Author Topic: How do index funds determine the amount of each holding in the portfolio?  (Read 1202 times)

Dancin'Dog

  • Handlebar Stache
  • *****
  • Posts: 1759
  • Location: Here & There
I've realized that index funds favor the top listed holdings in the portfolios based on weighting, which I assumed meant the portfolios held more shares, or higher dollar values, of the biggest winners. 


I have recently been interested in individual stocks, and it's obvious that NFLX has dominated the market over the past decade.  When I checked VTSAX's list of holdings I was surprised to see that NFLX was only ranked in the 27th spot. 


Can someone explain why NFLX isn't towards the top of the list?  There's $35.5B of MSFT and less that $5B of NFLX. 


According to portfoliovisulizer.com $10K invested in NFLX in 2002 would be worth almost $5M vs $94K for MSFT.  In fact, $10K invested in MSFT in 1985 still wouldn't be worth $1M yet.

Steeze

  • Handlebar Stache
  • *****
  • Posts: 1205
  • Age: 36
  • Location: NYC Area of Earth
VTSAX tracks a market capitalization weighted index, meaning the fund holds shares in proportion to the size of the company (price x # of shares). Apple and Microsoft are much larger companies than Netflix so the fund holds more of Apple and Microsoft shares. If Netflix grows so much that it is the largest company in the US, then VTSAX will hold more Netflix than anything else.

And yes - if you invested in Netflix in 2002 you are doing well. Imagine investing in Microsoft in 1987 for a similar comparison.

secondcor521

  • Walrus Stache
  • *******
  • Posts: 5503
  • Age: 54
  • Location: Boise, Idaho
  • Big cattle, no hat.
    • Age of Eon - Overwatch player videos
Most index funds, including VTSAX, are weighted by market capitalization.

The market capitalization of each stock is simply the number of shares outstanding multiplied by the stock price.

The market capitalization of the index is simply the sum of the market caps of all the individual stocks that make up that index.

Each stock in an index fund is held in an amount proportional to that stock's market capitalization compared with the total market capitalization of the whole index.

So as a random example, if Microsoft's market cap is 5% of the market cap of the total of all funds in the index, then the fund would invest 5% of their assets in MSFT.

As the relative proportions of the market caps change, the mutual funds adjust their weighting.  However, note that a lot of this is take care of automatically.  If Microsoft's stock price were to double tomorrow, then it's market cap would then be closer to 10% of the index.  But the shares of MSFT that an index fund holds today would double in value tomorrow as well, so would similarly represent 10% of their holdings.  It's not like the doubling of the stock tomorrow would cause the index fund to have to go out and double the amount of shares that they own.

Dancin'Dog

  • Handlebar Stache
  • *****
  • Posts: 1759
  • Location: Here & There
That's for the explanations.


It looks like you could just choose about 20 of the better performing stocks for the past 5 or 10 years and equally weight them to make a portfolio that would beat the S&P most years.  But I guess it's not quite that simple?




Steeze

  • Handlebar Stache
  • *****
  • Posts: 1205
  • Age: 36
  • Location: NYC Area of Earth
If you figure out the magic recipe you can get a 7 or 8 figure salary and handsome bonus structure here on Wall Street. Just keep in mind you are competing against all the smartest financial quants, mathematicians, physicists, computer scientists, and AI on the planet who have all been working to solve this problem and crack the code for decades.

Dancin'Dog

  • Handlebar Stache
  • *****
  • Posts: 1759
  • Location: Here & There
If you figure out the magic recipe you can get a 7 or 8 figure salary and handsome bonus structure here on Wall Street. Just keep in mind you are competing against all the smartest financial quants, mathematicians, physicists, computer scientists, and AI on the planet who have all been working to solve this problem and crack the code for decades.




Oh, I know that it's supposed to be almost impossible to beat the market long-term. 


Just for a simple test, I just now took the first 20 holdings listed for VTSAX and put them into portfoliovisualizer and weighted them equally for the years 2008-2020 and they yielded twice what VTSAX did, beginning with a fixed amount.  With monthly contributions VTSAX did a little better, but not much.

