Fabio Pulizzi: 00:09
Hello, this is How to Save Humanity in 17 Goals, a podcast brought to you by Nature Careers in partnership with Nature Water.
I’m Fabio Pulizzi, chief editor at Nature Water. This is a series where we meet the scientists working towards the Sustainable Development Goals agreed by the United Nations and world leaders in 2015.
Since then, in a huge global effort, thousands of researchers have been using those targets to tackle the biggest problems that the planet faces today.
In episode 10, we look at Sustainable Development Goal number 10: to reduce inequality in and among countries. And meet an academic works to improve opportunity to level the playing field.
Francisco Ferreira 01:03
So I’m Francisco Ferreira, and I’m the Amartya Sen Professor of Inequality Studies at the London School of Economics, where I also direct the International Inequalities Institute.
Sustainable development goal number 10 is about reducing inequalities, both within and among countries. And it’s got, you know, like all the Sustainable Development Goals, it’s got different targets and indicators and so on.
As to how aware I am of it when I’m conducting my research, you know, I’ve been studying inequality and working on inequality since before the Sustainable Development Goals. But obviously, when they came up, you know, they’re a huge sort of unifying force, a kind of a rallying cry for policymakers around the world.
And so that did become important to those of us working on it. And in fact, I was a little bit involved in the, in the design of some of the targets, because I was working at the World Bank at the time.
And this concept that target 10.1 has, which is about sustaining income growth for the bottom 40% of the population at a rate higher than the national average, is something that we came up with at the World Bank under the heading of “shared prosperity.”
Yeah, inequality can mean different things to different people. But I think there are two main definitions that I like to think about.
One, which is perhaps the primary definition, it’s just a measure of dispersion in a distribution of something among some group. So the distribution of income amongst people, or the distribution of years of schooling amongst countries, or the distribution of wealth amongst households.
So we just measure how far apart those incomes or wealth are from one another, or from the average. And that’s really what most inequality is. Now the secondary definition is that people say, individuals or countries that have other characteristics as well.
So if we’re thinking of individuals they have races or genders or ethnicities. And so we can take the same measures of inequality, but among groups, rather than between people. And whereas the between people is often called vertical inequality, the measures amongst groups are often called horizontal inequality.
And one is not better than the other. They’re just different. And they inform us about different things. But often when people think about inequality between men and women, for example That’s an example of horizontal inequality.
So as I was saying, inequality is, is about how things are distributed, right? So in a sense it’s important to understand the causes of that. Because nobody lives on the average, right? If you think of yourself, you’re not exactly on the average, whoever you are listening to this, you’re not exactly on the average, or maybe close to the average, and maybe richer than the average and maybe poorer than the average.
So averages don’t matter on their own. When a country produces a certain amount of output or generates a certain amount of income, what really matters in the end is who gets to benefit from it, to what extent.
And there’s quite a lot of evidence, scientific evidence based on experiments, on experiments with people and experiments with monkeys. And there are surveys. There’s all kinds of different evidence that on the whole people prefer equality to inequality. Even monkeys get uncomfortable if they see an uneven distribution of food in their group.
And human beings tend to have a preference for fairness, if not exactly equality, at least for fairness. And so because of those preferences, and because also inequality can offset how an economy functions. You know, if a lot of people are very poor, they can’t get their children into good schools, or they can’t invest in their own learning or in owning their own business ideas, and the economy ends up being less vital and less vibrant than it might otherwise have been.
So both for intrinsic reasons that we care about fairness, and also instrumental reasons that, you know, more equality tends to be associated with better functioning outcomes for everyone, it’s important to understand where that inequality comes from, and whether we can or should do anything about it.
Francisco Ferreira 05:48
I was born in São Paulo, Brazil. São Paulo is the largest city in Brazil, the largest city in South America. And Brazil is a incredibly unequal country.
And I grew up in a middle class to upper middle class household in São Paulo, and I was exposed very early on to the huge gaps, right. So I would drive around or drive to school, for example, and, and see people my age selling chewing gum on the streets.
And it occurred to me that there was something odd about that. And I had a sense straightaway that it wasn’t only about how different we were then, but that our whole lives would be different.
