vote tallied have Bolsonaro down to 46.1% and Haddad up to 29.2%.

    • The impression I get from your comment is that the other three Portuguese candidates were left-aligned, whereas in Brazil about 8.6% of the vote appears to have gone to right-leaning non-Bolsenaro candidates, which would seem to make Haddad’s task much harder

      • Not only “left-aligned”, but they were both further left than Soares (Zenha was supported by the Communists, and Pintassilgo was a kind of “new left”/”participatory democracy” candidate); in contrast, in Brazil, even some candidates that can be considered “left-wing” are probably to the right of Haddad, while Soraes was the more centrist of the left-wing candidates (until some months ago, he was the prime minister of a government who put practice strong austerity measures backed by the IMF, and in the campaign he was even physical assaulted by Communist sympathizers – who probably voted in him in the second turn…)

      • I was going to say that perhaps someone from Portugal (or more familiar with Portuguese politics) could clarify where former PM Pintasilgo stood ideologically in 1986: my vague recollection was that she was at the very least in the vicinity of the left, but it’s been thirty-two years and I no longer have the issues of Spanish magazine Cambio 16 featuring extensive coverage of that election. At any rate, Miguel has now clarified that matter.

        That said, this is a good moment to point out a very important difference between that vote and yesterday’s election in Brazil, namely the more than ten million blank or invalid ballots, which constitute 8.8% of the total number of ballots cast (and match the entire present-day population of Portugal), but are not reflected in the candidate percentages, as these are based solely on valid votes. In the last four presidential elections the blank/invalid vote has been noticeably smaller in the runoff election than in the first round – there were no runoffs in 1994 or 1998 – with an average drop of three million between rounds. However, I have no idea in which direction they go, except for the fact that in 2014 the percentage change in the blank vote only, over the total number of registered electors, had at the state level a fairly high correlation (0.80) with the abstention rate change (that is, the larger the increase in the abstention rate between rounds, the smaller the change in the percentage of blank ballots); in that year the blank vote alone dropped by nearly 2.5 million between the first round and the runoff. However, the correlation between the percentage change in the invalid vote (which fell by just under 1.5 million) and the abstention rate change was negligible.

    • Yes, 1986 Portugal is always the case I think of when someone misses a first-round victory by a few percentage points, despite such a big lead.

      My first thought when I read Manuel’s comment was, “Haddad is no Soares”. My second thought was that Bolsonaro is not as isolated on the right as Freitas do Amaral was.

      • One historical account notes that few polls gave Soares a better than even chance of prevailing in the runoff, Freitas do Amaral’s ideological isolation notwithstanding. In fact, in the aftermath of the first round, it appeared very unlikely that Soares would manage to line up the far left behind him, not least because they vehemently detested him and his Socialist Party.

        There’s also the case of Peru’s 2016 presidential election, in which PPK ultimately prevailed despite arriving a distant second to Keiko Fujimori. To be certain, the latter’s share of the vote in the first round (just under forty percent) was quite a bit lower than Bolsonaro’s showing on Sunday, but as in Portugal’s case thirty years earlier, PPK didn’t have at first a clear path to victory, not least because of the strong third-place showing of the leftist Broad Front. And then there’s Carlos Alvarado’s unexpectedly large victory in Costa Rica general election earlier this year, despite polls showing a substantial lead for Fabricio Alvarado, who had topped the poll in the first round.

        To be certain, Haddad is no Carlos Alvarado, and most certainly he’s no PPK. But like both men, he’s facing a seemingly uphill battle against a highly polarizing, right-wing populist rival. However – and without losing sight of the vast differences between Brazil and Peru or Costa Rica – in light of what ultimately came to pass in those two countries, I’m of the view he should not be written off yet. That said, ever since Lula was prevented from running I’ve been of the view that a Bolsonaro victory could not be written off either, and the outcome of Sunday’s vote has certainly reinforced that perspective. In fact, he would have been elected outright on Sunday had it not been for his dismal showing in Brazil’s northeastern states, where Haddad won an absolute majority, beating Bolsonaro by nearly two-to-one.

