Plain English with Derek Thompson - Why Are the Police So Bad at Solving Murders?

Episode Date: July 6, 2022

Today’s episode is about the story of the moment—gun violence. There’s been a surge of violent shootings, mass shootings, and gun-related murders in the last few years. Today, Derek investigates a mystery behind this surge of violence: Why are the police so bad at solving murders? According to FBI statistics, in the 1960s nearly 100 percent of all murders were "cleared" by police, typically by arrest. In 2020, the clearance rate hit an all-time low of nearly 50 percent. Today, half of the murders in the United States go unsolved. Why? Today’s guest is Jeff Asher, a crime analyst, writer, and cofounder of AH Datalytics, which analyzes data for local government agencies like police departments. We talk about seven possible explanations for this alarming trend before settling on one particular explanation that's probably the most important. Host: Derek Thompson Guest: Jeff Asher Producer: Devon Manze Learn more about your ad choices. Visit podcastchoices.com/adchoices

From the host that brought you to Coding Westworld. And Westworld, the recapables. Comes the Ringer Prestige TV podcast on Westworld. I'm Joanna Robinson. I'm Danny Hyfitz. And I'm David Shoemaker. Welcome to Westworld Season 4 in the Prestige TV podcast feed, where we're going to break down every episode of Westworld season four.

Every Monday, the day after the show comes out on the Prestige TV podcast feed. Wherever we get your podcast, but get them on Spotify. Bye. Today's episode is in a way about the story of the moment. There has been a surge of mass shootings, violent shootings, gun-related murders, not just this year, but in the last few years. Crime and gun violence has become one of the top issues for Americans in this midterm election year. And this episode is about a mystery behind that story. It's the mystery of why fewer and fewer of these murders are,

ever solved. So according to FBI statistics, 60 years ago in the 1960s, just about all murders were cleared by police. Cleared typically means someone is arrested for the killing. But since then, since the 1960s, in every single decade, 70s, 80s, 90s, the clearance rate has declined. The share of murders that police solve has fallen in every single decade for 60s. years. That is awfully strange. A crime went way up in the second half of the 20th century and the clearance rate fell. Then crime went down in the early 2000s and the clearance rate fell. And now murders are rising again and the clearance rate just keeps going down. It recently hit an all-time low in 2020 of nearly 50%. I mean, put together these statistics. 60 years ago, almost every murder.

ended in an arrest. Today, half of murders in the U.S. go unsolved. Committing a murder today basically leads to a coin-flip chance of getting arrested. So today's guest is Jeff Asher. Jeff is a crime analyst, a writer, and the co-founder of A.H. Datalytics, an organization that analyzes data for local government and agencies like police departments. And as you're about to hear, as bad as America is at solving murders, we might be even worse at collecting data on crime. So why do the police seem to be so bad at this central job, solving murders? Why does it seem like they're getting worse with every passing decade? And what might an answer do to point us to a solution?

I'm Derek Thompson. This is Plain Image. Jeff Asher, welcome to the podcast. Hi, thanks for having me. So, Jeff, I brought you on the show to solve a mystery. And that mystery is why police seem to be so bad at solving murders. And why statistically speaking, they seem to be getting worse at solving murders decade after decade after decade since the 1960s.

And the data point that we're starting with is that every five to ten years since the 1960s, with every passing five to ten years, the national murder clearance rate has declined. So let's start with the methodology. What is the clearance rate? So when the FBI calculates clearance rate, what they're doing is they're saying, their numerator, is what are the number of murders that were solved either by arrest? Obviously, we all know what arrest means or by exception. And what exception means is they've identified the murderer. They've identified who they think did it.

But they, for whatever reason, they can't arrest that person. So it might be because it's a murder suicide. it might be because the person died a decade ago and you're solving a 15-year-old murder and there's no you know who did it, but that person is long buried. So that's the numerator. And it's not just this year. It's any year that the murder occurred, the year that they make the clearance. So if they solve a 1968 murder in 2022, that counts as a 2022 clearance.

