3 Reasons to Think Twice About Concentration

In a recent Kiplinger piece, James K. Glassman argued that it is better to own fewer stocks and concentrate your portfolio. This is an increasingly popular opinion among (value) stockpickers. There’s some truth to it, but at the same time I’d call concentration a weapon of portfolio destruction.

Glassman supports his argument in three ways: diversification in 20 stocks works just as well as a thousand, it doesn't make sense to invest in your 20th best idea and studies show returns would rise if managers would concentrate further. I’ll go over the arguments, including a key quote, one by one and argue against them:

20 stocks diversify just as well as a a thousand

"For instance, Edwin Elton and his colleagues conclude in their textbook, 'Modern Portfolio Theory and Investment Analysis,' that if you own 1,000 stocks, your portfolio will be 61% less volatile than if you own just one stock. But if you own 20 stocks, your risk falls 59%, or nearly as much as it declines with 1,000 stocks."

This is an important notion for sure. But instead of worrying about what doesn’t work, do something that does work. If you just add a lot of stock, it doesn’t work if you don’t evaluate that investment on its ability to diversify your portfolio. What you’d like to do is add investments that have an idiosyncratic profile or a very different profile from your existing portfolio profile. The S&P 500 isn’t that diversified itself. It is about 24% tech, for example, and only 3.19% basic materials:

Data: Morningstar

If you are adding stocks based purely on valuation, it’s true you aren’t very likely to diversify your portfolio very much beyond the first 20 stocks. It’s even worse if you are adhering to your “circle of competence” (a popular idea) and add stuff from within just a few industries.

If you generate a lot of alpha because of that practice, as many great investors do, you are still in an awesome position. But if you aren’t adding much if any alpha, you are doing yourself a terrible disservice.

The risk of your portfolio could be much lower without you giving up much. Random stocks don’t add much in terms of diversification but if you add something like real estate, a shipping stock, an early stage biotech, a patent troll, liquidation play, an M&A play or a commodity stock (especially one based on precious metals), you are likely diversifying your portfolio much more effectively compared to adding your 21st stock randomly.

The aforementioned type of investments tend to be idiosyncratic in nature. meaning they beat to their own drum. A biotech becomes more valuable on FDA approval whether the market is good or bad. The M&A is only correlated so much with the general market. A liquidation play distributes its cash when it can, not when the markets are booming. A lawsuit is won or lost when the judge rules. Commodity stocks tend to do well when the price of the underlying commodity soars, which isn’t a 1:1 correlation with the market.

There are many more examples. One of my favorite investments is Oaktree Capital (OAK, Financial), which is a private equity firm specializing in distressed debt investing. The firm, led by Howard Marks (Trades, Portfolio), gathers a lot of assets in downturns and reaps the rewards in the years after. Since asset managers are judged on assets under management, that makes it an interesting candidate to hold up well in a recession. If there is no recession, it isn’t so bad with a trailing 12-month yield of 8.68%.

Why bother with your 20th best idea?

"As Warren Buffett (Trades, Portfolio) wrote in the 1993 annual report of Berkshire Hathaway: 'If you are … able to understand business economics and to find five to ten sensibly-priced companies that possess important long-term competitive advantages, conventional diversification makes no sense for you. It is apt simply to hurt your results and increase your risk. I cannot understand why an investor of that sort elects to put money into a business that is his 20th favorite rather than simply adding that money to his top choices.”‰… In the words of the prophet Mae West: ‘Too much of a good thing can be wonderful.’”

Buffett is right of course. I rarely argue against him about any investment issues, and when I do I’m probably wrong. Suppose you are able to understand business economics and find five to 10 sensibly priced companies (good luck on that one) that possess important long-term competitive advantages (again, good luck) then conventional diversification makes no sense.

Buffett, in a nutshell, is saying that if you are a superinvestor, conventional diversification makes no sense.

You don't just have to be a superinvestor, but one with a very specific strategy (e.g., Buffett's GARP).

Let's say for example you specialize in stub investing as described in Joel Greenblatt (Trades, Portfolio)’s book, "You Can Be a Stock Market Genius." Stubs are highly leveraged companies that only have a very small piece of equity as part of the overall enterprise value. These represent a zero-or-hero type of situation. If you invest in only five zero-or-hero-type situations and continue that practice, a string of bad luck is going to destroy your long-term results. That’s an extreme strategy as an example but if you invest heavily into microcaps, this is another type of stock of which I would buy more than five.

Trimming portfolios improves returns

"These funds all support the notion that fewer choices are better choices. Further proof is an elegant study by Danny Yeung, of the University of Technology in Sydney, Australia, and three other researchers. They found that returns would rise sharply (with little effect on risk) if fund managers would just pare the number of stocks they hold to their 20 to 30 favorites."

This could be explained in part because of the narrow mandates of mutual funds. These guys often have to add stocks from a certain universe or select from within a benchmark. The 21st idea doesn’t make a big difference for diversification purposes.

But if you focus on diversifying your portfolio, it can make a big difference. For example, you could buy a closed-end fund trading at a discount to net asset value that holds bonds.

Maybe that’s cheating, but it goes to show how a little creativity and effort can accomplish things academic research would suggest are impossible. It also works partly because these managers know they would do better by concentrating but if they go down 20%, and they own 10 stocks, they could be fired. On the other hand, adhering to a more conventional approach may keep them behind the steering wheel in such a situation.

This is a fringe case, but some funds have too much assets to be able to concentrate them without problems, like triggering the 10% ownership threshold. Of course funds focused on U.S. large caps don’t suffer from this, but others can.

Disclosure: Author is long OAK.

• Howard Marks Undervalued Stocks
• Howard Marks Top Growth Companies
• Howard Marks High Yield stocks, and
• Stocks that Howard Marks keeps buying
• Joel Greenblatt Undervalued Stocks
• Joel Greenblatt Top Growth Companies
• Joel Greenblatt High Yield stocks, and
• Stocks that Joel Greenblatt keeps buying