Brazil’s 1992 Administrative Improbity Law, which authorized severe sanctions on government agents who commit “acts against the public administration,” was the first Brazilian statute specifically targeted at government corruption. Last year, Brazil adopted extensive amendments to this law, many of which were controversial. In a recent post, I criticized the amendment that reduced the number of institutions responsible for enforcing the Improbity Law. But other controversial amendments to the law are, in my view, positive developments. In particular, I want to defend two other amendments that critics have asserted weaken the law:
- First, under the original version of the Improbity Law, a public official could be sanctioned for negligent behavior that caused damage to the public treasury. Under the amended version of the law, only intentional acts can be considered administrative improbity punishable under this statute.
- Second, the original version of the law listed ten forms of administrative misconduct that would constitute “violations of the principles of public administration,” but, importantly, that list was not exclusive. Rather, the listed forms of misconduct were presented only as examples. This meant that law enforcers could, and often did, bring an action under the Improbity Law for conduct that, in the enforcer’s view, violated a “principle of public administration,” such as morality and equity, even if the particular form of alleged improbity was not included as one of the specifically listed forms of misconduct in the statute. The amended law constrains enforcement discretion by establishing a well-defined and restricted list of acts that qualify as violations of the principles of public administration.
Critics, including many anticorruption advocates, assert that these changes unduly narrow the scope of the law, thereby undermining one of Brazil’s most important anticorruption instruments. These concerns, while understandable, are misplaced: Both of the above amendments improve the law by ensuring that it is administered fairly and used to target serious corrupt acts, rather than being wielded as a political weapon to punish partisan adversaries for good-faith mistakes.
Consider first the addition of a requirement that administrative improbity offenses must be intentional rather than negligent. Critics argue that this change will make it too difficult to prove misconduct—because intent is harder to prove than negligence—thus weakening the law’s deterrent effect. The problem, though, is that under the previous version of the law, public officials could be, and often were, sanctioned for administrative mistakes and other minor misconduct that could not plausibly be characterized as corruption. Treating all such mistakes as “improbity” was counterproductive, wasting resources on costly lawsuits involving minor cases, and often leading to the imposition of unfairly harsh sanctions on government officials who made administrative errors but who did not seek to enrich themselves or otherwise behave dishonestly. This severe approach is especially inappropriate in a country like Brazil, where many government entities, especially at the local level, have limited resources and lack specialized personnel. In this environment, administrative mistakes are relatively common, and often result from lack of legal and management knowledge rather than bad faith or serious malfeasance.
Moreover, the waste of resources on Administrative Improbity cases that do not involve corruption is not costless. According to an official report released last year by the Brazilian Judiciary, barely more than a third of the 177,921 administrative improbity suits filed up to 2018 have been resolved. Judicial and prosecutorial resources should be targeted on cases of actual corruption, not dissipated on cases that involve mere mismanagement. (it’s also worth noting that this amendment to the Administrative Improbity Law does not mean that negligent public officials won’t face any sanctions. Depending on the circumstances, a negligent official might still face disciplinary action under the Civil Servants’ Statute, and there are other judicial mechanisms for ordering restitution for damages to the public treasury.)
As for the amendment narrowing the scope of administrative improbity offenses to a list of specifically enumerated forms of misconduct, here critics worry that depriving law enforcers of the flexibility to go after new and perhaps unforeseen forms of misconduct will limit the law’s effectiveness. That argument might sound plausible in the abstract, but Brazil’s experience under the original version of the Administrative Improbity Law has highlighted the price of giving law enforcers such flexibility. Law enforcers can, and unfortunately do, use this discretion for purposes other than fighting corruption—such as harming political adversaries or punishing minor misconduct that could be addressed through administrative disciplinary actions. A recent study of Superior Court of Justice decisions found that over half of administrative improbity suits filed against mayors between 2005 and 2018 were for unspecified “acts against the principles of public administration,” a figure that suggests that, rather than using the law’s flexibility to go after novel and unforeseen forms of misconduct, law enforcement bodies have instead been using the open-ended prohibition on “acts against the principles of public administration” as a license to go after government officials who could not credibly be accused of specific forms of misconduct. Limiting the application of the Improbity Law to a list of punishable conduct will allow a more rational use of the punitive power of the law, prevent abuses, and provide legal certainty.
For these reasons, these two amendments to the Administrative Improbity Law are welcome changes. While it is understandable that anticorruption advocates would be worried about any changes that limit the scope of this important law, these particular restrictions will increase the law’s overall effectiveness by ensuring that scarce enforcement resources will be targeted toward the investigation and prosecution of actual corruption cases.
GAB | The Global Anticorruption Blog 
Thank you for this post Rafael! With regards to the first point concerning elevating the standard from negligence to intent, you make a compelling point that such a change would limit the over-penalization of administrative errors by political adversaries. This makes me wonder if, in a country like Brazil, you would posit that the weaponization of corruption allegations is as big, if not bigger, an issue than corruption itself. How common are such allegations weaponized, particularly in the local level?
Hi, Victoria. Thanks for your comment!
In my opinion, corruption is still the biggest issue.
