Talking With Jeff Asher

My go-to guy for crime data

Urban crime is a longstanding obsession for Donald Trump. Inaugural addresses are normally devoted to uplift and hope; his first inaugural was about “American carnage,” the wave of violent crime he claimed was destroying our cities. Last week he seized control of the DC police and sent in the National Guard to deal with what he claimed was out-of-control crime.

But what do the data say? I’ve come to rely heavily on Jeff Asher, whose Substack is an invaluable source of analysis and whose Real Time Crime Index lets us track recent developments in many cities. Jeff wrote presciently about DC just before Trump moved in.

So this week I talked with Jeff about crime trends and his views on the recent apparent plunge. Transcript follows:

. . .

TRANSCRIPT:
Paul Krugman in Conversation with Jeff Asher

(recorded 8/13/25)

Paul Krugman: Well hi everybody. Paul Krugman again. This week I'm going to be talking with Jeff Asher, who is certainly my go-to guy, and I think a lot of people's go-to guy now on crime statistics. It's a subject that is important. Lots of statistics, lots of data to be interpreted. And Jeff's Substack, Jeff-alytics, is now where you can find a lot of the number crunching analytics interpretation. He also has some views on what's been driving swings in crime and this seems like a good week since we've just had something like martial law declared in the nation's capital—or something like that—on the grounds of crime.

So, hi Jeff.

Jeff Asher: Hi. Yeah, it was a weird week to randomly write about DC. And then all of a sudden, everybody cared.

Paul Krugman: Oh, that’s right. You had written about DC. You had read up about DC already. So, yeah, now everybody is caring about DC. I want to talk about all of these things. But first I just want to know, you are kind of unique in providing these perspectives. Tell me a little bit about your professional life story, how you got to this point.

Jeff Asher: So I'm in New Orleans. I was born and raised here and went to the University of Texas and then moved to DC for grad school and got my master's in DC and started working for the Defense Department. I did that for about a year, and then moved over to the CIA, where I was for five years and worked as just an analyst over there. They taught me to write and think and all that good stuff, and then I met my girlfriend who became my wife, and decided to move back home to be near my family.

You know, in DC, everybody can write and has a master's degree and can analyze and everything. And in New Orleans, it's a very anecdote driven city, but I was looking for employment. The place that I ended up looking was the police department, because they hadn't had a crime analyst in decades. And so I started working for NOPD. I was there about 18 months and decided I wanted to start consulting. And, one of the challenges of having worked first for the CIA and then for a police department is that you're sort of a nameless, faceless blob.

And if you want to be successful in business as an entrepreneur, you need to not be a nameless, faceless blob. And so the way that I sort of decided I would get myself out there was by writing. I got introduced to Nate Silver and started freelancing at 538 and then Upshot, and I've just sort of been writing for ten years, ten plus years on this subject, which feels under-reported from a pure data perspective. It's a very difficult subject to cover from a data perspective.

Paul Krugman: And that was crime analysis for the CIA?

Jeff Asher: No, that was just general analysis. The skill set is shockingly applicable to other subject matters. Once you can kind of learn to analyze stuff and to communicate it effectively, then you can do it elsewhere.

Paul Krugman: Okay. That's a useful insight. But it also means that you're basically a deep state agent, so you can't be trusted on anything.

Jeff Asher: I've never denied that!

Paul Krugman: Okay, so, before I started being able to draw on some of your stuff, I found it is hard to do the numbers on crime. So why don't you tell us a bit about what's involved in that?

Why is it so hard? And why do we need AH Datalytics, basically.

Jeff Asher: The challenge is, we have expectations of precision with a lot of things. And I think we're kind of learning about this with the labor statistics as well, that our guesses are kind of guesses and they're decent guesses and we revise them. We can kind of accurately tell a story, but we don't have precision. And the comparison I like to make is living in DC in my 20s, I was right near the ballpark. I went to lots of Nationals games, so I became a big Nationals fan. And I loved in the last few years following James Wood, who's their big superstar coming up now. And I love going to Baseball Savants’ website, where you can say, you know, “I want to see all the pitches in April 2024 that James Wood hit with an exit velocity off the bat of 100 plus miles per hour last night.” He hit a ball 118mph off the bat. So that's an insanely hard hit baseball. You can do that within seconds of the pitch being thrown and hit.

