« Latest Gallup poll indicated slight decrease in (still strong) support for death penalty | Main | "Gimme Shelter: Mass Incarceration and the Criminology of the Housing Boom" »

October 30, 2013

Do lead exposure realities continue to best explain modern crime-rate realities?

Every time I see reports new reports about crime rates in the United States or in certain regions, I cannot avoid continuing to think about the interesting research connecting crime rates and childhood exposure to lead.  Against that backdrop, I was pleased that Rick Nevin, a Senior Economist at ICF International, sent me this lengthy e-mail discussing his research and writing on this topic:

I want to thank you for yourJanuary blog about the Mother Jones article discussing my lead and crime research.  I also want to let you know that I have several posts at www.ricknevin.com that update my earlier analyses, and are closely related to recent posts:

Your October 28 post about the NYT editorial on "Why Prisons Are Shrinking" is related to my paper on The Plummeting USA Incarceration Rate showing that the recent incarceration rate decline reflects much steeper declines for younger adults (ages 18-30) born across years of declining lead exposure, partly offset by rising incarceration rates for older adults born across years of pandemic lead poisoning. 

Your October post on NYC murder rates is directly related to my post on Why is the Murder Rate Lower in New York City?

You had two posts in October about 2012 FBI and BJS data showing relatively stable crime rates related to my recent Lead Poisoning and Juvenile Crime Update paper showing that juvenile arrest rates are falling to record lows since 1980, reflecting ongoing declines in lead exposure over the 1990s, while arrest rates since 1980 have increased for older adults.  This paper also updates my crime trend graphs for Britain and Canada showing the predictive power of earlier lead exposure trends, with the same relationship between lead exposure and crime trends and the same shifts in arrest rates by age observed in the USA.  I also have a recent paper showing how lead exposure trends can explain Juvenile Arrest Rate Trends by Race and Gender

I also have a post on Lead Exposure and Murder in Latin America and a longer paper called The Answer is Lead Poisoning that updates and integrates findings from several of my related peer-reviewed studies.  All of the questions at The Questions link to this same paper.

I know the Kevin Drum story in Mother Jones seemed new and speculative to most readers, but there is actually a large body of research now supporting this relationship, and I have links to many peer-reviewed studies in my posted papers.  I don’t know of any other criminology theory that can explain both the rise and fall of crime in so many places -- and different trends by age, race, and gender -- or any theory that has so accurately predicted ongoing crime trends in so many different places for so many years.  I hope you will consider bringing some of this information to a broader audience through your blog, and I would welcome your use of any text or graphs from my posted papers.

Some recent related posts:

October 30, 2013 at 10:12 AM | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d83451574769e2019b007691a8970d

Listed below are links to weblogs that reference Do lead exposure realities continue to best explain modern crime-rate realities?:

Comments

I took your intellectual property course in 2002 or 2003. I want to thank you for writing about the relationship between lead exposure and crime rates. I grew up in (and have recently returned to) extreme NE Oklahoma where lead and zinc were mined for almost 100 years. Much of the lead used in WWI and WWII came from what is now the Tar Creek Superfund Site. The EPA is still decades away from remediating the former mine fields.

It's good to see broad research into the ongoing social costs of lead exposure. So thank you!

Posted by: Martin Lively | Oct 30, 2013 10:47:19 AM

The proposition that society's reaction to crime has nothing to do with the incidence of crime is preposterous on its face. If true of the United States, it would be the first time in the history of civilization.

Posted by: Bill Otis | Oct 30, 2013 10:53:58 AM

Bill, you know very well that this is NOT the upshot of this research or the point of this post. Nobody is saying punishment or prison "has nothing to do with the incidence of crime," rather this is all part of an effort to figure out WHAT ESLE might have a very big influence on crime. (Notably, as you should come to see, if this lead story proves true, it might well better justify mass incarceration for those populations most subject to lead exposure than most would concede.)

I always hope and expect serious folks who are opinion leaders in this arena like you, Bill, will be serious and nuanced in reaction to evidence and data and ideas whether or not those sources of information confirm or contest your pre-existing views. I hope I am not expecting too much from you in this context.

Posted by: Doug B. | Oct 30, 2013 11:51:12 AM

Rick.

Always appreciate empirical data here.

Many questions. Start with two.

China. High lead levels, low crime rates.

Passive smoke. Children in smoking households will have both high lead levels and more impulsive, addiction prone parents.

Posted by: Supremacy Claus | Oct 30, 2013 2:02:54 PM

Doug --

Since last week's unfortunate news of the rise in crime coinciding exactly with the decrease in the prison population -- after a virtually unbroken streak of 20 years of DEcreasing crime coinciding with INcreasing prison population -- there has been a frantic if scattershot effort by liberals to explain this away. One (Grits) questioned whether the recently released figures were calculated in the same manner as the ones before; another (Fred) said that the entire crime decrease was due solely to better police tactics ("Obviously in the afflicted communities, things needed to change to make them safer. But the ONLY changes needed were improved police tactics, such as the aggressive stop and frisk recently practiced in NYC. For the rest of the country and for the country viewed as a whole, ABSOLUTELY NOTHING NEEDED to be done)(emphasis added).