Beginning with $10K & adding $1K per month = $478K vs $300K
&
Beginning with $10K & adding nothing else =  $45.5K vs  $22.6K


(Note: Facebook & Visa data didn't go back to 2008, so replaced them with the next symbols down the list)

I don't know exactly what this experiment proves, but it does raise questions about whether VTSAX is all that hard to beat.  Can I move to Wall St & get paid the big bucks??  I think I could tweak my secret "list of 20" a little and do even better.  Heck, Netflix isn't even on that list!  Where do I send my resume??  ;)

I know VTSAX is comprised of 3500 different stocks, but I'm not sure how much the lower tier ones contribute.  From what I gather the large-caps tend to outperform the mid & small caps over the long-term. 

Have you guys played with portfoliovisulizer?  I'm sure picking a winning portfolio isn't this easy.  How's this thing misleading me? 

Travis

  • Magnum Stache
  • ******
  • Posts: 4219
  • Location: California
If you figure out the magic recipe you can get a 7 or 8 figure salary and handsome bonus structure here on Wall Street. Just keep in mind you are competing against all the smartest financial quants, mathematicians, physicists, computer scientists, and AI on the planet who have all been working to solve this problem and crack the code for decades.




Oh, I know that it's supposed to be almost impossible to beat the market long-term. 


Just for a simple test, I just now took the first 20 holdings listed for VTSAX and put them into portfoliovisualizer and weighted them equally for the years 2008-2020 and they yielded twice what VTSAX did, beginning with a fixed amount.  With monthly contributions VTSAX did a little better, but not much.

Beginning with $10K & adding $1K per month = $478K vs $300K
&
Beginning with $10K & adding nothing else =  $45.5K vs  $22.6K


(Note: Facebook & Visa data didn't go back to 2008, so replaced them with the next symbols down the list)

I don't know exactly what this experiment proves, but it does raise questions about whether VTSAX is all that hard to beat.  Can I move to Wall St & get paid the big bucks??  I think I could tweak my secret "list of 20" a little and do even better.  Heck, Netflix isn't even on that list!  Where do I send my resume??  ;)

I know VTSAX is comprised of 3500 different stocks, but I'm not sure how much the lower tier ones contribute.  From what I gather the large-caps tend to outperform the mid & small caps over the long-term. 

Have you guys played with portfoliovisulizer?  I'm sure picking a winning portfolio isn't this easy.  How's this thing misleading me? 

Important to remember there's a difference between biggest company and fastest growing company.  Apple and Amazon have higher market caps than the GDP of many nations, but if their stock growth is flat, then you're not making much money off of them.  The trick is to find these companies before they make it big and ride that wave to the top.  There was a term in the 1970s called the "nifty fifty." It was thought that buying the largest 50 stocks in the index was the safe move and you could just sit on their dividends or enjoy the growth they'd continue to have. Only a couple of those 50 are still in business today.

Every company in the Dow or S&P500 was a small cap company once upon a time.  We buy them all because we don't know which ones will rise to the top. They may not "contribute" much today, but in 10 years you might have a golden ticket or two in there.  Unless you can predict which one it'll be, you take the good with the bad and results still tend to be satisfying.
« Last Edit: April 07, 2020, 10:44:32 PM by Travis »

maizefolk

  • Walrus Stache
  • *******
  • Posts: 7400
The biggest companies today are naturally going to be ones that grew a lot (or at least didn't shrink) over the last decade. The biggest companies of 2008 haven't done as well over the 2008-2020 time frame as the companies that ended up the biggest in 2020.

Back in 2008 the two biggest companies by market cap, and the two companies that made up the biggest shares of VTSAX, were ExxonMobile and General Electric.

Michael in ABQ

  • Magnum Stache
  • ******
  • Posts: 2626
If you figure out the magic recipe you can get a 7 or 8 figure salary and handsome bonus structure here on Wall Street. Just keep in mind you are competing against all the smartest financial quants, mathematicians, physicists, computer scientists, and AI on the planet who have all been working to solve this problem and crack the code for decades.




Oh, I know that it's supposed to be almost impossible to beat the market long-term. 


Just for a simple test, I just now took the first 20 holdings listed for VTSAX and put them into portfoliovisualizer and weighted them equally for the years 2008-2020 and they yielded twice what VTSAX did, beginning with a fixed amount.  With monthly contributions VTSAX did a little better, but not much.

