Weekly Puzzle #3: To diversify or not to diversify

The Set up

Almost every risk and return model in modern finance is built on the presumption that ratiional risk taking requires you to spread your bets, i.e., to diversify. Some models, like the CAPM, take this to its logical limit, if there are no transactions costs and you cannot pick stocks, to supreme diversification, where you own a little piece of every traded asset in the market. In practice, though, there is substantial evidence that many investors not only choose not to diversify but believe that a more concentrated portfolio helps rather than hurts them. They are defended by some well known value investors who also believe that diversification is an admission that you cannot value companies and that you have no faith in your investing abilities. Since the measures of risk that we get from financial theory, i.e., beta or betas are all based upon the diversification argument, I would like you to think about where you stand on this question

The Arguments for and against diversification

If you have taken an introductory finance class, the argument for diversification is usually couched in statistical terms. As the number of holdings in your portfolio expands, the law of large numbers kicks in and risks that are holding-specific average out, yielding a portfolio that delivers a better expected return for any given level of risk. If you prefer a smarter and more qualified authority than I explaining this concept, here is the Nobel Prize talk delivered by Harry Markowitz, who initiated modern portfolio theory with his insights on diversification.

The push back against diversification is as much against passive investing as it is against diversification. In fact, it is the old time value investing camp that is most attached to concentrated portfolios, drawing on the deacon of value investing himself (Warren Buffett) to make its case. The argument is a simple one. If you can value companies, you should load up on your most under valued companies, and diversifying only results in diluting your winnings.

There is another argument against diversification and it comes from scanning the Forbes list of wealthiest Americans, many of whom ended up on that list because they founded companies that they own big chunks of still. Mark Zuckerberg and Larry Page are clearly not diversified, holding the bulk of their wealth in Facebook and Alphabet respectively, and the pathway to being like them seems to be do the opposite of diversification.

The Evidence on Investor Diversification or its absence

Individual investors may not read Markowitz or listen to value investors, but the evidence is overwhelming that the typical individual investor, who trades actively, holds few stocks (perhaps as few as three) in his or her portfolio. In this article, the authors chronicle some of the key characteristics of individual investors including their propensity to sell winners too soon and hold losers too long, to extrapolate past history and to hold concentrated portfolios.

Investment advisors and teachers have often viewed this as a problem and attributed it to either a lack of basic education on the mechanics of risk taking or to behavioral factors including hubris, greed and inertia. Whatever the reason, it is not restricted to just individual investors since institutional investors who cannot claim to be uneducated about diversifistation's benefits seem to be willing to also put all their eggs in a few baskets.

The Trade Off

I believe in diversification but I also think that some financial theorists are far too rigid in their defense of it, viewing any one who does not meet their wide diversification requirement as irrational or worse. To me, the question of how much you should diversify boils down to two simple questions:

  1. How much faith do you have in your assessments of value for assets? (Other things remaining equal, the greater your faith, the less you need to diversify)
  2. How much faith do you have that the market price will adjust to this value, and how quickly do you expect this adjustment to happen? (Other things remaining equal, the greater the faith that the price will adjust to the value and the sooner you see the adjustment, the less you have to diversify)
The limiting case where you don't have to diversify then is one where you are absolutely certain about both your value and in the market's imminent correction, in which case you can put all your money in that stock. As you get more uncertain about one or both these variables, the more diversified you need to get. This blog post expands on the implications. Put simply, if you choose not to diversify, I can see reasons why you may arrive at that conclusion rationally, though I would question the basis for your certainty.


Questions/ discussion issues

  1. Both sides of this debate use overwrought claims to make their case.
    - Some proponents of diversification claim that diversifying across assets and asset classes eliminates all risk from your portfolio. Explain why this is not true.
    - Some proponents of concentrated portfolios claim that if you diversify, all you can ever do is keep up with the market. Explain why this is not true.
  2. How many stocks/assets would you hold in your portfolio?
    a. One to two stocks/assets
    b. Three to five
    c. Five to ten
    d. Ten to Twenty
    e. Twenty to Fifty
    f. More than Fifty

  3. Explain why you chose the number that you did in the last question and how it may change over time (as you age, gain more investing knowledge, get wealthier).
  4. If you have an entrepreneurial streak and want to start your own business, you will end up undiversified, investing all your human and financial capital in that start up, at least to begin with. How do you reconcile this action with the argument for diversification?