Accordingly, you are simply incorrect is saying (emphasis added) that, "NOBODY is saying punishment or prison 'has nothing to do with the incidence of crime'." One of your regular readers -- and not a head case -- said exactly that. You did not contradict or even question him; in fact, I'm the only one who so much as asked him for evidence to prove his claim (at which point he went silent).

"...rather this is all part of an effort to figure out WHAT ESLE might have a very big influence on crime."

I have said from the getgo that there were multiple causes of reduced crime -- which is more than I see in the excerpt from Mr. Nevin you have posted.

If, as I have said and most readers have not disputed, vastly increased imprisonment has played a significant role in vastly decreased crime, it follows, as the recent figures suggest, that less imprisonment will result in more crime. Thus the relevant question, for those opposed to what they call "mass incarceration," is specifically how much more crime, and what types, they are willing to see visited upon our citizens in order to implement their agenda.

That is what I would say is a nuanced approach to this subject. I would also note that no one has given an answer to that question.

Posted by: Bill Otis | Oct 30, 2013 3:46:37 PM

Bill, again you create straw men in your imagination: I do not think any regular reader would (or should) assert that mass executions or mass LWOP for all drug offenders or drunk drivers or child porn offenders would have "nothing to do with the incidence of crime." Rather, as I hope you also recognize, every reasonable person sees that there are considerable (huuman/social/fiscal/crime-impacting) costs as well as benefits that surely flow from "society's reaction to crime," and a wise society seeks to maximize the benefits while minimizing the costs.

The lead research, as I see it, is an effort to highlight that there may be an unexpected but very important factor impacting crime rates. I do not understand this research as an effort to prove that punishment of offenders has no impact, but rather to enhance our ability to assess your claim that "vastly increased imprisonment has played a significant role in vastly decreased crime." What lead researcher suggests is that crime decreases vastly when lead decreases vastly EVEN in places that have not vastly increased imprisonment rates. If that is true, and especially if there is a close lead/crime link, then to have a better cost/benefit experience, we should consider mass incarceration of the lead-infected and less incarceration of the lead-free.

As for what types of "crimes" could go up --- I am generally happy to see Colorado and Washington becoming states full of federal marijuana criminals, and I am hoping that a robust marijuana industry might provide more jobs for others coming out of prison to find "legitimate" work in the drug trade. More broadly, I want to be doing more sophisticated risk assessment --- not court-ordered scrambling as in California --- to better figure out who can/should be safely in the community and who cannot. If lead proves to be a very big part of this story, so be it. But the fact that SC has a more reasoned and moderate reaction to this post than you do should tell you all you need to know about nuance.

Posted by: Doug B. | Oct 30, 2013 5:05:37 PM

Bill, it does not follow that simply because greater use of imprisonment has helped to reduce crime that less reliance on imprisonment will increase crime. I have never heard anyone who cited Steven Levitt's studies in which he argued that greater incarceration in the 1980s and 1990s helped reduce crime by as much as 25 percent explain how he is wrong today when he says that the costs have so far outpaced the public safety benefits that prison populations should be reduced by one-third. It's the same guy using the same methodology, one presumes.

Posted by: Thinkaboutit | Oct 30, 2013 5:21:37 PM

"the fact that SC has a more reasoned and moderate reaction to this post than you do should tell you all you need to know about nuance."

Ha! I was just thinking the same thing! Strange days, indeed.

BTW, speaking of SC's reference to crime in China, recently I ran across an interesting piece by Stephen Levitt suggesting based on anecdotal observations that their low official crime rate may be understated.

Posted by: Gritsforbreakfast | Oct 30, 2013 5:29:01 PM

Doug --

OK, you are willing to see pot crimes increase. This leaves a lot of questions unanswered, some of them from the past, but all relevant now.

1. Do you agree that what's happening with the crime rate is the single most important measure of the success of the criminal justice system?

2. Do you doubt that imprisonment has played a significant part over the last 20 years in reducing crime? In other words, do you doubt that prison works to reduce crime?

3. If not, what are your reasons? If so -- if imprisonment is effective -- specifically how much and what kinds of additional crime, apart from pot offenses, are you willing to see increase as we send fewer criminals to prison and for less time?

4. What guarantees do you offer that, once releases begin, only non-harmful offenders will be released? If it turns out that harmful offenders are released, will that cause you to recalculate any part of your position? Are you aware that, despite California's assurances, the Plata-related (and other) prison population reductions there have in fact resulted in the release of violent offenders who, once out, did it (or worse) again?

