April 25, 2015

Mail Bag: Index Investing vs. Picking Individual Stocks

This mail bag question has to do with index investing compared to investing in individual stocks when building a portfolio.

Hey Joshua, have enjoyed reading many of your posts for a while. You articles on about.com on the mysterious world of bond investing are some of the most lucid I have read — cleared a lot of things up for me.

Mail Bag Mail IconI read with interest a post linked today (Jan 10th, 2013) on the Wall Street Journal website, calling into question the entire “active trading” or “individual stock trading” approach advocated by many people, yourself included.

The post talks about Fortune Magazine’s “10 stocks to last a decade” — the sort of blue chip companies that you should just invest in and forget about. On that list were Nokia, Nortel, Enron, Oracle, Univision, Morgan Stanley etc. Of those 10 companies, half don’t exist any more, the others’ stock is decimated. If one had held the index, one would have been safer.

Do you not subscribe to this view? I’ve seen you talk about your brother and relative (flight attendant)’s retirement being secure since they have planted the seed of a nest egg in the blue chips, such as Apple. But who knows, Apple’s stock price could go the way of Nokia and Oracle and other seemingly infallible companies, wiping out the nest egg.

Perhaps Markowitz and Sharpe (and Fama et al.) were right when in the capital asset pricing model and its derivatives they said you can’t beat the market (or it is too risky to try to do so).

Would love to hear what you think, and how in your mind have you reconciled these two schools of thought.


A quick side note: I have never bought shares of Apple for any of my family members.  You are thinking of a friend of mine I once mentioned on the site.

That aside, before I get to the heart of your question, let me wax philosophical for a few moments.

Articles like the one you mentioned go back to the granddaddy of them all, the infamous Everybody Ought to Be Rich: An Interview with John J. Raskob by Samuel Crowther, published in the Ladies Home Journal, August 1929 edition.  In it, readers were urged to invest $15 per month (around $190 today) into stocks so that they, too, could someday have great wealth.  Unfortunately, the article forgot to mention that the price you pay matters – your return is going to depend on your base earnings yield, adjusted for growth, and be influenced by the total rate of inflation and taxation.

By way of example, you cannot pay $100 per share for a company earning $1 per share and expect to do well unless growth is at least 20% per year for a decade; an extremely difficult feat to accomplish.  Stocks, or any other asset class for that matter, are not a magical lottery ticket that will always do well for you.  Many Americans just learned this lesson in a very painful way when it came to real estate (although in the case of real estate, it is the ratio of monthly home payments to median monthly income for the given market area, adjusted for supply and demand, that matters, whereas with a stock or bond, it is the cash flows from the asset that drive the valuation).

The article was great as a lesson on compounding, but terrible for what it omitted.  In truth, had someone continued with the plan for the 50 years following the Great Depression, they would have grown very rich.  No one ever says that, though.  People point out the fact that shortly after beginning the program, stocks fell.  That’s actually good for a long-term buyer.

Addressing the article you cite, in particular: In almost every case of an essay like the one you mentioned – which are designed to get people to pick up the publication with the catchy headlines – the list of stocks given are nothing more than a list of popular stocks that everyone is talking about at the time.  Not one of those stocks you mentioned has ever been in any of my portfolios, nor any of the portfolios I’ve handled for friends and family.  The decade before that, I remember reading about all of the dot-coms everyone “had” to own.  None of them made their way into my portfolios, then, either, even though I was just a kid with a lot less money.

Why?  Because I can’t predict the cash flows.  I have absolutely no idea what Oracle will be earning ten years from now.  None.  If I can’t answer that question, I can’t value it.  Anything I do is speculation, which is fine – there is nothing wrong with speculation if you can afford it – but to call it “long-term investing” is a joke.  There might be people who can value Oracle.  I’m not one of them.  On the other hand, valuing Walmart or Amazon.com is much easier.  Looking at a firm like PepsiCo compared to those two is even simpler.

