A couple years ago, Rutgers historian David Greenberg noticed a defect endemic to books about social, political and economic problems: The last chapter always sucks. "Practically every example of that genre, no matter how shrewd or rich its survey of the question at hand, finishes with an obligatory prescription that is utopian, banal, unhelpful or out of tune with the rest of the book," Greenberg noted.

And it's not just books. I'll be the first to admit that the possible fixes with which I finished off my series on the alarming rise in college tuition were pretty vague and utopian. But helpfully, the good folks at Third Way have noticed that the conversation about how to reign in tuition has gotten a little too small-minded. "For both parties, in particular Democrats, our solution to the problem of rising cost of college has been to subsidize the rising cost,"  the think tank's president, Jonathan Cowan, says. "That's been our official policy, to subsidize the rising cost, and that has to be seen as a fairly intellectually bankrupt approach. We need a dramatically different approach that is about driving down the rising price."

To that end, Third Way is publishing a new report by Anya Kamenetz, one of the most interesting writers on higher ed innovation in the game, that lays out a detailed plan for pushing the total cost of a public bachelor's degree down to $10,000. Not $10,000 a year, mind you: $10,000 total. She's not the first to have this idea, as Govs. Rick Perry (R-Tex.), Scott Walker (R-Wis.) and Rick Scott (R-Fla.) have all proposed $10,000 degrees.

But she thinks their plans are inadequate. "I've spoken to some university leaders in Florida, Texas and Wisconsin," she writes in an e-mail. "Unfortunately, with the exception of the UW Flexible Option, which looks like a real innovation in delivery, many of these new programs are pretty gimmicky and unrealistic. They rely on dual enrollment and acceleration, and they are limited to small regional campuses. They are unlikely to make a big impact."

So how would she do it? Six simple steps.

1. Reduce administration

This is probably the most important cost-saving in Kamenetz's plan. That's as it should be. As you'll recall, administrative expenses are a significant driver of college spending. Kamenetz makes the point vividly: "Roughly 70% of a typical college budget goes to compensate faculty and administration; but ironically, the real cost increases have not been in the teaching side of personnel costs but in the administrative side."

So she proposes ending that practice by flattening the responsibilities of colleges, both instructional and administrative, into three new professional roles. Academic advisers/coaches would be professionals, hopefully with subject knowledge relevant to the student's interests, who would be responsible for regularly checking in with students, focusing on those who are falling behind academically. They would be evaluated based on how persistent and successful the students they mentor are, much like K-12 teachers are currently evaluated. Meanwhile, instructors/instructional technologists would effectively work like K-12 teachers and, unlike traditional faculty, have specialized training in pedagogy and effective teaching practices. Half their weeks would be spent directly on students in a classroom setting, and the other half meeting with students and designing curricula. Both professions would work closely with professors/instructional designers, who would design the curricula the instructional technologists teach and oversee MOOCs (massive open online courses) that hundreds of thousands of students could enroll in. As in current practice, they'd be encouraged to conduct research as well.

Notice that, more than just reconfiguring how classes are taught to make more efficient use of professors, and to improve teaching quality, Kamenetz's plan would also render a lot of the administrative responsibilities of colleges and universities related to student counseling, academic advising, instructional design and development, student counseling and so forth, obsolete.

2. Reduce perks

Kamenetz's personnel flattening proposal doesn't eliminate the bulk of administrative costs at schools, but her second idea does. Kamenetz would effectively end the use of tuition dollars or public subsidies on room, board, sports, extracurricular activities and food provision. Those amenities, even living in a dorm, are luxuries at this point. If students want to pay extra to live on campus or play sports, that's their business, but there's no reason for the government to subsidize that, or for the students to take out loans for that purpose.

Colleges ought to "adopt the same strategy used by airlines: They can offer
these services and charge extra for them if students so desire," she writes. "No sports subsidized by tuition or taxpayer aid. No extracurriculars beyond what students organize themselves. There’s no food service, or to be more specific, onsite food services are provided by outside contractors with no cross-subsidy from tuition."

