祸兮福所倚

“It’s An Ill Wind That Blows No Good” How The Mutual Fund Scandals Will Serve Fund Owners Keynote Speech by John C. Bogle, Founder and Former CEO The Vanguard Group Before the Society of American Business Editors and Writers Personal Finance Workshop Denver, Colorado October 27, 2003 When I conclude my remarks today, I hope that you’ll have a better understanding of what today’s mutual fund industry is all about—that you’ll see our industry as it is, and not as virtually every industry leader sees it. The gift, in the words of Robert Burns, “to see ourselves as others see us.”

In the autumn of 1996, the last time that I had the privilege of addressing you, I began my remarks by reminding you that I’d just had “a change of heart.” As most of you seemed to know, that was a literally correct statement. Only seven months earlier, I was the beneficiary of a heart transplant, a miracle that gave me a second chance at life. I’m now having my eighth extra year, and I assure you that my new heart is still in the same place that it was when I spoke to you in 1996.

Then, I warned you that the “spirit of trusteeship, professional competence and discipline, and a focus on the long term, are rapidly losing their role of the driving force—in the long run, the life force of this industry.” Today, the three principal points I made seem almost prescient:  “The industry’s traditional focus on trusteeship, implying placing the interest of fund shareholders as our highest priority and charging a reasonable price for our services, is being supplanted by a focus on asset-gathering—on distribution—as we worship at the shrine of the Great God Market Share, the exorbitant cost of which is borne by our own fund shareholders.

 “The industry’s traditional focus on professional competence and discipline has moved from long-term investment to what is really speculation, with rapid turnover in our investment portfolios (averaging almost 100% per year!), funds concentrating on ever-narrowing segments of the stock market, and far too many gunslinger portfolio managers.

 “And the industry’s traditional focus on the eminent suitability of mutual funds for long-term investors is quickly becoming a focus on investing in fund portfolios for the short-term (a second level of speculation) and, even more baneful, a focus on enticing fund shareholders to use their mutual funds as vehicles for rapid switching, either for the purpose of market timing or for the purpose of jumping on the bandwagon of the latest hot fund (and that’s called speculation, too).”

Shocked, Shocked What we now know, of course, is that the consequences of these three baneful trends have come home to roost in the most painful sort of way: in damage done to the pocketbooks of the shareholders who placed their trust in mutual funds.

The recent market timing scandals are but a midget manifestation of the problem. But the industry’s response can be best characterized by a classic line spoken by the police chief in the film Casablanca, Claude Rains: “I am shocked, shocked to find (timing) going on here.” We’ve already been told that the misdeeds are akin to “parking at a meter and not paying. Nobody is being bankrupted by this.” And we’ll doubtless be told (if we haven’t already been told) that these breaches of fiduciary duty are attributable to only “a few bad apples,” although as these scandals continue to come to light, we may need to liberalize our definition of “few.”

Even as the spotlight that shined on the specific acts that brought notoriety to corporate America’s bad apples—the Ken Lays, the Dennis Kozlowskis, the Sam Waksals, the Jack Welches, the Richard Scrushys, to name just a few—illuminated all the nibbling around the edges of proper and ethical conduct that, absent the intrusive spotlight, could otherwise have persisted for another decade or more, so the spotlight that shines on the scandals perpetuated by the bad apples of the mutual fund industry reflect the frequent willingness—nay, the eagerness—of fund managers to build their own profits at the expense of the fund owners whom they are honor bound to serve.

“It’s an ill wind that blows no good.” By illuminating the inherent conflict of interest between fund managers and fund investors, these scandals will ultimately prove a blessing for fund owners. This conflict is hardly a secret. Indeed in that very 1996 speech, I urged this industry to move to a system in which “the focus of mutual fund governance and control is shifted . . . to the directors and shareholders of the mutual funds themselves, and away from the executives and owners of mutual fund management companies (where it almost universally reposes today), who seek good fund performance to be sure, but also seek enormous personal gain.”

If such a shift of control and governance had taken place, the market-timing scandals detailed in the Spitzer-Canary settlement may well never have occurred. The Attorney General’s seemingly airtight case was built, not only on covert practices, but on open motivations—on the receipt of payola, for the want of a better word. The managers received that payola in the form of side banking deals, lending money at high interest rates; large investments in other funds on which the manager earns high fees (“sticky assets” in the vernacular of the trade); and the like.

One manager’s e-mail could hardly have made the motivation clearer. “I have no interest in building a business around market timers, but at the same time I do not want to turn away $10-$20m[illion]!” (Yes, the exclamation point was there.) The writer emphasized that allowing the timing trades would be in the manager’s “best interests.” Lest his colleagues be complete nincompoops and fail to get the point, he explained in a parenthetical aside what that meant: “increased profitability to the firm.” Another e-mail (God bless e-mail!) also told the truth: “Market timers are a big problem . . . it’s very disruptive to the operation of the funds. (But) obviously, your call from the sales side.”

A Study in Corporate Incest It can be little surprise that the mutual fund industry has not escaped the same kinds of scandals that have faced Wall Street and Corporate America. For in no other line of business endeavor is the conflict between owners capitalism and managers capitalism more institutionalized, and therefore more widely accepted. Yet by its very structure, this industry, for all its protestations about its dedication to Main Street investors, seems almost preordained to give the managers total control over the fund shareowners.

Consider how the typical fund organization operates. Even when their assets are valued in the scores of billions of dollars, fund complexes do not manage themselves. They hire an external management company—with its own separate set of shareholders—to manage their affairs. The management company runs the fund’s operations, distributes its shares, and supervises and directs its investment portfolio. It decides when to create new funds, and it decides what kinds of funds they will be. When the funds are badly run, the company replaces the portfolio manager . . . but with one of its own employees. And when a fund outlives its usefulness, it is the management company that decides how to dispatch it to its well-deserved reward: simply liquidating it, or, much more likely, merging it into another fund with a better past record. . . but a fund that it also just happens to manage.

What’s more, this typical management company graciously provides all of the fund’s officers, who are employees of the company, not the fund. And while the executives of the manager usually have a miniscule investment in the funds they run, they select themselves for the fund board, and until recent years, also selected most of the funds’ “independent” directors, who by law must now compose at least a majority of the board. In the typical case, furthermore, the chairman of the board of the management company also serves as, you guessed it, the chairman of the board of the mutual funds.

Given the Gordian knot on the rope that binds the fund and the manager together, it is impossible to imagine that at one of the fund’s four annual board meetings the less-well-informed independent directors can stand up to the steeped-in-the business management company minority. Small wonder that an early law review article about this industry’s structure was, as I recall, entitled: “Mutual Funds: A Study in Corporate Incest.”

The Emperor’s Clothes How can it be that the industry takes this bizarre governance structure as the natural order of things? How is it that its leaders couldn’t see that this structure was an accident waiting to happen? It must have something to do with what Hans Christian Andersen wrote about in 1837 in The Emperor’s Clothes: “When the little child said ‘But he has nothing on,’ and the whole people agreed, the emperor shivered, for they were right. But he thought ‘I must go through with this procession.’ And he carried himself still more proudly, and the chamberlain held on tighter than ever, and carried the train, which did not exist at all.”

But the ability to ignore the reality of our industry’s existence goes back even further than that. Hear Descartes in 1650: “A man is incapable of comprehending any argument that interferes with his revenue.” And even 1000 years before that, in 350 B.C., hear Demosthenes: “Nothing is easier than self-deceit. For what each man wishes, that he also believes to be true.”

The fund industry, in the vernacular of the day, “just doesn’t get it.” So I urge you journalists not to be persuaded by the self-aggrandizing comments offered by the Investment Company Institute’s president to the audience at this year’s General Membership Meeting: “Your unshakable commitment to putting mutual fund shareholder interests first has served our shareholders and our companies well. In a nutshell, we have succeeded because the interests of those who manage funds are well aligned with the interests of those who invest in mutual funds.”

But the interests are not well aligned. And to ignore—indeed, to deny—the obvious and profound conflicts that are manifest in this industry is hardly the beginning of wisdom. The fact is that whatever alignments of interest may exist are far outweighed by the misalignments. Consider just four of the major areas in which what is good for the managers is bad for the shareholders: 1) Market timing, which brings in temporary assets that provide higher fees to managers, but only at the cost of dilution in the returns for fund owners.

2) Management fees, which are—unarguably—inversely related to fund performance. The higher a fund’s management fees and expenses the lower the returns earned by its shareowners.

3) Growth in a fund’s assets to elephantine size, which enriches managers but destroys the fund’s ability to repeat the performance success that engendered that very growth. The bigger the fund, the bigger the fee, and the more likely the fund’s reversion to the market mean.

4) The industry’s marketing focus, which seems inevitably to demand the creation of new and often highly specialized funds to meet the heated investment passions of the day, creating huge capital inflows, huge fees for managers, and—far more often than not—huge losses to investors.

1. Market Timing Becomes Rife In view of its topicality, the market timing issue is the first obvious conflict of interest I’ll discuss. I’ve already laid bare the obvious conflict in the “late trading” scandal, the brazenness of which astonished even an industry reformer like me. But late trading is only the small tip of a big iceberg. “Time-zone trading” is likely even larger in its negative impact on fund shareholders. Yet the shocking thing about time-zone trading—usually, taking advantage of a free (to the timer!) arbitrage between an international fund net asset value calculated at 4 PM in New York, but based on closing prices across the Pacific 14 hours earlier—is that it has been going on for so long, without significant defenses being erected by managers. It has hardly been a secret; academics have been publishing papers about it at least since the late 1990s.

A prescient article in the Financial Analysts Journal1 carefully described the time-zone trading strategy, quantified its effectiveness, and showed, with specific examples, how easy it was to make money by gaming the system. It also berated the industry for its benign neglect of the market-timing issue: “When the gains from these strategies are matched by offsetting losses incurred by buy-and-hold investors in these funds . . . why haven’t more funds taken stronger actions to restrict short term trading?” What is more, the four authors cited fully 20 other academic studies on the same point, and, especially prescient the light of the Canary hedge-fund settlement, noted that 30 hedge funds had blatantly listed their investment strategy as “mutual fund timing.” If industry participants were fast asleep before, that article sounded the alarm, and it surely answered the question: “What did we know and when did we know it?” Fund shareholders, if not fund managers, owe these four academics a major debt of gratitude.

Yet the sole published response to the revelation was a screed from a representative of the manager whose funds were mentioned in the article. He berated the Journal: “Publishing such a piece in a publication that is aimed solely at financial professionals is a bad idea in the best of times, but is abhorrent when investor confidence is already shaken by corporate greed . . .” Nonetheless, just nine months later, the very firm that employed the respondent initiated a 2% redemption fee on its international funds. At long last!

“General” Market Timing in Funds But general market timing—not the illegal late trading, not the unethical time-zone trading— suggests that investors, using the finest vehicle for long-term investing ever designed, are doing too much short-term speculation in mutual funds. There’s a lot of money sloshing around the mutual fund system.

“Stale Prices and Strategies for Trading Mutual Funds,” by NYU Professors Boudonkh, Richardson, Sulrahmanyam, and Whitelaw. The Financial Analysts Journal, July/August 2002.

How much market timing is there? We simply don’t know. But we do know a great deal about what is going on.

First, there is much more timing activity than the industry acknowledges. By failing to acknowledge that redemptions whose proceeds are invested in another fund within the same family—so-called “exchanges out”—are actually, well, redemptions, the ICI substantially understates fund redemption rates. Such intra-family redemptions are the clearest—though hardly the only—example of a market timing strategy; i.e., frequent moves back and forth between a stock fund and a money market fund in the same family. While the ICI reported an equity fund redemption rate equal to 29% of assets in 2002, the actual rate, including exchanges-out, was 41%, half again higher.

Next, we also know the (true) redemption rate has soared—from 5% to 15% in the 1950s through the mid-1970s, to the 30%-35% range into the late 1990s (excepting a 60% rate in the turbulence of 1987), and to the 40%-50% range thereafter. The average fund investor, who not all that long ago held fund shares for an average of more than ten years (the reciprocal of, and proxy for, a 10% redemption rate), now holds shares for less than two and one-half years (proxy for a 41% redemption rate).

Redemption Rates Soar 50% Investor Turnover of Equity Fund Shares 41% 40% Actual 33% Reported by ICI 31% 30% 20% 11% 10% 5% 0% 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 00 Robust Timing by the Minority Interestingly, industry studies of investor behavior show that the typical (i.e., median) fund investor doesn’t do much trading. During 1998, according to an ICI survey, fully 82% of equity fund owners made not a single redemption. Even if that figure is accurate (and assuming that the ratio holds for 2002), however, consider what it implies: the 41% total redemption rate, spread over only the remaining 18% of investors, indicates that this small segment of investors has an average holding period of just 160 days. And if we assume, arbitrarily, that one-half of these investors maintained, say, a ten-year holding period, the remaining half would have an average holding period of about 90 days—a redemption rate of an astonishing 446%.

When we look at fund objectives, it’s easy to see what is going on. International funds win the prize, with an average redemption rate rising from 94% a year in 2000 to 97% in 2002. During the same period, sector funds redemptions averaged 57%, and aggressive growth funds 51%. While the ICI’s major understatement of redemptions requires us to do our own calculations, these data are all actually reported, so it takes little effort to observe where the worst investor behavior is going on.

What is more, the annual report of each mutual fund is required to report total redemptions. It is a revelation to examine some of the funds involved in one aspect or another of the recent timing scandals, where the numbers approach the brazen. The Alger equity funds, with total assets averaging $2 billion in 2002, reported redemptions for the year totaling $9 billion(!)—a 440% redemption rate. Bank of America’s Emerging Markets Fund had a 295% annual redemption rate, and Janus Adviser International Growth fund had a 372% redemption rate.

The dollar amount of redemptions (but not the turnover rate) are clearly set out in each fund’s financial statement without comment, and sent to shareholders (and, we must assume, to directors as well). This redemption activity, then, is not only going on with the tacit knowledge of the managers, directors, and regulators, it is happening right under the noses of the shareholders, the press, and the public as well, fully disclosed for anyone interested enough to look. Yet I have never seen it questioned or challenged.

The solution to the problems of excessive market timing by fund traders is straightforward: 1) Close the funds’ transaction window at 2:30 PM instead of 4:00 PM for everybody. If the assets for 401(k) plans can’t meet the deadline, they’ll just have to execute the orders on the next day. 2) Impose a redemption fee of 2% for shares held for less than, say, 30 days. Alas, with the fierce competition to attract assets, few firms will have the courage to take these two steps on their own. It would cost them business! So, I urge the Securities and Exchange Commission to impose these standards on this reluctant industry. I also urge a firm like Morningstar to regularly publish and comment on the redemption rates of individual funds. As it almost invariably does, the sunlight of disclosure would quickly modify the behavior of both managers and traders.

2. The Conflict of Interest in Fund Fees I have no trouble in postulating that both the fund directors and the management company share a common interest in providing good returns to the fund shareholders. But when it comes to how good, their interests diverge. Why? Simply because the higher the management fees and other fund expenses, the lower the fund’s return.

Sometimes, this relationship exists on a virtual dollar-for-dollar basis. For example, the correlation between the yields that money market funds deliver to their shareholders and the expense ratios of these funds is an inverse –0.98, almost, well, perfect. When money market yields are 3%, for example, a high-cost fund will deliver as little as 1¾% to its owners; a low-cost fund will deliver as much as 2¾%—fully 50% more. Indeed, whenever fund gross returns are commodity-like (for example, in stock index funds and bond index funds), the same kind of “locked-in” relationship of returns to costs prevails.

But even in actively managed funds, costs clearly differentiate the superior performers from the inferior performers over the long run. Consider a study we recently prepared quantifying the relationship between the total costs of equity funds and their returns. Using all 803 diversified U.S. equity funds in the Morningstar database in existence over the full ten-year period ended August 30, 2003, we compared each fund’s investment returns with its costs. The average expense ratio for these funds was 1.3%, and their average portfolio transaction costs were estimated at 0.7%, for a total of 2.0%. (We conservatively assumed that transaction costs totaled 1% of turnover, equal to only ½% on each side of the trade.)

Results? The high-cost quartile of funds, with all-in expenses of 3.4%, provided an average annual return of 6.8%.2 The low-cost quartile, with expenses of 1.0%, provided an average annual return of 10.2%, earning an advantage of 3.4 percentage points per year. On a fund-by-fund basis, the inverse correlation between cost and return was remarkable: minus 0.60%. So yes, cost matters.

Cost Matters!

10-Year Returns, All Equity Funds 12% Low-Cost Quartile 10.2% 10.3% High-Cost Quartile 10% 8.3% 8% 6.8% 6.5% 6% 4.3% 4% 2% 0% Actual Return* Risk-adjusted Return After-tax Return* *Source: Morningstar. Avg. Ann. Returns for 10 yrs ended 8/03.

What’s more, the funds with the highest costs also assumed the highest risks (a standard deviation 30% higher than the lowest-cost funds); generated the highest turnover (160% vs. 22%); and produced the poorest tax-efficiency. As a result, the low-cost group had an even greater advantage (3.8% per year) in risk-adjusted return, and an amazing advantage of 4.0% per year in after-tax return. It’s hard to imagine presenting a more persuasive case about the relationship between fund costs and fund returns.

Reinforcement of the Low-Cost Thesis But, of course, I’ll present a more persuasive case anyway. For when we sort the funds into their nine Morningstar style boxes, the consistency of the performance margin (even without risk-adjustment and tax-adjustment) was little short of astonishing. The low-cost quartile provided a consistent edge in the remarkably narrow range of 2.3% to 2.7% per year among the nearly 500 large-cap funds, and in two of the three mid-cap styles. In the remaining (smaller and therefore less statistically reliable styles; the small-cap value group had a total of only 28 funds), the excess returns achieved by the low-cost funds were even higher (averaging 5.6% per year). With this reinforcement from the segment data, it is simply impossible to argue that the link between lower costs and higher return isn’t about as strong as the suspension cables on the Golden Gate Bridge. Cost matters, and it matters everywhere.

We omitted the impact of initial sales charges, and few of these established funds have 12b-1 fees, so the expense ratios of this select group were significantly below industry norms. Further, since we made no adjustment for survivor bias, the average return was also overstated.

Cost Matters Everywhere!

10-Year Returns: Low-Cost vs. High-Cost Funds 14% Value Blend Growth +2.3% +2.7% +2.5% 10% 6% Large 2% LC HC LC HC LC HC 14% +2.5% +4.3% +2.4% 10% 6% Mid 2% LC HC LC HC LC HC 14% +3.6% +6.9% +7.7% 10% 6% Small 2% LC HC LC HC LC HC The establishment of a fair cost structure—including management fees, portfolio turnover expenses, operational expenditures and sales loads—must be the categorical imperative of the fund board. Given the circumstances of fund governance, however, the directors affiliated with a management company have a compelling interest in the reverse. They seek the highest fees that public opinion and traffic will bear. The manager not only places a high priority on its own profitability; but it’s arguable that it has a fiduciary duty to its own shareholders to do exactly that. And the independent directors seem reluctant to challenge that interest. (Hear Warren Buffett: “When the managers care about fees and the directors don’t, guess who wins? Negotiating with oneself seldom produces a barroom brawl.”) Fee negotiation is a myth, and the fund shareholders suffer, not only accordingly, but as the data make clear, both measurably and substantially.

3. Let the Asset Growth Roll It must be obvious that as the assets of actively-managed funds grow, the challenges of implementing their investment strategies increase. And when the assets grow exponentially, so too do the challenges. The number of stocks available for the portfolio manager to choose from shrinks, and portfolio transaction activity tends to either become more expensive as average trade size increases or, for better or worse, to diminish. Further, as the fund grows, investment returns have a powerful tendency to revert to the market mean—three negatives, at least for the investors who chose the fund because of its demonstrated ability to generate superior returns. But exceptional growth generates exceptional fees, and the managers are enriched accordingly.

Yet it is only in rare cases that managers summon the courage to close funds to investors, which suggests that the pressures to let funds grow beyond their ability to be effectively managed exist. Fund closings are the exception that proves the rule: Only ten of the 3,363 domestic equity mutual funds in existence today have completely closed, although another 127 are closed to new investors. In an industry where average assets of the fifty largest equity funds have burgeoned from $4.6 billion to $23.3 billion in a decade, why have the closings been so rare? It seems reasonable to assume that it is because the manager’s interest in ever-higher fees carries the day, and outweighs the shareholder’s interest in sustaining superior returns.

Let’s consider a single (extreme!) example of how the interaction of these trends works in practice. I’ll call it Fund X, (although its actual name will be no mystery to you). During its early years, it turned in an astonishingly successful record, outpacing the Standard & Poor’s 500 Stock Index by an average of 26 percentage points per year from 1978 through 1983. With such success, its assets burgeoned from a mere $22 million to $1.6 billion during that period. While its performance then reverted toward the mean, its excess return from 1984 through 1993 remained a healthy four percentage points per year. By then, its assets had grown to a staggering $31 billion, and the excess returns came to an abrupt halt. Four years of losing to the S&P followed, and then three small gains and two small losses. Since 1993, it has fallen an average of almost two percentage points per year behind the 500 Index—a far cry from the success of its earlier years.

With soaring management fees leading the way, the expenses borne by the shareholders of Fund X kept growing. And growing, and growing. From $400,000 in 1978 to $17 million in 1984, to $166 million in 1991, and $500 million in 1996, expenses peaked at $763 million in 2001. At the outset of the period, small fees for large returns. At the period’s conclusion, awesome fees for mediocre returns. Obviously, the fund’s asset growth was wonderful for its managers, but the exact opposite was true for its owners.

4. The Marketing Focus—We Make What Will Sell A recent article3 in The New Yorker described Hollywood as exemplifying, “the most joyless aspects of capitalism. The ‘industry,’ as it insists in calling itself, packages ideas and images as ‘products,’ and then values them according to how they ‘penetrate’ markets, support ‘platforms’ of ancillary products, and ‘brand’ a company as a reliable purveyor of similar products.” If that makes Hollywood sound like what the mutual fund industry has become, you are paying attention to the march of our history.

When I came into this business 52 years ago, fund management companies were relatively small, privately-held professional firms, and we focused on stewardship. Marketing had yet to rear its ugly (in the context of fund management) head. Those managers provided their services to just 75 mutual funds, of which 66 were essentially what today we would call “large-cap blend funds,” holding a widely diversified portfolio of blue-chip stocks and providing returns that generally paralleled the returns of the stock market itself, as measured by the Standard & Poor’s 500 Index. Most of these fund managers ran but a single stock fund. In short, we sold what we made. We sold what we made.

What a difference a half-century makes! Today, the industry is dominated by giant financial conglomerates that focus on salesmanship. There are 3,650 equity funds large enough to be tracked by Morningstar, and only 560 of them closely resemble their blue-chip forebears. What are these other funds? 1,890 funds are diversified equity funds investing in one of the eight remaining “style boxes,” that make bets away from the total market—bets on large-cap growth stocks, or small-cap value stocks, and mid-cap blend stocks. Another 450 funds invest in specialized industry segments—technology, telecommunications, and computers, are (or were!) the most popular examples. And 750 are in “international” funds, an odd locution that applies largely to funds that invest in foreign markets (albeit sometimes with a seasoning in U.S.-based companies).

In general these other fund categories assume higher risks than the market-like funds of yore. In 1951, an investor could throw a dart at the (tiny!) fund list and have nine chances out of ten of picking a fund whose returns would parallel the return of the market itself. Today, the investor’s chances of doing so are just one out of eight! For better or worse, selecting mutual funds has become an art form, and “choice” rules the day.

What a Difference a Half-Century Makes 1949 2002 Number of Equity Funds* Large-Cap Blend 66 560 Large-Cap Blend Specialized 9 1,890 Other Div. Equity 450 Specialized 750 International Total 75 3,650 Total *2002 total is number covered by Morningstar “Penetrating Markets” The fund industry has become a business school case study in marketing—packaging new ancillary products in order to penetrate new markets and to expand penetration of existing markets. Modern marketing has played a major role in the burgeoning profitability of investment managers, and, in that sense, it has worked. The hundreds of billions of dollars poured into these “new products,” along with appreciation in the value of “existing products” during the great bull market, created a bonanza for fund managers. From 1980 to 2002, total mutual fund assets rose 60 times over—from $115 billion to about $7 trillion. Yet despite the staggering economies of scale in this industry,4 fund management fees and expenses rose far faster—90 times over, from $800 million to $72 billion.

Where are the Economies of Scale?

Fund Assets vs. Fund Expenses, 1980 and 2002 Bil Bil $8,000 Fund Assets $80 Fund Expenses $72 $6,700 Increase: 60x Increase: 90x $6,000 $60 $4,000 $40 $2,000 $20 $115 $0.8 $0 $0 1980 2002 1980 2002 Why? Because rather than creating sound investment choices, our Great God Market Share, to repeat an earlier phrase, demanded that we create funds that the investing public wanted to buy. And what the public wanted to buy—and was willing to pay higher fees for—was the hot idea of the day. In the late bubble, of course, it was the “new economy” internet funds and technology funds and telecommunications funds, and the aggressive growth funds that concentrated in those stocks. Indeed, these risky sectors also dominated the portfolios of even the more diversified traditional growth funds. The public had little interest in the more sedate value funds. So we created these risky new funds, promoted them, and sold them. Why? We made what would sell. We made what would sell.

The trends are easily measured: From 1998 through 2000, the public bought $460 billion(!) of high price-earnings-ratio growth funds, at ever ascending prices and redeemed a net total of $100 billion in lower price-earnings ratio value funds. Then, after the market neared its lows, investors switched gears, and in 2001-2002, these growth funds experienced net redemptions of $46 billion, and value funds took in $89 billion of additional capital—proving, once again, that this is a market-sensitive industry.

“Remake Man” by Tad Friend; June 2, 2003. The word “products” read “commodities” in the original article. The average expense ratio of the Vanguard funds, which are operated on an “at cost” basis, declined 54% during the same period, from 0.59% to 0.27%.

Managers Win, Investors Lose This sensitivity worked to the advantage of fund managers. That huge flow of additional capital to new-economy-oriented funds produced some $30 billion(!) of additional management fees and costs during 1998-2000, and, even though these costs tumbled as the bubble burst, an additional $20 billion during 2001-2002—total revenues of $50 billion, accompanied by only modest incremental expenditures by the managers. The focus on marketing was a remarkably profitable strategy for managers.

Investor Returns, For Funds, for Shareholders But not so for investors. That same marketing strategy cost our shareholders hundreds of billions of dollars. Aided and abetted by the aggressive sales promotion of the managers, investors moved their money into the most vulnerable areas of the market and withdrew money from the least vulnerable areas, as we now know, precisely the reverse, of what they should have been doing. And we now have the tools to recognize just how badly these investors fared. For in my 1996 speech, I talked to you about the need for mutual funds to report not only their “time-weighted” returns (our standard measure for the return a mutual fund earns on each share), but their “dollar-weighted” returns (the measure of what the fund earns for its shareholders as a group).

While that suggestion for a second measurement never materialized, it is instructive to consider the dollar-weighted returns earned by the shareholders of a whole variety of the largest and most popular growth funds during the past five years. Roughly speaking, the $460 billion that investors poured into this group of “new-economy”-oriented funds resulted in a loss of some $300 billion in the decline that followed.

It’s fair to say that (a) fund managers made huge profits by their willingness, indeed eagerness, to make what would sell, and (b) fund investors absorbed huge losses from (literally) buying into that strategy. Put another way, the change in the mutual fund industry, in which the managers have come to consider money management a business focused on their desire for profits, rather than a profession focused on the interests of their fund shareholders, could itself well be considered a certain kind of scandal.

