← Plain English with Derek Thompson
Loading summary
Transcript42 lines
- [00:01]
A
What's up everybody?
- [00:02]
B
Chris Vernon here and welcome to a.
- [00:04]
A
New season of the NBA and the Mismatch.
- [00:07]
B
And huge welcome as well to my.
- [00:09]
A
New co host, Dave Jacoby. I can't wait to link with you twice a week, every Tuesday and Friday right here on the Mismatch to break down everything that's happening in the league. Who's playing well, who we loved, who we loathed, trade rumors, team dysfunction. We've got you covered right here. So follow us, subscribe and hit us with those five star ratings on Spotify or wherever you get your podcasts. And also don't forget to follow us on social media. That's Ringer NBA. And check out the full Mismatch episodes with the two handsomest podcasters in the history of podcasting right on The Ringer NBA YouTube channel. This episode is brought to you by Lifelock. The New year brings new health goals and wealth goals Protecting your identity is an important step. LifeLock monitors millions of data points per second. If your identity is stolen, LifeLock's restoration specialist will fix it, guaranteed or your money back. Resolve to make identity, health and wealth part of your New year's goals with LifeLock. Save up to 40% your first year. Visit LifeLock.com podcast terms apply build a routine with Ollie that supports your wellness.
- [01:18]
B
Needs, like getting your daily vitamins and minerals with Ollie's multigummies or keeping your mood upbeat with all the vitamin hello happy. Give your gut health some support with probiotics and wake up feeling refreshed after taking Ollie sleep. Do wellness on your terms.
- [01:32]
A
Find Ollie at a Walmart or Target.
- [01:34]
B
Near you or@olli.com these statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure or prevent any disease today. Why are Americans so unhealthy? It's a question lurking behind news headlines everywhere you look these days. There's RFK Jr. Make America Healthy Again movement, which has focused a lot of attention on the unhealthiness of American lifestyles and food systems. And then there's the young man who killed the CEO of United Healthcare, whose manifesto called out, among other things, the infamous gap between the costs of the US Health system and the fact and this is a fact, that Americans have shorter lives than almost every similarly rich country. But while the US Healthcare system is overflowing with problems ranging from the maddening and often cruel behavior of private insurers to hospital incentives that drive up the cost of care, an underrated and under reported reason why American lives are so alarmingly short has less to do with our hospitals and our insurers, and more to do with our lives outside the health care system entirely. To put things starkly, Americans die young, inexcusably young, because life in America is inexcusably dangerous. According to data analyzed by the Financial Times and The Atlantic, a 29 year old in the United States is four times more likely to die in the next 12 months than a 29 year old in another OECD country like Australia or France or Japan. In fact, if you take 25American 5 year olds, a typical kindergarten class, on average, one of them will be dead before their 40th birthday. This is not true of any other developed country on the planet. This is not normal. And while it is certainly related to our system of healthcare, it has more to do with our guns, our drugs, our cars and calories. The US has more fatalities from gun violence, drug overdoses and auto accidents than just about any other rich country. And its obesity rate is about 50% higher than the European average. You put all that together and the US is rightly considered a rich death trap for its young and middle aged citizens. That's the bad news. Now here's the good news. In the last 12 months, quietly, quietly and without much media fanfare, the US government has reported that drug deaths are down, murders are down, traffic fatalities are up, are down, and the standard measure of obesity is down. As I recently reported in the Atlantic, this little inside straight of good news is remarkable for a couple of reasons. First, I'm not sure where else this 1, 2, 3, 4 of good news has ever been reported. The US news media, like most news media, is very, very good at telling its audiences when something happens that's really bad and very sudden. Hurricanes, mass shootings, falling bridge, celebrity scandal. We got you covered. But front pages and homepages and news feeds don't often tell people about important news trends that are the opposite of bad and sudden, which is to say, good and slow. Between 1950 and today, the age adjusted mortality rate of cardiovascular disease declined by almost 10 70%. Heart disease is the number one killer of Americans, right? And in the last 70 years, the age adjusted mortality rate has declined by 70%. It might be one of the best stories of the past century, but it's not the sort of thing that news tends to cover. It's rather the sort of thing that the news tends to occlude. The second reason that this health wave is so important is that by my count, it marks the first time in the 21st century, that obesity, overdose deaths, traffic fatalities and murders all declined in the same year's official data analysis. And I'm sorry for sort of the legal technicality of that last sentence, but I had to put it that way because, to be totally honest with you, some government estimates, like the rates of obesity and overdose deaths, have reporting lags of one to two years. So I cannot guarantee you, nobody can guarantee you, that all these things are declining at the moment you've chosen to listen to this podcast. But I can tell you that they declined the last time the government looked all this stuff up. And it's never happened before in the last 25, maybe 30, even 40 years that all these things have declined in one year's analysis. That is a health wave, and I don't see it being reported. So this week on the show, I want to bring this news to your attention. And then, because it's not a plain English episode, if we don't ask why, I want to ask why. Why is this all happening simultaneously? Are we looking at something monocausal? Is it all GLP1 drugs? Is it all just the pandemic ended and everything got better? Is it all brilliant public policy by the Biden administration? Today's guest is Charles Fane Lehman. He's a fellow at the Manhattan Institute and someone who, in our last conversation, I jokingly referred to as the Prince of Darkness, because his expertise is unpacking complex trends in truly the most gruesome areas. Drugs, murder, death in America. Most of the time when I reach out and read Charles, it's to explain why things are getting worse and worse, which is why I wanted to get his mind on this moment, when it seems like the US Is experiencing a sort of mysterious health wave and when the causes of premature mortality in America weirdly seem to be getting better. I'm Derek Thompson. This is plain English. Charles Fane Lehman, welcome to the show.
- [08:32]
A
Thanks so much for having me on. Glad to be here.
