I’m thrilled to hear that the Economist has just launched a new column about business, innovation and entrepreneurship in honor of Joseph Schumpeter (1883-1950), the brilliant Austrian economist who,

argued that innovation is at the heart of economic progress. It gives new businesses a chance to replace old ones, but it also dooms those new businesses to fail unless they can keep on innovating (or find a powerful government patron). In his most famous phrase he likened capitalism to a “perennial gale of creative destruction”. For Schumpeter the people who kept this gale blowing were entrepreneurs. He was responsible for popularising the word itself, and for identifying the entrepreneur’s central function: of moving resources, however painfully, to areas where they can be used more productively. But he also recognised that big businesses can be as innovative as small ones, and that entrepreneurs can arise from middle management as well as college dorm-rooms.

Schumpeter’s work on the dynamism of high-tech markets (later married with Clayton Christensen‘s concept of “disruptive innovation“) is one of the most persistent themes across cyber-libertarian thinking of all stripes on a wide variety of issues. You can listen to an interview with the new column’s author on the Economist podcast here (MP3). One important point the author makes is that Schumpeter realized that celebrating capitalism did not preclude criticizing individual capitalists when justified and vice versa—something all too often forgotten today.

The Wall Street Journal reports today that student loan borrowing for college “in the 2008-09 academic year grew about 25% over the previous year, to $75.1 billion,” with the average student borrowing $13,172 to pay for college. So it should come as an enormous relief that one Internet start-up, StraighterLine, has essentially made the university fully virtual, offering classes for just $99/month.  While this may seem like a boon for students, especially the millions of Americans for whom even community college tuition seems an insurmountable obstacle to climbing up the economic ladder, such “e-Learning” offerings are already, predictably, coming under attack by entrenched interests in “Big Ed” (the professoriat!) as the “media-software–publishing–E-learning-complex.”

In Washington Monthly, Kevin Carey explains why “The next generation of online education could be great for students—and catastrophic for universities.” In a nutshell, the story is the same basic theme of Chris Anderson’s book Free!: digital distribution of information will ultimately drive costs down to zero. Carey shows how universities are essentially facing the same sorts of pressure from disruptive innovation as newspapers—except with more capital costs:

Colleges are caught in the same kind of debt-fueled price spiral that just blew up the real estate market. They’re also in the information business in a time when technology is driving down the cost of selling information to record, destabilizing lows. In combination, these two trends threaten to shake the foundation of the modern university, in much the same way that other seemingly impregnable institutions have been torn apart. In some ways, the upheaval will be a welcome one. Students will benefit enormously from radically lower prices—particularly people like Solvig who lack disposable income and need higher learning to compete in an ever-more treacherous economy. But these huge changes will also seriously threaten the ability of universities to provide all the things beyond teaching on which society depends: science, culture, the transmission of our civilization from one generation to the next. Whether this transformation is a good or a bad thing is something of a moot point—it’s coming, and sooner than you think.

Carey tells the the fascinating tale of StraighterLine, whose founder, Burck Smith, predicted the rise of digital higher education in his graduate thesis at Harvard’s Kennedy School of government in 1999. He predicted that we would look back from 2015 and see that:

Technological change was the spark that ignited the wildfire of change. Like a hole in a dike, cheap and instantaneous Internet-based content delivery and communication nibbled away at barriers to institutional competition… . Suddenly, a student seeking an introductory statistics course could choose from hundreds of online courses from anywhere in the world… . Feeling the effects of low-cost competition, site-based education providers started cutting course costs and prices to attract students.

Carey’s tell would resonate well with anyone who’s read George Gilder’s brilliant works Microcosm and Telecosm, about how innovation in general and digital economics radically drive down costs.

Even as the cost of educating students fell, tuition rose at nearly three times the rate of inflation. Web-based courses weren’t providing the promised price competition—in fact, many traditional universities were charging extra for online classes, tacking a “technology fee” onto their standard (and rising) rates. Rather than trying to overturn the status quo, big, publicly traded companies like Phoenix were profiting from it by cutting costs, charging rates similar to those at traditional universities, and pocketing the difference. This, Smith explained, was where StraighterLine came in. The cost of storing and communicating information over the Internet had fallen to almost nothing. Electronic course content in standard introductory classes had become a low-cost commodity. The only expensive thing left in higher education was the labor, the price of hiring a smart, knowledgeable person to help students when only a person would do. And the unique Smarthinking call- center model made that much cheaper, too. By putting these things together, Smith could offer introductory college courses à la carte, at a price that seemed to be missing a digit or two, or three: $99 per month, by subscription. Economics tells us that prices fall to marginal cost in the long run. Burck Smith simply decided to get there first.