If I were to put together a list of stocks that would “last a decade”, it would have to pass a series of draconian checklists:

  • Does the company generate a majority of its revenues and profits from an activity that is simple to understand, has easy to calculate cost inputs, and a demand equation that is unlikely to change in the foreseeable future?
  • Are there any competitors that have a greater than small likelihood of significantly disrupting the existing market share allocation among established competitors?
  • Are there any unique regulatory issues that are likely to cause after-tax net margins to change, either positively or negatively?
  • Is there any significant chance of equity dilution?
  • Is the balance sheet strong with little relative debt and a lot of cash and / or cash generation?
  • Are the products and services protected by strong barriers to entry, patents, trademarks, or other significant safeguards?
  • Is the company so boring that few people will ever be attracted to it?
  • Do the after-tax net earnings of the business consistently exceed the total maintenance capital expenditures required to keep unit volume steady?  If the answer is “no”, the business is really a Ponzi-scheme of sorts, paying dividends to equity holders, which are funded by new bond issuances raised from the debt markets, with profits going to maintain inflation and not much more.  That doesn’t interest me.

As we discussed yesterday, the companies that would make this list would be firms like General Mills, Clorox, Colgate-Palmolive, Nestle, Procter & Gamble, or Coca-Cola.  Next would come starting valuation.  As I pointed out many times in the past, a great business is not necessarily a great stock at any given time.

Short of some drastic market upset, it’s easy to get a general idea of what each of those companies will look like a decade from now.  I have no idea what Apple will look like a decade from now.  None.  It could be lucrative.  Or not.  It could be the largest company in the world.  Or not.  Clorox hasn’t changed in a century.  The iPhone changes every year.

These types of businesses tend to cluster around certain markets.  The food industry is relatively stable thanks to trademarks – if you want an Oreo, no one else can sell you an Oreo except the firm who has ownership of the name and ingredient list.  The restaurant industry is not.  The banking industry doesn’t qualify because you are dealing with a lot of debt and relatively small amounts of equity.  Consumer staples fit perfectly but consumer durables don’t.  Then there are specific companies that have unique advantages – Compass Minerals of Kansas City, for example, is one of only a tiny handful of businesses in the world that provide salt to highways to de-ice following winter weather.  That’s not a business that is likely to change much in the next ten years.  There aren’t many competitors.  If the price is right, it’s not that hard to figure out whether you will make money or not.

Addressing Your Broader Question About Using an Index Strategy vs. Individual Stock Selection

Whenever someone says that a particular portfolio did, or did not, beat a particular index, it always amuses me.  Logically, what they are saying is, “This basket of stocks did not perform like that basket of stocks.”  Of course it didn’t.  They are not made up of the same assets, at the same time, at the same price points.

Imagine that I woke up at 25 years old, with no debt, and a suitcase full of $100,000 in cash.  I was given a choice between putting it in the Dow Jones Industrial Average or investing it myself.

First, you have to understand that there is no such thing as the Dow Jones Industrial Average.  There is an imaginary list of companies chosen by the editors of The Wall Street Journal.  Were I to buy an index fund that matched the Dow, my portfolio would consist of the following holdings:

  • $11,000 worth of IBM
  • $6,300 worth of Chevron
  • $5,520 worth of 3M
  • $5,420 worth of Caterpillar
  • $5,210 worth of McDonald’s
  • $5,080 worth of Exxon Mobil
  • $4,840 worth of United Technologies
  • $4,390 worth of Boeing
  • $4,240 worth of Travelers
  • $4,120 worth of Johnson & Johnson
  • $3,950 worth of Procter & Gamble
  • $3,900 worth of Wal-Mart Stores
  • $3,630 worth of Home Depot
  • $3,470 worth of American Express
  • $3,030 worth of UnitedHealth Group
  • $2,900 worth of Disney
  • $2,630 worth of DuPont
  • $2,630 worth of JP Morgan Chase
  • $2,480 worth of Verizon
  • $2,440 worth of Merck
  • $2,110 worth of Coca-Cola
  • $1,960 worth of AT&T
  • $1,530 worth of Pfizer
  • $1,510 worth of Microsoft
  • $1,240 worth of Intel
  • $1,210 worth of General Electric
  • $1,170 worth of Cisco Systems
  • $930 worth of Hewlett Packard
  • $670 worth of Bank of America
  • $510 worth of Alcoa

When you say, “Why wouldn’t you want to own the index?” what you are asking me is, “Why wouldn’t you want you assets allocated exactly like this?”.  To repeat: There is no index.  It’s a fictional idea to help rationalize the concept of equity ownership in a way people can perceive more easily.

Looking at that list, there are certain types of companies that perform better over long periods of time, provided the price is rational, than other types of companies.  I am not, for example, interested in owning Alcoa at any price, unless I am speculating with it as a short-term trade due to cyclical factors.  That is not a firm I am going to buy and hold.  The returns on equity are not high enough to offset the necessary capital investment and you cannot get a decent increase in real, inflation-adjusted earning power over time.

On the other hand, I’d love to own Johnson & Johnson because it generates real, inflation-adjusted increases in wealth.  Whether I wanted it added to my portfolio would depend on whether the base earnings yield, normalized for one-time events and tax law, at least exceeded that available on the 30-year Treasury.

Likewise, why would I want to own $5.21 worth of IBM for every $1.00 I owned worth of Coca-Cola?  Both are good businesses but I do not find IBM 5.21x cheaper or more attractive than Coke at present levels.  That seems like an irrational allocation decision; the numbers do not warrant it.  The only reason it is this way is because the Dow weights stocks based on share price, which is entirely arbitrary!  IBM is now $192.88 per share and Coca-Cola is $36.96 per share due to Coke’s stock split last year.

For me, given my ability to understand advanced accounting, and my joy of owning specific businesses, I would be happier if I were to split the $100,000 into no fewer than 20 piles of $5,000 a piece and no more than 50 piles of $2,000 a piece (depending on how cheap stocks were in general at the time), allocating the money to businesses that were not likely to change, that had long histories of making a lot of money for owners, and that returned a significant portion of cash in the form of dividends.  Those dividends would serve as a counterbalance during bear markets, as well as provide fresh capital to reinvest to take advantage of the lower prices, assuming I wasn’t allowed to add additional outside money to the portfolio.

For Most People, Indexing Is Still the Best Strategy

This may sound like an odd statement given what I just said, but for most people, indexing is the superior strategy.  If something were to happen to me, I would want my family to put everything into a low-cost index fund at Vanguard, reinvest or live off the dividends, and ignore it for long periods of time.  I wouldn’t want them selecting individual equities.  That’s because the list of stocks in the Dow Jones, overall, is very good.  It is.  For someone who can’t value a business, owning that basket makes a lot of sense.  (With the portfolio I own now, they wouldn’t have to make the switch right away.  The stocks held in our accounts are among the most stable, blue chip firms in the world, bought at attractive prices over the past few years.  The portfolio should continue to run itself with no turnover for the next decade.)

While we are on the subject, consider this paradox: If a significant majority of investors began to index, indexing would no longer work for a lot of structural reasons as the underlying equity market would break down and become non-functional.  It’s also interesting that indexes are far less diversified than people like to think.  The Dow Jones Industrial Average contains 30 stocks, but 56% of the assets are held in the top 10 of those stocks.

I sleep well at night 1.) knowing what I own, 2.) knowing why I own it, 3.) being able to compare the yearly performance of the underlying business to the expectations built into my valuation model to determine how accurately I discounted the cash flows and thus, calculated intrinsic value, 4.) being tied to specific businesses in the same way I love the private companies I own.  I love the fact that when I walk into a movie theater, it is almost impossible for me to get from the ticket booth to the seat without encountering at least four companies that are paying me cash dividends every year.  I’m just as much an owner, and feel that way, for those firms as I am the ones under my control.  I know their sales, their profit margins, their returns on capital, who is running the management team, and the price I paid for my stake.

If I had to buy a basket of stocks chosen for me by the newspaper editors, it would be boring.  I certainly wouldn’t be writing and this blog would never discuss finance.  I love understanding working systems and economic models.  Businesses let me do that while increasing my purchasing power.  I find it enjoyable, and I’m good at it.  Why change?

