In a new research paper, Christian Terwiesch, professor of operations and information management at Wharton, and Karl Ulrich, vice dean of innovation at the school, examine the impact that massive open online courses (MOOCs) will have on business schools and MBA programs. In their study — titled, “Will Video Kill the Classroom Star? The Threat and Opportunity of MOOCs for Full-time MBA Programs” — they identify three possible scenarios that business schools face not just as a result of MOOCs, but also because of the technology embedded in them. In an interview with Knowledge@Wharton, Terwiesch and Ulrich discuss their findings.
An edited transcript of the interview appears below.
Knowledge@Wharton: Christian, perhaps you could start us off by describing the main findings or takeaways from your research?
Terwiesch: Let me preface what we’re going to discuss about business schools by saying that Karl and I have been in the business school world for many, many years. We love this institution, and we really want to make sure that we find a sustainable path forward for business schools.
Business schools in the world of these massive online courses are somewhat threatened, and a lot of that has to do with our cost structure. We are very expensive organizations. There are two main reasons for that. We do two things. We teach and we do research, but only the teaching part comes with revenues, and so often, the research work that we do, all this great research work that is funded for us, is funded by our students. The second thing is, honestly, like most non-profits, we don’t always have an eye on efficiency. If you and I were running an airline together and we were to fly our planes half empty, very quickly, bad things might start to happen. Yet that culture of efficiency and productivity is something that we haven’t had in the business schools. As these MOOCs come along, the cost pressure on our institutions is going to change because suddenly, there’s a very serious alternative to coming to a two-year degree program at Wharton.
Knowledge@Wharton: Karl, anything to add to what Christian just said?
Ulrich: The other thing I would say is a key finding of our analysis is to look at what the benefits are that MBA students derive from their full-time MBA experience. We point out that teaching and learning, the more traditional academic topics, are probably only a quarter or a fifth of the reason that students come and get an MBA. And that’s the piece we think is most susceptible to change from this emerging new technology but that probably doesn’t impact other elements of the MBA program.
Knowledge@Wharton: Your research paper talks about the fact that in a MOOC, what is relevant to the business school is not the MOOC per se, but the technology, the video-based technology that is embedded in the MOOC. Twenty years ago, there was a lot of hype around the fact that lectures were available on video tapes, or later that they were available on CD ROMs. Each time, there were high expectations about what this would do to business schools, but none of these [technologies] disrupted the business school model in any fundamental way. Do you think a MOOC is fundamentally more disruptive? If that is the case, I wonder if you could explain why.
“What can we do in the classroom that can only be done when 60 people are together sharing the same time and location?” –Karl Ulrich
Terwiesch: Let’s start with the technology. We find it’s helpful to separate the MOOC from the technology that enables the MOOC. That technology is a combination of short videos, smart testing with potentially automated grading, social networking, online communities — all of these things wrapped together, as we’ve seen with courses at EdX and particularly with Coursera, are creating a technology that is applied now to the MOOC, not given away for free as we’ve seen it over the courses that we’ve taught here at Wharton. But that MOOC and that technology are two separate pieces. The MOOC is just one application, yet we can imagine many other ways in which these new technologies — we call these the SuperTexts — can be used. There are many other ways in which we can deploy that technology. These and other applications are a much bigger threat to institutions like ours, compared to taking these courses, teach them to 100,000 students at once and giving that away for free. We don’t feel threatened by the MOOCs, [but] we do feel threatened by the SuperTexts.
Ulrich: I’ll make a few other observations about video. Video has been available to consumers for 30 years, but there’s something fundamentally different about the availability today of video. I think that’s evident if you look at the behavior of consumers online. They’re watching Khan Academy, they’re watching TED Talks, they’re spending a lot of time on YouTube. When I watch my teenagers do their math homework, they do it with their laptop and their smart phone so that they can text their friends, and then they watch a YouTube video to [learn] how to do a math problem. Video is fast, and it’s available everywhere with just a click. The big insight in the first MOOCs was that by chunking the video, by making it just a few minutes long and making it semi synchronous — that is, available on demand during a week of the course — you get a much greater adoption and viewing of that video as opposed to, for instance, having to take a video cassette out of a library, check it out, take it home, plug it in, fast forward to the right section. That friction substantially diminished the usefulness of video when it was first introduced.
