f you are fortunate enough to live in a home you own, you probably remember the mix of emotions you felt during closing. Pride, anxiety, and excitement are all likely candidates, but finding a word to perfectly describe what you felt is probably quite difficult. If describing the effect of homeownership on an individual is difficult, imagine how much more difficult it is to determine the impact it has on a community as a whole.
It seems logical that more homeowners in a neighborhood results in more civic engagement and personal investment, but to what degree? Does one lead to the other? Which one? This “chicken-or-the-egg” dilemma lies at the heart of many community development efforts. It seems obvious that many indicators of neighborhood quality are interrelated, but it’s difficult to determine the exact nature of their interplay. That is why this study, showing a clear, causal link between elevated levels of homeownership and lower rates of violent crime demands attention from policy makers, and the public at large.
From 2013 to 2017, a one percentage point increase in homeownership rates in Baton Rouge census block group resulted in a statistically significant decrease in combined assault, homicide, and robbery incidents of over 4.5% in relation to the average. More plainly: increasing the amount of people who owned their home caused a reduction in violent crime in their neighborhoods. This finding is noteworthy enough to warrant a brief exploration of the methods used to reach this conclusion.
The Statistical Analysis
The statistical technique used harnesses instrumental variables, and is based off of a two stage least squares approach. First, a linear model is developed, including various controls ranging from block group property value to employment rates. The purpose of this is to remove bias, or the “hidden” impacts of these controls. This is done so that the impact of homeownership on violent crime can be isolated from the impact of these other, related factors. Yet, no matter how many controls you include—this study uses nine—no list can be completely exhaustive. So, secondly, fixed effects based on year and zip codes are included. This allows the model to capture any “spillover” effects; this is where a change in one neighborhood affects an adjacent neighborhood. The additional inclusion of yearly effects allows larger societal trends to be accounted for. These might be trends like a state-wide increase in median income, something that could not be appropriately accounted for using controls in step one.
The most interesting part of the model, however, is the third step: the use of instrumental variables. By identifying factors that are related to homeownership, but are exclusive from violent crime rates, the impact of homeownership on violent crime in a neighborhood can be “teased out” and isolated from the reciprocal impacts of neighborhood violent crime on homeownership rates. Using a novel method to determine the geographical concentration and proximity of lending institutions, instrumental variables representing access to mortgage lending were calculated. Mortgage rates, of course, have a strong impact on homeownership rates, but are independent of violent crime rates. Finally, these instrumental variables, the controls from step one, and the fixed effects from step two, are combined to develop the complete model. Using this model, the impact of a percentage point increase in homeownership rates, holding all else constant, was found to cause a decrease in violent crime of over 4.5% in that particular neighborhood.
Conclusion
Indicators of neighborhood quality are tightly interwoven and interact with one another in complex and mutually reinforcing ways. The ability to leverage the extrinsic effect of increased homeownership on violence offers a very real opportunity to stimulate a self-amplfying cycle resulting in an improved quality of life for residents. Focusing on increasing homeownership as a means to kick-start this snowball effect is especially advantageous. This is because low rates of owner occupancy can be addressed using proven methods, like providing financial counseling and access to funding. Other related neighborhood quality indicators have proven more difficult to address directly.
This study, available in its entirety below, shows that encouraging homeownership directly impacts one of the most important factors in any neighborhood: violent crime. This, in turn, will promote improvements in other neighborhood indicators. In this way, providing residents with achievable homeownership opportunities increases the well-being of the greater community. Expanding efforts to place people in their own home therefore has the ability to start a chain reaction improving the quality of life in local neighborhoods, and should be considered an integral part of any neighborhood revitalization reduction program.