So you get some funky statistics sometimes where agencies, will report regularly a hundred and fifty percent murder clearance rate where they solved three murders, but two occurred. Which looks incredibly goofy, but you're saying it's just a function of the fact that clearance rates are calculated in the year when the arrest or exception is made. So clearance rates are kind of like a police batting average. They're kind of like a very simple batting average, but they just have these two exceptions. And it should be clear, the first. The fact that the clearance rate has declined from like 90, 95% in the 1960s to a significantly lower level, like in the 50 percentage point level today, that you would agree is a real trend.

Yeah, absolutely. And that's a very real trend. That's a very real drop. And we have no reason to doubt that this trend exists, that agencies are reporting fewer and fewer murders being cleared each year. Let's get right into it. Let's talk about some of the reasons why this might be happening. Let's start with the 1960s. When the clearance rate was sky high, when it seemed like practically every single murder was being solved by police agencies across the country, was the clearance rate in the 1960s so high because mid-century police were so incredibly good at their jobs? Or was it because of something else, like the fact that these statistics were highly unreliable?

I think one of the certainties that everybody that studies this has is that you should not rely on the 1960s data and really much of the 1970s data for reliability. It's, it's, you're getting 90, 100% clearance rate in dozens or hundreds of cities that are reporting lots of murders, places that probably aren't solving it. And because we know that they're saying what they're clearing, but they're not giving any more details on that, that there's a lot of question marks about what exactly these numbers mean. And just to be clear, are you saying the police the 1960s, 1970s, were therefore much more likely than today to be arresting people who were known to be innocent, or are very likely to be innocent? Or is it because they were simply making

up the statistics when reporting to the FBI? Again, it's hard to say with a ton of certainty which of those answers it is, but I think it's probably both of them. We know, obviously, and especially in the pre-Miranda days, even today it's a problem of arresting people that are innocent of the crime. But even, especially pre-Miranda, the 60s, 50s, 40s, police departments had a lot less scrutiny on them. It's hard to say that what percentage of these are bad arrests and what percentage of them are bad exceptions. But there's likelihood, so likely some combination that's leading to the vast majority of big cities reporting 90, 100 percent clearance. rates for 100, 150 murders in a year. It's just, it's implausible that the numbers were as high as they

were in the 60s and the 70s. And I think that we've gotten several decades of much lower clearance rates at a level that is much more believable, I think, that makes us even more certain that those figures that were being reported in the 60s and the 70s are not that good. So basically, one reason why this is happening is that we are declining from a 60-year high It was basically a fabrication. It was basically a lie. It was either a lie because the police were obviously arresting lots of innocent people or because they were wildly misrepresenting their statistics, the FBI.

Just one of those statistics, I think this was from one of our conversations online. It's just unbelievable. In the 1960s, among agencies dealing with more than 50 murders, a third of them had clearance rates over the 90%. In 2020, zero agencies dealing with 50-plus murders had a clearance rate over 90%. So to put that in a slightly different term, in the 1960s, it was very common for agencies dealing with lots of murders to basically solve all of their murders, or at least claim to the FBI that they were solving all their murders.

Today, the number of agencies doing that is basically zero. And your sort of first big statistical piece of evidence for, you know, what is happening is, look, that was a myth. That was a lie. We should not believe the 1960s, 70s numbers. That's the most important place to start. To move on to number two, something else happened in the mid-1960s that we absolutely have to talk about, and that is the 1966 U.S. Supreme Court case, Miranda v. Arizona.