So I guess I was interested from very early on in ideas of justice, and, and ideas of what determine the distribution of life chances and outcomes in Brazil. And from there I ended up going to the London School of Economics where I did my PhD, which was very fortunate because at the time, inequality wasn’t a central question in economics, so much as it had been in the classical days.
But in the 1980s and early 90s. It wasn’t yet but. But at LSE it was. And we had some really important thinkers and professors who had been there like Amartya Sen or Tony Atkinson. So I learned a lot from these these guys. And it was hugely important. Then I spent a lot of my career at the World Bank where we in the research department worked a lot on the measurement of global poverty and global inequality.
And so I worked there for many years before eventually returning to my alma mater, to the LSE, now as a professor on this topic.
Inequality of opportunity is basically differences in in people’s life chances for reasons that they don’t control.
So for example, when the COVID pandemic hit the UK, and, you know, 17% of all workers in April 2020, lost their jobs, there was a almost five percentage point difference in terms of women having lost their jobs, five percentage points more than than men. You know, another example is in South Africa, in 2017, the average income for a white household was 5.6 times larger than an African household.
So these are examples of, these are examples of huge differences that are determined by factors that people don’t control themselves like race, or gender, or parental background, or where you come from.
So inequality of opportunity is about differences that we observe in society, which are not due to, say, “Ah this person worked harder than the other one, or was more responsible than the other one.” But really have to do with things we inherit, and cannot control race, ethnicity, family background, where we were born, our earliest experiences, and so.
There are a number of people out there who are really interested in an opportunity. But the reason I think it deserves more attention than it gets is that in a sense, I like to think of inequality a little bit like cholesterol, you know, there’s two kinds. And one is worse than the other.
You could think of inequality of effort or inequality that rewards responsibility in some senses, the good cholesterol. The really bad cholesterol, the really bad inequality is the one that we inherit, that we’re not responsible for.
And of course, it shapes also how much effort we put in and how much responsibility we have. I’m not suggesting and at least that they’re, they’re separable in that easy way.
Of course, we are shaped by the circumstances we inherit. So quite a lot of the inequality that we see really is inherited and is inequality of opportunity.
And it’s the bad cholesterol in two senses. It’s the bad cholesterol in the sense that it is the one that people find most objectionable. And again, without getting into the details, there are experiments out there. There’s a lot of work done by people in Norway, for example, in Bergen, at the Fair Institute, they’re looking at, you know, how do people in games, how do players in games with real money. What do they really object to, when are they prepared to pay to have a more equal outcome in their game?
And it’s typically when the inequality arises from something the players do not control, like, they will randomize, they were randomly given a lower wage rate and a game or something like that.
So there’s a lot of evidence that people object to that kind of inherited or, or or, you know, arbitrary inequality that we cannot control. It’s also the bad cholesterol in the instrumental sense, how inequality can harm the economy as a whole.
Because, you know, the example I like to think of this as a Brazilian is, if you think of all the kids growing up in the Brazilian slums, in the hills in Rio, or in various areas in São Paulo, or other cities, you know, how many engineers and scientists and great writers and so on, that could have been there, and some of them will make it, but many, many fewer than would otherwise make it. Just because of the quality of the schools they are confined to, and other hardships and having to get out of school early to help their parents, all sorts of other other things that happen to them. Crime, violence.
So there’s a lot of wasted human talent, a lot of human potential that goes to waste, because it is not given those opportunities that other that other people have. So inequality of opportunities is, you know, the bad cholesterol in both of the senses. And so I think it deserves even more attention than it already gets.
Francisco Ferreira 11:58
Another reason, perhaps why people talk less about inequality of opportunity than, than I think we should, is because there isn’t yet full agreement on on how we should measure it, and on the idea of data that, that we need to observe it. You know, in a sense, inequality of opportunities quite close as a concept to the inverse of intergenerational mobility.
So there is a lot of work out there on on measuring the association between parents and children, say the incomes of fathers and sons or mothers and daughters.
And that association measured in different ways and elasticity, or a correlation, or a rank correlation, those measures are popular and are, are often widely available.
They’re very close to what inequality of opportunity seeks to measure because it is about the influence of inherited circumstances on life today. I find that by looking only at income, we’re ignoring some of the other factors that are important in shaping opportunities.
