A 40-Stock Portfolio: Not Just for the Bold Types

What is the first word that comes to mind when you see a portfolio with fewer than 40 stocks? Risky? Intuitively, it makes sense to think that a portfolio of 40 stocks is inherently riskier than one that owns 400. Managers who choose to hold more stocks usually point to "risk control" as a key reason for spreading their bets across a broad swath of stocks. At the same time, however, managers of concentrated funds argue that they minimize risk in other ways such as demanding a low price, digging deeper on each company, or combining stocks that have different fundamental drivers.

Concentrated Funds Are Not Necessarily Riskier
Our recent research on this topic suggests that risk-averse investors needn't reflexively shy away from concentrated funds. We took a look at the broad universe of funds over the past 15 years and found that the most concentrated 25% of funds in each category wasn't any more or less volatile--as measured by standard deviation--than the peer-group norm.

That said, we did find that those concentrated funds that trade infrequently (and thus exhibit a long investing horizon) have been less volatile, on average, than their category norm. We looked at funds that landed in both the lowest quartiles for concentration and turnover. That subgroup exhibited a lower standard deviation of returns than the category average in 76% of rolling one-, three-, five-, and 10-year periods combined from 1992 through the end of 2006.

For example,  Ariel  (ARGFX),  Jensen  (JENSX) and  Sequoia  (SEQUX) are highly concentrated and have exhibited low volatility over the years. Each has its own method of controlling risk that doesn't involve owning a large number of stocks. John Rogers at Ariel only buys stocks trading at a discount of 40% or more to his estimate of their intrinsic values. He also leans toward companies in stable industries with healthy financials. That posture has smoothed the bumps that may come from the occasional stock-specific miss.

While volatility is one measure of risk, a fund's performance in a down market is arguably a more important consideration for most investors. The early-2000s' bear market was thus a good test. The most concentrated quartile of funds held up better than their nonconcentrated peers in 2000, 2001, and 2002. In 2001, for example, the funds in the most concentrated quartile of funds landed in the 39th percentile of their respective categories on average, while the least concentrated funds averaged a 54% ranking that year. That implies that concentrated funds have more tempered downsides, on average. There are other explanations, however. Several value-oriented skippers couldn't find stocks that fit their criteria in 1999 and early 2000 and were thus sitting on some cash when the bear market hit.

To be sure, not all focused funds are going to be less bumpy. Take  Janus Twenty  and  Baron Partners (BPTRX), for example. Both are highly concentrated and have performed well over long stretches, but they have exhibited very high levels of volatility over short and longer time periods.

Unless You Expect Them to Act Like a Benchmark
When thinking about relative risk, however, the tables were turned. We looked at funds' R-squared data, which measures the percentage of a fund's returns that can be explained by movements of a particular index. For example, an R-squared of 100 implies that 100% of a fund's returns could be explained by the movements in a benchmark. So, the closer R-squared is to 100, the more closely its returns have resembled an index.

We first calculated R-squared based on relevant market benchmarks for each capitalization range. For example, we used the S&P 500 Index for large-cap categories and the Russell 2000 for small caps and MSCI EAFE for international funds. We found that the average R-squared was meaningfully lower for the most-concentrated quartile of funds than it was for the least-concentrated quartile.

This suggests that concentrated funds are more likely to dance to the beat of their own drummer, rather than mimic an index. Meanwhile, funds that hold a large number of stocks have tended to move more in sync with a relevant market index. Take  Vanguard Explorer (VEXPX). Its multiple-subadvisor approach leads it to hold more than 1,000 stocks, and it sports an R-squared of 97 with the Russell 2000 Growth. On the flip side,  Oakmark Select (OAKLX) has a very concentrated portfolio. And while it hasn't been volatile in terms of standard deviation, it has a low R-squared of 66 with the S&P 500. Oakmark Select has lagged the S&P 500 by a substantial margin in recent years, exhibiting the heightened relative risk that concentrated strategies face.

Relative rankings tend to bounce around for concentrated funds. For example,  Fairholme Fund  (FAIRX) was behind 73% of its mid-blend rivals for the year to date through the end of June 2007. Just one tumultuous month later, Fairholme leapfrogged many of its peers to the top third of its category for the year to date through the end of July.

Perhaps the best reason to own a concentrated fund is its returns' low correlation to other core funds in a portfolio. While concentrated funds' returns may fluctuate relative to their peers, such funds can help mitigate overall portfolio volatility because funds with low correlation don't move in lockstep with other holdings. But for that to work, investors have to be willing to hold on through sometimes erratic relative results.

What Concentration Won't Tell You
Although we found that focused funds aren't necessarily riskier on average, we did not find a connection between the level of concentration and long-term performance. That's not a huge surprise because there's a lot more to the equation than how many stocks a fund owns. A manager's skill, execution, and a fund's fees remain critical factors. A poorly executed strategy and hefty expenses can and will weigh on any fund, focused or not.