For a more efficient front line

New Zealand's headed for a pretty austere budget; the government's looking for front-line efficiencies:

The Government's "general strategy" remained trying to move money on to front line services, but English signalled pressure would go on for savings there too.

"Where the front line is inefficient, of course, it will have to be sorted out," he said. 

"We're not offering any guarantees ... Just by way of example, if you can give the police mobile technology, then you could get more front line hours with the same number of police - so it's more policing, safer public and we cut costs."

If we're really looking for front-line police efficiency, allow police to stop worrying about marijuana and instead focus on violent crime. Reallocate resources currently devoted to big grow op busts and put them into areas where social costs are real and substantial.

The Law Commission reported that cannabis enforcement cost the police about $116 million in 2005/06; it would be surprising if current figures were lower. Back of the envelope, but if that were shifted over into violent crime, we might expect some pretty serious savings. Treasury estimated a 1% decrease in violent crime saves $25 million in social costs; if the elasticity of crime with respect to police expenditures follows American estimates, which I'd take as a lower bound given that they're further out on a likely concave production function, we'd reduce violent crime by about 11% and social costs of violent crime by about $250 million.

Even better, legalize marijuana but regulate it similarly to alcohol and tobacco with restrictions on age of purchase and on permissible sales venues. Set an excise tax to keep the selling price to consumers about where it is now if demand-reduction is government policy. It's not simple to figure out what that tax should be because it's not obvious what production efficiencies could be achieved in a legalized production environment. But back-of-the-envelope numbers suggest excise revenues a bit over $150m (with a very wide confidence interval; I'd be very surprised if it came in at less than $100 million or at more than $300 million). If production costs in a legal environment were $30-$50 per ounce, and if we assume distribution costs roughly on par with production costs, then a legal market would deliver product to consumers at about $100 per ounce; if weed currently sells at about $300 per ounce, then excise and GST* could total about $200 per ounce. A 2001 estimate put the black market in marijuana at around $190 million in 2001; if marijuana follows the CPI, that's about 830,000 ounces. At $200/oz in tax, that's about $165m.

Recall that drug use in Portugal has gone down, not up, since they abandoned prohibition.** In that case, tax revenues are less than I'm here ballparking.

* Normally we'd ignore the GST revenues collected as they'd just be a shift in spending from other GST-subject goods. But moving to a legal market would bring in net new GST revenues. I discussed some related issues here.

** They have not gone to full legalization; read the Cato summary.

Confounds, alcohol and violence

It's plausible that increased alcohol consumption in a neighbourhood directly affects assault rates. But I don't think you can tell it from this study (HT: Bakadesuyo).

Methods and Findings

We performed a population-based case-crossover analysis of all persons aged 13 years and older hospitalized for assault in Ontario from 1 April 2002 to 1 December 2004. On the day prior to each assault case's hospitalization, the volume of alcohol sold at the store in closest proximity to the victim's home was compared to the volume of alcohol sold at the same store 7 d earlier. Conditional logistic regression analysis was used to determine the associated relative risk (RR) of assault per 1,000 l higher daily sales of alcohol. Of the 3,212 persons admitted to hospital for assault, nearly 25% were between the ages of 13 and 20 y, and 83% were male. A total of 1,150 assaults (36%) involved the use of a sharp or blunt weapon, and 1,532 (48%) arose during an unarmed brawl or fight. For every 1,000 l more of alcohol sold per store per day, the relative risk of being hospitalized for assault was 1.13 (95% confidence interval [CI] 1.02–1.26). The risk was accentuated for males (1.18, 95% CI 1.05–1.33), youth aged 13 to 20 y (1.21, 95% CI 0.99–1.46), and those in urban areas (1.19, 95% CI 1.06–1.35).

