Sandra Susan Smith

How to Reform the Police

Air Date: October 4, 2021

Harvard scholar Sandra Susan Smith discusses the next phase of criminal justice reform and public perceptions of mass incarceration.


HEFFNER: I’m Alexander Heffner your host on The Open Mind. I’m delighted to welcome our guest today, Sandra Susan Smith. She is faculty director in the program in criminal justice policy at Harvard and professor of criminal justice. Thank you so much for joining me today.


SMITH: It’s a pleasure, Alexander.


HEFFNER: A lot of Americans are familiar with the First Step Act in terms of reforming the system as it exists today. Now there were some days and even years approaching removed from the First Step Act. How can we evaluate what that accomplished and how can we evaluate what needs to be the Second or Third or Fourth or Fifth in that Act?


SMITH: This is a great question. I mean, we should think about what it is that we continue to struggle with. There have been some successes, some improvements, I would say frankly, minor, moderate at best in terms of how it is that we deal with people who are brought into the criminal legal system. We have seen declines for instance and significant in some places in the number of people that we’re holding in our nation’s jails and prisons. We have seen improvements in terms of the access to opportunities to become integrated or reintegrated into our communities once people are released from the system or at least released from prison or jail. So they’re provided opportunities to be able to kind of get a foothold and be able to reintegrate back into their families and in their communities. But to be honest, many of these, if we would call them successes have been relatively minor on some level.


I think we always have to celebrate when we are holding fewer people in our nation’s jails and prisons, because that often does more harm than good. We are fairly harsh in terms of the kind of punishments that we meet out. We hold people for much longer than we need to in order to protect public safety. And when they’re released, we often make it very difficult for them to be able to reestablish themselves or establish themselves because many enter the system before they’re really able to get a foothold.

So we have to work on what we do on the front-end: reduce the number of people that we’re arresting for very minor offenses and then bringing into the system thereafter. We also have to be very careful about how it is that we engage with residents, with citizens. We don’t want to engage with them in ways, so for instance, do we think about policing strategies. We don’t want to engage in aggressive policing strategies that in the end do more harm than good because they affect how it is that community members see law enforcement and also impact the social, psychological, and in some cases, economic wellbeing of people who are having these engagements. So work on the front-end in terms of reducing the kind of contact that has been typical over the past 45 years.


And that has often been characterized by aggressive enforcement of low-level non-violent offenses. We also have to work on reducing the amount of prosecution that happens, especially with regards to those kinds of offenses. A recent report came out in March that showed that the non-prosecution of low-level, non-violent misdemeanor offenses actually produces better results than if we were to prosecute the same. So they’re less likely to have new criminal complaints, are less likely to be arrested, they’re less likely to spend any time in prisons or jail within the next one or two, three years after. A lighter touch where crime is concerned actually has much better outcomes and benefits for the broader society than the harder, tough on crime approach that we’ve been taking. And we see this all the way through to probation and parole where often for instance, with regards to probation, people will be, oh, we’ll have 10 years that are on the order of three to five years.


What we’ve learned fairly recently is that anything over 18 months actually does harm to people. 18 months seems to be a sweet spot where you can make it so that people are accountable, you’re watching to be sure that they’re kind of becoming fairly well integrated back into society. After that you just increase the likelihood that they’ll get caught up in a technical violation or some such thing and brought back into the system. It really doesn’t do much good. And so what we’re learning through social science research is that a lighter touch, especially with regards to, with regards to nonviolent, low-level offenses actually does much, produces much better results in the short and long-term than this kind of tough on crime approach that we’ve been taking. And I think increasingly we will learn as research begins to engage on questions about more severe crimes.


Are we being too harsh on even, you know, more violent offenses? What we know to date is that with regards to the more non-violent, lower-level offenses, we’ve done way too much. We brought way too many people into this system. We’ve made it very difficult for them afterwards to get jobs and to keep those jobs, to stay in stable, affordable housing, to take care of their children with a lighter touch, we can enable people to get back to their lives, hopefully be able to get and get a foothold in economy and be able to take care of their families. And until we address these issues until we stop criminalizing, especially these low-level, fairly minor offenses, stop prosecuting them and then stop with the very harsh punishment once they get well into the system, we will continue to have the poor outcomes that we’ve had for roughly since mid 1970s, up until the mid 2000s.


