Contact FutureLearn for Support
Skip main navigation
We use cookies to give you a better experience, if that’s ok you can close this message and carry on browsing. For more info read our cookies policy.
We use cookies to give you a better experience. Carry on browsing if you're happy with this, or read our cookies policy for more information.

Poverty and Crime

A high degree of residential segregation is observed across the United States.

In 2014, per capita income and poverty rate in Compton, California were $13,847 and 26.7 percent. Meanwhile, a neighboring city of Gardena had per capita income of $23,230 and 15.6 percent poverty rate. Not surprisingly, Compton also had a lot more violence than Gardena. The 2010 homicide rate in Compton was nine times as large as in Gardena (27 per 100,000 in Compton and 3 per 100,000 in Gardena).

Crime data across different time and places consistently show that crime is heavily concentrated in economically disadvantaged cities and neighborhoods. Over the years, researchers spent tremendous efforts on understanding the causal link between poverty and crime and whether government anti-poverty programs can help reduce crime. We will take a brief look at existing findings.

More Poverty and More Crime? Really?

Surprisingly, the relationship between poverty and crime is rather weak when looking at the historic data.

As we know, there was a historic crime rise and fall in the U.S. during the 1980s and 1990s. Yet, the change in national poverty rates during this period of time was relatively small. (Poverty rate was equal to 13 percent in 1980, 13.5 percent in 1990, and 12.7 percent in 1998.) More recently, poverty rates rose from 12.5 percent to 15 percent between 2007 and 2011 (the “Great Recession”), but crime just continued to fall.

A cross-country comparison also suggests that the criminogenic effect of poverty may be modest. In 2010, the United States, United Kingdom, South Korea, and Japan all had similar poverty rates: 15.1 percent in the U.S., 15 percent in the U.K., 14.6 percent in South Korea, and 16.1 percent in Japan. However, there is a great discrepancy between their crime rates. The 2011 homicide rate (per 100,000 individuals) was 4.7 in the U.S., 1.0 in the U.K., 0.9 in South Korea, and 0.3 in Japan. Similarly, the 2011 larceny rate (per 100,000 individuals) was 1,969 in the U.S., 2,638 in the U.K., 570 in South Korea, and 458 in Japan.

Given these data, one may wonder whether the observed crime concentration in poverty-ridden places is really driven by criminogenic effects of poverty, or by some other factors associated with both poverty and crime. For example, high-poverty and low-poverty neighborhoods may differ in many other aspects, including the quality of public schools, types of jobs available, presence of successful adult role models for youths, police effectiveness, and ethnic and cultural heterogeneity. Similarly, those who live in low-poverty, low-crime neighborhoods may be systematically different from those who live in high-poverty, high-crime neighborhoods. Can we really attribute the difference in crime between high-poverty and low-poverty neighborhoods as the causal effect of poverty on crime?

Similar problems emerge when trying to find the causal effect of a government anti-poverty program on crime. Each year, more than 4 million Americans under poverty receive cash assistance from government. What is the effect of this government program on recipients’ criminal risk? Does the cash assistance lessen their need to look for additional money from illegitimate work?

A simple comparison of offending rates between welfare recipients and non-recipients is unlikely to give us the answer because it is likely that they are systematically different in many other aspects (aside from the difference in their income levels).

Ideally, a random assignment of government benefits among equally eligible individuals would help researchers to recover the causal effect of the benefit program on crime. Then, we would simply compare the offending rates between those who receive the benefit and those who do not, and the difference in the offending rates can be credibly interpreted as the causal effect of the benefit on crime. But most government benefits do not work this way for an obvious reason, making a causal analysis on the effect of the program on crime very much difficult.

But there are always exceptions. For example, while most income-eligible Americans can expect to receive cash and food assistance from government, receiving housing assistance is not as easy. Since the number of available public housings and housing vouchers is usually a lot smaller than the number of eligible applicants, it can easily take several years for the applicants to hear back from the government and receive housing assistance. But what if there are just too many applications? Government may then run a housing lottery and provide the housing assistance to a randomly selected group of households. In the next step, we will see how such a randomized lottery can help researchers to recover the causal effect of a housing assistance program on crime.

Share this article:

This article is from the free online course:

Economics of Crime

Hanyang University

Course highlights Get a taste of this course before you join: