MOOC platforms are the new startups.  Literally. We are closing in on a half billion dollars pouring into online education companies like Coursera, Udacity, and edX. Tens of millions of dollars are flying out the door of places like MIT, Stanford and U Penn to produce new instructional materials.

Nobody really knows how it will all turn out, but  these are experiments that need to be given time, space, and dollars to to incubate innovation.  But what exactly does that mean?  And what models are available to institutions that want to try to create such safe spaces for innovation?

When the residents of Al Capp’s mystical, mythical Uncertain Hamlet of Dogpatch needed to brew up a batch of Kickapoo Joy Juice, they did on the edge of town.  It was a smelly, messy process and the factory was better suited to the environs of Skunk Hollow — “worse than the badlands” — than the otherwise proper City of Dogpatch.  By the way, being worse than the badlands is a real black mark because in the Badlands “it’s no good here.”

The vision of Skonkworks, perched on the edge of Skunk Hollow, belching the byproducts of producing exquisite joy juice has been a metaphor for internal innovation ever since.

The approach at Georgia Tech has been to create an internal laboratory to try things out. Other places have tasked educational technology groups, CETLs, or distance education departments.  They are all  Skunkworks.  Just don’t call them internal startups. That is a sure path to failure.

When it comes to skunkworks, there are ideas to try out and ideas to avoid.  A colleague of mine once started a discussion by saying, “Let’s begin by figuring out what the administration will allow us to do.”  What a terrible idea — a rookie error. It defines your design space by all kinds of parameters that have nothing to do with success.

But it is easy to fall into this kind of trap.  There are plenty of examples of “internal startups” that failed in exactly this way.  It’s not just that online courses at elite campuses are brewing their own brand of joy juice.  New developments like MOOCs exist to bend perceptions and blur boundaries, so using traditional perceptions and boundaries to explore MOOC potential doesn’t make a lot of sense.

I want to repost a series of articles I wrote last year about this topic.   I think the lessons apply to academia.

Next:  The Internal Startup

OK, so it’s one thing to know what forces are driving tuition increases.  It’s another thing to know where money is being spent.   That is a problem of different proportions because universities do not report their spending in neat categories like

  • cost of scaling beyond capacity
  • making up for lost subsidies
  • cross-subsidies with loss leaders

And, as The Delta Project’s Jane Wellman is  fond of pointing out,  there is no category representing value that ends up on a student’s diploma.

It’s not a completely satisfactory approach, but you can look that the “Where does the money go?” in question in two ways.

The Accounting View

As I pointed out in a previous post, overall spending has not substantially increased — even as tuition has risen.  What has increased is spending in the following categories:

  • institutional grant aid
  • public service and research
  • energy
  • facilities
  • contingent faculty
  • residence hall operations
  • groundskeeping
  • online
  • admin
  • bookstore
  • general staffing
  • financial services

These are interesting categories because increases in these areas generally compensates for decreased spending in others.  Take contingent faculty, for example. Increased spending for part-time or adjunct professors compensates for an overall reduction in tenure-track faculty. Contingent faculty carry  a completely different cost platform.  In many cases health care costs can be trimmed by reducing the number of full-time faculty. adjunct professors do not require expensive startup packages. Capital requirements also change, because office and lab demands are not the same.

This brings us to a second way of tracking dollars: what behaviors result in large shifts in spending priorities.

The Behavioral View

Behaviors are interesting because you can imagine controlling them.  Here are the behaviors that institutions report as affecting their current spending priorities.

  • Making up for lost revenue
  •  Lost productivity
  •  Capital expenditures
  •  Cost of non-core activities
  •  Administrative bloat
  •  Giving it away
  •  Compliance
  •  Health Care and other costs of an aging workforce
  •  Unquantified “quality”
  • Overshooting markets like internationalization
  • Stealing revenue  from academic programs
  • Operational inefficiency
  • High cost of materials
  • inappropriate skills utilization

There are others of course.  Stanford president John Hennessey has recently pointed to labor costs tied to the professorate drive cost increases, but I can’t find anyone who will show me a price increase that is due to increased faculty costs.

I can find examples of capital projects where initial funding plans have collapsed. Few of these projects are drop-dead institutional requirements.  There are a fair number of vanity projects and many examples of non-core activities like athletic practice facilities or mixed-use facilities that were once thought be revenue-producing opportunities but for which a real market never materialized.  When funding profiles change dramatically, one response might be to re-evaluate the need for a new building, but that rarely happens.