To the extent that we’ve seen shifts in this, better outcomes it’s actually been associated with being less harsh. It’s actually been associated with being showing much more compassion towards people with having empathy and with understanding importantly, where crime comes from; that so much of crime, including violent crime is as a result of breakdowns in community organizations and institutions that support individuals, families, and their communities. And to the extent that we can invest in those, we can actually do a lot of work towards reducing the kinds of crimes that we should be concerned about and producing much better results overall.


HEFFNER: And you’re involved, Sandra in that research that is explaining the carceral state of this country, that what has driven incarceration to be so pervasive. But the operating thesis that you are articulating goes against the current of a lot of the kind of media narrative that has criminalized people of color, that has sensationalized crime and that would lead you to believe in the most simplistic way that we are safer today, even though, of course, this contradicts the narrative now about increased crime, but that we are safer today as a result of incarcerating more people. So separate from the moral dilemma that as fellow human beings we should not want to imprison fellow human beings there is the false argument but still grounded in that basic truth that we are safer today than we have been. And, and the anti-justice reform voices would say, correct me if I’m wrong, some would say at least that’s because we do incarcerate more people and more people than ever.

SMITH: Yeah. So even from the beginning and when I say the beginning, I mean, in 1980s and we think about broken windows, theories about crime and justice. There had been very little research to support the argument that by aggressively pursuing these low-level minor offenses, we can hold off, we can, we can make it so that the more serious offenses won’t even occur. Even at the beginning of discussions and debates about that, there was almost no research whatsoever to support it. Now, what we’re finding with much greater access to data, from many criminal legal system agencies, and also much more sophisticated methods that what tends to be true is a lighter touch goes a long way. With regards to incarceration, for instance, people do argue that while we do see declines, because we have incarcerated so many people. What we’ve seen in a number of states and in some smaller municipalities since the, since about 2002 is even in those places where there have been dramatic declines in the number of people who’ve been held in the upstate prisons in jails, there’s also been a corresponding dramatic decline in crime rates, right? So incarceration hasn’t truly protected us from anything. There’ve been shifts, other shifts that have occurred in a broader society that makes it so that people are less likely to engage in crime, and those have also occurred. So for instance, in New York, in California, in New Jersey, we’ve seen declines in incarceration since 2002 or thereabouts on the order of 25 to 30 percent. And in each of those places, we’ve also seen major declines in crime. New York is a major, an incredible example of how this has happened. And you’ve seen this in a number of other places. And one could argue that by moving or shifting to less harsh penalties, such that there’ll be fewer people held. The expectation that I’m going to be held in jail or prison for long periods of time no longer exists because we become a bit softer on, I hesitate to use the term soft on crime because that will be read very poorly. We have had been much more, I think, prudent about how we punish. And, so don’t hold people as long. Don’t put them in jail at all in some instances. Some people will argue that that should have driven and people have argued, don’t move in this direction because if we move in this direction, we can expect an explosion of crime. And to some extent, we’re seeing that happen again now. But in, in those places where this, there have been these shifts. We didn’t experience increases in crime at all. It actually went and the, the rates of crime went down along with rates of incarceration. And so, it’s really difficult in these contexts that we see across the country and not just in California, New Jersey and New York, we see major declines across this period that correspond with increasing the rate, increasing declines in the use of incarceration as a form of punishment.


What that tells me is that incarceration is not the solution to solving all of our crime problems. And the problem is from the mid, 1970 forward, that’s how we’d been using it, incarceration, penal punitive penal policy, generally speaking, have been the way to address what we have been thinking of as crime often what lies behind that, however, has been these social problems, right? Increases in poverty and struggling, increases in mental health issues, increases in substance abuse disorders, increases in difficulties finding housing, et cetera. The way we as a society dealt with that from the early 1970s forward, was by attacking it with our penal system, with our criminal legal system, as opposed to buttressing up our social welfare systems in ways that would support people. And to the extent that we’re shifting away from punishment in the context where that’s happening, what we see is not an explosion of crime. What we see is actually a reduction of crime, especially in cases where the kind of punitive approach has been replaced by more substantive social supports to help people get by. And you see increasingly a number of initiatives, programs across the country that are seeking to deal with those problems in that way, with much better results, and also far cheaper than it has been using law enforcement to deal with these problems.


HEFFNER: That may be the reality, but do you see and believe that the perception of the public has changed markedly and substantively in terms of what you’re describing, which is, that punishment or a punitive penal approach is not going to yield the societal outcomes that you want.


