April 2014 Part 2
By Philip J. Koehler, Esq.

Watching CSPAN’s coverage of General Motors’ new CEO, Mary Barra last week, a Yogi Berra sense of ‘déjà vu’ hung over the proceedings. Ms. Barra (a curious irony, but, apparently, no relation) was testifying before the Senate Commerce Committee and the House Energy and Commerce Committee about GM’s recall of 2.6 million vehicles (mostly Chevy Cobalts and Saturn Ions manufactured before 2008) with a defective ignition switch. Prior testimony had established the fact that back in 2001 GM had discovered that the switch, which adds fifty-seven cents to the cost of the car, had a flaw that caused it to move into the “off” or “auxiliary” position while driving. This would disable power steering and braking and prevent airbags from deploying in the event of a collision. GM admits that the defect has been linked to 13 deaths.

Haven’t we seen this movie before? Fifty years ago Ralph Nader incurred the wrath of the auto industry by writing “Unsafe at Any Speed: The Designed-In Dangers of the American Automobile.” Nader’s book accused car manufacturers of slow walking the introduction of safety features, like seat belts, and an institutionalized resistance to make their cars safer regardless of cost. In retrospect, his observations, which helped to foster similar Congressional investigations at that time, seem almost uncontroversial.
The Committee members are in a nonpartisan uproar, especially since Ms. Barra, a thirty-year GM employee, expressed the view that the New (post-bankruptcy) GM is not the same company that existed before the court-approved reorganization agreement went into effect in 2009 (the “Old GM”) and, therefore, it’s immune from civil liability for defects that were in production on or before then. Congress is incredulous that neither the Old/New GM, nor the National Highway Traffic Safety Administration, did not act more swiftly to fix this defect and potentially, at least, prevent some of these deaths.

Under general product liability law an auto manufacturer is strictly liable for the damages to customers or bystanders that are caused by a defect in its product, regardless of whether or not the defect was caused by the manufacturer’s negligence. Many members of Congress have warned Ms. Barra that, while the protections of bankruptcy law may shield the New GM in this case, they believe criminal statutes apply to a corporation that waited (i) seven years to replace the fifty-seven cent defective part in new cars coming off its assembly line, and (ii) 13 years to recall affected vehicles on the road. The sense that “something must be done” is palpable.

How quickly Congress’s moral outrage wanes when the focus shifts from protecting the general public from defective cars to protecting their retirement security from defectively constructed 401(k) plan fund line ups. In both situations we have large populations of consumers who are at risk of suffering significant damage. While the defect that is the subject of the GM recall is associated with at least 13 fatalities, the systemic diminution of the retirement security of hundreds of thousands of American workers is no small matter either. Consider a generally well diversified 401(k) plan fund line up that includes mostly cost-efficient funds, including an S&P 500 index fund with fees of 25 bps or less. Imagine that the plan’s fiduciaries decide to add an actively managed, large cap fund with an investment style and objective that closely tracks that index, but charges fees of 150 bps. The probability that the large cap fund’s performance will on average exceed the performance of the index fund by more than the 125 bps spread in fees is virtually nil, making the large cap fund ex ante an objectively poor investment choice given that line up.

The sophisticated participants in such a plan who seek a fund with that style would presumably recognize that the large cap fund has negative relative upside potential and they are better off investing in the index fund. But we know that the vast majority of plan participants are not sophisticated and that the mere inclusion of the fund in the line up will be perceived by some as an endorsement that standing alone it is a potentially sound investment with at least some upside potential. Modern behavioral analysis shows that invariably some participants will switch from the low cost index fund to the large cap fund or spread their investment between both funds in the erroneous belief that they are achieving the benefits of greater diversification. Thus, the mere inclusion of the large cap fund ensures that some participants will unnecessarily suffer the erosive effects of the high fee large cap fund. The obvious question, of course, is: who is responsible for these losses? Is it the participant who made the decision to invest in the large cap fund, or should the fiduciaries who included that fund in the plan’s line up bear responsibility?

In his recent study, “Beyond Diversification: The Pervasive Problem of Excessive Fees and “Dominated Funds,” Professor Ayers of Yale University describes a “Dominated Fund” as a fund that is so clearly inferior to other funds or groups of funds offered in the same line up that participants are categorically better off avoiding it. Such funds are, in effect, “dominated” by the risk-adjusted, net-after-fees performance of all other funds in the line up.

