This time the polls in Brazil were right. Former president Luiz Inácio Lula da Silva has scored just the narrow win that they predicted (see my preview here): he won with 50.9% of the vote, a lead of 2.14 million votes over incumbent Jair Bolsonaro. Official results are available here.)
It’s as good an illustration as you could want of the evil effects of the electoral college. Two years ago, Joe Biden beat Donald Trump by a considerably larger margin – 52.3% of the two-party vote, a lead of just over seven million – but because the result depended on individual states it was not declared for three days. That period of doubt gave Trump the opportunity to sow distrust in the result, with destructive consequences.
No such problem in Brazil. It’s a single nationwide ballot, and counting proceeded quickly and smoothly. Bolsonaro led early on, as expected, since his strongest areas report first, but with about half the vote counted Lula took the lead and never relinquished it. At the time of writing (about 11.30pm Brazilian time) Bolsonaro is yet to concede defeat, but the electoral authority has confirmed Lula’s victory and foreign governments have been sending congratulations.
If Bolsonaro was planning a military coup, there would probably have been some sign of it by now. Instead, a sympathetic news outlet reports that he does not intend to question the result, although he also does not plan to call Lula tonight. Lula is due to take office on 1 January, and hopes are rising for a peaceful transition.
Nonetheless, Lula is unlikely to have an easy time of things. He does not have a majority in Congress; the congressional election, held in conjunction with the first round four weeks ago, gave the right the advantage in both houses, although no single party will have control. Lula’s image is much more tarnished than the first time around, and the country is more deeply divided.
It’s impossible to say whether a different centre-left candidate with less baggage would have done any better, but the late drop in Lula’s support is alarming. A year ago he led in the polls by about 15 points; in the first round it was more than five points and he was endorsed by the next two placegetters. In the circumstances, a gap of 1.8 points is much too close for comfort. Turnout was 79.4%, up just slightly on four years ago and almost identical to the first round.
The geographical pattern of the result is clear (the Guardian has a nice map): Bolsonaro’s strength is in the south, Lula’s is in the north-east. Although there are more people in the south, Lula won because he was more competitive there – 43.5% in Rio de Janeiro, 44.8% in São Paulo, 53.6% in Rio Grande do Sul – whereas Bolsonaro was getting less than a third of the vote across most of the north.
So far, Lula has said all the right things about fostering unity and governing for all Brazilians. But he will need to find ways to defuse the animosities that Bolsonaro has fostered, while curbing the authoritarian tendencies on his own side.
5 thoughts on “Lula wins it”
> It’s impossible to say whether a different centre-left candidate with less baggage would have done any better, but the late drop in Lula’s support is alarming.
I would have though that’s pretty clear cut. Bolsanaro soundly defeated a generic centre-left candidate four years ago. Lula came back because Lula is a winner.
Thanks David! I’m not entirely convinced. An awful lot has happened in Brazil in the last 4 years; I don’t think you can infer much from what happened in 2018. I think Lula was a risky choice – I’m glad it paid off, but it was too close for comfort. But you could be right; maybe no-one else would have done any better.
The appearance created by the latest reports is that Jair Bolsonaro is attempting the political equivalent of what Americans call a nolo contendere or ‘no-contest’ plea in court (an option which doesn’t exist in Australia): he’s not admitting that he lost, but he’s not going to resist the consequences of that result being declared.
Yes, that seems about right. Although Trump 2020 looked a bit like that at first.