Quarter 2, 1998
by Peter Fortune

From December 1994 through December 1997, stock prices surged. The Standard & Poor's 500 index rose at an annual rate of 28 percent, well above the 10.3 percent average increase that took place between 1989 and 1994. A dollar invested in the S&P 500 in December 1989 would have accumulated to $3.49 eight years later, with almost three-quarters of that increase occurring after December 1994.

The acceleration in stock price increases has been accompanied by rapid growth in stock mutual funds. Net new money flowing into stock mutual funds totaled $917.4 billion between 1990 and 1997. Of this total, 61 percent arrived after 1994.

This rapid growth of money flows to mutual funds has attracted considerable interest. Stock market forecasters use these flows as an indicator for future stock price changes, arguing that money put into a stock fund today is stock purchased by the fund tomorrow. Economists refer to these flows when considering the potential for massive redemptions and the possibility that a reversal might destabilize the stock market. Public policy analysts express concern about the ability of the infrastructure serving the mutual fund industry to handle the order flow volume in a major market downturn, and about the implications for the size of the stock price decline.

For more than two years, I have been engaged in a study of these questions. Over the course of this research, I have identified a short list of ideas about mutual funds that are au courant, but which my work suggests should be discounted. While definitive answers on these ideas are not yet in, I call them "myths" because there is so little evidence to support them.

Myth 1: Flows into mutual funds are driving up stock prices

How should one interpret the simultaneous surge in stock prices and mutual fund assets? One commonly voiced hypothesis is that flows into stock funds push up stock prices. For example, a recent Wall Street Journal article argues, "Most of the torrent of money has been going, as usual, into stock funds rather than bond funds, and creating a kind of feedback effect: the roaring stock market attracts more cash from investors, and that helps the market soar further." This feedback scenario would suggest that one should view stock funds as contributors to stock price instability and, perhaps, as a threat to the impressive record of economic growth enjoyed in the 1990s.

In spite of the frequency and firmness with which this view is expressed, its validity is only in the mind of the beholder. A look at the historical record strongly suggests we should reject this notion for several reasons. First, much of the increased money flows to stock funds has come from sales of directly held shares of stock. Asset allocation decisions, once handled by institutions with little input from beneficiaries, have now shifted to management by individuals. And with this shift has come a corresponding shift from direct stock ownership to shared ownership through mutual funds. For example, retirement fund contributions have switched from defined-benefit pension plans, with the money managed by life insurance companies and pension funds and held in individual securities, to defined-contribution plans, such as 401(k)s, managed by the individual employee who chooses from among mutual funds.

Second, households have actually been net sellers of directly held stock in recent years, and much of the money flowing into stock funds has come from the sale of stock to—guess who—the very funds that individual investors are favoring. Finally, the available statistical studies show that while stock prices affect money flows into stock funds, there is no evidence that money flowing into stock funds affects stock prices.

Confusion between correlation and causation is often at the heart of strongly held, but incorrect, beliefs. In the 1990s, a robust economy, low inflation, and high earnings growth have made common stocks the security du jour, encouraging investment in stocks. At the same time, investors found an investment vehicle they prefer to direct ownership, the mutual fund. Thus, investors have been shifting their existing portfolios—and their new saving—toward common stocks because of high expected returns, and they have chosen to use mutual funds as the primary investment vehicle. A correlation between stock prices and mutual fund flows is created, but the causal relationship is from high expected stock prices to mutual fund flows. Thus, if you are searching for the cause of an impending decline in the stock market, look for other culprits, like our love affair with stocks, whether held directly or through mutual funds.

Myth 2: Mutual fund shareholders are less likely to sell on bad news than are direct investors

As common stock ownership has shifted from direct ownership by individuals and institutions to shared ownership through mutual funds, some have argued that this switch has served to stabilize stock prices. Mutual fund investors are less likely than direct investors to sell on bad news, the argument goes, so the average dollar invested in stocks is thought to be less sensitive to current conditions. The growing education and financial sophistication of individual investors is one reason cited, but this applies to both direct investors and mutual fund investors. Perhaps a more persuasive reason is that much of the new money flowing into mutual funds is from retirement plans, and this money is guided by a long-term perspective not unsettled by short-term fluctuations.

