Tuesday, October 21st, 2014
Half-True
Kilmartin
For every $1 spent on afterschool programs, "we can save over $5, almost $6, in crime costs down the line."

Peter Kilmartin on Monday, June 28th, 2010 in a speech at the state Democratic convention

Kilmartin claims $1 spent on afterschool programs saves more than $5 in crime costs

During a speech seeking the endorsement of the state Democratic Party in his race for attorney general, lawyer and former police officer Rep. Peter Kilmartin promised to work hard to prevent crime as well as prosecute it.

"It's about the attorney general, when I'm attorney general, advocating and increasing afterschool programming for children," he said. "For every dollar we can invest and parlay from private charitable foundations and the federal government, we can save over $5, almost $6, in crime costs down the line."

Sounds like a good investment. But claims based on the so-called multiplier effect always make us suspicious. While they often have the ring of authority, they are sometimes based on a pile of assumptions. We wondered where Kilmartin got his numbers.

His spokesman, Brett Broesder, quickly produced a copy of "A Report to the Legislature: Rhode Island Afterschool and Summer Learning Program Act" released in May by the Rhode Island Department of Elementary and Secondary Education.

The report, which argues for financing afterschool and summer programs for at-risk children and youth, says in two places: "For every one dollar invested in expanded learning opportunities, there is a savings of up to $5.29 in decreased criminal justice costs."

The RIDE report said the savings come in "criminal justice costs," which we took to mean the costs incurred by the police, courts and corrections departments. "Crime costs," the term used by Kilmartin, would also include the broader costs to the victims and their families, both tangible and intangible.

When we looked into the RIDE number, our suspicions were confirmed.

The RIDE report says its information came from a 2001 analysis from the Washington State Institute for Public Policy, a 174-page examination that tries to assess how much various social programs cost and how much they would save the state of Washington over the long term.

It's clear from Page 15 that the $5.29 figure comes with some significant caveats, none of which were mentioned by RIDE in its pitch for more money:

* The Washington analysis is based on just two studies of mentoring programs conducted outside the criminal justice system, one run by Big Brothers Big Sisters of America. The Big Brothers study never documented a drop in the crime rate as a result of mentoring; instead it asked participants for the number of times they hit someone. From that, the Washington researchers estimated that the two mentoring programs reduced the crime rate among participants by 4 percent.

* The Washington analysis estimated that there was no direct benefit to the taxpayers from the mentoring programs when they looked at how much would be saved in decreased criminal justice costs. "Overall, taxpayers just break-even for this investment," the authors concluded.

In other words, even though the RIDE report used the qualifying statement "up to $5.29," which technically covers every dollar estimate down to zero, it strikes us that RIDE misstated the findings. Twice.

So where does the $5.29 figure come from?

The Washington authors factored in their estimates of the costs of compensating victims of a crime, including the cost of lost quality of life. (Sample costs that they used: $8,734 for robbery victims and $3,137,793 for murder victims.)

The authors then plugged in several other numbers -- including the $1,054 cost of mentoring one person and their estimate of the cost of the crimes that mentoring prevented -- to arrive at $5.29.

Fortunately for Kilmartin, his mischaracterization of RIDE's conclusion (saying a $1 investment reduces "crime costs" instead of "decreasing criminal justice costs") makes his characterization of the Washington analysis more accurate than RIDE's.

Steve Aos, director of the Washington State Institute for Public Policy, said citing the 2001 analysis in 2010 -- instead of a later update -- would be like investing in the stock market today using investment advice given in 2001.
 
The most recent analysis, from 2004, estimates that society saves $3.28 for every dollar spent, not $5.29. If a mentoring program uses paid workers instead of volunteers, according to Aos, "the program breaks even."

We contacted RIDE spokesman Elliot Krieger to ask why the department cited an outdated study and failed to alert the legislature to some significant caveats. The department responded that "we are not persuaded" that the 2004 study refutes or negates the 2001 analysis.

"There are obviously many studies examining the value of afterschool programs, using many different methodologies, assumptions, and definitions. Experts in the field are free to debate the validity of the various methodologies. RIDE cited several studies in the report on afterschool and summer-learning programs. All were cited accurately and properly footnoted to enable readers to check the sources."

Well, the one cited source we checked -- because it impressed Kilmartin -- turned out to be a study that RIDE misrepresented.

Kilmartin was quoting a report he thought he could trust. So we'll give him a Half True.