Knowledge@Wharton: One of the things you mention in your paper is that of the three pathways that you described in your research, the first one is that B schools will be able to serve more students better and more efficiently. Since each of you teaches a course on Coursera, I wonder if you could draw upon your own experiences and show how you teach better and more efficiently via the MOOC platform than you might in a classroom.
Ulrich: I teach product design at Penn. I’ve taught it for 20 years, and I’ve recently taught a MOOC in which maybe 100,000 students over four offerings of the course have participated in that subject. I made 65 short videos that explain some of the key concepts in product design. When teaching product design two years ago, I found myself in class opening up my web browser, going to my Coursera course and showing my students a video, [saying]: “Oh, I have this great example, let me show it to you.” And then I realized how silly that was and that what I really should be doing is having the students watch those videos without me before they get to class. So, in the last year, when I’ve taught product design, I’ve posed to myself the challenge: What can we do in the classroom that can only be done when 60 people are together sharing the same time and location? I have them watch the video offline, and then when we get together, we do a simulation or an exercise or presentations or group work — things that can only be done in that location when we’re all together.
Knowledge@Wharton: So, that’s the flipped classroom model. Christian, are you using the same model through your MOOC?
Terwiesch: Unlike Karl — his videos are actually funny — I am in a somewhat more constrained environment. The question you’re posing is an important one — that we think about our brick-and-mortar classroom offerings. I teach a core course in operations management, among other things. It’s 12 times 80 minutes that every student at the Wharton School has to take — 960 minutes in total. My online course on Coursera is about 45 video segments of about six to seven minutes. So you see that there’s almost a compression of 50% as you go from the traditional classroom to online media.
Now the question is, of course, what happened to the other minutes? It’s helpful to break those up into a couple of buckets. The first bucket is just there are certain things that we do in the classroom that we unfortunately cannot replicate on the MOOCs. These are case discussions, games, exercises. Those are moments that Karl describes; those are things that can only happen when we are together. But I also have to point out that there is a fair bit of waste in what we do in the classroom. That has to do with the fact that different students learn at different rates. When a student struggles on the MOOC, they can just rewind or they can just read something and then catch up, whereas in the classroom, you’re constrained by a common pace with everyone. So you really do gain efficiency. You gain productivity as you take the same group of students and have them moved over to a hybrid model. Karl has been at the forefront of this. At the core, I don’t think we’re quite as far, but the direction is certainly clear. We have to benefit from the new technology.
Knowledge@Wharton: Despite the evident advantages, one of the challenges that all MOOCs face is that the completion rate typically tends to be pretty low. I wonder both as teachers and as researchers, what you have learned through your experience that might be relevant to business schools?
Ulrich: The completion rate statistic is a red herring. People like to trot it out, but it doesn’t seem terribly relevant to me. If you think about what the barrier is to registering for a MOOC, it’s literally one click in a Coursera environment. So many people enroll to check it out, to watch a few videos, to see what it’s all about.
My MOOC requires hundreds of hours of effort to complete a substantial design project. Very, very few people are willing to put in hundreds of hours, but many people are interested in learning a little bit about customer needs or aesthetics in design or prototyping, so they watch a few videos. We think pretty carefully in MOOCs about three categories of learners. Those who are just browsing; those who want to view the material but won’t do the work; and then those who will do all of the work. So as I say, that narrow completion statistic is not meaningful in terms of evaluating the success of the MOOCs.
Terwiesch: Yeah, it’s a classic example of what oftentimes people do when an innovation comes out. They take a performance metric that is designed to evaluate the old technology — which is brick-and-mortar learning — and try to apply it to the new technology. Completion rates are important if you pay $120,000 for your MBA degree or even more dollars when you send your kid to college. There, the completion rate is a really, really important number to track, and we are anchoring this on the kind of 95-plus percent completion rate that we have in that world. Taking that number and comparing it to the 5% in the MOOC just makes no sense, as Karl describes.
Most of these people, since they don’t have to pay anything, just click and register in an effort [similar to browsing at] a store. You check out an item and you wonder, “Oh, do I want that?” This is not in some sense purchasing; it’s just looking at something and it’s the nature of the web that every looking instance is actually recorded as if you were enrolling. That statistic does not tell us anything. We should focus much more on the participation — and again, we see the segmentation that most students really just want to watch some videos, and then there are those students who really do the work. There [are also those who] drop out at the beginning, because they find that this is not what they are looking for.