- [08:35]
B
So, in the open, I said that obesity and the murder rate and the drug overdose deaths and traffic fatalities all declined in recent government analyses. Perhaps the first time this has happened in one year of government analyses in the last few decades. The easiest two of these to explain, I think, are obesity and traffic fatalities. Like here you have the rise in GLP1 drugs, and with traffic fatalities, you had this explosion of chaotic and weirdly antisocial behavior during the pandemic. And that sort of weird and chaotic antisocial behavior seems to have abated in the last few years. I want to spend Most of our time talking about the decline in drug overdose deaths and murders, because these are both incredibly important trends, and I think they're somewhat mysterious trends, which is the right intersection, I think, for a show like this. So first set the stage for us with a brief history of the drug waves of the 21st century up until about 2020 and the pandemic. Because I know about the prescription drug crisis and the Sacklers, and I've heard of fentanyl, but I'm not sure that I have in my head a neat story that connects all of these dots. So what has been the story of drugs in America in the 21st century?
- [09:57]
A
Absolutely. And the way that I think about this and other people think about this is this concept of the triple wave epidemic, although at this point you talk about it as the quadruple wave epidemic. This is a phrase coined by Densha Garon, who's a researcher at ucsf. But the idea is that basically we've seen a series of changes in the kinds of drugs that people use. And with each change, the deadliness of the crisis has increased. So roll the clock back to the 1990s. We're talking about prescription opioids and the rise of prescription opioids, primarily through iatrogenic, that is doctor driven abuse. This is the story about the Sacklers. There's a huge explosion in the use of opioids based largely on the false perception that they're not addictive. This is exacerbated by the perverse incentives of Big Pharma. Purdue Pharmaceutica, the company that you alluded to, rolls out OxyContin, a highly potent oxycodone pill that contains a huge dose which is supposed to be delayed release, but you can just crush it and then snort it. These flood the market. We have good quasi experimental evidence that this is a major driver of the early stages of the overdose crisis, probably through two channels. One channel is the sort of story that you hear a lot about, which is like somebody goes to the doctor and gets a 30 day script, and at the end of the 30 day script, they're hooked. That's probably part of the story, but I think it's probably less part of the story than the risk of diversion, which is like your aunt gets 30 days of pills for her dental surgery, but she only needs 10, and the remaining 20 are in the cabinet. You steal them and you sell them and that floods the market with pills. You also have these pain clinics. So there's a big increase in opioid use and opioid addiction that's seeded by unethical doctors in the pharmaceutical industry. Midway through the 2000s and up into the early 2010s, there's sort of a focus cracking down on this. So Purdue reformulates OxyContin, so it's abuse resistant. When you try to crush it, you can't snort it afterwards. The Obama era DEA goes after a bunch of these guys in a story that almost nobody ever tells. And there's this initial settlement and the number of prescriptions starts declining. In about 2012, you roll out prescription drug monitoring program. So that first wave of drugs recedes. There are a bunch of people out there who are hooked on opioids. They can no longer get their supply, and so many of them substitute to heroin. That's the second wave. Heroin's been around long time. It comes and goes. You can go back to the 19th century and see heroin epidemics. Heroin is pretty quickly. You see heroin deaths start to rise, but then they start to decline again because they're pretty rapidly replaced by synthetic opioids. What you alluded to earlier is fentanyl very briefly, and we talked about this more if you want to. A synthetic opioid is distinguished by its method of production. If you want to make heroin, you got to grow poppies, and then you have to drain the SAP, and then you have to refine the SAP into morphine. You have to refine the morphine into heroin, synthetic opioids, and run all of that. You don't need plants. You just need simple precursor chemicals to produce something that is equivalent to, or in many cases, much stronger than morphine or heroin. So humans have known how to make these drugs forever. When my wife gave birth, she got an epidural and she got fentanyl in the epidural. We've been able to make this since the 60s, but it took a really long time for the illicit market to catch up to the illicit market. There were sort of little patches of fentanyl, but it's really in about 2014, 2015, that the Mexican cartels finally get this technology online. Fentanyl is just better on every dimension than heroin. It's cheaper, it's easier to smuggle, it's easier to manufacture, it's much more potent, it's much more appealing. And so very rapidly, fentanyl replaces the tired of the opioid drug supply. Moving from the east coast to the west coast. That's phase three. That gets us up to about 2020. Although I think the other dimension on this is that there are and we'll talk about sort of adulteration since then. The key thing about fentanyl is that it implies sort of a level of chemical knowledge that means that there's no longer natural limits on what the cartels can synthesize. So, like meth, if you remember Breaking Bad, meth is a thing in the early to mid 2000s, meth is back. Now meth is back and way worse than it was in 2000. For reason. There are all sorts of other adulterants. We talk about nitazines, we can talk about xylazine, we can talk about all these different drugs. That's really the fourth wave is now we're in this sort of polysubstance poly drug crisis, where it's not just opioids, it's all of these different drugs often taken together. Downstream of that change in how drugs.
- [14:29]
B
Are manufactured, that was a fantastic history of a very grim story. So I appreciate that. To summarize, entire 21st century drug overdose deaths are skyrocketing. First with prescription drug abuse, then you've got heroin's resurgence, then you've got fentanyl, and now the fourth wave poly drug crisis.
- [14:48]
A
However, the thing I should add, because you offered the context, is, you know, that all looks like an exponential curve in the increase in drug OD deaths, that we're going from like 10,000 OD deaths to 100,000 OD deaths in the course of that process. Right.
- [15:02]
B
And if you just look at the line, it looks like a smooth line, which is why I think it's so interesting and important to X ray that line and see that hiding within it is not one trend, but rather something more like four different trends. However, May 2024, the US government reports that drug overdose deaths fell by 3% between 2022 and 2023. This was a real surprise. This is a rare bright spot in a century of escalating drug deaths. Let's try to explain, just as you've explained, why the crisis has unfurled, let's try to explain why this might have been a rare bright spot in American drug history. Let's start with the possibility this is a policy success story. What is the steel man version of the idea that local, state or federal governments got together, changed policy, and made a dent and saved people's lives?
