The geography of voting in the 2022 elections

Image: Magali Magalhães


The country is much richer and more nuanced and does not fit into a model of “two Brazils”: the conservative Southwest versus the progressive Northeast

The “two Brazils” of the 2022 presidential election

Apparently, both the left and the right have adopted a version about the geographic distribution of votes in the country that, from our point of view, is too simple. From this perspective, the Brazilian northeast, being “responsible” for the election of Lula, would be the main electoral base of the left in the country. By contrast, today, the Midwest and South regions would be the conservative nucleus in Brazil. And the North and Southeast regions would find themselves divided, despite having a more conservative than progressive bias.

This reading is far from gratuitous or mistaken. In fact, if we only take the presidential election as a reference, it is essentially correct. A single example is enough to demonstrate the point: Lula's victory over Jair Bolsonaro in the second round of the 2022 national elections was defined by a difference of 2 million and 140 thousand votes. But in Bahia alone the pro-Lula difference was 3 million 740 thousand votes. It is worth saying: the pro-Lula difference in this single northeastern state surpassed the difference in all of Brazil: if it weren't for Bahia, Jair Bolsonaro would have been victorious. And this spatial concentration of the pro-Lula vote was already manifested in the first round.

Chart 1, below, seeks to translate the geography of the vote for President. It was ordered by the voting percentage of Lula and Jair Bolsonaro in the various Federation Units (UFs). Both obtained, in the first round, more than 50% of the votes in eleven FUs. Only five of them – Amazonas, Minas Gerais, Amapá, Rio Grande do Sul and São Paulo – did not give an absolute majority to any of the two candidates in early October. Among the eleven states in which Lula won an absolute majority are the nine states in the Northeast and the two states in the far east of the North Macroregion (Pará and Tocantins), on the border with the Northeast.

By opposition, Jair Bolsonaro’s vote has a “Southwest” inflection, but its dispersion is greater. He obtained more than 50% of the votes: (i) in all four UFs in the Midwest; (ii) in two of the three UFs in the South: PR and SC; (iii) in three of the seven UFs in the North: RO (on the border with the Center-West), AC (southwest of the North region) and RR (UF in the disputed Yanomami reserve); and (iv) in two of the four UFs in the Southeast: RJ and ES. The highest percentage of votes for Jair Bolsonaro is in the extreme North, in RR, with 69,57% already in the first round. However, given the very low demographic density of the north of the country, the total number of votes obtained by Jair Bolsonaro in RO, AC and RO corresponded to 1,19% of his national vote.[I] His vote in SC (fourth highest percentage of votes for Bolsonaro in the country) corresponded to 5,29% of the national total for this candidate.

But Jair Bolsonaro's great electoral advantage over Lula will manifest itself in a UF that didn't even give him a victory in the first round: São Paulo. Almost a quarter of Bolsonaro's votes (24,2%) came from this state which, equally well, guaranteed him the greatest absolute advantage over Lula: 1 million and 750 thousand votes. This difference is greater than that obtained by the candidate in the four UFs of the CO combined (1 million and 400 thousand votes) which, in turn, is equivalent to Bolsonaro's advantage in Santa Catarina. Thus, despite the greater geographic dispersion, if we take the absolute and relative expression of the vote in the country, the southwest inflection of the Bolsonarist vote is confirmed.

Table 1: Percentage of votes for candidates for President by UF in the First Round of the 2022 presidential elections


This inflection is also noticeable when we observe the distribution of votes in the five States that did not win by absolute majority to any of the two candidates in the first round. Amapá (in the north of Pará, almost an appendix of this state), Amazonas (in the center of the North Region, on the border with Pará) and Minas Gerais (in the north of the Southeast region, on the border with Bahia) gave partial victory to Lula. While São Paulo and Rio Grande do Sul gave partial victory to Jair Bolsonaro. Figure 1, below, presents these results even more clearly.

