"All the market return comes from just 4% of shares"

F

Fella

Registered User
Messages
524
Copied from another thread

There was lots of discussion on how many stocks you need to own , I remember Brendan (the site owner) suggesting 10 stocks is enough.
Sarenco popped up some stats that indicated holding 10 stocks is not enough . The point Sarenco made very well was that the bulk of stock market gains come from a tiny percentage of the stocks and if you miss one of these your likely to do no better than investing in bonds or just saving money in the bank.

I didn't just take Sarenco's word for it but researched it myself and as always he was in fact correct , that was enough for me to avoid individual stocks and go with the investment trust route.

Stock markets have delivered big long-term returns because of the stupendous returns earned by a tiny minority of companies. Bessembinder calculated that the US market has earned $32 trillion (€30 trillion) for investors since 1926. Of the 25,782 stocks in his sample, some $16 trillion (€15.1 trillion) in wealth creation was generated by just 86 companies. The top 1,000 stocks accounted for all of the $32 trillion wealth creation.

Put another way, one third of one per cent of stocks accounted for half of the overall market gains; less than 4 per cent accounted for all of the market gains; the remaining 96 per cent collectively generated lifetime dollar gains that merely equalled the amount earned through treasury bills.
 
Last edited by a moderator:
When Sarenco made this point originally, I had meant to research it, but never got around to it.

It's very important. My understanding was that from a portfolio of 10 blue chip stocks, you would pick something like 2 dogs which would fall by 100%, 6 shares which would give a reasonable return, and 2 stars which give a great return well outweighing the losses on the two dogs.

This research quoted by Fella seems to suggest that most shares generate nothing at all. And just 4% of shares generate a positive return.

If this is correct, then 66% of people who pick 10 shares would not pick one of the 4% stars. If my 20% stars is correct, 90% of 10 stock portfolios would include at least one star.

So it would be clear that despite the costs and despite the tax complexities, one should buy a collective fund. Hopefully if they have a few hundred shares, they will pick a few of the stars which will make up the return.
 
So, is Besseminder correct?

I presume that he is factually correct - 96% of all US stocks generated no return between 1926 and 2016. I am surprised by the figure, but does it tell us that much? Most of these would have been judged from their Initial Public Offerings. So if you picked 10 shares at complete random over the years, most would have done poorly.

It just does not seem consistent with the experience of those I know who have invested in shares. Most seem to have made a lot of money through shares. But maybe I have some sort of bias and am excluding all the people who have lost money.

I know people who have lost money. One person told me that his diversified portfolio of shares had fallen by 90%. I was shocked at this but when I investigated, his "diversified" portfolio was comprised of AIB, BoI, ILP, Anglo, CRH, Readymix, (the company which made radiators?) and UTV. This was, in my view, a three stock portfolio and not diversified.

What about my own experience? Over the years, I have had two shares which lost 80% - AIB and Power Securities (The capital disappeared almost completely but I did get a stream of dividends.) I have had a few which disappointed me. Sold at a loss or not much of a gain. But I have a "problem" now with capital gains. I have two loss makers. Aryzta which is down over 50% and an ISEQ ETF which is down about 20%. ( Oddly enough, Aryzta is a fallen star. It had risen to such an extent, that I reduced my holding as it accounted for over 10% of my portfolio. )

Was I just lucky? If so, maybe my luck will run out and my portfolio will generate no further return over the next 20 years.

I did not engage in any stock picking, so any outperformance would be luck instead of skill.

But I suspect it's because I was not picking from a universe of 25,782 shares. I have always advocated picking ten blue-chip shares. So I probably had screened out most of the speculative shares which go bust eventually.
 
Last edited:
I wonder if Bessbinder's statistics have been replicated in another mature stock market like Europe?

Does all the return come from just 4% of shares?

If one picked the top 100 shares by market cap in Europe 20 years ago, how would it look today?
 
Last edited:
Now, here is the embarrassing bit. I attach the Chapter I wrote in May 2001.

In summary, I recommended picking 10 of the Top Irish Companies
upload_2018-6-12_7-46-23.png


The top Irish companies in Table 5 are well diversified in terms of industry sectors and overseas earnings. Pick any ten of these shares which take your fancy. It's impossible to tell which will be the best performers over the coming years.



