The Horse Race

A Practical Test
There's not much point in further pursuing the American Funds versus Vanguard comparison, which I discussed on Tuesday and again on Wednesday. That territory has been covered: American Funds made business decisions that helped it during the 1980s, 1990s, and early 2000s but which have since held it back. The performance of the company's funds, by and large, has been just fine.

Let's take that last part further. Today, the nearly unanimous advice from the academic community, institutional investors, and the financial media is to index. Often, the recommendation is made aggressively, with the suggestion that to do anything but index is irresponsible at best and borderline criminal at worst. (The most aggressive of attacks compared mutual fund companies to cigarette manufacturers.) Two years ago, Yale University chief investment officer David Swensen penned a highly influential The New York Times article recommending that the SEC require that financial advisors show prospective investors an index option. Wrote Swensen, "What about a requirement that every mutual fund offering be accompanied by an index-fund alternative, with the burden of proof on the vendor to justify the sale of a high-cost product?" 

We can test how that would have played out with American Funds. Let's take a good long time frame, say 25 years--when I first started at Morningstar. If prospective investors in 1988 were given the choice between owning an American Fund and owning an index fund, how would that have played out? We'll look at domestic-stock funds because that's what people mean when they attack actively managed funds. (Nobody says that  PIMCO Total Return (PTTAX) isn't worth owning.) Besides, that has the possibility of being a true competition. There's no battle with either international-stock funds, where the American Funds' two entrants blew all index funds out of the water (yes, they did), or with bond funds, where American Funds is not particularly successful.

In mid-1988, among U.S. stock funds, there were seven American Funds and 10 index funds (from all companies, not just Vanguard). Remarkably, all 17 funds still exist. I dropped their 25-year annualized rate of return, through June 30, 2013, onto a spreadsheet. This was the rank order:

  - source: Morningstar Analysts

Yes, there is a catch. Typical mutual fund calculations assume that the investor holds the fund both entering and exiting the time period. Thus, the American Funds' funds benefit from this analysis because they are not hit with their load charges. 

Let's adjust. This is the rank order after the American Funds are charged with their maximum front-end load at the time, 8.5%. 

  - source: Morningstar Analysts

Score that as a very mild victory for index funds. It's close, though. 

At this point, the academicians, institutional investors, and financial press conclude their analysis. They shouldn't. They just compared the price of a mutual fund plus financial advice with a mutual fund and no financial advice. That's not a complete analysis, from two directions. First, some American Funds' investors can own that company's funds in a no-load, no-advice format--for example, within a 401(k) plan. Second, if the recommendation is that a financial advisor place people in an index fund rather than an actively managed fund, then the analysis needs to acknowledge that the advisor will be paid for that activity. The index fund will not come for free. (Nor should it.)

That's easy enough to model, because while rates do vary, the industry standard asset-based fee for financial advice is 1% per year. So, let's adjust the study by charging the index funds 1% annually to reflect the cost of financial advice. The rank order becomes:

  - source: Morningstar Analysts

My goodness. 

However, there is a final item to address. Swensen argues that the advantages of index funds include not only lower costs but also lower tax bills. So, let's adjust the returns for the effect of taxes. That's harsh on those who own American Funds in 401(k) plans or in IRA accounts, but it's certainly a reasonable argument. So, we will assume: The investor buys the fund on July 1, 1988 (thereby paying the maximum front-end load, if applicable); owns the fund for the next 25 years (thereby paying the annual advisory fee, if applicable); pays taxes each year on income and distributed capital gains, at the maximum tax rate of that time; and pays taxes on the gains on the liquidation of the fund on June 30, 2013.

The one catch is that I can't easily generate aftertax returns for the past 25 years. I can do it, but it will take a lot of manual effort on a spreadsheet. So, I will take a shortcut. For each fund, I have the annualized difference between the pre- and post-tax returns over the trailing 15-year period. I will assume that the same relationship holds for 25 years. In other words, if  American Funds AMCAP's (AMCPX) pretax return was 1.76 percentage points higher per year over the trailing 15 years than its post-tax return, then I will assume that the same 1.76 percentage point gap held forth for the 25-year period.

Not perfectly correct, but the inaccuracies should be small. Certainly, the direction is correct; as expected, the actively managed American Funds tend to be less tax efficient than the indexes, in some cases by more than 1 percentage point per year. 

This is the rank order for the final, post-cost, post-tax list:

  - source: Morningstar Analysts

I've done my best in this analysis to be fair to both sides. I've charged the highest possible load (in fact, a load of that level is now extinct). I've assumed fully taxable accounts. I've assumed the maximum possible tax rates. All three of those items hurt American Funds' numbers. On the flip side, I've included all index funds available for sale at the time, rather than selecting the lowest-cost example. And I've included the effect of an advisory fee for financial assistance. Those two items hurt index funds' numbers.

One remaining criticism that could--and should--be levied is that I've cherry-picked this study by selecting American Funds, rather than another fund company that had weaker investment results. Yes and no. By the late 1980s, American Funds had become the clear choice for U.S. stock funds for independent and/or experienced financial advisors. In its distribution group of nonproprietary load companies (meaning funds that appealed to independent advisors), American Funds ran the largest, third-largest, fourth-largest, and sixth-largest funds in existence. It was the dominant player. 

This column does not argue against indexing. As I've written elsewhere, my largest fund holding is  Vanguard Total Stock Market ETF (VTI). Rather, this column argues against stridency. Nearly every financial commentator argues stridently for indexing and against active management. Some are evangelical in their fervor. Look at this column's final list. Is that so deserving of certainty? Is that a surefire indication of the righteousness of the cause?

I say no. I say, if the matter were so simple, these tables wouldn't look the way that they do.

John Rekenthaler has been researching the fund industry since 1988. He is now a columnist for Morningstar.com and a member of Morningstar's investment research department. John is quick to point out that while Morningstar typically agrees with the views of the Rekenthaler Report, his views are his own.