Knowledge@Wharton: Let’s push a little further on that $120,000 number you mentioned that MBA students pay to attend an on-campus program. And as you correctly said, joining a MOOC costs nothing at all. But given the fact that the content you produce for a MOOC is expensive in terms of time and money, how do you think MOOCs can be made financially sustainable?
Ulrich: It’s actually not very expensive. If you look at what it costs to develop a MOOC, in a sustainable mode, in the long run it would be about $70,000, but we reach with a MOOC several hundred thousand students. If we really look at it – if you look at it on a per viewer basis — it runs to about 50 cents per person. At 50 cents per person, that’s cheaper than almost any other form of outreach. Fifty cents for that kind of engagement is very, very inexpensive.
Terwiesch: A little bit of further math might help here. We, elite universities, elite business schools, are in the business of creating reputation. Reputation is important. It will drive our demand; it will drive how our graduates are viewed in the market; so reputation is key. How do you create reputation? The traditional vehicle of reputation building was research. It is useful to look at some numbers on how much it costs us. When Karl and I sit down and we write a scholarly article that makes it into the best academic journals, we find in our report that it takes us somewhere around $300,000 to $400,000 of research investments to just get one scholarly article out. For that money, for one single paper, we can basically create somewhere around three, four, or five MOOCs with the enrollment that we just talked about.
“The first thing we should do is distinguish between MOOCs and other forms of online education. MOOCs are fundamentally about outreach and social mission, and they’re not very expensive.” –Karl Ulrich
The other reference point that we might use is [a comparison of the school’s] operating budget and what we’re spending on MOOCs. We’re spending about 0.1%, if not a little less, of our budget on MOOCs that apparently we all believe is the key technology for the future. So this notion that we’re over-spending I don’t buy. Other forms of disruptions, other forms of concerns we need to discuss, but the notion that this is expensive and we’re stealing money from our traditional customers and giving that content away for free, that threat I think is not real.
Knowledge@Wharton: Clearly, there is a huge value to the knowledge that is disseminated through MOOCs. The question, of course, is that different schools are trying to monetize them in different ways. At Harvard, for example, they have introduced a paid pre-MBA program. If you were to look at some of the different economic models that are being wrapped around MOOCs, could you give me your thoughts on where you think this is going and which models are likely to be more sustainable?
Ulrich: Well, the first thing we should do is distinguish between MOOCs and other forms of online education. MOOCs are fundamentally about outreach and social mission, and they’re not very expensive. We can easily justify them based on the social value that’s created. Now, that’s not to say that there aren’t some nice economic opportunities for online education. At Wharton, there are two that we think could be important. One is executive education. If you look at how expensive it is to send someone physically to Philadelphia to spend a week in executive education, it’s a $5,000 to $10,000 cost, and we may be able to deliver some of the benefits of that experience much less expensively using technology. So that’s one. The other is that we think that a place like Penn and Wharton can be the originator of content that’s used in instruction by other institutions, and as with textbooks, there’s a way to charge per use or per user for that material, and so that could be a potential revenue source.
Terwiesch: There’s a lot value on the table, especially going back to Karl’s executive education example. There are lots of people in corporate settings who come to Wharton, and we’re talking about tuitions of $7,000 to $8,000 per week. These people have to travel and also be away from work — and so by design, this is only a privilege that you get if you’re working in top management. There is a large number of other people in executive positions at the echelon below whom we are right now not reaching but whom we could be reaching through technology. They would love the learning. We can certainly deliver through the technology at cost points way south of maybe $1,000 or $500 for that one-week learning experience online. If we bring those two things together, we’re going to generate a lot of value. There’s no reason to believe that, especially with a strong brand like we have at Wharton, we wouldn’t be able to also then capture some of the value for ourselves.
Knowledge@Wharton: In other words, you can spread the ripple effects of executive education much farther to the lower tiers of an organization by combining executive educations and MOOCs. That’s a great point. The other interesting point in your research paper is that you said you can teach existing students with fewer faculty members. Given the fact that most things tend to happen in academia when they get enough faculty buy-in, do you think that faculty members are likely to support something that threatens to cut down their numbers in the future?
Terwiesch: This one is a scary thought, but let’s put this in context. What we do in the paper is we articulate three scenarios that we believe can play out in the world of SuperText. As I said earlier on, I’m not too wild about the term MOOC because we’re talking about the technology here. There are three scenarios, and the reason why we work with three scenarios is that academia is full of bad forecasts and predictions, so we didn’t want to join that club. We know that social systems are so complex that if you and I and Karl sit here and make predictions about the future in 10 years, we’re going to be wrong.