Figure 1 – Geographical Distribution of the Vote for President in the First Round of the 2022 Elections


The red areas correspond to the municipalities where Lula was the most voted candidate; the blue areas correspond to the municipalities where Bolsonaro was the most voted candidate. From the outset, the map reflects the polarization of the 2022 elections: there is not a single point in another color; that is to say: the “third way” candidates were not the most voted in any municipality in Brazil. But, for that very reason (and contradictorily), the map hides something that the previous table reveals: the geographic distribution of votes for the different “third way” candidates.

As a general rule, the order of the vote percentage of the four most voted candidates in addition to Lula and Jair Bolsonaro was the same in Brazil as a whole and in each of the states. Tebet and Ciro are, respectively, third and fourth placed in all of Brazil and in 20 UFs. Soraya and D'avila are, respectively, fifth and sixth in Brazil and in 22 UFs. However, some exchanges of position between these two “pairs” emerge. They are hatched with a light green background in Chart 1. Ciro surpasses Tebet in seven FUs. All from the Northeast. And D'avila outperforms Soraya in the three states of the extreme south, as well as in São Paulo and Minas Gerais. This reveals two dimensions: (1) Ciro was unable to consolidate himself as a “PDT” candidacy and remained in fourth place even in UFs where this acronym has a long tradition and strong roots, such as RS and RJ; and (2) the Novo is an essentially southern party.

The advantage of the map over Table 1 lies in the fact that it highlights the regions of each UF where Lula and Bolsonaro won a simple majority. In this sense, attention is drawn to the homogeneity of the broad “red” spot that runs from the north of MG and ES to the east of AM, passing through the entire NE, north of PA and AP. The blue dots in this big red spot are very rare. The predominantly blue area, in the southwest of the country, has numerous red spots. And it's not just any stains. Some are wide and continuous as in northwest MS and southwest MT. Others are smaller in spatial terms, but very expressive in demographic terms. This is the case of the red stain in the Metropolitan Region of São Paulo (including the capital itself) and the stain that starts in Porto Alegre and continues towards the south of RS, including municipalities such as Pelotas, Rio Grande and Bagé, to then , turn northwest, encompassing the wide strip between Santa Maria and São Borja.


The geography of the vote for governor: first dissonant chords

The geographic distribution of the vote for governor is not very different from the presidential vote. But neither is its faithful reproduction. In Table 2, below, we present the party of governors elected in 2022 and compare it with the party of the governor in office (at the end of the term) in the same year. We ordered the FUs according to the same criteria as in Chart 1, with a view to facilitating comparison. In addition, we hatched the cells according to the parties' position in the 2022 elections. The red background was adopted for the parties that supported the Lula-Alckmin ticket; the yellow background for parties that supported third way candidates; and the blue background for the parties that supported the Jair Bolsonaro candidacy.

Although the PSD has not presented its own candidate, nor has it supported Lula or Jair Bolsonaro, we classify it as a “third way”. This classification is due to the fact that, despite this being an essentially conservative party (from the so-called “Centrão”), throughout the electoral process Gilberto Kassab even signaled his support for Lula in a negotiation that involved, in exchange, the Lula's support for the party's candidate in MG. The national agreement was not signed, but it was enough for the PSD to adopt a more equidistant stance between the two presidential candidates that polarized the 2022 elections.

Table 2: Caption of Governors who ended their term in 2022 and those elected in the same year


In only 6 of the 11 UFs in which the Lula-Alckmin ticket obtained more than 50% of the votes were elected governors of the parties that were part of the composition. In one of these states, PE, governed by the PSB until 2022 and where Lula won 65,3% of the votes in the first round, a PSDB governor was elected. In Tocantins, where Lula also obtained more than 50% of the votes, Governor Wanderlei Barbosa, from the Republicans, was re-elected in the first round. In MG, where Lula obtained 48,3% in the first round and where a broad front is formed with PSD, PT, Rede, PSB, PCdoB and PV in support of Alexandre Kalil's candidacy, governor Zema, do Novo, was re-elected in the first round with 56,18% of the votes. The dispute in SP went to the second round, but the victor was Bolsonaro's former minister, from the Republicans. On the other hand, in ES, where Bolsonaro obtained 52,23% of the votes in the first round, Renato Casagrande, from the PSB, was re-elected.


