Target-date funds have piqued the interest of many by satisfying qualified default investment alternative regulations implementing the Pension Protection Act of 2006. They have also provided two key benefits to participants in defined contribution retirement plans such as 401(k)s. First, TDFs offer a turnkey solution to the challenge of participant inertia. By reducing risk as retirement approaches, the need for participants to rebalance their accounts is diminished. Second, they provide diversification and mitigate the risk of oversized allocations to company stock or cash.
While target-date funds have the potential to improve outcomes relative to concentrated portfolios with too little or too much market risk, several challenges remain for plan executives who desire an understanding of them beyond the obvious. It has been widely observed that TDFs sharing the same target year differ vastly with respect to their asset class exposure. Questions abound as to which glidepath is most appropriate for a given plan and how to measure the results of particular TDFs.
Analysts have been using performance attribution to understand equity portfolio returns for a long time and, with a few changes to the traditional technique, this type of analysis may be adapted to multiasset-class portfolios such as TDFs.
In equity portfolio attribution, the idea is to break out investment returns into a series of components ascribed to various investment decisions such as security selection and industry exposure. The technique shows which decisions added or detracted value relative to a market index for a given period. The same thing can be done for target-date funds with an adjustment to the scope of the analysis which aligns it with the scope of investment decisions falling into the purview of TDF managers, namely asset allocation and fund selection. Numerous investment theorists, from Benjamin Graham to Brinson, Hood, and Beebower, have written about the primacy of asset allocation in the investment decision-making process so it is fortunate that we can observe its contribution to TDF results.
To apply this technique one should use a non-prescriptive glidepath that is derived from market observations as the proxy for asset allocation policy. This framework facilitates a comparison of actual asset allocation of a given target-date fund manager with a consensus asset allocation derived from market observations. The term “consensus” applies because the proxy asset allocation policy is obtained from observations of the holdings of a group of TDF managers (for a given target year) and therefore represents an overall market view of appropriate investment policy. A consensus glidepath comprises a set of consensus target year asset allocation policies.
Why is using a consensus glidepath important? Many TDF managers would protest the idea of being compared to a consensus. They may feel their asset allocation policy is uniquely suited to a set of particular investment goals and therefore inappropriate to compare it to a market proxy. While it is best practice for fiduciaries to take into account workforce characteristics like average working age, compensation and other traits deemed necessary to making a prudent TDF selection, it should be remembered that no single fund can completely address the asset allocation requirements of all individuals within a plan. Selecting a target-date fund is more akin to estimating the best fit for a plan, given its workforce characteristics, than it is to optimizing investment exposure. Deviations from the market consensus glidepath ought to be considered in light of balancing a desire to find the best fit with the risk of overengineering toward the expected outcome. In light of this fact, plans should seek common sense solutions that can help them to understand TDF performance.
To the straightforward question, “Is a TDF manager more or less aggressive than their competition?” one can compare the manager’s glidepath to a consensus glidepath. For more complicated questions such as “Is the TDF manager our plan hired doing what we hired him to do?” one can employ target-date fund performance attribution using a consensus glidepath. Consider a plan that has come to the conclusion that its workforce would be best served, on average, by a conservative glidepath which manages its asset class exposure up to the target year with the expectation that funds will largely be withdrawn and annuitized. Manager X responds to the plan’s RFP with a pitch highlighting its “To” investment philosophy and its record of success in helping other plans with similar needs.
After the decision to hire manager X has been made and implemented, how can the sponsor measure X’s success? First, in target year portfolios approaching their target year, manager X should have more of its portfolio allocated to fixed income than the consensus glidepath. This is a check on whether the manager is holding true to its mandate. But we can go much further in our assessment. In any given performance period, manager X should underperform the consensus glidepath if more-risky assets have outperformed less-risky assets. It is clear then, that underperformance, in and of itself, is not a reason to turn negative on manager X. In fact, it may be demonstrating that he is doing exactly what he was hired to do.
But how would the plan know whether manager X underperformed because of proper asset allocation policy or from poor fund selection? TDF performance attribution using a consensus glidepath has the answer. If manager X is doing a good job, his allocation effect will be negative, following from the fact that he was overweight less-risky assets, which underperformed in the period. But if the analysis shows, for example, that manager X was overweight more-risky assets and therefore underperformed because of poor fund selection, then he might be a candidate for replacement.
In the world of single-asset-class performance attribution any underperformance is negative, but this is not so in the multiasset-class world. Perhaps this is why many oppose target-date fund performance attribution relative to a market based proxy. But we shouldn’t throw the baby out with the bathwater. If we adjust our analytical techniques to suit a multiasset-class environment, TDF performance attribution using a consensus glidepath can be an invaluable tool.
Philip Murphy is a vice president at S&P Indices in New York.