For more information, you may want to read this mail bag question from the past.

  • Paul Sowden

    The choice of index to base this post on doesn’t seem to me like it’s representative of how mutual fund investors actually invest. The Dow Jones Industrial Average is relevant primarily due to its longevity[1]. By far the most popular index is the S&P 500, with 34% of index fund assets invested in funds indexed to the S&P 500[2].

    While the S&P 500 is also selected by a committee (rather than being strictly based on measurable stock attributes), the biggest points of contrast to the Dow Jones Industrial Average are that it is much broader, representing around 75% of the total market by capitalisation, and that it is cap-weighted, so each company is held proportional to its market capitalisation. The typical investor that wants to invest in a mutual fund that represents the US equity market invests in a mutual fund that tracks the S&P 500.

    In general, in my opinion, when you move outside of indexes that represent broad market segments (e.g. equities by capitalisation and geography; bonds by duration and quality) then you start moving away from the spirit of passive indexing, that is attempting to “match the market”, and into speculative indexing, by trying to find indexes that “beat the market”.

    [1] According to Wikipedia it’s the second oldest US equity index after the Dow Jones Transportation Index http://en.wikipedia.org/wiki/Dow_Jones_Industrial_Average
    [2] At year end 2011, see Figure 2.11 http://www.icifactbook.org/fb_ch2.html#index

    • http://www.joshuakennon.com/ Joshua Kennon

      Your statement is absolutely true when you say, “The choice of index to base this post on doesn’t seem to me like it’s representative of how mutual fund investors actually invest. ”

      However, this is one of those rare cases where it doesn’t matter. (Anytime you hear someone say that, you should be skeptical unless they provide the reasoning so I’ll explain.)

      Going back over the past half century, the Dow Jones Industrial Average and the S&P 500 have returns that correlate roughly 95% of the time. During those 50 years, the S&P 500 came out on top 26 times and the Dow 24 times, while long-term returns are within a few hundred basis points.

      The reason is simple: The S&P 500 is not equally weighted due to the impracticality and impossibility of achieving that. The smaller capitalization stocks could not absorb the kind of asset levels that are currently parked in S&P 500 based index funds.

      Although people often know this academically, they fail to grasp the implications. There is this naive idea that the S&P 500 is somehow safer than the Dow Jones Industrial Average because the former contains 500 stocks whereas the latter “only” 30 – I suffered from this misconception myself when I was younger and less experienced. Yet, the S&P is not really diversified in the way most people think it is (a fact to which you alluded when you discussed the size of the enterprise being the weighting variable rather than the stock price, as it is in the Dow).

      To illustrate: For the S&P 500, the top three stocks – Apple, Exxon Mobil, and General Electric – accounted for 10% of the portfolio value as of last year!

      To put it in raw numbers, a $100,000 investment in the S&P 500 means that you are going to have $10 invested in Big Lots, $20 invested in Pitney Bowes, $20 invested in J.C. Penny, $10 invested in Advanced Micro Devices, etc. Meanwhile, you will have $10,000 parked in Apple, Exxon Mobil, and General Electric combined. That is not diversification as most people envision it.

      Another point: Looking at the top 10 holdings in the S&P 500, you see Apple, Exxon Mobil, General Electric, Chevron, Microsoft, IBM, AT&T, Google, Procter & Gamble, and Johnson & Johnson.

      All of those stocks except for Apple and Google are in the Dow Jones Industrial Average.

      If something were to happen to cause those key stocks -the ones that make up the Dow Jones Industrial Average and the top 10 holdings in the S&P 500 – collapse, the relatively small diversification advantage of the latter falls apart because people flock to the strongest, biggest companies when the world goes to hell in a handbasket. Those other 490 stocks aren’t going to do much, if any, good. If anything, they’d be the first to fold or get taken private, locking out equity holders from future recovery.

      The bottom line is there is absolutely no meaningful difference between the Dow Jones Industrial Average and the S&P 500, either in risk profile, historical returns, or even the type of equities to which the fund itself is exposed. (This would not be true if you were talking about an equally weighted S&P 500 portfolio, which does not exist in a sustainable, accessible way.)