Parental income is hugely important. There is evidence, there are a few studies that suggest that in many countries race, or parental wealth, are important independently of income, in addition to income. Economists say even when you control for income, these other things are still important. And so you’d like if you’d like in some sense, to get a measure of the overall extent to which inequality that we observe today is inherited these past circumstances, you’d like to observe things beyond income and to go beyond the studies just mobility, which are very important and informative in their own right. But you’d like to go beyond that.
But then we get into this question of, in some sense, where do we stop? Does any dataset contain all of the information that we need? And how do we put that together? And there are some statistical issues, but they pose some challenges. And that’s some of the areas in which my group at LSE has been working.
Francisco Ferreira 14:26
In the end, we’ve come I think, to accept that these measures of the extent to which inequality that we observe is inherited, are never going to be perfect, because you’re never going to observe all of the data on circumstances.
So we need to think of them as flaws, as indications of at least how much of the inequality that we see today, is inherited. And now we have some different techniques and approaches to just select the circumstances that are most important.
And these use machine learning tools. These techniques are not magic, that they’re just designed to look at a data set and extract the most powerful predictors, the most salient divisions in society in some sense.
These are machine learning techniques, I mean, in some very basic sense, they are artificial intelligence in the sense that they are supervised learning, you know, the computer is learning from, from the data that it’s, that it’s looking at. What it does is fundamentally, look at the way income, say is distributed and how it’s related, how it’s associated with different characteristics.
So when it looks at South Africa, for example, just looking at the sample, it finds, it’s very heavily associated with race.
So it’ll use race as a first splitting point. And then it it keeps going. And so then it will find that, you know, mother’s education is actually the thing that’s most associated in that area. Father’s education in their other area. Maybe occupation, maybe place of birth, and it keeps looking for the most salient variable that is inherited, that is beyond people’s control, that seems to predict income differences.
And it tells us then, how to partition the population in a way that is most predictive of income, most salient. And from that division, we can we can obtain thesei measures of inequality. And that’s what they’re designed to do.
And when we apply those to, to this problem, we tend to find much larger numbers. In a recent paper we did for Latin America, we find an average about 50%. And some countries in the 60% range. And for South Africa, we find a number nearer 70%, suggesting that, you know, even with the imperfect, imperfect information that we have, in a country like Brazil, or Guatemala, about two thirds of the inequality that we observe, is fully inherited.
And that’s likely to be just a base. And in South Africa, the two thirds really are inherited. In richer countries, in Europe, in the United States, the numbers are a bit lower. In some of the more egalitarian European countries, they are in the 20 to 30% range. In the United States, they are under sort of 40% range.
And then so again, is just an indication of, at a minimum, how much of the inequality that we observe is deeply unfair.
The second one is really about whether there’s a relationship between these kinds of inequality and the aggregate performance of the economy.
For example, economic growth. And there are a number of studies of that, some of which have found quite a substantial impact, actually. There’s a very nice study in the Journal of Development Economics by by two Spanish economista, Mateo and Rodrigues. who find that if they look across states in the US, and the US Census data for the states over a 30 year period, and they look at growth in in GDP per capita of each state.
And they find that if you look at the association between just aggregate overall inequality, income inequality, and, and growth in the States, they don’t find much.
But when they separate out between these two components, right, the bad cholesterol, inequality of opportunity, and the residual, some of which is an equality of effort, and some of which is just things we don’t observe, and in terms of circumstances. But anyway, when they split it between those two, they find it as something of a positive association between inequality of effort and growth, and a very negative association between inequality of opportunity and growth.
That is to say, those states, where more of the distribution seems to reflect inherited factors like race, like differences in family background, and so on, and so forth, grow more slowly than those states, which have less of that kind of inequality.
And this has been replicated in Brazil and a number of other countries And then there are other kinds of studies that don’t use this kind of cross country regression method.
But there have also found, for example, that the opening up of opportunity in the medical and legal professions in the United States, to women, and then to African Americans, over the course of the 20th century, contributed a great deal to growth in the output of those professions. That also has model-based simulations in it and so on.
But it’s a, you know, it’s an influential paper that also reflects this idea that that more opportunity means more talent can be used, and out of that comes more production and more efficiency.
There are at least two basic ways in which we can use that information on measurement that we have created. To reduce the rate, right, to help generate policies that will make a difference and improve the lives of those of the bottom.
