Question regarding impact of owner operators on Indexes

[racemize]

Hi all, I've been semi-obsessed with indexation and the float adjustment that was introduced in 2005 since reading a lot of Murray Stahl's commentary (I think he is more obsessed).  In particular, I've been discussing this change with an indexer friend of mine, as it concerns me for future returns of indexes.  However, it has not been easy to reproduce the data in Stahl's claim (see below).  I was curious if anyone on the board has looked into this or if they have access to the data, in which case I would like to perform the calculation.  In any event, this is an email I sent to Stahl that I'm interested in answering:

 

Hello, after reading all of the FRMO transcripts available, and various other articles/commentaries from Horizon Kinetics, I was hoping to ask Murray some follow up questions, mostly with regard to indexation and the effects of owner-operators:

 

1) In the 2012 Annual Meeting, you said:

 

"If you tried to replicate that exercise, you would look at the S&P 500, look at all the holdings since the day it began in 1957, extract all the positions in which there were owner-operators that you could identify, and then recalculate the S&P as if there never were any owner-operators.  In every time period, there would be a different number of owner-operators but, for ease of illustration, let's just make believe that there were an average of 50 companies of this type.  You'd be calculating, on average, the S&P 450 instead of the S&P 500.  I absolutely promise you, if you did that calculation, you would never buy the S&P.  What I'm telling you is that the bulk of the return of the indices--and not just in the United States, but in all nations, the bulk of the return was earned by these owner-operators."

 

I found the above quote to be very compelling.  However, it is extremely difficult to do that calculation without access to quite a bit of data, which I have been unable to get.  Do you have any sort of write-up that provides the data or results of that calculation?  e.g., what the returns would have been or what magnitude the reduction would be from the S&P 500 returns?  While it seems you have satisfied this question for yourself, it would be nice for some of us followers to get a look at the data and/or results of the calculation versus simply taking your assurance.

 

2) Following on from the above, do you have any idea of what the returns of the S&P 500 would have been if the float adjusted method was used from 1957 on, or perhaps what the S&P 500 results would have been using the unadjusted method from 2005 on?  Or said another way, how much impact does the float adjustment have on returns, on average?  e.g., are we talking reductions of 1% a year on average? more? less?

 

3) Was the wealth masters index always an owner operated index?  i.e., does the outperformance of the wealth masters index over the S&P 500 represent the general outperformance of owner-operators versus the S&P or are there/were there other companies in the wealth index as well? 

 

In general, I have found the indexation discussions fascinating, but would like to get some more specifics on what we are talking about.

 

 

Thanks very much!

 

If anyone else is interested in this question or has already answered any of the above, I would love to hear it.

[Packer16]

Jeremy Siegel in his book "The Future for Investors", examines the firms with largest returns in the S&P 500 going back to 1957.  In the appendix he provides some details.  You may want to read the book and send an e-mail to him to see if he has looked at the S&P 500 this way.  I am not sure what he describes as an owner operator but with the list in the book you can test the hypothesis.  Siegel also develops some rules for out performance based upon his observations.

 

Packer

[Dustin T]

I can't produce the data you are looking for, but I have been interested in the theory that owner operators outproduce the rest of the market for about 5 years and towards that end I created a marketocracy portfolio that is in the 97.2 percentile of funds on that site over the last 4 years based only on having a basket of owner operators as it's portfolio.

 

I also randomized the S&P taking the first 30 stocks that had either a CEO or Chairman owning at least 2% and have compared it to two different random control groups on an excel spreadsheet. Four years in the Owner operators are beating one control group and losing to the other, but what I find most interesting is the top stocks in both control groups are Ralph Lauren and Whole Foods which are both ran by the founder which owns them at 1% without them the owner operator stocks are the clear winners.

 

Running a correlation on the data, it showed a 20% positive correlation between ownership % and total return.

 

I know enough about statistics to know that my data isn't a perfect scientific test, but it has certainly convinced me of the validity of the owner operator idea and within 24 hours of learning about Murray Stahl I had invested in his company which is I think a record for me.

 

 

[racemize]

Thanks for the responses guys.  I also got one from HK.  Here's what they said, with attachments:

 

Hi Joel,

 

Thanks very much for your thoughtful email.  Attached please find further information on our Wealth Index.  If I may recommend a next step, I think you will find speaking with one of the co-portfolio managers on our Wealth Index useful in addressing your thoughts/questions.  Please let me know if there is a particular time that would work best for you and I’d be happy to coordinate.

 

In the interim, please feel free to give me a call should you have any questions or need any additional information.

 

 

The Index Liquidity Riddle: More Is Less

By JUSTIN LAHART

 

You would think that the whole point of a stock index is to be, well, an index of the stock market's performance.

 

But thanks to the popularity of exchange-traded funds, or ETFs, stock indexes have in recent years been doing double duty as investment vehicles. At the same time, there have been subtle but important changes in the way indexes are constructed.

 

Bottom line: The indexes aren't measuring exactly what they used to.

 

It is a lot easier to manage an ETF if the stocks that underlie it are easily traded. If, instead, the stocks are illiquid, there is a risk their prices will get artificially inflated when money flows into the ETF. The opposite can happen when money flows out.

