All Robo-Advisors Are Not Created Equal: What to Look out for

This is a guest post written by the CEO of Scalable Capital, Adam French. All views and statements are representative of the author.

Here’s what to watch out for before choosing a robo-advisor

At first glance, robo-advisors all look the same. They create a low-cost, diversified portfolio for long-term investors. They offer nice interfaces and smooth client onboarding processes.  However, behind the scenes, robo-advisors can vary greatly in their investment approach – the actual service they deliver to their clients. Retail investors now have the choice between a lot of different propositions, and it may be tricky to choose one.  So what should investors watch out for when choosing a robo-advisor?

1. Do humans interfere with the investment model?

All “robos” offer an online sign-up process, most also have an app, but the extent to which they use technology at the core of their service – the investment process – varies widely. Some online investment managers rely on human wealth managers to decide where to invest and how to rebalance the portfolios, adopting a traditional rather than a technology-enabled approach to investing.

There is ample evidence that human interference with an investing model doesn’t add value over time. In 2012, American academic & hedge fund manager Joel Greenblatt tried to outperform an algorithm by choosing stocks from the pre-approved list that went into the algorithm and then exercising discretion about the timing of the trades. He ended up underperforming the algorithm – despite being one of the most successful investors ever. So before you entrust a robo-advisor with your money, ask yourself the question: is the robo-advisor purely data-driven, or do human emotions enter the picture?

2. Is the investment model based on modern capital markets research?

Most asset managers and robo-advisors like to mention that they are using “Nobel Prize-Winning Research” for their portfolio allocation process. What they refer to is typically a framework called “Modern Portfolio Theory” (MPT), which was, in fact, developed in the 1950’s and uses many simplifying assumptions about how capital markets behave. They were necessary to put the theory into practice before computers could work with the large amounts of data required to simulate the real, rather more complicated, behaviour of financial markets.

Most investment managers, both traditional and online, still use this theory today, ignoring the negative repercussions for their clients. The problem with this approach is that after several decades of observing and analysing how markets behave we now know that many of the simplifying assumptions of MPT do not hold true (e.g., losses of more than 2% are a lot more likely than the model suggests, and diversification doesn’t work equally well in all market environments). So by continuing to rely on MPT, most investment managers operating today construct portfolios and allocate assets in a way that neither diversifies risk as effectively as it should nor generates the best possible returns.

Instead of using this important, yet by now in many ways outdated theory, a new generation of robo-advisers use technology-driven investment models based on the latest empirical research. They use modern computing capabilities (“cloud computing”) to optimise portfolio allocations without having to make simplifying assumptions. By running tens of thousands of projections on how portfolios might behave in the future, based on recent as well as historical behavioural patterns of each asset class, they achieve a better representation of the complex financial markets. This enables them to continuously optimise their clients’ portfolios with the objective of achieving a better return for every unit of risk they clients are exposed to.

3. Do you get an accurate understanding of your portfolio risk?

In times of high volatility, it is important that retail investors understand exactly the amount of risk they take when investing. In the financial industry, risk is often described in vague terms such as ‘moderate’, ‘conservative’ or ‘opportunity orientated’, instead of giving a precise indication of the downside risk. The problem is that whilst a moderate portfolio may indeed on average have a moderate risk profile over the course of a 15-year investment, its risk can fluctuate significantly from year to year and even from month to month. During periods of turmoil, risk levels can multiply quickly and if the investment manager doesn’t work with dynamic weights across asset classes, the actual risk in a ‘moderate’ portfolio could easily increase to what most people would perceive as ‘adventurous’.

When you select a robo-advisor, ask yourself: can they tell you how much you could lose in a bad year? If so, how do they come up with that figure; can they back it up with data?

4. What are the fees of the robo-advisor?

Fees can easily eat into an investor’s returns. But when it comes to cost, the devil is in the details. There are 3 main elements to consider:

  • Annual costs: Most robo-advisors charge an annual fee as a percentage of the investor’s account value. But watch out: are the fees based on the average daily investment amount, or is the fee charged on the value at the end of the month? Since your investments, on average, increase in value, the former is more advantageous for the customer than the latter.
  • Performance fees: Most retail investors thinks these are “good” fees as they seem to incentivise their investment manager to do a good job. However, they should keep in mind that these charges can account for a substantial portion of the total costs, as a performance of 6% in a given year – which is not unheard of – will add 0.6% of additional costs. Additional problem: Performance fees charged in the past are not refunded in a year of negative performance.
  • Costs of the underlying funds: The costs of the funds/ETFs in which the clients are invested can vary significantly. If the online investment manager uses actively managed funds, the cost can amount for double or even triple of the cost of a pure ETF portfolio. Even if only ETFs are used, there may be differences of 0.1% to 0.3% per year depending on the ETFs.

5. Can you influence the investment universe?

Some robo-advisors in the U.S. and now in Europe allow investors to meddle with their investments and advertise that as a benefit. For instance, you might be interested in socially responsible investing.

While the desire of investors to have a say in how their money is invested is understandable, robo-advisors who offer this flexibility should be able to explain to their clients clearly how their risk/return profile is affected by this, as investors inevitably move away from the “Efficient Frontier”, i.e., the allocation which will give them the best return for the risk they’ve taken. Due to a lack of data on certain types of investments, it might not be possible to quantify how far the portfolio will move away from an optimal portfolio, which should then be clearly spelt out to the investor.

6. Is your robo-advisor independent?

More and more incumbents are starting to build their own robo-advisors. With this in mind, it has become even more important to check whether a robo-advisor selects funds independently or not, as you want to make sure you’re invested in the most cost-efficient and suitable funds/ETFs.