SMITH: You know, I mean, it, what seems very clear, especially now when we see spikes in violent crime. And coming out of California, there are lots of reports about the kind of shoplifting raids of places like target and Walmart, et cetera. You see people pointing to these spikes in crime as evidence that these policies don’t work. And I think that there’s a substantial percentage of Americans who do believe that tough on crime is the way to go. And if it’s not working, go tougher, but I also think over the last five years or so, and especially the case after the George Floyd murder, people took a step back and started to reconsider whether or not these tough on crime policies, where the general public of really work. And so we’re willing in a lot of cases to consider or be open to more progressive policies over the past. I would say four to five years, you’ve seen an increasing number of progressive prosecutors being elected into office, running on platforms that say, we need to be smarter on crime, not tougher on crime. And they’ve been elected and they’ve done some really important work. The problem is that in American society there is always a huge amount of fear, right under the surface, right? So even for those who might be willing to try this new approach, especially in a period where things seem to be going fairly well, there’s always the fear that crime will come up and that we will be under threat or something. And I think that this is what’s happened in the context, both of COVID, and of this this kind of moment of racial reckoning, where there has been increased focus on police brutality and which I think itself has sparked some, or maybe contributed to some measure of violence across the country.


And so you combine the destabilization that’s occurred in part because of COVID and all of what COVID has done or shown with regards to how weak some of our systems are, and this this kind of a moment where the racial reckoning is, has happened. We see all of these destabilizations that are happening, that are producing these kind of outcroppings of, I would say there are lots of different ways in which it’s showing itself, but crime is one of the ways it has shown itself, and specifically violent crime. And I think people see that in a context where there have been these progressive policies, what do you point to in a situation where there are these spikes in crime during a period where these law enforcement authorities have shifted to more lenient policies, you point to the policies. It must be the policies that are making us unsafe. But what we’re not paying attention to is how utterly destabilizing this pandemic has been to the broader society, in such a way that has made it very difficult for organizations and institutions within the context of communities to support individuals and families, the way they have been. And it’s led to just kind of a breakdown in social order that has manifested at least in part through the kind of violence, spikes in violence that we’ve seen. And instead of paying attention to those kinds of de-stabilizing forces and attacking those, we instead attack the individuals who have gotten caught up in crime and including violent crime. We point our fingers at the progressive prosecutors who put forward these policies because that is what seems to make sense in our minds. I think unless we come to a much deeper understanding about the root causes of some of these behaviors and attack those, we will find ourselves right where we were 10 to 15 years ago with high rates of incarceration, higher rates of arrest and criminalization and making it so that people are constantly going in and out of the system without any real kind of resources to mend their own lives, mend the lives of their families.


HEFFNER: In this climate Sandra, in which you have described you know, there being kind of two roads we could take. Do you think that it is more important to change the cultural practices of the police departments or to actually change the laws based on your research? It’s not to say you can’t do both, you know, we know that they are related and also related to investments in communities. But right now I started the conversation by asking you about the First Step Act. So one of the things that made that marginal was it largely pertained to federal. We have a lot of incarcerated people at the state level. And so there, and there’s a lot of incoherence in state policy governing who is incarcerated, who gets locked up, in what states, there’s no leniency in what states, you know, three strikes you’re out, or two strikes for that matter, or one strike for that matter. So knowing that patchwork of incoherence in state policies around the country, what’s more important based on your research right now, changing the practices of police departments, we talk about reforming the incentives, the importance of community policing, or actually changing laws in states governing sentencing, parole, and probation?


