Brazil holds the first round of its presidential election on Sunday, and if you want to understand how global finance is processing this event, the shortest path is a quote, via CNBC, from Fernando Marengo, chief economist at Black Toro Global Investments: “The Brazil trade is: Does Lula win or does Bolsonaro win?”

That is the whole thing. On one side is Luiz Inácio Lula da Silva, the 80-year-old leftist president, running for a fourth term. On the other is Flávio Bolsonaro, a 45-year-old senator who is, as you may have inferred, the son of the man Lula narrowly beat in 2022 — Jair Bolsonaro, now serving a 27-year sentence for plotting a coup to overturn that loss, a scheme that according to police included a plan to assassinate Lula. If neither candidate clears 50% of the valid votes on Sunday — and polls suggest neither will, with twelve candidates on the ballot — the top two go to a runoff on Oct. 25. Brazilians, nearly 160 million registered voters of them, are also electing the entire lower house, a third of the Senate, governors and state legislatures.

The positioning is not subtle. CNBC reports that, according to a recent JPMorgan note to clients, the MSCI Brazil “rose by 0.25% on average each day that Flavio gained in the polls” over the last few months. Which is a tidy way of saying the stock market has been turned into a poll tracker with a preferred answer.

Kalshi, the prediction market, has Bolsonaro favored 60% to 39%. Two caveats apply. Prediction markets are banned in Brazil, so the prices may not reflect local sentiment. As Richard Lapper, senior advisor at Aurora Macro Strategies, put it in a client note quoted by CNBC, “the balance has shifted toward Flavio over the past month, but not nearly as far as the prediction markets are pricing.” (The other caveat, per CNBC’s own disclosure: the network has a commercial relationship with Kalshi.)

The fiscal fantasy

Why does Wall Street want Bolsonaro? One phrase: fiscal discipline. Brazil’s debt-to-GDP is 81.9%, up 10% since Lula took office, and the received wisdom among economists is that something has to give. “We need a 3-3.5% fiscal adjustment to stabilize the public debt in relation to GDP,” Leonardo Porto, Citi’s head Brazil economist, told CNBC, adding that it can’t come from one-offs like privatizations: “Brazil needs a permanent fiscal adjustment.”

Now, imagine you run a country where roughly 90% of the budget is mandatory — some of it constitutionally locked — and where the tax burden, at 32%, is already the highest in Latin America, according to the OECD. You are told to close a 3-to-3.5-percent-of-GDP gap, permanently, by cutting spending you cannot cut or raising taxes that are already the region’s steepest, in a low-growth economy. Sure.

The historical template, per JPMorgan, is the father. Under Jair Bolsonaro, Brazil passed pension reform — a minimum retirement age of 65 for men and 60 for women, where previously men could retire at any age after 35 years of work and women after 30, with average actual retirement ages of 56 and 53. That saved hundreds of billions of dollars. During that reform period, JPMorgan says, two-year yields fell almost to 4.7% and the equity market gained 130%.

Extrapolate, then. If Bolsonaro fils wins and delivers what the firm calls a “robust reform agenda,” JPMorgan figures interest rates could fall to neutral — 6% in real terms, 10% nominal — and “we would be thinking about the MSCI Brazil upside potential between 21% and 41%.” The forward P/E could go from 8.6 to 13.3, a level last seen in 2020. The currency call is “bimodal”: USD/BRL at 4.90 if Bolsonaro wins, 5.50 if Lula wins. Note the asymmetry. The model has two modes, and each mode has a president attached.

Marengo points to precedent: Colombia’s pro-business win compressed its risk premium by “about 200 points, and it was one of the stock markets that rose the most — something similar to what happened in Peru.” Though he cautions some of the move is already priced into Brazil. The risks, meanwhile, are the usual emerging-market kind — rising global rates, and El Niño wrecking harvests. Also, the composition of Congress matters enormously for whether any of this reform actually happens, and nobody has a hedged ratio for that.

The things the model doesn’t price

Here is where the bimodal distribution gets awkward. Flávio Bolsonaro has promised to pardon his father if elected. He campaigns on hardline security and Milei-style spending cuts, brands Lula “incompetent,” and is himself under investigation for potential money laundering connected to a transfer of money from Daniel Vorcaro — the jailed banker at the center of one of Brazil’s biggest financial scandals, which also financed “Dark Horse,” a biopic about the elder Bolsonaro, and which has tangled up Supreme Court justices. Bolsonaro denies wrongdoing. (The scandal has been hanging over this whole race, oddly enough hurting the left more than the candidate actually being investigated.)

Then there is the foreign policy leg. The sources disagree slightly on the arithmetic, but agree on the direction: last year, Trump raised tariffs on Brazil — NPR says by 40%, Euronews puts the figure at 50% — citing, among other things, the trial of Jair Bolsonaro. Euronews reports a further round of tariffs earlier this year, imposed despite a U.S. trade surplus, plus the designation of Brazil’s biggest drug gangs as foreign terrorist organizations shortly after Flávio visited the White House. Trump does not formally endorse the son, but says he is following the election “very closely” — close enough that Flávio’s rallies feature giant inflatable Trumps, and Lula’s campaign countered with a Portuguese version of Bad Bunny’s Lo que le pasó a Hawaii, a song about what U.S. colonization did to Hawaii. As campaign economics go, that is a thesis.

Lula has made sovereignty the centerpiece — “Brazil belongs to Brazilians,” per the banners at his São Paulo finale — and in July publicly accused Washington of planning to meddle in the vote, which the State Department called “a baseless lie.” That matters because of the scenario Yuri Sanches of the polling firm AtlasIntel sketched for NPR: a narrow Lula win, and “then almost for sure we would have some Bolsonaro supporters questioning the results. They may ask for help from Trump, from Rubio, from the United States, saying that the election was rigged, that the electoral system, the electronic voting machines are not reliable.”

So the trade, as priced, has exactly two states of the world: a rally on 4.90 or a sell-off to 5.50. The actual distribution includes a third state — a close count, a crying foul, and an outside power with a recent record of tariffing countries over court cases getting involved — and nobody’s P/E multiple has a cell for that. We noted yesterday that the market’s clean binary may be the misprice.

One way to summarize the market’s position: it has strong views on what Flávio Bolsonaro would do to the budget, and no views whatsoever on what he would do to the institutions that enforce budgets. The first candidate in modern Brazilian history to run on pardoning a convicted coup-plotter — his own father — is, to Wall Street, the discipline candidate. Which tells you something about what the word means in that sentence.