Comments on: Index Funds Don’t Match Indices Exactly

By: jim

jim — Fri, 08 Jun 2007 20:35:05 +0000

I compared YTD returns but those expense ratios are annual expenses, I compared apples to oranges… sorry about that.

By: Brandon

Brandon — Fri, 08 Jun 2007 18:45:41 +0000

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.

By: MossySF

MossySF — Thu, 07 Jun 2007 15:31:29 +0000

Some index funds loan out shares for shorters and earn a small bit of interest on it.

By: KMC

KMC — Thu, 07 Jun 2007 14:25:51 +0000

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.

By: jim

jim — Thu, 07 Jun 2007 14:17:08 +0000

It doesn’t matter but it’s certainly interesting to know.

By: Mike

Mike — Thu, 07 Jun 2007 14:15:38 +0000

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.

By: jim

jim — Thu, 07 Jun 2007 13:26:38 +0000

Oh really… I had no idea!

By: plonkee

plonkee — Thu, 07 Jun 2007 07:15:52 +0000

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.

By: dong

dong — Thu, 07 Jun 2007 02:29:58 +0000

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.

By: hustler

hustler — Thu, 07 Jun 2007 02:17:58 +0000

I’m with Vanguard retirement fund 2045. Highly recommended.