Two implications for older men from the lead paint theory regarding prison sentences and supervised release


Design and Image Credit: Stantasyland and Cafepress

Design and Image Credit: Stantasyland and Cafepress

If you click on his website, Rick Nevin looks like a movie star in the photo that appears and not the statistics aficionado that he is. Nevin’s site has much more than a glossy photo. It contains lots and lots of thoughtful statistical analysis and well-written reports about crime rates. Although he has no training as a criminologist, Nevin has a Masters in Finance, Managerial Economics, and Strategy from Northwestern University and a Masters in Economics and Bachelors in Economics and Mathematics from Boston University, according to his profile at Healthy Housing Solutions, Inc., where is on the staff.

Nevin has an idea that has captured the attention of many serious people including Professor Doug Berman at Sentencing Law and Policy (SL&P). Basically, that idea is this: If you crunch the data, lead exposure levels better account for variations in violent crime rates than any other single variable including especially stiffer sentences. Most recently, Doug posted about an additional analysis that Nevin sent to him. See SL&P, Finding an age-based silver lining — or lead lining — in latest BJS prison data (September 17, 2014). Rather than summarize what Doug wrote, I find it more accurate to quote him including the hyperlinks and “emphasis added” in his post. That is:

Regular readers know I am very intrigued by the (often overlooked) social science research that suggests lead exposure levels better account for variations in violent crime rates than any other single variable.  Consequently, I am happy an eager to note this new data and analysis sent my way by researcher Rick Nevin who has been talking up the lead-exposure-violent-crime link for many years.

This short new piece by Nevin, titled “Prisoners in 2013: The News Media Buries the Lead,” responds to yesterday’s report from the Bureau of Justice Statistics that the US prison population increased in 2013 for first time since 2009. Without vouching for the data, I am eager to highlight Nevin’s interesting and encouraging age-based data discussion (with bolding in original and a recommendation to click through here to see charts and all the links):

The news media is reporting on U.S. incarceration data from the Bureau of Justice Statistics (BJS), but the media and BJS have ignored the important news: From 2012 to 2013, the male incarceration rate fell 21% for men ages 18-19, 6% for ages 20-24, and 5% for ages 25-29, but increased by 5% for ages 50-54, 7% for ages 55–59, and 8% for ages 60–64.

BJS Prisoner Series data show an ongoing incarceration rate decline for younger males and an increase for older males that has been ignored by the media for more than a decade.  From 2002 to 2013, the male incarceration rate fell by 61% for men ages 18-19, 34% for ages 20-24, and 25% for ages 25-29, but increased by 30% for ages 40-44.

BJS data for older age groups, reported since 2007, show the same trend through the age of 64. From 2007 to 2013, the male incarceration rate fell 37% for ages 18-19, 28% for ages 20-24, 14% for ages 25-29, and 7% for ages 30-44, as the male incarceration rate increased 22% for ages 45-49, 50% for ages 50–54, and 57% for ages 55–64.  In 2007, men ages 18-19 were twice as likely to be incarcerated as men ages 60-64.  In 2013, men ages 60-64 were almost 20% more likely to be incarcerated than men ages 18-19.

The BJS Prisoners in 2013 report ignores the detailed data on trends in male incarceration rates by age, and highlights an increase in the total prison population of about 4,300 from 2012 to 2013, but notes that the overall incarceration rate (per 100,000 U.S. residents) did fall from 480 in 2012 to 478 in 2013….

The actual BJS data show a long-term trend of falling incarceration rates for younger men that has continued from 2002 through 2013. That decline was the inevitable result of a shift in violent crime arrest rates by age since the 1990s. From 1994 through 2011, the violent crime arrest rate fell by 64% for ages 13-14, 61% t0 52% for ages 15-18, 44% to 39% for ages 19-21, 37% for ages 22-39, and 19% for ages 40-44, as the violent crime arrest rate increased by 6% for ages 45-49, and 13% for ages 50-54.

What is the causal force behind the shift in age-specific violent crime arrest rates and incarceration rates?  The Answer is Lead Poisoning.