The Supreme Court says in that case that the Fifth Amendment guarantees citizens certain rights when they're being questioned by police. This creates the famous norm of Miranda rights, which anybody who's ever watched a half-second of law and order, or any crime procedural can tell you, you have the right to remain silent. anything you say or do can be used against you, et cetera, et cetera. Jeff, how important is Miranda in helping to explain the declining clearance rate because police, say, had a higher standard in terms of arresting people and getting information from them when they do arrest them? So it was clearly important, and you can see that if you look at the years before Miranda,

you're talking 92, 93, 94, 91, 901 percent murder clearance rates. And then every year after Miranda, you see pretty much a 15 or 20 year decline in the nation's murder clearance rates. So it was clearly an important factor. But also, I think that if it was the only factor, you would have expected basically, you know, things to have fallen off a cliff and then fall into a really low level. The fact that it was more of a gradual decline suggests that it was a very good tool, obviously, for improving. the way that police departments are going after people that they think did murders and improving the rights of people that may or may not be innocent. But it doesn't seem like it was the only factor. And it's one of those things where we really wish we had data from the 50s and

the 40s and could go further back. It's plausible that there was just something weird in the water in the first couple of years of 1960 that led to this really high artificially high murder clearance rate, or it's possible that Miranda was the biggest kick in the butt to get police departments to be more forthright and honest about who was being arrested and what murders were being cleared and led to this gradual decline into what was the 70, 60 percent range for murder clearances for much of the last three decades. I want to hold on this point because I think a lot of people in seeing the fact that police are solving fewer and fewer murders every single decade will immediately jump to the conclusion to say, ah, this means that police are getting worse.

And we're going to get to that potential explanation in a second. But a lot of people might jump immediately to the explanation that this is just because police are getting worse. This is purely bad. But when you put explanations one and two together, it strongly suggests, again, that a lot of these mid-century convictions were completely false or attained by practices that many people today would regard as highly unethical, which means that the decline in the clearance rate partially, and I want to emphasize this, partially conceals an increase in police ethics.

Do you think that's a fair explanation or a fair interpretation from explanations one and two? I think it's certainly a fair assertion. I think that when we talk about this problem, we really have to look at it in two ways. One is the big six. year decline in the clearance rate from the 1960s, and the other is the 20 or 30 year decline from the 90s and the 2000s that we're currently experiencing. And I think that when explaining the big 60 year decline, something changed in the 60s in terms of how police departments were reporting murder clearances, and it matches up time-wise where you see this sort of accelerated decline in terms of when Miranda came down. And so it's certain

would make sense that Miranda required police to think more about who they were identifying as culprits and how they were reporting clearances to the FBI. Again, it's hard to say with 100% confidence of that, but it certainly does make sense and the timing does match up as far as a potential factor there. Yeah. So let's go on to explanation number three. And here we get to maybe the real meat of your theory. Explanation number three deals with guns. Jeff, how much of this is about guns and why would guns explain the decline

and the clearance rate? So, you know, usually when you're doing these types of analysis, it's the guns is oftentimes a solid explanation for a lot of the problem. The guns doesn't solve everything, but the share of murders that have been committed by firearm,

has steadily crept up at a nearly identical rate to over the last 40 years to how the percentage of murder clearances has fallen. So you're basically seeing, and again, correlation does not equal causation, but if you plot the two together, you see a very strong correlation in the last 40 years between the percentage of murders that happened via firearm. And for that, we use the FBI's supplementary homicide report data where they give us that share of murders with firearm each year and the share of murders that are cleared each year. And last year, in 2020, the last year that we have available data for, 77% of murders were via firearm and 54% of murders were cleared. In 1961, it was 53% of murders via firearm and 93% of murders were cleared.

we know that 93% is a little wonky and was probably off, but it was a much lower share in, say, 1980, 62% of murders were via firearm and 72% of murders were cleared. So we've seen as the share of murders that are committed with firearm go up, the share of these murders that are solved goes down. And the reason is that firearm murders are much harder to solve. And why? Well, so you have oftentimes fewer witnesses. You have, there's, they take place from further away. So, you know, most murders with a firearm are not right next to each other. If you're stabbing somebody, you've got to be right on them. And so there's more likely to be physical evidence there. You know, there's more likely to be screams or whatnot that's going to

clue people in on who the perpetrator is. And there's more likely to be stuff that the detectives can then look at and help to solve the murder. And so as we see, the, you know, beatings and drownings and knife-based murders go down and firearm murders skyrocket, it just each individual case makes it harder on police. And there's a great Los Angeles detective, LAPD, retired detective now, John Skaggs, who was the protagonist in Ghetto Side, which is this terrific book about murder in Los Angeles, but really is about murder in any big American city. and I interviewed him for a piece a couple of months ago, and he talked about these, what he calls,