And you compare that to crime stats, where the FBI last week published the 2024 crime stats. They won't publish an estimate of 2025 until around this time next year. And this was really early. August is really early for them to publish crime data from the preceding year. By that point, your trends have changed. Your understanding of what's happening is kind of useless if you're going off of year old data.

And the other factor that I think lends itself now to analyzing data much more effectively, is that a lot of places are just publishing the data. A lot of cities, a lot of states, they're publishing the data daily, they're publishing it weekly, they're publishing it monthly. And the FBI is even going to start publishing their raw data monthly. It's going to be, for a lot of people, I think, a disaster to use just because it's going to be, “here's the raw data, unaudited, prone to being updated.” It's going to be difficult. But if you know how to use it, it can give really good insights. And so being able to evaluate monthly data, weekly data, daily data puts you in a position to understand, to know what the trends are basically as they occur, which for crime is kind of the first time we've ever been able to do that. And in a very nerdy way I think that’s very exciting.

Paul Krugman: Okay, so crime data all ultimately comes from basically local police departments, right? It's not like how manual labor has BLS, that does its own surveys. They do interviews. But in this case we're relying on reports from local agencies mostly.

Jeff Asher: There's also BJS. The Bureau of Justice Statistics runs a National Crime Victimization Survey, which will get published in September for 2024. We know that not every crime gets reported to police. So it tells us, kind of, what are the parameters of non-reporting. What were the trends, basically. But like BLS, it's a survey. It's got a margin of error. It's prone to correction. So all of those things are surveys. But for the most part, the FBI data is coming from 18,000 police departments. There's no federal mandate. According to the FBI's lawyers and all the lawyers I've talked to, they can't mandate reporting because it would violate the 10th Amendment. So for the FBI, they basically rely on agencies and agencies, have these state UCR programs that kind of coordinate everything and hopefully make sure that everything gets reported.

Paul Krugman: UCR is the Uniform Crime Report?

Jeff Asher: Yeah.

Paul Krugman: That's correct. So, there are standard categories. Okay. Actually I just want to share a slide with you for a second. This is just from four days ago.

Paul Krugman: Basically, the same people who have been saying that the labor statistics are fake are also saying that the crime statistics are fake. And, then, you know, there's political stuff which we don't need to talk about here. But tell me, would it actually be even feasible to fake this stuff? It seems to me that it would be really, really hard to be faking this crime data.

Jeff Asher: Within a city, within a district, there are times, either by mistake or by intention, that an agency will manipulate a certain type of crime. There are times where things will get underreported. There will be mistakes. There are times where things will get over reported and there will be mistakes. But because there's 18,000 individual agencies reporting data, usually when the data is wrong, it's obviously wrong.

It's like Chicago reporting, you know, six murders in a month when the city averages over 20, and in some years more than that. Way more than that. It's when you've seen these sharp drops in crime all of a sudden when a data reporting system changed. But to manipulate national crime data would be virtually impossible. I think that’s the value of being able to go and get it from each individual city. You can draw your conclusions and audit agencies that look wrong and still come to the correct conclusion. And I'll note the FBI has seven major categories of crime that they collect. And there are ten different population groups. Every one of those population groups in every category of crime reported a decline in 2024, per the FBI. So it's not a big blue city thing. It's a small city thing. It's a suburb thing. It's a rural county thing. It's a big city thing. It's everywhere.

Paul Krugman: I live in New York. And the idea that the New York Police Department would make crime look good to help a liberal Democratic mayor, just if you know anything about New York now, that’s not going to happen. They might do it the other way around. So, yeah.

Jeff Asher: And you can even say, okay, I'm not going to believe the theft numbers. I'm not going to believe the assault numbers, but I'm going to believe the murder numbers. And murder is the thing that went up the most in 2020. And it's the thing going down the most now. And we have a bunch of independent sources. The CDC, the gun violence archive, there's too many places where we can verify that if murder was going down in the FBI data, but up in the CDC data, we'd say, “Oh, we might have a problem.” But they're in perfect lockstep.

Paul Krugman: Okay. And murder. It's kind of hard to conceal or fabricate a murder. Literally habeas corpus. So, your big thing, which I now look at a lot, is the Real Time Crime Index, which you've set up. Hopefully you've got this at least largely automated, but it is collecting data from a large number of departments and producing crime data on a monthly basis, although usually, I guess as 12 month moving averages. How hard was that to create?

