Why Are Actively Managed Funds Still So Popular?

For most readers of The Index Investor, the evidence against investing in actively managed funds probably is already convincing, and the previous two articles have further added to it. Unfortunately, this is not the case for most investors. Index products still account for only about ten percent of total individual investment in U.S. mutual and exchange traded funds. The question we have to ask ourselves is why this is so.

Explanations for apparently irrational behavior generally fall into two categories. Either investors lack adequate information about the advantages of index products, and/or they fail to process the information they have correctly.

Various researchers have suggested a number of different explanations for investors' lack of adequate information. First, investors tend to be myopic, and evaluate their funds' performance annually. This works against index products, as over one-year periods, roughly fifty percent of actively managed funds will outperform index funds in the same asset class. However, because superior active performance does not persist from year to year, over longer periods of time fewer and fewer active funds outperform their indexed competitors in the same asset class.

A closely related issue is the nature of information (also known as "cues") that attracts investors' limited attention. Research has shown that readily available information often has a higher impact on one’s beliefs than information that is more costly (in terms of time and cognitive resources) to obtain and process. Seen in this light, it is no surprise that many actively managed funds are supported by heavy advertising that emphasizes their gross returns over one year, rather than their longer term returns after sales loads, expenses, and taxes have been taken into account (and trading costs too, but that’s another story). Research also has shown that the way we allocate our limited attention to different information cues is both an individual and a social phenomenon; we naturally tend to pay attention to the cues our friends consider important. This is another reason that actively managed funds spend so much money on advertising that emphasizes their short-term gross returns.

Investors' lack of adequate information also has its roots in what is known as the "confirmation bias." In essence, the confirmation bias has two key elements. First, it takes much more information to change a belief than it does to form one. Second, because we prefer consistency in our beliefs, we have trouble perceiving information that contradicts them, and find supporting cues much easier to recognize and absorb. This means that people who have invested in actively managed funds are unlikely to seek out information that supports index investing, and will require repeated exposure to it before they change their beliefs.

Finally, many investors who rely on advisors to provide them with fund recommendations fail to pay enough attention to the way in which these advisors are paid. They neglect the important information that the incentives of advisors whose compensation comes from commissions on the products they sell are not well aligned with the goals of the investors who rely on them for advice. At the margin, these advisors' incentives will lead them to prefer actively managed funds with high sales loads (out of which their commissions are paid) rather than no-load index funds.

The second major explanation for investors' apparently irrational belief in the superiority of actively managed funds is that they fail to correctly process the information that is available to them. For example, research on the prevalence of over-optimism and overconfidence is quite extensive; it has even been suggested that these traits are, in moderate amounts, evolutionary adaptations that enhance a person's social attractiveness. (Anybody who has tried to decide whom to talk with at a party intuitively recognizes the potential validity of this view.) Practically, this causes many investors to be overoptimistic about the future returns they expect on their actively managed funds, and overconfident that past performance is actually a good predictor of a fund's future success. Overconfidence also works to active funds' advantage via another route. It leads some investors to believe that the best way to minimize risk and maximize returns is to quickly sell losing funds while buying recent winners, rather than holding a well- diversified portfolio of index funds. In practice, this belief leads to higher trading volume, taxes and transaction costs, and lower returns than would have been realized using the index approach.

Another cognitive processing issue is related to the way investors choose between competing alternatives. Some researchers have proposed that people take into account not only rational factors, but emotional ones as well. For example, regret theory suggests that investors measure their performance relative to others, and have a strong aversion to appearing inferior when the task involved doesn't appear to be hard. Given a myopic focus on one year gross returns, this can easily lead an investor to prefer actively managed funds to an index product that will, in any given year, under-perform about half the funds in its asset class.

Finally, when it comes to processing information, most human beings also have a tough time learning from their mistakes. As previously noted, research has shown that people derive emotional benefits from maintaining consistent beliefs. To enable us to do this, we tend to change our stories in retrospect in order to justify our past decisions, even if their results have turned out poorly. This is known as "hindsight bias", which in some ways is the mirror image of the previously discussed confirmation bias (the former being backward looking, and the latter forward looking). For example, how many people do you know who explain the negative returns on their actively managed funds by saying "the whole market has been down", while avoiding the fact that their returns have lagged behind those of index funds in the same asset class (even on a gross return basis)?

While enlightening, these explanations leave unanswered an important question: which of these two factors (inadequate information or incorrect processing) provides the best explanation for investors' continued interest in actively managed mutual funds?

