Global Development Institute Blog

By Chris Lyon, PhD researcher at the Global Development Institute. The working title of his thesis is Exploring a relational conception of social justice: liberals, radicals, and Brazil’s ‘new social contract’.

Much of IRIBA’s research has been cautiously optimistic, analysing changes that saw Brazil embark on a path of pro-poor growth and consolidate its newly redemocratised institutions. Now, the distinctive development ‘model’ that emerged in the early 2000s faces its first major challenge, in the shape of the current political crisis.

A key aspect of this crisis is corruption; widespread, elite-level, blatant, endemic. As this working paper details, the main corruption problem in Brazil is ‘grand’ corruption, as opposed to petty street-level shakedowns. This is bad for business, growth, government performance, social solidarity, inequality, democracy. Corruption is bad news. However, in Brazil, extremely low conviction rates have historically meant almost de facto impunity.

Matters have recently come to a head. While economic downturn is one tidal force behind the wave of general dissatisfaction that culminated in the removal of President Dilma Rousseff in mid-2016, the more proximate cause is public disgust at a conveyer belt of sordid revelations of greed, dishonesty and abuse of power among Brazil’s political and business elite. The media has succeeded in corralling much of the outrage in the direction of the Worker’s Party (PT, in power 2003-16), but the situation is not that simple. Corruption is rife across the political landscape; no less than 48% of senators and 55% of federal deputies face corruption charges, as do many business figures. New President Temer and allies face considerably more serious and personal corruption allegations than Rouseff.

Each new day seems to bring another bombshell from the Lava Jato anti-corruption probe, with a demoralising and sickening effect on the public. Sat in stationary traffic in Rio de Janeiro recently, my taxi driver looked up from the latest of these allegations, splashed across his newspaper, to declare that the only solution was probably the return of military dictatorship. Is he right?! Let’s hope not. One strand of IRIBA’s research has analysed both the extent and nature of corruption in Brazil, as well as efforts to tackle it – which are beginning to show ‘green shoots’. These in turn suggest ways forward, and also hold potential lessons for other countries fighting similar problems.

Federal government departments neatly lined up on the Esplanada dos Ministérios in Brasília. Photo by author.

In this article for the special issue journal collating much of IRIBA’s research, Lindsey D. Carson and Mariana Mota Prado shed light on both how to understand corruption in Brazil and elsewhere, and what to do about it. Corruption, they argue, has typically been conceived as a ‘principal-agent problem’. This picture involves a ‘principal’ (P), who entrusts an ‘agent’ (A) with power. The principal could be a higher-ranking superior in a department or in a business, but could also be the public insofar as the public bestows elected officials with power. In this model, cases of corruption occur when the principal lacks sufficient information to oversee the agent’s activity, and thus the agent is unaccountable and is able to abuse their power. So, a member of staff hoodwinks their boss, or a government minister trousers campaign finance unbeknownst to voters. Consequently, measures informed by this model have aimed at changing the behaviour of and incentives open to ‘agents’.

However, as Carson and Prado argue, there are major questionable assumptions here. One is that the ‘principal’ is well-intentioned, whereas the ‘agent’ acts out of pure self-interest. What about cases where, for instance, the CEO (‘P’) of a private company that takes on government contracts actually benefits from and facilitates corrupt practices among subordinates (‘A’s)? A second questionable assumption is that, in cases where the power-entrusting ‘P’ is the public, when a politician’s misdeeds become widely suspected, he will be held accountable; for instance, by removal from office. But what if the public sees little point in booting out particular corrupt politicians because the replacements queueing up are just as bad? In other words: what about when everybody’s at it?

In such cases, the authors argue, corruption has become systemically endemic, and we face not a principal-agent problem, but a collective action problem. In this sort of scenario, a strategy that appears to be individually beneficial contributively leads to an outcome that is collectively inferior. Society overall would be better off if all individuals eschewed corruption, but particular individuals making a lonely stand might well be worse off. Partially this has an intersubjective basis; everybody expects everybody else to engage in corruption. For any individual, taking the leap of faith into more principled behaviour threatens significant costs – ostracism; risks of tackling the powerful; foregoing access to resources that competitors access corruptly – with few benefits. Even somebody who might otherwise ‘want’ to act honestly may well not, under these constraints. As such, framing this as a collective action problem helps us understand how corruption can become a durable and self-reinforcing feature of a society.

It also helps identify different possible anti-corruption strategies. In particular, Brazil’s recent experience suggests the possible benefits of ‘institutional multiplicity’ in anti-corruption. Institutional multiplicity is where multiple institutional arrangements or options exist within a given field, such that different institutions’ functions may overlap, duplicate, or even compete with one another. This is not necessarily a positive feature, and in some contexts is argued to lead to inefficiencies and conflict.

However, Carson and Prado argue, in the case of endemic corruption, institutional multiplicity can help fight the problem – now re-understood as a collective action problem. Powerful stakeholders may be invested in existing institutional status quos and have little incentive for change, but institutional multiplicity can develop subtly without risking intense political resistance from such vested interests. Moreover, it reduces the need for powerful ‘champions’ in existing elites to step forward and lead anti-corruption drives, by making it easier for less powerful actors to strategically move around different institutions to effect change. This is particularly useful given the possibility of an organisation with a monopoly on anti-corruption efforts itself becoming compromised. Institutional multiplicity also can mean simply more likelihood of catching offenders, and that potential offenders are threatened with an overlapping array of possible sanctions. Functional overlap may be the best strategy in the face of the inherently complex and secretive phenomenon that is corruption.

This is, of course, not a silver bullet, but, as the paper argues, in Brazil’s case the initially rather uncoordinated simultaneous empowering of several anti-corruption institutions has actually bolstered anti-corruption efforts, fostering collaboration, complementarity and positive reinforcement among initiatives.

The current crisis is painful and demoralising for Brazilian citizens. In particular, former president Lula is the figurehead of significant advances against poverty and inequality, and left office in 2010 with approval ratings almost unheard of for political leaders anywhere in the democratic world. He remains a much-loved icon for many. Should the ongoing investigations into his affairs during and after the presidency reveal significant wrongdoing, serious emotional hurt would be inflicted on the national psyche. Even if they don’t (he denies all charges), trust in political leaders and institutions in general is at rock-bottom.

However, a crumb of comfort may be found in the fact all this might at least reflect what Carson and Prado term the anti-corruption ‘web of accountability’ attaining effectiveness, thanks in part to its institutional multiplicity. In my personal view there are also irreducibly political factors that further complicate the picture: what interests drive particular anti-corruption movements, targetting whom, at what time, and why? But, at least, Brazilians should have no need to call for the military anytime soon.



Note: This article gives the views of the author/academic featured and does not represent the views of the Global Development Institute as a whole.