Jeff Asher: It was both surprisingly easy and surprisingly hard. Our expectation was that we'd build relationships and individual agencies would give us data. People would be more than willing to hop on board. And the reality is, while we have maybe a dozen agencies that send us data, the vast majority of our data is coming from the state level and the open data portals.

So, it was hard to kind of come up with the vision. Arnold Ventures, who supports the project, could not have been easier to get on board. They were like, “yes, we should support this. This is a great idea.” And they've been extremely supportive of it, just as a data infrastructure project that can help us to know these trends.

We got started, I think, with about 300 agencies’ data. We're now up to 420 with complete data. How do we kind of make another leap into that 500 to 600 agency level, where we're very closely mimicking the national trends? I think we're getting there now. I'd like to get a little higher. The initial goal was 500 to 1000 agencies. So if we could get about 500, I’ll feel good about having accomplished that.

So, it was hard in that it's not necessarily easy to get the data, but states like Texas and Ohio publish it. We've just said, “hey, can we have your data?” And it’s like, “Great, we'll do that.” And so Ohio sends us the data monthly. Other places have been difficult, either because they don't have the resources to do it, or they want outrageous sums of money or to have us go through FOIA, and things like that. So, it does vary, which adds to the challenge.

Paul Krugman: Okay. So, as you say, the FBI data comes in late, which might not be so much of a problem except that we've had some pretty wild swings in crime, during Covid and post Covid. And suddenly, for a little while there, things looked much, much worse. Now, they look incredibly better. And you're catching it. So you're showing that the FBI already shows a big decline in crime for 2024. And your data is showing further large declines this year, right? So it looks as if crime is kind of falling off a cliff. Is that the right way to say it?

Jeff Asher: Yeah, in a lot of places. Obviously it's not everywhere. But of the 30 cities that reported the most murders to the FBI in 2023, murders are down in 26 of them. We're seeing a 20% drop in murder, a 10% drop in violent crime, a 13% drop in property crime. Whereas in 2024 murder fell a lot and auto theft fell a lot, now it's pretty much that everything is falling a considerable amount.

Paul Krugman: Okay. So let's talk for a second about DC, and then I want to come back to stories. So, you had just written about Washington and its crime. And, basically, I gathered that there were some data issues, not with murders, which is always solid, but on some of the other statistics. But DC did have a really pretty bad crime wave, I guess post-Covid. But that has now come way down. Is that correct?

Jeff Asher: I think that when you talk about DC, we can talk about the level. The level of crime is too high, the murder rate is too high. There's still work to be done. You know, nobody should say that DC is a city where things are perfect and that DC is a place without crime. But at the same time, we have to talk about the trend. And if we're going to be making policy decisions based on the trend, we should be getting the trend right. And the trend is that DC, like a lot of places, had a big surge post-Covid. It lasted a little bit longer in DC through 2023.

Murder fell a bunch in 2024. It's falling again in 2025. Carjackings surged a ton in 2023 and fell a lot in 2024, and are falling again in 2025. So, we're seeing all of the trends heading in the direction that we'd want to be heading in. I don't think that anybody should say that things are perfect in DC, but if you're going to learn from what's driving things down, what can we do better? It's best not to say everything's a hellhole and requires an aggressive touch to reverse an increasing trend, because that's not what we're seeing.

Paul Krugman: From a New York or Boston standpoint, DC still has very high crime, right? It's way up in trying to do quick and dirty stuff, but it does look like the homicide rate is several times what it is in New York City. But, you’re sitting in New Orleans, which also has quite a high crime level. We'll talk about the decline in a minute, but the level looks hard. Now, the question I hope you can answer at this time in your role as a citizen, as a resident of New Orleans, is what does it feel like? I mean, I showed you the Stephen Miller quote saying that the statistics are false and everybody is, in fact, terrified and that they build their lives around fear of crime.

Does it feel that way in New Orleans?

Jeff Asher: I mean, it certainly doesn't feel that way in New Orleans. I think that everyone is aware that crime exists and aware that there are behaviors in places and at times of day that will expose you potentially to higher levels of crime. At the same time, I think people also feel that things have gotten a lot better, and that's something that's been reported a lot here.

It's something that you see. They do an annual survey of the police department, and you see peoples’ satisfaction with the police department going up, people feeling more safe in the city. And I think it's that dichotomy where, if you're going to live in a city, especially in a major American city right now, you're probably going to live in a city that has a decently high level of violent crime and has a higher level of crime than we would want. But at the same time, things have gotten a lot better. And we can acknowledge both of those facts as being true.