Crossing the samba: the composition of the Chamber of Deputies

The distance between voting for Lula-Alckmin and the governors is real, but somewhat subtle, and could be attributed to regional and local political idiosyncrasies. This reading, however, does not hold when we analyze the election for the Chamber of Deputies by UF. To simplify the analysis of the results, we aggregated the 22 parties that managed to elect federal deputies in four groups. In the first (with 122 deputies), we included the 9 parties that supported the Lula-Alckmin ticket, namely: PT, PCdoB and PV (Federação Brasil Esperança), PSOL and Rede (also federated), the PSB, Avante, Solidariedade and PROS . In the second group (with 198 deputies sworn in), we included the 3 parties that supported Bolsonaro – PL, PP and Republicans – as well as 2 parties that, despite not having formally supported this candidacy, operated as a base of support during his presidential term and in the campaign: PSC and Patriota.

In the third group (89 deputies), we included those parties that supported the candidacies of Tebet (MDB, PSDB, Cidadania and Podemos) and Ciro (PDT) and that, in the second round, with greater or lesser commitment, supported the Lula-Alckmin ticket . In the fourth group (104 deputies) we include the “conservative third way” parties, which abstained from supporting any candidate in the second round, but whose electoral bases and various regional leaders aligned themselves with Jair Bolsonaro: União Brasil, Novo and PSD. The result is below

Table 3: Structure of the Chamber of Deputies in 2023 by UF and Party Political Orientation


As would be expected, almost 50% of the Central-West bench is made up of deputies from Jair Bolsonaro’s base. But already here surprises emerge. The third way that supported Lula in the second round elected almost a quarter of the region's bench, despite Ciro and Tebet having won, together, only 7,2% of the votes in the Midwest. In addition, 25% of the new DF and MS benches are deputies from the parties that supported the Lula-Alckmin ticket. In total, the “left bloc” elected 6 deputies in the Midwest, almost 15% of the regional bench. This result may seem insignificant. And indeed it is, if we compare the vote for the House[ii] with the vote for Lula in the first round (which was 37,83% in the Midwest). However – and this is the point we want to draw attention to – the vote in the “leftist bloc” in the Center-West was much higher than the vote received by this same bloc in the North of the country. And this is not trivial.

As you can see on the map above, most of the Northern Macroregion is “stained in red”. However, the left-wing bloc elected only two deputies in this region, both from Pará. Of the eight deputies elected by Tocantins (where Lula obtained 50,4% of the votes in the first round), two are from the PL, two are from the PP, three are from the Republicans and one is from União Brasil. The performance of Amazonas and Amapá (where Lula obtained more than 45% of the votes in the first round) was not so conservative: the two UFs elected deputies from the “progressive third way” parties. But no candidate from the “left bloc” was elected in these two UFs.

In a way, the situation in the Northeast is even more surprising. In the first round of elections, Lula obtained 68,84% of the votes in Maranhão, 65,91% in Ceará and 64,21% in Paraíba. But of the 18 deputies elected by Maranhão, 9 are from Bolsonarist parties, and the rest were divided equally between the other three blocks: that is to say: the parties of the “leftist block” received less than 20% of the votes in this State Ceará elected 22 deputies; of which 5 are PL, 5 are PDT (Ciro Gomes' party), 4 are União Brasil and 3 are PSD. The left block elected 3 deputies in this UF; all from PT.

Of the 12 deputies to which Paraíba is entitled, 9 of those elected are from the Bolsonarist camp and 1 is from União Brasil: that is to say: 83,3% are from conservative parties. It is quite true that there are UFs in the region where the left's performance was better. The extreme example is Piauí, where of the 10 deputies elected in the UF, 5 are from the left field. But the other 5 are from the conservative camp (PP and PSD). Which is in contrast to the 74,3% vote for Lula (as opposed to the 19,9% ​​vote for Bolsonaro) in the first round. In an intermediate position between the pattern of MA, CE and PB and of PI we have Pernambuco and Bahia. PE deserves 25 deputies and elected 11 from the leftist block, 5 from the PSB and 1 from each of the other parties (except PSOL and PROS). But the Bolsonarist block was not far behind: it elected 10 deputies in this UF. Bahia elected 12 deputies from the leftist bloc. But it also elected 10 deputies from parties supporting Bolsonaro, 6 from União Brasil and 5 deputies from the “third via Simone-Ciro”.