For the sake of argument, I am about to assume that Nevin is right. But before I do, and without intending to be all-inclusive, I next list four caveats because I am skeptical that Nevin is correct that the absence of lead in our environment was or is a “causal force” in crime rate declines:

  1. Nevin does not doubt that increased rates of incarceration contribute to the decrease in crime rates. See Exchange between Bill Otis and Rick Nevin at SL&P (November 1, 2013).
  2. Before he died, Professor James Q. Wilson, “one of the foremost authorities on crime” and the man who authored the “broken windows” theory of policing that was until recently followed by the New York City Police Department, believed that “greater incarceration can explain about one-quarter or more of the crime decline.” Id. By the way, Wilson’s estimate was put forth in an essay that forms the basis of my fourth caveat.
  3. Graham Farrell at Simon Fraser University in Canada tested the lead paint theory plus 14 other theories, including the theory of increased incarceration, to determine whether a particular theory adequately explained crime rate declines in the United States and elsewhere. After validity testing, he found that all but one theory flunked. Graham Farrell, Five Tests for a Theory of Crime Drop, Crime Science Journal (2013). The only theory that passed was a theory related to increased security practices of law abiding citizens (car alarms, for example). According to his profile, Farrell holds a BSc (Surrey) and PhD (Manchester) and is a Professor and Research Chair in Environmental Criminology at Simon Fraser.
  4. Near the end of his life, Professor Wilson wrote an an essay attempting explain declining crime rates and he gave a nod to Nevin’s research, and the research of many others. Wilson came to the conclusion that crime rate reductions were driven by various factors with a big change in the culture being the foremost explanation. James Q. Wilson, Hard Times, Fewer Crimes, Wall Street Journal (May 28, 2011) (“At the deepest level, many of these shifts, taken together, suggest that crime in the United States is falling—even through the greatest economic downturn since the Great Depression—because of a big improvement in the culture. The cultural argument may strike some as vague, but writers have relied on it in the past to explain both the Great Depression’s fall in crime and the explosion of crime during the sixties. In the first period, on this view, people took self-control seriously; in the second, self-expression—at society’s cost—became more prevalent. It is a plausible case.”)  If “culture” was the ultimate explanation, “[c]ulture creates a problem for social scientists . . . We do not know how to study it in a way that produces hard numbers and testable theories. Culture is the realm of novelists and biographers, not of data-driven social scientists.” Id.

With these caveats in mind, let me finally get to the main thrust of this post. Roughly and simply speaking, here is Nevin’s reasoning:

Excessive lead exposure due to paint chips, car and truck fumes and otherwise can permanently drive down the IQ of a child and, at the same time, it can permanently reduce the child’s ability to control his impulses. Those twin harms tend to produce a criminal when the child grows up. Before lead reduction efforts by the government, crime rates rose, but after government lead reduction efforts, crime rates fell. That’s why you see reduction in crime rates for younger men, but no reduction, or even an increase, in crime rates for older men. Presumably, many of these older men have suffered the deleterious effect of excessive lead exposure when they were children, but the younger men have been spared.

Now, assume that Nevin’s reasoning is correct and the data support him. What should a judge do with this information?  In my opinion, there are two things a judge should strongly consider if Nevin’s conclusions are correct.

First, particularly for drug and gun charges (that are often associated with violence), judges should be far more skeptical of the ability of older men (40 and above) exposed to lead as children to reform themselves in prison or conform their conduct to supervision when released. Therefore, increased incapacitation–longer prison sentences–for older men exposed to lead as children may be necessary plus additional supervised release time may be required to assure public safety. For example, instead of sentencing at the low-end of a Guideline range, an older man convicted of a drug or gun charge and exposed to lead as child might warrant a high-end prison sentence plus several years of additional supervised release as compared to that period the judge would have normally imposed.

Second, particularly for drug and gun charges, judges may wish to instruct the probation departments to seriously consider whether older men exposed to lead as children should be subjected to intensive supervision after they are released from prison. In other words, judges may wish to treat older men exposed to lead as children as posing an increased risk to violate the conditions of supervise release and insist that the probation departments act aggressively when supervising such a person in order to protect the public from the offender.

In summary, Nevin’s theory, if believed, may warrant tougher sentences and tougher and longer supervision for older men who commit drug or gun crimes, offenses that are often associated with violence, if the offender has been exposed to lead as a child. While Nevin may not have intended this result, it is a logical consequence of his research. And the increased risk of reoffense that Nevin’s theory implies, is precisely the type of risk factor J.C. Oleson, Senior Lecturer in Criminology and Director of Research, School of Social Sciences, University of Auckland, proposes that we consider and that I advocate. See herehere and here.