Three of the stocks are financials and three of the stocks are in the food industry. Don't select more than two from any sector or you could by overexposed to that sector.

(There seems to be an error here. 4 of the stocks were financials, not three. It says that three were in the food industry, although only two from that list. So I suspect that Waterford Foods might also have been in that list.)

In retrospect, I am very surprised at how many dogs were in that selection of 16.

It's hard to follow what happened to some of them, but I think that the following lost most of their value:

  • Elan
  • BoI
  • AIB
  • Anglo
  • Independent
  • Riverdeep
  • Iona
That a 50% complete failure rate.

  • IAWS
rose dramatically, converted to Aryzta and has since fallen by 80%. I don't know what the return has been since 2001.

I don't know how the following have done because of share splits, takeovers,etc. Does anyone know?

  • Smurfits
  • Galen
  • Viridian
  • CRH
The following have generated a good return:

  • Kerry Group
  • Ryanair
  • DCC

So, in retrospect, it's closer to

40% failure
40% OK
20% stars
 

Attachments

  • Chapter 5 How to invest in the Stockmarket.doc
    48 KB · Views: 265
Last edited:
How would an investment in the Index have done?

upload_2018-6-12_8-4-55.png

If you had restricted 20% of your portfolio to financial services, your return since 2001 would have been 68% (80% of 110% return on ISEQ General =168%)
 
Last edited:
Brendan Burgess said:
If you had restricted 20% of your portfolio to financial services, your return since 2001 would have been 88% (80% of 110% return on ISEQ General)
Click to expand...
Doesn't sound like you're allowing for negative return in that calculation? Would it not be closer to 70%?
 
Brendan Burgess said:
If this is correct, then 66% of people who pick 10 shares would not pick one of the 4% stars. If my 20% stars is correct, 90% of 10 stock portfolios would include at least one star.
Click to expand...
You had me struggling for a while there Boss but yes your math is correct. Fascinating to read that list of shares from 2001. I remember them well. And absolutely astounding that so many have tanked. I thought the banks were a once in a lifetime freak but Elan, Riverdeep, Iona. If ever there was a cautionary tale against a concentrated portfolio, this is it.
 
Duke of Marmalade said:
Fascinating to read that list of shares from 2001. I remember them well. And absolutely astounding that so many have tanked.
Click to expand...

It was very surprising for me too. I would love to get the data on all shares in Ireland which had a market cap in excess of €1 billion to see how they fared.

Here are today's top companies
CRH €27B
Ryanair: €18B
Kerry Group €16B
AIB €13B
BoI €8B
Paddy Power: €8B
Smurfit Kappa: €8B
Kingspan: €7B
Glanbia: €5B
Tullow Oil: €4B
Cairn Homes €1.4B
Dalata : €1.2B
Irish Continental :€1 B
Aryzta €1B


Will 5 of these crash to zero? Two or three probably will. And, there could be another banking crisis which takes out AIB and BoI.

Brendan
 
Duke of Marmalade said:
You had me struggling for a while there Boss but yes your math is correct. Fascinating to read that list of shares from 2001. I remember them well. And absolutely astounding that so many have tanked. I thought the banks were a once in a lifetime freak but Elan, Riverdeep, Iona. If ever there was a cautionary tale against a concentrated portfolio, this is it.
Click to expand...

Elan was always a high risk proposal given their fortunes were highly conditional on unpredictable clinical drug trials.
Riverdeep was a bottle of smoke even in 2001.
Iona was effectively an overgrown tech startup.

Brendan Burgess said:
It was very surprising for me too. I would love to get the data on all shares in Ireland which had a market cap in excess of €1 billion to see how they fared.

Here are today's top companies
CRH €27B
Ryanair: €18B
Kerry Group €16B
AIB €13B
BoI €8B
Paddy Power: €8B
Smurfit Kappa: €8B
Kingspan: €7B
Glanbia: €5B
Tullow Oil: €4B
Cairn Homes €1.4B
Dalata : €1.2B
Irish Continental :€1 B
Aryzta €1B


Will 5 of these crash to zero? Two or three probably will. And, there could be another banking crisis which takes out AIB and BoI.
Click to expand...

My money would be on Paddy Power, Tullow Oil and Dalata.
 
