Low-cost index funds will beat the average actively managed fund after expenses. But can advisors identify superior active funds to overcome this disadvantage?
Advisors who believe they can choose those funds will be challenged by the results of two studies from the defined-contribution industry.
Comparing advisors to plan sponsors
Unfortunately, there are no comprehensive performance studies of funds chosen by advisors. But there are studies that examine choices for 401(k) plans. With defined-contribution plans, like 401(k)s, the plan sponsor decides which funds to offer to participants. Such funds typically span a range of fixed-income and equity offerings and may also include alternative asset classes.
Plan sponsors may assign this selection responsibility to a single individual, use an internal committee or rely on outside consultants. Many advisors face similar decisions in choosing, and periodically updating, the stable of funds to recommend to clients. Similar to plan sponsors, advisory firms may rely on individuals, committees or outside consultants.
Although there are certainly differences between plan sponsors and advisors, there are enough similarities in the processes they use and the skills they possess that plan sponsor results give an indication of how advisors fare with fund selection.
The Elton, Gruber and Blake study
I'll discuss two studies carried out in different timeframes. The first is the Elton, Gruber and Blake (EGB) study based on 1994-1999 data. Their study, which was published in 2007, received recent recognition in a January 2013 briefpublished by the Center for Retirement Research at Boston College.
These researchers examined the month-by-month performance of 43 separate 401(k) plans and compared risk-adjusted returns for each plan against two separate measures: a passive portfolio of indexes and a random selection of similar funds. To compare, the researchers calculated the contribution to overall return from systematic market factors (such as large-capitalization or small-capitalization index performance), which they called beta. Alpha represented the positive or negative value added by investment management skills like security selection and market timing. In the second comparison to similar funds, the researchers calculated a differential alpha equal to the alpha for the selected funds minus the alpha for the similar non-selected funds.
This study showed that the average differential alpha for the 43 plans was 0.52% per year — plan sponsors selecting funds for the 401(k) plans were able to add value, on average, compared to random fund selection. Additional analysis showed that average fees in the funds selected by plan sponsors were 0.23% per year lower than in the random set of funds. Plan sponsors offered roughly equal value in their ability to choose funds with lower expenses and choose managers who performed better.
But despite the positive differential alphas, the selection process did not add enough risk-adjusted return to beat low-expense, passive investing. The average alpha for the 43 plans was -0.31%, which was "larger than normal expenses for low-cost index funds,” according to the report. The report suggested “performance would be improved if passive funds had been substituted for the active funds that were selected."
The EGB study provides a general indication that those selecting funds add value over random selection but not over low-cost index funds. But, the sample size was small for this study and the data quite old, so one has to be cautious about reading too much into the results.
The Brown and Harlow study
The Brown and Harlow (BH) study used more recent data and a much larger sample size. Their research covered the period from January 2000 to June 2007 and utilized data from more than 27,000 plan sponsors. They examined fund performance rather than overall plan performance and focused their investigation on domestic equity funds. The performance measures they used were similar to those used by EGB – alphaanddifferential alpha – but they did much more testing for the robustness of findings.
Brown and Harlow tested regression formulas with different numbers of beta variables. To calculate the differential alpha estimates, they used different methods for matching funds chosen by plan sponsors against randomly selected funds. They also split the sample period into three sub-periods. In general, they found that these various approaches yielded similar results, giving additional support to their conclusions.
This chart summarizes data from the BH study and includes a column of comparison data from EGB. The EGB analysis was based on a sample of 45 plans, so the BH numbers carry much more statistical significance. I've annualized the data that was presented in the studies.
|
Plan sponsor performance results from two studies
|
|
EGB Mean
|
BH Mean
|
BH Median
|
BH Percent Positive
|
BH Sample Size
|
|
Alpha: all options
|
N/A
|
-1.32%
|
-1.11%
|
36%
|
10,368
|
|
Alpha: plan options
|
-0.31%
|
-0.30%
|
-0.18%
|
46%
|
1,350
|
|
Alpha: non-plan options
|
-0.83%
|
-1.47%
|
-1.26%
|
35%
|
9,018
|
|
Differential Alpha
|
0.52%
|
1.17%
|
1.08%
|
The BH differential alphas of more than 1% per year support the authors' conclusion that "plan sponsors do appear to possess superior skills when designing the set of investment options offered to plan participants." Brown and Harlow did not provide information on the difference in expense charges for the selected and non-selected funds in their study. But if we assume a number roughly equal to the 0.23% reported in the EGB study, that would leave a substantial annual benefit of more than 0.75% from selecting more skilled managers.
