• Page :
  • 1
  • Text Only
Voting History
rated:
I had some idle time in my hand this weekend (a rare phenomena) and hence was just looking into what is termed "Excess Return" in the financial industry, specifically for the two trackers that I have a bunch of money in (FUSVX/FUSEX for my 401k and VFINX for HSA).

Even though I was familiar with the concept of excess return (long time ago I wrote some code to model excess return for an entirely different market with very different issues involved) - I never realized just how big a deal this could be for S&P Trackers.

Cases in point:
VFINX : https://fundresearch.fidelity.com/mutual-funds/summary/922908108 
FUSEX : https://fundresearch.fidelity.com/mutual-funds/summary/315911206 
Identical charts are also available from Morningstar - however, I used the charts from Fidelity since it clearly mentions that the charts do not include any fees.

Notes:
1. They both have identical return charts. Do they use the same underlying investment vehicle? (i.e. vanguard managing Fidelity's money, or vice versa)?
2. If you invested $10,000 in either of them back in 2006, you would have 18,436 in your account as of 7/31/2016. A pure S&P Tracker with no "Excess Return" would have $21,089 in it. i.e. the tracker underperforms the index by $2653 over 10 years. If we ignore the time value of money - that is > 25% of the initial investment!!!

Calculating the geometric mean - it comes out to 1.505%/year in excess return. Calculated as a difference between the two returns - it comes out to be 1.61%.
Excel formula used to calculate geometric mean is =((end_val/start_val)^(1/(10-1))-1)).

Here I was obsessing about management fees of 0.02% vs 0.1% - while losing at least FIFTEEN times that amount to excess return!!!

Any ideas what causes this tracking error or "excess return"?

Footnote:
I have some familiarity with the reasons for "Excess Return" in the commodities market as I had to deal with them at work. Most of the "excess return" there is driven by:
1. "Roll Return" when you have to roll forward the spot futures as they expire. Usually this is an expensive affair as the spot futures markets tend to move against you as many trackers start the roll at around the spot expiry.
2. Either Contango or Backwardation - as the market situation may be.
We used to use another general catch-all bucket called "operational excess return". Usually #1 and #2 dominated the excess return in the commodities market.

Assuming S&P 500 trackers don't use futures - none of these causes of inaccuracy really apply!! So what causes the 1.5% underperformance??

Member Summary
Staff Summary
  • Also categorized in:
Thanks for visiting FatWallet.com. Join for free to remove this ad.

I'm not sure exactly what data you're looking at since you didn't say what dates, etc, but I'm pretty sure you're misunderstanding what you're seeing.  Things that track the S&P or other large cap index don't outperform by 1.5% over a decade.  Maybe some of them have dividends and some are just price indexes?  That could be about 1.5% off.

You are right - I actually misinterpreted the data!

When using the right data - FUSVX/FUSEX/VFINX only underperforms the market by 4-13 basis points, which is far more understandable as tracking errors.

Editing out the "sensationalist" headline in a bit.

------------------ Details of the data -------------------------
The links I have posted in the OP has a chart titled "Hypothetical Growth of $10,000" in both of them.
In this chart - if you mouse over the chart then it shows you the date, hypothetical growth of 10,000 from 7/31/2006 for various dates. The data point I was using are:
Start Date: 7/31/2006
Initial Amount: 10,000

End Date: 7/31/2016
FUSVX: $21,029 (IRR: 8.61%)
FUSEX: $20,952 (IRR: 8.565%)
VFINX: $20,850 (IRR: 8.506%)
S&P 500: $21,089 (IRR: 8.644%)

"Large Blend": $18,427 (IRR: 7.033%) - this is the average of all funds that Morningstar/Fidelity considered as "Large Blend".
I misread the chart colors to think that this $18,427 represents the ending value for FUSEX/FUSVX/VFINX and hence the problem.

ETFs should have a very minimal tracking error. Read this brief description how the authorized participants trade with the ETFs: https://www.ici.org/viewpoints/view_12_etfbasics_creation. The authorized participant arbitrages the difference between the ETF market price and NAV, and at the end of the day exchanges creation units with the ETF. The ETF should not lose money on this (assuming they can agree what shares and how many belong to the tracked index). However, the retail client may lose some money whenever they buy / sell the ETF shares, but that is expected and somewhat avoidable with limit orders and a buy and hold strategy.

A mutual fund will have a larger tracking error. Whenever investors buy or sell shares in the fund, the fund manager needs to buy or sell shares of the companies in the index directly on the market. For large funds with large turnover this is rather difficult to do right. They have the need to trade a large number of shares and they must buy them from other market participants such as day traders and hedge funds who are better equipped to play games (e.g. identify which market orders belong to mutual funds).

Is this tracking error guaranteed to always be a "loss"? Or is there some scenario where randomness comes into play?

If it is always a loss - then I sense some arbitrage opportunity right there. ETF -> less tracking error. Mutual Fund -> more tracking error -> returns always lower than ETF. Find an ETF/Mutual Fund pair and arbitrage the heck out of the spread between them with as much leverage as possible.

Is there some flaw in this line of reasoning? Or did I just strike (and published to FWF) an idea that will make all of us billionaires? :-D

If I understand your last idea correctly, it would require the ability to short the mutual fund...

KingPeter said:   If I understand your last idea correctly, it would require the ability to short the mutual fund...

I was searching for option chain on vanguard index tracker mutual funds but did not find any in 5 minutes of googling.

With something so widely used - there must be some way to cobble together something that mimics the cash-flow from a short.

It's a different topic if the arbitrage opportunity is:
A. At all present.
B. wide enough to allow crossing so many bid-ask spreads (kills you on not-so-liquid markets) and fees.

Obviously all the above are brain-f*** (errr - loud thinking) on my part! If an arbitrage is there to be found - I bet it has already been exploited down to oblivion. As far back as 10 years ago - I heard a program trader who made a living on arbitrages complain that he is being driven into less and less liquid markets by disappearing arbitrage gaps in more popular markets.

I don't think they diverge enough to make money with it. And if a pair was reliable enough- the short side would likely be expensive to borrow.

puddonhead said:   
KingPeter said:   If I understand your last idea correctly, it would require the ability to short the mutual fund...

I was searching for option chain on vanguard index tracker mutual funds but did not find any in 5 minutes of googling.

With something so widely used - there must be some way to cobble together something that mimics the cash-flow from a short.

It's a different topic if the arbitrage opportunity is:
A. At all present.
B. wide enough to allow crossing so many bid-ask spreads (kills you on not-so-liquid markets) and fees.

Obviously all the above are brain-f*** (errr - loud thinking) on my part! If an arbitrage is there to be found - I bet it has already been exploited down to oblivion. As far back as 10 years ago - I heard a program trader who made a living on arbitrages complain that he is being driven into less and less liquid markets by disappearing arbitrage gaps in more popular markets.

  
That's good creative thinking to use options to mimic a mutual fund short position. Unfortunately, I have never heard of an option on a mutual fund. (NB: Options on ETF's exist, but not options on the underlying mutual fund)  



Disclaimer: By providing links to other sites, FatWallet.com does not guarantee, approve or endorse the information or products available at these sites, nor does a link indicate any association with or endorsement by the linked site to FatWallet.com.

Thanks for visiting FatWallet.com. Join for free to remove this ad.

While FatWallet makes every effort to post correct information, offers are subject to change without notice.
Some exclusions may apply based upon merchant policies.
© 1999-2017