And so what we did is articulate three scenarios. The first scenario is that we’re going to have this new technology that makes us more productive. We have the same faculty, the same business schools stay in place, and so now that they’re more productive, they can reach more students.
The second scenario is the one that you’re pointing to, where we have a constant student population because really the demand for top MBA programs almost by definition is not without limits. So, if we have fixed demand but a much more productive production technology, by definition, there are going to be fewer of us [faculty]. And the third scenario is so disruptive that the business schools fall to pieces.
So that second scenario that we’re talking about, we call that the clowns and the movie theaters. In the 18th, 19th, 20th century, if you wanted entertainment, you would show up at the local village square, you would watch a clown, they would do some funny things in front of you and you would have a laugh. Well now, you watch a movie and unfortunately, most clowns have lost their jobs. What we’re debating is to what extent Karl and I are going to be the clowns of the 21st century.
Ulrich: You asked about whether faculty would resist the change. Clearly they would, under that scenario, but I think we already have pretty good evidence that in the face of cost pressure, there will be a reduction in faculty. If you look at higher education more generally over the last 10 to 20 years, the number of student credit hours taught by tenured and tenure track faulty has been steadily declining. There has been an increase in the use of adjunct and part time faculty. That is a trend driven by cost pressure. Yes, we all didn’t like it, but it didn’t mean it didn’t happen. And so we can not like it, but in the face of a competing technology that may be better and that’s definitely cheaper, it’s very hard to argue that we can completely resist it.
“Is there another way of delivering knowledge very much on demand when you want it — instead of buying this library of videos, have much more of an iTunes or Netflix model?” –Christian Terwiesch
Knowledge@Wharton: Let’s talk about the third scenario before coming back to all three of them, which is: When you talk about the complete unbundling of the business school or the emergence of other alternatives to business schools, what do you think might play out in that option?
Ulrich: The first thing to recognize is that the primary reason that an MBA student matriculates at the Wharton School, or at any other top business school, is because they think it’s going to advance their career. To some extent, they want to learn some specific skills and knowledge, but there really are other benefits that they look to achieve and to obtain. These are access to careers and the development of a social network that will benefit them throughout their lives, and to some extent, a credentialing or a stamping of approval that distinguishes them from other people in the labor force.
If those functions of credentialing, of social network and of access to employers are now provided by others outside business schools, then I think our role is quite threatened. Right now they’re bundled together and they include this academic learning piece. But if they ever become unbundled and are started to be provided by other institutions, then I think business schools are threatened.
Terwiesch: The metaphor that Karl and I use in our report is the idea of a Swiss Army Knife. When you purchase a Swiss Army Knife, you don’t know yet when you will use it and you don’t know which part of the knife you’re going to use first. That’s very similar to our MBA students. They come here, they get all this knowledge that one day in their life and sometimes 10, 15, or 20 years into their career, they might have to access but they don’t know when and they also don’t know what part. That begs the question, is there another way of delivering that knowledge very much on demand when you want it — instead of buying this library of videos, have much more of an iTunes or Netflix model where you say, “Well, I want to watch that movie today,” or just click on it and get it when you need it. If you face a situation at work, you’re doing some M&A work, why don’t you take the M&A course right now when you need it?
Knowledge@Wharton: Of the three scenarios that you outline in your research, which ones do you think are most likely for business schools and why?
Terwiesch: The nature of scenarios is that they are all somewhat likely or we wouldn’t have created them. Let’s add to them that these scenarios to some extent are scenarios like we would roll the dice and it’s going to be one, two, three, four, five, six, right? It’s a shock, it’s random and it’s just hard to predict.
But to some extent, these scenarios are also active choices. They’re strategies, especially at the top schools, especially at the leadership level. We can make some choices that will influence the future and that will basically fall into these three pockets. I feel that we are well underway [in terms of] playing out scenario one. We have done the MOOC for outreach; we’re experimenting with executive education. We need to do more, but we’re clearly underway on scenario one. I would love us to play a little bit with scenario three — that’s the furthest out. As for scenario two, right now we’re in this privileged situation here at Wharton, so the cost pressure has not been quite as big. I would [bet on] a combination of one and three.
