This past Monday, Brazilians voted to reinstall as president Luis Inácio Lula da Silva, the leftist labor leader who previously served in the same office from 2003 to 2010. Lula defeated the far-right Jair Bolsonaro, elected in 2018 – which means that for the first time in Brazil’s history as a democracy, an incumbent president failed to win reelection.
The election was interesting for a whole host of reasons. It’s part of a wave of leftist resurgence in South America in the last few years, as leftists have also defeated far-right opponents in Peru, Chile, and Colombia. Lula’s story is particularly remarkable – his return to the presidency after more than a year in jail is astonishing.
But the election was also interesting because in an important way, it mirrored recent contests in the United States and other Western countries.
Education polarization:
Education polarization is the political science term for the divergences in political preferences of college-educated and non-college educated voters. In the last few decades, highly educated voters across the Western world have consistently moved toward left-wing parties, while their less educated counterparts have moved to the right. The below chart, from the French economist Thomas Piketty’s research, illustrates this trend nicely:
Thus far, education polarization has mostly been studied in the context of the countries in the chart—Britain, France, and the United States. The results in Brazil, however, provide support for the idea that this trend is a global phenomenon, affecting democracies in developing countries as well.
What happened in Brazil:
The charts below are based on municipality-level data from Monday’s runoff election, and were compiled by the Twitter user (and political gambler) Iabvek. All three charts are different ways of showing the same thing: that the more educated an area was, the more, on average, it shifted toward Lula relative to the 2018 baseline.
Here’s swing compared to the proportion of people who lack literacy:
As you can see, there’s a negative correlation between illiteracy and swing: eyeballing the chart, it looks like areas with less than ~5% illiteracy swung towards Lula by 5-10%, while areas with 40% illiteracy moved towards Bolsonaro by about 5%.
Here’s swing compared to the college-educated proportion:
And here’s swing by the percentage of people with a high school degree:
Together, these charts are convincing evidence that Lula gained the most relative to his party’s 2018 result in whiter, more highly educated locations, while slipping a bit in poorer areas with less education and higher illiteracy.
The funny thing about this is that in the Brazilian context, this result actually represented education depolarization. That’s because, unlike in the US or UK, college-educated voters still, on average, vote for right-wing parties in Brazil, while working class voters vote for the left. College-educated Brazilians moving left and working-class voters moving right brings the two groups more in line with each other, given their current levels of support for each party: thus, depolarization.
A global phenomenon:
If the history of other developed countries is any indication, this is just the beginning of a shift to the left for Brazil’s educated classes.
The Brazil results also suggest that any explanation for education polarization must be a truly global one. People come up with all sorts of just-so stories for why educated voters are moving left, and less educated voters right, in a given country. In the United States, to give one example, some have tried to blame the loss of working-class voters from the Democratic party coalition in the United States on the party’s supposed betrayal of its union base.
However, if a similar phenomenon is happening to Lula, perhaps the most labor sympathetic politician in the entire world, it seems like something deeper must be going on.1
As for what that deeper thing is, I don’t think anyone knows for sure, though I have a few guesses myself – more on this soon.
Death, taxes, and education polarization
I don't understand the point being made here about unionization. Fall in union density and labor power has been a pervasively global phenomenon over the last 40 years; far from weakening that argument, the facts objectively support it.
I am so tired of everyone who is somewhere the to right of moderate being called “far-right” while actual communists are only called leftists. The message I get is that the media believes that there is no such thing as merely right or right-of-center. There are the good guys to the left of center, the wishy-washy moderates, and then the far right. It undermines everything said after “far-right.”