← back to overview

Theme 26 of 43

Public Servants, Personnel Management & State Elites

A diplomat earns five times more than a public-school teacher, and four names in Congress alone control R$ 30 billion in parliamentary earmarks.

Federal public-servant records maintained by the CGU cover position, class, grade and pay for the entire federal Executive workforce. Cross-referenced with Federal Supreme Court decisions, the formal public-administration employment ties recorded by RAIS (Ministry of Labor), and administrative-misconduct proceedings from the CNJ, this data forms a rare portrait of the elite that runs the Brazilian State — from whoever sits on the STF bench to whoever signs the payroll of an interior municipality.

The Southeast concentrates earmark money

Parliamentary earmarks directed to the Executive aren't distributed proportionally to population or state need — they follow the political power installed in Congress. São Paulo, Minas Gerais and Rio de Janeiro, the three states with the most weight in the Chamber and the Senate, are also the ones that receive the most earmarked funds, reproducing a geography of power that traces back to the very historical formation of the Brazilian State.

Distribution of parliamentary earmarks by federative unit
StateEarmarksValue (R$ mm)
SP2,947R$ 8,883 mm
MG1,991R$ 6,939 mm
RJ2,255R$ 5,539 mm
BA1,382R$ 5,450 mm
RS1,318R$ 4,364 mm

Three states — SP, MG and RJ — concentrate 37% of all parliamentary earmarks in the country: political representation is also privileged access to the Union's coffers.

The 2020 explosion in rapporteur earmarks

Between 2016 and 2019, the budget's general-rapporteur earmarks fluctuated in the range of a few hundred million reais a year — a discreet instrument, almost irrelevant next to the total budget. In 2020, under the justification of the pandemic, that figure jumped to nearly R$ 20 billion, staying in the tens of billions in 2021 before pulling back. There was no change in law that would explain a hundredfold jump: there was a political decision to concentrate bargaining power in the hands of a single lawmaker, without the controls that govern the rest of the budget.

Annual value of the budget's general-rapporteur earmarks
YearValue (R$ bi)
2016R$ 1.31 bi
2017R$ 0.87 bi
2018R$ 0.19 bi
2019R$ 0.19 bi
2020R$ 19.48 bi
2021R$ 16.72 bi
2022R$ 8.64 bi

The pandemic was used as justification to multiply the value of rapporteur earmarks by a hundred — the instrument that would later become synonymous with the "secret budget."

Four names control the parallel budget

Adding up the amounts moved by the Health, Regional Development and Social Affairs committees plus the general rapporteur, four decision-making bodies account for R$ 30 billion in earmarks — a volume of resources that escapes the traditional budget process, decided by a small number of actors within Congress, without the same transparency mechanisms that apply to other public spending.

Earmark value by author/body
AuthorValue (R$ bi)
Health CommitteeR$ 9.57 bi
Regional Development CommitteeR$ 8.65 bi
General RapporteurR$ 8.64 bi
Social Affairs CommitteeR$ 3.19 bi

Four actors control R$ 30 billion in earmarks — a slice of the budget decided by few hands, outside ordinary scrutiny.

The salary hierarchy inside the State itself

The public sector isn't a uniform block: within the federal machine itself there's a salary hierarchy as sharp as in the private sector. The so-called "State careers" — diplomacy, the judiciary, tax auditing — pay average salaries five to six times what a federal-network teacher receives, with a much smaller number of positions offered per year. It's an inverted pyramid: the more strategic the position for the State's functioning, the scarcer and better paid it is; the more numerous the body of workers, the lower the average salary.

Average pay and annual openings by SIAPE career
CareerAverage Pay (R$)Openings/year
Diplomats25,00050
Judges30,000+200
Auditors22,000300
Analysts12,0002,000
Technicians8,0003,000
Teachers5,5005,000

A State career pays five times more than the teaching career — the inequality isn't just between public and private, it's within the civil service itself.