16 responses

  1. Three thoughts:

    1) The sentencing guideline ranges already reflect penal norms that arose during the crime wave, which we’re assuming for sake of argument was caused by lead. So older men who are “lead casualties” don’t need longer sentences or increased supervision because the standard (long) sentences already incorporate the presumption of their incorrigibility. What’s needed, instead, is shorter sentences for the young and unleaded.

    2) This ignores the central demographic fact that Judge Posner so often points out: old people (even old criminals) just don’t commit many crimes (See e.g. This almost certainly overwhelms the lead-paint demographic issues. We’d end up with a lot of harmless old men in prison.

    3) Consider the moral aspect. Assuming that lead poisoning left certain people with poor impulse control and low IQs, which has given them a propensity toward crime, they are less morally culpable for their actions than the young and unleaded. It seems monstrous to punish them MORE for the same conduct.

  2. Bart,

    Well argued. Regarding point 1, young get bottom end and old get top end of the same Guideline range–your problem solved. Regarding point 2, men in their 40 and 50s are not old. Regarding point 3, they may be morally less culpable, but they are more likely to reoffend too–public safety trumps fairness issue.

    All the best.


  3. This is a tough one. On the one hand, we should punish miscreants. On the other hand, if it is true that a group of people have an anti-social tendencies due circumstances beyond their control, i.e., exposure to lead as a youth, is it really fair to sentence them to punish them more harshly? We give leniency to soldiers with PTSD who commit crimes. But, then again, PTSD is treatable (arguably). Why don’t we just have a policy that gives more lenient sentences to college graduates, for they, as a group, commit fewer crimes?

    I would hope that, as a society, that we can offer remedial measures outside of incarceration to those whose lives so adversely affected by lead in the air.

  4. The debate on lead-based sentencing or probation changes is just part and parcel of a longstanding debate about (1) punishing people based on society’s disapproval of the offense they have committed, and their mens rea (retributive theory); and (2) punishing people based on preventing danger to society and enhancing public protection (utilitarian and to some extent deterrence theory).

    Under (1) we might punish lead-exposed defendants less harshly since their diminished impulse-control makes their behavior seem less culpable; under (2) we might punish or monitor them more severely for the protective reasons the Judge has outlined.

    The debate has been around since Kant’s categorical imperative (“never use a man as a means to another man’s end”) and even longer, and I think we’ll never know the right answer. It’s an extraordinarily difficult task to sentence and we should be grateful for the state and federal judges who have to do it.

    In my view, the lead hypothesis is interesting, and there’s data to support it, but we could use a lot more proof (either way) before truly relying on it. Correlation and causation are very different things.

  5. Ancel Keys was on the cover of Time and we all ate carbs and avoid eggs, now I have a morbid fear of eggs and I am told they are good for me and I should watch my bread intake. His science was better than most of what passes for science in criminology. The publish industry in connection with tenure probably generates a mountain of bunk.
    In theory sentencing for public safety is appropriate, but if disconnected from desserts does not support putting people in US prisons, and tends in fact to produce sentences probably more consistent with the phobias of judges than real danger.

  6. Ex-Tacoma Owner Ex-Clerk,

    Congratulations on becoming an ex-clerk. And, you get no disagreement from me on your comment. Indeed, it is clearly and nicely put. But that’s what I would expect from you.

    All the best.


  7. Anon.,

    Please realize that a low IQ and impulse control problems are deficiencies I see very often whether caused by lead or chance. By definition, I don’t sentence anyone unless they are blameworthy. Thus, while I appreciate your sentiment, and agree that we could do a much better job as a society extending a hand to the unfortunate, the stark reality is that the people we are talking about are criminals in every sense of that word.

    All the best.


  8. You say “public safety trumps fairness issue” as though it’s self-evident, but it isn’t; some argument is required there. The research on lead shows that people whose behavior is affected by childhood lead exposure are literally brain damaged — their brains are physically changed in ways that make them less able to regulate their actions than people who haven’t been exposed to lead. Arguably they are qualitatively less culpable than people who commit the same acts but do not suffer from lead poisoning. Given the excessively punitive nature of incarceration as actually practiced in this country, the implications of these facts need to be grappled with in a serious way before it’s possible to conclude that the appropriate response would be to increase the sentences of people whose criminal acts have lead poisoning as a contributing factor.

