Shorter Kevin Drum:
1. Lead is remarkably nasty stuff. In minuscule quantities (measured in the single digits of micrograms per decliter of blood) it damages both IQ and the capacity for self-command.
2. The rise of lead exposure from gasoline, and its subsequent decline, accounts for a large share of the crime boom of 1960-90 and the crime decline that started in 1994.
3. Children are still being exposed to lead from two major sources: residual lead in soil, and lead paint in homes, especially window-frames.
4. Substantially eliminating those sources of exposure would require a one-time expenditure of about $400 billion. (That sounds like a lot of money, until you look at Treasury bond rates and translate it to $12 billion a year.)
5. The return on that investment would be at least in the high tens of billions of dollars per year.
Kevin’s piece, the cover story in the current Mother Jones, is a model of science/policy journalism. He carefully chases down both the biology and the social science supporting the claims about the effects of lead, identifies the main remaining sources, and documents the abatement processes, with cost estimates. (The quantification of benefits is mostly hand-waving; neither the outcome predictions nor the valuations are convincing. But it’s more than adequate to show that the benefit-cost analysis would come out hugely positive.)
It remains to be seen whether a crime-control and educational initiative that doesn’t fit the intellectual categories or feed the budgets of the criminal-justice and educational systems can obtain any political purchase. The answer so far seems to be no.
I think this would be a great thing to do, but I don’t the $12 billion a year is the right cost estimate. That is just the interest cost, but you probably should add in some repayment of principal eventually. If you amortized the principal over the 30 year life of the bonds, it would be more like $20 billion/year. Still worth it.
If the benefits are recurring forever vs. the do-nothing option (probably not quite true, but pretty close) then the way the costs/benefits are framed here is probably right. There is a $12B annual cost in perpetuity and a much higher annual benefit in perpetuity.
It’s a wonderful example of the scientific and moral bankruptcy of those who oppose environmental health regulations or the precautionary principle.
Mark: How does lead account for the drop in crime in NYC being so much larger than anywhere else — in some years as you know it was as much as 1/3 of the murder rate.
I assume that its because NYC had a higher spike, due to greater urban density and thus greater environmental lead levels. The murder rate in NYC was very high to begin with (i.e. at its peak), and thus had a long way to drop.
And besides - nobody is really arguing that lead explain 100% of the violent crime changes (I would guess that even those who show a 90%+ explanatory power for lead in their analysis suspect that they are picking up some impact from an as-yet un-modeled third variable). NYC did implement generally good policing practices, and did benefit from a generally strong economy during the period in question too.
The great source of bafflement is why my home town of Chicago has fared so poorly. I suppose its possible that environmental lead is higher here. But the violent crime clusters strongly by neighborhood (see the map in today’s NYT for evidence). Hyde Park cannot plausibly have that much less lead than the neighborhoods immediately north, west and south of it. The fact that the University of Chicago supplements the Chicago PD with its own University Police Force (one of the largest police forces in the state of Illinois, IIRC) has to have some explanatory power as well.
Neither Kevin nor I is a mono-causalist on this. I don’t doubt that the way Bill Bratton and Ray Kelly used the huge police force Giuliani inherited from Dinkins made a difference. It’s possible - I don’t know the facts - that NYC’s density meant higher lead exposure there than elsewhere. But crime fell like a stone almost everywhere.
Drum has “Gasoline lead may explain as much as 90 percent of the rise and fall of violent crime over the past half century.”
and speaks of the *real* criminal element, so he is writing as a virtual mono-causalist even if he isn’t one. The charts he produces shown the big drop in lead is since 1979, but that’s where the lines depart (leads plummets, crime drops only modestly). If one wants to argue it’s a lagged effect, lead exposure in children leads to crime 10 years later, why is the shape of curve so similar in the 1970s?
UPDATE: APOLOGIES TO KEVIN AND READERS HERE, I MISREAD HIS CHART, THE CURVES OF CRIME AND LEAD ARE OVERLAID BUT REFER TO DIFFERENT ERAS, SO IT IS A LAGGED EFFECT.
Not saying lead doesn’t matter it probably does, but I think in the quest to psuh back on other over-sold explanations, this one is also being over-sold.
