Sunday, February 18, 2018

Accountability for What?

About fifty years ago, the sociologist David Riesman -- famed for The Lonely Crowd -- published a compilation of essays he titled “Abundance for What?”  It’s hard for contemporary readers to imagine, but at the time, serious American scholars were focused on what they considered the dangers of affluence.  At the height of the postwar economic boom, they were concerned that the central organizing principle of the economy -- scarcity -- was losing its pull, and that the culture would fall victim to entropy if it were not held together by material need.  Many of his cohort took the subsequent seeming chaos of the counterculture years as evidence for their thesis: with material scarcity a thing of the past, all hell broke loose.

In retrospect, the postwar observers’ faith that the gains of affluence would be evenly shared comes off as naive, even cute.  They couldn’t see the assumptions on which they based their analyses.  In the last forty years, the gains from increased productivity have gone almost entirely to the very top, with scarcity becoming much more real for most people.  As seriously as their work was taken at the time, now it reads as a dispatch from a forgotten era.

I was reminded of Riesman’s cohort in reading Robert Kelchen’s new book, Higher Education Accountability.  Kelchen is a scholar of higher education at Seton Hall, and his book is both an overview and an argument.  As an overview, it provides a valuable and concise introduction to many of the accountability regimes to which higher education has been subjected.  As an argument, it holds that whatever the flaws of existing regimes, we’ll gradually get better at measurement, to the eventual good of all.

As with Riesman, I can see where he’s getting it, but the larger issues underneath it all make me wonder.  

Kelchen’s historical overview is clear and helpful.  He calls attention to a long-forgotten effort at a federal ranking of colleges by the Taft administration (!), and traces the evolution of the American systems of regional, national, and programmatic accreditation.  His account of the last ten years or so is particularly strong, with nuanced readings of the increased scrutiny on for-profits, the trials of the City College of San Franscisco, and the relationships among various accrediting agencies and the Federal government.  I would have preferred more attention to the “outcomes assessment” movement as applied to individual courses, but that’s a quibble.

The argument is trickier.  In outlining theories of accountability, Kelchen helpfully lays out the principal/agent distinction and calls attention to its various dangers, but glides over the fundamental conflict over who is who.  In the context of public higher education, who is the principal?  

That’s not an abstract question; it’s at the heart of most of the issues I deal with daily.  

The “shared governance” model on which most colleges are run are built on the assumption (or aspiration) that a college is a closed system.  It was built specifically to blunt the influence of funders, and to allow academic freedom and relative institutional autonomy.  Kelchen correctly notes that autonomy can cover a great many sins, but the idea was that the faculty delegated operational authority to administrators while maintaining curricular authority to itself.  In that model, the faculty are the principal.  But Kelchen’s approach starts with the assumption -- defensible, but unargued -- that the state is the principal.  To the extent that we take the state as the principal, then matters like shared governance have to be profoundly rethought.  If the state is the principal, then faculty preferences become far less important.   As I’ve argued elsewhere, what may look to policymakers like accountability may look to faculty like usurpation.  There’s an intelligent argument to be made to the effect that the state should be the principal, but it goes unmade here.  

Kelchen does note in passing that increased state accountability has come at the same time as decreased state funding, which makes the “principal/agent” frame that much harder to sustain.  If he who pays the piper calls the tune, then states should be getting more circumspect, rather than more directive.  

Clearly, something else is going on.

“Accountability” implies both a clear judge and a clear task.  Kelchen notes correctly that many existing accountability systems, such as performance-based funding, are subject to predictable pathologies.  Colleges can game systems, such as by structuring remedial curricula to ensure that any students who place into remedial courses are removed from the “first-time, full-time” cohort that determines headline graduation rates.  He cites Campbell’s law, noting that any single quantitative measure used as a proxy can take on a life of its own and lose its validity as a proxy.  (For example, if pass rates are taken as the sole measure of success, then grade inflation will look like real improvement.)  Data aren’t always verified, which allows for cheating, and even good-faith analysts can define the same term differently.  

That’s all correct, as far as it goes.  Kelchen is admirably thorough and careful in delineating the ways that measures can go awry.  Full credit there.

But even as we refine the data -- and I agree with Kelchen that the analytics are improving -- we’re still left with the questions of judge and task.  

I’ll provide an alternate reading.

Kelchen notes in passing that most federal aid to colleges, especially outside of the research university sector, comes in the form of voucher-ized financial aid.  But he leaves that observation hanging out there.  I think it’s key to the whole thing.

As public college budgets have moved from mostly-subsidy to mostly-tuition, non-elite higher education has shifted from a seller’s market to a buyer’s market.  In other words, enrollments drive decisions.  Any countervailing force that pushes back -- whether it be regional accreditors, faculty unions, state governments, or anyone else -- quickly learns that the power has shifted to the students.  

The students are the principal.  Not just morally, but financially.  They pay the bills.

As a sector, we weren’t built for that.

The for-profits figured that out first, and demonstrated the dangers of taking that logic as far as it would go.  But now we see much of traditional higher education adopting perspectives and tactics that for-profits pioneered.  The economic incentives are too powerful not to.

For institutions to maintain academic integrity and improve performance, they need some autonomy (there’s that word again) from the marketplace.  They need to be able to take a long-term perspective, meaning that they need the material resources to survive a short-term hit.  That’s the logic behind endowments, reserves, tenure, and trustees.  But a more market-based system discounts the long term much more severely than our systems were built to do.

That’s at the root of much of the suspicion around performance-based funding, risk-sharing, and the other accountability schemes gaining currency now.  They all discount the future, and raise the cost of short-term risk.  They’re about increasing the power of the market, and doing away with any buffers.

In the case of performance-based funding, for instance, what happens if your college is identified as underperforming?  It’s deprived of resources.  Apply that same logic to, say, fire departments.  If we respond to an outbreak of arson by cutting resources for fire departments, what do we think would happen?  What starts as an effort to prod can quickly become a death spiral (or, as Kelchen notes, would if not for political leaders intervening).  PBF schemes almost never involve significant new money; they’re zero-sum at best.  That means they rely on creating death spirals to work.  To its supporters, that’s not a bug of PBF; it’s a feature.

That’s why I find Kelchen’s faith that internally generated measures will head off more interference unconvincing.  We’ve been generating measures internally for the last twenty years, yet the demands keep accelerating, even as funds flatline or drop.  The strongest supporters of PBF are also the strongest supporters of the student-as-consumer model.  We aren’t going to assess our way out of this.

At its base, the issue is political.  Graduates’ earnings are much more a function of the economy than they are of the English department.  And decisions about how to pay for higher education -- whether by shifting ever more of the burden to students, or by recognizing the need for actual operating funding -- are inherently political. Just like the decisions made over the last forty years to let all of the gains of productivity accrue to a lucky few.

Shortly after writing about endless abundance, Riesman co-wrote a classic book on higher education, “The Academic Revolution.”  It was about the emergence of a self-contained world of higher education, relying on the abundance of the time.  That world is gone, replaced by one in which we’ve adopted the market as a judge of all things.  The self-contained world of higher education is struggling in the more hostile setting, and has been for some time.  It can change, and I share Kelchen’s hope that it will find intelligent ways to do that.  But let’s not lose sight of the setting itself.  He who pays the piper calls the tune.  If we don’t like the tune, we know what we have to do.