Summing up: Earning a Fair Share of Stock Market Returns?

What has been described as “a pathological mutation”5 in corporate America has transformed traditional owners capitalism into modern-day managers capitalism. In mutual fund America, the conflict of interest between fund managers and fund owners is an echo, if not an amplification, of that unfortunate, indeed “morally unacceptable”5 transformation. The blessing of our industry’s market-timing scandal— the good for our investors blown by that ill wind—is that it has focused the spotlight on that conflict, and on its even more scandalous manifestations: the level of fund costs, the building of assets of individual funds to levels at which they can no longer differentiate themselves, and the focus on selling funds that make money for managers while far too often losing money—and lots of it—for investors.

The net results of these conflicts of interest is readily measurable by comparing the long-term returns achieved by mutual funds, and by mutual fund shareholders, with the returns earned in the stock market itself. You may have seen my work on these returns earlier in the year, but I’d like to take this opportunity to evaluate the data and expand on the reasons why they are what they are.

William Pfaff, writing in The International Herald-Tribune on September 9, 2002.

During the period 1984-2002, the U.S. stock market, as measured by the S&P 500 Index, provided an annual rate of return of 12.2%. The return on average mutual fund was 9.3%.6 The reason for that lag is not very complicated: As the trained, experienced investment professionals employed by the industry’s managers compete with one another to pick the best stocks, their results average out. Thus, the average mutual fund should earn the market’s return—before costs. Since all-in fund costs can be estimated at something like 3% per year, the annual lag of 2.9% in after-cost return seems simply to confirm that eminently reasonable hypothesis.

But during that same period, according to a study of mutual fund data provided by mutual fund data collector Dalbar, the average fund shareholder earned a return just 2.6% a year. How could that be? How solid is that number? Can that methodology be justified? I’d like to conclude by examining those issues, for the returns that fund managers actually deliver to fund shareholders serves as the definitive test of whether the fund investor is getting a fair shake.

The Stock Market, Funds, & Fund Owners 1984 - 2002 100% 12.2% Annual Return Market Return 9.3% Fund Return 75% Investor Return 50% 25% 2.6% 0% The Cost The Timing and Penalty Selection Penalty Is the Dalbar Study Accurate?

Let’s begin by using some common sense. It is reasonable to expect the average mutual fund investor to earn a return that falls well short of the return of the average fund. After all, we know that investors have paid a large timing penalty in their decisions, investing little in equity funds early in the period and huge amounts as the market bubble reached its maximum. During 1984-1988, when the S&P Index was below 300, investors purchased an average of just $11 billion per year of equity funds. They added another $105 billion per year when the Index was still below 1100. But after it topped the 1100 mark in 1998, they added to their holdings at an $218 billion(!) annual rate. Then, during the three quarters before the recent rally, with the Index below 900, equity fund investors actually withdrew $80 billion. Clearly, this perverse market sensitivity ill-served fund investors.

Lipper data show that the funds that were in business throughout the period earned an annualized return of 9.8%. We estimate that survivor-bias reduced those returns by at least 0.5% per year, to 9.3%. Even that number overstates the fund record, because it ignores the impact of sales charges.

The Dalbar study calculates the returns on these cash flows as if they had been invested in the Standard & Poor’s 500 Index, and it is that simple calculation that produces the 2.6% annual investor return. Of course, it is not entirely fair to compare the return on those periodic investments over the years with initial lump-sum investments in the S&P 500 Stock Index and in the average fund. The gap between those returns and the returns earned by investors, then, is somewhat overstated. More appropriate would be a comparison of regular periodic investments in the market with the irregular (and counterproductive) periodic investments made by fund investors, which would reduce both the market return and the fund return with which the 2.6% return has been compared.

But if the gap is overstated, so is the 2.6% return figure itself. For investors did not select the S&P 500 Index, as the Dalbar study implies. What they selected was an average fund that lagged the S&P Index by 2.9% per year. So they paid not only a timing penalty, but a selection penalty. Looked at superficially, then, the 2.6% return earned by investors should have been minus 0.3%.

Worse, what fund investors selected was not the average fund. Rather they invested most of their money, not only at the wrong time, but in the wrong funds. The selection penalty is reflected by the eagerness of investors as a group to jump into the “new economy” funds, and in the three years of the boom phase, place some $460 billion in those speculative funds, and pull $100 billion out of old-economy value funds—choices which clearly slashed investor returns.