What's the problem? They don't seem to be controlling for day-level fixed effects or, even better, day-city fixed effects. Suppose there's a big hockey game on Saturday night that both brings a pile of folks onto the street and increases alcohol purchases. You can get a correlation between increased alcohol sales (relative to the week prior) and assaults entirely as artefact of the underlying variable driving both assaults and alcohol sales. A big hockey game, a holiday long weekend, even a big concert in town - none of those are addressed by comparing alcohol sales with those a week prior.

How do you fix this? Controlling for simultaneous alcohol sales in a similar part of town that's far enough away that it's unlikely to have had effects on the part of town in question would be a start, but might not catch localized effects of events that drive both alcohol sales and violence.

School quality and crime

David Deming argues in the Quarterly Journal of Economics (HT: @Crime_Economist) that high risk kids experienced lower likelihood of later criminal activity if they won random lotteries allowing them to attend their first choice public school.

How does it work? High-risk youths spread out over more schools rather than being concentrated in a few bad schools. If peer effects work with thresholds or if there are increasingly negative peer effects as concentrations of high-risk kids increase, this makes a lot of sense.

Lessons for us in New Zealand?

• The lottery design in North Carolina resulted in higher-risk youths being more, not less, likely to flip into competitive schools. This cuts against arguments that school choice and charters necessarily result in pools of worse kids left behind. 
• The gains from moving to charters here might be rather more extensive than just educational outcomes. This is also consistent with evidence from things like the Perry Preschool project where educational outcomes washed out over time but social outcomes, like crime reduction, persisted.

Here's an ungated working paper version.

Certainty and Severity

We usually expect crime to be more responsive to the certainty of punishment than to its severity in terms of fine or sentence imposed. The best explanation I've seen is that having any criminal record comes with a very large fixed cost regardless of sentence imposed: your chances for professional employment diminish and social sanctions apply. Consequently, differences in delivered sentences have only minor effect on the likelihood of offending if the informal penalty is large relative to the formal penalty and is not particularly increasing in the judicial penalty assessed.

A Canadian Ministry of Justice report finds that differences in first offence drink drive sentencing do not predict an offender's likelihood of re-offending. A straight punishment severity story would have that a more harshly punished offender will be more nervous about reoffending than will someone let off with a light sentence; that wasn't found.

The CTV piece concludes this is due to the severity of sentencing having little effect; they then conclude that Conservative crackdowns on repeat drink-drive offenders will not affect the likelihood of reoffending. But I'd be a bit nervous about drawing that conclusion without seeing the internal report cited. A few things I'd worry about:

• Suppose a judge assesses costlier penalties on offenders who are more likely to re-offend and lighter ones on those less likely to re-offend. In the absence of any penalties, there would be strong differences in likelihood of re-offending; in an equal-penalty treatment, there would also be reasonably large heterogeneity in recidivism rates. The higher penalties assessed to those perceived to be high risk may bring their recidivism rates down to those exhibited by lower risk offenders. You need some instrument to sort out causality.
• Drink driving convictions in Canada come with a one-year driving suspension for first time offenders. The costs of this are massive relative to a $600 or $1000 fine. How much would you be willing to pay to avoid a one-year license suspension? $5k? More? We'd need to know a lot more about the monetized value of the variance in imposed first-time penalties to know whether it's washed out by the fixed cost of losing your licence.
• If the variation in penalties for first offenders is small relative to the increase in penalties assessed for recidivists, it's hard to draw conclusions from the former about the latter.

HT: @DanGardner

Sentencing floors

Dan Gardner is right that mandatory minimum sentencing often yields perverse outcomes.

Now look at that first mandatory minimum sentence again: It means that anyone who grows six marijuana plants with the intention of sharing even a single joint with a friend will be guilty of an offence punishable with a mandatory minimum sentence of six months in jail.

And remember the phrase "real property that belongs to a third party"? That's what a rented apartment is. Imagine a university student living in a rented apartment with her boyfriend, suggests University of Toronto criminologist Tony Doob. She grows a single marijuana plant. She rolls a joint for her and her boyfriend. And just like that she's a "trafficker" subject to a mandatory minimum sentence of nine months in jail.