The over-penalization of administrative errors has several possible reasons. For instance, legislatures may approve overinclusive punitive laws to demonstrate to their constituencies that they are rigorous with matters related to the Public Administration. In a similar way, law enforcers may use overinclusive punitive provisions to demonstrate to society this same rigor.
Under a Law and Economics perspective, a better approach would be to invest in education of public officials, because many administrative errors are caused by lack of knowledge instead of bad faith.
Local level governments are less equipped and often lack specialized personnel and the appropriate checks and balance mechanisms. Therefore, intuitively, I would say that local level governments are more susceptible to the weaponization of corruption allegations. However, a definitive answer would require extensive empirical research.
The weaponization of corruption allegations in the local level was a matter discussed in the public hearings of the Brazilian Congress. For more information, I would refer to the previous post: https://globalanticorruptionblog.com/2021/12/17/the-end-of-institutional-multiplicity-a-drawback-in-the-new-administrative-improbity-law/
Kind regards.
Thank you Rafael, and my sincerest apologies for my delay in response. I completely agree with you that Brazilian local governments generally suffer from a lack of mechanisms to properly prepare their employees, which may lead to administrative errors. A huge issue is the sheer size and scale of Brazil, as well as its diversity. It would be difficult to institute a uniform procedure for educating local government employees and leaders that would be equally as applicable and useful to local governments in Santa Catarina, Acre, and Bahia, for example. The solution therefore can’t be centralized – accountability should come from the local citizens themselves.
I am somewhat reminded of Brian Wampler’s 2004 article (https://scholarworks.boisestate.edu/polsci_facpubs/136/) where he examines civic organizing as a form of local political accountability through the lens of participatory budgeting. I wonder what a more updated proposal, focusing specifically on administrative errors and corruption, would look like.
Hi, Victoria. Thanks for your comment. I am really glad with it. You have a remarkable knowledge about Brazil and you made really pertinent observations.
Brazil has invested a lot in transparency. There are transparency portals on the Internet that allow citizens to oversee public expenditures. Civil society organizations are better equipped than individuals and have been performing an effective job on using this open data to supervise the several levels of governments (federal, state and local).
Recently, a coalition of government agencies named ENCCLA (National Strategy to Combat Corruption and Money Laundering) has identified the issue of unequal access to information and knowledge among Brazil’s several municipalities. As a result, ENCCLA has been sponsoring educational initiatives (courses on subjects like public procurement, ethics, and public management) to overcome it.
For more information:
https://latinno.net/en/case/3261/
http://enccla.camara.leg.br/resultados
This was a fascinating analysis, Rafael. You provide convincing defenses for both amendments serving important purposes for anti-corruption efforts. To your second argument about narrowing the scope of administrative improbity offenses, I am still curious about your assessment of this new specifically-enumerated list. I remain skeptical that a narrower list would meaningfully prevent concerted attacks against political adversaries or elevations of minor misconduct to higher offenses. I worry that law enforcement bodies would find ways to creatively interpret this list to charge certain individuals they deem a threat anyway. How narrow is this list exactly? And, do you think the amendment has narrowed the list in a way that adequately captures the forms of misconduct that really need to be targeted?
Hi, Sandy. Thanks for your comment!
Before the changes, the law had an exemplificative list of 10 punishable conducts labeled as violations of the principles of public administration. After the amendments, the law has a list of 8 conducts characterized as violations of the principles of public administration. Broadly speaking, the list of conducts includes violations of the duty of confidentiality, violations of the duty of transparency, frauds in public procurement, frauds in entrance exams for public careers, and nepotism. I think the current list encompasses important duties that public officials should observe.
It is a much narrower and well-defined list. It may not eliminate all the possibilities of using the law for other purposes, but it certainly made it way more difficult.
Some congressmen wanted to remove the entire class of violations of the principles of public administration. I think the restricted and well-defined list was a much more productive outcome.
Kind regards.
This was a really fair and insightful assessment. You identify the wasted resources that went into prosecuting corruption under the pre-amendment law, which certainly does seem like a concern. I do wonder, however, if this approach might nonetheless yield benefits that outweigh the costs. Perhaps there were a lot of costs involved in going after acts that were difficult to prove under a negligence standard, but maybe casting a wider net eventually would outweigh these costs in the long run. And if this is the case, could it be that the negligence approach produces a higher net gain than the intent approach, even if in the former there are greater overall costs? This is an empirical question, of course, and most likely there isn’t enough data to come to any definite conclusion. But I wonder how this consideration might affect your view?
Hi Logan. Thanks for you comment!
That would be an interesting empirical research.
In the following lines, I will explain my rationale about the conclusions.
Before the amendments, administrative improbity lawsuits were very costly since they took on average 6 years to be adjudicated. Furthermore, the Administrative Improbity Law imposes severe sanctions (e.g., suspension of political rights and loss of public position), and a conviction results in a considerable social stigma. Given the generated social costs, under a Law and Economics perspective, the punitive provisions of the Law should only be used when necessary. In that regard, administrative disciplinary actions constitute a much less expensive way to deal with administrative errors caused by negligence. Disciplinary actions have deterrent effect, are faster, and demand less human and technical resources.