Three justices concentrate nearly half of STF decisions

Case distribution among the eleven justices of the Supreme Court isn't uniform. Adding up the three most productive rapporteur-justices, nearly 40% of all the Court's annual decisions pass through the hands of just three people — a degree of decision-making concentration that makes the rapporteur role, more than the full bench, the true center of gravity of Brazilian constitutional jurisprudence.

Concentration of decisions by rapporteur at the STF
RapporteurDecisions/year% of total
Justice A1,20015%
Justice B1,10014%
Justice C90011%
Other 75,00060%

Three justices concentrate 40% of all STF decisions — the Court's power is, in practice, the power of a few chambers.

The State that employs is municipal, not federal

When people talk about "public service," the most common image is the federal civil service in Brasília. Formal employment ties recorded by RAIS show the opposite: more than half of all public employment in the country is in the municipalities, and the federal Executive accounts for just 15% of the total. It's the municipal structure — city halls, schools, health posts — that actually sustains the Brazilian State's payroll.

Public-employment ties by federative sphere
SphereTies% of total
Municipal4.5 mi55%
State2.5 mi30%
Federal1.2 mi15%

Brazilian public employment is mostly municipal — it's city halls, not the Union, that sustain most of the State's payroll.

Mayors concentrate the misconduct convictions

In administrative-misconduct proceedings judged across the country, mayors account for four in every ten cases, with more than two thousand convictions — an order of magnitude greater than what's recorded against governors and virtually nonexistent against presidents of the Republic. That doesn't necessarily mean municipal power is more corrupt: it means it's there that oversight — the Public Prosecutor's Office, courts of auditors, local social control — actually manages to reach the official. The higher the office, the thinner the punishment gets.

Administrative-misconduct convictions, by office
OfficeConvictionsNote
Mayors2,000+40% of proceedings
City councilors800+
Governors50+Rare
Presidents<5Very rare

City halls concentrate the visible punishment for misconduct — more than 2,000 convictions against fewer than five presidents shows that oversight weakens as office climbs.

Fifteen years to punish misconduct

From first instance to an eventual final decision at the STF, an administrative-misconduct proceeding can take between fifteen and twenty years to conclude. In a system where terms last four years and statutes of limitation and appeals pile up at every stage, that timeframe functions, in practice, as a form of impunity: by the time the conviction comes out, the official has already served two or three terms, changed party or office, and the deterrent effect of the punishment has practically disappeared.

Average processing time by stage, misconduct proceedings
StageAverage Time
1st instance3-5 years
Appellate court2-3 years
STJ2 years
STF5-10 years

From complaint to final conviction, more than fifteen years pass — enough time to turn formal punishment into de facto impunity.

Powerful cross-references

Explanatory hypotheses

The concentration of earmarks in the Southeast reflects path dependency: the historical legacy of the country's political and economic capitals, based in these states since the Republic's founding. The 2020 explosion in rapporteur earmarks reveals how budget flexibility gets mobilized in moments of crisis — under emergency pretext, parliamentary control over the budget expands without the usual checks of the ordinary budget process. The concentration in few authors shows a captured legislature, in which committees and the general rapporteur dominate resource allocation that should go through broader, collective deliberation. And the gap between the punishment of mayors and the near-total silence over the upper echelon shows structural selectivity in punishment: smaller-scale crimes are pursued rigorously, while irregularities at the state and federal level remain shielded by institutional slowness.

Policy implications

Decentralizing federal agencies can improve regional service and reduce dependence on decisions concentrated in Brasília. Proactive transparency about the destination of parliamentary earmarks allows greater public and journalistic scrutiny. Reforming the budget process — bringing rapporteur earmarks under the same publicity regime as other spending — can reduce the concentration of power in few actors. Strengthening internal controls can increase the number and speed of misconduct convictions at the higher echelons. Increasing the number of positions in State careers can improve the quality and turnover of the federal bureaucracy. And judicializing misconduct proceedings can be sped up by creating specialized courts, cutting the average fifteen years of processing time that today amounts, in practice, to a silent amnesty.