The Timing Penalty: The Selection Penalty: Equity Fund Cash Flow Follows the Stock Market 1600 Quarterly Flow into Growth and Value Funds, Net Flow 1400 (bil) and the Nasdaq’s Close Nasdaq $160 5000 $150,000,000 1200 $140 Growth 4500 Total Flow $130,000,000 $120 Growth: $421b Value 4000 $110,000,000 Net New Cash Flow $100 Value: ($10b) Nasdaq Close $90,000,000 S&P 500 800 $80 $70,000,000 $60 600 2500 $50,000,000 $40 400 2000 $30,000,000 $20 $10,000,000 200 $0 -$10,000,000 -$20 1000 -$30,000,000 -$40 500 -$50,000,000 -$60 0 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1'98 Q3'98 Q1'99 Q3'99 Q1'00 Q3'00 Q1'01 Q3'01 Q1'02 Q3'02 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 Source: Strategic Insight Dollar-Weighted Returns: How Did Fund Investors Fare?

Now let me give you some dollars-and-cents examples of how pouring money into the hot performers and hot sector funds of the era created a truly astonishing gap between (time-weighted) per-share fund returns and (dollar-weighted) returns that reflect what the funds actually earned for their owners. So let’s examine the astonishing gap between those two figures during the recent stock market boom and subsequent bust.

Consider first the “hot” funds of the day—the twenty funds which turned in the largest gains during the market upsurge. These funds had a compound return of 51% per year(!) in 1996-1999, only to suffer a compound annual loss of –32% during the subsequent three years. For the full period, they earned a net annualized return of 1.5%, and a cumulative gain of 9.2%. Not all that bad! Yet the investors in those funds, pouring tens of billions of dollars of their money in after the performance gains began, earned an annual return of minus 12.2%, losing fully 54% of their money during the period.

Now consider sector funds, specific arenas in which investors can (foolishly, as it turns out) make their bets. The computer, telecommunications, and technology sectors were the favorites of the day, but only until they collapsed. The average annual returns of 53% earned in the bull market by a group of the largest sector funds were followed by returns of minus 31% a year in the bear market, a net annual return of 3% and a cumulative gain of 19.2%. Again, not too bad. Yet sector fund investors, similar to the hot fund investors I described earlier, poured billions of dollars in the funds as they soared, and their annual return averaged –12.1%, a cumulative loss of 54% of their capital, too.

How Selection and Timing Can Destroy Investor Returns Dollar-Weighted Time-Weighted Returns Returns Avg. Ann. Return Cum. Return Cum. Return 97-99 00-02 97-02 97-02 Hot Funds 51.4% -32.0% 9.2% -54.0% 5 Major Sector Funds 53.3% -30.8% 19.2% -54.0% S&P 500 Index Fund 27.5% -14.6% 29.2% 15.3% While the six-year annual returns for these funds were hardly horrible, both groups did lag the 4.3% annual return of the stock market, as measured by the largest S&P 500 Index Fund, which provided a 29% cumulative gain. But the investors in that index fund, taking no selection risk, minimized the stock market’s influence on their timing and earned a positive 2.4% return, building their capital by 15% during the challenging period. Index investor +15%; sector fund and hot fund investor –54%. Gap: 69 percentage points. It’s a stunning contrast.

Given these caveats about methodology, and these actual examples of how shareholders actually invested their dollars, the 2.6% return cited by Dalbar certainly overstates the annual return earned by the typical fund shareholder during a period in which the stock market return of 12.2% was virtually there for the taking. But let’s assume that their figure is accurate, and wrap up the issue by calculating the cumulative compound return earned during the period: $1000 invested in the S&P Index at the outset would have grown by $7,900, $1000 in the average fund would have grown by $4,400; and $1000 for the fund shareholder would have grown by just $600—less than it would have grown in a savings account. No, most fund investors have not been given a fair shake.

The Stock Market, Funds, & Fund Owners 1984 - 2002 100% $7,900 Profit on $1,000 Market Return 75% Fund Return Investor Return $4,400 50% 25% $600 0% The Cost The Timing and Penalty Selection Penalty It is the myriad conflicts between the interests of fund managers and the interests of fund owners that exist in this industry that bear so much of the responsibility for this staggering gap between the stock market’s return and the returns earned by fund investors, and even the returns earned by the funds themselves. While the unacceptable, and partly illegal, market timing scandal has gained a great deal of well-deserved attention, it pales in significance when compared with the powerful impact of high costs on reducing fund returns, on the force of fund size in diminishing fund returns, and on the marketing focus that tempted too many investors to purchase funds that they now wish they had never bought.

That conflict is severe, and it is unacceptable. It can be resolved only by implementing reforms in fund structure that create a governance model that puts the shareholders in the driver’s seat, where all those years ago the Investment Company Act of 1940 insisted that they belong. Such structural reform must be our highest priority, and the sooner we get about the task, the better.

Note: The opinions expressed in this speech do not necessarily represent the views of Vanguard’s present management. © Copyright 2005 by John C. Bogle