The world is full of other countries that also had leaded gasoline and then stopped. (And a few bizarre holdouts that STILL have leaded gasoline, though that’s not relevant here. I was horrified when I went to South Africa a few years ago to see that they still had leaded gasoline, and plenty of kids walking to school along the side of the road, though SA has recently phased out leaded gasoline.) The phaseouts occurred over a wide range of time.
Given the importance of the issue, it seems like there is value in performing an equivalent analysis on crime rates in all these other countries. This should provide enough data to tease out lead vs other effects (eg crack, legal abortion, etc).
My understanding is that this has in fact been done. And the effect shows up in every country that moved away from leaded gas. The timing of the drop in crime is different in every country but by a similar gap to the timing of the removal of lead from fuel. Kind of spooky result.
Kevin discusses this in the article and provides a link to this paper:
http://pic.plover.com/Nevin/Nevin2007.pdf
I’m not a statistician but it looks pretty damn good to me.
“the *real* criminal element”
Surely here you’re placing too much weight on a pun?
If you look at Rick Nevin’s paper (which is what Kevin Drum refers to, p. 32), you will see that he found an R^2 of .9 for all violent crime combined, but a lower R^2 for each individual type of crime: .46 / .6 for murder; .89, .87, .79 for assault, rape, and robbery, respectively (with varying assumptions for lag). Thus, you can have a higher unexplained variance in the murder rate compared to all violent crime combined.
I wouldn’t be so taken with the aggregate-level statistics if we didn’t have individual-level prospective data to back them up.
http://www.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal.pmed.0050101
Am I wrong in thinking that the aggregate-level statistics are themselves fairly impressive? Because while I know these things are very tricky, we’re talking about correlations that appear on multiple scales in multiple countries and come with an obvious causal mechanism. That sounds pretty good to me but then I know nothing about social science so that’s not worth too much.
But when I add that paper into the other ones I’ve looked at today, this seems at least modestly slamdunky. Kevin talks a lot about this being ignored by the criminology community but is there any active opposition to it?
But, but, but. Regulation! Freedom of contract! Coasian bargaining, etc.
Or maybe not.
I would note (on this subject, I can speak from experience-I made a living at construction for a decade) that the estimates of costs and benefits either overestimate the costs or underestimate the benefits-this is one of the rare subjects where the analysis isn’t as good as the truth.
Every study I’ve ever seen assumes the costs as just remediating existing housing. That is needlessly expensive; in most cases (and the most problematic cases, since good maintenance reduces lead exposures dramatically-basically keep the lead paint covered with lots of other paint and don’t let paint dust stay around in the house), it’s not much more expensive (assuming-the key assumption-that you can rebuild the same sort of buildings on the same lots) to just tear down the houses and rebuild. Houses from pre-1978 are nearly always poorly insulated, poorly wired by modern standards, and have worn components generally; in a lot of cases, you can build a new house with modern-standard insulation and wiring for only a little more than lead removal in an old house.
Lead-based paint is wonderful stuff - dense, so it covers quickly and completely; highly reflective, so it makes rooms bright; smooth, hard, and flexible, so it cracks and chips less than alternatives; once dry, easy to clean with soap and water, so it’s great for kitchens and children’s rooms. It’s a pity it’s poisonous.
Indeed. I recall some years ago a friend in the remodelling business making much the same complaint.
There are “unexplained” decreases in child maltreatment beginning in the early 90’s that could also be attributed to lead reductions. During this time there was increased emphasis on reporting yet the rate of incidents have steadily decreased since then. As far as I know nobody really knows why, but I’ve been a quiet proponent of the lead theory since I heard about it a few years ago (from Mark Kleiman).
Re: “It remains to be seen whether a crime-control and educational initiative that doesn’t fit the intellectual categories or feed the budgets of the criminal-justice and educational systems can obtain any political purchase.”
Will it be a prominent Democrat that stands up and explains that, through no fault of their own, the urban poor (disproportionately black, obviously) are disproportionately lead-poisoned and brain-damaged, and consequently have reduced IQs and are more prone to violence? Or are we waiting for a Republican to deliver that hot potato?