groundball murders, where basically... You said groundball murders? Yeah, ground ball murders. Yeah, ground ball murders. They're self-solvers. The police walk in and they find the husband, and the husband has the bloody knife in his hand, and the spouse's body is below him,

and this is a cleared murder. The police didn't do anything to solve this. It was solved itself. And so he talked about how 15 to 20 percent of all murders are going to be these self-solvers. And the majority of the self-solvers are more likely to be non-firearm murders. So as we have more and more firearm murders and a higher share of murders or firearm murders, the likelihood of them basically being self-solvers goes down. And the likelihood of them ultimately getting cleared goes down with it.

and you get individual cases one after another, year after year, it adds up. And that, I think, is a large part of the trend that we're seeing in the overall national clearance, right? That is so interesting. Walk me through again the statistical change. The share of murders that were firearms murders, you said, maybe 30, 40 years ago versus the share of murders that were firearms murders in the last data set that we have. I just want to make sure that I have top of mind how stark that transition has been.

Yeah, so in the 80s and early part of the 90s, most of the 90s, really, up until 2015, it was usually in the low 60% range were via firearm, and usually in the mid-to-high 60% range, low 70% range, were being cleared. Then in 2015, it sort of switched, where 2015 was the first year on record that we got 71% or more of murders were via firearm. And each of the last six years, we've seen that increase from 71% in 2015, all the way to 77% of murders in 2020 were via firearm. At the same time, murder clearances in, say, 2014 were 65% murder clearance rate. That fell to 62% in 2015 and 59% in 2016, which was the first time ever recorded under 60%. And then it sort of hovered around

there, that 60% range for the next few years, before really taking a nose dive in 2020. And one thing we know about 2020, there was a big surge in murders, but almost all of it was firearm. Right. It was a 30 plus percent increase in firearm murders and like a 12 percent increase in every other type of murder. One thing that I'm putting together, which you didn't explicitly say, but you definitely strongly suggested it, is that, you know, we've talked a couple times in the show about

how the murder rate and the violent crime rate has resour. surged in the last few years. But you're pointing out that just about all of that seems to be caused by gun violence. And it seems to me that it's possible that the share of non-firearms murders, right, the share of murders that are committed by anything other than a gun, a knife or whatever suffocation might be at like an all-time low. That the share of ground balls, so to speak, for police units might be at an all-time low. I mean, is that true? Are we basically looking at sort of a historic low of murders that aren't carried out by firearm?

But firearm murders have surged so much that it's accounted for that previous decline? Yeah, we're not necessarily at the bottom, but we're near the bottom relative to where we were, especially in the 80s. In 1986, for example, there were 7,743,000. non-firearm murders. And in 2020, there were 4,650. So that's a pretty significant decline over the last 35 years. But there were about 3,000 more firearm murders in 2020 than there were in 1986.

We haven't quite hit the peak of the 90s in terms of firearm murders. but the vast majority of the increase in murder, and the vast majority of what we're talking about when you hear crime wave or violent crime spike or whatever, we're talking about firearm murders increasing. And we don't count, we don't specifically count non-fatal shootings, which is we could do a whole other episode on that. But so the best measure of gun violence in America is firearm murders.

And that's really what we're talking about whenever we can. talk about this increase in violent crime, this increasing crime that we're dealing with. I find this so interesting, right. So guns are a major part of this mystery. Gun murders leave less physical evidence at the scene. They're less likely to be ground balls. Maybe they happen in better locations to get away with the murder. Is there anything to the argument? I saw a little bit of this online, but I was in absolutely no place to adjudicate it. Is there anything to the idea of that there's an increase in gang-related crimes, which might, A, make these gun murders more likely,

but also make it harder to sort of penetrate the network to figure out who committed the gun murder because that criminal is surrounded by a culture of silence, a group of people who very much want to not talk to the police and conceal the murderer's identity. Anything to that argument that I'm not. I saw flitted around quite a bit online. I think it's a really hard question to answer for a couple of reasons.

One is that we really don't collect data on it. And on it, you mean we don't collect a lot of gang-specific shootings and murders. Nationally, there's just, there's no good data collection. Jeff, it's becoming very frustrating in this conversation. I mean, and it's a frustration that I've had in previous conversations about crime, and to a certain extent this is central to, I guess, why you do what you do. The data on crime in America just seems to be absolutely horrible. Oh, it's terrible.

The murder data comes out 18 months to two years late sometimes. We don't collect nitty-gritty, fine-tuned data. We don't even track gang-related crime nearly to the extent that we should. And I'm not making a point that's going to be particularly innovative to anybody who's studied this for a few years. But it does clearly seem to be a major, major problem here. Well, and again, this is something we could do a whole other show on. But if you want to talk about how it's about to get worse, there's switching crime data collection systems so that the whole problem is about to get much worse.

So I'll come on the show next month and we'll reiterate on that or something. The other issue with the gang question is that frequently it's sort of, there's not really a rhyme or reason. There's no standardization. You frequently get people put on gang lists that don't belong there. that there's a lot of argument that the gang lists themselves are racist and that there's, you know, with that terrible phrase, racial undertones in them. So it's very difficult to say, even if we were collecting the data, that departments were appropriately evaluating whether or not gang violence was a part of the increase. That said, it's logical that every type of violence is a part of the increase. domestic violence, you know, gang-related violence, road rage incidents, every type of firearm

murder that you can think of is probably contributing to the increase. And there's no way to say what that exact recipe looks like, but the strong likelihood is that it's not just one thing or another that's causing it. Just a review, we've talked about the fact that the 1960s and even 1970s data is mostly bunk. We've talked about Miranda. We've talked about guns. We've talked about guns, which I went to highlight and emphasize with italics in bold, guns, guns, guns. I want to move on to explanation number four, which is higher standards from district attorneys and juries. And here I'm quoting from a response that I got from Chris Hasig, who's commander of Houston Police Department's Southwest Division. He told me he spent five years investigating

homicides. He absolutely agrees that the mid-century rates were inaccurate. He also said that greater accountability and higher standards today lend to more accurate data and better tech can be beneficial, but it creates higher demand in the eyes of the district attorney and in the eyes of juries. Is there any way, sort of building from that explanation, that police officers working a murder case anticipate that the bar has been raised? And if the bar of evidence has been raised, then it is harder to clear the murder by finding someone who only meets, let's say, 70 or 80 percent of that newly high bar, essentially that all these juries have been watching CSI and CSI Miami and they expect DNA evidence and they expect, you know, forensic evidence,

and they expect all this different stuff. And that as a result, it's just harder to meet that bar on a consistent basis. And so you have fewer suspects arrested and brought before a jury in the first place. anything to this higher standard explanation? I think it makes a lot of sense, and I think it sounds like it's a bad thing that it's an onus on police departments, but I would see it as only a good thing. Yeah, and I apologize if I represent it as a bad thing.

I'm obviously not interested in it. It goes to my previous comment that to a certain extent, a statistic that is seen by some as, oh, well, this merely shows that police aren't doing their jobs, might show that some police aren't doing their jobs, but might also show that ethics and standards are improving over time. So sorry to interrupt and maybe even steal the point you were about to make, but please continue. It's that you've got decades, plausibly have decades where prosecutors want to

win. That's what they want to do most of all. And if they've been refuted by jurors, if they have a more difficult time making cases these days without a higher bar to clear, then it's not that, and again, I want to be clear that the clearance, rate has nothing to do with whether or not the DA accepts the charges, but it certainly would make sense that the police, when they're making the arrest, are less likely to make the arrest if they don't think they have the evidentiary base to where the DA will accept the charges. All of that has been baked into the calculus that the police are using when they're trying to clear a case. And given all of the issues of innocence with the American justice system over, you know, the last,

