In this interview, Marco Ranaldi discusses his work on measuring the shares of income that come from capital and labor, and what this means for the structure of the economy.
 
Marco Ranaldi is a postdoctoral scholar at the Stone Center who is also affiliated with the GC Wealth Project. An Italian economist who received his Ph.D. from the Paris School of Economics, Ranaldi devised a way to measure income composition inequality (ICI), which is featured in his recent Stone Center Working Paper.
 
What is income composition inequality?
 
Income composition inequality, or ICI, tells us something about the structure of the economy —whether social classes exist, or not, in the way that classical economists like Smith, Ricardo, and Marx defined them. Classical economists described an economic system or society as composed of workers and capitalists. Workers are those whose income depends entirely on their labor, and capitalists are those whose income does not depend at all on their labor but on their capital.
 
Income composition inequality tells you to what extent societies are still composed of social classes in this way, and to what extent this is no longer the case — that is, to what degree all individuals are part “capitalist,” part “worker.”
 
ICI is different from income inequality, which indicates the magnitude of the distance between the rich and poor within a population. Income composition inequality tells us to what degree we have a society composed of rich capitalists and poor workers, versus one in which everyone has the same income composition and therefore the society doesn’t have social classes.
 
And you’ve come up with a way to measure this?
 
Yes, after defining this concept, I constructed an indicator that measures the extent to which a society has a high or low level of compositional inequality. This measure is very similar to the way the Gini coefficient, the standard measure of income inequality, is constructed. It’s a bit more complex and uses different tools.
 
How has income composition inequality changed in recent decades?
 
To put this in context, Piketty has shown that the share of GDP that comes from capital income has increased in most economies since the 1980s, and this is mainly due to globalization and financialization. Most economists would agree that GDP is important, but its composition is also important.
 
If income composition inequality is high, that means that rich individuals primarily earn capital income, while poor individuals primarily earn labor income. Under this scenario — a high level of compositional inequality — an increase in the capital share of income will enrich only those at the very top of the income distribution, because those at the top are the ones who earn capital income.
 
Many scholars, including Piketty, have assumed that an increase in the capital share of income will be correlated with an increase in income inequality. But, to assess whether that is correct, we need to measure income composition properly. We want to know to what extent this capital share of income is at the top of the income distribution, and therefore whether an increase in capital shares translates to an increase in income inequality.
 
And using my composition measure, we do not necessarily find that the dynamics of compositional inequality parallel the dynamics of income inequality.
 
In Italy, for example, while the Gini coefficient has steadily increased from the end of the ’80s until today, composition inequality has steadily decreased over the same period. We see a structural transformation of Italy’s economy — it’s moving away from being a classical capitalist economy, composed of rich capitalists and poor workers, into something that is much more mixed: a society that is not polarized between capitalists and workers, but much more fluid in terms of its income components. In the Nordic countries, the situation is quite different, as Iacono and Palagi have found. The level of ICI has in fact increased in the ’90s, helped by the emergence of tax reforms throughout the Nordic countries which made capital income taxation less progressive; that had the effect of shifting capital incomes towards the top of the distribution.
 
So you’re finding in Italy that people below the top 1 percent are earning more income from capital. But what are the sources of this capital income?
 
This is an important point, because it’s about how we define capital and labor at the individual level. My collaborators and I rely on survey data, and the data come from surveys that are run by national statistical authorities, or by other qualified institutes that collect nationally representative data. Capital income is basically defined as rental income, interest, and dividends. Rental income is basically the income you draw from renting out an apartment or other property, and this source is most responsible for the rise in income composition inequality in Italy. Other capital income, known as financial capital income, comes from stocks you might have that generate interest or dividends.
 
However, there’s another component that should be considered as capital income, but it’s difficult actually to work with, and that’s self-employment income. It is standard in economics to ask: “If you earn $1,000 per month in self-employment income — if you own a business — how much of this $1,000 is capital income versus labor income?” The standard practice is to split the components, with one-third counted as capital income and two-thirds as labor income.
 
But this assumption is problematic if you are considering, say, Latin American countries, where self-employment can mean income from activity like selling cigarettes from a truck. In these countries, self-employment income is concentrated at the bottom of the income distribution, where, on average, less than one-third of income actually comes from capital income. Hence, by assigning one-third of self-employment income as capital income, we are increasing the average level of the capital share held by the bottom class. (The effect is milder in high-income countries, where self-employment is concentrated in the middle of the income distribution: an income class that is more capital abundant.)
 
What larger trends are you seeing?
 
Latin American countries, on average, have very high levels of both income inequality and composition inequality: high concentrations of capital income at the top and labor income at the bottom. Most high-income countries, including many European economies, Australia, North America, and Japan, display on average moderate levels of these two dimensions.
 
Branko Milanovic and I find a third cluster — basically, the Nordic countries — that display an extremely low level of income inequality but a relatively high level of compositional inequality, similar to the levels that we find in Latin American countries. The Nordic countries are exceptional in this respect, and this result confirms the findings of Iacono and Palagi mentioned earlier.  
 
What drew you to this area of research? And when did you start working specifically on this idea of measuring income composition inequality?
 
I started to work on this at the end of 2015, which when I started my Ph.D. in Paris. Originally what I was working on was a completely different topic: a mathematical model of political business cycles. But when I started to work on that model, one of the underlying assumptions was, “Suppose we have a society composed by rich capitalists and poor workers…”
 
And my first concern was, “Let’s start thinking more about this aspect of the model, about the actors — the actors of political economy dynamics.” The actors seemed to be two groups with completely conflicting interests: capitalists and workers. My question was whether this is still the case. Can we really assume something of this nature in today’s economy? That’s how I started thinking about this, and I never stopped.