I think the lead abatement argument is convincing and a bargain at $20 billion a year, but I don’t see how our political system can discuss it sensibly.
There are two effects: Lower IQs and lower impulse control (i.e. being prone to violence).
The former is a better sales pitch to the broad middle to the extent that they feel that their own children are at risk. The latter is a better sales pitch to the broad middle to the extent that other children are at risk. If lead exposure is perceived as a widespread problem that could impact middle class kids then middle class parents will support action to the extent that they fear that their own precious kiddos will not get into good colleges as a result. If lead exposure is perceived as a localized problem that mostly impacts lower class kids then middle class parents will support action to the extent that they fear that their own precious kiddos will be victimized as a result.
But I suspect that middle class parents can’t believe that their own children will be violence prone (after all, we raised ’em right) and don’t much care about negative IQ effects in someone else’s kids (we got enough problems of our own around here dammit). The IQ effect is a threat to the extent that it harms my own kids. The violence effect is a threat to the extent that it makes the streets less safe. In that nexus, I believe is the winning political strategy. But telling the average Joe that he needs to worry about his own kids turning into violent predators, or that he needs to worry about inner city kids having lower academic achievement, strikes me as a political loser.
I dunno. I guess we could just, like, not scream our heads off about black people? Maybe just focus on the general problem?
Crazy idea, I know.
Read a few of the other pieces by Nevin and he makes it pretty clear that it is a black problem. But if the messaging strategy will be to persuade people to put up $20 billion a year to aid the urban poor in blighted areas without mentioning the likely race of the beneficiaries, well, good luck with that.
Re: “…a better sales pitch to the broad middle to the extent that they feel that their own children are at risk. ”
Not to be tedious but people have been talking about the importance of avoiding lead for, well, centuries, but lately since the 80’s or 90’s. The link to crime is new and presumably not widely accepted; as a general health hazard, Drum is hardly breaking new ground here.
FWIW, a link to a 1988 NY Times article:
http://www.nytimes.com/1988/09/01/us/environmental-hazards-after-years-cleanup-lead-poisoning-persists-threat-health.html?pagewanted=all&src=pm
Back in the late 1990s I read an article that mini-blinds made in China gave off leaded dust. Our kid was a newborn then and we were living in a very nice rental townhouse in Kemp Mill, Maryland (just north of Silver Spring and DC) with yes, mini-blinds everywhere.
We got a lead test kit from the hardware store and the paper turned whatever color it did when lead was present. It took some effort but we finally got the landlord to replace all the blinds. And then I washed the windows, sills and the walls, and vacuummed the floors around the windows like a maniac.
On a slightly different note, nowadays one group in particular pediatricians keep their eye on for possible high-lead levels are the children of middle-class parents who are “urban pioneers,” you know, hipsters who move into old houses or tenements and rehab them. All that wood-stripping and sanding, wall-knocking-down and window-replacing makes for a lot of dust full of lead. Obviously, the poor people who are tenents in the building just next door are not doing big interior redesigns.
When the inner-city middle-class kids go to the doctor, their parents are told to make sure they get plenty of calcium (to help stop the lead from being absorbed), and they’re probably sent for a blood test, which isn’t part of the usual well-child prevention regimen.
My point is, middle-class kids aren’t exempt from the risk of lead exposure but for a variety of reasons, their parents are in a better position to protect them.
One of the things you learn on a submarine is to stop the flooding before attending to the water already taken aboard.
In this context, means focusing first on ongoing sources … The big one apparently being aviation gas, still leaded after all these years, and uniquely evil because the toxic waste is nicely dispersed widely and not bonded in a matrix that fixes it in place.
I just read a scientific American piece that suggests that two thirds of the small air craft using it could just as well use high octane unleaded.
Proposal
1) outlaw leaded avgas
2) do a “Cash for Clunkers” for civil aviation, buying up all the small aircraft engines that cant convert to uneaded. It would only cost a few hundred millions, tops.
3) sue all leaded motor fuel and avgas makers, dealers, and distributors under a new Superfund law for all dispersed lead cleanups.
4) profit from better health, and a boom in small engine sales.