- [16:02]
A
Yeah, there have been a couple of prominent policy changes in the past several years that plausibly contributed to the phenomenon that you're identifying. You can tell a story where that's relevant. One is many listeners will probably have heard of the overdose Reversing drug naloxone, also known as Narcan. It's the drug that if you administer it to people who have overdose, you can reverse the overdose. That has become much more widely available. There have been efforts to make it more widely available for years and years. Making loxone available over the counter has been a major step, step in that direction. And so, you know, I think there is a strong argument that naloxone, I mean, I have, I have naloxone in my bag, which is like five feet away. Naloxone is much more available than it was when it was a prescription drug that was hard to come by. It's much easier to administer because it's a nasal spray now as opposed to something you have to inject. And so we, we've sort of made strides there. And in theory naloxone ought to reduce overdoses and so thereby reduce overdose Deaths. That's Thing 1. Thing 2 is a series of steps to try to increase the availability of medication assisted treatment. The most obvious one is that a couple of years ago the federal government to get really in the weeds. There's a drug called buprenorphine which is probably the best medication for addressing opioid use disorder. It helps calm cravings. It's relatively hard to abuse it itself. It doesn't have many of the nasty side effects of some of the other drugs. It's really probably the front line. It used to be very hard to prescribe buprenorphine. You used to need a special training in order to prescribe buprenorphine for your clients. The federal government killed what's called the X waiver. So the federal government killed what's called the X waiver and as a result, in theory is much easier to distribute buprenorphine through all sorts of other steps. During the pandemic, we made mobile prescribing much easier so you can drive buprenorphine truck around. That's thing two. Thing three is sort of a supply side story, which is that through some series of actions on the part of the federal government, the DEA is very eager to take responsibility for this. The cartels have at least publicly signaled, some of cartels have at least publicly signaled that they're going to stop producing and selling fentanyl. Whether and how credible that story is perhaps a subsequent note in this conversation. But there has been some aggressive action on the part of the feds, particularly trying to stop the flow of drugs at the border. There's been some movement on trade negotiations with the Chinese to Try to get them to schedule precursors of fentanyl. We'll see how well that plays out. So there are things that have happened where you could be like, these things have happened. And this therefore explains why there was this sudden dramatic reversal in the overdose death rate. That's like a story one could tell.
- [18:48]
B
And it's a nice story. I mean, you want to believe. I want to believe that government does good things and that government looks at problems and diagnoses the causes of the problems and comes up with policy solutions to stop the problems. And then you get progress as a result of smart policy. When you look at the statistics, do you believe that the decline in drug overdose deaths is primarily a story about policy success?
- [19:15]
A
No. And that's for a couple of reasons. Reason one is sort of geographic in its character, and then reason two is, I think the evidence that I find the most persuasive causes me to seriously doubt the plausible efficacy of some of the interventions I just listed. When I say geographic. Okay, so I said earlier that the OD death increase spread from east to west. You see broadly this geographic pattern. Interestingly, the OD death decrease is also spreading east to west. And so many of the western states that were late to fentanyl are also not experiencing the downturn that we're seeing pretty dramatically in some of the eastern states. That's not a uniform pattern, but it tells you something important because it means there can't be national policy variation fully explaining this. But also, if it's local or state policy variation, has to be local or state policy variation that follows that gradient. And if that's what's going on, then it is probably more likely that whatever caused the gradient in the first place is in some way shape or form contributing to what is causing the gradient now. So that's the first point, is it's hard to square the geographic variation with any particular policy story. And then point two is, very bluntly, I am not persuaded by a lot of the efficacy of the policy interventions that I just listed. Let me go through them. We have a whole body of literature on what naloxone distribution does. It can work, but a lot of the estimates are pretty small. It's probably better for reversing non fatal overdose, which doesn't mean that naloxone distribution isn't a good thing to do. But it means that as we're dealing with more and more potent drugs, the availability of naloxone can only do so much to reduce people's risk of overdosing. And the reality is you only need to overdose and die once in order to be dead. There's actually a recent paper by some of the folks at RAND where they show that naloxone distribution probably did make a difference for non synthetic opioids. And it's really hard to find an effect for synthetic opioids. And that tells you where we are, even over the counter naloxone. So that's, you know, that's, that's one the X waiver. I'm, I'm, I'm actually quite bullish about buprenorphine distribution. But I think that the simply making treatment more available on the dimensions that federal policy has, hasn't done a lot to get people into treatment. This drive says. Recently a major $300 million federally funded multi site study throws a bunch of interventions at treatment and control communities, including LOCKS and including buprenorphine access to treatment. No statistically significant difference between those who receive the treatments and those that don't. Why is that? It's because just sort of like saying, hey, treatment is available is not on its own enough to get many people who are most severely addicted to drugs off of drugs. And that's because drugs are reinforcing. It's really hard to quit. And also in many regards, people, and people want to quit, but they also don't want to quit. It's a complicated experience. It's not like my friend Keith Humphrey says, it's not like, you know, chronic pain where if you could cure it tomorrow, you would. People are more ambivalent about cures. That's thing two. And then thing three is the story about enforcement. And to me that comes down to historically border interdiction is not a very effective strategy for reducing the flow of drugs, particularly against the current drugs that we're dealing with, which are extraordinarily easy to smuggle. One piece of evidence that I have, and this probably segues to later stories, is that the drugs on the street are today more adulterated than they were four or five years ago. And that is important because it means that the cartels are not optimizing for smugglability when they bring the drugs across the border because you want the most compact product. This is something called the iron law of prohibition. The idea is that the more heavily controlled the drugs are, the more potent they are because they're more compact per, they carry more potency per unit dose. And so the fact that the drugs are getting adulterated says that there's less pressure on the cartels than there was five years ago because the Drugs are. They're prioritizing less the smugglability of the drugs. So that's one argument against that position.
- [23:42]
B
So to summarize, you're saying if this is truly a story about successful national policy, we should be looking at declines in drug overdose deaths that are subtle and that are national, widespread across the country. And instead we're seeing a decline that is large and geographically concentrated, and that should rule in other explanations. Let's turn from successful policy to a drug supply story. And I think this is sort of where you were headed just with the end of your last answer. You told us a few minutes ago that the drug industry is, you know, innovative is kind of a gross word to use here, but it might also be the accurate one, like drug makers innovate all the time, the product evolves, and historically the product has evolved to become more general, dangerous generation over generation. Is it possible that people doing drugs on the street or just people doing drugs generally today in America are dose for dose taking drugs that are less deadly?
