Results First program guides states to smarter spending; Pa. doesn't use it
With Pennsylvania facing a $1.4 billion budget deficit, lawmakers are spending the weekend in Harrisburg squabbling over which programs to preserve or slash.
But they won't be taking advantage of a free tool that a growing number of legislatures use to identify the smartest cuts and investments.
Fourteen states and four California counties joined the Pew-MacArthur Results First Initiative since its debut in 2010. It equips resource-strapped state governments with research and a customizable tool for doing systematic reviews of programs and policies.
“Typically, governments have made decisions much more on inertia and anecdotes, rather than data,” said Gary VanLandingham, director of Results First. “Programs get established, and they go out to agencies or private providers to administer them, and then they tend to go on autopilot.”
Pennsylvania isn't part of the initiative.
Gov. Tom Corbett might consider looking into it, but he has directed state agencies to get leaner and more efficient and use data to gauge performance, Corbett spokesman Jay Pagni said.
Criminal justice, education and health care are among major cost-drivers in Pennsylvania that could benefit from a sophisticated review of long-term costs and benefits.
A 2012 Vera Institute of Justice and Pew Charitable Trusts report found that Pennsylvania spent around $460 million more on prisons than was stated in the $1.6 billion prison budget, when accounting for costs the state folded into other budget categories — such as fringe benefits for employees, underfunded pension and retiree health plans, and spending on inmate health care and education.
VanLandingham said officials in states that participate in Results First often are surprised at how their programs stack up.
In many cases, the data show that programs simply don't work, or make it clear that other research-based methods generate much more for taxpayers' money.
Iowa Corrections officials discovered in 2011 that their approach to curbing domestic violence not only failed to lower recidivism rates — nearly a quarter of men who went through the program got caught re-offending — but cost the state $3 for every $1 invested in the program. The findings spurred officials to dump the program for one that's getting early results: The state's domestic violence re-offender rate plummeted from 22.9 to 13.4 percent.
“We were really wasting taxpayer money,” Corrections Director John Baldwin said. “The new program is much better for the offenders who take it, and there's fewer victims.”
Sometimes state-funded programs do more harm than good.
Those once-popular “scared straight” programs — in which children interact with prison inmates — proved to increase the likelihood that a child would commit a crime by 1 percent to 30 percent.
New Mexico used the Results First tool to scrap a substance abuse treatment program when discovering participants had worse outcomes than peers who didn't. The model showed New Mexico could save $8.3 million by lowering recidivism rates by 10 percent — data that persuaded lawmakers to set aside money for adult education and other programs proven to yield up to a $25 return for every $1 spent.
“We're basically monetizing the cost of crime in New Mexico, which cuts across multiple agencies and programs,” said Charles Sallee, deputy director of an independent research arm called the Legislative Finance Committee.
In theory, cost-benefit analysis can help bridge the chasm of partisanship by presenting lawmakers with facts, said J. Wesley Leckrone, a political science professor at Widener University in Chester. The trouble is, lawmakers often are interested only in facts that reinforce what they want to do, Leckrone said. And every program has advocates who will fight to preserve status quo.
“Doing the numbers is only half the story,” said Steve Aos, director of the Washington State Institute for Public Policy, whose work inspired Results First. “The other half is the institutional apparatus, and the motivation on the part of a number of people in state capitols.”
A successful Results First-type effort requires support from executive and legislative branches, and technical work should be carried out by an independent body “viewed as an honest broker,” VanLandingham said.
In July, Results First will release a one-stop online database enabling those even in states that aren't formal participants to peruse research available in key policy areas.
“Scarcity and crisis create ingredients for actually changing things,” said Robert Strauss, economics and public policy professor at Carnegie Mellon University's Heinz College, “but if the Legislature can't agree on things like getting a budget done on time or keeping the credit agencies happy by not going over fiscal cliffs, it might be wishful thinking.”
Natasha Lindstrom is a Trib Total Media staff writer. Reach her at 412-380-8514 or firstname.lastname@example.org.