Mutual funds that pay revenue sharing to recordkeepers of defined contribution (DC) plans are more likely to be added to plan investment menus and are less likely to be deleted from them, according to researchers from Vanderbilt University; the University of Texas at Austin and the National Bureau of Economic Research (NBER); and the Board of Governors of the Federal Reserve System.
In a paper, the researchers note that Investment Company Institute (ICI) data shows that in 2018, mutual funds managed more than 60% of the $5.2 trillion invested in 401(k) plans. “Retirement plans are among the most important distribution channels for mutual funds,” they say.
The researchers used data on mutual fund level revenue sharing, which became available in 2009 when the Department of Labor (DOL) extended reporting requirements about fees. They collected investment menu options for the 1,000 largest 401(k) plans in the U.S. for the 2009 to 2013 sample period from annual plan-level Form 5500 disclosures filed by plan sponsors with the DOL.
In approximately 54% of the plans, recordkeepers received a rebate in the form a revenue-sharing arrangement from at least one fund on the menu. Recordkeepers also receive revenue sharing when included as third-party investment managers in plans administered by other recordkeepers. On average, about 55% of the third-party (unaffiliated) funds in revenue-sharing plans offer revenue-sharing rebates.
The analysis found that funds that revenue share are economically and statistically significantly less likely to be deleted from revenue-sharing plans. For example, the average deletion rate is about 20% for revenue-sharing plans and 28% for non-revenue-sharing plans. The research also showed that funds that pay a higher rebate are significantly less likely to be deleted.
In addition, the researchers found that funds that tend to revenue share on other menus and tend to pay higher rebates are significantly more likely to be added to revenue-sharing plans. For example, funds with an above-median propensity to revenue share have an average addition rate of 0.16%, whereas funds with a below-median propensity have an average addition rate of 0.1%. “Overall, our results suggest that revenue sharing affects the investment choices offered to plan participants,” the researchers wrote in the paper.
The researchers next investigated whether revenue sharing also affects plan costs. The main hypothesis is that, since indirect compensation is less transparent, intermediaries might be able to overcharge some of their customers. They note that a study on 401(k) plans by the Government Accountability Office (GAO) suggests that due to sponsors’ and participants’ lack of understanding of indirect fees, recordkeepers might not reduce direct fees sufficiently. They hypothesize that if direct and indirect payments do not offset each other, recordkeepers may collect more revenue in the presence of indirect compensation and participants may pay higher fees in these plans. Additionally, if recordkeepers are better off when they receive compensation indirectly, they may influence 401(k) sponsors to include and subsequently keep funds on the investment menu that pay a higher rebate, even when these funds are dominated by peer options.
The researchers first found that revenue-sharing plans and non-revenue-sharing plans are similar with respect to most observable plan and participant characteristics. They calculated total fees paid by participants as the sum of direct plan fees (i.e., administrative fees) and the value-weighted average expense ratios of the investment options in the plan. The results indicate that rebates translate into higher expense ratios in the retirement setting, while direct fees are not significantly different across revenue-sharing and non-sharing plans. Consequently, participants face higher all-in fees in revenue-sharing plans.
The researchers then looked at whether higher fees are offset by higher returns. They found that is not the case, and, instead, the future performance of revenue-sharing funds is weaker than that of non-sharing funds. “The bulk of the under-performance is driven by higher fees, though revenue-sharing funds display lower performance even after accounting for fees,” they note.
One final analysis of the paper examined whether the market power of the recordkeeper affects the nature of the compensation scheme. The researchers say their results have suggested that recordkeepers receive higher fees when their compensation includes indirect payments, implying that they may prefer these arrangements. Since recordkeepers also act as gatekeepers in the retirement channel, third-party funds may agree to revenue share to gain access to retirement assets.
They hypothesize that third-party funds are more likely to revenue share and pay a higher rebate when the plan has a more powerful recordkeeper. The analysis shows that the influence of the recordkeeper on the management company—which the researchers measured by calculating the percent of the total 401(k) assets of the management company included in plans administered by the recordkeeper—is significantly positively related to the size of the revenue share. Using network measures of centrality to capture the market power of the recordkeeper, they find similar results: Recordkeepers that are more central in the network of families that service 401(k) plans receive higher rebates.
Additionally, the analysis shows that third-party funds with investment mandates that correspond to those also offered in the recordkeeper’s own proprietary lineup are more likely to pay a rebate. And, the researchers found, reciprocal relationships that arise when two management companies work together in two or more plans in which they take turns being the recordkeeper reduce the size of revenue sharing.
Revenue sharing has been the subject of many Employee Retirement Income Security Act (ERISA) lawsuits against DC plan sponsors. But research shows the inclusion of funds that pay revenue sharing on DC plan investment menus is on the decline. There has also been a push to levelize fees among DC plan participants.
The researchers say the threat of lawsuits and the desire to make fees more fair for participants were considered in their analysis. A copy of the research paper may be downloaded or purchased here.
« PBGC Announces Settlement to Benefit Multiemployer Pension Plan