Are these outcomes simple, clear, and predictable? Hardly. They're shocking as hell. But mandatory minimums have a nasty tendency to do that.

He argues mandatory minimum sentencing has no deterrent effect. I'm not familiar with that literature, but mandatory minimums in the form of three-strikes legislation have substantial deterrence effects. They just may not be worth the candle.

Gardner nails one other point though:

But mandatory minimums don't actually do away with discretion.

They merely transfer it from judges, by restricting their ability to choose the sentence, to prosecutors, who choose the charge. The system is still ambiguous, uncertain, and unpredictable. It's just ambiguous, uncertain, and unpredictable in a different way.

Discretion lets prosecutors extract plea bargains to lesser charges from risk-averse defendants facing ridiculous minimum sentencing if convicted.

Deterrence or encouragement

One reason it's tough to identify deterrence effects of the death penalty: the long and variable lags between sentence and execution can be under conditions more pleasant than those in the general prison population.

AS AN ORANGE COUNTY jury debated in 2009 whether the white supremacist Billy Joe Johnson should live or die for murdering a fellow gang member, he asked to be sent to death row. Not because he felt any sudden remorse for the five people he’d killed over the years—“I commit crimes when people piss me off,” he once explained, matter-of-factly—but because Johnson believed he’d have better living conditions, including liberal phone privileges, a bigger cell, and daily human interaction, at San Quentin’s death row than he would at Pelican Bay, one of the state’s toughest maximum-security prisons, where he was serving a 46-year-to-life sentence, primarily in solitary confinement.

He also knew that the odds were good that he might never be executed. Bogged down by constitutional challenges and appeals, California’s system takes an average of 20 years to move a prisoner from conviction to execution.

Experts on both sides of the death-penalty debate have long agreed that California’s system is the nation’s costliest and least efficient. This June, a landmark report by Paula M. Mitchell, a professor at Loyola Law School, and Arthur L. Alarcón, a senior judge on the Ninth Circuit Court of Appeals, unearthed new data that reveal just how bad the system is.

Their report showed that since the current death-penalty statute was enacted in 1978, taxpayers have spent more than $4 billion on only 13 executions, or roughly $308 million per execution. As of 2009, prosecuting death-penalty cases cost upwards of $184 million more each year than life-without-parole cases. Housing, health care, and legal representation for California’s current death-row population of 714—the largest in the country—account for $144 million in annual extra costs. If juries continue to send an average of 20 convicts to San Quentin’s death row each year, and executions continue at the present rate, by 2030 the ranks of the condemned will have swelled to more than 1,000, and California’s taxpayers will have spent $9 billion to execute a total of 23 inmates.

I'd want to see that as a per-case excess cost rather than as an aggregate. But it's hard to see how the system makes sense.

Drinking culture [updated]

Some sense from Kate Fox at the BBC:

The problem is that we Brits believe that alcohol has magical powers - that it causes us to shed our inhibitions and become aggressive, promiscuous, disorderly and even violent.
But we are wrong.

In high doses, alcohol impairs our reaction times, muscle control, co-ordination, short-term memory, perceptual field, cognitive abilities and ability to speak clearly. But it does not cause us selectively to break specific social rules. It does not cause us to say, "Oi, what you lookin' at?" and start punching each other. Nor does it cause us to say, "Hey babe, fancy a shag?" and start groping each other.

The effects of alcohol on behaviour are determined by cultural rules and norms, not by the chemical actions of ethanol.

There is enormous cross-cultural variation in the way people behave when they drink alcohol.

...the variation is clearly related to different cultural beliefs about alcohol, different expectations about the effects of alcohol, and different social rules about drunken comportment.

This basic fact has been proved time and again, not just in qualitative cross-cultural research, but also in carefully controlled scientific experiments - double-blind, placebos and all. To put it very simply, the experiments show that when people think they are drinking alcohol, they behave according to their cultural beliefs about the behavioural effects of alcohol.