  9. Pingback: Safety First? | Simple Justice

  10. Griff,

    You’re right–more explanation is required. First, as an elaboration on your ultimate point, please see Scott Greenfield’s excellent post entitled, “Safety First” at Simple Justice. Second, I would consider a “diminished capacity” argument under the Guidelines. See, e.g., United States v. Follett, 990 F.Supp. 1172 (D. Neb. 1998) (where I imposed a sentence of probation for a young deaf woman who suffered from a bipolar disorder and who was guilty of being an accessory after the fact in a bank robbery). Third, as I have said to another commentator, please understand that I can only sentence such a person to prison if that person meets the legal test for culpability. Low I.Q. and low impulse control are typically not sufficient legal reasons to excuse sanctions for criminal behavior. Fourth, I see the criminal justice sysem at the federal level quite differently than you do. It is a large part of my job, as I see it, to protect the public from criminals who are likely to reoffend even if they are otherwise a sad lot. In this sense, “fairness” is trumped by “public safety concerns” because the defendant has first been proven to be legally culpable even given his or her deficiencies and therefore it is more important to protect the innocent public from the guilty defendant than it is to be “fair” to the defendant. Bluntly, my job is not to be a social worker. I don’t have the training, and I don’t have the resources to serve in that capacity. Fifth, I agree that I must seriously grapple with the facts and that to me means being very serious about assessing risks in a clear-eyed manner. The sentencing process is not and should not be a one-way ratchet that constantly looks for ways to execuse the conduct of a legally culpable defendant and lower his or her exposure to a prison sentence. Sentencing is a process of balancing the defendant’s interests with the public’s interest.

    All the best.


  11. Pingback: ‘Positively’ Utilitarian Approach | Lawyers on Strike

  12. One question which the idea of sentencing people based upon their presumed exposure to lead poisoning as a child is that of best evidence. I gather that the main causal argument is that minimal brain damage leads to poor impulse control. The thing is, there are tests for minimal brain damage, even simple tests for lack of muscular coordination. If brain damage is the question, much better to test for it instead of relying upon exposure to lead in the past.

    Also, if brain damage is a consideration, surely testosterone level is, as well. And, as we all know, testosterone level drops with age. And it is also something that can be tested. BTW, that was why I was opposed to the California three strikes law. Doesn’t it make more sense to gives longer sentences to young men, so that when they get out their testosterone levels are significantly lower?

  13. Billikin,

    Fascinating points. Thanks for making them.

    At bottom, the more we know about whether risks to reoffend are medically based, the more intelligently we can exercise our sentencing discretion in a targeted way. There is a sense, however, that this begins to sound like “A Clockword Orange” world. I can guard against that but I can’t do much without good data.

    All the best.


  14. Judge Kopf,

    Thank you for your thoughtful responses. A few more thoughts from me:

    1) I think your job as sentencing judge is to implement the (open-ended, contradictory) policies set out at 18 USC 3553. One of the factors there is the one you focus on: the need to protect the public from further crimes by the defendant. That’s very important, often paramount. But you are also supposed to consider the “history and characteristics” of each particular defendant. I consider brain damage to be among the characteristics of a particular defendant worth taking into consideration to arrive at “just punishment,” whether or not it arises to “diminished capacity.”

    2) You say the sentencing process should not be a one-way ratchet to excuse the conduct of criminals. Of course not. But there’s a sense in which your post seems to be arguing for a one-way ratchet that just works in the opposite direction: you’ll consider the effects of lead on the developing brain, but only to the extent that it leads to longer, more punitive sentences, and not to any extent that it argues for lenience. You admit you’re not a social worker, and you’re not a scientist, but (to be blunt) you seem willing nonetheless to play that role if it means harsher sentences.

    Sentencing is indeed a balancing act, and I know that it is the most difficult thing trial judges do. One of the reasons it is so difficult is because you’re permitted (even required) to consider so many factors that point in different directions, and sometimes even the same factor points in opposite directions depending on which way you look at it. Lead exposure is probably one of those two-faced factors. It’s at least far from clear that a defendant’s socially-acquired brain damage inevitably tilts the scales of justice toward even more punitive sentences.

  15. Bart,

    To the degree you say I am required to be balanced, I agree. I am perfectly willing to look at diminished capacity as that idea is reflected in the Guidelines, but I am also perfectly willing to consider the risk of recidivism when diminished capacity either does not exist or even when it does, when offending again seems to be likely despite the diminished capacity. All the best.


%d bloggers like this: