Swensen completely remade it. Harnessing a rare investing talent, but also adapting it to the efficient-markets doctrine that had become an article of faith in academia, Swensen redeployed close to 90% of Yale’s fund into a diverse mix of assets stretching across the investment universe: venture capital, leveraged buyouts, hedge funds, foreign equities, real estate, natural resources such as timber. The result was a revolution in the quiet but competitive world of university endowments. Over the three decades through June 2015, ­Swensen’s portfolio grew at an annual pace of 13.9%, outpacing those of his peers and all the relevant benchmarks.

His success enabled the endowment to increase its support of the university at a rate far above inflation; last year it contributed a third of Yale’s budget, just over $1 billion. The outsize performance, moreover, reassured alumni that their gifts would be preserved, persuading many to open their checkbooks. Today the endowment stands at $25.4 billion.

Not surprisingly, other schools rushed to adopt Swensen’s approach—now known as the ­“endowment model.” Today more than half (52%) of endowment assets are invested in so-called alternatives (everything from private equity to real estate) while only 16% are deployed in what was formerly the bedrock investment class, U.S. listed stocks, according to the 2015 National Association of College and University Business Officers (Nacubo)/Commonfund report.

Swensen, for his part, still has the magic. Over the past two decades Yale’s endowment has outperformed the average school’s by a whopping five percentage points. Over 10 years, Yale earned 8.1%—three points better than the average.

But elsewhere in academia, people are beginning to wonder whether that remarkable record has less to do with Swensen’s model than with the manager himself. That’s because few schools have been able to pull off what Yale’s investing wizard has accomplished. Through 2016, the average university endowment has had a poorer record—over one year, three years, five years, and 10 years—than the average public pension fund, according to the Wilshire Trust Universe Comparison Service. Even over 20 years, endowments and pensions are in a virtual dead heat (the latter trailed by 20/100th of a percentage point), meaning that over a generation, academia’s vaunted edge has vanished. For that matter, colleges also trail a passive stock and bond index, which requires no management talent at all. Through 2015, according to data from Harvard and Nacubo, which tracks 812 schools, the average endowment earned 6.3% a year over the previous 10 years, compared with 6.8% for a 60%/40% blend of U.S. stocks and bonds.

Even Yale’s ancient Ivy League rival has struggled of late. Under the tutelage of Jack Meyer, its endowment chief from 1990 to 2005, Harvard’s returns sizzled. But in the past decade, the $35.7 billion endowment fund has returned only 5.7% a year, significantly underperforming the passive 60%/40% blend it uses as a yardstick. Harvard Management Co., which runs the endowment, has been rocked by turmoil. It recently hired a new CEO—its fourth in a decade—and talented investors have departed amid embarrassing publicity over disappointing performance and eye-catchingly generous bonuses. (Harvard Management declined to comment for this story. Recently it revamped its compensation scheme to better align bonuses to performance.)

Photos, Swensen: Peter Foley—Bloomberg via Getty Images; Narvekar: Courtesy of Harvard

While Harvard’s challenges are, to some degree, sui generis, they have reinforced growing doubts over an investing style that has all but taken over America’s ivory towers. And coincidentally, one of the earliest warnings was signaled by Harvard itself. As far back as 2010, Jane Mendillo, who took over Harvard’s endowment in the dark days of 2008 and managed it through the end of 2014, wondered aloud in the fund’s annual report, “Has the ‘endowment model’ run its course?”—before responding with a self-­assurance not unknown along the Charles River: “Our answer to that question is ‘No.’ ”

Today Harvard is blushing crimson. Many endowment chiefs defend their performance, saying it’s unfair to grade them against U.S. benchmarks because U.S. markets have been top performers since the financial crisis. They also contend that central bankers have distorted equity returns by lowering interest rates. However, if low rates have benefited stock prices, they should equally have benefited the private equity and venture capital assets owned by colleges. So why isn’t the endowment model performing?

There seem to be three parts to that answer—all beginning with the letter “C,” which is a fitting grade for the investment performance of universities today.

The first is unnecessary complexity. It isn’t just that schools are chasing ever more elaborate portfolios; it’s that they’re doing so with an ever more complex roster of overseers. The average endowment has more than 20 outside managers, yet fewer than two full-time staffers to track them (many rely on consultants). The biggest endowments, those with more than $1 billion in assets, employ roughly 75 managers each just in “alternative” strategies. “Portfolios have gotten so complex, I worry about it,” says the head of a high-performing New England college. “It’s not clear that everybody knows what they own.”

Second is competition: It has whittled away the edge, even for the bigger and elite schools that have typically racked up the best results. Capital has poured into private equity funds, raising deal prices and reducing returns, which now mirror those of public markets. Hedge funds are further along the curve toward mediocrity, a result of high fees that attracted a tsunami of hungry capital, often managed by ordinary talent. “You could probably replace the term ‘alternative’ with ‘high-fee,’ ” says a cynical trustee on the investment committee of a prestigious New England school.

And the third, and perhaps most important, is conceptual. The endowment model—which emphasizes broad diversification—emerged from the dogma of efficient markets: the notion that, generally speaking, markets are good at internalizing available information, and therefore they generally price shares appropriately. That makes it hard to beat the market consistently over any long period. Nonetheless, many endowment managers try to outperform anyway.

Not everyone considers this a conceptual muddling. Andrew Golden, a Swensen disciple who manages the endowment at Princeton, says it’s not contradictory to respect the market as mostly efficient while trying to beat it. To the extent there are at least a few inefficiencies scattered about, presumably somebody can make money off of them.

And this is where the rub on that third “C” comes in: Swensen’s idea has always been that he could outperform by exploiting an edge. “If you don’t have an edge,” says Bruce Zimmerman, who just stepped down as head of the mammoth University of Texas endowment, “you ought to be simple and passive.”

[fortune-brightcove videoid=5229908449001]

Zimmerman contends that the biggest advantage of universities is “perpetuity”—endowments are investing for so far in the future that they’re not concerned with near-term volatility. Such thinking has enabled Yale’s endowment and others to take the plunge into private markets and venture capital and to invest in tangible assets—which, because of the lack of frequent trading and broad public vetting, are likely to be less efficiently priced than, say, equities on the Nasdaq. Furthermore, managers at Yale—which, like other elite universities, benefits from a cross-pollinating alumni network of money managers, entrepreneurs, and venture capitalists—have had the ability to cherry-pick the best of such investments. In Yale’s case, that alumni network has helped seal early investments in Oracle (ORCL), Dell (DELL), Amazon (AMZN), and Google (GOOGL). Yale’s approach in that regard is perhaps better labeled the “relationship model” than the endowment model—and the former is much harder to copy.

Larger schools also have the resources to scrutinize investments around the world (Golden, when Fortune caught up with him, had just returned from a tour of Asia). And some have advantages that are literally irreplaceable—the University of Texas reaps the bounty of a gushing, 2-million–acre field of oil and gas. Then there is the edge that flows from research. Mendillo, Harvard’s former endowment chief, discovered that pulp companies were eager to get timber assets off their books. Realizing she was dealing with motivated sellers, she scooped up timber at an attractive price.

The problem for smaller or less connected funds that copy the strategies of the big endowments is that price inefficiencies last only a short while. By the time that copycats invest in a well-marketed “timber fund,” the asset is likely to be fully priced.

In fact, even with their apparent edge, the relative outperformance of large endowments has been shrinking. Over the decade that ended in 2015, these endowments barely beat the passive stock and bond blend—edging out the latter by a mere four-tenths of a percentage point. (Nacubo has yet to update results for 2016.) Harvard is a case in point. Thanks to early investments in Baupost Group and Sequoia Capital, among others, it trumps a passive blend over 20 years, but underperforms over the past 10.

