VoxEU Column Development Politics and economics

What drives protests in Brazil? Corruption, ineptitude and elections

After decades of macroeconomic stability, structural reforms and declining inequality, mass political protests unexpectedly erupted in Brazil. This column reports new empirical evidence on protests in Brazil over the long-run to shed some light on recent events. It argues that they were driven by three main factors: corruption in public services delivery, political ineptitude in the run-off to major international events, and the political-economy effects of the electoral cycle.

The conventional wisdom is that Brazilians are a laid-back people who can always find a way (‘jeitinho’) even if that involves dribbling the rules. If that was actually true, political protest would be extremely rare in Brazil. Because of almost two decades of macroeconomic stability, successful structural reforms, respectable growth rates (Amman and Baer 2012), as well as declining income inequality (Lopez-Calva and Rocha 2012) and the consolidation of democratic institutions (Feres and Kerche 2013), the mass political protests taking millions of people to the streets (that are still going on and may well continue) were utterly unexpected. The recent protests were driven by corruption, ineptitude and elections.

Blood, sweat and tears: Measuring protest in Brazil since 1870

Important recent research argues that institutions are a fundamental cause of long-run economic growth (e.g., Acemoglu and Robinson 2006). Recognising that institutions can be formal (e.g., laws and regulations) or informal (e.g., customs and traditions), this line of inquiry has moved towards identifying which ones are the most important institutions and how they interact with each other. Economic historians excelled at this and highlight the role of both political institutions and financial institutions (Haber 1998, Musacchio 2009, Haber et al. 2008).

In a recent paper, my colleagues Menelaos Karanasos, Jihui Zhang and I provide new econometric evidence on the dynamics of political protest from 1870 to 2003 in Brazil (Campos et al. 2013). We construct a new data set on formal and informal political institutions in Brazil and study how these institutions relate to per capita economic growth. We construct eight new measures of ‘formal’ (i.e. government crises, legislative effectiveness, legislative selection, major constitutional changes, size of the cabinet, number of cabinet changes, purges, and changes in the executive) and seven new measures of ‘informal’ political institutions (anti-government demonstrations, political assassinations, general strikes, guerrilla warfare, coups, riots, and revolutions). These variables were chosen mainly because growth research often uses them and, as most originate from the Arthur Banks data set, economists recognise these variables, are familiar with their definitions, advantages and limitations. Two well-known limitations are that data start in 1919 and exclude all world war years. Based on archival research and the extensive historical literature (which includes, among many others, Abreu and Lago 2010, Bethell 2008, and Fausto 1986), we constructed new time series based on the definitions from Banks that fill in the periods 1870 to 1918 and 1939 to 1945 and also allows us to check the accuracy of the original (shorter) Banks series.

What do these new data reveal? One message is that there has been substantial protest in Brazil since 1870. Clearly, this squares badly with the notion of a laid-back people. One can see considerable, significant and dramatic changes in both formal and informal political institutions.1 Figure 1 shows, for example, data for riots and anti-government demonstrations revealing the extent of protest that has taken place in Brazil since 1870.2 The figure also suggests that their trends may have reversed, with non-violent anti-government demonstrations becoming more, and riots (violent demonstration) becoming less prevalent over time.3

Figure 1. Number of riots and anti-government demonstrations in Brazil from 1870 to 2003

Source: Campos, Karanasos and Zhan (2013).

We take these data to a simple estimation framework developed in Campos, Karanasos and Tan (2012). This framework generates, from a combination of the Power GARCH and Pesaran-Shin ARDL models, four types of effects:

  • Direct (on growth).
  • Indirect (on unanticipated growth volatility).
  • Dynamic (short and long-run).
  • And structural breaks.

We find evidence for negative direct effects on real-GDP growth from changes in both informal and formal political institutions. We also find that almost all measures have negative effects on per capita output growth in the short run, but almost none display long-run impacts. The analysis suggests that legislative effectiveness and cabinet changes seem key among formal institutions, and anti-government demonstrations among informal.4 

Past, present and future: What explains the current mass protests?

The evidence above suggests that the propensity to protest in Brazil is not low and, if anything, seems to have increased over time. Although this makes the recent protests less surprising, they were still largely unforeseen. Why did they happen? I can venture three main, somewhat related, reasons:

  • Perceived inefficiencies and corruption in public services delivery.
  • Political ineptitude in the run-off from major international events.
  • And political economy (electoral) factors.5

One first reason is the widespread perception of corruption in public services delivery. Recent research tends to support this perception: for example, Ferraz et al. (2012) report a negative association between corruption and the performance of primary-school students. Brazil will host the 2014 football World Cup and the 2016 Olympic Games, among other major events. When awarded, the agreement was that the private sector would cover substantial costs and that there would be new investments in public services. It is now clear that none of these will take place. Moreover, it is also clear that the benefits from these projects may become more concentrated than expected. The increase in bus fares was brandished initially as the trigger but the deterioration in other services (health, public safety, etc.) may have been equally important.6 

The second reason is political ineptitude in the run-off to major international events. Increasing bus fares just before a major international event was not smart. Yet it was also the culmination of a number of decisions contrary to public opinion. One example is the government of the State of Rio de Janeiro deciding to knock down the Maracana swimming complex and the Museu do Indio (Native-Brazilians Museum) in order to build new car-parking structures (which will be needed given the expected low investment volumes in public transportation.)

