Ranking the Robos

A new service evaluates the investment performance and user experience of all the new automated online investment platforms.

Have you ever wondered what kind of returns the machine algorithms and online platforms (aka “robos”) are actually getting for their clients?  How do they measure up with your own portfolio performance?  Are some of them noticeably better than others?  What kind of user experience do you get from them, and what can we learn from that?

These are some of the questions that Ken Schapiro, of Condor Capital Management in Martinsville, NJ, was asking himself several years ago when, out of curiosity, he started investing minimum amounts of his own money in these interesting new platforms.  Naturally, he tracked the results, and as robos proliferated like toadstools after a warm Spring rain, he suddenly had a significant amount of data that he wanted to share with consumers and the financial planning community.

The result is something called The Robo Report, accompanied by The Robo Rankings, both of which are available on the website of Schapiro’s subsidiary startup, Backend Benchmarking.  With his associate David Goldstone’s evaluation of user experience related to more than three dozen different qualitative factors, the two reports satisfy your curiosity in pretty much any area where it might have arisen.  In all, the company now has 47 different accounts on 28 different platforms.  “Our most recent additions were UBS, Wells Fargo and Morgan Stanley,” says Goldstone.

The Robo Report provides the most obvious comparison—performance, year to date (the second quarter report is out but not yet the third), one-year trailing, and two-year trailing (annualized).  Only ten of the portfolios currently have two-year track records, led by Schwab Intelligent Portfolios (9.48%), SigFig (9.33%), Wealthfront (9.25%), WiseBanyon (9.17%) and Vanguard (9.11%).  At the lower end of the spectrum, the outliers are FutureAdvisor (6.58%) and Acorns (6.73%).

Looking at the broader array of 18 portfolios with one-year trailing performance, you find everybody grouped between 4.85% returns (FutureAdvisor) on the low end, to 8.76% (Wealthfront) at the top.  Each of those are actually outliers; most of the others range from 6% to 8%.

In addition, the returns are broken out into equity and fixed income returns over the same time periods, the returns for an IRA account (19 robos now let you open an IRA on their platform), plus the fees that are charged (90 basis points a year for the Capital One platform, 89 for Personal Capital up to the first $1 million, as high as 50 for FutureAdvisor, United Income, USAA, Wealthsimple and Wells Fargo, down to zero for some smaller accounts and the most basic packages on other platforms).  The report graphs the Sharpe Ratios (none are over 1.0) and standard deviations for all accounts with at least a one-year and (separately) two-year history, plus the upside/downside capture ratios.  With a 2-year track record, Vanguard leads with both the highest Sharpe ratio (roughly 1.30) and lowest standard deviation (roughly 6.2%).

Of course, in some cases you’re comparing apples to oranges, which makes the return comparisons less relevant.  For instance, Schapiro notes that Wealthfront has made a relatively large bet on energy, which hasn’t served it well.  Others have decided to take on greater international exposure, and not all have a standard 60/40 equity/fixed-income allocation.  “We tried opening 60/40 accounts for everybody, but they don’t all offer that option,” Schapiro laments.  “So we’re now doing some normalized benchmarking calculations.” 

The benchmarking analysis looks at the asset mix of each portfolio, and then approximates the return of each asset class with a comparable ETF, and then calculates the robo’s return above or below the benchmark.  So far, all of the robos with two-year track records are negative compared to their custom benchmarks, ranging from -2.87% (Acorns) to a “high” of -0.02% (Vanguard). 

Qualitative Measures

The Robo Ranking report is a little different; it provides an overall ranking, plus sub-rankings, to show where different platforms fare over a variety of measures, including size and tenure, performance, costs, customer experience, interesting features, transparency and conflicts, the quality of financial planning services and access to advisors, and how large the account minimum is. 

You will not be astonished to learn that Vanguard Personal Advisor Services won the first “Best Overall Robo” award, followed by Betterment, with SigFig taking home honorable mention.  Vanguard was cited for offering full-service financial planning (by live advisors manning a phone bank), plus automated rebalancing and the best-performing portfolio.   Betterment won points for its superior user experience and technology.  Reading between the lines, it appears that Schapiro’s team found the Vanguard online experience a little clunky.  Betterment also allows its users to invest in fractional shares, and automates rebalancing and tax-loss harvesting.  SigFig was cited for its user interface and ability to aggregate outside accounts into the performance statement.

There are finer breakdowns.  Personal Capital is cited as best for complex financial planning needs and for digital financial planning, while Betterment was recommended as the best robo for first-time investors.  For advisors, it seemed to be a tie between Schwab Intelligent Portfolios and TD Ameritrade Essential Portfolios, with E*Trade and Fidelity Go analyzed at some length.

     Interestingly, Schapiro received some push-back from some of the automated portfolio providers as soon as he started publishing his statistics and analyses.  “We had Vanguard try to close their accounts,” he says.  “Wealthfront actually did close our accounts, but we had backups.  Others sent legal letters.”

The objections have died down recently, but Schapiro thinks that the entire financial services world is a bit sensitive to having its performance dissected.  “We audit our own performance, and provide it to prospective clients,” he says, “but most advisors don’t.  If you put your money at Fidelity, Schwab, Morgan Stanley, Merrill Lynch or XYZ Wealth Manager, you probably are not getting any type of track record that is audited, or a basis of comparison.”

Now that the subject is broached, how is Condor doing with its client portfolios, compared with the robos?  “We’re competitive,” says Schapiro.  He adds that his firm tends to have a growth bias, which has done relatively well recently, while most of the robos tend to be either neutral or have a bit of a value bias.

Is this a commercial venture, or just something designed to satisfy Schapiro’s curiosity and allow the profession to look over his shoulder?  “We’ve been working on an app, and written a business plan around it—seeking $10 million of funding in the Fall to execute the plan,” he says.  The target audience would be millennial do-it-yourself investors, validators and anyone who wants to benchmark the online platforms against their advisor’s performance. 

Schapiro thinks that a segment of his own clients would be interested in the data.  “They want to know, how do we stack up?” he says.  “How do our fees compare?  How does my investment selection compare with these platforms?  The plan,” he adds, “is to make our reports the arbiter of all that information.”