Rethinking Active Management
Active equity management has been much maligned over the past few years as index funds continue to lower their fees and the track record of active managers has been less than impressive. There are many factors for the underperformance of active managers, fees, high equity correlations, markets not driven by fundamentals etc. But let’s make it simpler, beating your benchmark on a consistent basis is hard, really hard. That’s just the nature of investing. With fees on index funds approaching zero, perhaps it is time to rethink what compensation for active management should look like.
Starting with Morningstar’s Large Cap Value mutual fund universe of 1390 funds, I filtered for those funds with 10 years of returns to analyze and I cut out all but the lowest fee share class, leaving me with 163 Large Cap Value mutual funds. Now these are mainly institutional shares that most regular investors are not able to access. I compared this set of managers to the Russell 1000 Value from 2005 to 2014. Overall, only 43.5% of managers were able to outperform the Russell 1000 Value during that 10 year period. What’s more Dimensional Fund Advisors the broad universe of equity mutual funds and found that outperformance over long time periods was even lower than 43.5%. Their data ends at the end of 2014 and shows over a 15 year time span, if take survivorship into account, 19% of mutual funds outperform their benchmark.
*Source: Dimensional Fund Advisors
These findings are coupled with a paper out of Oxford[1] examining the performance of investment consultants. These consultants help the largest pensions in the United States choose which managers to hire and fire. The authors found that, when selecting US equity managers, the consultants not only do not add any value, they actually detracted value with their selections. To put it briefly, the highest paid consultants in the US cannot add value by picking managers, why should we expect different from advisors.
With results like these, it is no wonder investors have been jettisoning active managers in favor of cheaper and often more successful index funds. As you can see below the trend towards passive management has been growing for over a decade.
*Source Morningstar
The question to consider is whether we should abandon the idea of active management or if there is a way forward where active management works for investors?
I would argue the latter. I believe reframing the way we think about active management fees can make active funds more attractive to clients and improve performance. With index fund fees reaching for zero, I believe active managers should only be compensated for performance in excess of the benchmark. In the days of yore managers could justify charging an annual fee for active management because you could not go out and buy the index ETF. I propose a fee structure where the manager is paid a percentage of their annual alpha (outperformance compared to benchmark), and that fees are paid over three years to smooth out manager income.
Using the group of 163 Large Cap Value managers I mentioned above, I compared the performance and fees collected from 2005 to 2014 using the standard fees of each fund and using a 0% base fee and 25% performance fee paid over three years. Overall, if the 163 funds used the proposed fee structure, 95% of funds would charge less, and 95% of funds would have earned their investors more over the 10 year period. The chart below shows how asset growth and fees using the performance fee only methodology compare to the current model.
This chart breaks out the 163 funds into four groups, with the first quarter having the best performance during the 10 year period. Fee savings are shown on the left axis and the difference in asset growth is shown on the right axis. For the top quarter of managers, using a performance fee only fee structure would have, on average, lowered investor fees by 38% and increased total asset growth by 3.5%. As you can see the results become magnified as we move down to managers with lessor performance. Switching to a performance fee only model would increase the percentage of managers who would outperform their benchmark from 43% to 53%. This is a significant increase, but still leaves something to be desired.
As you can see below, across each performance quartile transitioning to a performance only model would have increased 10 year returns by a meaningful margin. The lines labeled T represent the average performance of the quartile using the current fund fees. The lines labeled GF represent the average performance of each quartile using the performance fee only formula proposed above.
There are caveats for manager and client. For the client there are financial constraints as it would currently only be open to qualified investors per SEC rules. For the manager; they would have to assess the reward of providing a product more in line with investor goals and more honestly priced, with the risk of lower revenue and potentially less consistent revenue.
I for one believe in a performance only model as it better aligns the interest of the investor with the interest of the manager and makes compensation commensurate with actually adding value for the client.
Nathan Wallace
[1] http://www.umass.edu/preferen/You%20Must%20Read%20This/PickingWinners.pdf
Great article. Would this model mean that mediocre managers would be paid mediocre sums and rock-star managers would be paid like rock-stars? And in either case the client would make out better than with traditional fee plans?
As it says in the article, in 95% of cases clients would see more asset growth vs the traditional model. Out of the 163 managers there was one case where the client would pay more for lower performance, AMGIX who is certainly not a rock-star manager. The reason they have higher fees and lower performance is they have violently erratic performance. On two occasions from 2005-2014 they under-performed their benchmark by more than 15%. Conversely, they also beat the benchmark by more than 10% three times.