Paul Krugman: Yeah. I spent a lot of my career up in the Boston area. But I knew New York back when. I know New York now. And, my God, New York is extremely low crime, at least according to the numbers. But this is not the city where as a teenager, I would actually occasionally be stopped by police officers saying, “I wouldn't walk in that direction if I were you.”

Jeff Asher: So, we launched a YouTube channel. Part of our effort is communicating this stuff better. It’s called Aggregated: the numbers behind crime.

Paul Krugman: Okay.

Jeff Asher: And one of our next videos is going to be on New York City, just sort of looking through the history of it. I've also decided to rewatch all of the old Saturday Night Live episodes and I was noticing, like in 1981, Saturday Night Live started saying, “And now from the most violent city in the world, New York, it's Saturday Night!” And they had a sketch with Jim Belushi where Brad Hall says, “Somebody gets robbed or mugged in New York City once every 11 seconds. And now to meet that man here's...” and it's Jim Belushi and they do an interview with him getting mugged.

So we're looking at all these things. And in 1981, almost 20% of the nation's robberies were in New York City. Just an insane amount of crime. And it's gotten so much better. New York is such a unique case because of how bad it was and how much better it's gotten. But it's gotten better in a lot of places.

Paul Krugman: But tell me about New Orleans which has had an even steeper drop from the pre and post-Covid era. Something like a 100% fall in violent crime, a 100% fall in murders. I'm not sure which it was.

Jeff Asher: Not quite 100%.

Paul Krugman: Right. It’s still a city.

Jeff Asher: Yeah, but it's more than 60%. In 2019, New Orleans had 121 murders, I believe, which was a modern low, the lowest in 50 something years. Post-Katrina, in 2011, I believe it was 200 murders. It’s a much smaller city than New York, but from 200 murders being the murder capital of the country, down to 121 in 2019 was real progress. And like everywhere else, it surged in 2020 and 2021. By 2022, the city hit 266 murders. That was the highest murder rate since Katrina. The highest number of murders since Katrina by far, just horrific. A 120%, 130% increase in murders from 2019 to 2022. [There was also] an enormous drop in the number of police officers on duty and enormous increases in vehicle burglaries, carjackings. Response times through the roof.

So there were all of these problems. Then it started getting better. In 2023-2024, the city had 125 murders. This year, I don't have the exact numbers, but we’ve had fewer murders than through 2019, through July. And that's including 14 murders on January 1st on Bourbon Street. So, that’s just an incredible turnaround for the city of New Orleans.

Now, it's not over. Last night, there was a triple murder, I think, or a double murder. There was a triple murder earlier last week. So these issues are still with us. They're still horrific. They're still tragic, but it's gotten a lot better than it was three years ago.

Paul Krugman: And so, you've just presented a story. Do you have an explanation for the big drop and what went right in New Orleans over the last couple of years?

Jeff Asher: I think what happened in New Orleans is instructive for the decline as a whole. You can see the same kinds of things happening. And the challenge with explaining why murder is falling now is, there's a number of different factors that you have to be able to explain that don't fit easily on that kind of left-right, red-blue continuum.

First it started in 2023. You could also say it was a 2024 or 2025 thing. But really, the factors that drove this decline started in 2021, 2022. We have fewer officers. Nearly every big city has seen a large decline in officers since the pandemic. New Orleans had 1200 officers and recruits in mid-2019. I think it has 940 officers and recruits now. So, a large drop in officers, large drop in recruiting, and that's typical. We haven't fixed guns, we haven't fixed poverty, we haven't fixed education. All of these root causes that you might point to and say “The first step is to fix the root causes of crime.” That's not what we've done.

And yet we're seeing a decline. We haven't done a gun control [law] here. The country is still awash in guns. Last year clearance rates were about 60%. So they largely returned to 2019 levels. But it's not like we've seen some enormous increase in clearance rates, or enormous increase in police efficiency that's enabled this decline to happen.

