It turns out the optimal equity portfolio contains only 30 to 40 stocks. Beyond that, the advantages of diversification fall dramatically. In a nutshell, if you own 100 stocks or more, you're not much better off than just investing in an index.

Real world results – not merely academic studies – bear this out. As found in a rather straight-forward study using Morningstar data, the average "Optimal Holding" mutual fund (those with fewer than 50 holdings) outperformed the average "Overdiverse Holdings" mutual fund (those with more than 500 holdings) by anywhere from 70 basis points to 1.4% annually over various time periods. These real world results are consistent with the conclusions of the academics studies.

So, is overdiversification a lurking liability within the bowels of the DOL safe harbor provision? Prudence demands 401(k) plan sponsors cannot overlook this Deadly Sin. Too many 401(k) plans offer too many options in the name of "diversification." Unfortunately, too many 401(k) plans offer too little guidance for the employee. The DOL has opined on the dole of diversification in 401(k) plans (see "Why Overdiversification Matters to the ERISA Fiduciary," Fiduciary News, November 1, 2011). But the DOL doesn't define prudence, other than to suggest it is a process rather than a result. 

This leaves 401(k) plan sponsors to wonder what is the best way to discourage overdiversification. Fortunately, a recent article outlined two easy actions plan sponsors can take right now to give 401(k) investors a better chance to achieve their retirement goals – one where they retain control and one which relies on the employee. We'll offer a third choice here, one that benefits from recent research in the field of behavioral economics. We know one of the big problems with 401(k) plans (and the main reason for overdiversification) occurs because plans have far too many choices for the typically employee to properly analyze. While simply reducing the choices to a bare few may satisfy researchers, this may be too draconian a step for the average 401(k) plan sponsor to swallow.

Instead, what if the plan design creates investment options that are divided into different categories of investors? For example, "do-it-yourself" investors might be offered a limited number of index funds; automatic enrollees would be offered a limited number of lifestyle funds; and the "in-between" employee would be offered a limited number of diverse optimal holdings funds, say two equity funds and a stable value fund.

As a practical example, let's say each category would contain three funds, giving the plan a total of nine funds. In the standard model, the employee sees nine funds with no guidance. In this suggested model (which, granted, can't be called "novel" but is in no way "standard"), each employee, once they determine their personal category, would only have to consider three funds. And if the component funds are selected properly (i.e., with the optimal number of holdings) by the plan sponsor, even a naïve diversification strategy (i.e., putting a third of one's assets in each fund), would produce an equity portfolio of only 60 or so stocks, which, while not quite optimal, remains a far cry from the overdiversification found in equity portfolios in excess of 500 stocks.

Obesity is a national health problem, but it's also a problem for many of the nation's 401(k) portfolios.