Colloquy Podcast: Is the US Economy “Fair Enough”?

Economic inequality has skyrocketed in the West since the 1970s, especially in the United States. So why have US politicians been unable to build sustained support for policies of redistribution? In this episode of Harvard Griffin GSAS’s Colloquy podcast, the Toulouse School of Economics social scientist Charlotte Cavaillé, PhD ’14, author of the recent book Fair Enough, points to an underexamined cause: the complicated ways people think about fairness.

The transcript below has been edited for clarity, correctness, and style.

How do you define economic inequality? How steeply has it risen over the last 25 or 50 years—a generation or two? And most of all, why does it matter?

Let's keep it simple, because there are big debates on what inequality is and how to measure it. We're just going to call it the uneven distribution of economic resources. Usually we think of two types of economic resources: one is income, which is the flow or cash you receive over a given period; the other is wealth, which is what you would get from the market at a specific time if you were to sell everything you own.

In terms of impacts, unfortunately we don't have great wealth data, so a lot of the longitudinal data is about income. We're going to stick to that for the moment. How do we measure inequality? Let me take everyone who lives in a country in a given year—let me talk about the United States—and see how much income they receive over this period. We rank people from super poor to super rich—from the person who gets the least to the person who gets the most. Then I cut this population into chunks: the bottom 10 percent (the poorest) and the top 10 percent (the richest). Then I do the same thing the next year, the year after, and the year after that, comparing this top and this bottom. We're not comparing the same individuals over time; we're comparing the same chunks of what we call the income distribution.

So what has happened in the US? Inequality has risen in most post-industrial democracies, usually starting in the late 1970s, but taking off in the 1980s. The US is unique for what I like to call the "fanning-out" pattern, which is: the higher you go in the income distribution, the quicker the growth. If I'm at the bottom—again, comparing these bottom chunks—since the 1970s my average income went down, controlling for inflation. Then it started growing—not much, but today, on average for this group, it's about 25 percent richer than in 1978. If I take the middle people—meaning half of the population is richer than them, and half is poorer—their income didn't decline; it stagnated, and then it grew at a similar rate, so now they are also about 25 to 30 percent richer than they were in the 1970s. Again, we're talking about groups of people in similar positions in the income distribution.

Now let's talk about the top 10 percent. They've had a much better time. What is unique is that, not only are they doing better, but instead of the 10 percent, if we take the 1 percent or the 0.1 percent, the higher you go, the better they're doing. For instance, the top 1 percent had an increase of basically 175 percent—so it doubled and more. If you go to the 0.1 percent, last time I checked, it was something like 800 percent richer than in the 1970s. Another way to think about this: back in the day, their income represented about 8 percent of total income generated that year. One percent of the population had 8 percent of the income; today, it's about 19 percent.

That is what makes income inequality quite unique in the US: it's this way in which the richer are getting richer, faster. That is not observed in similar democracies.

The last part of the question I asked was, why does this matter? Why should we be concerned if, even if you take into account wealth transfers and redistribution, the people on the bottom part of the income distribution are doing better than they were in the 1970s? Does it really matter that the people at the very top are doing so much better, and what impact does that have on our economy and our society?

That is usually the debate in my first class when I teach about these topics. Why should we care? Conservatives would usually say it's fine: there hasn't been a big drop, and most people are earning a living. And by the way, you're not comparing the exact same people over time, so where does inequality fit into real people's life experiences? I think there are three ways to think about this.

First, you are exposed to information from this data that I just described—which 99 percent of people agree on—and it shocks you from a moral perspective. You are usually on the left if you have that gut reaction: you bring a set of priors, you see this data, and it can shock you for several reasons. One is thinking there is no way these people are so much better than the rest of us that they deserve that much income, or thinking they are truly despicable people.

Another reaction, which is more common among economists, I call the utilitarian reaction: saying, “I'm neutral, but it can have bad effects.” There is a large body of research trying to document the effect on growth and democracy. To be honest, I would argue that it's hard to assess because we don't have enough countries with the kind of large inequality that the US has to compare and contrast. We have some evidence that the mechanisms tied to high income inequality are there, but someone who is not morally revolted by this data could also say that the evidence from a utilitarian perspective is not that strong. That's why some people don't worry about inequality.