Interesting point. Any data on how much avgas we are talking about? Sometimes it is better to pick a fight worth winning (and sometimes it is better to establish a principle…).
Hey Brett? How would you solve this problem?
Dispersing a clearly toxic chemical into the environment in amounts which have measurably bad effects is a classic violation of other peoples’ rights under libertarian theory, getting rid of lead in gas was the obvious thing to do, and I’m shocked it’s still used in aviation fuel.
But suing people for selling legal products people didn’t have to use? Why didn’t occur to JMG to sue the people who *used* the fuel? Or sue the government for exempting it?
I do find myself wondering if this is why my son is clearly brighter than I was at that age, or at least contributes to it.
Paint is a bit more complicated, in as much as it isn’t marching off your property and attacking other people. I’d concentrate on disclosure to home buyers, the problem can be taken care of in place, and will eventually disappear entirely as the housing stock turns over.
Dispersing a clearly toxic chemical into the environment in amounts which have measurably bad effects is a classic violation of other peoples’ rights under libertarian theory, getting rid of lead in gas was the obvious thing to do, and I’m shocked it’s still used in aviation fuel.
Doesn’t the same apply to dispersing gases that cause global warming?
Only if they are toxic.
But isn’t the ‘solve this problem’ part left un-addressed here?
The question is how to stop exposing American children to lead. Not about who should sue who. As a public nuisance, it hurts all of us by increments. Disclosure to home buyers does not help renters, accumulated lead that blows down the street, or lead pipes at the house your kid’s friend lives at. I don’t break out the assay kit when vetting possible sleep-overs.
Here’s a question: Drum talks about impulse control and notes that most of the people exposed didn’t become violent criminals — “Everyone over the age of 40 was probably exposed to too much lead during childhood, and most of us suffered nothing more than a few points of IQ loss.”
But. What has the political and economic history of the US for the past half-century been if not an orgy of poor impulse control and lack of ability to sacrifice present gratification for future avoidance of loss, with a side order of rhetorical violence that’s almost always uncalled-for. Change the lags to account for the fact that people reach their political and corporate peaks at later ages, and I wonder how much variance you could account for.
This hypothesis does offer the hope that people like Eric Cantor are the tail end of a cohort rather than whole new generation.
This is an interesting point. The effects of lead-based paint on a community are going to be marginal to other major factors involved in socioeconomics. Children exposed to lead paint, but otherwise with high levels of societal capital (educated, emotionally and cognitively healthy parents with stable incomes, likewise family and friends, safe neighborhood, etc., ) are going to vastly outperform those exposed to lead, but without the societal capital. Yet measuring things like impulse control - or, ADD? - might be possible if we look for it in other ways.
Structurally, neighborhood segregation by income is still going to compound the effects of societal capital in profound ways. Interventions designed around targeting societal capital deficiencies merely by waiting for the criminal justice and education system to deal with them are going to be incredibly expensive and marginally effective at best.
I don’t understand why Kleiman is so pessimistic. This is a problem that’s mostly been solved! We banned leaded gas and paint and did various things to mandate lead-abatement. A big win for liberalism and environmentalism!
If there’s still a bit more to do, history would seem to predict that we’ll do it. Even if you’re pessimistic about federal action, this is a problem that can mostly be solved at the state and local level. I’m no expert, but my understanding is that local governments have done a lot over the years to reduce lead exposure.
Can local governments ban the use of aviation fuels containing lead? Maybe, if they bar airplanes using such fuels from landing at local airports. I don’t know if they have the authority to do that. It still looks like it’s easier at the federal level, but maybe not.
Local governments probably can’t ban leaded aviation fuel, but they could clean up lead-contaminated soil.
State governments certainly could. California banned lead gas for cars before the EPA did. However, that may have been possible because of special rules for CA air quality problems.
Maybe not, though I wonder how much control they, or the states, have over airports, or the types of fuel that can be sold, or used by planes landing or takimg off. Is that all federally controlled?
Probably need to match the colors of your walls white and white, similar to your wall, you should be able to choose the colors of the curtains. Chicago treatments you do not like the dark tones Window Treatments Chicago
are canWindow choose a liner that matches the wall color. Alternatively