5. How do you propose to factor into the cost/benefit analysis the increased private costs to crime victims of their victimization?

And in other news.........

"Bill, again you create straw men in your imagination: I do not think any regular reader would (or should) assert that mass executions or mass LWOP for all drug offenders or drunk drivers or child porn offenders would have 'nothing to do with the incidence of crime.'"

Who did you say was creating straw men??? Could you please quote my supporting "mass executions?" Or LWOP for drunk drivers and "all" drug offenders?

What actually happened is you claimed that, "Nobody is saying punishment or prison 'has nothing to do with the incidence of crime...'" But that's not so. One of your regular readers said point-blank that smarter police tactics were the SOLE reason crime decreased, which necessarily means that increased incarceration had nothing to do with it.

"What lead researcher suggests is that crime decreases vastly when lead decreases vastly EVEN in places that have not vastly increased imprisonment rates. If that is true, and especially if there is a close lead/crime link, then to have a better cost/benefit experience, we should consider mass incarceration of the lead-infected and less incarceration of the lead-free."

No we shouldn't. We should "consider" imprisoning people for what they DO, not for the content of their blood. Liberals used to know this.

"I am hoping that a robust marijuana industry might provide more jobs for others coming out of prison to find "legitimate" work in the drug trade."

And I am hoping to have the body of an Olympic gymnast. C'mon, Doug. Drug dealers don't want a "legitimate job" AT ALL. They want a quick buck WITHOUT HAVING some yuuuchy, bourgeois, stuffed-shirt thing like a legitimate job. That's why they went into drug dealing to start with. You can't be in the USAO for long at all before you learn this.

Posted by: Bill Otis | Oct 30, 2013 6:03:16 PM

Thinkaboutit --

Levitt has never backed off his conclusion that increased imprisonment reduced crime by at least a quarter, and his present view doesn't back off it either. His present view is about the cost efficiency, not the effectiveness, of incarceration. But opinions about whether it's worth the cost are for those paying the cost, to wit, the taxpayers.

As Doug noted in an earlier entry, elected judges and DA's virtually always do better by campaigning for more incarceration of criminals, not less.

What does that tell you about what the taxpayers actually want?

Posted by: Bill Otis | Oct 30, 2013 6:20:08 PM

From my paper on The Plummeting Incarceration Rate: "The theory linking rising incarceration to declining crime rests on two assumptions: (1) the incarceration of convicted criminals keeps those criminals from committing more crimes (the incapacitation effect); and (2) the threat of incarceration keeps others from committing crimes (the deterrence effect). The incapacitation effect cannot explain the ongoing steep decline in arrest rates for those under the age of 30, because incarceration rates by age show that we are now “incapacitating” a much smaller percentage of those under age 30. It also seems unlikely that the 91% decline in arrest rates for those under the age of 10 reflects a deterrence effect associated with rising incarceration of adults over the age of 49 when there is no apparent deterrence effect seen in the rising arrest rate for those who are over age 49."

Posted by: Rick Nevin | Oct 30, 2013 7:43:11 PM

Mr. Nevin --

What is your view of the contribution of increased incarceration over the last 20 years to the very substantial decrease in crime over that period? Would you say it has had no effect? That 10% of the decrease is because of it? 20%? More than that? Less?

Do you think that there are any additional causes, other than lead-based paint and gasoline levels, that have contributed to the drop in crime? Among the candidates, several have been suggested: better private security measures, the hiring of more police, more sophisticated methods of policing, even the readier availability of abortion.

Do you have a view of the amount of contribution, if any, of these factors?

Posted by: Bill Otis | Oct 30, 2013 11:06:50 PM

Bill, I have no idea what you are saying. Levitt's an economist. He is always talking about cost-efficiency. He thought increasing reliance on prisons in the 1990s was positive. He now thinks we've gone too far and should reduce prison populations dramatically. You're not saying you support his position, are you? My sense is that you supported his view of the 1990s and do not share his prescription today.

Posted by: Thinkaboutit | Oct 31, 2013 12:23:01 AM

Bill - My paper acknowledges that incarceration prevents prisoners from reoffending, and studies show that about two-thirds of released prisoners are rearrested within 3 years, but this is not the same as saying that incarceration trends explain crime rate trends. The USA incarceration rate rose about 400% from 1970-1994 as the violent crime rate soared, because rising incarceration was an effect of the rising crime trend. That same pattern of crime and incarceration rising for two decades or more is evident in other nations. The fall in USA crime since the early-1990s has been led by stunning declines in arrest rates for young males, even as the male incarceration rate fell from 2001-2011 by 43% for ages 18-19, 29% for ages 20-24, and 17% for ages 25-29.

I think the abortion-crime theory is nonsense. It was a coincidence that abortion in the USA was legalized when air lead levels peaked. Abortion was legalized several years earlier in Britain, but Britain reduced lead emissions much later than the USA, and the violent crime rate in Britain doubled over the 1990s as it plummeted here.

