It’s long been said that portfolio diversification is the closest thing to a “free lunch” that an investor can get. However, the suggestion that more is better may be somewhat misleading. Even assuming that diversification provides the risk-reduction benefits that proponents claim, it is not just a question of how many positions are involved. In fact, it is possible that a portfolio perceived as being well diversified may, in fact, be anything but.

There is some debate about what it means to be diversified. In terms of equities, economist and author Burton Malkiel, who once advocated a 20-position portfolio, now says the total should be 30 or perhaps more because stock volatility has risen over the past several decades. In contrast, financial theorist William Bernstein maintains that even 200 may not be enough. In his view, the Irish Times noted, only “a handful of ‘superstocks’ have historically accounted for the bulk of market returns.”

That said, researchers E.J. Elton and M.J. Gruber, who studied diversification in the late 1970s, found that most of the benefit of increased diversification, measured by the standard deviation of returns, goes away once a portfolio has between 20 and 30 securities in it, according to Forbes. The magazine added that the pair “included data for 500- and 1,000-security portfolios but the drop in risk was so slight that, when graphed, it looked like a straight line just below the 20% mark.”

Diversification in Action

More vs. less

To some analysts, having a large number of positions means something else. According to portfolio manager Brett Carson, “overdiversification guarantees average performance, at best. A portfolio comprised of even 100 stocks is spread so thin [that an] investor can only hope to achieve returns [that are] consistent with the broader market. Factor in the costs associated with owning so many securities, and even average performance isn’t possible. This is a core reason why many mutual funds underperform the broader market.” For equities, in particular, “each company is unique, and very few offer a combination of fundamental attributes at a price attractive enough to warrant…investment.”

Proponents of diversification counter by noting recent blow-ups at well-known hedge funds. Many had outsized exposure to formerly high-flying stocks and sectors that suddenly came crashing to earth. The roster includes Bill Ackman’s Pershing Square, which was hard hit by the plunge in Valeant Pharmaceuticals, Edward Lambert’s ESL Investments, which has owned the long-faltering Sears, and Greenlight Capital, which has had significant exposure to energy shares, a sector pummeled by the collapse in oil prices.

To money manager Cullen Roche, the problem with these types of funds and the “gurus” that manage them is that they aren’t well diversified; as a result, they tend to live and die by their concentrated risk. As he sees it, they are specialists in certain areas of the market; but since “nothing works all of the time in finance, they often go through long periods of looking brilliant and long periods of looking silly….Given the dynamism of risk and the cyclical nature of the financial markets,” he says, recent developments demonstrate why “it can be very dangerous to get too hung up on one guru’s ideas.”

Maybe he is right, though at least one guru who is still going strong would probably disagree. In a spring 2015 market commentary cited by ThinkAdvisor, Bill Nygren, legendary portfolio manager of several Oakmark funds, wrote:

“any benefit from reducing risk by adding more stocks to our portfolio is outweighed by the return lost from diluting our best ideas. We are trying to maximize our probability of outperforming by a meaningful amount…. That’s why our funds range from a low of 20 stocks to a high of about 60.” He’s not concerned with diversification because Oakmark investors “are already taking steps to diversify their assets.”

Not a prerequisite to good performance

Some academic research appears to bear such a perspective out. MarketWatch columnist Mark Hulbert recently highlighted a decade-old study that maintained “less-diversified mutual funds had better performance, on average, than the actively managed funds that were most diversified.” While the less-diversified funds “were riskier than the most diversified ones,” he wrote, “even on a risk-adjusted basis, they still came out ahead.”

Another study also concluded that diversification isn’t necessarily a prerequisite to good performance. On the contrary, PLANSPONSOR reported, portfolios can be under-diversified but optimal if they are formed on information advantage. Using data about the holdings of nearly 11,000 institutional investors from 72 countries, the researchers found that institutional portfolios that were more concentrated in a few countries and industries, particularly in foreign markets and industries, performed better than those that were more diversified, suggesting that investors “have some information advantage when forming concentrated portfolios, which results in better portfolio performance and risk-adjusted basis.”

Regardless, how a portfolio is constructed is probably just as important. Whether investors have few or many positions, their capital might not be as cushioned against certain risks as they expect if their portfolios are not exposed to a sufficiently broad range of investments–that is, asset classes other than stocks and bonds. The same may hold true if portfolios include an array of equity, bond or other funds with overlapping holdings, or which are overly concentrated in certain sectors, indexes or regions.

A crowded trade

Even if a portfolio is diversified within and across asset classes, it might not mean what it once did. Investment advisor Dave Merkel maintains that one reason why portfolio diversification is less effective than in the past is because so many investors are emulating one another. While exposure to, say, small cap stocks, REITs, international stocks and emerging markets once served to enhance risk-adjusted returns, much of the benefit dissipated when those subclasses became crowded with institutional money seeking the same. Consequently, the “native return drivers” of many of these assets–most of which are 60-90% correlated, in his view–are what remain.

Another thing to remember, of course, is that diversification is not alchemy. If a portfolio is exposed to securities and asset classes that are intrinsically poor values, the combination is unlikely to produce the kinds of returns that investors are striving for. The primary focus should be on adding assets with superior prospects, not simply for the sake of diversity.

Finally, it is worth keeping in mind that any sort of diversification may not be that helpful with respect to mitigating volatility when upheaval occurs. As an old market saying goes, correlations tend toward one in a crisis, reflecting the fact that investors become almost blindly focused on reducing risk and preserving capital, regardless of security or sector-specific fundamentals. Whether investors have a few positions or many, it is not uncommon to see their portfolio losses multiply when the crowd is rushing for the exits.

The Disconnect Between Theory and Practice

An alternative approach

In the end, the portfolio management approach advocated by Merkel sounds like a decent compromise. In his view, investors should “spread [their] exposures, and do it intelligently, such that the eggs are in baskets [that] are [as] different as they can be, without neglecting the effort to buy attractive assets.” Beyond that, he says, investors should try to hold “dry powder”–in his own case, that means 20-30% of assets–such as cash, which doesn’t earn or lose much, as well as some longer, high quality bonds, such as Treasurys, that “do well when things are bad.”

Michael Panzner

Michael J. Panzner is a 30-year Wall Street veteran and the author of three books, including Financial Armageddon, which predicted the 2008 global financial crisis.