how old are we now, 270 years, whatever our age is. I should know. America just celebrated her birthday, but I do not have a cap of mind. You know, in that entire frame, we have had severe criminal justice issues. So in this case, you know, in that framing, it would be a good thing to have better evidence being required and that being baked into the calculus. That said, going back to the point that we keep raising, is that we don't know for sure the degree to which that is. It really is anecdote and feeling. And from somebody that, Chris, who is

obviously very knowledgeable about the subject and has lived it, but we don't have what we would want, which is the data to be able to say, you know, here's the chart going down or here's the chart going up. But it certainly makes a lot of sense. Number five, explanation number five, I have to ask about racism. The same way that for some people, guns explain that. everything is their universal theory. There are other people, especially in analyzing the criminal justice system and the carcels system, for whom racism is or should always be the first order explanation. So got to talk about race here. I've seen evidence that the clearance rate for black victims in the last few years or decades has gone down while the clearance rate for white

victims has gone up. That is, that cops maybe have always been better at solving crimes when the victim was white and that disparity is getting worse. There's that potential piece of evidence that I'd like you to comment on. But at a more broad level, I'd love you to comment on the possibility that what we're looking at is just that a lot of American police officers can't say all of them, but a lot of American police officers are just racist, whether that racism is implicit or explicit it, and that we have to deal with that explanation for the declining clearance rate over time. What do you see as the best way to bring race into this calculation? It's obviously, it's very tough. It doesn't help that we don't really have clearance rates

broken down by race. The best we can do is using the FBI supplementary homicide data and guesstimate where they've identified an offender if that was a solved case or an unsolved case. It's absolutely right that we have seen a steady increase in both the number of black murder victims and a steady decline in the share of those victims that are seeing their cases likely solved. whereas white victims have tended to, I mean, if you squint, you can maybe see a slight increase, but it's largely been more of a plateau over the last two decades or so in terms of the share that are likely solved. And it's hard to say, you know, the degree that is it just that black victims are being shot and killed more often and white victims are just not,

that that's sort of at a plateau. certainly plausible. We saw an increase in both white shooting victims and black shooting victims in 2020. But whereas we saw a really small increase in white shooting victims, there was a huge surge in black shooting victims in 2020. So that might help explain some of the decline last year, but it doesn't explain two decades worth of decline there. I guess the question is always, you know, is racism involved, and I'm not one here to say that there's not racism in the criminal justice system or policing, that is obviously the case still, but is that responsible for the problem getting worse? And I don't think that you're seeing worsening clearance rates because police

are getting more racist than they were 60 years ago, or even 30 years ago. Right. It's very hard to argue that America's police. departments are whatever, 25, 35% more racist than they were 30 years ago, the sort of number that would be necessary for racism alone to explain this entire picture. At the same time, how do you feel about an explanation that says that as black victims of gun murders increase as a share of all gun murder victims, the poor relations between black Americans and police officers might make it harder for the police to get evidence about who might have committed these murders, and that therefore, the impression, often accurate, of police being racist, feeds into black Americans being less

responsive to, less cooperative with police, which then feeds into a lower clearance rate for black victims of gun homicide. So this is an explanation that doesn't try to, you know, blame anyone's specific group for the entirety of this picture, which just points out that the equilibrium of bad relationships between black Americans and police officers might be a contributor to the decline of the clearance race. Yeah, and that's the entire thesis that Jill Leovian Ghetto Side has. And I think that it's certainly plausible. And again, I'm pitching the book.