The third reason, I would argue, is from a purely scholarly perspective: why is the US so different from other Western democracies? Why are we only seeing this in the US?

How do you define and really quantify in an accurate way the contributions that people make to a society or to an economy, and to the value of the services or the products they create? If you're somebody in private equity, for instance, how do you gauge whether the contribution you make to the economy is greater than someone who's a surgeon or someone who's an EMT (emergency medical technician)?

Ultimately, every time we talk about an uneven distribution of an economic resource, we are bringing to the table some theory about what a just allocation or process of creation of these inequalities is. So it's inherently political.

Ultimately, every time we talk about an uneven distribution of an economic resource, we are bringing to the table some theory about what a just allocation or process of creation of these inequalities is. So it's inherently political.

So in a sense, political economy is a redundant term. All economics are political.

Yes. And I think when economists say “political economy,” they just mean politics. But it doesn't look good for them to study politics, so they do “political economy.”

Given the increase in inequality that you've described, one might expect an equally steep increase in support for redistributive policies like taxing the rich or strengthening the social safety net. What does the data actually show about attitudes in the US toward redistribution?

Before I tell you what the data shows—and it's imperfect data—let me unpack this expectation that we should see an increase in support. By saying we should, you are already implicitly assuming that anyone who looks at the data should be concerned for moral, comparative, or consequential reasons.

Let's start with the first one: you see evidence of the concentration of income and accumulation of wealth at the top, and you find this morally problematic enough that it makes you want to change your opinions and potentially vote for someone who runs on this platform. Under what conditions would we see an increase? We would see it if a majority of the population had that reaction.

So then we have to explain who has that reaction. If we assume that people are rational, self-interested individuals, the people most likely to have that reaction are those who would not be hurt by an increase in redistribution—they enjoy a policy affecting a group that is not them—or those who could directly benefit because this money would be taxed and used for public goods provision like universal health care or free education.

Implicit in your question is the sense that people should find this increase morally abhorrent, tied to an implicit theory about self-interest. So under what conditions should we see an increase if a majority of people would benefit from redistribution? In the US, if we start taxing wealth or top incomes, there is quite a lot of resources—though it won't single-handedly eliminate the federal deficit. But there is another set of conditions: you must trust that when the government taxes, it will actually implement a policy that benefits you. The US is different when it comes to trust in government capacity. It is a large federal state where things are decentralized, and it is not always the same level of government helping you, so things get a bit messy in the American context.

So what do we see in the US? In most places, we don't see much movement. On average, that is true of the US as well. But while in other countries attitudes are pretty stable (with some exceptions like the UK), in the US, stability at the aggregate level hides extreme polarization. Polarization defines everything now: Democrats are becoming more pro-redistribution, and Republicans are becoming more opposed.

The problem with that general statement is that we only have one survey item on this topic that has been asked repeatedly since the 1970s, so you need to trust that this item is measuring something meaningful. This item asks: to what extent do you agree that the government should redistribute wealth, versus to what extent do you think the government should not get involved in this business?

Right now, more than 40 percent of Democrats strongly agree that the government should get involved, compared to closer to 20 percent in the 1970s—it has more than doubled. For Republicans, back in the early 1980s, about 15 or 16 percent were strongly opposed to government involvement; now, it's closer to 35 percent. So while there is still a good chunk of people somewhere in the middle, we now have a lot of people in these two extreme response categories, perfectly predicted by partisanship. They cancel each other out at the aggregate level, making the trend line look flat.

We've talked about the attitudes. Now let's talk about the “why” behind all of this. What does your research show about the forces that shape people's attitudes toward redistribution?

I started this book wanting to argue that the US was not so unique. It is a book about the US in comparison with other countries. Because I am looking at longitudinal data, I am dependent on questions people have asked in the past. The UK is a good place to look at attitudinal change over time, and I also have data from Germany, France, Scandinavian countries, and elsewhere. So I was able to write a book that explains both the US and Europe, though the US is ultimately quite unique.