SMITH: So I think so it would be a problem for me to say both, and. I think, here’s what I think we definitely need reform with regards to sentencing et cetera. But one of the things that I think that we need to do as a broader society and frankly, the cultural change that I think it needs to happen is in good part with regards to society itself. And that is a change around how we understand what public safety is and how we achieve it. I think as a broader society, we see police and public safety as one and the same. I mean, you throw of course firefighters in, for instance. But, when we think about crime, for instance, what we’re thinking about are police, that the solution to crime is police. You have more crime; you want more police. And I think what we’re finding is that we can actually do a much better job of addressing issues with regards to crime and the factors that lead to it by de-centering police, by moving away from police. This might not be a full-on abolition of police per se, but there are a lot of areas where we focus on police and it’s completely unnecessary for us to do so. It’s actually quite expensive for us to do so. And it’s ineffective to do so. So for instance what many of us don’t know, and why would we, is that with regards to what it is that police should do, and forgive me for maybe taking a little bit of a detour here, there’s very little, if we were to look at how police spend most of their working days. There’s very little that police actually do that relates to crime fighting or crime solving. Even in some of our most crime-ridden, dangerous cities, let’s take Baltimore, as an example. Prior research has shown that police officers spend about 11 percent of their time on the job focused on crime fighting or crime solving, right? And only half of that. So about 5.5 percent is focused on violent crime. That’s in a very violent city. If you get to the cities that are maybe small, or maybe less violent, that percentages decline significantly. Much of what police are doing through the course of their day is focused on administrative work. They might be patrolling in their cars, which has been shown to not affect crime rates in any way, shape or form. It doesn’t make people feel safe either. And they’re also dealing with like traffic stops, et cetera. So the overwhelming majority of activities that police are engaged in have very little to do with crime fighting or crime solving. And that is in the most dangerous of cities. So when we imagine pulling them out or decentering them, if we keep in mind, they’re not doing all that much with regards to crime in the first place I think we could take a little bit of comfort in it. We should also think about the extent to which what is happening in some communities, so in addition to police, not doing very much towards crime solving crime or reducing the likelihood that crime will occur, in some communities they’re actually doing a great deal of harm. So, marginalized communities, communities of color, low-income communities, what we have come to learn is that a lot of the aggressive policing tactics that police deploy in order to presumably to bring down rates of violent crime, even in cases where those rates might decline, and in most cases they don’t significantly, but it might decline somewhat. What we know is that it produces an impact on the people who live in those communities. We have learned, one of my colleagues Desmond Ang, here at the Harvard Kennedy School has a terrific and amazing paper that shows that in a context, in a neighborhood where a police officer has killed a resident, that has an impact on adolescents who are going to school, who live right in that community. It doesn’t matter if they knew the person or not. What happens is that their GPA declines for three semesters in a row. They are less likely to graduate high school. They’re less likely to go on to college. In other words, it affects the academic achievement and the educational attainment. We also know that because of aggressive policing approaches, not just killing, but also stop and frisk, et cetera, has a detrimental effect on the social and emotional health and wellbeing of adults as well. It’s having an impact on the residents of people who live there, who aren’t directly affected by these encounters. So, and we already know that reduced educational attainment, academic achievement, increases the risk of having future criminal legal system involvement. So even the strategies that we think we’re deploying, that police are deploying in order to bring down rates of crime and especially violent crime, in the few instances where it actually makes a difference, the costs associated with it are often higher than the benefits associated with it. And increasingly we know of other options that are available to bring down those rates of violence, to bring down rates of other kinds of crime, without having to engage in those very aggressive practices. And so I won’t, unless you ask, I won’t go into detail about what those approaches are. They’ve been in the news quite a bit over the last 18 months. It includes Cure Violence. It includes CAHOOTS crisis intervention. There are a number of other programs that are available to address mental health issues, violence, that have been shown to be far cheaper, that have been shown to be far more effective, that are, have been shown to instill trust in community members and that build trust among community residents that we can look to, to address the issues that we are concerned about without bringing in police who often, especially in these communities, again, do far more harm than good.


HEFFNER: I’m glad you did expound a little bit because we are out of time, but it should, it should be said that, you know, those police practices that you describe and their effect on the cultural and well-being of communities, that’s something that may or may not ever be able to be achieved through legislation. And that’s why I drew that dichotomy about reform around, but there can be legislative reform about the role of policing. I mean, you can literally rename your department to reflect that it is not a conventional police department of the 90s or 2000s, that it is a community driven policing unit. So it can be done through legislation. So I gather you’re, in answering the question and the seconds we have left, you would say that if legal mechanisms are going to be used in Minnesota or Florida, wherever the state legislature is considering rewriting the law, that policing and the role of police and the investment in police all should be analyzed and considered as part of that reform, just as much as changing bail, parole, probation, sentencing laws, that that police should be right up in there with respect to areas for reform.


SMITH: Absolutely. But we also then have to increase investments, which also will be done through the legislature, increase investments in the community organizations and institutions that do actually lead to much safer communities. So, yes.


HEFFNER: Sandra, thank you so much for your time today.


SMITH: Thank you so much Alexander.


HEFFNER: Please visit The Open Mind website at to view this program online or to access over 1,500 other interviews. And do check us out on Twitter and Facebook @OpenMindTV for updates on future programming.