I don't get a check every month for pitching it. It's just a terrific book. But she talks about what she calls the absence of the absence of the... the state monopoly on violence, where if there's a 50-50-percent clearance rate, or in a lot of neighborhoods, Wesley Lowry and The Washington Post a couple years ago did a bunch of pieces on places where they talk about it's just free to murder, that certain neighborhoods where the police just don't solve murders. So like in New Orleans, if you're murdered in the French quarter, 90% of those are going to be solved. If you're murdered a mile away in the seventh ward,

you know, maybe 15% of those are being solved. And so how important it is the geography of murder. And so certain communities have baked in the likelihood of this murder being solved. And because they don't trust the police, they don't trust the state to go out and actually do it and solve it, that they take things into their own hands. And it sort of creates this cycle of violence that's very difficult to interrupt because you don't have. have that initial layer of trust in the police to be, to have that monopoly on violence, to be the only ones that can make arrests, to be the only ones that can be their own criminal justice system. And when we talk about murder going up in 2020, I think that it's, it's a logical

explanation that after the murder of George Floyd and all the protests and, you know, police violence and dep policing and everything that happened, that communities that have never had high-level of trust in policing to begin with, had that level eroded even further, which led to this cycle of violence that we've had for the last two years that's been very difficult to interrupt, virtually impossible and interrupt pretty much anywhere. You're not seeing anywhere that's having sustained declines in gun violence, really. I mean, you're getting a couple of places that are having had declines in 2021, and then they're seeing increases this year. Like Dallas was touted as this place seeing this great decline in 2021 and now murders up a bunch.

in Dallas. So, and this is very much like the Patrick Sharkey thesis, Patrick Sharkey, who's a researcher and professor at Princeton University and one of, I think, the great authorities on crime in the last 20 years. This is a part of his thesis, which is that you have a really unstable equilibrium. Even in the 1990s, early 2000s, when crime was going down, this unstable equilibrium where, yes, crime was going down, but relations between the public, especially black communities and the police were extremely fragile, such that it seemed to him relatively inevitable, that the equilibrium would break somehow. There would be some kind of Ferguson moment or George Floyd moment, which would ruin relationships, create distrust, maybe cause some police to pull back, cause a sort of acute depolicing of an area,

and then crime would bloom there and worsen relations between the community and police even further, and the disequilibrium that you're describing just becomes very, very hard to settle down. I think it's a complicated thesis, but I think in heart, because it is so sophisticated, it really appeals to me. So we're reaching the very end of our explanatory jambalaya here. And I just want to review before we get to the very last potential explanation. 60s clearance rates were hogwash. Number two, we got Miranda.

Number three, the rise of guns. Number four, the rise of standards. And number five, racism and or poor relations between police and communities creating this unstable equilibrium. Finally, number six, are police just not doing their jobs? Is it possible the quality of detective work is just worse? And that's sort of the simplistic way to suggest that maybe just police aren't doing their jobs. But also, I've read a lot of studies suggesting that police today spend much more time than they used to, doing stuff like clearing homelessness or responding to mental health crises in downtown areas.

That's not really like a detective work. that's kind of just doing the job of a social worker, but, you know, with a gun in your back pocket. So is it possible that the quality of police work itself is declining? It's certainly plausible. It's, again, not something that we can necessarily measure in a satisfying way. I think that the issues are we've seen, certainly we've seen a drop in the number of police and a lot of departments, especially over the last couple of years, are struggling. So as your numbers go down, and you're seeing a rise in violence, you're going from, let's say you have 30 detectives who are investigating and, you know, 150 murders. So that's five murders per detective, which is close enough

to what the standard should be. Well, let's say that doubles and you're looking at 300 murders. Or you're looking at 250 murders. And now each detective, and you've only got 25 murders, because five of them left to go work, you know, in the suburbs where there are no murders. So now you've got 25 detectives working on 250 murders, and you've basically gone from five cases per to 10. And so it's not inherently the detectives' fault that they're doing a worse job, but if you double their workload and you reduce the number of them, then you're going to have a harder time solving murders. And I think that you're probably seeing some of that. And especially in 2020, you're seeing that, and we don't know what the 2021 numbers will look like, but I'm guessing that

they're low again. Well, in 15 years, we'll have the good data on that. What about technology? Is it possible that technology might play a role in terms of, I mean, you know, it's funny. Like, with journalists, like, being able to do a lot of work over Twitter means that we place fewer phone calls. Is it possible that police detectives have a sort of similar, but, you know, different issue where the amount of information that's available to them on their computers