The older version of the Law had the following disutilities: an overload of the judicial system, an exacerbated chilling effect on public officials, the potential misuse of the law to go after non-corrupt conducts, among others. Therefore, I think that is very difficult that the older version of the Law could be economically more efficient.
It would be interesting to assess the data in the future about improbity lawsuits to check if the new provisions had any positive effect.
Kind regards.
Hi Rafael,
Thanks for this a highly informative post! I also appreciated your focus on positive aspects of Brazil’s recent amendments, as opposed to an exclusive critique of these efforts. You highlight particularly well the tradeoff between flexibility in pursuing corruptive practices with the discretion to use enforcement authority for purposes other than corruption. Aside from its problems with overinclusion, how well you think the Improbity Law does to combat actual instances of corruption? Furthermore, do you think these amendments do more than just more accurately channel resources to fighting actual corruption?
Hi, Katherine. Thanks for your excellent comment!
The Administrative Improbity Law is a crucial tool in the fight against corruption in Brazil. Overall, I think the changes are positive and will make this important legal statute more effective. The two amendments discussed in the post will also alleviate the workload of the Judiciary and will likely produce a rapider response to society. Lawsuits with endless durations may cause a sensation of impunity.
There had been procedural changes that will speed up improbity lawsuits. In the past, lawsuits took a long time to be adjudicated by the Judiciary (on average, 6 years).
Therefore, in my opinion, resources will be more wisely deployed in the fight against corruption.
Regarding what I consider a major drawback of the new law, the removal of institutional multiplicity, the Brazilian Federal Supreme Court will decide the constitutionality of it. Currently, this new provision was suspended until the Court makes a definitive decision about this matter. Reference: https://www.conjur.com.br/2022-fev-18/alexandre-veta-exclusividade-mp-propor-acao-improbidade
Kind regards.
Thank you very much for your very insightful and well-written blog post, Rafael! Your analysis was spot on. Notably, your blog post showed how important practical considerations are as well as understanding the realities of a country when it comes to analyzing legislative amendments.
I have a few questions in relation to the amendments under consideration, viz.:
(1) With respect to the first amendment, you mentioned that the same does not equate to negligent public officials being left unsanctioned. In this light, do you think that the disciplinary actions under the Civil Servant’s Statute are sufficient to address the issue of negligence in the administration of one’s public duties?; and
(2) As regards the second amendment, do you know how Congress was able to come up with a well-defined and restricted list of acts concerning violations of the principles of public administration? I believe it is important to know how Congress was able to come up with such a list.
Thank you so much. I look forward to your answers to my questions.
Hi, Teri. Thanks for your comment! Excellent questions. Here are the answers for them:
(1) Yes. I think that the disciplinary actions of the Civil Servant Statute produce an adequate deterrent effect and would be a proper way to handle negligent behavior of public officials. In Brazil, lawsuits are slow and the Judiciary, as demonstrated in the statistical data cited in the post, is overloaded. This compromises the punishment function of the accountability system and creates a sensation of impunity. The administrative proceedings are faster and less costly than judicial lawsuits. As an example, in Brazil, a policy decision made to address corruption committed by legal entities was to create administrative sanctions that could be applied by specialized agencies, such as CGU (the Office of the Comptroller General) and TCU (Federal Court of Accounts).
(2) Initially, the government created a committee comprised of judges, legal scholars, and lawyers to study the reform. This commission prepared a proposal, which was used as a basis for the bill. This committee likely focused on the most common illicit conducts verified in practice. Frauds to public procurement is an example.
Best regards.
Hello, Rafael!
I would like to ask that if you think that Brazilian institutions have enough tools to find enough evidences to achieve the new evidenciary standard required by the new law, since it is now required that prosecutors should prove that the misconduct was willful.
My point is: if that is not the case, then the law will result in a significant setback.
In fact, that was precisely the critique that International Transparency made to the bill: https://comunidade.transparenciainternacional.org.br/brazil-setbacks-2021
By the way, according to some sources, the ammendments made to the improbity law would result to the dismissal of 40% of ongoing lawsuits, including some rather famous cases (https://noticias.uol.com.br/ultimas-noticias/agencia-estado/2022/08/19/cpi-da-covid-pgr-quer-arquivar-investigacao-sobre-prevaricacao-de-bolsonaro.htm).
Moreover, the bill established that there is no improbity when a public agent does not present accounts, even when obliged, if the necessary conditions to do so are lacking. Since there is no clear specification on what it is this “lack of conditions”, this might be a loophole to decrease transparency and accountability.
Last but not least, the bill has now restricted the legal concept of nepotism. According to the ammendments made, there is nepotism only when a nomination is made in order to achieve an illicit purpose. In other words, the nomination of a family member would not be enough to characterize improbity, it will be also required for the appointment to be permeated with a special, illicit purpose, which may be a quite challenging burden of proof..
Another famous case dismissed by the new legislation is this one: https://www1.folha.uol.com.br/equilibrioesaude/2022/05/mudanca-na-lei-de-improbidade-livra-pazuello-de-acusacao-sobre-oxigenio-em-manaus.shtml