- [24:40]
A
Yes, and that in fact appears to be what's happening. And if you talk to people who do drug testing on the ground, who report on this and are great, they will tell you that the composition of drug supply is really radically different, even what it looked like a year or two years ago. I want to highlight sort of one substance, which is the animal tranquilizer, Xylazine, which listeners may have heard about. Xylazine is a sedative. Again, it's used for tranquilizing animals, but it is increasingly added to fentanyl. The reason why is that it sedates people. And fentanyl is sort of a very fast, up, down high. It's not like heroin, which is sort of more extended. But if you want to sort of extend the period of time that people are out, you can add Xylazine. And what you see is that the thing about Xylazine, it has all sorts of harmful effects. Like people who inject xylazine often develop necrotizing wounds, wounds that will not heal. But because your overdose risk is lower with Xylazine and the quantity of fentanyl is reduced by xylazine adulteration. And because people are dosing less, because they are, are out for longer, because they're sedated, the frequency with which they're exposed to highly potent drugs declines. And so it is totally plausible that xylazine adulteration is reducing the overdose death rate even as it's yielding all of these other health harms. One piece of evidence for this, John Byrne Murdoch at the Financial Times looked at trends in Xylazine exposure, and he showed that the spread of Xylazine follows the same geographic pattern that I alluded to earlier, that Xylazine is most common on the east coast and is only sort of starting to hit the West Coast. But I do want to underscore that isn't the only substance that we're talking about here. Xylazine is already getting part of why Xylazine was getting added to the drug supply is relatively easy to obtain. It's a veterinary drug, and so you can buy it pretty easily. State legislatures have moved to schedule it and that makes it harder to obtain. And so already you're seeing substitution to other drugs. If you go to Philadelphia, where I've been, and Philadelphia was ground 0 for Xylazine 10 years ago, at this point, Philadelphia, they now have something called metatomidine, which is a different tranquilizer. This is the whack a mole that you play with drugs. And it gets to this bigger point about this is all downstream of a market that is intrinsically innovative, as you allude to, but also has just gotten much more innovative in the past 15, 20 years. That they have this real human capital here and in Mexico that just didn't exist in the 80s and 90s. And that makes a big difference for the shape of the crisis.
- [27:20]
B
To try to connect two points you've made. One, you mentioned the iron law of prohibition. The idea that when prohibition laws are stronger, the dose of the drug, the strength of the drug per dose goes up because it's harder to smuggle or move the same unit of drugs, so you have to pack more punch into a per unit basis. You also mentioned the fact that the supply of drugs on the ground, so to speak, is becoming a little bit less deadly is the story here. Tell me how this gloss of the story matches your understanding of it. That while my first question was about how policy successes could explain the decline in drug overdose deaths, there's a very strange way in which the famous policy failures of the Biden administration to secure the border have in fact made some cartels realize it was easier to ship product across the border, which, according to the iron law of Prohibition, meant they could make each individual unit of the drug moving across that border less potent than it was in a previous border control regime.
- [28:40]
A
Yeah. So I will very carefully say that I think that that story is plausible. The reason I'll be described as plausible is because the reality is all of this is fundamentally speculative. Because the sort of people who really know don't answer Reporters questions because they're drug kingpins if they don't answer surveys? No, I think that that is a very plausible component of the story that there has been this very large increase in uncontrolled migration crossing the southern border that has dramatically swamped CBP's U.S. customs border protections resources that if you are working on the border, you're simply processing people's asylum claims. And the result of that is that it is easier to move drugs across the border just because there's less capacity to address the drugs issue as a matter of like, man hours. There's a paper from, I think earlier this year that shows that places that are more exposed to trade also have higher overdose death rates, controlling for a bunch of other factors in a way that is at least plausibly finding a causal connection. And that gets to the fact that drugs are goods, they flow across the border, their products, the people who are selling them are making rational financial decisions. And so the calculus of the cartels is influenced by how easy is it to move drugs. When it becomes easier to move drugs, they can optimize on other variables. For example, to go back to the point about, you know, the decline in potency, part of what has happened five, six years ago, if you went and bought drugs on the streets of, you know, San Francisco or whatever, you were probably buying powder and then you probably injected it. Today you are probably buying little blue pills. They are cheaper than they used to be, and you are probably smoking them. That is on some dimensions, an improvement in the user experience, right? Like it's easier to smoke than it is to inject, and it's of a known dose size. And people like pills. It's improving the user experience. They're able to make that trade off against potency in part because it is easier for them to move drugs across the border.
- [30:52]
B
So we've done policy. We've explained why good policy might be responsible. We've explained why, ironically, bad policy might be partly responsible. We've also done this supply side explanation of how innovation on the drug making side of this whole equation might be filling the market with slightly less deathly drugs because the high produced by drugs mixed with xylazine and animal tranquilizer is going to be more tranquilizing than it's going to be. Sort of like, you know, filling your blood with chemicals that are going to cause you to die immediately. Let's finish off this section on drug overdose deaths by talking about the demand side explanation. That is, is it too grim to think that maybe so many people died around the pandemic of drug overdoses, that a snapback in deaths was practically inevitable. That that little drug wave within the drug wave just burned out. And what we're looking at when we look at the data is a pandemic effect winding down.
