Wednesday, January 28, 2009

Economic Stimulus and Crime

I'm back, after not having posted for several weeks. The reason for my absense was that I was studying to take a comprehensive exam as part of the Ph.D. program I'm in. I took the exam last Saturday! I didn't feel that confident, but it's over and that's all that really matters to me right now. Whew, what a process. If I pass this one then I'm ABD and it should be smooth sailing from here.

In any case, what I wanted to write about today was President Obama's proposed economic stimulus bill. If anyone hasn't taken a look at it yet, this 647 page bill has plenty of pork in it. Further, much of the proposed spending has little to do with stimulating the economy (call me crazy but I thought that was the primary purpose of an economic stimulus bill).

Obviously I was most interested to see if there was any crime-related spending in the bill. I was somewhat surprised to see that there was. The bill includes a proposed $3 billion to be spent for "state and local law enforcement assistance" in order to continue the Edward Byrne Memorial Justice Assistance Grant Program. Byrne JAG grants provide funding to states and localities to conduct a broad range of crime control/prevention activity. This funding stream replaces what was previously called the Local Law Enforcement Block Grant (LLEBG) Program. The impact of LLEBG grants was the subject of a recent issue of Criminology & Public Policy (Volume 7, Number 3). While the primary paper by John Worrall presented results supporting the conclusion that LLEBG grants had a significant impact on lower subsequent crime rates, two other papers in this same issue provided a critical response. The paper by Al Blumstein points out that LLEBG funding is so broad and that the heterogeneity of programs/initiatives funded should be examined to determine what specifically is worth funding and what is not. It could be that some LLEBG-funded programs in some particular localities have a null impact or, even worse, actually increase crime. The aggregate impact may mask important differences.

The other paper by Thomas Marvell and Carlisle Moody points out that this is the first study to specifically examine the impact of LLEBG on crime, and that one study is hardly definitive. They point to the methods used by Worrall and show several examples of how other authors have used similar econometric methods and come to widely varying conclusions on a number of other criminological topics. As a prime example, Marvell and Moody point to Community Oriented Policing (COP) funding in the 1990s. According to an initial study on the impact of COP funding on crime rates (Zhao et. al., 2002), Community-Oriented Policing was found to have positive effects. However, a couple of later studies (including one conducted by Worrall himself) came to contradictory conclusions on the impact of COP funding. So we can hardly say that this one study on LLEBG provides conclusive evidence that further LLEBG/JAG funding will have a postive impact on reducing crime. It's way too early to make such a conclusion, and certainly to put $3 billion worth of trust in such a conclusion. One final historical point on LLEBG...it was implemented at a time when most cities nationwide were experiencing unprecedented drops in crime rates. This also complicates Worrall's findings. I'm not sure that he adequately accounted for this general drop, which may have occurred (even at the same rate) regardless of LLEBG funding.

And speaking of Community-Oriented Policing, President Obama's economic stimulus bill also includes a proposed $1 billion to continue funding COP. We've already seen that the existing studies on the impact of COP funding are mixed. Also mixed are the findings on the impact of COP practices. Several reviews, including a recent review of policing practices by the National Academy of Sciences (2004), conclude that community-oriented policing mostly has a null impact on crime. These studies and reviews also point out the wide degree of variability in what is considered COP practices, which makes program evaluation all the more difficult. The research is hardly exciting for the prospects of community-oriented policing. Especially when more promising, evidence-based policing innovations exist, like Hot Spots Policing and Problem-Oriented Policing, why is the Obama administration proposing to pour $1 billion into COP funding?

So bottom line is that President Obama is proposing to pour $4 billion into criminal justice initiatives that we don't know work. This is a shame when we have evidence of some concrete criminal justice practices/programs that have a demonstrated track-record of effectiveness. But even this is missing the broader picture here. Does it bother anybody else that this proposed funding is tacked on to an economic stimulus bill? What does this have to do with economic stimulus? Today's news was full of other examples of proposed funding in this bill that is unrelated to the economy (e.g., funding for STD prevention, etc.). I don't care if the proposed funding is for effective criminal justice practices or not...don't tack it on to an economic stimulus bill and try to sell us on the need for this bill to help turn our economy around. This is just crazy.

And the late breaking news this evening is that the bill has now successfully passed in the House, despite not one single Republican voting in favor of it (so much for bi-partisanship). I know, I know, you may say that a poor economy leads to rising crime rates and thus reducing crime is an important bi-product of an economic stimulus package. The evidence on this is tenuous, however. Any relationship between the economy and crime rates is a moderate one at best. In Al Blumstein's book "The Crime Drop in America", he examines several explanations for the large crime drop in the late 1990s. One chapter is on the impact of an improving economy on the crime drop. In this chapter, the author concludes that it was not so much that the economy writ large had an impact, but that an improving legitimate labor market combined with the increasing cost posed by illegal crack markets in terms of violence led to a shift from illegal employment to legal employment for a large group of young males. Thus the relationship is highly contextualized to what was occuring with the crack epidemic of the 90s and didn't represent a simple relationship between the economy and crime. Further, the impact was moderate compared to other factors examined in Blumstein's book. In another recent article by Steve Levitt (2004), he examines explanations that can and cannot explain the crime drop of the 90s. One of the six explanations he finds cannot explain the crime drop was a strong economy. He points out that from previous research, any identified relationship between the economy and crime is a very small relationship. He further points out that it is probably only reasonable to believe that certain types of crime would theoretically be impacted by the economy, particularly economically-motivated property crimes and drug markets. Certainly there may be indirect effects, such as found from the experience of waning crack markets, but targeting the economy should still be a very small component of any systematic attempt to reduce crime rates. It definitely isn't worth $4 billion in crime fighting or prevention outcomes.

So what gives? Is anybody else bothered by this? Why don't they ever listen to us criminologists when they want to throw money out? I can think of much better ways to spend that money if we're gonna spend it on crime. Or better yet, let's don't spend at all until we get a handle on this economy. Our government spending is out of control and we're not gonna spend our way out of it. Don't get me wrong, I think a core function of government is to protect its citizens, and I'm glad I've made a career in a policy area (criminal justice) that seeks to meet this core function. But when things are tight, spending has to tighten up.