3. Boost graduation

While going to college is still worth it even if you drop out, it's a much worse deal if you don't finish, or if you take five or six years to finish. So Kamenetz emphasizes increasing the share of students who graduate, and shrinking the period during which they do this. That means emulating self-paced universities, like the online Western Governors University, that allow students to finish faster, or embracing competency-based diplomas, where your degree is based on the knowledge base you can demonstrate, not how many years you sat in a classroom.

It also means using academic advisers, like those Kamenetz proposes creating, to push students to graduate. ASAP, an intensive, accelerated remediation program at the City University of New York (CUNY) is a good example of how this would work. "ASAP students were placed in smaller face-to-face classes with extra access to advisers, tutors and other services," Kamenetz writes. "All the extra features meant that ASAP costs the college a total of $49,358 over three years for full-time-equivalent students, compared to $29,521 on average for the typical CUNY community college program — two thirds more. But of the original cohort who entered ASAP in 2007, 55 percent  earned their associate’s degree in three years, compared to 24.7 percent  of similar students in the broader CUNY campus, and just 16 percent  of urban community college students nationally. These graduation rates were so much higher that the intensive ASAP program ended up costing about 10 percent  less per graduate than the standard CUNY program."

Let me repeat that for emphasis: Even though it cost more upfront, the ASAP students were so much likelier to graduate that it cost less per graduate. That's pretty amazing, and something other schools should be emulating.

4. Blended learning

The choice between MOOCs and traditional, brick-and-mortar universities is often presented as an either/or, but it's really not. A meta-analysis from the Department of Education suggests that "blended learning," in which computerized lectures are combined with personalized in-person instruction, can save money but also boost achievement. And the more scaled-up text, video and other online content is, the more cost savings you'll find. Imagine if every school in the country, for their freshman mechanics course, only showed lectures by the country's most effective physics teacher, the one whose lectures are the most engaging and best at keeping students' attention. It stands to reason that most students would learn more and millions of dollars would be saved.

So Kamenetz wants to scale up blended learning through MOOCs, but not through relying exclusively on online classes, which tend to have huge dropout rates and abysmally low pass rates. More typical could be what Anant Agarwal, the founder of the MOOC platform edX, calls "SPOCs" — small, private online courses. The National Center for Academic Transformation (NCAT) has been experimenting with these approaches for a while now. Between 1999 and 2004, they redesigned courses for 55,000 students at 30 colleges.

"Many of the NCAT’s course redesigns do away with the traditional structure of lectures, sections and office hours altogether in favor of a model called the 'emporium,' " Kamenetz writes. "Students come to a computer lab and work on their assignments, while getting help on demand from faculty members and undergraduate learning assistants or peer TAs, whether in person or online. In some cases, they also had more group assignments that encouraged them to interact and form support systems for each other — more than is typical for an intro-level course — both online and offline."

That means that you're not losing the benefits of college that come from interacting with your peers, effects that online-only courses often lack. And because so much of the work is online, there can be more interaction in forums and chatrooms, and teachers are better able to track students' progress, seeing when and for how long they read various documents, how long they worked on certain assignments and so forth.

And it worked! The NCAT redesigns reduced spending by 37 percent on average, and almost all led to higher exam scores and final grades.

5. Fewer majors

Yes, Folklore & Mythology is awesome (although I think Dutch Studies might actually be my favorite). But in practice, students are mainly interested in only about a dozen subject areas, and plenty of the sundry departments and majors that many universities offer can often be combined. Neuroscience and cognitive science can go together, international studies and political science can be merged, the sundry biologies can live peacefully together, the Spanish and French and Italian departments can become one Romance Languages department, and so forth.

To that end, Kamenetz proposes offering only 10-12 majors, including engineering, biology, education, computer science, English, communication, accounting and economics. Around 80 percent of students major in one of  these majors anyway, so the amount of curricular damage relative to the savings would be minimal. She would also abolish the major of "business," the single most popular undergraduate major, but perhaps also the least rigorous, and which produces relatively poor-achieving students. Instead, she'd fold practical business classes into the economics major.