- [31:55]
A
Yeah. And, you know, it's certainly grim. And, and this is a lot of my beat is unfortunately grim. Things are also often true. And I think this is again, at least a very plausible story for what is going on here. And the way to think about this is there's always a relatively limited supply of people who are at meaningful risk for overdose death. Right. Like most people do not dose fentanyl. If you're addicted to fentanyl, you use it, I don't know, three to five times a day. You are using compulsively. You spend a lot of your time and money dedicated to focusing on fentanyl. Most people are not doing that with their time. And so there's a relatively small pool of users who are at risk of overdose. It's also the case that we have probably gotten through the period in which we were adding people to that pool. That was sort of the late 90s, early 2000s story. If you go look, for example, at rates of past year opioid use among high schoolers, they pretty precipitously declined. That's not a surprise because we're sort of at the point in the epidemic where people are like, wow, opioids. Those don't sound nice. I know what happens when people get hooked on opioids. So you have a relatively small pool of people who are at risk for overdose death, and you have fewer people being added to that pool every year, and you have an extremely high overdose death risk. The result is that at a certain point, the number of people, for lack of a better term available to overdose, is going to decline. There's a paper that came out six years ago that shows that overdose death rates have sort of been following this exponential curve since the 1970s. What you find is that if you sort of fit that same curve to the Data up through 2023, 2024, there's a big, sharp upwards deviation in about 2020. Many more people overdosed in 2020, 2021 than the underlying trend would imply. We should have expected to overdose. That's not surprising. People were isolated. The price of drugs was declining. The opportunities for drug use were rising. All sorts of different stories you can tell there about the pandemic, but the result of that is that there was clearly this sort of surge in overdose deaths. And again, it's very grim. But mechanically speaking, people who sadly died earlier were not available to die later. They mechanically could not. And so at some point there is sort of a grim dividend of that phenomenon where the trend has to return to sort of lower because you can't continue to burn at the rate that you were. That again, may not be the whole story. And I think it's quite possible that in three years the trend persists, in which case I'll say something else is going on. But given what we know in the data right now, I think it is very plausible that that's at least part of what's happening, is that we're getting sort of a. A Covid dividend, for lack of a more sensitive term.
- [34:50]
B
Yeah, I mean, one way to sum all of this up, because there's a lot of moving pieces, but there's a way in which I think a lot of the moving pieces cohere into a kind of one sentence explanation, which is that for a variety of reasons, America's drug mix is less potent than it was five, 10 years ago. There's more Narcan available, there's more Xylazine mixed into the drugs people are using on the street. There's maybe less Fentanyl per dose because of the ironies of the iron law of prohibition. I know you're not going to want to loudly endorse the story because I'm very familiar with your opinions about marijuana, but it's at least plausible that as marijuana gets stronger that maybe some people that are like marginal prescription drug users might stick with very strong weed instead. I'm not asking you to endorse that theory. It's a part of the idea that if a certain aspect of America's drug mix is a little bit less potent, even while being maybe more widespread.
- [35:57]
A
Right.
- [35:58]
B
Like maybe more people are on drugs, but the per unit hit of those drugs is weaker than it was five years ago. Then what you would expect from that change is more people being high, but fewer people getting high in a way that will inevitably result with their death. How does that work as an attempt to provide a kind of meta narrative on top of all these explanations you've been talking about?
- [36:25]
A
Yeah, I mean, something I might say in response is that in Many regards the 2020-2023 status quo was. It's very hard to believe that it was sustainable for some of the reasons you've identified for demand reasons, supply reasons, population reasons. I mean, you know, when I say 100,000 drug deaths a year, that's never happened in human history. Like, you can go back to the 19th century. There were no meaningful drug epidemics before about 1860. And you can. We have good overdose death rate data from the early 1900s on. We've, okay, overdose death rate data from the early 1900s onwards. It's not just that the rates have never gotten this high, it's that they haven't gotten within an order of magnitude of this. And so in some senses, because what is happening is we were at 100,000, 110,000 deaths, and now we're down to 90 or 80,000 deaths. And people are celebrating this sort of great change. And I'm like, yeah, you're picking around the margins, but that's still more deaths in a year than used to happen in a decade.
- [37:33]
B
Right? Still four times more than the number in, say, 1999 or 2000, where people were already concerned about the number of drug overdose deaths in America at the turn of the century. And now people like me are talking about relatively good news, right? We're talking about the rate, not the level, but relatively good news, meaning the number of drug overdeaths are four times higher than they were back when we already thought they were a problem. So I actually think that's a very important way to close the conversation, that this is a silver lining of good news. But we are not anywhere close to the moment of celebration, A, because the level is way too high, and B, because I think that a really good faith interpretation of your explanations is that we don't exactly understand what's going on and therefore there's not a lever that has revealed itself to us that we can just pull harder every year throughout the 2000s and 2000s so that we know that number is going to keep going down. I want to move from drugs to murders because this is Charles Van Leemon and all you do is talk about the most depressing trends in America. But this is again, a sad story where the rate line tells us something that's a little bit hopeful. We've seen, especially since 2021, the murder rate in America decline meaningfully and very quickly. And the first thing I want to do on this subject, the national homicide rate, is talk a little bit about the statistic itself. There has been some controversy in the news because the US Is not very good at collecting up to date national murder statistics. And so when some people report that the FBI says that murder rates are down, some other people say, nope, the FDI isn't counting all of the cities. This statistic is Made up. This is propaganda to help the incumbent administration, help us understand A, what this debate is really about, and then B, where do you stand on it? Is it wisely accurate to say that given what we can see in the numbers, we should think that the national murder rate is declining?
