“John Arnold” isn’t a household name, even though he’s made a ton of money as “a wunderkind natural-gas trader at Enron who later founded his own hedge fund.” Enron? Well, even Enron made some real money in its day, and Arnold was their killer trader. With more than he could ever spend, John and Laura Arnold have moved on to doing something worthwhile with their money.
From the Wall Street Journal :
Most billionaires tend to write checks to good causes they’re part of, hospitals where they were treated or universities they attended. These are the so-called “grateful-recipient” donors. Or there are donors who make sizable gifts to meet an obvious need in a community, such as hunger or education. But at a time when charitable giving in the U.S. is still down from its peak in 2007, the Arnolds want to try something new and somewhat grander. John says the goal is to make “transformational” changes to society.The Arnolds want to see if they can use their money to solve some of the country’s biggest problems through data analysis and science, with an unsentimental focus on results and an aversion to feel-good projects—the success of which can’t be quantified.
What the idea lacks in immediate gratification it makes up for in scope and breadth. Intransigent systemic problems tend to be ignored because the cost/benefit ratio sucks. Even if a fortune is dedicated to fixing such a problem, there is no assurance that any benefit will ever be derived. Sometimes, intransigent problems are just, well, intransigent. But then, if you never try, you never know.
This is where the Arnold’s effort gets curious.
Along with obesity, the Arnolds plan to dig into criminal justice and pension reform, among others. Anne Milgram, the former New Jersey attorney general hired to tackle the criminal-justice issue, has a name for all this: She calls it the “Moneyball” approach to giving, a reference to the book and movie about how the Oakland A’s used smart statistical analysis to upend some of baseball’s conventional wisdom. And the Arnolds are in no hurry for answers. Indeed, they believe patience is a key resource behind their giving.
If the goal is to “tackle the criminal-justice issue,” what message comes from the choice of former New Jersey attorney general Anne Milgram to lead the effort? Her relatively brief career in the law has ranged from New York prosecutor to New Jersey prosecutor.
Then again, this only matters relative to the gist of the criminal justice reform. If the question is how can we convict more people and send them to prison longer, it’s one issue. If the question is how can we prevent the conviction of the innocent it’s another. The issue under scrutiny here is a bit different:
The Arnolds hired Milgram to analyze criminal justice, with the mandate to “take resources off the table for a while”—in other words, don’t worry about what the research will cost. She zeroed in on how, despite New York City’s success with the CompStat crime-reduction system, the influence of empirical data has barely trickled down to the local level. “It’s hard to think of an area that is less data-driven and analytical than local government,” Milgram says.
Huh?
In particular, she and Laura became fixated on how the country spends $9 billion to keep nonviolent, pretrial defendants behind bars, even though there is little data on the risks those accused pose to society. As Milgram points out, most baseball teams know more about their backup shortstops than judges know about pretrial defendants before locking them up at great cost to society and the accused.
Oooh. Didn’t see that coming. As for the “Moneyball” analogy to shortstops, well, it’s pretty darned poor. There are a whole lot more criminal defendants than major league quality shortstops, and not every defendant has a relevant statistical record to rely on. “So what team did you commit burglaries for in college, Joe?”
“It struck us as a uniquely bad mechanism to decide who should stay in jail,” says Laura, who also sits on the board of The Innocence Project, which uses DNA testing to clear prisoners on death row. Using data from more than 1.5 million cases, Milgram and her team created a risk-assessment tool for judges that will be tested in three jurisdictions later this year. The Manhattan district attorney’s office is planning to try a similar tool designed for prosecutors.
[Edit: This was omitted from the original post this morning. Sorry.] Risk assessment in pretrial bail/detention is relatively easy at the low ends and relatively pointless at the upper ends of the crime spectrum. At the low end, where poor defendants can’t make $500 bail, they sit until they plead to avoid spending the year awaiting trial. At the upper end, solid citizens who have no reason in the world to flee but for the seriousness of the charge, are held without bail lest the judge be impugned for letting some accused murderer out on the streets. It really doesn’t require a billion dollar study to empirically demonstrate that non-violent defendants pose no risk of violence if released on or without bail.
