How can we improve social mobility?

Chicago Booth’s Amir Sufi and University of Chicago’s Ufuk Akcigit and Ariel Kalil discuss how early-childhood development, innovation, and inequality affect social mobility

Sep 19, 2017

Sections Economics Video

Collections Social Impact

How do the early-childhood years affect your ability to climb the socioeconomic ladder?         

Kalil: Children need both cognitive and what we call noncognitive skills to succeed in life, to perform well in school, to perform well in the labor market, to get along with other people. These skills are crucial both for improving social mobility and for mitigating inequality. 

Although there is opportunity for skill development throughout the life course, the early years are critically important. Early childhood is a period of rapid brain development, when the path to skill development is laid down. So it’s a formative period, and parents and the home environment play an important role in helping children acquire those key skills. Children’s ability to think and to solve problems, as well as their preliteracy and premath skills, and their understanding of the world and how to get along with others all have to do with the quantity and quality of time that caregivers spend with children on particular activities, such as reading books, playing puzzles, simply talking to children, helping them solve problems, etc. 

That kind of interaction typically comes from parents, but it could come from anyone. It need not be only the mother; it need not be only the father. There is a range of important caregivers, typically, in a child’s life.

How does innovation relate to social mobility?

Akcigit: Along with colleagues, I’m studying the importance of innovation for the macroeconomy. And our overarching research question is: Why do we care about innovation, from a macroeconomics standpoint? And we find three answers to this question.

First, innovation is the sole driver of long-term economic growth. In the long run, economies are growing only due to technological progress and innovation. The second answer we find is that when we look at the relationship between well-being, or happiness, and innovation, we see a strong association between innovation at the regional level and the happiness of the region’s individuals. And the third aspect of it is that innovation is leading to growth in average incomes, but how does that growth evolve across generations and how does it get spread? And what we see in the data, starkly, is that innovation, or the turnover in the economy, is strongly associated with social mobility. In regions where there’s a lot of turnover, there’s also high social mobility. And this is particularly true if innovations
are coming from new entrants, young entrepreneurs. 

And one more interesting result we find is that this strong association between entrant innovation and social mobility gets much weaker in regions where incumbents are spending more money on lobbying activities. Because, as you can imagine, innovation is a process of creative destruction, where new entrants are coming and replacing the incumbents. And this, of course, creates the chance for the child of an assembly-line worker to become the business owner. If incumbents are slowing down this process, that is going to have a negative impact on social mobility.

In a parallel study, we try to also understand how politically connected firms affect innovation. Where incumbents are politically connected, where they are hiring moonlighting politicians, we observe a much lower entry rate into those industries. And, of course, that is going to have negative consequences on social mobility as well.

Who are the innovators? What do we know about them?

Akcigit: What we find across numerous studies is that, first, education is an extremely important ingredient for becoming an inventor. The second important ingredient is having rich parents—that helps a lot. But this strong association between having rich parents and becoming an inventor disappears completely once you control for a child’s education. Parental income matters because [rich parents] are able to afford a better education for their children. So an immediate conclusion suggests the importance of education for innovation, which later on also leads to social mobility.

Kalil: A child’s skills are jointly produced, however. We can’t assume that a lack of opportunity for learning in the home environment can be made up for by schooling. For one thing, children who arrive at school with fewer skills will get less out of the learning environment that is provided to them there. Secondly, the sheer amount of time that children spend in home environments, or in the company of caregivers, trumps the time they spend in school. There certainly can be cases where there’s a fabulous school environment; but in general, we should not think that better education is a solution for improving social mobility on its own. We can certainly think about a joint effort. But it’s really not one versus the other; it’s those two environments combined.

How does inequality relate to social mobility?

Sufi: Social mobility is a dynamic concept referring to how people move throughout the income or wealth distribution, whereas inequality is more a static statement about, currently, how does the top 10 percent of the wealth distribution compare to the bottom 10 percent? They’re related in that we might think high inequality may exacerbate the decline in social mobility, but the evidence is a bit mixed on that. 

Akcigit: If we have a lot of social mobility, that allows new entrants to catch up with incumbents. If there’s a lot of turnover in the economy, that’s going to have a negative impact on inequality because it’s going to shrink the income distribution. On the other hand, as I highlighted earlier, there is a connection between being from a rich family and becoming an inventor, and that has an impact on social mobility. It looks like we are not giving an equal chance to everybody. 

Sufi: We’ve seen a sharp rise in measures of inequality over the last 40 years. A lot of people have focused on a “keeping up with the Joneses” story, because simultaneous with that rise in wealth inequality, we’ve seen a sharp rise in household-debt levels, measured just about any way you want to measure them. And a lot of people say, “Oh, that’s because lower-income households, people in the median of the income distribution, they need to borrow more to get an education, to buy a car, to buy a home, in order to catch up, in some sense, with more-wealthy people.” 

