Update of Evaluating Performance Essay

[racemize]

I fixed some minor errors, updated for 2017 year-end results, and added Tom Gayner to the list (equity returns). 

 

https://drive.google.com/file/d/0BxTPR9eP5nWebFdOTnZSR3l1eVk/view

 

[crastogi]

Thank you for compliling!

[John Hjorth]

Thank you for sharing, Joel!

[chrispy]

Very well done. Looking forward to digging more into this

[racemize]

I just realized I screwed up my own record (that's what I get for not using XIRR), so updated-updated version now (same link).

[gjangal]

Hi Joel,

 

Thanks for sharing. Very generous of you given the time and effort it takes to compile such a study. Really appreciate it. The rolling annual out performance statistics gives a great idea of how the manager performed over time. Do these numbers include the effect of dividends? for the SP500 dividends form a good part of the total return, but maybe it harder to include in the study as compiling dividend data for all the funds may be too difficult, timing of cash flow etc make it a hard problem. Thanks again

[racemize]

Hi Joel,

 

Thanks for sharing. Very generous of you given the time and effort it takes to compile such a study. Really appreciate it. The rolling annual out performance statistics gives a great idea of how the manager performed over time. Do these numbers include the effect of dividends? for the SP500 dividends form a good part of the total return, but maybe it harder to include in the study as compiling dividend data for all the funds may be too difficult, timing of cash flow etc make it a hard problem. Thanks again

 

Net returns for hedge funds include dividends as they just keep them in the account.  I believe all of the returns from the mutual funds are all total returns as that is the standard reporting method and lines up with morningstar's numbers who are TR.  The S&P numbers are all total return, so I believe dividends are accounted for in the results.

[chrispy]

This very much reinforces the idea that time is the best friend of a patient investor.  After 5 years most outperformed the S&P 50-70% of the time.  After 10 years almost every single one had outperformed the S&P 100% of the time.

 

One bit that jumped out at me was Davis' having 0.77% outperformance over 47 years which lead to 33% more in total return.  Every little bit counts!  This is similar to the underperformance that high fees will cost an investor over long periods of time.

 

How many of these funds do you think benefited greatly by holding BRK the whole time?

 

What were some of your takeaways from doing the study racemize besides what you mentioned in the first couple of pages?

[racemize]

This very much reinforces the idea that time is the best friend of a patient investor.  After 5 years most outperformed the S&P 50-70% of the time.  After 10 years almost every single one had outperformed the S&P 100% of the time.

 

One bit that jumped out at me was Davis' having 0.77% outperformance over 47 years which lead to 33% more in total return.  Every little bit counts!  This is similar to the underperformance that high fees will cost an investor over long periods of time.

 

How many of these funds do you think benefited greatly by holding BRK the whole time?

 

What were some of your takeaways from doing the study racemize besides what you mentioned in the first couple of pages?

 

I think I'd call a few things out.  First, there's a huge amount of selection bias in this essay (although this wasn't a statistical paper for all funds), so I would caution the conclusion that waiting gets you outperformance--it does, so long as the investor is actually good.  It can be very hard to know whether they are good while the underperformance is happening.  But, that is the point of this type of analysis, to see how often they do well and whether they are worth being patient with, but perhaps that was what you meant already? 

 

The other thing I'd say is that many times the percent outperformance at the 10 year mark wasn't as good as I had hoped it would be coming in.  To me, if you are an outside investor and you were patient for 10 years, you need to be rewarded almost all the time (say 95%), but there are investors with clear skill that are much less than that.  Even more surprising is seeing Schloss, who absolutely crushed the markets over long periods of time, have underperformance over 20 year periods.  Imagine being the one guy who joined then.  Ouch.  That has led me to the conclusion (and I think this is coming out both from the essay and how the markets have been the last few years) that opportunistic value investing is quite lumpy but high quality investing is very consistent.  If you are taking outside money, then that consistency is important.  If you are doing it on your own, then it probably doesn't matter (i.e., outside money, the statistics over various holding periods matter because people go in and out, but for the investor himself, only the overall matters).

 

Also, I'd caution you a bit on the 0.7% outperformance--these don't include taxes, so unless you are in an IRA, I would bet that the after tax performance is actually less than the S&P.  I have a couple of essays on that topic that may be interesting.  I'd expect a 1% drag on most long-term results.  On that point, Russo actually calculates his tax drag (and he holds most of his stuff forever) at something like 0.7%, so almost everyone will be above that.

 

Further, the Davis family is really interesting to read about.  The first one had an insane record (I think right up there with Buffett) and had caught on to insurance companies around the same time.  I think his son said he kept complaining that Buffett would buy his companies, but he ended up with a fair amount of Berkshire the whole way through as a result.  The Davis mutual fund was his son first, who had a great record.  Now the third generation took over and the results are not that great.  Of course it is a financial fund and there was a financial crisis, so Chris may well pull it out.  Anyway, changes in PMs at funds make me nervous about making any conclusions using numbers from the previous PM.

 

 

[stahleyp]

With Davis I bet there was probably a load too. 4.75% right off the top hurts a bit.

 

Since Chris Davis took over the fund has barely beat S&P 500 (before load and taxes).

[racemize]

Oh and the Berkshire question--I think it depends on the time period and the person, but I'm not sure how easy it is to quantify it.

 

Schloss didn't own it. 

Munger was contemporary. 

I doubt the tweedy partnership owned it (and their mutual fund didn't outperform, so it doesn't matter). 

I think Sequoia benefited from it quite a bit, but they generally did well otherwise, except for this recent Valeant debacle. 

Russo definitely had a big position throughout. 

Longleaf hasn't owned it in a while. 

Gayner is pretty diversified, but Berkshire is consistently 10% I think? 

Looks like Oakmark doesn't own it. 

Akre never had a huge position in Berkshire (he always had MKL). 

It isn't a position in Yacktman.

It's a 20% position for Giverny

I think it is consistently 12%+ for Aquamarine

Pabrai only held it as a cash replacement in between ideas

Fairholme owns other stuff generally

Arlington--I'm not sure whether he's always had it--he took a huge position when it hit book value in 2011 or 2013 (whichever it was), so that was more than just a buy and hold type of deal.

 

[Jurgis]

First, there's a huge amount of selection bias in this essay (although this wasn't a statistical paper for all funds), so I would caution the conclusion that waiting gets you outperformance

 

Hear, hear.

 

Anyway, changes in PMs at funds make me nervous about making any conclusions using numbers from the previous PM.

 

Which might be an issue/question with new Sequoia...

[chrispy]

Thank you for taking your time to share all of this!