But if this risk assessment makes any sense, it would seem to have far greater application to sentencing, a far larger systemic problem, than pretrial detention. After all, at arraignment, everyone is presumed guilty, even though bail determinations don’t necessarily reflect it. After conviction, everyone is guilty. What then is the risk? What then is the sentence? That’s where some empiricism could do us a world of good.
Sentencing, as discussed here numerous times, remains a black hole of empiricism, the last bastion of religion in the law. It can only be explained in broad, meaningless strokes, without any hint of justification for the length of time a defendant is taken from his family, or society is made to pay for his warehousing. It’s voodoo.
And yet, one can’t help but wonder what bases predicated the 1.5 million cases of data? On the one hand, Laura Arnold sits on the board of The Innocence Project, which suggests a concern for the wrongful conviction of the innocent, and in turn the use of weapons in the fight for convictions that lend themselves to the securing convictions at all costs.
On other hand, former New York County prosecutor Milgram has been chosen to take the lead. So Manhattan District Attorney Cy Vance is planning to try using a similar empirical tool? Is that offered to suggest that if it’s good with prosecutors, it should be good with the rest of us? Are prosecutors the bellwether for a healthy criminal justice system?
The outcome of this project may well prove to be beneficial to everyone, providing an empirical basis and maybe sentencing that will end the imposition of meaningless numbers that leave people who pose no risk of recidivism or harm to society behind bars for years, for decades. But every effort at reform begins with a bias, and there is nothing in this article to suggest otherwise here.
If anything, it’s particularly scary as it has the potential to offer what appears to be an empirical basis for what’s not done with eyes closed which may prove to be impossible to argue against in the future (unless you have a few billion to do a study of your own), and yet infected with the same assumptions about the bad guys that have given rise to such brilliant solutions as mandatory minimums, three-strikes laws, zero tolerance and life without parole for children caught with their shirt tails out.
So will money end recidivism? I dunno, but I do know that whatever it comes up with, there’s no one around willing to engage in a counter effort to dispute the results.
H/T Stephanie West Allen
As they say at Sammy’s, “It couldn’t hoit.”
I’m still amazed at the striking resemblance between you and the piano player. Like twins.
The probability of incarceration is inversely proportional to wealth. Why do they have to spend tons of money to find that out?
That’s one of the variables I wonder about.
We should all be wary of anything that will get called the Milgram Experiment. (No link, ’cause I know the rules.)
How could you possibly do a double-blind study on something like this?
Don’t be ridiculous, Frank. There is no shortage of double-blind people in the criminal justice system.
“It couldn’t hoit.” Indeed. A few notes on your essay:
1) 1.5 million cases is enough to smooth out most biases.
2) The focus on pre-trial detention doesn’t seem to lead directly to your discussion of sentencing.
3) In the original WSJ article John Arnold acknowledges that “. . . many—if not most—of his foundation’s projects will fail or come to nothing.”
I’m not sure how much worse our justice system could get, but I see the direction it’s headed. (“Prison Nation: Assume Nothing”, Simple Justice 4/25/13) Maybe statistical analysis by smart people, influential but non-stakeholders, can give us the courage to overcome some of our biases. It’s been done before.
Arnold’s acknowledgement that his efforts may produce nothing is one of the most important virtues of his endeavors. It’s the need to get results that does the most harm.
The original Milgram experiments were highly useful (if controversial) in discovering a rather unpleasant aspect of human society. And if nothing else, whatever might come from this effort, it’s likely to create similar projects to refute whatever is presented. Properly done, this could prove to be highly illuminating (and most likely very unsettling).
Sure. Stanley Milgram was a serious guy. And his work was important (though I don’t know how his subjects ultimately felt about their participation).
Still, one is wary.
I have many physical defects, but I lack a comb- over and spare tire.
Well, of course. How else would you come in second in the Sexiest Guy in the Harris County Public Defender’s Office, after Frank Bynum?