Hello and welcome to the EisnerAmper podcast series. I’m your host, Elana Margulies-Snyderman, and with me today is Bob Elliott, Co-Founder, CEO and CIO of Unlimited Funds. Elliott, a former executive at Bridgewater Associates will share with us his outlook for using machine learning to create products that replicate the index returns of alternative investments, including the greatest opportunities, challenges, and more.
EMS:
Hi, Bob. Thank you so much for being with me today.
Bob Elliott:
Thanks so much for having me. Looking forward to be here.
EMS:
Absolutely. To kick off the conversation, tell us a little about the firm and how you got to where you are today.
BE:
Well, I’ve been a systematic investor for most of my career, and what I increasingly realized, sitting on the manager side of the 2/20 universe, was that many 2/20 businesses are very good for the manager and not that great for the investor. And the core reason why that is, is managers generate lots of alpha, but they also take it away in fees. And so that got me to start to think about whether there was a way to leverage technology, particularly modern machine learning approaches, to help build replications of how those two and twenty managers are positioned in markets. And because I’m using technology instead of hiring highly paid star PMs, I can do it at a much lower fee point than many other 2/20 strategies. And that’s really what we started at Unlimited, that’s what it’s all about, which is this idea of technology driven diversified, low-cost indexing of two and twenty strategies like hedge funds, venture capital, private equity, et cetera.
EMS:
That segues nicely into the follow-up question I have for you, Bob. I would love to hear your overall outlook for using machine learning to create products that replicate index returns of alternatives.
BE:
I think there’s been a heck of an evolution in terms of the techniques that are commercially available to start to do this sort of replication work. Back 20 or 30 years ago, the work by Andrew Lowe, the original work related to replication was using three year or full history rolling regressions of returns, looking at manager returns relative to asset returns. And the trouble with that is that hedge fund managers, as an example, move their positions around a fair amount and relatively agilely at any point in time. And so, the challenge that most replication strategies had was that long back-end window. Well, fast-forward to today, we have much better techniques available, Bayesian Machine Learning style strategies that allow us to probabilistically determine what is the most likely portfolio these managers are holding as close to today as we possibly can see in the context of that history. And so, what it allows us to do from a technology standpoint, is get a much more timely read of how managers are positioned than really was even commercially viable three or five years ago.
EMS:
Bob, as a follow-up, what are some of the specific greatest opportunities you see in your space and why?
BE:
Most investors want some allocation to alternative assets, and there’s a good reason why that is that actively managed strategies can often generate alpha, meaning returns in excess of standard index investing. The challenge is always that the fees are too high and so the managers earn all the benefit, and the investors don’t get much of the benefit. The typical hedge fund manager takes 80% to 100% of the alpha that they generate. And so, by using replication approaches, we have the ability to generate portfolios that are similar to how those hedge fund managers are positioned. But because you don’t have to pay people, highly paid folks who engage in the research and come up with their own proprietary views, you don’t have to pay them in the way that you do if you were running a hedge fund if you use technology. That really creates an opportunity to generate something I call fee alpha, which is one of the most durable alphas that exist, which is if you can get a lot of the benefit of how those hedge fund managers are positioned, but at a cost that’s much less than what a typical manager would cost, you as the investor can take advantage of that gap and get the excess returns in a way that you never have been able to.
EMS:
Bob, on the other hand, what are some of the greatest challenges you encounter in your investing space and why?
BE:
I think one of the biggest challenges whenever you’re using systematic approaches to investing, is that you have to carefully craft the investment technology and approaches based upon depth of experience. In the replication space, for instance, there’s been a lot of made by pure technologists without a lot of investment management expertise. And so, I think that’s a challenge because you can often build the technology that is disconnected from the reality, for instance, of how managers actually manage their portfolios, think about investing, think about exposures, et cetera. And so the key thing whenever you’re developing any systematic approach is to marry the intuitive common sense understanding of how markets or economies or managers work with the tools that are available and have an expert crafted solution to the problem instead of one that just either goes entirely discretionary, based upon a manager’s whim, or one that’s entirely systematic, not rooted in the realities of how money gets managed.
EMS:
Bob, we’ve covered a lot of ground today and wanted to see if you have any final thoughts you’d like to share with us.
BE:
Well, I think the overall industry is moving towards a world where fees are getting compressed. It was started 50 years ago with Vanguard bringing diversified low-cost indexing to stock investing and bond investing, and that. Frankly, that saved investors billions and billions of dollars. And so now it’s time to turn our attention to other areas of the market, those sophisticated strategies that are still too expensive relative to what investors get out of them and figure out a way to give investors better returns with lower fees and better risk return profiles than they have available today.
EMS:
Bob, I wanted to thank you so much for sharing your perspective with our listeners.
BE:
Thanks so much for having me. It was great.
EMS:
And thank you for listening to the EisnerAmper podcast series. Visit eisneramper.com for more information on this and a host of other topics. And join us for our next EisnerAmper podcast when we get down to business.