The Contemporaneous Impact of Unemployment and Income Inequality on Violent vs Property Crime Rates in New England: A Panel Data and Machine Learning Approach

This paper investigates the possible impact of unemployment and income inequality on property and or violent crimes in the New England region in the United States from 2003 to 2022. Using a panel data analysis looking at unemployment rates and factors of income inequality in New England states, Connecticut, Maine, Massachusetts, Rhode Island, New Hampshire, and Vermont, along with demographic factors like race and gender aim to find a relationship between unemployment and income inequality with crime rates. Economic distress, usually caused by unemployment, can cause an individual to commit crimes due to hard times. Using machine learning decision trees and cluster graphs, we then investigate the connection within different Massachusetts counties from 2008 to 2010 aiming to see if the independent variables can be used as predictors. Overall, there was significant evidence that unemployment rate, race being non-white and minimum wage affect whether someone is likely to commit a property and violent crime. The machine learning findings did find that demographic factors like race being Hispanic can be used as predictors but also found certain results to contradict regression findings.