 

One step index providers have taken to bolster liquidity has been to move from capitalization-based indexes, where the weight of each member is determined by the value of its total shares outstanding, to float-adjusted indexes. In the latter, shares that are unavailable to the public (such as stock held by company directors) don't count toward a company's weighting. Britain's FTSE Group made the move to float-adjusted indexes in 2000, followed by MSCI in 2002 and Standard & Poor's in 2005.

 

But while switching float adjustment may improve an index's liquidity profile, argue researchers at New York money manager Horizon Kinetics, it may also cut into its ability to generate returns. That is because many of the companies that have added oomph to indexes like the S&P 500 in the past did so at a time when a great many of the shares were held by insiders.

 

For example, when Microsoft entered the S&P 500 in June 1994, the combined holdings of co-founders Bill Gates and Paul Allen, along with then-Vice President (and now chief executive) Steve Ballmer, represented about 40% of the company's shares outstanding. By the end of 1999, the three had whittled their stakes down to a still-substantial 22% of the company.

 

From the end of 1994 to the end of 1999, Microsoft's market capitalization swelled from $35.5 billion to $604.4 billion, a gain of about 1,600%. That accounted for about one-sixteenth of the S&P 500's 267% increase in market capitalization over the period. If the S&P had been float-adjusted back then, Microsoft's weight would have been lower, and it wouldn't have helped lift the index as much as it did.

 

It wasn't just Microsoft. Several S&P 500 companies with limited floats, including Wal-Mart Stores, Oracle and Gap, served up outsize gains in the 1990s.

 

Indeed, Horizon Kinetics has constructed an index composed of companies whose wealthy insiders hold substantial stakes. This has beaten the S&P 500 by 2.7% annually on average over the past 20 years

 

One reason could be simply that company founders hanging on to big chunks of their companies do so for a reason. So, by curtailing the S&P 500 weighting of companies whose executives have substantial skin in the game, Standard & Poor's may have also limited the index's future gains.

 

David Blitzer, who heads S&P's index committee, points out that at the time the S&P 500 made the switch to float adjustment, most of the stocks in it had high floats. He said that under an informal rule dating back to at least the 1970s, stocks whose floats counted for less than half of their capitalization weren't included in the index.

 

That said, while the S&P 500 has risen 26% since it completed its transition to float adjustment in 2005, it would be a percentage point higher if it was still a market-cap-weighted index. And this was a fallow period in the annals of corporate America.

 

If the U.S. enters a more dynamic phase, and recently minted companies flourish again, the S&P 500 may leave gains on the table. The same is true for investors in ETFs and other funds that track the index.

Owner_Op_Paper_March_2013.pdf

Horizon_Kinetics_Wealth_Strategy_11052013_CALL.PDF

[Hielko]

Thanks for posting.

 

Based on:

 

Indeed, Horizon Kinetics has constructed an index composed of companies whose wealthy insiders hold substantial stakes. This has beaten the S&P 500 by 2.7% annually on average over the past 20 years

I would expect that removing the owner operators from the S&P 500 isn't going to have a huge impact on the performance. Owner operators are - I think - just a small part of the S&P 500, and if they outperform on average 2.7% you have to figure that the average S&P performance is reduced by a percentage substantially less than 2.7%. Would love to see some to do the real math though, but not easy to get access to the right data.

[racemize]

Thanks for posting.

 

Based on:

 

Indeed, Horizon Kinetics has constructed an index composed of companies whose wealthy insiders hold substantial stakes. This has beaten the S&P 500 by 2.7% annually on average over the past 20 years

I would expect that removing the owner operators from the S&P 500 isn't going to have a huge impact on the performance. Owner operators are - I think - just a small part of the S&P 500, and if they outperform on average 2.7% you have to figure that the average S&P performance is reduced by a percentage substantially less than 2.7%. Would love to see some to do the real math though, but not easy to get access to the right data.

 

So there's a couple of interesting questions:

 

1) How much effect should we assume is happening from shifting from market cap to market cap float adjusted index returns?

 

One answer, is that since 2005, there's has been a 1% effect (27% versus 28%), so it is not substantial.  However, it may be larger in a different time period, e.g., because of the large correlation we have had over the recession and recovery.  Overall, I get the impression that the change isn't good, but isn't devastating either.

 

2) How much would the returns be affected historically if the S&P 500 did not include the owner operators?

 

We don't have a good answer for this, but they do state this (not terribly satisfying, however):

"We have not performed the exercise of calculating the S&P as if those companies never existed but we are confident that the return would be dramatically lower than what it has been, and would not have been an attractive investment."

 

There's an implication of perhaps 2% loss in returns, per annum here, I would assume.  i.e., for it to become not attractive, it would need to be significantly less than ~9% returns.  7% is still decent, compared to bond yields say, so I wonder if they are implying more than that.

 

That's a very large implication.  I wonder if it has to do with the particular companies in the past, e.g., Wal Mart, MSFT, etc? versus the ones we have now?  Generally, hard to reconcile.