- [39:53]
A
Yeah, the very narrow explanation, and then I'll give to the bigger picture explanation, the very narrow explanation is, is that so we have this sort of very odd system for generating crime statistics where but since the 1930s, all of the 18,000 or so, or many of the 18,000 or so police departments in the United States have reported counts of different kinds of crimes to the FBI. And the FBI has sort of added them all together and said, this is your national crime rate and we can get into the limits of that. But that's not really the point. The official FBI numbers, and somebody says the official numbers, they're usually alluding to the FBI numbers, seem to imply that there was a very sharp increase in homicide in 2020. And then 2021, 2022, it sort of starts to come down in 2023. It's really started to come down quite aggressively. And we'll talk about some other data sources in a second. There's been a great deal of debate over this, a lot of which I think is fundamentally political, as you alluded to, where you know, depending on your political valence, and I'll say for the benefit of your audience, you know, I work for a center right think tank. I consider myself right leaning, but I've been critical of the right on this. If you're on the left, your tendency is to downplay crime because you want to say that crime is not really a big issue. And if you're on the right, your tendency is to play up crime because crime is a really big issue. And the most recent iteration of this discussion has involved some actors on the right saying the crime data are fake. They looked at a revision to the FBI's numbers, which is really about the fact that of the reporting in 2021 was reliable. For they changed the system that they have been trying to change for several decades. It was scheduled to do that. It wasn't a great big conspiracy. But it means that the 2021 numbers are largely garbage on all dimensions. And everybody who's reasonable kind of agrees the 2021 numbers are garbage. But that has been interpreted to mean they were faking it. And really crime increased in 2022 when you said it decreased. Here's what I say is I think that there are challenges to every data generating process that's necessarily true. All measurement is measurement of error. That said, I think we have a fair amount of signal that murder specifically is in decline. You have the FBI's numbers. You can look at agency specific numbers reported to the FBI and they seem to follow the general story. There's sort of some dissent that's focused on the National Criminal Victimization Survey, but that doesn't measure homicide. You can look at the CDC's numbers, that counts homicide, and that's showing a slightly different level, a still elevated level, but a similar trend of steady decline. And my favorite, and I'll plug here, there's a new service called the Real Time Crime Index, run by a guy named Jeff Asher, who is an Atlantic contributor sometimes. And I like former guest of the show. Yeah, well, I always tell people it's really insane that the best crime data in America comes from a guy named Jeff, but it does. And Jeff has been aggregating data from catching strays. Jeff is great. No, but Jeff has been aggregating data from several hundred cities. And the trend in those much more reliable records follows the national trend, more or less. And that tells me, yeah, it's probably true, it's probably real. There's a meaningful decline in murder. I want to underscore that I'm saying murder because people often say, is the murder rate up? Is the murder rate down? That's not the same as crime overall. And so something can be happening with murder that doesn't necessarily mean it's happening in other kinds of crime or the aggregate national murder figure can be down while murder is up in some places or down a lot in others. That heterogeneity is very important for the conversation.
- [43:40]
B
So establishing that the murder rate seems to be declining in the last few years, I want to trace two specific theories for why this might be happening. And the first theory I think you've already directly alluded to that what we're looking at is essentially a pandemic story that violent crimes surged when the lockdowns and other social disruptions. Really unmoored young men who commit the vast majority of violent crime and murder in America. Unmoored young men from the routines in 2020 and 2021. And then people's routines caught up. School resumed as normal for America's teenagers. People started leaving their homes more and as a result, they were filling public spaces and therefore adding potential witness eyes to various corners of the country that had previously been dark to the experience of bystanders. What part of this pandemic story America's was weird and then our behavior normalized. Do you believe?
- [44:46]
A
Yeah. So my conventional answer is that can't be none of the story. And I think there are some people who dismiss it. And the sort of strong argument for dismissing it, we'll talk about it in a second, is, well, America's murder spike was much more pronounced. Indeed, America's murder spike was almost unique among international comparison. And that tells you like everybody got hit by the pandemic. America was the one with this really big weirdo problem. Something else has got to be going on here. That said, it is extraordinarily hard to avoid that story as just sort of a very plausible explanation. And there are two different ways of telling the story. One version is, you know what I mean? Think of as the pandemic made people sad, which is like you have people who are at risk for homicide offending and the experience of the pandemic was traumatic to them. And I think this is projection on the part of people who did comply with lockdowns and did follow the rules and you know, were good law abiding citizens. And that's not the kind of person who commits a homicide. Like the average guy who's going to murder somebody doesn't really care about lockdown orders, does not affect his day to day life. But there's another component of the story which is like what criminologists talk about is social control, which as you alluded to is the idea that formal and informal institutions constrain our ability to act in antisocial ways. As you alluded to young men, the population of people who commit homicide is like really small. It's really small in any given city. It's usually like tightly concentrated pockets of young men with interconnected social networks, a tendency to honor based beefing and access to guns. That's the special, that's the recipe for homicide. Those guys sometimes are controlled by school, sometimes they're controlled by work, sometimes they're controlled by the formal criminal justice system, sometimes they're controlled by eyes on the street. All of those things went away in 2020. Whether it was going to school, going to work, whether it was the courts that we shut down, whether it was the jails that we emptied out, whether it was just the fact that particularly the most law abiding people, the most socially conformist people were not out in public. There is a decline in that level of social control. And so if you're the kind of guy who's in a feud, you're like, hey, now is my opportunity to settle the beef and maybe you go out and do it. And so that is a very straightforward explanation for I, I, unlike some other people on the right who are more dismissive of this, I think it's impossible to hear that story and not go, yeah, that's probably part of the story, but I think it's not the whole story.
- [47:16]
B
So the other part of the story that I want to get your mind on is what some sociologists and criminologists call the Ferguson effect. What is the Ferguson effect? Where does the name come from? And why do some people find it even more persuasive than the pandemic story?
- [47:35]
A
Yeah, so the Ferguson effect alludes to the large increase in murders in Ferguson, Missouri after the protests there in 2014. The phrase refers more generally to the idea that large scale protests that are critical of the police and or aggressive policy action meant to curb police behavior will result in a reduction in police activity. Either police proactivity, the amount of time you're sending walking the beat, or the level of the number of police officers. The level of policing will be reduced by that. And that will in turn make it easier for people to go out and commit criminal offenses because there will be fewer police on the block. The origin of the term, there was a sizable increase in homicide in 2015, 2016, and then it sort of came back down. That's when we first had the Ferguson effect debate. Was that caused by the racial justice protests in 2014 and 2015? Back then the social science was not great. My view is that the social science is a lot stronger now and we have pretty persuasive estimates that say, yes, all else equal, these protests yield a decrease in police productivity and can lead to an increase in particularly violent crime. Which I'll say for the benefits of your listeners is not a dispositive argument against the protests, but is at least an input into decision making at protesting. The very concrete reason that I think that this is a key part of the story in the United states, that in May 25, 2020, Derek Chauvin murders George Floyd in Minneapolis. If you look at the data on homicides and on shootings in the United States, that is the week that the spike happens, happens, that there's sort of a slight increase in April, things were sort of trending up a little bit, but there was an immediate and sharp discontinuity in the end of May of 2020, and violence and homicide remain elevated from that point onwards. That's consistent with a large reduction in police productivity immediately partially in response to major protests. Partially in response to the messaging that was sent by major protests. It's consistent with what we know, which is that particularly big city PDs have struggled to hire and retain police officers over the past several. And it's consistent with a story in which cops are much warier about doing the proactive things that we know are important for controlling violent crimes specifically. So, you know, my view is at this point, we've been through 10 years of this, we've seen it happen twice. You don't have to believe that, you know, it's never just tried to protest the police or that you don't have to leave the police above approach to acknowledge. I think there is pretty strong evidence that, that certain aggressive forms of criticism of the police do empirically seem to yield this effect. And then the question is, what do you do about it?