From our point of view, these results show that the Northeast is much more heterogeneous in the political field and has a more conservative (or at least “politically eclectic”) electorate than those who take the results of the presidential elections as a necessary expression intend. and sufficient understanding of the political-ideological profile of the region. In fact, the set of electoral results in the territory reveal a region that is more “Lulist” than “leftist”. It is worth noting that the PSOL did not elect a single deputy in the entire Northeast. And the PT only made a good bench (21 out of a total of 69 elected by the party) due to Bahia (7 deputies) and Piauí (4). In the other FUs, the performance of the PT was quite modest.

The southern region also brings some surprises, starting with its diversity. Of the 18 deputies elected in the region by the leftist bloc, 9 are from RS, 7 are from PR and only 2 are from SC. Almost 30% of the gaucho bench is from the left block, while in SC this percentage is just over 10%. There is a little more homogeneity in the representation of Bolsonaro's base parties: just over 30% of the representation of each of the states. And this is an important point to highlight: the federal bench of Bolsonarist parties in the South region corresponds to 32,47% of the total bench in the region. This percentage is high, but it is the lowest among all regions of the country.

If we take Jair Bolsonaro's base parties as a reference, the South is less Bolsonarist than the Northeast, whose federal caucus from Bolsonaro's base corresponds to 37,75% of the total. On the other hand, the heterogeneity of the region manifests itself again in the “two third ways”. In PR, the “progressive third way” elects only 10% of the bench, while it exceeds 25% in RS and SC. On the other hand, almost 37% of the deputies from Paraná are from the “conservative third way”, 4 from União Brasil and 7 from PSD.

The first element that draws attention in the analysis of the Southeast is its greater relative homogeneity and greater consistency between the votes in the different presidential candidacies and the votes in the parties that supported them. This characteristic is not surprising: SP, MG and RJ are the three largest electoral colleges in the country and the economic and cultural center of Brazil. As contradictory as it may seem, internal diversity feeds the group's homogeneity and convergence, as it depresses the relative expression of idiosyncratic dimensions, strictly regional and/or local. See, for example, the participation of the “leftist block” in the stands of ES (30%), MG (33,96%), RJ (30,43%) and SP (28,57%). Taking the region as a whole, 30,73% of the elected representatives are from the “left bloc”.

And here is the second point to note. The percentage of the “bloc of the left” in the Southeast is higher than the percentage of the “bloc of the left” in the country as a whole: 122 deputies out of a total of 513 correspond to 23,78%. More: this percentage is higher than that of the Northeast Region taken as a whole. The northeastern bench is made up of 151 deputies and the “left bloc” has 41 elected representatives, making up 27,15% of the total. Not for free, the Southeast was responsible for almost half of the national bench of the “left bloc”: 45,08% of the total. Even more: the representation of some parties from the left-wing bloc is based almost entirely on the three largest electoral colleges, SP, RJ and MG: 91,6% of the PSOL bench; 71,4% from Avante; 75% from Solidarity; 100% of the PROS bench; and 50% of the Network's bench is made up of deputies elected in these states. In short: without the FUs that elected Zema, Tarcísio and Castro as state governors, both the numerical expression of the “leftist bloc” in the Federal Chamber would be smaller (it would fall from 122 to 69 deputies), and its internal diversity would be negligible, since parties such as PSOL, Avante, PROS and Rede would not survive as viable organizations within the current barrier clauses.



The geography of the vote for the Federal Chamber reveals a much more complex and heterogeneous country than the one that emerges from an analysis limited to the presidential election. It brings to light a more conservative Northeast and North and a less conservative Southeast, South and Midwest than the map of Brazil in Figure 1 implies.