        Finally, my website’s Brazil page now has federal- and state-level results of Sunday’s first round voting.

  2. Bolsenaro’s Social Liberal Party has secured fifty seats in the 513-member Chamber of Deputies. This is obviously very far short of a majority, and unlike Rousseff in 2014, who had a PMDB member on the ticket, Bolsenaro doesn’t seem to have included any congressional allies in his presidential campaign. Luckily, Latin American presidents have always dealt effectively and cooperatively with unfriendly legislatures, and given Bolsenaro’s calm temperament and staunch commitment to democratic principles I can’t foresee any problems in the event that he wins.

  3. Also, somehow Marina Silva’s Sustainability Network won only one seat in the 513-member Chamber of Deputies but five seats in the much smaller Senate? Does anyone know how that might happen?

      • Yes, I suppose it is somehow due to alliances, which can be different in each house (and often need to be, given plurality election of senators). But it still seems like a pretty extreme case! I wonder if Sliva’s party wasted votes in Chamber alliances by not having popular enough candidates to make it near the top of their alliance lists.

      • Three of the five REDE Senate seats were won by party alliances, while the other two were outright victories. I’m now going over the legislative election results available on TSE’s website, which unfortunately appear to have only candidate totals, and no party vote totals at all.

      • The reason behind the noticeably different outcomes for REDE in the Chamber of Deputies and the Senate is that in the former the party won just 0.8% of the votes cast for candidates, while in the Senate their candidates received 4.2%. Interestingly enough, in the two states where REDE won Senate seats on its own – Rio Grande do Norte and Sergipe – the party fared poorly in the Chamber vote. Meanwhile, REDE’s single Chamber seat came from Roraima state, where it won 4.2% of the vote but secured one mandate (out of eight) in an alliance with PTB, PT and PV, which polled 9.8%.

        As previously noted, the results currently published by TSE do not list party vote totals. Moreover, the published figures do not include a breakdown of the votes cast for parties only in the Chamber of Deputies – for lower house and state legislative elections, Brazilian voters may vote for either a party or a specific candidate – although the total number of such votes is included in the results. While most voters chose a candidate, 6.8% voted for a party; in Roraima state the corresponding figure was 3.9%, the lowest in all of Brazil. In the 2014 federal election – where the party-only vote stood at a slightly higher 8.4% of the total – the nationwide differences between Chamber party percentages calculated on the basis of all votes cast and those obtained from aggregating candidate vote totals reached up to 1.1% for the three largest parties, but stood at 0.1% or less for parties polling fewer than four percent of the vote.

        At any rate, on the basis of candidate votes, Bolsonaro’s PSL emerged as Brazil’s largest party with 11.4%, followed by PT with 9.7% (although the latter won four more Chamber seats than the former); no other party won as much as six percent, but there will be thirty parties with at least one seat in the Chamber of Deputies. In short, despite Bolsonaro’s unexpectedly strong performance in the presidential election, Brazil’s party system remains as atomized as ever, if not more so.

      • I was able to obtain full coalition-level Chamber of Deputies results for Sao Paulo from the state’s Regional Election Tribunal website, and while going over the numbers I came across a rather unusual result. In the state, Bolsonaro’s PSL topped the poll with 20.9% of the vote, but despite outpolling the PSDB / PSD / DEM / PP coalition, which won 19.6%, the latter obtained 17 of the state’s 70 seats, while PSL won just ten. Meanwhile, the PT / PC do B coalition won 9 seats, even though it polled just 10.2% of the vote, that is less than half of PSL’s vote but nearly as many seats.

        At first glance, this appeared to be at odds with the proportional distribution of Chamber seats among coalitions and single parties, but it happened for a reason: a 2015 electoral reform now requires Chamber of Deputies candidates to poll a number of votes no fewer than 10% of their state’s Hare quota in order to win a seat. As it happened, this so-called performance clause capped PSL to ten seats, even though the party would have been otherwise entitled to 17 mandates. Meanwhile, the party’s forfeited seats were proportionally distributed among the remaining parties and coalitions.