...I would like to see a complete change of focus, with all alcohol-education and awareness campaigns designed specifically to challenge these beliefs - to get across the message that a) alcohol does not cause disinhibition (aggressive, sexual or otherwise) and that b) even when you are drunk, you are in control of and have total responsibility for your actions and behaviour.

Alcohol education will have achieved its ultimate goal not when young people in this country are afraid of alcohol and avoid it because it is toxic and dangerous, but when they are frankly just a little bit bored by it, when they don't need to be told not to binge-drink vodka shots, any more than they now need to be told not to swig down 15 double espressos in quick succession.

I'm surprised that the Beeb allowed talk about personal responsibility!

Read the whole thing...

HT: Anon.

Update: Harford notes the author has worked for the alcohol industry. Watch for one-sided scepticism, but it's also worth following up on the experiments noted. I'll see whether I can track down the cites.

Minimum penalties and prosecutorial power [Updated]

A reasonable critique of minimum sentencing laws, and consequently of three-strikes legislation: they let prosecutors force plea bargains for offences without such minima because accused parties fear erroneous conviction and high penalties. Here's the New York Times (HT: all over Twitter)

“We now have an incredible concentration of power in the hands of prosecutors,” said Richard E. Myers II, a former assistant United States attorney who is now an associate professor of law at the University of North Carolina. He said that so much influence now resides with prosecutors that “in the wrong hands, the criminal justice system can be held hostage.” One crucial, if unheralded, effect of this shift is now coming into sharper view, according to academics who study the issue. Growing prosecutorial power is a significant reason that the percentage of felony cases that go to trial has dropped sharply in many places. Plea bargains have been common for more than a century, but lately they have begun to put the trial system out of business in some courtrooms. By one count, fewer than one in 40 felony cases now make it to trial, according to data from nine states that have published such records since the 1970s, when the ratio was about one in 12. The decline has been even steeper in federal district courts.

I wonder whether this effect has contaminated the studies on California's Three Strikes law. Recall that we've seen a big decline in strikeable offences and a smaller increase in non-strikeable offences. The argument has been that some of the decline in strikeable offences has been due to criminals substituting into offences carrying lesser potential sentence. Some of that substitution could just be the exercise of prosecutorial discretion in coercing a guilty plea on a non-strikeable offence where the accused fears the risk of taking a potential strikeable charge to trial. The effect can't be that large though: if it were all that kind of prosecutorial substitution, we'd expect the increase in non-strikeable offences to be larger relative to the decline in strikeable offences. UPDATE: Iyengar's work uses arrest data rather than conviction, and Shepherd's is all on crime rates. So unlikely to be a problem. See comments below, thanks to Lemmus!

I don't know the extent to which plea bargain is used in New Zealand. But this will be something to watch for as our three-strikes legislation becomes binding on more accused.

Previously:

• Is three strikes working?
• Marginal deterrence for the incarcerated
• A good strike
• Against the death penalty
• In favour of flogging

Death penalty

I used to hold the position that the death penalty was wrong, despite that it likely deterred around eight murders per execution, on the basis that the State ought not have that right. I suppose if the likely deterrent effect were really large, I'd be pluralist enough for the utilitarian side to beat the libertarian side. But not at that deterrent rate.

Wolfers and Donohue had previously shown much of the empirics on the death penalty are fragile.

The latest from Manski and Pepper (HT: Chris Blattman) shows the results more fragile than I'd thought: you can pretty much choose your conclusion through appropriate choice of identifying assumptions.

...we study the identifying power of relatively weak assumptions restricting variation in treatment response across places and time.  The results are findings of partial identification that bound the deterrent effect of capital punishment.  By successively adding stronger identifying assumptions, we seek to make transparent how assumptions shape inference.  We perform empirical analysis using state-level data in the United States in 1975 and 1977.  Under the weakest restrictions, there is substantial ambiguity: we cannot rule out the possibility that having a death penalty statute substantially increases or decreases homicide.  This ambiguity is reduced when we impose stronger assumptions, but inferences are sensitive to the maintained restrictions.  Combining the data with some assumptions implies that the death penalty increases homicide, but other assumptions imply that the death penalty deters it.