Someone could counter-argue that the vote for the executive and the vote for the legislature follow different logics, and that it is the vote for President that best reflects the political-ideological profile of the territory. If a “simple dialectic” is allowed, we would say that this counter-argument is and is not legitimate. It is legitimate in the sense that the voter's option for this or that legislative candidate is mediated by determinations that greatly transcend the ideological profile of the party to which the candidate is linked. Elements such as personal knowledge, territory of origin, expected benefits for the region, for himself and/or for the economic sector in which the voter operates are as or more important than the ideological inflection of the candidate and his party.

Undoubtedly, this point is important and true. It is not for nothing that, from time to time, conservative ideologues and politicians rescue the project of parliamentarism in Brazil. Presidential elections have a “plebiscitary” dimension, where two projects are opposed; as a general rule, a project on the left (more interventionist, industrializing and distributive) and a project on the right (liberal inflection, privatizing and averse to public policies for income distribution). In an exclusionary country like Brazil, in democratic regimes the tendency is for the left to win. Hence the recurrence of coups (as in 1954, 1964, 2016-2018) with a view to restoring the right to power. The parliamentarism project seeks to eliminate the plebiscitary dimension of the presidential election under the assumption that the electorate will continue to elect a Chamber based on “clientelistic” criteria, as opposed to specifically utopian-ideological criteria.

However, there is also another side to the issue. As Putnam argued in his Community and Democracy, one of the main expressions of the opportunist political culture of southern Italy in contrast to the high social capital of the north is found in the voting pattern: clientelist in the south and utopian-ideological (partisan) in the north[iii]. That is to say: a vote in proportional elections that is based primarily on personal relationships and professional interests and/or local benefits is not necessarily the norm. As, by the way, all those who vote for criteria defined from the social project of the party know. Voting for a candidate for what he “promised to do for me” is also a political option. And it has an opportunistic and conservative dimension.

Finally, the fact that the vote for President has a plebiscitary dimension and is marked, fundamentally, by utopian-ideological elements, does not make it exempt from elements of personal and regional interest. Lula is from the Northeast and has clear commitments to tackling regional inequalities and promoting socioeconomic development in the Northeast at all levels: from infrastructure (transposition of the São Francisco River, Luz para Todos, Água para Todos, etc.) to health services (qualification of SUS, Samu, etc.) and education (internalization of Universities and Federal Technical Schools, Pronatec, etc.). And, from our point of view, such commitments explain a non-negligible portion of northeastern (and northern) “Lulism” as opposed to the “anti-Lulism” of the Southwest.

The Brazil that voted for Jair Bolsonaro is, fundamentally, an included Brazil that fears any kind of redistributive public policy, whether between social classes or between regions. This point is very clear in the “red spots” in the UFs in the South, Southeast and Midwest. In RS, Lula had the majority of votes in the Southern Half, characterized by large estates and very low economic dynamism. In PR, the big red spot is found in the center-west of this UF, around Guarapuava, the poorest and least industrialized region of PR. In MG, the red spot is found in the north and northeast of the UF, also the poorest and lacking government policies to support regional development.

The great lulist spot in the Center-West corresponds to the Pantanal region, with the lowest IDHM in MS and MT. The “Lulism” of these regions does not seem to be based on any political-ideological hegemony of the “left”, but on the belief that public policies are needed to support job creation and federal public investments in the territories with a view to confronting economic stagnation . Without being backward or stagnant, the capital of São Paulo and its industrial surroundings can also be included in this group: due to the country's deindustrialization process, which has now completed three decades, the RMPS has been losing participation in the Brazilian GDP and GVA every year what happens. And it will only be able to recover its previous dynamism and productively include part of its unemployed population if active public policies are adopted to support the national industry. In short: Lula's Brazil is, above all, a Brazil that demands public actions to support development. But it is not necessarily a Brazil committed to the entire spectrum of the left agenda.

Our intention, evidently, is not to deny the political inflection to the left of the Northeast. All Brazilian progressives owe a political debt to this region of the country, which saved us from another 4 years of Bolsonaro misrule. Our intention is just to warn that the country is much richer and more nuanced and does not fit into a model of “two Brazils”: the conservative Southwest versus the progressive Northeast. Without the South and Southeast, the bench of the Federal Chamber's “left bloc” would correspond to a mere 40% of what it is today. So it is also worth saluting and thanking the important political contribution of these regions to the confrontation of Bolsonarism in Brazil.

