Can Hero Funds save portfolios? S&P DJI’s Joe Nelesen takes a closer look at our special SPIVA report examining the performance of multi-asset portfolios of funds versus weighted blends of indices.
[TRANSCRIPT]
Paul Murdock:
Can Hero Funds save portfolios? For over 20 years, S&P Dow Jones Indices' SPIVA Scorecards have reported rates of success and failure among active managers seeking to beat their category benchmarks, but there’s been less focus on the resulting challenges for managers or advisors who select and allocate across multiple funds to build portfolios.
Hello, I’m Paul Murdock, and today, I’m joined by Joe Nelesen from S&P DJI’s Index Investment Strategy team for a closer look at our special SPIVA report examining the performance of multi-asset portfolios of funds versus weighted blends of indices.
Joe, thanks for being here today.
Joe Nelesen:
Pleasure to be here, Paul.
Paul Murdock:
So, Joe, to start, can you walk us through how you went about measuring active portfolio performance versus index blends in your research, Heroes in Haystacks?
Joe Nelesen:
Absolutely. And, as you mentioned in the introduction, we’ve been doing SPIVA for more than 20 years now. And, those data are organized category by category. So, looking at managers relative to a benchmark. What we hadn’t done up to this point was look at a portfolio, which is the real world where investors live. What we did to accomplish that was to take nine of our SPIVA categories, which are most commonly represented in a lot of individual portfolios, allocate those in a 60/40 blend, very non-controversially in proportions that match global issuance of those categories. We’ve got fixed income. We’ve got U.S. equity, international, emerging, etc. And, then, take that 60/40 and scale it up anywhere from a 10% equity and 90% fixed income, all the way to a 90/10 on the other end.
So, we have nine portfolios of active where we would then randomly pick active managers from each of those fund universes over the last 10 years, those that were eligible, and compare each of those blends to the exact same allocation of indices in terms of total performance and risk. We repeated that simulation over a million times. So, we have a very robust sample of simulated portfolios of real live active manager performance as well as the associated index blends.
Paul Murdock:
And, what were your key findings?
Joe Nelesen:
The key findings is building a portfolio of categories that each individually have difficult rates of underperformance doesn’t necessarily lead to a better outcome. In fact, we found with the 60/40 example a 97% underperformance rate. Most of those blends of actual active funds underperformed the equivalent blend of indices over 10 years.
Paul Murdock:
And, what appears to be driving this underperformance at the portfolio level?
Joe Nelesen:
I think there are two things happening. One is just the probability when you have nine categories where, most of the time, there’s underperformance. That really starts to compound. Even if it were 50% in each of these categories. Think about it if you were to flip a coin nine different times, or nine different coins, more likely. What’s the chance of getting all of them heads or all of them tails? It’s 1 in 512. And, so, when those probabilities are higher than 50% of underperformance, we’re seeing a very low rate of success.
The second part that is really important to mention is the difference in outcomes. Outperformers and underperformers aren’t necessarily fitting a normal distribution. In other words, the outperformance alpha among funds that beat their benchmarks tends to be a little bit lower than the underperformance of the other end of the tail. So, when you have funds that are skewed that way, the chances are less and less of picking a fund that outperforms in the upside that would outweigh the underperformers on the other end.
Paul Murdock:
How about when you drill down to top-quartile funds or looked at the performance of specific market cap or fixed income sleeves versus their respective index combinations?
Joe Nelesen:
This is essential. And this is, I think, one of the points of this paper that starts to get interesting for me and others is thinking about the real world. Fund selectors don’t typically pick randomly from the universe of active managers. They have some sort of criteria. We thought a very fair one is top quartile. So, if we limited our selection universe to just the top-quartile managers from each of those nine categories in the five years prior to our sample period, choose only from those funds and build portfolios, we found really a similar outcome. Underperformance rates at around 90% or higher for all of the allocations.
But, what also stood out was the fact that in the fixed-income-heavy allocations, more around the 10/90 and the 20/80, there was a remarkable increase in risk, and we can see that, in some of the charts in the paper, without a subsequent increase in performance. And, so, that tells you that a lot of the managers, particularly in fixed income, are taking on equity-like risks without the equivalent rewards of doing so.
Paul Murdock:
Interesting. So, were there any specific blends where the results differed?
Joe Nelesen:
I think we could look at blends and, then, back from those blends, into which categories made the biggest difference. And, this is where you get into that final part about the Heroes, which is in the title. Can you find a Hero Fund that can lift up all of the others, thanks to its magnificent outperformance leading to a winning portfolio? And, what we found was that, more often, the portfolios that had the biggest difference versus the benchmark, and it’s not a big difference, but the biggest were in the heavier equity categories, particularly those that managed to locate outperforming international managers.
In the Heroes section, we actually take a look at each of the nine categories and say, which one had the biggest impact if you had a crystal ball? You could predict exactly which funds would have outperformed over that 10-year period, only pick from those funds in one of those nine categories and have the other eight be comprised just of the benchmark return. So, in essence, we were constructing hypothetical portfolios that could not lose, that were guaranteed to outperform the benchmark. What was the impact? Even at the best case with the international fund of picking that median international outperformer, it’s around 50 basis points a year. It’s not a lot. And, so, this goes back then to a bigger question around how much time and energy one spends on fund selection, manager selection, is paid off in terms of outsized returns. It’s not impossible, but statistically, we’re finding it’s very difficult, which points to how readers of the paper might think about allocating their resources and time.
Paul Murdock:
Interesting. Thank you so much for these insights today, Joe.
Joe Nelesen:
Thank you, Paul.
Paul Murdock:
To take a deeper dive into Heroes in Haystacks: Index Comparisons for Active Index Performance or to get our latest SPIVA Scorecards, visit us at the link below. Thanks and have a great day!