Ulrich: Scenario two is pretty likely. There’s going to be some decline in faculty size. The cost pressure is unavoidable. I don’t think it’ll be a catastrophic decline in faculty. My prediction would be a modest decline in faculty. To build on Christian’s point, we need to aggressively pursue scenario one, pathway one in order to mitigate pathway three. In particular, I think we have an opportunity at Wharton and at other elite business schools to create an unrivaled educational experience for students, one in which they learn a lot of the rote subjects, the more mechanical subjects online and without the use of expensive faculty resources. Then when they’re here together, we challenge ourselves to give them an experience that they couldn’t get anywhere else. By doing so, we reinforce the value of them participating in the Wharton experience, and that helps us mitigate against the unbundling that could happen in pathway three.
Knowledge@Wharton: What surprised you most about this research?
Ulrich: I have to say, our analysis is just arithmetic. It’s just adding and subtracting and a little division, but we were quite surprised by what scholarship costs, just how expensive it is and how expensive it is relative to the educational activities of the school. That surprised me.
Terwiesch: For me, the biggest surprise has been how seriously threatened our institution and the business school community is. As I said at the beginning, this is the institution where we’ve been employed for more than 20 years. It has made our careers. This has been the perfect job, [and it’s an] institution that we love. Nevertheless, when you run those numbers, you feel how threatened that ecosystem is. There is no guarantee that 20 years from now, we’re going to have the same business model going as it is now.
Knowledge@Wharton: Your research focuses primarily on business schools and the MBA program. What do you think in your research could apply more broadly to higher education as a whole?
Ulrich: Higher education sees the same pressures and the same threats and opportunities. In fact, to some extent, other parts of the university have felt them before business schools. My wife is a professor in the English department at Penn. They have already seen tremendous cost pressures, and they’ve seen that impact on the faculty ranks and the way they do instruction. I don’t really think business schools are terribly unique. They are somewhat unique — I would say professional schools are somewhat unique — in that the primary purpose of paying the tuition dollars is to advance yourself and your career. That is unique to professional schools as opposed to, say, an undergraduate experience which I think is much less threatened. An undergraduate experience is much more about going away, finding your identity, having enabling experiences, building a social network. Those things are much harder to substitute, at least for the elite universities, the Ivies and so forth.
Terwiesch: Certainly, we as business schools are particularly threatened because of course, again, I think we are used to this very high tuition, very high salary type of lifestyle. We could afford ourselves the luxury of not looking at efficiency; in many of the second and third tier schools, and many of the other fields outside of business schools, they have been under that cost pressure for so many years that I think they are better prepared. I think the small liberal arts colleges don’t have this dual purpose that we talked about earlier on — the research and the teaching. For them, it’s all about the education of the student. So as long as they do that well, as long as they serve their customers and students well, people will be willing to pay for it. We are threatened in particular by that scenario of unbundling the teaching and the research — that is something that is unique to law schools and business schools. I have an appointment at the medical school, where if you want to spend time doing research, [you need to] have somebody pay for it. Bring in the research dollars. This cross-subsidy is actually something that is not that broadly applicable other than in the social sciences. That is one area where I would feel we’re going to see some big pressure mounting.
Knowledge@Wharton: What future topics for research do you think are sort of thrown up as a result of what you’ve studied so far?
Terwiesch: I would love to see more insights into scenario three, into the unbundling [of business schools]. We make it sound so nasty in many ways, this unbundling, the business school falling to pieces. I think it’s actually not that bad, and in many ways quite exciting. When you do research, other than the peer evaluation, there’s no measurement of quality. There’s no feedback other than what we’ve researched. If you think about scenario three where the students learn on demand what they need and practice, that would actually create an enormous pull from the market that would direct us where we develop new content, where we develop research. I think it would actually be a very interesting development in terms of guiding our research where it matters the most in practice. I would love to see more research understanding of scenario three, understanding exactly that business model and how that activity system is falling into different pieces, but then these pieces are potentially managed by different organizations.
Ulrich: My personal agenda is that I’d very much like to see a quarter to a third of the content of our MBA program move to more self-directed, asynchronous learning online. One of the real challenges there is how you assess competence. It’s actually an opportunity because right now, we don’t assess competence in business education. We simply assess [whether] you complete the degree. Did you complete these courses? Get a passing grade in the courses? There’s an opportunity to improve what we do by developing some methods of assessing how good you are when you graduate from the Wharton school. If we can do that, we can both improve the quality of the education a student gets, and we can possibly make it better by moving some of the instruction online.