6. Four levels of college

Perhaps Kamenetz's most radical proposal is to emulate the old University of California model, which had three tiers of institutions: research universities at the top, which accepted the best students and award PhDs; California State Universities, which accepted students who didn't reach the UCs and offer master's degrees; and community colleges open to all, whose students had a right to spots at CSU or UC campuses once they got their associate's degrees. That made sense in the 1960s, but Kamenetz argues that a four-tier model is more fitting these days:

• Cohort colleges: These would be commuter campuses for traditional-age students who need either remediation or other forms of support, and aren't ready to study online or at a flagship campus. They would provide weekly academic coaching, meant to develop "non-cognitive" skills such as motivation and persistence, and use blended learning for their classroom instruction. Students would get an associate's degree in 18 months, and a bachelor's in three years, or could merge into online universities (a version of Cohort College would be offered online as well). Second-year undergrads would be required to serve as TAs, tutors and support staff, which keeps their tuition low and strengthens the community.

• Adult online universities: These would be self-paced programs where students, particularly older or working students, would gain degrees for demonstrating knowledge or conducting research, rather than attending classes or paying for a required number of years. There would be a supervised internship or co-op component to boost the school's ability to place students in the workforce. Generous credit would be given for prior coursework. Colorado State University-Global is already doing this, and has had great success. It's entirely self-funded, needing no state money, and 95 percent of its graduates get a job, with over half earning in excess of $55,000 a year. And depending on the number of credits students bring to it, a bachelor's at CSU can cost as little as $10,500.

• Flagship campuses: These would still serve as research centers with a large faculty, but with some twists. They would be reoriented around developing MOOCs that could be used for blended classes at the flagships or at cohort colleges, or for online-only classes at Adult Online U. The flagships would also be responsible for producing research and course materials that would inform instruction throughout the system. Kamenetz also wants flagships to stop being snobs and get radically less selective. There's no reason for public schools to act like elite private schools like Harvard or Yale and exclude from the outset nine out of ten applicants. So flagships in this model would accept many more students, including many who wouldn't qualify under narrow SAT/grade/extracurricular admissions criteria.

• Micro/popup schools: These are tiny, intensive, face-to-face programs that focus on particular subjects and use practitioners to teach them. A great example would be Treehouse, a online school that teaches Web and mobile application development, or App Academy, an in-person Ruby-on-Rails/Javascript course in San Francisco and New York where tuition is paid as a percentage of income once you get a job programming. Kamenetz would encourage the public sector to encourage these kinds of start-ups and be more willing to offer credit for students who attend them.

So why not do it?

Kamenetz's plan doesn't lack for details. An as-yet unpublished appendix lays out exactly where spending at Cohort Colleges and Adult Online Us would go, down to the instructor salary level. But it also would uproot thousands of workers in the higher ed sector and radically change the careers of those who remained. Kamenetz acknowledges the concern, but thinks reform is too important to not do something drastic.

"I recognize the political pain involved, but the purpose of public higher ed is to educate people, not to employ them," she e-mails. "One of the travesties of the current system is the percentage of classes taught by low-paid adjuncts. I envision a system that might have fewer jobs, but they're better jobs."

And she doesn't think much of the suggestion of some MOOC critics, like the San Jose State philosophy professors who opposed using lecture videos from Harvard political philosopher Michael Sandel's MOOC, that online education is just a cost-cutting measure that will worsen the gap between the educations the rich and poor receive. "Class-based stratification is entrenched in the existing model. It could hardly be worse. I'm talking about differentiating experiences based on whether someone speaks English as a second language, hasn't passed algebra, or is 35 with two kids," she writes. "Right now the kids at San Jose State don't even get to watch the videos. Is that somehow better?"

Conwan emphasizes that Third Way and Kamenetz are decidedly not calling for a reduction in the amount that states spend to subsidize higher education; they simply want the money to be redirected to activities that impart learning more efficiently and cheaply from the student's perspective. "It is no longer sensible to finance the university system on the backs of middle-class families," he says. "And that is what people opposed to these kinds of reforms are for."