To measure the pervasiveness of Dominated Funds, Ayres’s study looks at two categories of funds. The first category includes any fund where:

  1. There is another fund in the line up with the same style but with fees at least 50 bps lower;
  2. The fund has fees 25 bps higher than the mean fees of all funds with the same style in the study’s 3500+ plan sample of 401(k) plans; and
  3. The fund has a relatively insignificant weight (less than 1%) in the optimal portfolio that can be achieved using that plan’s line up.

The second category includes any fund where:

  1. There is no other fund in plan menu with the same style,
  2. The fund has fees 50 bps higher than the mean fees of funds with the same investing style in the study’s 3500+ sample of 401(k) plans, and
  3. The fund has a relatively insignificant weight (less than 1%) in the optimal portfolio that can be achieved using that line up.

With these two categories in hand (the first category identifying a specific alternative fund in the same style that is a significantly lower cost alternative, and the second category considering the line up as a whole and ensuring that an investor would be better off allocating among other funds in the menu to achieve the same style), Ayres surveyed the 3,573 plans in his study’s sample. He found that 1,842 plans (52%), holding 22.6% of total plan assets, included at least one Dominated Fund. Of those, 15% of the funds in the line up were dominated, holding 11.5% of plan assets. Looking across all plans, 8.9% of the funds are dominated, holding 3.4% of plan assets. Since, by construction, an optimal portfolio weight for Dominated Funds was less than 1%, Dominated Funds were significantly overweighted in the sample.

The study computes the one-year returns for all funds in the sample beginning each month for the period January 2010 through January 2013. The average returns on a Dominated Fund during that period were more than 60 basis points lower than the other funds in the line up. Dominated Funds that fall into the first category are outperformed by their lower-cost same-style alternative by more 1.5% on average. Looking at the second category, Ayres found that if Dominated Funds were mapped over pro rata to the other funds in a plan’s line up, participants would save more than half of the investment management fees currently charged by the Dominated Funds, or 67 bps, the total plan cost of plans with Dominated Funds would fall by about 7 basis points, or 11%, and the savings to the participants holding the Dominated Funds would be even more substantial.

It can always be argued that the losses from Dominated Funds are the result of participant investment decisions. It’s an obvious truism that but for the participant’s decision, there would be no losses. But, does that seem to be a reasonable application of ERISA’s fiduciary standard of prudence in choosing to include/retain a fund in the line up that is ex ante a poor investment choice? Ayres uses the example of a car manufacturer who builds a car with a special button on the dash. If the driver pushes that button, the car will malfunction. So, even if it’s foreseeable that some percentage of customers will push the button anyway, is the car still nondefective, because, absent pushing the button, nothing bad would have happened? If Ms. Barra’s litany of mea culpas on Capital Hill was any indication, even GM would shrink from such a suggestion and Congress rightfully would never stand for it.

The Department of Labor has consistently argued that menu construction is prior to investor choice, and therefore the ERISA Section 404(c) safe harbor is unavailable as a defense for fiduciary breaches in menu design. Yet, there is a line of cases out of the Fifth, Seventh and Eighth Circuits that reject this position. Langbecker v. Electronic Data Systems Corp., Hecker v. Deere & Co. and Braden v. Walmart. These cases support the notion that as long as a plan’s line up includes some good options, the fiduciaries will not be liable merely because they also included some bad options. The Ninth Circuit recently declined to follow these cases and the Supreme Court is considering whether to grant cert. in Tibble v. Edison International.

Every year since 2001 there have been over 200 different models of new cars available in the U.S. marketplace. No reasonable person would argue that, as long as some of them were nondefective, a manufacturer of a defective car should be absolved of responsibility, yet this is essentially the level of protection afforded 401(k) plan participants under applicable case law. In fact, the situation is much worse if 401(k) plan participation is treated as a consumer product. The typical 401(k) plan line up includes only 10-15 funds.  With 52% of plans that limit fund options to mutual funds including at least one Dominated Fund, it is far more likely that a participant in such a plan would invest in a Dominated Fund than he or she would buy a defective automobile.  Since the cost of scrubbing Dominated Funds from plan line ups is relatively de minimis, Congress should spare some of the moral outrage it leveled at GM last week and save it for the 401(k) plan industry.