I will tell you what I've learned on average about how people think about redistribution, and then we can talk about what is unique about the US.

As a first foray into what I'm arguing in the book, you should think of people as thinking about redistribution in principle versus in practice. They might give you different answers depending on whether you ask about their inclination toward the general principle of redistribution versus a specific policy.

Second, people rely on what I call a self-interested leg and a moral leg (a fairness leg). The self-interest leg is: “Would I benefit from this, or would it hurt me?” The moral leg is: “If we implement this policy, would it move the world in a direction that feels fairer? Am I contributing through my attitude to moving the status quo in a way that feels better morally?”

Those are the first two things. Then we need to understand under what conditions people lean more on one leg or the other, as well as what they actually mean by “fairness.”

People care about fairness as well as self-interested material benefits. But there's more than one way that they think about fairness. What did you find?

That was the biggest finding for me. There were two main insights. First, self-interest matters, but only under conditions where people understand the consequences of a policy. Some people understand specific policies very well: rich people understand the tax code really well, while low-income people understand welfare benefits much better than the tax code. People are informed on policies that are important to them. But for all the times when they aren't directly affected or don't have time to invest in understanding long-term implications, that is where thinking in terms of fairness comes in.

In my field, we often have a basic understanding of fairness defined as: to what extent do people think effort pays? That is, to what extent do people get what they deserve because they worked for it? But looking at ethnographies, focus groups, and survey data, I found that “effort” means very different things depending on the type of redistribution you're talking about.

The best way to synthesize five years of staring at the data is that people think about the fairness of policies affecting market income—income generated through a capitalist, market-based process—differently from how they think about the fairness of policies that pool resources to help others, such as unemployment insurance in Europe, Medicare and Medicaid, or welfare in the US. With pooled resources, they are not thinking about the fairness of markets; they are thinking about the fairness of shared resources: “Are we giving this to the right people?”

To my surprise, I discovered that these two ways of thinking need not be correlated. In Europe, they are completely uncorrelated; in the US, they are somewhat more correlated.

In your book, you talk about policies that are “redistribution from” and policies that are “redistribution to.” What is the difference between the two, and how do they play into these different ways of thinking about fairness?

There is a debate among some of my colleagues on whether this was the most helpful way of describing these two, but they are helpful in the following sense. “Redistribution from” is taking money away from someone, while “redistribution to” is giving it to someone.

Policies that take are usually policies that interfere with market income—the creation of market income—either by literally taking it through taxation or through antitrust policies that try to change the labor market to make it more competitive and prevent rent-seeking. In that sense, I call them “redistribution from” because you are preventing or limiting the accumulation of market income.

To simplify, back in the 1980s, the distinction was whether you were a communist or whether you trusted capitalism. That era is gone; most people do not see an alternative. Within capitalism, the question is: do you think the “rat race” is fair or not? Most people think it's not perfectly fair. But the degree to which people think the market deviates from pure, fair competition shapes their views. Even people who never took economics understand what fair market competition means because of our shared culture. If you think the market deviates significantly from fairness, you will be much more supportive of “redistribution from” policies that prevent the concentration of income among specific industries or individuals.

“Redistribution to” is about social solidarity and means-tested benefits. What I discovered is that people's perceptions of these policies are shaped by whether they think free-riding and abuse are prevalent. Abuse can mean cheating to claim benefits, but it can also mean taking advantage of collective resources—such as declaring unemployment by agreement with an employer to take a paid year off. Perceptions about the prevalence of free-riding are not predicted, especially in Europe, by what people believe about the fairness of the capitalist system.

To give an example: French President Emmanuel Macron was elected by a centrist electorate that believes capitalism works, favoring minor tweaks to ensure fairness. They look right-wing on market issues, but because they trust there isn't much welfare abuse, they look left-wing on social safety net policies. Conversely, the traditional far-right in Europe doesn't trust anyone: they think capitalism is an unfair race and that welfare abuse is rampant. So on welfare, they look right-wing, but on taxing the rich, they can look left-wing.

In the US, it's a bit different.

How is it different in the US?