We searched for data that could help us answer this question, and came up with some interesting findings. On the information front, many surveys show that in general investors have a low level of "financial literacy." For our purposes, however, we chose to use the findings from an annual survey that Vanguard does in conjunction with Money Magazine. The ninety five percent confidence interval for this survey is plus or minus three percent (that is, we can be 95% confident that the true value of a variable is within plus or minus 3% of the survey's estimated value). Last year, this survey found that only sixteen percent of investors understood the concept of an index fund.

Finding data that reflected the extent of investors' cognitive processing problems was more difficult. After looking a lot of alternatives, we settled on the Retirement Confidence Survey, which has been conduced for the past twelve years by the Employee Benefits Research Institute and the American Savings Education Council. The estimates from this survey also have a 95% confidence interval of plus or minus 3%. Our objective was to derive an estimate of the percentage of overconfident investors, which we took as a proxy for the broader set of processing problems. To begin with, the survey found that seventy two percent of investors were confident they were doing a good job of preparing financially for their retirement. However, the survey also found that only twenty four percent of investors had actually calculated (or had an advisor calculate) how much money they would need to save in order to live comfortably in their retirement. We took the difference between these two figures (72% - 24% = 58%) as our estimate of the percentage of investors who are overconfident.

Combining our estimates for the percentages of investors suffering from information and processing related shortcomings gives us this estimate of the underlying dynamics driving investors' continued over-reliance on actively managed mutual funds:

	Well - Informed(16%)	Not Well - Informed (84%)
Overconfident (58%)	9%	49%
Not overconfident (42%)	7%	35%

If one assumes that investors who are well-informed and not-overconfident are the ones most likely to invest in index funds, then the 7% estimate from this analysis aligns quite well with our previous estimate that index funds represented 7.8% of non-money market mutual funds in the U.S. at the end of 2001 (see "How Big is the U.S. Indexing Market?" in our December, 2002 issue).

Given this picture of the underlying dynamics, it seems clear that when it comes to the continued popularity of actively managed funds relative to index products, lack of investor information is a more important obstacle than cognitive processing problems. As you can see, better information seems likely to raise the probability that at least 35% of investors will consider using index funds. Moreover, improving information may also indirectly act to reduce overconfidence in the remaining groups of non-index fund investors. Therefore, we will now turn to a closer examination of how we might overcome the information problem faced by many investors.

The first, and perhaps most important question to ask is whether or not this response should be left to market forces, or whether government involvement is justified. The latter requires that some type of "market failure" has occurred -- in other words, that significant costs or benefits exist that are adequately incorporated in market prices. We believe that this is indeed the case, and have written about this issue more than once (see "The Financial Scandal Nobody Wants to Talk About" in our November, 2002 issue, and "We Need More Prudent Experts" in our January, 2003 issue). And we are not alone in our view. The existence of government regulations regarding the presentation of performance data in mutual fund prospectuses confirms that the existence of some degree of market failure has already been accepted by the U.S. Securities and Exchange Commission.

This has led to the SEC's recent requirement that in mutual fund prospectuses, historical returns must be shown not only on a gross basis, but also net of sales loads, expenses and taxes. Recent regulatory changes also forced the disclosure of performance over periods of time longer than one year. However, the survey data cited above suggests that these changes have not substantially improved investors' understanding of the critical differences between actively managed and index mutual funds. More changes are still needed. Here are some of the ones we'd like to see:

First, not only prospectuses, but also mutual fund advertising should be required to present returns after sales loads, expenses and taxes are taken into account. They should also be required to show returns over longer periods of time, as well as the returns on a comparable asset class index. Trading costs should also be disclosed in a manner that makes them comparable across funds.

Second, the current "health warning" on mutual fund advertisements ("past performance is no guarantee of future results") needs to be strengthened. The weight of evidence shows that good past performance is unlikely to persist in the future.

Third, the regulations governing defined contribution pension plans (e.g., 401k plans) should require that at least one index fund be offered in every asset class included in the plans’ range of possible investments, and that the default allocation for plan participants is into these index funds. This would bring these plans into better alignment with the Thrift Savings Plan offered to all federal employees, which offers only index funds (see Thrift Savings Plan ).

Fourth, to make information about the growing popularity of index investing more visible, the Investment Company Institute should be required to separately present data for index and actively managed mutual funds (today, it only breaks out data for exchange traded index funds).

Finally, given the potential future cost to taxpayers of investors' continued over-reliance on actively managed funds (see our November, 2002 article "The Financial Scandal Nobody Wants to Talk About"), the SEC should begin an advertising campaign to strengthen public awareness of the advantages of index funds and portfolio diversification across asset classes. This will not only help investors to make better decisions in their pension plans, but also help them to better evaluate the recommendations they receive from their financial advisors.

Some may see this as an overly aggressive reform agenda. However, given the substantial improvement in investor welfare it could produce, it is one we should pursue with vigor, both in the United States and in all other countries where index products are available to investors.