Index Funds Don’t Match Indices Exactly
The Vanguard 500 Index Fund (VFINX) has a svelte expense ratio of 0.18% with a year to date performance of 11.67%, compared to 11.83% increase made by the S&P 500, the index the fund is modeled against. The difference of 0.16% is less than the expense ratio of 0.18%, which would mean that the index actually beat its benchmark by 0.02%, not something to write home about… until you realize that the fund isn’t meant to beat the index, it’s meant to mirror it exactly.
Let’s take a look at Fidelity’s Spartan 500 Index Fund (FSMKX), their version of the S&P 500 index fund with an even lighter expense ratio of 0.10%. Year to date, the Spartan 500 has gained 8.73% compared to 8.77% by the S&P 500 (ignore the difference in the two YTD numbers, Vanguard updates quarterly whereas Fidelity updates monthly, but they’re still apples to apples comparisons). So the 0.04% difference meant that the Fidelity fund beat the S&P index by 0.06% year to date.
Why the disparity? I mean the index funds are supposed to match the index right? Well the problem, which has been well documented but something I just didn’t think about until now, is that the differences in performance reflect how quickly a particular fund can react to changes in the index. Companies being added, dropped, changing in allocation; all those events need to be acted on and when you talk about billions of dollars, it’s now like they can change at the drop of a hat. The different returns reflect this lag time.
Is there anything you can do to take advantage of this? Nope, in fact you can’t even look at the 0.06%, how much the Fidelity Spartan fund beat the index after fees, and make a decision because it’s likely (though I haven’t the inclination to calculate for sure) to be statistical noise.
(For Fidelity and Vanguard, I selected the Investor class shares, not the “better” class of shares which required much higher balances)
Subscribe
del.icio.us
reddit.com
11 Comments - Share Your Thoughts
[...] just wrote up a post on Blueprint for Financial Prosperity about the differing performance returns of index funds, specifically a Vanguard and a Fidelity S&P index fund, that may be of interest to folks who [...]
I’m with Vanguard retirement fund 2045. Highly recommended.
Actually it’s more complicated than that. Not all index funds are the same. Vanguard 500 index fund does not actually hold the S&P 500. It uses complicated math to hold a basket of funds that closely match the S&P 500 to mimic it at a lower cost. Basically sampling technique
link
Other index funds such as the Fidelity Spartan 500 hold exactly what the S&P 500 is. They should all be pretty close thought. Vanguard and it’s ilk use it’s method to lower expenses.
This is tracking error right?
I guess that you can’t do anything much about it. If I’m comparing two or more index funds (for the same index) I usually look at past performance in tracking the index to see if one of them has been much worse than the other. Other than that, I just don’t check the values very often - you miss more of the noise that way.
One of the things that affects the tracking error is the regular updates of the index (companies being added, dropped etc). I bet that near this point in time, the tracking error is larger, and further away it gets smaller.
Oh really… I had no idea!
Why does this matter? With a difference of 0.06% or less, and the statistical likelihood that this will approach zero over time, this seems like financial trivia to me.
It doesn’t matter but it’s certainly interesting to know.
Like Dong said, tracking error can play a role.
An index fund can go one of two ways. It can statistically sample the index (as Dong describes) or it can hold substantially all of the indexed stocks. In the former case, transaction costs are kept down, resulting in lower fees. In the latter case, the fund more closely tracks the index.
Some index funds loan out shares for shorters and earn a small bit of interest on it.
Maybe I am misinterpreting what you are saying, but you said YTD returns, or ~5 months. Wouldn’t the expense ratio be adjusted accordingly, i.e. the expense ratio on the Vanguard should be closer to .075% YTD (.18%*5/12) while the Fidelity expense ratio is slightly less than 0.05% (.1%*5/12).
Overall, good points… These tracking errors can be even greater when dealing with larger groups like the Russell 2000/3000 or Wilshire 5000 which rely more on sampling than actually holding every stock.
I compared YTD returns but those expense ratios are annual expenses, I compared apples to oranges… sorry about that.