Beginning with $10K & adding $1K per month = $478K vs $300K
&
Beginning with $10K & adding nothing else =  $45.5K vs  $22.6K


(Note: Facebook & Visa data didn't go back to 2008, so replaced them with the next symbols down the list)

I don't know exactly what this experiment proves, but it does raise questions about whether VTSAX is all that hard to beat.  Can I move to Wall St & get paid the big bucks??  I think I could tweak my secret "list of 20" a little and do even better.  Heck, Netflix isn't even on that list!  Where do I send my resume??  ;)

I know VTSAX is comprised of 3500 different stocks, but I'm not sure how much the lower tier ones contribute.  From what I gather the large-caps tend to outperform the mid & small caps over the long-term. 

Have you guys played with portfoliovisulizer?  I'm sure picking a winning portfolio isn't this easy.  How's this thing misleading me? 

Go pick the top 20 holdings from some prior year and you'll see that some of them have shrunk considerably or gone out of business, while others that have grown thousands of percent are nowhere to be seen. Companies like Amazon, Apple, Google, etc. have crushed the returns of old "blue chip" companies like GE or Ford or Exxon.

Just take it as another lesson that it is very very hard to consistently beat the market and just by buying that index fund and matching the market you're well ahead of almost everyone in the long run.

Telecaster

  • Magnum Stache
  • ******
  • Posts: 3551
  • Location: Seattle, WA
You absolutely can come up with stock screens that outperformed over 10-15 years, or whatever period you like.

The problem is the out of sample performance.  That is, how did the screen work after it was discovered?  What tends to happen is that you discover a screen that was red hot for the last ten years, but then becomes ice cold going forward.

That said, I absolutely encourage you to take a deep dive into this.  Philosophically, the advice is to buy the index.  Why is this index better than that index?  Looking at stock screens greatly sharpened my viewpoint in this regard and made me a better investor.

MustacheAndaHalf

  • Walrus Stache
  • *******
  • Posts: 6633
Vanguard Total Stock market weighs each stock by multiplying price/share by the number of shares.

Over the past 5 years, the S&P 500 gained +7% per year, while Amazon stock (AMZN) gained +40% per year.  Amazon didn't start at the #3 position, it grew into the #3 spot on the S&P 500.  If you picked the previous #3 stock, you would have missed Amazon's growth.

Dancin'Dog

  • Handlebar Stache
  • *****
  • Posts: 1759
  • Location: Here & There
Vanguard Total Stock market weighs each stock by multiplying price/share by the number of shares.

Over the past 5 years, the S&P 500 gained +7% per year, while Amazon stock (AMZN) gained +40% per year.  Amazon didn't start at the #3 position, it grew into the #3 spot on the S&P 500.  If you picked the previous #3 stock, you would have missed Amazon's growth.




The "top 20" list that I used was today's leaders, which would have required a time machine or crystal ball to possess ten years ago.  I'm don't know where to find the VTSAX portfolio lists from previous years, but it would be interesting to see how they would compare using the same top 20, equal weighting backtesting. 


I've been using https://www.finscreener.com/screener/stock-screener to view the performance rankings for various timeframes.  But I haven't figured out how to review the rankings from previous dates.  That seems like valuable info to study, so I'll ask their developers how to do that. 








maizefolk

  • Walrus Stache
  • *******
  • Posts: 7400
The "top 20" list that I used was today's leaders, which would have required a time machine or crystal ball to possess ten years ago.  I'm don't know where to find the VTSAX portfolio lists from previous years, but it would be interesting to see how they would compare using the same top 20, equal weighting backtesting. 

Here are the companies with the biggest market caps in 2008:

https://money.cnn.com/magazines/fortune/fortune500/2008/performers/companies/biggest/

Dancin'Dog

  • Handlebar Stache
  • *****
  • Posts: 1759
  • Location: Here & There
The "top 20" list that I used was today's leaders, which would have required a time machine or crystal ball to possess ten years ago.  I'm don't know where to find the VTSAX portfolio lists from previous years, but it would be interesting to see how they would compare using the same top 20, equal weighting backtesting. 

Here are the companies with the biggest market caps in 2008:

https://money.cnn.com/magazines/fortune/fortune500/2008/performers/companies/biggest/




Thanks.  I'll plug those in later and see how they did. 

Steeze

  • Handlebar Stache
  • *****
  • Posts: 1205
  • Age: 36
  • Location: NYC Area of Earth
Maybe if you want to catch the growth story you use the 50 newest companies to be listed on the S&P 500