*Carlos Águedo Paiva is a doctor in economics and professor of the master's degree in development at Faccat.


[I] In addition to the national vote, the vote abroad must be taken into account. Despite not being expressive, it discreetly changes the percentages of each FU in the total.

[ii] Note that we are using the percentage of deputies elected in each of the “four blocks” as proxy of the percentage of total votes for the parties included in each grouping. It is proxy it is far from perfect. When using it, we disregard the votes attributed to those parties that did not reach the electoral quotient. A more rigorous research would also involve taking these votes as a reference. However, we understand that the costs of gathering this information would not compensate for the benefits obtained. And this for several reasons. But two are fundamental. Firstly, because our focus is not on individual parties, but on large blocks. In this way, the deviations associated with disregarding the votes of parties that did not reach the quotient tend to be evenly distributed among the four blocs, as they all have small parties with an essentially regional base. This is the case of Solidariedade, PROS, Avante and Rede in the left block; PSC and Patriota in the Bolsonarist block; from Cidadana and Podemos, in the “third progressive way”, and from Novo, in the “third conservative way. On the other hand, all blocks have large parties, nationally structured, such as the PT and the PSB in the victorious coalition, the PL and the PP, in the Bolsonarist group, the MDB, the PSDB and the PDT in the progressive third way and the União and the PSD in the final block. These parties tend to receive the vote of their ideological counterparts in territories where “runt” parties have little or no chance of gaining representation.

[iii] Putnam's research took place before the operation Hands clean, the “Italian Lava Jato”, which criminalized politics and virtually destroyed traditional parties such as the Christian Democrats, the Communists (later the Left Democrats) and the Socialists.

See this link for all articles


  • João Cândido and the Revolt of the Whipwhip revolt 23/06/2024 By PETRÔNIO DOMINGUES: In the current context, in which there is so much discussion about State reparations for the black population, the name of João Cândido cannot be forgotten
  • Fear and HopeJoao_Carlos_Salles 24/06/2024 By JOÃO CARLOS SALLES: Against the destruction of the public university
  • The collapse of Zionismfree palestine 80 23/06/2024 By ILAN PAPPÉ: Whether people welcome the idea or fear it, Israel's collapse has become predictable. This possibility should inform the long-term conversation about the future of the region
  • Franz Kafka, libertarian spiritFranz Kafka, libertarian spirit 13/06/2024 By MICHAEL LÖWY: Notes on the occasion of the centenary of the death of the Czech writer
  • A look at the 2024 federal strikelula haddad 20/06/2024 By IAEL DE SOUZA: A few months into government, Lula's electoral fraud was proven, accompanied by his “faithful henchman”, the Minister of Finance, Fernando Haddad
  • Return to the path of hopelate afternoon 21/06/2024 By JUAREZ GUIMARÃES & MARILANE TEIXEIRA: Five initiatives that can allow the Brazilian left and center-left to resume dialogue with the majority hope of Brazilians
  • The society of dead historyclassroom similar to the one in usp history 16/06/2024 By ANTONIO SIMPLICIO DE ALMEIDA NETO: The subject of history was inserted into a generic area called Applied Human and Social Sciences and, finally, disappeared into the curricular drain
  • About artificial ignoranceEugenio Bucci 15/06/2024 By EUGÊNIO BUCCI: Today, ignorance is not an uninhabited house, devoid of ideas, but a building full of disjointed nonsense, a goo of heavy density that occupies every space
  • Theological manual of neoliberal neo-PentecostalismJesus saves 22/06/2024 By LEONARDO SACRAMENTO: Theology has become coaching or encouraging disputes between workers in the world of work
  • Chico Buarque, 80 years oldchico 19/06/2024 By ROGÉRIO RUFINO DE OLIVEIRA: The class struggle, universal, is particularized in the refinement of constructive intention, in the tone of proletarian proparoxytones