So what explains all of this? I think that the answer is, in the immediate post-Covid period, we saw this enormous influx and spending on a lot of stuff. And some of that is directly related to public safety. Some of it is not directly related to public safety, but might have public safety implications. Things like enormous increases in local and government construction on streets and highways, on street lighting. We know street lighting has an effect on violence. We know that it reduces crime if you install better street lighting. We've done that, and we potentially have seen big drops. [There’s also been] something like an 80 to 90% increase (adjusted for inflation) in social and neighborhood spending, construction from local and state governments, 50 something percent increases in public safety related construction from state and local governments, big increases in local and state government hiring, enormous increases in DOJ spending on grants to local governments, to local communities to do everything from community violence interruption, to doing things like hiring analysts, supporting police departments, acquiring technology, all sorts of things that DOJ spends a lot of money on that probably had some effect. And so you can't say which of these things was most important. But we did everything all at once, and I'm guessing all of it was probably important. And so in New Orleans, a big increase in street construction that was helped along by the Super Bowl coming about and led to even more street control.

I got a new street. Earlier this year I taught my kids to ride their bikes thanks to a street that wasn't enormously potholed like it was before. So, there’s been big increases in local spending.

The DOJ gave a $2 million grant to the city of New Orleans to start a community violence interruption program, based in the hospitals first. Every time a gunshot victim came in, they responded to it. They tried to interrupt beefs. They made sure that people had access to resources. Then, because the city had gotten all of this American Rescue Plan Act money, they spent half of it on basically shoring up the city's budget. So New Orleans, being a tourist city, saw enormous drops in revenue because of the pandemic. The city spent an enormous amount of money just making sure that it was on sound financial footing. Having done that, the city then invested its own money, $5 million, once it realized the pilot of this DOJ-led program was successful in expanding the program citywide. And so now they've got a much bigger program.

Paul Krugman: Just to be clear, what does “violence interruption” mean?

Jeff Asher: So, violence interruptors are often people that have been impacted by gun violence themselves, and they work with the families, they work with the victims, they work hard to prevent retribution shootings. They talk about, just exactly like it says, interrupting that cycle of violence by providing immediate support to the victims, going directly at the people that are most at risk of engagement with gun violence.

And there's all this research. This great professor in Chicago, Andy Papa Christos, talks about the social networks of gun violence and the degree to which, like, if you and I are friends, Paul, and somebody shoots me, that substantially increases your odds of being a future victim or perpetrator of gun violence. And so by providing that interruption at the point of gun violence, you're potentially interrupting a large stream of future shootings. In theory.

Paul Krugman: You're saying that big government drove the crime drop. That it was just a lot of public spending on multiple fronts that really did it.

Jeff Asher: I think that was the big picture driver and it sort of filtered down to the local level. And so it wasn't some random program that any particular city implemented, because I think if you did a survey of 100 cities, you'd find they did 100 different things, but it's that all of them were doing things. And all of these things 1) were enabled by this availability of funding from the government and 2) they were all doing things that had various effects on violence. And so having done all of these things, we're now starting to see all of these declines.

Paul Krugman: You know, in my bailiwick, there's endless discussions about the American Rescue Plan and costs and benefits. Like, was it a mistake or was it too big? And, I have never heard anybody say it helped reduce crime. And yet, it sounds like that was a pretty big spillover.

Jeff Asher: I think so. Taking a step back, criminologists aren't sure why murder fell in the ‘90s. So I think it's something that can be debated for a long time. It's a difficult thing to know at this point. I'm just a guy who has a theory. But I think that if you lay out all the things you have to explain, this is the only thing I found that really explains all of it.

Paul Krugman: I know I'm pushing this, but, federal budget policy has done a pretty massive U-turn now. We've got the One Big Beautiful Bill Act, and state and local governments are going to be quite strapped as a result of the cuts that have just been enacted. Should we be expecting a major uptick in crime?

Jeff Asher: I don't think it works quite like that. I certainly don't think that seeing a sudden drop off in spending, all of a sudden, the button will turn off and people will start shooting each other more often. I think the bigger risk is that at some point this drop will well stop. We're not going to reach murder zero in the nation. At some point we will see the decline level off. I have no idea if that's six months from now or six years from now, or if it's under a Republican president or a Democrat president. I don't know that the president can control that. But at the same time, if it happens, knowing that these tools are important, we've put ourselves in a less stable spot to respond to an increase in these crimes. If your response is, we're going to send in the National Guard everywhere that murders are increasing, you're hurting your ability to respond in a way that was successful when there was this almost once in a lifetime major surge and then enormous reduction in crime.