The narrative that we actually believe is more relevant is more of a push rather than a pull factor coming from the financial sector. Because people who are extremely wealthy consume, or spend, so little of their wealth, as you get a rise in wealth inequality, more and more money comes into the financial system. That system, of course, is charged with going out and trying to find good investment opportunities, good ways of making use of those funds. In our view, perhaps there just aren’t enough good investment opportunities in, say, the private sector. So a lot of that savings gets transformed into borrowing by lower-income people: subprime mortgages, subprime auto loans, more education loans. 

The ease with which households below the 70th percentile of the wealth distribution can get credit has increased over the last 30 years. One prominent example is the ease with which people can borrow against the value of their homes. Thirty or 40 years ago, it was almost impossible to take out what’s called a home-equity loan, or cash-out refinancing, where you literally go to the bank and say, “I want to take out more money because my home’s value has gone up.” Now, and especially in 2004–06, it was incredibly easy to do that. 

Subprime auto loans are another innovation. People far down in the credit-score distribution can now borrow almost the entire value of the car when they buy a car. So that’s one way in which inequality is affecting the financial system and thereby affecting borrowing patterns.

How does that compare to the long-term view of innovation and social mobility?

Akcigit: We digitized all the patent records since 1836 in order to uncover the inventors in each of those patent files. Then we merged them with the decennial US census records between 1880 and 1940, which made it possible to see all these historical inventors of this golden age to whom we owe bicycles, refrigerators, safety pins, you name it. 

And it’s an interesting question: Who are they? There are some facts that are similar to today: for instance, back then, innovation was strongly associated with social mobility as well. But historically, innovation was a little bit more of a democratic activity in the sense that most of the inventions in the beginning of the century were done in garages. When you look at the patent records, about 70 percent of patents were assigned to individuals. Whereas today, more than 90–95 percent are assigned to nonindividuals, to corporations or universities. 

Kalil: We’ve taken a similar historical perspective to look at change in inequality in learning opportunities between rich and poor kids, for example. One thing we’ve noticed is that, over time, there has been an increase for rich and poor families alike in the amount of time that they spend with kids. And to such a great extent that, at present, the home environments of low-income kids look a lot like the home environments of rich kids 30 years ago. So you could say, well, that’s a good thing. But just as low-income caregivers have increased their time, so too have high-income caregivers. And so, in fact, the gap has shrunk very little, because high-income parents are doing orders of magnitude more than they did 30 years ago.

What are some policies or solutions that can decrease inequality or foster innovation and social mobility?

Akcigit: One policy conclusion coming from our analysis is that access to education is a very, very important tool. I also mentioned that innovation has an unusual relationship with happiness, and the reason is that, through innovation, you’re also replacing some companies, which means there can be many workers who are going on unemployment. And especially these days, when technology’s evolving at a fast pace, we observe that it’s typically the older workers who are not able to find a new job quickly. We see that the relationship between innovation and happiness gets much tighter in regions where unemployment benefits are higher. If we take into account not only the winners but also the losers in innovation, we can create bigger gains from this entire process. This doesn’t mean we necessarily have to increase unemployment benefits, but introducing some training programs for the people who are losing their jobs could be an important policy tool.

Another important policy conclusion coming from these studies is that we need to adapt procompetitive policies, which will make it easier for new entrants to come in and create this turnover that we want.

Sufi: Debt, as a financial contract, really is a pretty terrible contract if you’re a middle- or low-income person, because it essentially forces all the risk on you. The example I always give is: you have a $200,000 house and a $100,000 mortgage, and house prices drop by, let’s say, 50 percent. Well, then, you end up eating the loss, not the bank. Your mortgage is still going to be worth $100,000. The way debt contracts work is that they concentrate those losses on the debtors, who tend to be lower-income people. One idea is to try to put more equity-like financing into the market so that risk is shared more equally between
the rich and the middle- and low-income households.

Kalil: A puzzle that we have to solve is: How much cognitive and emotional stimulation is necessary for healthy childhood development? Is there some absolute level that will indicate, “This child is ready to succeed”? Even college-educated and/or rich parents are not spending five hours a day reading or playing puzzles with their kids, but what they’re doing is spending, let’s say, 20–30 minutes a day on a particular kind of learning activity that’s relevant for children’s development, and they’re doing it every day. What we see in lower-income households is that these activities happen much less frequently. When low-income parents do spend time with their kids, the actual number of minutes does not differ that much from their high-income peers; it’s just that it happens, as I said, less frequently. There are many different demands on low-income parents, many fewer resources, much less social support, many fewer peers around them
who are reinforcing these kinds of activities. 

Our research suggests that they aspire to interact with their children, but somehow, for many people, life gets in the way of the things we want to do. My colleagues and I have thought about how to design an intervention that helps parents set goals, reminds them of the goals they’ve set, and gives them some sort of social reward or feel-good feedback for meeting their goals. We send them text messages to congratulate them on meeting their goals. It’s really as simple as that. We do these simple, light-touch, low-cost kinds of interactions, and we find dramatic increases in the behavior of interest.