Last edited:
Brendan Burgess said:
So, is Besseminder correct?
Click to expand...
While it certainly true that the bulk of stock market gains come from a small minority of stocks, I would be inclined to treat the Bessembinder paper with some caution.

Firstly, he is saying that the best-performing 4% of publicly traded companies when stated in terms of lifetime dollar wealth creation, explain the net gain for the entire US stock market since 1926, as other stocks collectively matched Treasury bills. That's not the same thing as saying that "all market return comes from just 4% of stocks" and strikes me as a slightly odd way of measuring returns. In contrast, Dimensional Fund Advisors found that from 1980 to 2008, the top-performing 25% of stocks were responsible for all the gains in the broad US Stock Market and the bottom 75% of stocks collectively generated annual losses of around 2% over that period.

Secondly, the statistics reference the broad US stock market (as measured by CRSP). Currently roughly 80% of the stocks that constitute the total US stock market by number make up less than 10% of the market by capitalisation (i.e. there are a heck of a lot more small caps than large cap stocks). If you compiled the statistics using only large caps (S&P500), you would get materially less dramatic results.

Again, I don't want to detract from the general point that the bulk of stock market gains come from a small minority of "super-performers" (think Apple, ExxonMobile, etc.) but I do think the Bessembinder paper exaggerates the position somewhat.

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2900447
 
  • Like
Reactions: PMU
"All the market return comes from just 4% of shares"

Is this not a follow on from the fact that very few companies that existed in 1926 are still in existence today. Therefore is all the wealth generated since that time unfairly attributed to the "survivors". For example only two companies Microsoft and Cisco are still in existence today from the dot com era (or are still dow 100 companies). Therefore is all the wealth generated in the tech boom unfairly attributed to them now because they survived. In other words if you were lucky to have only invested in Microsoft and Cisco back in 2000 you would not have earned the big wealth earned since the dot com bust.
 
Are you sure about the 2 out of 100 figure?

I don't think that the other 98 companies disappeared. They probably merged and changed their names.

Brendan
 
Hi Colm

That is a fascinating article.

How have these original 30 companies fared? Quite well overall. Only four have gone bankrupt, with Eastman Kodak the latest example in January 2012, although, as mentioned above, the position was sold in 2011's third quarter after the share price fell below $1. Eighteen of those stocks still remain in the portfolio in some form. But many of these have morphed through mergers and acquisitions. Only five companies remain in their original incarnation: [broken link removed] [broken link removed], [broken link removed] [broken link removed], [broken link removed] [broken link removed], [broken link removed] [broken link removed], and [broken link removed] [broken link removed].

13% bankruptcies in 80 years is very low. In my sample there were 40% falling close to zero in 17 years.

And it's quite possible that the likes of Eastman Kodak produced a positive return on investment over the 70 years it was in the portfolio.

Brendan
 
Just wondering if investing in some of the different managed funds would give much the same %'age results?
 
A good article on the issue here

The Math Behind Futility
An overlooked statistical concept shows why it’s so hard to beat a benchmark.

It suggests that the under-performance by active managers is due to the positive skewness in the market.

I don't fully understand it. I would have thought that the if a few fund managers picked a higher than average proportion of the stars, their performance should be stellar. But I don't think that there are many stellar performers.

Don't most actively managed funds have around 100 stocks in them anyway? If so, they should pick a few of these star performers.

Positive skewness is a risk for someone who picks only 10 shares. They might miss out on the stars completely.

Brendan
 
Yes interesting stuff. The article states that the average growth of the S&P was 1.5% higher than the median growth. That is one mathematical definition of skewness. We are most familiar with it in the distributon of incomes. The average income is considerably higher than the median as the Donald Trumps easily outweigh the down and outs.

The premise of the article is that stock picking and indeed active management is purely a game of chance, and I am not going to quibble with that. So ignoring charges the active management population as a collective should match the benchmark. But that means the average is the same as the benchmark and so too is the median. The problem is that in comparing active management with passive management we mix the calibration. When we say the S&P grew 10% we are talking about an average. When we say the average active manager underperformed the S&P we are talking about the median.
 
Last edited:
I get it now. If 10 managers have the following performance
9 managers score a 4.
1 manager scores a 10
The average performance will be 4.6 so 90% of them will be worse than average.

But that is a side issue. The main point for investors is whether 10 shares is enough.

Brendan
 
You must log in or register to reply here.
Back
Top