Paul Krugman: Okay, so let's talk about the big decline. It varies by city but, circa 1990, crime really was very high. And then there was this epic decline. That sort of bottomed out circa 2015, or thereabouts. But it's stayed quite low. I'm sure that you will tell me nobody really knows. But what are the stories? What went right? What happened? What happened to “escape from New York” and that kind of landscape?

Jeff Asher: I think it's an excellent question. There are a lot of things, like reducing lead paint and ideas like that that probably have some merit. And there are things that we probably couldn't do today. There was an enormous increase in police hiring. And we do know that if you increase the number of officers, you generally reduce the number of crimes in a city. At the same time, you also increase the number of arrests for low level offenses that generally target people of color. So it's a trade off. I'm not arguing one way or another in favor of it, but I'm saying that there is evidence that it tends to work.

In New Orleans, the city had 420-something murders in 1994 for a city of 500,000. They brought in a new police chief that year. He fired half the department as corrupt or incompetent and rehired and grew the department by two years later. We can't do that now. But basically there's a major argument we made that they hired a lot of cops in the 90s in a way that is not feasible today.

There's a guy at Princeton, Patrick Sharkey, a sociologist there that does a lot of great work. He had a book out where he talked about how the more you invest in community nonprofits, you see a correlation between a causal relationship between investment in each group and reductions in crime, reductions in murder. So I think very similarly, there's probably no one who can answer that. But, it's still something that is debated. And all we know is that it did go down, not that we know exactly why it went down.

Paul Krugman: Whatever your political orientation is, there's something to hate in what you just said. If you're a ‘defund the police’ person, well, turns out more police is good. If you hate squishy stuff like community groups, well, those turn out to be good, too.

Jeff Asher: [Laughs.] My goal with every answer is to hit that pissing-off-everybody angle.

Paul Krugman: And maybe I have the advantage of having grown up around New York, but then having spent a good part of my professional career in other places, and then in some ways seeing it with fresh eyes, coming back to the New York area and finding that it just bears no resemblance. But there is this consistent story, which is that if you ask people what's happening to crime, they've been through that epic decline, yet people keep on saying that crime is bad and getting worse. And I've said a few things about that. But I'm just curious for your take on why the public perceptions, or least what people say is their perception, differ so much for that.

Jeff Asher: That's the billion dollar question. I hate to keep pitching my stuff, but I'm launching a podcast soon where that is the theme. The perceptions of crime are so off. Why is that the case? What can we learn and what are people doing about it? Gallup had a poll last November that showed something like 70% of the country thinks crime rose in the last year. And it was heavily skewed partisan. 80 something percent of Republicans, only 28 or so percent of Democrats, but 60 to 70% of independents said they thought that it had risen. YouGov.com just came out with a poll that found pretty large majorities of every age group, every gender group, every political group that crime is higher now or that murder is higher now than it was in 2020 or in 1990.

So, these are knowable facts that it's just incorrect. Not even close to being correct. It's like answering “is the sky blue or green?” It is not green. Or if it is, something's gone wrong. But these are knowable facts that people get wrong every time, I think every year but one—2003, I think—the people have told the Gallup poll that crime has risen. And obviously we've seen increases during that period, but we've also seen a lot of declines, including 2014, which is the modern low for the nation's crime rate. And yet a majority of people said they thought crime had risen. So it's something that we haven't conquered yet. And I take it as a big challenge of how can we better communicate this and reach people so that they won't believe the worst ideas about crime?

Paul Krugman: Yeah, 2024 especially. I mean, last year saw a big decline in crime, as the FBI confirms and it's astonishing that people thought it was going up. One thought I have and just tell me whether I'm being stupid, is that there really is an issue of people having heard stories about the national stuff and, you know, direct experience may tell them one thing, but they are willing to believe that crime is skyrocketing someplace else.

Jeff Asher: Yeah, which was what made the YouGov poll so interesting and so depressing. I think you could reason out: the Gallup poll is about “crime.” Crime isn't very well defined. And the past year? Okay. That's maybe a little bit nebulous. You can't necessarily expect people to be able to compare rates of this year versus last year.

The YouGov poll is one higher than two. And people did not know the answer to that. They had no idea how high number one was compared to number two. 2020 was the highest murder rate that we recorded in 30 years. And 1990 was one of the highest murder rates we've ever recorded. And yet a majority, for both questions, said we think crime is higher now. Murder specifically is higher now than it was at times when it was undeniably much higher.

