Brazil's Bolsa Família programme has been credited with helping to significantly reduce levels of absolute poverty and inequality in Brazil. Started in 2003, and funded by less than 0.5% of the country's GDP, it now facilitates small cash transfers to 46 million Brazilians whose current income places them below the poverty line. By providing the transfers, which require school attendance and health checks for children, Bolsa Família helps to support families in times of hardship, including against unexpected economic shocks and fickle labour markets.
Luis Henrique Paiva worked on Bolsa Família for five years, running it for three. A permanent civil servant, Luis has worked in the Brazilian public administration since 1998, including for the Chief of Staff in the Brazilian presidency between 2005 and 2007. He currently works as a researcher at the Institute for Applied Economic Research (Instituto de Pesquisa Econômica Aplicada - IPEA), with a focus on pensions and non-contributory social programmes.
I am currently helping to tie up the IRIBA research project, an international, collaborative effort helmed by the Global Development Institute (University of Manchester), that sought to investigate the nature of Brazil's impressive record of pro-poor, inclusive economic growth over the 2000s. I asked Luis to reflect on the success of Bolsa Família, and the implications of Brazil's current recession and recent political turmoil for the longevity of the programme.
You've worked through three presidencies, what are your reflections on their differences?
It's difficult to make comparisons. Both leading parties consider themselves centre-left. During the eight years of Fernando Cardoso's administration a lot of effort was put into stabilising the economy. Brazil had hyperinflation till 1994. During the period 1995-1998 it was quite difficult to find a good, solid ground for fiscal policies. The country took quite a long time to reach this more stable monetary and fiscal situation. A lot of reforms in social security, the labour market and in the public sector were conducted during this period. Lula and the Labour administration, starting in 2003, put a lot more emphasis on social policies. Keeping the country in a solid fiscal and monetary situation but also addressing the social questions, with social programmes etcetera. My impression was that Lula was able to put all this emphasis on social policy while at the same time keeping a solid economic foundation for the economy.
It was out of that background that Bolsa Família was started?
The very idea of conditional cash transfers emerged in Brazil and Mexico in the second half of the 1990s. Latin America is probably the most unequal region in the world, in economic terms, and so in Mexico and in Brazil, in the second half of the 90s, this idea of cash transfers with a component addressed to improving the human capital of children emerged. Firstly in local administrations in Brazil, and in 1997 as a national rural programme in Mexico. In the beginning of the 2000s Brazil had three different conditional cash transfer programmes, and Bolsa Família unified them as a more coherent, strong programme. With higher coverage, and eliminating all the overlaps that we had.
So it replaced those programmes?
Yes, that's it. Basically, it unified all the members that still complied with the eligibility criteria. Those who were members of other programmes continued to be members of the Bolsa Família programme. The programme reached 11 million families by 2006, and around 14 million families more recently. Which means basically 1/4 of the Brazilian population. This kind of programme, like Bolsa Família, or Mexico's Prospera, and other programmes like these were addressed to part of the population that was not covered by traditional social insurance and social assistance.
Can you explain the difference between social insurance and social assistance?
Social insurance is basically a kind of social protection for those that can pay for it. It's traditionally related to the formal labour market, where people contribute to social security on a monetary basis. They are covered against a lot of social risks, like unemployment, sickness and old age. Obviously, as it is related mainly to formal jobs - and in developing countries a relatively small number of jobs are formal - a large part of the population remain uncovered by these schemes. I think it is the first time that this region had a programme addressed to people that are in their working age, and their families, and children especially. Who are able to work, are frequently working, but continue to be poor and with no way to contribute to social insurance. It's the first time that social protection reached this quite large group of the population.
Was it quite a battle to convince the non-poor population that this was a valid, important thing?
This whole story about how these programmes are assessed by the non-beneficiaries is quite complicated. In the very beginning I think that there was resistance to the programme, the programme suffered a bit. But after a while, I think that we can tell this story as the story of turning good policy into good politics. These programmes in general, in several different countries, were evaluated a lot and they frequently present excellent results in terms of impacts on reducing poverty and improving educational and health indicators. My impression is that the resistance decreased a lot. But as the programme was very associated with the Workers Party, there seems to be signs that criticism towards the programme increased again.
As the results started to show?
Yes, I think so. First, there are lots of good results to show - these programmes work. Secondly, they are becoming part of the landscape. They are there. As time goes people tend to accept the idea of protecting the poor population, and having some sort of safety net for that. During the 2014 election, three out of each four Brazilian voters declared support to the programme. It used to be one of the highest rates of approval among government programmes. My feeling is that approval went down since then, but fundamentally because the programme was associated to the former Central administration. I expect that the rate of approval will increase again with time.
Why are children considered an important group to target?
