Skip to main content
automated investing: software

Eric Kirzner, professor of finance at the University of Toronto’s Rotman School of Management, is also a consultant to robo-adviser Wealthsimple in Toronto.JENNIFER ROBERTS/The Globe and Mail

Robo-advisers don't care about your children (and their costly university tuition), or about your parents (and their need for home care), or the kitchen reno that's eating into your investment budget.

All that plain-vanilla robo-advising services care about is the asset allocation of a client's investments, based on risk tolerance and age. Plain and simple.

That's both the purpose and the problem with the algorithms running today's automated investing services, say finance specialists.

"Honestly, it's not very sophisticated," said Moshe Milevsky, a professor of finance at York University's Schulich School of Business. "When you compare it to what Google's got, driving your car on the road without a driver, it's pretty basic."

The algorithms only learn so much about a client, which some argue is a good thing. It means investing is freed of emotion, which can lead to losses. Instead, the programs quietly rejuggle clients' portfolios periodically to maintain investment goals and risk preferences.

The most popular automated investment services will take investors through a series of questions to determine their risk tolerance, investing time horizon and what kind of return on investments they are looking for. This may also include a call from a broker from the service to verify the information and portfolio suggestions. The algorithm then begins a process of adjusting the assets periodically to try to hit the investors' preferences.

This shouldn't be mistaken for algorithmic trading programs, which buy and sell securities based on complex technical analysis of patterns and correlations within the market. Automated advisers' tasks are simpler, matching different kinds of portfolios with investors' preferences. "In fact, this is 30-year-old technology," Dr. Milevsky said.

It's even older if you consider that the basics come from Nobel Prize-winning economist Harry Markowitz's Modern Portfolio Theory introduced in 1952, if not from concepts of risk tolerance that are far older. The assumption is that greater risk should have the potential of bringing greater reward, and that diversifying a portfolio is a way to reduce risk.

Yet, like technical analysis, which studies big-picture market movements and not the fundamentals of individual companies, Dr. Milevsky argued that robo-advising doesn't look closely enough at the individual, either.

"I think that's the concern that I and many academics have looking at these algorithms. They haven't really stepped up to the plate to deal with who I am individually. They are dealing with me as an age, as an individual who has a certain time horizon. There are so many other things going on in my personal balance sheet," he argued.

Take a physician and a teacher, he said. Both are 35 years old, and both earn the same. The teacher may make less than the doctor throughout his career, yet he has a pension and job security. The physician may make more down the road, but she's self-employed and without a pension. "His portfolio, by virtue of what he does for a living, by virtue of his job, should look very, very different than the physician's," Dr. Milevsky said.

Mark Yamada, president and chief executive officer of Toronto-based Pur Investing, which designs algorithmic investment software, said that there are very sophisticated algorithms being used in finance. Yet those aimed at everyday retail investors "are pretty simple."

He noted that "all of the robo-advisers offer a simplified risk profiler that tries to get to not only the investing time horizon but also the short-term risk tolerance of every individual client."

Yet, because of the simplified nature of the product, automated advisers are tied by limited knowledge of clients. "More information is always better, but there are some practical limitations to that. And certainly with robo-advisers, you can't ask them [clients] 45 questions," Mr. Yamada said. "Maybe seven or eight is the max."

The main selling point is that robo-advising is cheaper, especially for clients who need less hand-holding. Yet the algorithms are based on the same kinds of assumptions as all investing. For instance, Mr. Yamada noted that despite the assumption that higher risk should bring the possibility of higher returns, there is the low volatility anomaly in the market, in which some low-volatility stock can have higher risk-adjusted returns than highly volatile stocks.

And then there is diversification. The assumption is that a wide variety of securities should help dampen risk. Yet when the market tumbles, diversification can be of little help. When the market falls, many diverse sectors fall. "We know that when they need it the most, it doesn't work," Mr. Yamada said.

But this isn't just a flaw with automated advisers, he said. "It's a flaw with all investing. What robo-advising is doing is they're taking the flaw, and they're selling it to you at a fraction of the price of a real live adviser."

Ultimately, it's about whether a plan is suitable for an investor. There are some who can completely figure out investments for themselves, some who may need a little prompting with a robo-adviser, and some who may need a lot from a financial adviser. Automated investing and its algorithms have a niche squarely in the middle of this spectrum.

"Some people really do need to have advice, they need to have direct advice, and they need to have it very tailored to their needs.

"But not all investors need that, of course," said Eric Kirzner, professor of finance at the University of Toronto's Rotman School of Management. He is also a consultant to robo-adviser Wealthsimple in Toronto.

Automated advising offers an asset-allocation product to investors who do not need further help, and it is attractive to many because of its low cost, said Dr. Kirzner.

"They do do some suitability analysis, but not anywhere near the depth of analysis that a full-service broker is supposed to do," he said.

But for all its limitations, it offers a service that can be tricky for individual investors to replicate, in terms of readjusting assets, keeping risk steady, or moving some assets for tax reasons.

"A little knowledge is a dangerous thing. Certainly, if someone said, 'I understand Modern Portfolio Theory and I apply Markowitz,' I'd be really suspicious," he said.