Two counterarguments can be raised. First, even if it is true that the average dollar managed by individuals is more stable than it used to be, it is not clear that the average dollar invested in stocks is also more stable. Pension funds and life insurance firms were also slow to shift from stocks to other securities, and the decline in their share of stock ownership might mean that the stock market is more volatile because of these institutional changes. Second, there is no compelling evidence that a dollar invested in mutual funds for retirement purposes is less likely to be shifted out of the stock market than is a dollar directly invested in stocks. And there is some evidence on the other side. For example, one pension benefit manager examined switches among its funds during the July 1996 sell-off, comparing the switches out of stock funds by its clients to all stock funds. The retirement money was just as "hot" as the average mutual fund dollar. In discussions with mutual fund managers, I have found no agreement on whether a dollar held for retirement is more or less likely to be redeemed than a dollar held for other purposes.

So we really don't know whether the shift from direct investment in individual stocks to shared investment through mutual funds has resulted in increased investor willingness to hold during downturns. Moreover, with just a few, brief downturns in recent years, history provides little guidance on the issue.

Myth 3: The mutual fund infrastructure is insufficient for massive redemptions

During the October 1987 stock market break, many shareholders could not reach their mutual funds by phone to get information or to submit redemption requests, leaving perhaps an indelible impression of inaccessibility. My research confirms that mutual fund senior management have been concerned about accessibility, feeling that continued growth of money under their management rests on the shareholder's belief that the fund is accessible under all conditions. To ensure accessibility, funds have supported a significant expansion of facilities to improve the ability to handle surges in shareholder calls.

Advances in telecommunications technology have also helped; improvements in both local and national telephone networks allow larger peak volumes as well as rerouting of calls around congestion points, and an electronic mutual fund share-clearing network has been established that allows brokers to trade shares for their clients without adding to the congestion at the fund's transfer agent. In addition, mutual funds are better prepared to add telephone capacity quickly. Contingency plans at many of the funds involve bringing staff from their regular duties and assigning them to answer the phones when response times slip below preset standards.

Expansion of capacity doesn't guarantee an improvement in performance, just as widening a highway doesn't guarantee faster travel speed. As capacity increases, so does use of a facility; so normal loads will increase and peak loads can still be excessive. But the market break of October 1997 suggests much improved performance under stress: Over 1.2 billion shares were traded, twice the volume on October 19, 1987, with no sign of the disruption encountered 10 years earlier, although there were isolated stories of shareholders not able to contact their mutual funds. This experience suggests that mutual funds are better prepared for a surge in telephone and electronic contacts.

Myth 4: In a stock market free-fall, shareholders might not get their cash back

Investors in mutual funds can get their money back by redeeming their shares, which the funds buy back at the next posted daily net asset value. The Investment Company Act of 1940 allows funds to delay payment of redemptions for up to seven days or, in extreme cases, to adopt payment in kind by distributing shares of individual stocks to the fund shareholders. Clearly, however, shareholders expect to redeem shares quickly and for cash.

Can shareholders feel assured that cash redemption will prevail? The historical record back to the 1970s suggests that redemptions on a magnitude that might threaten the ability of shareholders to get cash are both rare and short-lived. Even in October 1987, funds faced only a brief spike in redemptions to 3 percent of their assets, double the normal 1.5 percent but well within manageable limits. This was followed by a return to usual redemption levels. True, in the frenzy of October 19 to October 21, redemptions at some funds were especially high, but these quickly abated as investors realized the episode was not the beginning of a major debacle. True, also, that some funds face more redemption pressure than others, particularly international funds and funds in less developed countries.

Any mutual fund that contemplates either redemption delays or in-kind redemptions is in deep trouble. Sensing this, funds have prepared themselves to satisfy their shareholders' thirst for cash. The typical stock fund holds about 5 percent of its assets in cash equivalents, and most fund families now have lines of credit with banks. Some large fund families have received Securities and Exchange Commission approval for intrafamily lending, allowing a fund short on cash to borrow from a fund with lots of cash, like a money market fund. Thus, the size and duration of redemptions would have to be quite large by historical standards before a fund would have to rely on its authority to delay redemptions for up to seven days or to redeem in kind.

Can it happen? The answer is yes. But the likelihood of encountering a payment delay or payment in kind is extremely low for those funds holding marketable securities in countries with healthy security exchanges and wide investor interest.