Asking the Right Questions about Risk and Robos
FORT LAUDERDALE, Fla. -- As regulators begin examining robo advice offerings, firms developing financial software and digital platforms say advisors who are transparent about how they use such tools have nothing to worry about.
Some executives and industry insiders even question the value of relying on a questionnaire to determine risk tolerance and make recommendations -- a method that has attracted scrutiny from critics of automated advice.
"[A questionnaire] is insufficient by itself," says Praveen Ghanta, co-founder of portfolio stress testing developer Hidden Levers. "At best, it answers half the question, which is, 'What are you exactly comfortable with?' If you don't know where you are today in terms of your portfolio, how do you determine that match between what you're comfortable with and what I'm recommending for you?"
Recently, FINRA officials said they have started reaching out to broker-dealers who offer robo advice, asking how they determine client risk tolerance, among other factors. SEC officials have also publicly questioned whether a fiduciary duty applies to automated advice, and if laws need to be revised to reflect new innovations.
Software firms attending the Technology Tools for Today conference agreed that regulatory oversight of robo advice is necessary.
"Five years ago, robo advisory wasn't really discussed, and now everyone is coming out with a robo advisor. They have to find a way to regulate that and protect the innocent investor," says Kevin R. Miller, CEO of performance evaluation tool E-Valuator.
Regulatory consulting firm NRS distributed a white paper about robo advisors at the conference with an ominous title: Embrace the technology, but be aware of the regulatory scrutiny. Among the paper's recommendations is that advisors using robos still be involved with clients in the fact-finding process at the start.
"Relying entirely on a questionnaire provided by a robo vendor without supplemental interviews or additional fact-finding is not a good idea, particularly because many of the questionnaires are not very robust and there is a chance that clients will answer incorrectly, leading to allocations that do not adequately meet their needs," says NRS consultant Amber Tatman.
The group's recommendations echo a critique of robo advisor questionnaires raised in an asset manager-funded whitepaper that argued robos fall short of fiduciary law, partly because they may not be comprehensive enough to generate appropriate advice for a client.
- Read more: How Robos Fall Short of Fiduciary Law
However, conference organizer Joel Bruckenstein argues such critiques don't have merit.
"I don't understand what the concern is," Bruckenstein says. "Advisors use the same process. Technology has made it more efficient, so why should anything change? If it is suitable using paper it's suitable with technology."
Bruckenstein compares such analysis to industry concerns with the advent of social media. "It's the same that it's always been," he says. "If it can't be in a letter or testimonial, it can't be on social media. It's common sense."
The only problem Bruckenstein could see is if a client were able to ignore the risk score generated by a digital form and be able to opt for another portfolio allocation level.
'MANAGE IN THE TECHNOLOGY'
A number of technology vendors pointed out such concerns ignore customization. Firms will incorporate client questionnaires into their digital solutions and offer a variety of approaches to score client risk tolerance and generate allocations.
"We have our own questionnaire, but if [a client] has a questionnaire we'll put it right in our tool," says Michael Wilson, president and chief operating officer of AdvisoryWorld.
Still, NRS's Tatman says advisors should be as critical of the questionnaires used to direct robo advice as she expects regulators will be.
"How robust is the questionnaire?" she says, listing a variety of concerns: "What happens if there are inconsistent responses in the questionnaires? How does the information get updated? What about anomalies?"
AdvisoryWorld's Wilson says there's a shared responsibility when advisors seek technology solutions from a vendor. "We always coach [our clients] to be as transparent as possible in the scoring methodologies. If the questionnaire scoring actually changes how we score existing and proposed accounts, that's all stuff we can document."
That ability to provide exact documentation instantly is one of the powerful aspects of moving to digital solutions, Wilson says.
"Transparency means you can say, here's our questionnaire, based on these answers, here's how things map out. It goes to a certain category, here's how it lines up with a certain risk tolerance level, and it's repetitive. We score every existing account, every proposed account, every model account the same way. We don't deviate."
Scenario-based stress testing and data visualization is a better way to gauge an investor's risk tolerance anyways, argues Hidden Lever's Ghanta, one of several providers offering advisors analytic tools to develop portfolio recommendations.
"We were talking to one advisor who was recommending to us to record the market past six month's volatility, along with a timestamp on a client survey when they took it because if they took it during a calm market then their answers are likely going to be different in a period of volatility," Ghanta says.
The debate about oversight and technology provides an opportunity for advisors to reflect on how they deploy robo advice in their practice, says Alexey Sokolin, chief operating officer of wealth management platform Vanare.
"There's this categorization that a robo advisor is to the side of my business, autonomous from me, therefore I don't see into it; it just kind of automatically rebalances things and hums along," Sokolin says. "If a client's not big enough, I'll put him there.
"In that case, there is a challenge. You have to open it up and manage your clients in that technology. You need to see everyone that signs up, you need to look at how they answer questions, how they profile themselves, what are they invested in. You’ll need to build all those tools to have a look-through into the algorithm and have real oversight."