One is simply targeting if you like, I mean, what is simply getting a map, a social map of society. So these, these trees, they generate, effectively a map of social groups. And we can see what their average income is and what their distribution is.
And we can see which groups, again determined purely by circumstances they inherited, which groups do systematically worse than others and persistently so.
And then presumably, if you wanted to target a program with early childhood development, or a program of improving teacher practices in certain schools, you might want to do it in the, in the areas where a majority of these people live first. And you may want to put most, most resources into those groups.
Because the other thing about inequality of opportunity is that we know that people who have the least advantage, you know, whose circumstances hold them back the most, in a sense, those are the people for whom it will be hardest to advance.
So in some sense, they should get not just equal treatment, but better treatment, because the intervention needs to compensate for the drag of the background, if you understand so.
So this is one way. And in fact, some of our funders and foundations that we work with are very interested exactly in that aspect of this kind of social map of societies in Latin America in particular, that we’re working on. That’s one way, there’s another way, which is actually through, in some sense, through the political process, which is, we feel that by highlighting the extent to which inequality is unjust and reflects things that have nothing to do with people’s responsibilities that people cannot control, that people simply inherit, we may contribute to an understanding of just how unfair inequality is, and therefore enhance political support for redistribution more broadly.
And there are many places in the world. Nowadays, in Europe and the United States, there’s already a great deal of understanding of the costs of inequality, which was not necessarily true in the US 30 years ago. But it is more so now after the, you know, the 1% Movement and the global financial crisis and Tomhas Piketty’s work and so on. But in many places, you know, I remember talking to an African president once when I was at the World Bank.
And he wanted to reduce poverty, he wanted to understand what more he could do in policy terms to accelerate poverty reduction in his country. And, and when we talked about inequality, his eyes glazed out a little bit. And he said, “Well, you know, inequality is not something we can worry about in Africa, because, you know, we have so little. We have to, we have to work our way from the bottom up for everyone. Handouts won’t work.”
And then when you sort of get across to him, the idea that actually by giving more opportunities to some of these poorest people, their contribution to the economy, their own output, their own incomes, what they do for themselves, will improve. It’s not about handouts, it’s about chances to produce, chances to lift yourself up.
You know, he changed his whole outlook. And in fact, I don’t want to give too many details. But in fact, I’m told by my former World Bank colleagues that they were policy changes and programs created, in part as a result of his being convinced that that there was something and I’m reducing inequality that went beyond handouts and had efficiency consequences.
Francisco Ferreira 23:58
So the thing about sustainable goal number 10 is it doesn’t have a specific numerical target like goal one has, right. So goal one says, well, this this the pure UN version of goal one says completely eradicate extreme poverty by 2030.
The slightly more realistic World Bank version of that says, let’s reduce it to 3% or below. And that goal will almost certainly not be met by 2030, in part, because of COVID, which represented a regression and a slowdown in poverty reduction. But in part, actually, because of inequality, interestingly enough. Because in some sense, the poverty that remains is the poverty that is hardest to reach, is the poverty in the poorest countries in the poorest people of the poorest countries.
And it’s harder to reach those parts of the poverty map if you like. But I think if we look at inequality between countries and inequality within countries, it will be difficult to conclude that it has been very satisfactory match.
Because there are many countries where inequality has grown over the last years, there’s a sense sometimes that inequality has grown everywhere. And this is actually not true. There are many countries where inequality has fallen. Pretty much however you measure it, a number of those are actually in Latin America.
But there are also many, many countries where it has grown, no matter how you measure it. And so the inequality within countries, and this is true in the US, and it is true in a number of other countries that, you know, we won’t be able to say that we met that.
In terms of inequality. between countries, there was actually a long trend of convergence and inequality, between countries driven primarily by the growth of places like China and India, which for the last half century or so has been much faster than than in the West. COVID produced a little bit of a setback to that, particularly because of India.
But by 2030, the process may have resumed. The issue there looking forward is a different worry. And that worry is now that China is around the middle of the global distribution, and its growth in the future will no longer contribute to a reduction in inequality, but in fact, possibly to an increase in between country inequality, then these trends may change considerably but that’ll be likely after 2030.
Fabil Pulizzi: 26:53
Thanks for listening to this series How to Save Humanity in 17 Goals.
Join us again next time when we look at Sustainable Development Goal number 11: how to improve our cities. See you then.