Recall that the advantage of big endowments is supposed to be that sense of perpetuity: a distant investing horizon that allows managers to snatch the best long-term opportunities and ignore short-term swings. But if you read through reports of various schools, it’s hard to escape the conclusion that endowments are beset by short-termism. Although Harvard has been properly cautious with regard to introducing major changes to its asset allocation model, its managers repeatedly pledged to “tune” or “refine” or “reposition” portfolios; they fretted one year over the “unusual uncertainty in the outlook” and in another that the “landscape” was full of “uncertainty.” (To paraphrase Warren Buffett, let me know when the future isn’t uncertain.)

Zimmerman says it’s unrealistic to expect universities to completely ignore short-term pressures. “The endowment may be forever,” he tells Fortune, “but the people aren’t. I’m not. The regents aren’t.” That said, it’s now likely that many big endowments are now so über-diversified—with so many investment positions—that they fall under the spell of short-term thinking nevertheless. At some point, the more securities and investment categories one holds, the greater the temptation to realign weightings at every market twitch.

What endowments have mostly abandoned is the old-fashioned, Graham-and-Dodd business of building a portfolio by picking attractively priced individual securities. Instead, they’ve adopted the academic conceit of building portfolios from the top down, assembling “risk factors” rather than stocks. Endowment investing has been reduced to a series of calculable ­tradeoffs—thus Harvard would tweak its portfolio in 2010 to add “unique low-beta opportunities” and now searches “computationally” for a portfolio “that maximizes our asset class specific return per unit of risk.” If your head spins, it should. No one knows what a “unit of risk” is until after the fact. Even Harvard admitted that its risk assumptions were couched in “high uncertainty.”

Other schools adopted a similarly slavish respect for formulas; small, smart Oberlin College targets future returns to the hundredth of a percentage point. (Endowments truly that prescient should go into political polling.) But setting ultraprecise targets carries a serious hazard: It increases performance pressure to the point where managers are constantly adjusting their ­portfolios—the kind of return-chasing behavior that’s a surefire way to lose your edge. Nacubo, curiously, nourishes this short-term mind-set by publishing volatility and “Sharpe” ratios (a measure of risk-adjusted return). Even to track such ephemera is to encourage endowments to shrink their time horizons.

“The idea that alternative investments are an ‘asset class’ just doesn’t make sense.” —Jesse Seegmiller, chief investment officer, Southern Virginia University

One of the few schools to shun the Yale model, Southern Virginia University, has one of the best records anywhere. Jesse Seegmiller, the chief investment officer who oversees the school’s tiny $1.1 million endowment, says simply, “We have a discipline and we stick to it.” The discipline consists of bottom-up selection of a relative handful of stocks. It does not own private equity or hedge funds. “The idea that alternative investments are an ‘asset class’ just doesn’t make sense,” says Seegmiller. Timber is a commodity; hedge funds are a legal framework for investing in any of a number of asset classes. Private equity is simply equity; indeed, over the past decade the average private equity fund has merely kept pace with public equities. And hedge funds—which account for a fifth of the investments in endowment portfolios—have significantly underperformed the S&P in recent years as a result of their high fees and efforts to tamp volatility and avoid market correlation, ridiculous aims for schools supposedly investing for forever. (According to HFR, the average hedge fund gained 42% over the decade through June 2016, compared with 104% for the S&P.)

What’s Southern Virginia’s annualized return over the past decade, you ask? That would be 10.3%—little short of exceptional.

In truth, many academic institutions—small and large—would be best off investing by way of low-fee index funds, as Yale’s Swensen has argued the average small investor should do. But then few people think of themselves as average, and few endowment managers do either. If only they knew.

This is part of Fortune’s 2017 Investor’s Guide, which offers advice about the best strategies for next year.

A version of this article appears in the December 15, 2016 issue of Fortune.