Bolsonaro's Dangerous Movement Will Outlast His Presidency

On Sunday, Brazilian voters will vote in a runoff election that will decide whether the far-right incumbent president, Jair Bolsonaro, is reelected, or if the leftist Luiz Inácio Lula da Silva, who served as Brazil’s president between 2003 and 2010, is returned to office.

That there will be a second round at all has come as a surprise to many, including in Brazil, where both many of Lula’s supporters and the pollsters had predicted that Lula would win in the first round. Prior to the election, some of the most important Brazilian polling firms, notably IPEC, Genial/Quaest, and Datafolha, had Lula defeating Bolsonaro by as many as 14 points, with support for him actually rising by several points in the week before the voting. Instead, Lula won the first round by only five points, a remarkable comeback by Bolsonaro, even in the highly volatile context of Brazilian electoral politics, and the tightest first-round vote in Brazil’s history since the return of democracy in 1985 after 21 years of military rule.

Brazilian pollsters have since justified this humiliating misreading of the electorate by invoking the so-called “embarrassed voter” theory, which holds that while voters on the left tend to be honest about whom they intend to vote for, voters on the right tend to be less so—though whether this is principally out of embarrassment about their intentions or because, encouraged by the Bolsonaro campaign, they mistrust and dislike pollsters whom they view as being on the left is a matter of debate. What is clear is that there is nothing especially Brazilian about polling organizations failing to accurately assess the right-wing vote, given that the same phenomenon occurred with British polling organizations in the run-up to Brexit and to their U.S. counterparts regarding support for Donald Trump in both 2016 and 2020.