Paul Krugman: Yeah. Now it's true that most people may not remember 1990. It sure is pretty vague in people's memory, but still. You should have some sense of what major cities were like then and now.

But, do you have a sense how many people actually really do visit other parts of the country to form an impression? I mean, it is a little bit hard for me to understand how people can believe that New York or Los Angeles are scary places when, if you just walk around for an hour or so...

Jeff Asher: Well, especially the places where tourists tend to go. Crime, especially gun violence and murders, are very concentrated. Criminologists call it sticky. It’s concentrated within population groups. It's very concentrated geographically, and it's usually not concentrated in the places where visitors are. And so I don't know what visitors are getting when they're going to New York if they think New York is dangerous, or what they're getting in New Orleans. But usually it's not the places where crime and violent crime is really occurring. Most substantially so. I think it's not coming from people's impression of having actually visited places, but just their impressions of anecdotes, their impressions of what the news says. There's that quote that the media never covers the planes that land. That I think is sort of applicable more or less to this.

You're seeing a lot of stories now about how crime is going down. But, you know, nobody ever wrote a story about how there were no murders last week or there were no robberies yesterday. But if there's a triple murder, that's very rightly going to get written up. And your impression of what's happening is going to be formed by the fact that you just saw there was a story about a triple murder two weeks ago, not that there were no murders in the two weeks preceding that and no murders in the two weeks after that. That shows the trend line is down. But you're going to form your impression on the anecdote of the one horrible incident.

Paul Krugman: Okay. Last question. Sadly, I have not been to New Orleans for a long time. But what we see in places like New York, Boston, is that whatever people may say about crime, you can clearly see that urban living became more attractive. Gentrification happened and that contributed, I think, to the fact that people are feeling a lot safer. Is that happening in New Orleans as well? Have we seen some revival of central city living?

Jeff Asher: New Orleans is a weird place. There's obviously been a lot of gentrification in New Orleans. We're coming up in a few weeks on Katrina's 20th anniversary, which caused a lot of changes here that you haven't necessarily had to deal with in a lot of the country. There's an enormous housing crisis here. The cost of housing has gone up. There's enormous issues with short term rentals and Air BnBs taking over blocks that have traditionally been held by renters and things like that. So I think New Orleans has a lot of those issues. Certainly the city has gotten safer in the last two plus years. It had gotten extraordinarily unsafe. And there were a lot of bad trends and bad things that were happening within the city. And those trends have thankfully reversed.

That said, there's still crime. If we're going to have 110, 115 murders in New Orleans, that's still an enormous number of murders for a city of 380,000 people. It's still going to be one of the highest murder rates in the country. So both of those things have to be true, and we have to be able to analytically work through those in our minds.

Paul Krugman: Yeah. New York has something like 350 murders a year with 8 million people. So that gives you some sense of just the disparity. But here it doesn't condition your life. You're not afraid to cross the street to buy bread, for instance.

Jeff Asher: No, it doesn't [condition your life here either]. Those aren't kind of the main drivers. The fact that Mother Nature has hurricanes trying to kill us, and the roads are demolished and our pumping stations are 120 years old for the pipes underneath that have to pump the water out, those are the things that we worry about. I'm not doing a good job of selling the city right now, so I'll say it's also a beautiful city with, you know, wonderful people, wonderful food, wonderful culture.

I'm not as impacted daily by crime. But I also can recognize that there are people in the city for which that sort of stickiness applies and they are impacted daily by crime, by violent crime, by murder, by gun violence, and they need support. They need citizen services and all of those government interventions as best we can deliver. So, your average citizen may not have to deal with it, but there are certainly places that feel very high crime, that feel very dangerous and feel difficult. And you're going to get that in every city, I think.

Paul Krugman: I think so. Although, again, there were areas that were no go when I was much younger in New York City that are now unaffordable because all of these high end apartments have been built. So, that happens.

Anyway, I think you're getting a lot of attention now because there's a lot of this fact-free stuff about DC. It'd be interesting to see how public opinion, whether they buy the story that we basically need to send in the military to bring our major cities under control from this crime wave that isn’t actually happening. In any case, I think you have a market for your research these days. Definitely.

Jeff Asher: Yeah. It was quiet for a few months, and now it's obviously gotten very loud.

Paul Krugman: Well, thanks so much for talking to me. This has been a great conversation.

Jeff Asher: Thank you, Paul.