In the US, I had to rely heavily on survey data to explain public attitudes. The US is unique in several ways. First, historically it has had only two major political parties, and these two parties have become strongly polarized. What we see now is that the US is one of the most ideological countries in the world. On the supply side of politics, the right-wing view of market fairness is bundled with the right-wing view of social insurance and solidarity, while the left-wing view of market attitudes is bundled with the left-wing view of social solidarity.

The US is one of the most ideological countries in the world. On the supply side of politics, the right-wing view of market fairness is bundled with the right-wing view of social insurance and solidarity, while the left-wing view of market attitudes is bundled with the left-wing view of social solidarity.

In Europe, these dimensions are much more disjointed because multiple parties offer different combinations of the two. In the US, when people answer surveys, they are not just expressing personal beliefs about a specific policy; they are reproducing these two party bundles. The polarized party system structures people into being consistent liberals or consistent conservatives.

However, when you analyze data that doesn't prime partisanship, the best predictor of who is a Republican versus a Democrat on redistributive issues is actually beliefs about social solidarity and the prevalence of free-riding. While there aren't as many survey questions on the fairness of capitalism—since the US is overall a more market-oriented country—beliefs about whether taxes are used to help the deserving poor have polarized sharply and strongly predict party identification.

You also say that self-interest trumps norms of fairness if it has a large and certain impact on a person's living standards. This was something that really struck me in reading your book, because the United States, among modern industrial developed countries, is unique in its relatively high level of medical bankruptcies, for example. Universal single-payer health insurance—a system like they have in France or Canada—would have an immense impact on the finances of millions of people, as would public daycare or free public college. If these programs would have a real and measurable impact on the lives of millions of people, why haven't we been able to institute them?

This is a great question. Let me first explain the example I had in mind when I came up with this idea. There is a common claim in the US and the UK that low-income people vote against their own interests by opposing policies that would benefit them. That is not what I see in the data. What I see is that while many low-income people express reservations about welfare in general, when you ask them if they want to cut welfare, they say no. In that sense, self-interest kicks in.

Crucially, it is much easier for people to understand the consequences of taking an existing policy away than to bank on the prospective benefits of a proposed policy—especially in the US context, where major new universal programs are rarely implemented cleanly. Obamacare, for instance, was a complex compromise where people still had to buy insurance on a market. The American political system requires an unusual degree of compromise. In the UK, by contrast, a party controlling Parliament can pass its platform directly, making it easier for voters to visualize future policies.

Irrespective of the system, it is always easier for a voter to evaluate the loss of an existing benefit than the uncertain promise of a future one. It falls on politicians to build the trust needed to convince people they will deliver, and that level of trust is currently lacking in the US. Even during the Great Depression, public opinion data shows that voters were initially skeptical of FDR's proposals. But once policies exist and take root, they become very difficult to remove.

Finally, reading your book, it struck me that the United States seems caught in a vicious cycle: increasing inequality paradoxically increases opposition, at least among a portion of the population, to the redistributive policies that could ameliorate that inequality. How can your research inform the work of politicians and activists who want to break out of this cycle?

There are two ways to think about this. Given my model, you could try to get people to put more weight on the self-interest leg. I'm currently reading Robert Caro's biography of Lyndon B. Johnson, reading about farmers asking the government for help. That is one path for self-interest. Historically, major shifts occur during periods of extreme hardship combined with the successful portrayal of the rich as undeserving, or when capitalism itself seems to fail—as happened during World War II in France.

My somewhat pessimistic answer is that, historically, the perfect storm where both legs (self-interest and moral fairness) work in the same direction—shaping people's beliefs about taxing market income and emphasizing self-interest—requires severe economic and social disruption. During the COVID-19 pandemic, for example, redistribution kicked in both in the US and Europe to cushion the blow.

There is a tension here: research suggests you need significant hardship to trigger major shifts, but existing social safety nets prevent society from reaching that extreme level of disruption. In that sense, I'm not overly optimistic about changing public opinion purely from the demand side by trying to persuade voters. Rather, it comes down to designing institutional reforms that build lasting support, if that is the goal you have in mind.