This is a true money pit.  There are really only three sources of discretionary revenue: tuition, government allocations, and private gifts. One way to make up for lost revenue is to increase income from other sources.  In most cases, this means tuition.

Critics seem hell-bent to fabricate exotic reasons that college tuition is rising. “It is a market response to free-flowing federal dollars”, say some.  “It is a conspiracy,” say others. “Declining productivity!” say those who are convinced that college professors are overpaid and underworked.

The 2011 annual spending report of the Delta Project on Postsecondary Education Costs, Productivity, and Accountability — the last annual report issued by Delta before it sadly closed its doors last spring — makes it clear that easy answers are likely to be wrong:

There is much public discussion and concern over rising tuition but much less attention to the intricate relationships among tuition and revenue sources–particularly state and local appropriations in the public sector — and spending.

Take the issue of lost appropriations, for example. It is true that from 1999 to 2009 tuition at public research universities rose at a much higher rate (56%) than the 32% increase at their private counterparts.  These increases are almost entirely explained by the loss of state appropriates, as opposed, say, to increased spending.

All other factors are incidental. Nor does the rate of increase tell the whole story, because private universities are starting from a much higher base for their sticker prices the average dollar increase at the private institutions was three times that of the private universities.  The situation is much worse at public community colleges which have seen the greatest increase in enrollment. This is because, when it comes to tuition, even small changes in enrollment headcount have a big effect on prices. And that is where capacity comes into play.

Private universities have been able to keep their increases in check because they do not accept as many students as they could otherwise afford.  The increasing enrollment pressures have been absorbed by community colleges, and public masters and research universities. Students, who cannot afford to attend private institutions flock to quality public universities.  These are the very institutions that have not added capacity over the last forty years.

It is seductive but entirely false to think of higher education as a single sided market — that is, a system in which there is a supplier of goods and services to a customer who is willing to pay a price determined by the marginal cost of production.  That is not higher education.  Higher education is a multi-sided market, a collection of stakeholders with often competing interests and cross subsidies that make it difficult to determine fair pricing schemes.

Local newspapers were a multi- sided market,  in which extremely profitable classified advertising subsidized news, subscriber fees, and print advertising.  It was a system that worked well until Craig’s List came along an undermined the whole value proposition that supported local news.  The result was rapidly dropping classified ad revenue, rapidly increasing print ad rates followed by a corresponding drop in demand for print ad, and ultimately a cash crisis in which subscribers fled to other media for their local news.  Newspapers were slow to recognize that they were a three-sided market. 

Higher education is taking the same path, and that is significant because it helps explain why the lack of capacity in American universities lies at the root of cost increases.

Success in a multi-sided market is determined by platform success in which a single horizontal layer allows many stakeholders to share common services at a total cost that is less than the cost of offering those same services in a vertically integrated business.

A multi-sided business without a scalable platform eventually succumbs to the nonlinear costs of growth.  Not only do the various services have to work out pairwise agreements with the other services, but individual stakeholders must also do the same.  As the number of stakeholders grow the overall cost of providing services grows in proportion to the ad hoc arrangements needed to keep everyone happy.

Universities have not invested in platform scalability for at least the last fifty years. Their processes and structures are locked into a set of capacity assumptions that are falling by the wayside. And they are rewarded for growing in a market where students seeking value are fleeing lesser institutions for those that are already overcrowded.

The obvious losers are the second and third tier institutions who desperately need students.  These are the schools that offer the steepest discounts. So steep that they cannot sustain them without falling into the financial danger zone I have talked about before.

The other losers are the students who are driven to higher quality by increasing prices.  Which institutions are they?  They are the very universities which have not added capacity but which have incentives to continue growing.

It is growth that comes at nonlinear costs.  There are no easy ways for most institutions to subsidize or cross-subsidize those costs, so they are passed on to students.

Next:  Where does the money go?

What is Driving Up the Cost of College?“, asked the question that many outside and inside academia want to know the answer to — especially this time of year as families contemplate writing college tuition checks that over four years will top $200,000   According to a recent Pew poll, the majority of Americans believe that a college education is becoming unaffordable and want to know why. Facile explanations are easy to come by, but there are few that match the facts.