In general, the BH study confirms the conclusions of the EGB study: plan sponsors appear to add value in selecting funds from the actively managed universe, but they do not possess enough skill to beat the performance of low-cost index funds. From the BH study, the negative alphas are in the 0.18% to 0.30% range, higher than the typical level of charges for low-cost index funds or exchange-traded funds.
Implications for planners and advisors
A familiar argument against active management is that the universe of professional investing is a zero-sum game before expenses and a negative-sum game after expenses are taken into account. These two studies put a dent in that argument by showing that skill in fund selection can make a difference. And since plan sponsors on average demonstrate skill, there's a good chance that those possessing above-average skills will product even better results.
This is good news for advisors who focus on active management and for those who follow a "core and satellite" approach with the bulk of client investments in low-cost index funds, supplemented by a few active-management choices. Advisors can feel some assurance that attention to fund selection can improve their clients’ investment performance prospects.
Keep in mind, however, some significant cautions. Recognize that, despite the skills they bring to fund selection, advisors’ success (or lack thereof) will be subject to considerable statistical variability. Luck and skill will each play a part, and success is not guaranteed.
Although these studies demonstrate that active fund selection can add value, they do not show that an active approach can beat a passive one. Average alphas for the plan sponsor funds in the studies are still significantly negative. Doing better than indexing on a consistent basis requires above-average fund selection skills. Also, to add value over indexing, active strategies would not only have to offset fund expenses but would also need to cover advisor charges for the active selection process. For assets in non-qualified accounts, performance would also need to cover the additional taxes generated.
Beating indexing is still an obstacle to many.
More research needed
These two studies explore an important area for further research on investment management and how advisors can best add value for clients. Some areas worth further study include:
-
Fixed income funds : Research to date has mostly focused on equity funds, and it would be useful to have similar studies on fixed-income experience. Advisors are feeling particular pressure these days to find more attractive fixed-income yields for clients without taking undue risk. A study similar to the BH research that focused on fixed income would provide valuable information on how much selection skill can improve risk-adjusted bond returns.
-
Advisor skills : It would be useful to have an extensive study like the BH study based on advisor fund selections rather than plan sponsor selections. Many advisory firms provide reporting on their own performance, but it would be useful to have industry research, perhaps with breakdowns by type of firm. Such studies would need to use methodology similar to the EGB and BH studies to properly measure risk-adjusted performance.
-
How to improve the fund selection process : The indications that fund selection skills can add value make it likely that even higher risk-adjusted returns can be achieved by improving the process of fund selection. There is existing research that may offer useful insights about what works best in fund selection. I plan to review some of this work for a future Advisor Perspectives article.
-
Other ways advisors add value : We're starting to see studies attempting to quantify the positive impacts that clients can derive from financial advice (discussed in a Jan. 29, 2013 Advisor Perspectivesarticle by Wade Pfau). Advisors can help clients with investment decisions and with a variety of non-investment matters. An estimate of how much value can be added by each type of advice could help advisors and financial planners prioritize their work and determine how much time and effort to devote to investment management.
A final word
Advisors with the right skills — and the ability to organize those skills into a top-quality selection process — stand a good chance of providing superior returns for their clients. But those with more average capabilities would likely serve their clients better by recommending index funds.
This state of affairs creates a dilemma for clients in terms of how much to rely on and pay for advisors that claim superior fund-selection capabilities. My personal view is to offer "core only" or "core and satellite" approaches and not to bet entire portfolios on active management. I realize, however, that others may feel justified taking a more aggressive stance. More research will help clarify the picture.
Joe Tomlinson, an actuary and financial planner, is managing director of Tomlinson Financial Planning, LLC in Greenville, Maine. His practice focuses on retirement planning. He also does research and writing on financial planning and investment topics.
Read more articles by Joe Tomlinson