        Although Chamber of Deputies seats in Brazil are allocated by the D’Hondt rule at the state level, the procedure used there calls for an initial allocation by Hare quota and the subsequent distribution of any remaining seats by the largest average method. Under the revised rules, PSL’s initial allocation of 14 seats by Hare quota was capped to ten, and separate counters were kept during the distribution of the remaining seats, one set for the number of mandates the parties and coalitions would have normally received without the performance clause, and another for the number they were actually assigned in accordance with said clause. However, I obtained the exact same result by carrying out a regular D’Hondt distribution; determining that PSL was to be capped to ten seats (down from 17); and then repeating the process without PSL’s ten seats or votes, that is for sixty seats among the remaining parties.

      • Hi Manuel,
        This “10% of quota” threshold seems to offer many of the same pitfalls and incentives as SNTV in Japan and Taiwan. A party needs to engage in vote-management for individual candidates and can’t just rely on pooling votes for the list to distribute its support among its nominees for optimal effect. Is this reading correct?

      • Tom, that would be correct. The problem with the PSL ticket in Sao Paulo was that nearly three-quarters of the votes cast for its candidates went to the top two vote getters, with almost half going to the top candidate alone, Eduardo Bolsonaro (son of party leader and presidential candidate Jair Bolsonaro). From what I’ve read, the party and its backers were elated about the fact he had been elected with the highest ever vote total for a Chamber of Deputies candidate; however, I’m quite certain they must have been none too pleased about the fact that his piling up a needlessly large vote total cost the party seven seats, which allowed PT to remain the largest single party in the Chamber of Deputies. But that’s precisely the intent of the performance clause, namely preventing the election of candidates who poll few votes but ride on the coattails of popular nominees.

  4. Here is the Glen Greenwald article on the results:

    Greenwald is probably looking for a country to relocate to. That wasn’t snark by the way, he has legitimate concerns for his personal safety under the upcoming regime.

    One thing that struck me from the article was the huge gap between the latest polls, plus past election results, and the published election results. How likely was electoral fraud a factor? Greenwald doesn’t consider this, but I think there is a prima facie case given the gap between the polls and the results.

    • What suggestions I’ve seen about possible electoral fraud in Brazil’s Sunday vote run in the opposite direction, namely that Bolsonaro was fraudulently deprived of a first round outright victory. At any rate, it seems to me that the outcome could be seen as the continuation of Bolsonaro’s rising numbers in later polls.

      • Yeah, it would seem rather unusual for a non-incumbent like Bolsenaro to be able to commit large-scale and otherwise seemingly undetected electoral fraud. Ockham’s razor would suggest that the polls were just wrong.

      • Brazilian polls proved to be flat out wrong in some legislative and state races: perhaps there was a shy Bolsonaro/PSL voter effect in play. And things like that lead one to wonder if there could be a shy GOP voter in play too for the upcoming U.S. mid-terms…

    • Brazil has a very insecure electronic voting system. While Bolsonaro may not be in control of the government, those who do have control are not precisely famous for their devotion to the rule of law or clean government. The supreme electoral court has refused any public testing of the electronic voting system since major flaws were exposed in 2012. Naturally the flaws were not addressed. There is an unfortunate worldwide tendency for election management bodies to both not understand electronic voting issues and insist that they do.

      In one famous public test of electronic voting, a team of University of Michigan University security researchers caused the tamper-proof machines to offer the voter a chance to play PacMan. On another the same team caused the tamper proof voting machines to play the University of Michigan football song every time a test vote was cast. More serious evidence from that team is here.

      The DRE machines used in Brazil provide literally no way to verify or validate an election.

  5. Pingback: Brazil’s open list is (a little bit of) a hybrid now | Fruits and Votes

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.