Brazil built, during 70 years, a relatively strong set of social security programmes addressed to the older population. Older people are in general very well-protected against poverty. Poverty is concentrated on children; they were unprotected. As I said, this is the first time that a programme addressed families that are struggling - they are working, they are able to work, but even so they are not able to make enough money to avoid poverty. When you address this public you reach the children. That's the first part. The second part is another aspect, that we should do something for these children in the long term. If they have better access to education and healthcare they will probably have a better prospect. Both in the short and the long term we could do something for these children, using cash transfers. On the other hand is the idea of conditionality, to reinforce the right to have access to education and healthcare.
How much does each family actually get? It goes to the household, right?
Whenever it's possible it goes to the mother. The average benefit now is 180 reais [approx US $55]. Which is a very modest benefit, per family, per month. Actually we now have four different kinds of benefits. If the family is extremely poor, independent of its composition they will receive a benefit at the level of 85 reais. For every child aged between 0 and 15, a benefit of 39 reais. Adolescents aged 16 and 17 will receive a benefit at the level of 46 reais. Overall, taking into account their declared income plus all these benefits, if they remain under the poverty line, which is R$ 85 per month per capita, they will receive a fourth benefit. A top-up benefit, basically.
To bring them over the poverty line?
Is there a limit to how long the family can receive it for?
No. We work with a minimum period. If, after their first year the family find a job and their income increases above the eligibility criteria they will continue to receive the benefit for an extra two years. Just after that, if their income is still above the eligibility criteria the benefit is cancelled. The idea is to protect them while we imagine that income is not secure, not stable. To have a minimum time but not a maximum time.
What motivated that?
Poverty and extreme poverty are very dynamic. People that are chronically poor are a very small group of the poor population. Most of the families are in and out of poverty because they are in a very unstable situation in the labour market. The idea is that, if you cancel the benefit in the first moment a family is out of poverty - the family may need the benefit again in the near future. The idea is to give a minimum length of time to be sure that that situation is more stable, so they can live without the benefit.
How do you define chronic poverty?
Basically, chronic poverty is defined by the fact that people are in poverty all the time, they are not out at any time -
Over the course of their lifetime?
Yes. That's why, to define a maximum period for families to be receiving a benefit doesn't work. For quite a few of these families, probably there is just a solution in the long term, with better education for their children, better healthcare. Hopefully in the long term it will work. These children will be better prepared to face the labour market and take advantage of it.
Are the jobs there, within the formal labour market?
They used to be for a long time. Between 2003 and 2014 almost 20 million formal jobs were created in Brazil. But since then the labour market collapsed. By the end of 2016, the country will have lost 3 million formal jobs in two years.
How were these jobs created and what happened more recently in the labour market?
Mid-term solid economic growth led to job creation and formalisation. What's interesting about that is that a much larger part of the Brazilian population became protected by social insurance, due to the fact that they used to be in the formal labour market. More recently the economic foundations worsened considerably and the country is facing a deep recession. Millions of formal jobs disappeared. Unfortunately, part of the progress has been lost.
Is there pushback from the government to the informal market? It seems to be that it's quite accepted.
Even developed countries have a small share of their labour market that is this grey economy in which no one knows exactly what is happening. They don't pay taxes, but they have no social security either. But in general, richer countries have a much smaller informal economy than poor countries. There is just one way to consistently reduce informality in the labour market. Which is growing, through economic growth. You can do lots of other things to improve formality in the economy, but even so, until the country's fully developed, I truly believe that a considerable share of the jobs will be there, simply because it's impossible to be the other way.
So for you, economic growth is definitely the path to poverty reduction?
Yes. Economic growth is probably the most important part of poverty reduction. First, because it generates new jobs that can generate a secure source of income, and second, growth generates revenues for the government to address poverty through using transfers. Economic growth cannot be considered the only solution, the thing that will solve everything, but it's an important part of it, no doubt about it.
Can economic growth create poverty and inequality? Even if it reduces it in some ways?
Find a very, very poor country - economic inequality is probably very, very low. Bangladesh has a lower Gini coefficient than the UK. But the fact is, the vast majority of people in Bangladesh are poor. When you start growing from this level, because only part of the population take advantage of the economic growth, inequality is going to increase. Eventually, if you have proper fiscal policies and transfers policies in place, economic growth will bring the poor population to an acceptable standard of living, and the country will start to be fairer again, or start to have lower inequality again. That's the whole expectation - it does not mean that all the countries fit in this scheme. It depends a lot on what country you are looking at. I agree with you, economic growth can result in increased inequality, but I think that in general it's better for the country to have economic growth than not, because you'll have a lot of different possibilities. It does not mean that economic growth will be pro-poor in all different situations. Even so, it's better to have economic growth than not to have.
There is criticism of modernisation processes leading to mass rural-urban migration in Brazil, and a rise in urban poverty. What are your thoughts on that?
Brazil now has a small share of its population living in rural areas. It's less than 15%. It's not a rural country anymore, it's basically an urban country. Obviously the mass migration from rural to urban areas tends to increase urban poverty. But you can't say that the urban poor are in a worse situation than the rural poor. People tend to be very smart. They know what their conditions were in the rural areas and what their conditions are in urban areas. If they insist on being in urban areas, they are basically saying to you that, despite their situation being very bad, it's better than it was in rural areas.