The third reason is the electoral cycle. The next presidential election is in October 2014 (between the World Cup and the Olympic Games). The economy has cooled off and the prospects for next year are mixed (Canuto 2013.) If not growth, political reform may be called upon to shore up the re-election campaign. Although presidential approval ratings declined slightly before the protests, with the protests they collapsed by 27 percentage points (57% before, 30% after), substantially increasing uncertainty about the results from this election. The swift reaction has been a set of proposals (by the president) including a referendum on political reform, oil royalties to education, and new health and transportation investments. The president seems to genuinely sympathise with the protester’s demands, which bolsters the protest movement but also antagonises with the legislative. This strengthens the risk of public-order disruptions in front of the world media during forthcoming major international events and suggests these protests may not be as short-lived as many believe.


Against the stereotype of a laid-back and peaceful people, the historical record suggests the propensity to protest in Brazil is high and may have increased in the last decades. The current wave of protests has multiple causes but three important ones are corruption and inefficiency in public services delivery, political ineptitude and the electoral cycle. These make for the possibility that protests may well continue as the executive and the protesters push for political reform and improved public services against forces that are well represented in the legislature.


Abreu, M and L Lago (2010), “A Economia Brasileira no Império:1822-1889”, PUC-Rio de Janeiro, Department of Economics, Discussion Paper No.584.

Amann, E and W Baer (2012), “Brazil as an Emerging Economy”, Brazilian Journal of Political Economy 32 (3), 412-423.

Acemoglu, D and J Robinson (2006), Economic Origins of Dictatorship and Democracy, Cambridge, Cambridge University Press.

Bethell, L (2008), “Politics in Brazil under Vargas, 1930-1945”, in L Bethell (ed.) The Cambridge History of Latin America, Volume IX, 3-86, Cambridge, Cambridge University Press.

Campos N, M Karanasos and B Tan (2012), “Two to Tangle: Financial Development, Political Instability and Economic Growth in Argentina since 1890”, Journal of Banking and Finance 36(1), 290-304.

Campos N, M Karanasos and J Zhang (2013), “Institutional Change and Economic Growth in Brazil from 1870 to 2003”, Brunel University, mimeo.

Canuto, O (2013), “Brazil: Chasing Animal Spirits”, Economonitor Blog, 18 June.

Chao, L (2013), “Brazil Protests Prompts Shift in Media Landscape”, Wall Street Journal, 29 June.

Fausto, B (1986), “Brazil: The Social and Political Structure of the First Republic: 1889-1930”, in L Bethell (ed.), The Cambridge History of Latin America, Vol. V: 779-829, Cambridge, Cambridge University Press.

Feres Jr, J and F Kerche (2013), “Social Unrest and Political Reform in Brazil”, NACLA Blog, 4 July.

Ferraz, C, F Finan and D Moreira (2012), “Corrupting Learning: Evidence from Missing Federal Education Funds in Brazil”, Journal of Public Economics 96(9-10), 712-726.

Haber, S (1998), “The Efficiency Consequences of Institutional Change: Financial Market Regulation and Industrial Productivity Growth in Brazil, 1866-1934”, in J Coatsworth and A Taylor (eds.), Latin America and the World Economy Since 1800, Harvard University Press.

Haber, S, D North and B Weingast (2008), Political Institutions and Financial Development, Palo Alto, Stanford University Press.

Lopez-Calva, L and S Rocha (2012), Exiting Belindia: Lessons from the Recent Decline in Income Inequality in Brazil, Washington DC, World Bank.

Musacchio, A (2009), Experiments in Financial Democracy: Corporate Governance and Financial Development in Brazil, 1882–1950, Cambridge, Cambridge University Press.

1 Various structural breaks in these series were uncovered, and were accounted for in the econometric estimation, in Campos, Karanasos and Zhang (2013.)

2 Banks defines riots as any violent demonstration involving more than 100 people, and anti-government demonstrations as peaceful public gathering of at least 100 people for the primary purpose of displaying or voicing their opposition to government policies or authority, excluding demonstrations of a distinctly anti-foreign nature.

3 Coups and revolutions were more prevalent before World War II than anti-government demonstrations and riots.

4 Further work (still preliminary) shows that political and financial institutions are more important drivers of Brazilian growth since 1870 than other factors highlighted in the literature (such as trade openness, government expenditures and budget deficits.)

5 Mind that because the protests are still ongoing, there is unfortunately little hindsight to benefit from. Moreover, we must at least mention the important role social media has played in coordinating and magnifying the protests in Brazil (Chao, 2013).

6 For instance, crime may have risen in the neighbourhoods in which Confederations Cup football was not played or far from international media and national teams’ hotels. We do not yet know whether data will be able to support this, but some anecdotal evidence exists: the Law School of the Federal University of Rio de Janeiro is not in a prime location. Law students were organising, weeks before any mass protest, to complain about the “sudden” lack of public safety and increase in robberies.

4,094 Reads