- [50:33]
B
Yeah, the Ferguson effect is very controversial politically and separate from whether or not the theory is true. Because to your point, I think we're still gathering data. I don't think it should be controversial politically. Which is to say, I don't think the left or the right should consider a theory like this to be out of bounds. I think sometimes the Ferguson effect is interpreted to mean that protesters volitional decisions cause declines in policing which cause increases in crime. Like a really simple three step process. The protesters cause police to pull back and then when police pull back, crime goes up. There's another way, I think, to tell this story which is that police killings cause protests which cause depolicing which cause increases in crime, which is a sort of left wing gloss on the Ferguson effect, which says that, look, we should be very clear and urgent about the idea that we need to reform police behavior. Because when egregious police killings are caught on video, it's almost inevitable that protests result from, from it. So I mean, this is, I don't know that I've tried out this theory that there's a more left wing gloss in the Ferguson effect that I think would make it less politically allergic among some people. How do you feel about that general theory that maybe there's another way to flip the Ferguson effect around to make it more appealing to people on all sides?
- [52:09]
A
Yeah, I mean, I would say a couple of things. The first thing I'd say is that certainly you can tell that is a coherent way to tell the story. Another variation on this is people sort of often implicitly acknowledge it and they'll say the cops are engaged in a wildcat strike and that they're, you know, they're just not Doing their jobs because somebody's being critical of them. And I'm like, if that's the balance that you need to give to the story, that's fine with me. I would say two things can be true. One is that the thing that you're saying is correct, that that obvious cases of police brutality will reasonably precipitate this kind of backlash. And you want to mitigate obvious cases of police brutality for that reason. In addition to the case that it's bad when public servants abuse people, that can be true. And it can also be the case that if you talk to police officers, particularly in big blue jurisdictions, their sense is that the public does not like them. And their sense is that many civilian leaders don't like them, that they view them, them in an unreasonably hostile fashion, that cops are treated as racists, as evil as, you know, the descendants of slavery. And when you tell cops that, they go, you don't want me to do my job and you don't like me. I think often about. There's a guy I was talking to, I visited Portland and there's an officer there who I did a ride along with and he told me a story about he was talking to a woman, long term history of meth abuse, lived on the streets, and he was sort of. They had a pre existing relationship and he was talking to her about. About talking to her about getting into shelter, talking to her about getting some help. And a Portland resident drives up and rolls down the window and starts yelling at her, you don't have to talk to him. You can just walk away from him. And she's like, no, he's my friend, I want to help him. And the guy in the car just ignored her and kept yelling at the officer until he left. So I guess what I would say is it is both your story is perfectly plausible to me and a good reason to pursue reform, in addition to the fact that, like murdering George Floyd is intrinsically bad. But it's also the case that I think many police officers with whom I have spoken feel a real sense of hostility. And it is worth asking, particularly on the part of political leaders, what are the costs of rewarding that kind of hostility? Because I do believe that many leaders did in 2020 not merely make reasoned, justifiable critiques of police officers, but kind of went overboard in a way that has real consequences.
- [54:35]
B
I think it sounds very reasonable. My interest here is that I think for a variety of complicated and ultimately highly political reasons, there's been this schism in America where the argument that policing tends to reduce crime, which is an argument that I not only believe but also I think is objectively true. When you look at both American history and this moment in history internationally, and do country by country comparisons, I think that sometimes that argument is seen as simping for cops. And I think it's surprising the degree to which the idea that policing is good for society can live alongside the idea that abuses of policing can be enormously disruptive to society because of this chain of effects that is described by the Ferguson effect. So what I'm trying to do here, to a certain extent, and this is like a political project that's sort of orthogonal to the reason I brought you onto the podcast, one thing I'm interested in is helping liberals see that policing makes society more orderly while holding alongside that the fact that they can still be just as critical, just as loudly critical of terrible abuses, of protesting brutality, as you said. So that's why I was trying to sort of reach this sort of supra partisan, trans ideological understanding of the Ferguson effect and the fact that I think policing overall tends to be a good.
- [56:21]
A
I'm always wary of sort of going, as a guy on the right, let me tell the left how to do things. But then on the other hand, like I've said this to a couple people, I would much rather have a left who like, agrees with me that like, crime is a serious problem and agrees with me about this stuff and you know, disagrees with me about the way to solve the problem. So I try to encourage it. What I would say is, you know, I think there's a key difference between seeing policing as an institution that can and ought to be improved versus seeing policing as a, as an illegitimate and actively hostile institution. That, that there is a deeply progressive tradition of police reform, police improvement, of seeing the police as something that sort of are a vital institution like any other government institution, but that because they're a government institution, we ought to sort of strive to always make better. That's, you know, I allude to every 30 years or so in America, we have a Democratic president and a Democratic Congress that passes some sort of major federal policing initiative. FDR did it, Johnson did it, Clinton did it. Theoretically, Biden should have done it, but the timing was wrong. And that gets the idea that policing is a public service and you should want public services to be better and you should demand that public services are better. It's a huge scandal when public services fail. But also it's a fundamental part of our society and progressives should defend the provision of public goods. And that's very different from this other idea, which is that policing is sort of a hostile, dangerous, often worse than crime institution. That I think is a separate stream in the American left that doesn't, you know, can be answered by a progressivism that says public services matter and are good, safety matters and is good. And because that's true, we're going to demand better cops. We're going to demand the bad cops are fired, we're going to demand that oversight matters. In which case, like I probably agree with 75 to 80% of the, that.
