Since 1990, the gap between the richest and poorest has been narrowing in Brazil. Which policies have contributed to this meaningful decline?
In this episode of VoxDevTalks, host Tim Phillips is joined by economist Alysson Portella (Insper) to discuss how Brazil has seemingly defied global trends. Between 1995 to 2015, Brazil has seen a significant and sustained decline in wage inequality. Drawing on a forthcoming paper co-authored with Sergio Firpo, Portella unpacks two decades of data and policy shifts that led to this surprising outcome.
Bucking the trend: Declining inequality in Brazil
During a period when most high-income countries saw rising inequality, Brazil saw its wage Gini coefficient drop by roughly 10 points.
“Brazil has always been like a champion of inequality in the world. Wage inequality was around 60 Gini points in the 1990s and in 1995 when it started decreasing, it was around 58 Gini points... it decreased by [an additional] 10 points [by 2015].”
This decline was highly visible on the ground. As Portella recalled:
“I remember watching TV on Saturday nights with those documentaries about people starving in the Northeast of Brazil, which is the poorest region here in the country, eating cactus and lacking water. We don’t see that anymore.”
The story stands in contrast not only to richer countries like the US, but also to Brazil’s own past. Historically, Brazil has experienced extreme inequality not just between regions, such as the poor Northeast and wealthier South, but also within cities and communities.
High-quality data reveals key economic trends in Brazil
Thanks to unusually robust datasets that include both formal and informal labour sectors, Brazil offers a rare opportunity to analyse inequality at a granular level.
“The data in Brazil actually is impressively good. So, we have been doing very good demographic census at least since the 80s. We also have annual survey data that covers the whole country that’s at least annually since the 80s. This data covers both formal and informal sector in Brazil.”
Portella and his co-author utilised decomposition techniques to understand the factors behind the drop in inequality. They found that the majority of the change was due to structural shifts — specifically how worker characteristics were rewarded in the labour market — rather than compositional shifts in the workforce.
“Most of the change was because of changes in the wage structure, so the way the market rewards different skills or different characteristics of the workers.”
Falling returns to education and experience
One of the biggest drivers of this wage compression was a decline in returns to education and experience.
“There was a very important decrease in the returns to education, but also [an] important decrease in their returns to experience. Older workers are not receiving as much as they were relative to young workers in the past.”
This reflects broader changes in Brazil’s education system. During the period studied, the country experienced a surge in college attainment and secondary education, increasing the supply of skilled labour and reducing wage premiums for such qualifications.
However, Portella cautioned that the effect of rising education is complex:
“Because there are more highly educated people coming into the workforce, they can’t demand such a high premium on their wages… On the other hand, jobs that highly educated people are getting, the dispersion of wages is higher in those jobs anyways. So, we’ve got these two competing effects.”
The role of policy: Minimum wage, trade, and gender equality
Government policy was another powerful force behind the shift. One of the most impactful policies was the sustained increase in the real minimum wage, which more than doubled between the late 1990s and mid-2010s.
“Minimum wage increased in real terms, more than 100% between the 1990s to the mid-2010s... [Studies] estimate that around half of the decline in the variance of wages in Brazil can be attributed to... the minimum wage.”
Portella also highlighted the effect of trade liberalisation in the 1990s, which reduced protection for capital-intensive sectors, thereby redirecting investment towards more labour-intensive industries that benefited low-skilled workers. Similarly, the commodity boom of the 2000s increased demand for low-skilled labour, narrowing the wage gap.
Race and gender discrimination also declined during this period, contributing approximately 15% to the reduction in inequality:
“The penalty that [women] receive in the labour market is reducing over time. We also have a reduction in fertility rates... We also observe a small reduction in the racial penalty in Brazil against Black people.”
However, Portella clarified that this historical trend was not primarily driven by affirmative action policies, which were introduced later, mostly affected access to elite education, and will take longer to unfold and understand.
Firm-level wage compression: An overlooked mechanism
Using employer-employee matched data, the researchers explored changes in how firms contribute to wage dispersion. They found a notable compression in the distribution of firm effects, that is, differences between what firms paid similar workers shrank significantly.
“The relative decrease in firms’ dispersion… is relatively much higher than the decreasing workers’ effect. This suggests that there was a compression in the way that productivity [of] firms transmits to workers’ wages.”
This reduction connects with the broader finding of falling unobserved skill premiums and structural wage compression.
Why the trend stopped and where research must go next
While Brazil’s progress was remarkable, the decline in inequality halted around 2015. Portella pointed to economic stagnation and the end of the commodity boom as likely culprits.
“It’s interesting to see that the decline in inequality in Brazil stopped more or less at the same time Brazil stopped to grow [in order] to decrease inequality.”
There remain unresolved questions — including how technological change has affected Brazilian inequality, why experience premiums declined so sharply, and how informality reacts to formal sector wage changes.
No silver bullet, but many valuable lessons
In conclusion, Portella emphasised that the fall in inequality was not the result of a single policy or event, but a combination of interlocking factors:
“There is no silver bullet… Increasing the minimum wage by itself would not have reduced the inequality. It was a result of a similar increase in the productivity of workers that is related, for instance, to their education.”
The Brazilian case shows how thoughtful, multi-pronged policies—spanning education, trade, wage regulation, and macroeconomic stability—can reshape inequality over time.