Wednesday, September 27, 2017

Debtor’s Prison and Performance Funding

In ninth grade, I remember hearing of debtor’s prison.  The idea immediately struck me as insane.  How do you pay off your debt if you can’t get to your job?  I wasn’t the only one to connect those dots; nearly everyone in the class did.  Yes, debtor’s prison was an incentive to pay up, but if you couldn’t, it functioned less as an incentive than as a death spiral.  The logic of incentives is powerful, but it only works to the extent that the desired behavior is actually under your control.  

As Slate noted in an uncommonly good article this week, we have a modern version of debtor’s prison.  It’s called the “suspended license.”  Outside of a few urban enclaves, if you can’t drive, you can’t get to work.  If you can’t get to work, you can’t pay your fines.  So people cheat and drive without licenses, at which point they incur more fines.  Or they don’t, they lose their jobs, and they can’t pay the fines they already have.  As the article notes, a disturbing percentage of suspended licenses are for behavior that isn’t dangerous.  It’s often parking tickets, or failing to show up for a hearing, or not paying a fine.  What sounds in the abstract like a slap on the wrist -- “driving is a privilege, not a right” -- ignores the material reality that in most places, if you can’t drive, you can’t work.  

It would be wonderful if we had enough, and good enough, mass transit to make a suspended license a mere nuisance.  But in most places, we don’t.  Bicycles can help sometimes, but they aren’t great in the rain, or in winter, or at night, or on the vast majority of roads that are built in ways that make bicycling dangerous.  For now, in most of the country, a car is a necessity.  That’s a problem, but it’s a different problem.  The article does a nice job of outlining the real-life consequences to some folks from having their licenses suspended; they seem wildly out of proportion to any ‘incentive’ value.

Which brings me to the latest peer-reviewed study on the effects of performance funding on public colleges.  The theory behind performance funding is that colleges respond to incentives, and have it in their power to improve dramatically their student outcomes; they just need to be prodded.  Kick ‘em in the budget, the theory goes, and they’ll find ways to improve.

Except that they mostly don’t.  In this case, I think the misunderstanding runs even deeper than debtor’s prison or suspended licenses.

In most cases that I’ve seen (and worked under), performance funding isn’t new money; it’s a reallocation of existing money or a little bit less.  In other words, the mandate for improvement is unfunded at the outset.  Given the high fixed costs and thin margins of most public colleges, especially in the two-year sector, it’s rare to find colleges making dramatic gains without significant external grants.  So there’s a bit of the debtor’s prison issue: if you can’t afford to improve, your funding will be cut, making you less able to afford to improve.  What was marketed as a prod becomes a death spiral.

The motivation is less direct, too.  The people who have to make the changes in what they do don’t get paid more.  If the college gets, say, a three percent increase in its operating subsidy, it’s immediately swallowed by that year’s health insurance increase.  The employees on the ground who made it happen don’t get any more than if they didn’t.  (There may be fewer layoffs, but that’s abstract until it isn’t.)  And public colleges aren’t for-profit; that’s not their motivation.  Yes, they have budgets to balance, but they don’t exist for their budgets.  

As Bailey, Davis, and Jaggars noted in Redesigning America’s Community Colleges, smart reforms can decrease the cost per graduate, but they increase the cost per student.  For colleges that are already barely above water financially, increasing the cost per student is a heavy lift.  In other words, the distance between the desired behavior and the ostensible reward is too great, and the reward feels too abstract to many employees to matter.  And that’s without even getting into the myriad of factors beyond a college’s quality that affect student completion.

The comparison to grants is revealing.  Grants can generate wonderful results because the means to achieve those results are supplied upfront.  You don’t have to raid other parts of the operation.  And colleges have some control over which grants they apply for, so they’re likely to pass on the ones they know wouldn’t make sense for them.  Grants spare you having to rob Peter to pay Paul, allowing you to devote more resources to improvement and fewer to infighting.  There’s value in that.  Significant new operating funding could achieve the same thing.

This argument is a difficult sell because it requires going beyond the first step.  Yes, improvement is possible.  Yes, incentives matter.  But so does investment.  When performance funding becomes effectively punitive, it reinvents a mistake hundreds of years old.  My ninth grade class could see the flaw in that.  I hope we still can.