      Given that basic truth, it saved time to limit the response to the 30 Dow stocks rather than try to enter all of the S&P 500 components. That’s what I did.

      But you’re right. Most investors opt for the S&P 500 over the Dow even though, frankly, there isn’t much reason to do so. Long-term, given the existing index methodologies and components, there isn’t going to be much, if any, difference between the ultimate wealth acquired through either investment strategy.

      • Rudy

        Great answer Joshua, thanks! Also, after all you recommendations I finally picked up Graham’s Security Analysis. Ordered a few others from your recommended reading list post on this site.


        • http://www.joshuakennon.com/ Joshua Kennon

          Let me know how it goes! Which edition year did you buy? Each is different. Although it’s now out of date, 1942 is one of the best but the most recent one is also very good. If you are dealing with all but the most recent edition, realize that the accounting rules today are vastly different. What you are interested in is his approach – how he thinks about certain things, like moving up the capitalization structure or comparing accruals to actual cash earnings. The theory is the important stuff. Once you understand modern GAAP, wedding the two is nearly instant.

        • Michael Starke

          My copy of sixth edition (2009 I believe) just arrived (literally two minutes ago). It has an accompanying CDROM with selected chapters from the second edition (1940). This is one book on the recommended reading list that I had never picked up, because I thought it might be a bit too deep for me, but I think I’m up to the challenge now.

  • Connelly Barnes

    In theory couldn’t indexing work if nearly everyone indexed provided that it is done on a fundamentally-weighted basis?

    • http://www.joshuakennon.com/ Joshua Kennon

      It would depend on the market capitalization parameters employed (which is one of the primary reasons it would be practically impossible for an equally-weighted S&P 500 index fund to ever work on any sufficiently grand scale without engaging some sort of synthetic derivative that mirrored results, which then introduces all sorts of risks).

      Imagine if the Vanguard 500 Index Fund – which has almost $183 billion in it alone and it is far from the only index fund in the world – weighted its portfolio so that the top component weighting was comparable to the current setup, only instead of market capitalization, what mattered was the aggregate score on some sort of passively-measured quantitative ranking algorithm (e.g., ROE, ROA, adjusted by debt-to-equity, growth in EPS, whatever; it’s not important for our purposes at the moment). Right now, the S&P 500 has a 3.43% weighting to Apple, meaning ~$1.00 of every $29.15 investors put in the index fund gets allocated to the tech firm.

      It is highly likely that many of the top weighted components under the new fundamental ranking system would have much smaller market capitalizations. Imagine if it happened to be Urban Outfitters, the smallest S&P 500 component. The entire company is valued at $4.64 billion. To get the 3.43% weighting, Vanguard’s index fund alone – not even counting all those other index funds from Fidelity, etc. – would have to buy $6.28 billion worth of Urban Outfitters. The stock doesn’t exist. It’s not possible.

      “Let’s lower the weightings, then!” you say. Okay. We do that. With each drop in the top weighting, though, you remove the advantage of a fundamental weighting calculation in the first place until you finally arrive at equally weighted, where every stock represents 1/500th of the fund, or 0.20%. That is still $366 million per position for Vanguard alone, not to count all of those other firms. There isn’t the float for that sort of thing, to have “dumb money” buying and selling automatically based on newly issued shares and redemptions in the index funds. You’d get these impossible-to-handle volatility situations and companies / their long-term shareholders wouldn’t stand for it.

      It’s one of those problems I can’t find my way around in any truly satisfactory way. Indexing is the best solution for a huge majority of society, yet it will be doomed to fail the moment a sufficient majority of the population adopts it because the aggregate assets exceed what the bottom segment of components can absorb in their share float.

      • innerscorecard

        Do you agree with what John Bogle said about the relative out-performance of Research Affiliate’s fundamentally weighted indexes, that this out-performance will only last until it doesn’t (due to becoming too big to allocate money this way)?

      • Robert from Ct.

        Joshua, Your thoughts on RSP which is EQUAL weighted the S&P 500 ?
        PS— another index worth looking into is SCHD