Police per capita in the USA and other nations looks like a mostly flat line across decades, with no apparent relationship with the large rise and fall in crime across decades. My statistical analysis also found that unemployment rate trends showed no significant relationship with violent crime trends and explained only 1% to 5% of the year-to-year variation in property crime rates.

I know this is hard to believe, but all the data I have looked at suggests that the impact of lead exposure on crime trends across decades is so strong that it is difficult to discern any clear consistent long-term impact from other factors.

Posted by: Rick Nevin | Oct 31, 2013 1:07:10 PM

Mr. Nevin --

Thank you for your prompt and very interesting response.

Your last sentence is this: "I know this is hard to believe, but all the data I have looked at suggests that the impact of lead exposure on crime trends across decades is so strong that it is difficult to discern any clear consistent long-term impact from other factors."

I would summarize that as follows: Lead exposure is the only factor that, over the last several decades, has been shown to affect the crime rate.

Is that a correct summary of your view?

Posted by: Bill Otis | Oct 31, 2013 2:13:20 PM

Yes, if you emphasize the words "several decades" and end your sentence with crime rate "trends". The spike in Chicago murders in 2012 is something I can only explain as random variation, but that spike just brought the Chicago murder rate to a little over half of what it was in the early-1990s, and year-to-date numbers show the 2013 Chicago murder rate will be a new multi-decade low. I don't know what the Chicago murder rate will be next year, but I am confident that Chicago will record many more record low murder rates over the next two decades.
I try to choose my words carefully because I appreciate the important work done by law enforcement, from police to prison guards. My father-in-law was a 30-year veteran of the NYPD, and knew more than I ever will about policing, and he was generally supportive of policies adopted by Bratton. That said, he also knew a lot about lead poisoning thanks to his daughter's choice in marriage, and he felt that my research answered a haunting question that he and his peers on the NYPD talked about from the 1950s to the 1980s: Any cop working in a large American city across those decades knew that something horrible was happening to our society, but didn't know what the root cause was. Better policing strategies and higher incarceration rates for repeat offenders have almost certainly made the crime rate lower than it otherwise would have been, but I don't think anyone would have seen any apparent benefit from those efforts if we had continued to increase lead exposure for another few decades after the early-1970s, because the best efforts from law enforcement would have been like trying to hold back a toxic tide.

Posted by: Rick Nevin | Oct 31, 2013 4:44:48 PM

Rick Nevin is ignoring two counter examples.

High lead levels in the children of China, low crime rates. The high lead levels in homes with passive smoking, but also containing parents transmitting genes for impulsivity and irresponsible behavior. Lead levels of children in Egypt and Mexico are similar. Both exceed those of the US. Until the Revolution, crime rates in Egypt were low by population survey, not official statistics. Crime in Mexico is higher than in the US.

There are dozens of kids with high blood lead levels for every one that becomes a violent criminal. Lead level may be a marker for social and economic disadvantage from the parents ADHD, addiction, and impulsive, irresponsible Roman Orgy lifestyle, rather than a cause of criminality.


As people may know, I have an alternative theory. The crime rate correlates with the supply of lawyers/population ratio. Japan low crime rate, active death penalty: 100 million people, 20,000 lawyers. Egypt low crime rate, low fraction of lawyers, many honor killings under Sharia, very active religious death penalty. Mexico, more lawyers than the US for population, high crime rate. Non-corrupt Mexican Navy special forces have begun killing drug dealers. The murder rate has leveled off. China in imitating the US has increased its number of lawyers. Crime is soaring, now. Very active death penalty, likely in the 1000's each year. Their death penalty applied for economic crimes comes closest to that I propose. They are not killing violent predators from ages 14 to 18, as I have proposed.


While the effect of lead level on crime rates (if real) is weak, that of bastardy is beyond question. The lead level may be a mere marker of bastardy, not relevant to causation. Lead level is left wing speak for bashing productive factories, landlords not spending millions to abate paint problems. It is a pretext for aggregate litigation.

Lead paint has been used for centuries, including periods of low crime rates.

One feature unifies very rich and very poor nations with low crime rates. Public self help. Criminals fear the neighbors far more than the worthless police. The police are the agents of the worthless lawyer prosecutor and completely hamstrung. I can say it. The police is worthless in high crime areas.

I live in a lawyer residential neighborhood. Try pulling a gun in a store. Three cruisers show up in 2 minutes. The college educated police come out blasting. The death penalty is at the scene for the minority thug who has not gotten the memo. Crime rates lower than Japan, China, Egypt, only 5 miles from a Fallujah like black ghetto. Blacks basically have no police protection. They do not show up, unless the caller lies and says, he thinks a police officer is down. Although lawyers live where I do, few work there.

Thank the lawyer for the crime rate, and its widest variation within a small geographical area.

It makes no sense for a Chinese criminal to think he can get away with anything when he has seen so many cohorts get beaten up by the civilians, and so many summarily executed by the authorities, no mater how high the lead in his blood.

Posted by: Supremacy Claus | Oct 31, 2013 8:48:47 PM

Rick is presumed to be a biased left wing, Democratic Party operative. I would like to know whom he voted for President in 2008 an in 2012. He also just ignores counterexamples, China, passive smoking. He just refuses to provide the intermediate steps from a blood test to a complex social behavior like crime. Contrast his weak data to that on alcohol. Half the murderers are drunk. Half the murder victims are drunk. The same for lesser batteries. There is a substance strongly associated with violent crime, alcohol.

The correlation between lead level and child psychopathology associated with criminality, is weak. An R of .15 needs to be squared to get the real weight of the factors. So we are talking about 1or 2 % strength of effect.

http://www.sciencedirect.com/science/article/pii/S0272735813000123

Here is the modern view of catastrophes. And a high crime rate should be viewed through this lens.

Multiple factors cluster in time and space, and cause the catastrophe. It is likely that the bigger the catastrophe, the greater number of factors and the more complex their interactions. The average number in a crash is around a dozen. Subtract just one, and the crash is prevented. So attention to each one is very useful for prevention, as in aviation.

Lead is likely a factor causing brain damage (never linked to profitable criminality, but linked to impulsive, violent behaviors) Police show up blasting. The neighbors will kick your ass. There are 1000's of executions in your country, including the numerous extra judicial killings by non-corrupt Navy special forces. Lastly, the lawyer profession will crush the neighbors who defend themselves against ultra-violent criminals, They will protect the criminal and allow 10,000's of thousands of murders of innocent people to keep their government make work jobs. Yes, Rick Levin, the Rent Seeking Theory is the most powerful theory of crime rates across time and space. It is possible stronger than the effects of alcohol, an established and undeniable cause of violent crime.

Posted by: Supremacy Claus | Oct 31, 2013 9:13:00 PM

As part of this multi-factorial approach to catastrophe management, I have praised the Sentencing Guidelines, as the greatest achievement of the Twentieth Century, by the lawyer profession. Genius stuff, and showing the potential of the profession to do good, rather than the pure evil they always do. Sentencing Guidelines were an all lawyer project, induced by the Comprehensive Crime Control Act of 1984(P.L.98-473).

They dropped the crime rate 40% without touching the happy go lucky, free wheeling ethos of our American culture. Judicial guidelines were definitely not a self evident remedy, and show a subtlety of mind and loft of intelligence totally out of reach to the ordinary civilian like me.

I want fairness credit for this praise of the lawyer profession. Bill is more correct than Rick here, at the gut and logic level.

Posted by: Supremacy Claus | Oct 31, 2013 9:28:25 PM

Criminality in Egypt has exploded. Has the lead level markedly increased in the past two years, or has the chance of getting an ass kicking after a crime dropped like a lead balloon?

Pollution control is being enacted in China. Crime is rising, along with the number of lawyers.

If Rick is a non-partisan economics researcher, he should mention and try to account for the exceptions and weaknesses in his arguments, without making the reader do research.

Posted by: Supremacy Claus | Oct 31, 2013 9:41:28 PM

Lead blood levels vs. Bastardy

Which could possibly be the stronger factor in the rate of criminality?

Bastardy is 100% percent caused by the lawyer profession, the vile feminist lawyer.

I would like to see a post on the remediation of the toxic effects of the lawyer profession. They are fully responsible for all social and economic factors that cluster to cause massive criminality in our nation. They protect the criminal. They crush all self help. They destroyed the black family, and are coming after the white family from all sides. The lawyer is winning on all fronts. The rate of white bastardy is now 40% up from 25%, headed to the black level of 70%. And because there is no difference between the races in predispositions to criminality, all black social pathologies will reach the white population.

The reason the lead level research has acquired such prominence is not any validity, since there is no linkage save weak correlations, just a bunch of ascertainment biases, misguided meaningless correlations, and ignoring of countless counterexamples, is that his propaganda is anti-corporation, anti-landlord, pro-tort litigator, pro-tax, pro-regulation left wing lying by omission.

I do appreciate Prof. Berman's highly provocative posts. Sorry Professor, but nuanced and moderate do not work with Rent Seeking, Democratic Party partisans. His organization is a front organization for Fannie Mae, and the Democratic Party. Fannie Mae was headed by a former black gang banger who rode it all the way down to its sorry state today. He was part of the gang that forced banks to give mortgages to totally irresponsible minority members, upon threat of revoking their charters. This black criminal has not even been interrogated about the damage done to our nation.

To blame lead levels for the massive effects of bastardy, black crime, the lack of police protection they endure, is ridiculous on its face.

It is a waste of time, but I wonder if Rick Nevin has the courage to debate Bill Otis, on the numbers, the statistics, not the law, never mind me. Bill would be completely justified in refusing to do so, since this is an advocate, uninfluenced by the facts.

Posted by: Supremacy Claus | Nov 1, 2013 1:04:29 AM

I understand that there is reasonable skepticism from people who haven't read much about lead and crime, which is why I have tried to make the data and research on this subject more accessible at www.ricknevin.com. Steven Hayward at Powerline is not someone you can call a "biased left wing, Democratic Party operative", so I recommend his post at http://www.powerlineblog.com/archives/2013/07/crime-and-punishment-get-the-lead-out.php that reflects a serious consideration of the evidence.

Posted by: Rick Nevin | Nov 1, 2013 2:46:59 PM

Mr. Devin --

I know Steve Hayward and the other authors at Powerline (one of whom was best man at my wedding (and was over for dinner last night)), and I read Steve's post.

Steve's post says, "There is some intriguing research [principally yours] that argues in favor of the conclusion that the decline in lead levels in the environment may have a significant but heretofore unrecognized role in the fall in crime." He then cites eminent criminologist James Q. Wilson: "Before he died last year, James Q. Wilson—father of the broken-windows theory, and the dean of the criminology community—had begun to accept that lead probably played a meaningful role in the crime drop of the ’90s. But he was apparently an outlier."

In that connection, I would note that Wilson wrote this in a WSJ piece two years ago (http://online.wsj.com/news/articles/SB10001424052702304066504576345553135009870): "Why is the crime rate falling? One obvious answer is that many more people are in prison than in the past. Experts differ on the size of the effect, but I think that William Spelman and Steven Levitt have it about right in believing that greater incarceration can explain about one-quarter or more of the crime decline."

I would also point out that for Hayward to say that the decline in the use of lead "may" have had a "significant" role in crime reduction; and for Wilson to say that "it probably played a meaningful role," is not to gainsay that other factors were also at work, and that one of those factors, to the tune (in Wilson's words) of "one-quarter or more of the crime decline" is the increased use of imprisonment.

To put it another way, what I'm saying here is not that your findings are erroneous. It's that they, and the sources you (wisely, in my view) employ to vouch for them, use careful and qualified language. They reach conclusions consistent with what Spelman and Levitt have said: That no one factor accounts for 100% the enormous decrease in crime over the last generation, and that the increased use of incarceration also helped to a not insignificant degree.

So let me ask: Do you view your work as contradicting Prof. Wilson's WSJ piece?

Posted by: Bill Otis | Nov 1, 2013 3:44:22 PM

Bill,
I actually traded emails with Wilson after his WSJ piece, and he was kind enough to make some last minute corrections to the City Journal version of the same piece to clarify that my 2000 study on lead exposure and violent crime predated the Reyes study. I was also tempted to ask for another clarification (but I didn't want to push my look) on his statement that one "oddity" of the lead-crime theory is that "the reduction related to lead-free blood included only violent crime, not property offenses". On pages 15-16 of my paper called The Answer is Lead Poisoning, I note that this oddity was unique to the Reyes study. Denno, Needleman, and Wright all found a relationship with property and violent crime offending among youths in Philadelphia, Pittsburgh, and Cincinnati. My 2007 study also found a very strong relationship with property crime trends through 2002 in every nation studied, and lead exposure trends in that study have since accurately forecast a more than 50% fall in the burglary rates in Britain and Australia since 2002.

With respect to incarceration, recidivism data makes it clear that the crime rate would spike if we let all incarcerated prisoners go free tomorrow, and the incapacitation of career criminals clearly has kept crime lower than it would have been if they had all been repeatedly released on probation. Teasing out how that logic translates into the specific percent of crime trend variation that is attributable to incarceration is much more problematic. I disagree with Wilson's conclusion that Levitt is "about right" that incarceration explains 25% of the crime decline, because that conclusion was based on crime data largely limited to the crime decline years. If you ran the same statistical analysis on incarceration rates and violent crime from 1960-1990 you could reach the irrational conclusion that rising incarceration increases the violent crime rate. In fact, the first paragraph of Conclusions from Levitt's paper on the 1990s crime decline states, with admirable candor: "The real puzzle in my opinion, therefore, is not why crime fell in the 1990s, but why it did not start falling sooner."

My earlier post was not meant to suggest that Hayward had endorsed my findings, but only to refer another commenter to someone who is not a liberal but who did take the time to visit my web site and seriously consider the research presented. Moreover, Kevin Drum is a self-described liberal but he has also been skeptical that lead exposure trends can explain the large percent of crime rate variation suggested by my studies. I have to admit that I was also reticent about that finding after my 2000 study, and even after my 2007 study, but as more years of crime data fall into place where earlier lead exposure trends anticipated it would, I feel more and more confident that I am not overstating the magnitude of this effect.
- Rick

Posted by: Rick Nevin | Nov 1, 2013 5:35:45 PM

Excellent commentary Mr. Nevin, thank you.

Posted by: Grant | Nov 1, 2013 6:58:06 PM

Well bill they could both be right. The lead in paint for decades could have damaged the brains of 10's of millions. Locking up a big part of them then controled them and brought the crime rate down!

Unfortunatley from what i remember reading there is not much that can be done with this type of brain damage.

Posted by: rodsmith | Nov 1, 2013 7:01:47 PM

rodsmith --

"Well bill they could both be right. The lead in paint for decades could have damaged the brains of 10's of millions. Locking up a big part of them then controlled them and brought the crime rate down!"

As I've said a few times, you're an underrated commenter.

Posted by: Bill Otis | Nov 1, 2013 7:05:45 PM

Rodsmith and Bill:

Don't forget that "paint" is only one source of environmental lead. We also removed tetra-ethyl lead from gasoline during the same time frames.

Posted by: albeed | Nov 1, 2013 10:10:24 PM

I have only now had a chance to review this comment thread, and I think it highlights (with a few rants notwithstanding) how valuable it can be to have an on-going discussion of these nuanced issues. And, to that end, I finally can respond in a (too) nuanced way to Bill's 5 questions to me from way above:

Bill asks:

"1. Do you agree that what's happening with the crime rate is the single most important measure of the success of the criminal justice system?"

Simple answer: NO. If it was, we certainly should endorse a criminal justice system calling for the execution of all males who commit the crime of dropping out of high-school (and/or have mental health or substance abuse problems) as data shows those guys always commit the most crimes. Instead, I think an array of American values (e.g., life, liberty, pursuit of happiness, due process, equal protection, democratic self-rule, etc) all are important measure for the "success" of any US public institution.


"2. Do you doubt that imprisonment has played a significant part over the last 20 years in reducing crime? In other words, do you doubt that prison works to reduce crime?"

Simple Answer: NO. But the real issue now, especially with US imprisonment levels so high, is at what cost to other values, particularly in a nation purportedly "conceived in liberty." Incarceration is very costly, and I thus I am always eager to explore less costly alternative to imprisonment as a means to reduce crime.

"3. If not, what are your reasons? If so -- if imprisonment is effective -- specifically how much and what kinds of additional crime, apart from pot offenses, are you willing to see increase as we send fewer criminals to prison and for less time?"

Answer: I am eager to see if we can find means to reduce imprisonment AND reduce crime. At least a few states have found ways to do that in recent years, including the diverse states of Texas and New York. With the help of good research and good government work, we should be able to figure out what works there and seek its replication nationwide.

"4. What guarantees do you offer that, once releases begin, only non-harmful offenders will be released? If it turns out that harmful offenders are released, will that cause you to recalculate any part of your position? Are you aware that, despite California's assurances, the Plata-related (and other) prison population reductions there have in fact resulted in the release of violent offenders who, once out, did it (or worse) again?

Life has few guarantees, and I am certain some harmful offenders will get released and do harm. But just knowing some bad people will in the future hurt others with guns (and cars and knives) does not alone justify banning all guns (and cars and knives). Similarly, I do not think just knowing some bad people will in the future come out of prison and hurt others does not alone justify giving everyone an LWOP sentence. (Keep in mind, there are already 700,000 releases every year.)

"5. How do you propose to factor into the cost/benefit analysis the increased private costs to crime victims of their victimization?"

I strongly favor seeking means to quantify the private costs of various crimes, though for many property crimes insurance, diminishes/spreads these costs. (E.g., my homeowners insurance covered for me the costs of having stuff stolen from my garage.) I think this is a critical part of the analysis that, sadly, economists have not often help provide tools needed for effective analysis.

Thanks, as always, for the engagement.

Posted by: Doug B. | Nov 2, 2013 4:57:07 PM

Prof. Berman: Watch out, your indoctrination is weakening in the face of some reality from here on, earth, "...for the execution of all males who commit the crime of dropping out of high-school (and/or have mental health or substance abuse problems) as data shows those guys always commit the most crimes." You know this already, the real remedy to crime, incapacitation. Because neither dropping out, nor addiction is illegal (I support the return of status crimes), 123D is and is a good substitute. I support massive funding into prenatal markers of Antisocial Personality Disorder, and their abortions in the first and second trimesters, the forced sterilization of promiscuous females carrying the gene. That would end the necessity for 123D, and the death penalty. One more piece of reality you and legal academia continue to ignore. There is a lively death penalty, of 17,000 murder victims a year, by gruesome methods, and of innocent people and strangers. The foreseeability of that rate makes the lawyer profession guilty of a crime against humanity, especially since there is a predominance of dark skinned people among the victims. Racist, and inhumane, morally disgusting, is the state of your profession. Add the bad faith motivation of rent seeking, and arrests are fully warranted, quick Nuremberg style trials, and immediate hanging are overdue for the hierarchy.

You want some moderate and nuanced? You will not like it too much. Say an air or car crash has a dozen factors meeting in one place and one time, and the prevention of any one would prevent the entire crash. The bigger the catastrophe, the more the factors, the more dynamic their interactions, the less certainty about causation. If you are talking about 20 million Index felonies, the factors likely number in the hundreds with complex interactions.

If lead intoxication from paint, from gasoline, from passive cigarette smoke increase crime by increasing brain damage and causes more impulsivity, decreasing IQ, including decreasing moral IQ, more ADHD, more presentism and inability to think about consequences, then it likely only increases nuisance crimes, fighting, assaults, even sexual crimes. It will not affect complex business like crime, such as organized crime, planned crimes, frauds, serial killers stalking victims, crimes requiring stealth, planning and skill.

Here are some other factors, with which lead levels may correlate and merely serve as a marker. Bastardy. Bastardy explains all racial disparities in victimization and commission of crime. You will agree it explains a lot more national statistics, and inter-state differences. China, Egypt, high lead levels, no bastardy, low crime.

Alcohol. Half the major crimes are committed by drunken people, murders, rapes, assaults, batteries. If lead is a spit ball, alcohol intoxication is a cannon ball lobed at our nation.

Compare the effect of incarceration, in the greatest achievement of the lawyer profession, and one that dropped outrageous, jungle like crime rates by 40%, without eradicating the criminal.

One has to account for the motivations, the planning, the fashion, the decision making of millions of criminals, pretty complicated.

Why does Rick Nevin not do a longitudinal study tracking crime rates in individuals and correlate them with blood lead levels the same year. He would need an immunity certificate from the department of Justice and ask the criminals themselves how busy they were recently, then comparing their blood levels at that time. The answer is his anti-corporation bashing agenda, and an agenda to provide potential fodder for massive aggregate claims against land lords, paint companies, and tobacco companies.

David Duke, the white supremacist, has a real PhD. He cherry picks news stories and statistics about black people. He is so biased, his web site would be called racist propaganda with zero credibility, despite his never telling a knowing lie. He lies by omission. When it is not the blacks, it is the Jews with him. He nevers says a falsehood.

http://davidduke.com/

How does it differ from this similar looking, big font, yelling site?

http://www.ricknevin.com/

Posted by: Supremacy Claus | Nov 5, 2013 12:11:49 AM

Prof. Berman: That format of your last comment, do you know where you got it?

Here is the tightest IRAC you will ever encounter.

http://www.lonang.com/exlibris/aquinas/sum12098.htm

He got the format from this guy:

Peter Abelard, who answered 158 questions in his Sic et Non.

Disputation was the method used by Schoolmen to resolve difficult questions, and that is the origin of the adversarial process. It was just the best in the 13th Century. Is it the best today? Everything from the 13th Century is thought to suck today, including the adversarial process. The judge and the court look ridiculous to anyone modern. Buffoons in Halloween outfits, playing Inquisitor, from an elevated bench, with a gavel. They look like really demented jerks, crazed clowns. All aspects of the modern court violate the Establishment Clause by their copy of Catholic Church practices. People will argue over a tiny cross in a flag of a city that really represents a famous tourist attraction. They do so in a copy of a church, before a clown in Inquisitor outfits behaving like a priest.

I know you studied philosophy, so this comment is for the historical edification of those who did not have a decent high school education.

Posted by: Supremacy Claus | Nov 5, 2013 8:10:21 AM

"5. How do you propose to factor into the cost/benefit analysis the increased private costs to crime victims of their victimization?"

I strongly favor seeking means to quantify the private costs of various crimes, though for many property crimes insurance, diminishes/spreads these costs. (E.g., my homeowners insurance covered for me the costs of having stuff stolen from my garage.) I think this is a critical part of the analysis that, sadly, economists have not often help provide tools needed for effective analysis."

How about allowing tort liability when a deviation from professional standards of due care of the judge, the jury, the police, as agents of the prosecutor, and the prosecutor result in damage to crime victims, in a foreseeable manner? These are the most incompetent professionals of all other professions. They allow 90% of serious crimes to go unanswered, and when they have a guy, it is the wrong guy in 20% of cases. Imagine a mechanic who fails to repair 90% of broken cars, and when he eventually repairs one, he fixes something that is not broken 20% of the time. How long would he be allowed to stay in business. Tort liability is absolutely necessary to help the criminal law improve.

Posted by: Supremacy Claus | Nov 5, 2013 8:18:46 AM

Post a comment

In the body of your email, please indicate if you are a professor, student, prosecutor, defense attorney, etc. so I can gain a sense of who is reading my blog. Thank you, DAB