- [58:11]
B
Yeah, we're not going to fix the American politics of policing on this episode. But I agree with an enormous amount of what you said. I think that the most important thing that you said is that fundamentally keeping society safe is a job for public institutions. And we should want the public institutions charged with that job to do their job. That's probably the least controversial possible way of stating what is, in fact, the obvious. I want to close by saying something a little bit broader on crime and pull in actually a column that you just wrote for the Atlantic today. This is Monday, December 23rd. Crime in particular, I think, is very difficult to describe as a national trend because nobody experiences crime as a national trend. Nobody's like, it's okay that my neighbor's house was broken into last week because nationally, Derek just told me that home break ins declined by 11% in the first two quarters of this year compared to 12 months prior. Like, no, statistics don't keep your kids safe. Right. And crime in particular tends to be very spiky. You can have a huge crime wave in, say, Chicago, and then 100 miles away, you can have crime plummeting in various suburbs, which pulls in the question, what the hell is going on with New York City right now? You have 11 murders in New York subways in 2024, the highest figure in decades. The Manhattan Institute, your Manhattan Institute showed that in analysis, citywide assaults are at their highest level since at least 2006. Crimes like robbery and auto theft remained significantly elevated over their levels before the pandemic. So this sort of obvious pandemic effect that you and I were just describing, it turns out it isn't relevant to the experience of someone living in Manhattan or Brooklyn. What's going on in New York?
- [60:03]
A
Yeah. And that point that you make is so important, that crime is hyperlocal. Your experience of crime is very determined by what you see. And by the way, your experience of crime is often determined by the things that we think about as petty crimes, but that are often quite serious. You experience, if somebody shoves me on the street, that's a simple assault. That doesn't even go into the national crime numbers because they only had aggravated assaults. But that was still a serious criminal victimization in my experience. And so I'm much more likely to be shoved on the street by some random guy than I am to be murdered. And that gets to the point about New York. New York is an interesting example. New York is the nation's bellwether for crime. New York is sort of proof that in my mind, that you can have a big, blue progressive city that is safe. It has a much lower homicide rate even now than many Southern cities, many cities that come in for much less criticism than New York. And this is partly because the city did so much in the 90s and 2000s to bring the crime rate down. I allude to in the column you're talking about. New York's crime decline is twice as long and twice as deep as the national average. That's Franklin Zimmering's estimate, the criminologist. So I think what's happened in New York City in the past couple of years, like many other cities, they saw a surge in homicides that we've talked about. That's mostly come back down. But there are all of these other problems that keep popping up. Whether it's this surge in aggravated assaults that nobody can explain, whether car theft that everybody has a problem with or robbery, whether it's shoplifting. And New York is actually kind of an outlier on shoplifting compared to other jurisdictions that they have a particular problem, whether it's. I'm obsessed with Roosevelt Avenue in Queens, has a sudden uncontrolled surge in prostitution. And what that says to me, to go back to this idea of public safety as a public service, is, you know, the city, like every other city, has a certain capacity to deal with a set of problems. And my view is that that capacity has been eroded in the past four to for a variety of reasons. And that decline in capacity means that even as they got the big flashing homicide number under control, lots of other things got away from them. When I talk about declining capacity, I mean, the NYPD used to have 36,000 officers. Now they have 33,000 officers. That doesn't sound. You know, it's still 33,000 officers, but it's a city of 8 million people. You're just able to do less. You're able to put fewer feet on the street when you have fewer officers. And by the way lots of officers are thinking about leaving the nypd, too. It's not just that it's. New York State implemented a series of reforms to its criminal justice system, which are in many ways well intentioned and which, as a general rule I don't want to get rid of wholesale, but left holes that have made it, for example, much, much harder for prosecutors to prosecute many cases because the burden on them for evidentiary discovery has gotten much, much higher. Higher that it was and still is the case that judges in New York cannot detain people pre trial on the basis of their risk of reoffending. The only state in the union where that is true, that they change the rules for juvenile offenders in a way that's probably good, but it's not great for serious juvenile offenders. So that's one component. And the third component is just to state very bluntly, it turns out that the guy in Gracie Mansion is probably pretty corrupt and has staffed his administration with people who are selected on the basis of nepotism rather than competence. And many of them have been forced to resign, including from the nypd. So the people running the show don't know what they're doing. So all of these factors have, in my mind, conspired to say the city just has less capacity to keep its crime problem under control. And until they address all of those issues, you're going to keep having this whack a mole thing. The problem is going to keep coming back.
- [63:51]
B
I think it's a great place to end because I framed this entire episode as being a conversation about understanding why good things are happening nationally to complex phenomena like drug overdose deaths and murders. And right here we just tried to narrow the question to why is the subway violence and assault rate in New York City not enjoying the kind of declines that we're otherwise seeing nationally? And in the process of two and a half minutes, you were like, here's four or five variables that might begin to explain it, but I can't guarantee you what the relative weight of each of those variables is. We're fundamentally talking about very complex phenomena with many, many causes and many, many implications. I don't want people to be afraid of the complexity. Complex things can still be important and you can still make progress in. In complex equilibria. But I do think it's important to end on the fact that these are important questions to ask. But not every important question to ask has a safe and easy and brief answer. So thank you for not being brief when you should not have been brief. I think that was a really, really lovely, comprehensive explanation for why some really important trends are moving the way they're moving. Charles Fayn Lehman thank you so much.
- [65:16]
A
Thanks for having me.
- [65:18]
B
Thank you for listening. Plain English is hosted, written and researched by me, Derek Thompson, produced by Devin Beroldi in 2025. We are coming back to you with our regular schedule of 2ish episodes per week. We've got some awesome features cooking. We're very excited to share them with you. Thanks for listening as always. And if you like what you hear, give us five stars on whatever podcast platform you listen to. Talk to you soon.