There is this romanticised view of the rural life. Once the government tried to take urban poor people from Brasilia, the capital, to their home city, in the countryside - they gave them some support to stay there, and to recover their former life. After a few years they were back in Brasilia. They can't disagree that their life is bad in Brasilia, but it can be even worse in rural areas. If people in general migrate to urban areas and stay there, I respect their view. I would do the same thing if I were them. I don't believe that we are generating poverty, by developing rural areas. Probably we are revealing some sort of poverty that people who live in urban areas didn't know existed.
Brazil's very big, and I imagine that what poverty looks like in one part of the country is quite different to how it presents in another, and the same for how people deal with it. Which must be difficult when you're running a programme like Bolsa Família - you can't have one strategy to reach everyone.
Yes, poverty is unequally geographically distributed in the country. Poverty is massively concentrated in the Northeast and Northern regions, in relative terms, and in absolute terms in the Northeast region. Naturally the programme tends to be concentrated in the Northeast and North as well. It's mainly addressed to poor people, where poor people are. Based on data from the census, we have an idea of how many poor families live in each municipality. We try to reach these families. The design is the same, but it works very well in poor regions. Actually, indicators demonstrate that the programme works a little bit better in poor regions than it works in more affluent municipalities.
What's interesting about that is Bolsa Família is managed in a decentralised context. In other words, it's a federal programme, but it relies a lot on the work of local governments. A lot. One could be concerned about how poor municipalities in poor states, in poor regions of Brazil can handle a programme like it. One could be concerned that these municipalities would not be able to run the programme locally, and so the poor people in these regions would continue to suffer just because of a lack of local capacity. But when you look at quite a few indicators of how well the programme is doing in every Brazilian municipality we perceive that - for instance, let's consider that those normal errors that you have in targetted programmes are basically the error of paying benefits for those that are not entitled and not paying benefits to those that are entitled. Inclusion and exclusion errors.
All these indicators are the best possible in the Northeast region. They know better who the poor population are, they are able to find them, they are able to reach them, they are able to collect their data, and you are able to pay the benefits. These indicators are systematically better for the Northeast region than in the South. We can't see any effect of what could be defined as a lack of administrative capacity of these governments. Maybe because, for these local governments, poverty is a much more important question. It's a much more important political issue than it is in Santa Catarina, for instance. In Santa Catarina extreme poverty is almost zero. For many local mayors, it's pretty much irrelevant to deal with this question. But for a mayor in the Northeast it's fundamental. So yes, they do address a lot of energy to deal with it.
Is there much knowledge-sharing between countries within Latin America, about social protection programmes?
We have been talking a lot with other countries that run the same kind of programmes. Countries like Mexico - or Argentina, or Uruguay or Chile. There is a general tendency to share experience, with Mexico especially, because they have a large conditional cash transfer programme as well. Progresa, that later on became Oportunidades and now is called Prospera. They changed the emphasis here and there, but it's basically the same programme. We share experience with them for sure.
What do you think the implications of the recent slowdown in growth in Brazil are for Bolsa Família, being a tax-funded programme?
The programme now faces a different situation to that it used to experience. The country is in the middle of a fiscal crisis and we have a right-wing government. One could think that this is the perfect environment for retrenchment. But, so far the programme has shown resilience. Benefits were recently adjusted and the number of beneficiaries hasn't decreased. On the other hand, one could argue that, in the current economic situation, benefits should be more generous and the number of beneficiaries larger. But in the middle of this fiscal crisis, it seems difficult to believe that it is going to happen. I really expect the programme to resist. Not because there is a general agreement about how good it is, but more because welfare programmes tend to be resilient.
How serious is the current recession?
It's very serious. GDP will decrease 8 percent in 2015 and 2016, the country is running a fiscal deficit of 9 percent of GDP, unemployment peaked at 12 percent and millions of formal jobs have just disappeared. Government revenues are at a record low and expenditures continue to rise. Between 2003 and 2010, economic foundations were very solid, but these foundations have worsened steadily since then. It is essential to recover a solid ground for the economy, but I don't believe we have consensus on how to do that.
2016 has been a particularly tumultuous year for Brazil, politically. Do you think this might have an impact on Bolsa Família's capacity to survive?
I believe that Bolsa Família is now part of the Brazilian social protection system and that it will survive. Paul Pierson once wrote that the political economy of welfare retrenchment is not the opposite of the political economy of welfare expansion. Left-wing parties tend to expand welfare programmes, but this does not mean that right-wing parties will be able to dismantle them. They also keep an eye on their political prospects. Maybe this is a bit of wishful thinking, but I believe the programme will be in place for a relatively long time.
For more on social protection in Brazil, read IRIBA's working paper: 'Antipoverty transfers and inclusive growth in Brazil'.
This is an updated version of an earlier interview.