The common root: how unequal earnings could be fuelling racial wealth and crime disparities

Associate Professor Cagri S. Kumru, a macroeconomist at The Australian National University Research School of Economics, studies inequality, public finance, and policy design. His new paper is the first to examine racial disparities in wealth and crime in the United States (US) within a single unified economic model. Published in the Journal of Monetary Economics, it is co-authored with Professor Ayşe Okten Imrohoroğlu from the University of Southern California and Assistant Professor Jiu Lian from Zhejiang University, who completed the work as a PhD student at ANU.

Rather than treating the wealth and crime gaps as separate problems, the research shows that unequal labour income over the life cycle can explain a large share of both disparities simultaneously, with implications extending beyond the US.

In a recent interview, Cagri unpacked the implications of his research.

Your paper explores racial inequalities in the US. What motivated you to study racial disparities in crime and wealth together?

• Two numbers from collected US data sit at the heart of this paper. Black individuals make up 12.4 per cent of the adult population but account for 33.8 per cent of the incarcerated population

• 42.7 per cent of Black individuals are in the bottom 20 per cent of the wealth distribution, holding less than US$3,400 in assets, compared with 15.1 per cent of White individuals.

These two facts are usually studied apart: crime literature looks at offending and enforcement; wealth literature looks at savings, income and bequests. Nobody had a single framework where crime and wealth were allowed to respond to the same underlying forces simultaneously. That gap is what we set out to fill.

Why focus specifically on the connections between labour, crime and wealth?

Low wealth leaves people vulnerable to income and health shocks, which can push some toward crime. But committing a crime and being incarcerated impacts income and job prospects, making it harder to save. Survey data alone cannot separate these stories, since both predict the same thing: poorer people, more crime. A model lets us put income, crime and wealth in the same system and then switch channels off one at a time.

How does the ‘overlapping generations model’ you develop work?

An overlapping generations model lets you follow people through an entire working life. In our version, individuals enter at age 20, work until 65 and live until roughly 85. Every year they decide how much to consume, how much to save, and whether to commit property crime. Property crime is a genuine choice: people weigh the extra income against the chance of being caught. Drug-related incarceration, by contrast, is treated as something that happens to people, capturing the possibility that enforcement itself falls more heavily on one group. We calibrate everything separately for Black and White individuals, matching real differences in education, unemployment risk, and life expectancy. The result is a model economy that reproduces the actual US racial gaps in both incarceration and wealth closely, which is what gives us the confidence to run experiments on it that we could never run on real people.

What are the key findings?

Mainly one factor, the gap in labour income over the life cycle, can account for a large share of both the crime gap and the wealth gap. When we give Black individuals the same labour market conditions as White individuals, both their crime rate and their share in the lowest wealth quintile fall together. Most existing work treats these as separate puzzles, each with separate bequests for wealth and enforcement bias for crime. One underlying driver generates both disparities through perfectly ordinary saving and crime decisions. People tend to think incarceration is dragging down Black wealth – our model says income inequality is the real engine, and the crime rates are mostly a symptom.

What happens when you equalise labour market conditions in the model?

When we equalise labour market conditions in the model, education, unemployment risk, age-income profiles and income risk, the Black crime rate falls by 8 percentage points, from 12.0% to 4.0%. The median wealth gap falls from 83.6 per cent to 25.4 per cent, and the share of Black individuals in the bottom wealth quintile drops from 37.7 per cent to 22.8 per cent.

The message for policymakers is that these are not two problems needing two policy toolkits: they are largely downstream of one labour market problem. The paper also shows that equalising age-income profiles for individuals without a high school diploma would require resources equal to 3.3 per cent of Gross Domestic Product (GDP), and for the high school group, where most Black workers are concentrated, 22.2 per cent of GDP. If you are serious about narrowing both gaps, the labour market is where the leverage is.

You find that eliminating crime entirely barely moves the wealth numbers. Why?

If we eliminate all crime in the model (zero property crime, zero drug incarceration), the share of Black individuals in the bottom wealth quintile barely moves, from 37.7 per cent to 37.6 per cent. The reason is simple: they are mostly young people who already have little or no savings. We see the same pattern when we isolate drug-related incarceration on its own or strip out the scarring effect on future earnings. Both produce only small movements in the wealth share. This cuts directly against the common narrative that mass incarceration is a primary driver of the racial wealth gap. Low income and the absence of any savings generate both the higher crime rate and the lower wealth.

What lessons might Australia draw from this research?

In this paper, we built a model calibrated to US data, since the racial wealth gap and crime rate differences are quite substantial there. The history and policy settings that impact Indigenous Australians are different, so the numbers do not transfer. However, the underlying mechanism is consistent. Where a group faces lower earnings over a working life, you would expect to see the same pattern: more contact with the justice system and a thinner wealth buffer. The lesson is to resist treating criminal justice contact as the problem to be solved in isolation. Policies aimed only at policing are unlikely to close a racial wealth gap that is rooted in employment and earnings. A serious response would look at earnings trajectories and economic opportunity over the life course, not just justice system outcomes.

What do you hope policymakers take away?

That crime and wealth gaps are not two unrelated problems. Our model suggests they share a common root, unequal labour income, and that closing that gap moves both outcomes together. But tougher criminal justice enforcement on its own will not get you there; doubling the clearance rate for property crime cuts crime sharply but leaves the wealth share almost exactly where it started. The evidence points squarely at the labour market, earnings, employment security and the capacity to save over a working life, rather than the criminal justice system.

Image credit: Gorlov Alexander, shutterstock.com