Essentially what Smith did was to unbundle content:

To anyone who has watched the recent transformation of other information-based industries, the implications of all this are glaringly clear. Colleges charge students exorbitant sums partly because they can, but partly because they have to. Traditional universities are complex and expensive, providing a range of services from scientific research and graduate training to mass entertainment via loosely affiliated professional sports franchises. To fund these things, universities tap numerous streams of revenue: tuition, government funding, research grants, alumni and charitable donations. But the biggest cash cow is lower-division undergraduate education. Because introductory courses are cheap to offer, they’re enormously profitable. The math is simple: Add standard tuition rates and any government subsidies, and multiply that by several hundred freshmen in a big lecture hall. Subtract the cost of paying a beleaguered adjunct lecturer or graduate student to teach the course. There’s a lot left over. That money is used to subsidize everything else. But this arrangement, however beneficial to society as a whole, is not a particularly good deal for the freshman gutting through an excruciating fifty minutes in the back of a lecture hall.

The analogy to newspapers is clear:

What happens when the number of students making that choice reaches a critical mass? Consider the fate of the newspaper industry over the last five years. Like universities, newspapers relied on financial cross-subsidization to stay afloat, using fat profits from local advertising and classifieds to prop up money-losing news bureaus. This worked perfectly well until two things happened: the Internet made opinion and news content from around the world available for nothing, and the free online classified clearinghouse Craigslist obliterated newspapers’ bedrock revenue source, the want ads. Suddenly, people didn’t need to buy a newspaper to read news, and the papers’ ability to subsidize expensive reporting with ad revenue was crippled. The result: plummeting newspaper profits leading to a tidal wave of layoffs and bankruptcies, and the shuttering of bureaus in Washington and abroad. Like Craigslist, StraighterLine threatens the most profitable piece of a conglomerate business: freshman lectures, higher education’s equivalent of the classified section. If enough students defect to companies like StraighterLine, the higher education industry faces the unbundling of the business model on which the current system is built. The consequences will be profound. Ivy League and other elite institutions will be relatively unaffected, because they’re selling a product that’s always scarce and never cheap: prestige. Small liberal arts colleges will also endure, because the traditional model—teachers and students learning together in a four-year idyll—is still the best, and some people will always be willing and able to pay for it….. Regional public universities and nonelite private colleges are most at risk from the likes of StraighterLine. They could go the way of the local newspaper, fatally shackled to geography, conglomeration, and an expensive labor structure, too dependent on revenues that vanish and never return.

The difference between newspapers and higher education is that, in education, the government ultimately holds all the cards through the power of accreditation:

Credits and degrees can only be granted by—and students paying for college with federal grants and loans can only attend—institutions that are officially recognized by federally approved accreditors. And the most prestigious accreditors will only recognize institutions: organizations with academic departments, highly credentialed faculty, bureaucrats, libraries, and all the other pricey accoutrements of the modern university.

Carey concludes by noting that government regulation ultimately threatens innovation and will likely protect “Big Ed” from competition, just as it has protected the Big Three auto manufacturers:

Smith’s struggle to establish StraighterLine suggests that higher education still has some time before the Internet bomb explodes in its basement. The fuse was only a couple of years long for the music and travel industries; for newspapers it was ten. Colleges may have another decade or two, particularly given their regulatory protections. Imagine if Honda, in order to compete in the American market, had been required by federal law to adopt the preestablished labor practices, management structure, dealer network, and vehicle portfolio of General Motors. Imagine further that Honda could only sell cars through GM dealers. Those are essentially the terms that accreditation forces on potential disruptive innovators in higher education today.

Carey’s story highlights not only what the future of universities might look like, but the dangers of subsidizing newspapers as well. Many who have bemoaned the demise of newspapers have called for indirect government support in the form of preferential tax treatment by relaxing non-profit rules to accommodate newspapers, letting newspaper subscribers deduct their annual newspaper spending or perhaps even borrowing from the French and giving income tax breaks to newspaper journalists—or direct subsidies like bailouts for newspapers or expanded public funding the model of NPR. The obvious problem raised by such proposals is the old Golden Rule: “He who has the gold, makes the rules.” Control necessarily follows funding.

The more subtle, long-term problem with such subsidies is that they require some form of accreditation to determine who or what is eligible for government largess or tax breaks.  This creates essentially the same problem Carey predicts with universities: Established interests will lobby government to protect them from competition by radically cheaper, innovative business models based on new digital technologies. This, in turn, will both retard the development of education, and therefore the advancement of knowledge, and increase the problem of government control: The lower down what George Gilder called the “Learning Curve” we go in terms of figuring out how to make e-Learning inexpensive, the greater Big Ed will depend on government for protection. If that happens, academic freedom will become as meaningless as freedom of a press dependent on government funding.

As a practical matter, I wonder:

1. How services like StraighterLine will compete with free offerings of course materials through services like iTunes University, which offers a growing compendium of free college lectures (but not course materials, tests or tutoring)
2. When we’ll finally see e-Learning costs drop enough that an ad-supported model might become feasible. You may laugh now, but today’s $99/month could well be tomorrow’s $9.99/month—at which point advertising models might prove competitive, especially if advertisers can get the revenue boost of delivering personalized marketing to users.