And so I revise: the death penalty is wrong, and it also likely has little measurable deterrent effect. There may still be a deterrent effect; we just can't show one given available data.

Update: Chris Auld has a nice intuitive explanation of the paper's results.

Prison cost minimization

The usual argument against privatizing prisons, at least in the econ literature, is that incomplete contracts combined with profit maximization gives prisons too strong of incentives to cut costs by worsening conditions for prisoners.

And so I was more than a bit surprised to hear this on Radio NZ this morning:

Mt Eden prison operator Serco is accused of bribing inmates with double-size meals and LCD televisions in their cells, so they are less likely to cause trouble.

The Corrections Association, the main prison officer's union, says that in addition to larger meals, Serco serves dessert every night, which is unheard of in the State prison system.

Association president Beven Hanlon says these ''luxuries'' allow Serco, a private operator, to get by with a skeleton crew.

But he says guards are feeling vulnerable and are leaving on a daily basis.

In a statement, Serco says the LCD televisions are small and must be paid for by the inmates.

The company won't directly respond to the other claims, but says the quantities of food served and the number of officers employed are both appropriate.

Yup, you heard that right. The private prisons here are accused of being too cushy: they've found that actually treating inmates well makes it easier to guard them.

Morning Report had the Sensible Sentencing Trust representative bemoaning that the nicer prisons aren't punitive enough - that harsher prisons are needed to teach inmates a lesson. Well, what literature we have on that suggests rather the opposite: harsher prisons correlate with increased likelihood of recidivism.

And so New Zealand's private prisons may be doing well by doing good.

Counting Lewd's utility

It's currently legal in New Zealand to post online racy pictures of your ex. The Law Commission recommends changing that:

The commission also recommends the exemption for personal or domestic information should not apply if the collection or disclosure of the information would be highly offensive.

The change would deal with situations such as when a person posts naked photographs of their ex-partner online without consent.

Privacy Commissioner Marie Shroff welcomed the report. "It will put, for the first time, the careless, the predatory and the criminal on notice," she said. "There will be consequences if you misuse our information."

I'm not particularly opposed to the law change. But I'm near certain that the Law Commission won't have weighed the benefit provided to voyeurs of the current legal framework. David Friedman argues, persuasively, I think, that we have to count all the benefits in the utilitarian calculus lest we wind up assuming our conclusion:

If instead of treating all benefits to everyone equally we first sort people into the deserving and the undeserving, the just and the unjust, the criminals and the victims, we are simply assuming our conclusions. Benefits to bad people don't count, so rules against bad people are automatically efficient. We cannot deduce moral conclusions from economics if we start the economics by assuming the moral conclusion.

It wouldn't surprise me if the utilitarian calculus found the law change to be efficient. Presumably, if the taking of private photographs has benefits for the couples involved, willingness to have such photos taken would depend on the likelihood of future dissemination. Further, some couples that would do better to split up may stay together, inefficiently, because of the hostage problem posed by the photographs' existence. It's entirely plausible that the losses to voyeurs are smaller than these gains. But I rather doubt that anybody's sought to measure either side.

And here is a test

Starting today, a zero blood alcohol limit applies to drivers aged 20 and under.

We'd expect this to show up in the drinking stats as a drop in drinking among folks in that cohort.

Any change in the crime rate (not counting drink driving) for 16-20 year olds relative to 21-25 year olds, and relative to the prior relationship between the two, and barring other anti youth crime initiatives coming on-stream, could then be used as part of an estimate of the elasticity of crime with respect to alcohol use among youths. Get the percent change in alcohol consumption among the cohort and the percent change in crime committed by folks in that cohort.

Give it a couple years...