might make them less likely to be integrated into their local communities to source up to solve some of these crimes? But definitely think technology is a double-edged sword for solving murders. Obviously, it's great if you, you know, you see the murder happening on a videotape, on a surveillance tape, and you're able to solve it that way. Great. Technology was key. DNA. DNA is great. Yeah, DNA helps to solve cases that 20 or 30 years ago might not have been solved or 40 years ago. But there's also the negative side, which is that you can go out and you can talk to witnesses and they say it was Johnny over there. And then Johnny says I was at Walmart and you go to Walmart and you've confirmed his ally that he was walking in at the time that the murder happened makes it more difficult. The other reason that it makes it more difficult in John Skaggs, the retired LAPD detective talks about this.

he says he's solving 90, 95% of all of his cases in a given year because his way is going out and talking to people and finding witnesses and gathering evidence and that that is 100% the key to solving murders. And too often these newer detectives, he says, are sitting on their butts. They're waiting for iPhone to respond to their warrant to get to open up the phone so that they can look in the phone. They're not going out there and doing this sort of old fashion detective work that is the way that and the more resources, the more time, the more witnesses you talk to, the more likely the case is to be solved. And so if you're not doing that, then you're probably going to solve fewer murders. And so technology presents this great

force multiplier, but in reality it basically, and I think someone like retired detective like Skaggs might argue, that it basically removes the traditional, way that you've been successful at on an individual detective basis or it doesn't teach the skills that you really need to be able to solve them. So I don't think it's detectives are lazy and not doing their job like they were decades ago, but I think that through a complex combination of metrics, technology may not be the way that that is the only thing that helps us solve these cases. And if we're only relying on technology, then that might help explain why we're not seeing as many cases solved. That's so interesting. Yeah, I find that so fascinating. In part because it's so

relatable, there's so many aspects of my job where it seems easy enough to answer, say, an economic mystery by just asking economists on Twitter or just posing the question on Twitter and hoping that economists will respond to it, rather than picking up the phone and reaching out to someone specifically and saying, hey, can I build this relationship with you such that you can call me and I can email you and we can answer these economic questions together. I don't want to suggest that, like, TweetDeck is a perfect metaphor for the job of detectives, which is frankly something that I only tangentially understand through talking to people like you, but this double-edged sword of technology where the convenience, the short-term convenience,

disguises a longer-term loss of knowledge, I think is absolutely fascinating. Jeff, this has been absolutely just really, really interesting to me. I want to ask you one more question, because we have what I see as like, seven different items on the menu. I've gone over them a bunch. The last few that we just added to the menu are fewer cops per murder, which seems pretty significant, and the rise of tech being a kind of double-edged sword. What do you think with the understanding that we don't have good data here, and these are all hypotheses rather than strongly held convictions, of everything that we talked about, what are the one or two issues that you think do the most work in terms of explaining why

clearance rates have declined, not since the 1960s, 1970s. I totally believe that those were fictitious numbers, but in the last 20 to 30 years, what are the most important variables to explain why the clearance rate has declined? I mean, I think it's the guns. It's the nature of murder in America is changing in ways that we don't really talk about enough, that it's not in the 90s, it was New York, Chicago, Los Angeles murders, not all firearms, but, you know, a significant portion, but not to the degree that is now, whereas now it's spread out over a whole bunch of cities. We've gone from, in the 90s, 20% of murders were in New York, Chicago, Los Angeles. Now it's like 6 or 7% each year. And so you've got a bunch of cities where firearms make up 80 to 90%

of their murders. And so that is the, I think the main driver of as that share creeps up and the number of cities that are having more and more murders creeps up, that you're seeing these are harder to solve. And for a whole host of reasons, it leads to lower clearance reigns everywhere. Jeff, thank you so much. I really appreciate all of this. Thanks for having me. Thank you very much for listening. Plain English is produced by Devin Manzi. If you have a comment, a concern, a question, an idea for a future show, please email us at Plain English at Spotify.com. That's plain, no space, English at Spotify.com.