So what’s going on?  Much of the popular discussion of tuition increases is based on three  “facts” that  are not true.  They may be easy to toss off on editorial pages or cable talk shows, but they are myths.

Myth 1: The rate of college tuition increases is abnormally high.

If you listen to many commentators, college tuition is hyper-inflationary.  In fact, students at many public campuses have seen tuition and fees double over the last decade, and–as the stubbornly persistent effects of the recession strain family budgets–become increasingly unaffordable.

Affordability is a real problem and prices are on an unsustainable path, but it is not the case that we are in the middle of some hyper-inflationary bubble.

In normal times, tuition rises at roughly twice the rate of inflation.  From 1958-1996 that averaged somewhere near 8%.  For most of those years, the increases were offset by increased spending for financial aid and by general growth in personal income, which hid many of the most serious consequences of this rise in prices.

Recent years are different, some argue. Let’s take a look at that claim. Annual inflation for 2011 was 3.16%, but that actually understates inflation for the months in which trustees and legislative committees approve tuition increases. The  annual rate of inflation for the last three quarters of 2011 was just under 3.5%.

Private institutions  raised their prices last year, but the rate of increase was actually in line with inflation.   2011-2012 tuition increases at nonprofit private institutions averaged around 4%.  For-profits increased their prices  3.6%. When it comes to holding the line on prices, these institutions actually performed much better than historical trends.

Public universities raised their prices at the much more dramatic rate of 8.6%, 20% above historical highs of the decade 1992-2001.

In short, the only increases that are above historical highs are at public universities, but only by a relatively modest amount.

Myth 2:  Easy availability of Federally-backed student loans is driving tuition increases.

This is a myth with many political consequences, but there is very little evidence that it is true.

The theory behind this myth is that a free market will cause prices to rise in proportion to a market’s ability to pay those prices, and the widespread availability of federal loans affects a student’s ability to pay.  The often unstated political corollary to this theory is that the students who are least able to pay are also the ones who are least likely to complete their degrees and pay back the loans.

In reality, the part of the higher education market that has control over its prices has been demonstrably immune from any such affect.

Less than 40% of all Federal aid is in the form of student loans (the rest is used for Pell grants and merit-based support).  These loans are distributed to public, non-profit and for-profit institutions, but for-profits tend to get a disproportionately large percentage of these loans.   These are the institutions that have raised their fees the least.

The remaining loans are distributed to nonprofit and public institutions, but since non-profits have also had historically low tuition increases, all of the factors affecting tuition would have to be concentrated in the sector that has the least flexibility in setting its own prices — the public colleges and universities. And as we have seen prices when viewed as a multiple of consumer price increases have risen only 20% faster than historical averages at these institutions.

Public institutions cannot retain earnings, so costs would have to have risen in proportion to loan availability, but that has not been the case.  Costs at public universities have been constant for several years.

So why have prices increased? In fact, total state support for public institutions dropped 7.5% in 2011-12. The total increase in tuition can be attributed to making up for lost income. Federal loan availability may be correlated with tuition increases but there is no reason for believing it is a cause.

Myth 3: Tuition rises because of declining productivity.  Professors are earning more and teaching less.

Productivity is not easy to measure in academia, but one thing is certain: unproductive professors are not a contributing factor to rising tuition.

One popular measure is the average ratio of students to faculty in general education courses.  This number has risen significantly at public universities whose state budgets have been squeezed.  In many industries this would be the very definition of increased productivity, which would mean that rich professors are skimming the cream off the top of rising revenues.

That is not true.  Faculty salaries have been stagnant at public colleges for almost a decade.  Many large state institutions have resorted to hiring part-time or “contingent” faculty to satisfy increased demand, which further depresses average annual compensation.

Another  measure of productivity is the number of courses taught per semester.  This number has been on a slight decline for over a decade, which might indicate that prices are going up even as productivity is decreasing.

However, the decline in average teaching load has been most noticeable at schools that are building sponsored research programs.  As I have pointed out in prior posts, sponsored research seldom pays for itself, so a decline in teaching loads represents, not  a productivity decrease, but rather a shifting of costs from academic programs to another income producing activity but one which often fails to cover even its own costs.

Next:  Three Reasons that Tuition is Rising: Capacity, Capacity Capacity