Jerry Yang’s departure as Yahoo! CEO opens the door to a renewed bid by Microsoft to buy Yahoo!’s search business (or Yahoo! itself).  Such a merger could produce a significantly stronger challenger to Google in the search market.  With this possibility in mind, the WSJ just ran a fascinating history of the “paid search” business—the placement of “contextually targeted” ads next to search engine results based on the search terms that produced those results.

In a nutshell, Microsoft failed to see (back in 1998-2003) the enormous potential of paid search—just as small start-ups (such as Google) were starting to develop the technology and business model that today account for a $12+ billion/year industry, which is  twice the size of the display ad market and which supports a great deal of the online content and services we have all come to take for granted online.  Microsoft first put its toe in the water of paid search with a small-scale partnership with Goto.com in 1999-2000.  But this partnership failed because of internal resistance from the managers of Microsoft’s display-ad program.  In 2000, Google launched Adwords and thus began its transformation from start-up into economic colossus.  By 2002, Microsoft realized that it needed to catchup fast, and approached Goto.com (by then renamed Overture) about a takeover.  But Microsoft ultimately chose in 2003 not to buy the startup because  Bill Gates and Steve Ballmer “balked at Overture’s valuation of $1 billion to $2 billion, arguing that Microsoft could create the same service for less.” 

Microsoft, meanwhile, spent the next 18 months deploying hundreds of programmers to build a search engine and a search-ad service, which it code-named Moonshot. The company launched its search engine in late 2004 and its search-ad system in May 2006.

But Microsoft’s ad system came too late:

Advertisers applauded Moonshot for its technical innovation. But Microsoft had trouble coaxing people to migrate to its search engine from Google; advertisers were unwilling to spend large sums on MSN’s search ads. By building a new system instead of buying Overture, Mr. Mehdi says, “we really delayed our time to market.”

What’s most fascinating about the piece is that it seems to suggest that Microsoft missed its opportunities to get into paid search not because it was “dumb,” “uninnovative” or a “bad” company, but for the same sorts of reasons that big, highly successful and even particularly innovative companies fail.  The reasons companies generally succeed in mastering “adaptive” innovation of the technologies behind their established business models are the very reasons why such great companies struggle to encourage or channel the “disruptive” innovation that renders their core technologies and business models obsolete.  This dynamic was described brilliantly in Harvard Business School professor Clayton Christensen’s classic 1997 book The Innovator’s Dilemma:  When New Technologies Cause Great Firms to Fail.  (Read chapter one here and Tim Lee’s recent discussion of the book here.)  

Whatever one thinks about the debate over whether antitrust intervention is necessary to restrain Google’s growth, I’m sure we’d all applaud the evolution of increased competition in the paid search market through market forces.  Let’s hope that Microsoft—as well as Yahoo!—have carefully studied the vast literature produced by business schools in the wake of Christensen’s book about how big companies can avoid the Innovator’s Dilemma by promoting—and capitalizing on—radical innovation from within.  Indeed, this seems to be precisely what has guided Google’s own strategy as it has grown from “disruptive innovator” to become the very sort of behemoth that cannot easily escape the Dilemma, even if corporate managers are fully aware of the problem on a theoretical level.  If Google can do it, Microsoft should be able to, too.  But let’s also not discount the possibility that, no matter how hard Google’s management might try to retain the innovative culture of a start-up, the giant  can’t do that well enough to prevent its own apparent market dominance from being disrupted by new upstart innovators in search and advertising technologies.  

The head of Google Research talked about some of these possibilities in July 2007 and the Google has recently covered other possibilities.  Here are my own bets—for what little they’re worth—as to what such “disruptors” might be:

• Semantic search and social search – whichever search engine masters these tools will likely dominate the market for search, and thus search advertising.
• Micro-payments to search users for using a search engine and discounts for clicking on ads – something Microsoft has pioneered with its Cashback system but which is probably still only in its infancy.
• Behavioral targeting that can make display ads competitive with search ads by making display ads as relevant to consumers as search ads (or even more so), rather than simply trying to target display ads based on the context of a page—which limits the economic value of the ad “display inventory” that websites try to fill with ads, especially for smaller websites in the Internet’s “long tail” whose subject matter might have little relevance to the keywords for products or services that are more highly valued by advertisers.  
• Technologies that allow contextual targeting of ads in or around videos based on the contents of the video (and associated discussion by viewers in comments). Even the imperfect ability to automatically create transcripts of a video, and then search for keywords, could hugely increased the value of advertising associated with video content.

I suspect we’d all be at least a little surprised if we could see what search engines—and online advertising—really looked like in, say, 2019.  But I won’t be terribly surprised if Google—for all its ingenuity—ends up making some of the same mistakes Microsoft made with Search 1.0 ( c. 1998-2005).