The Real Cost of Robotic Advice

June 20, 2019 | Investing, Guest Post, Robo Advisors

The Real Cost of Robotic Advice

Contributed by Markus Muhs - Portfolio Manager, Muhs Wealth Partners - Canaccord Genuity Wealth Managment

13 years ago, almost to the day, I first became licensed to sell mutual funds at a major bank. I always consider my financial planning/advice career to have started in late 2007, when I became a Financial Planner by job title, or in early 2009, when I actually obtained my CFP designation, though technically I remember becoming licensed to sell mutual fund securities right around when the Oilers were making their Cup run back in spring 2006.

It was a momentous occasion for me. Just around a year and a half earlier I started at the bank’s call center in downtown Edmonton. I worked on completing my Canadian Securities Course, hoping to transition from the banking call center to the discount brokerage call center on the floor below us. As luck would have it though, a few “Financial Advisor” (management-level personal banker) roles opened up in late 2005; a role two job-grades higher that I was completely unqualified/untrained for.

I realized quite quickly that the mutual fund selling role (which had its own title on a two-sided business card: “Investment Consultant”) didn’t really involve much in the way of financial advising or consulting. Each client interaction pretty much entailed going through a risk questionnaire, scoring it, then following the recommendation that the computer generated: an overpriced wrap fund or a combination of around 3-5 funds in a “growth”, “balanced”, “income”, etc. mix. The rest was paperwork and then usually you'd never see the customer again. Customers would then either transact on their account online or see a completely different representative the next time they were in the branch. It was very common to find customers coming in with mutual fund accounts laden with over 20 different funds, from them having seen a different rep each time they made an RRSP contribution, getting a different risk score each time, being put into a different favored fund or funds.

A fallacy, I think, in the way risk questionnaires and execution worked, was that when the customers meet a completely different rep each time, the manner in which the questions were asked - the guidance and explanation from the rep (or lack thereof) - the same customer could easily get a different risk score from one rep to the next. I saw this happen a lot, such as when I’d get a customer one day, who just last year saw the other rep in the next office, they’d score as “balanced growth” with me despite having scored “maximum growth” (two levels higher) with the other rep. If one rep really wanted to push their clients into 100% equities (for higher commissions, or at the bank, sales revenue points) it was easy to maneuver the conversation in such a direction.                                                                                                                                             

At that time I thought, “can’t a website simply do this all automatically?” You go to the website, fill out the risk questionnaire, send the company your money, and they automatically invest it in whatever automatically generated portfolio it comes up with. Throw in a retirement or education projection tool or something like that (something we already had available to the public on our retirement website) and save the cost of having a “Financial Consultant” doing all the machine input.

I think what I was describing is what we know today as “robo advice”. Robo advisors literally are an automated way to enter the info that the guy or gal at the bank enters for you (both in the risk questionnaire and the retirement projection tool), in turn charging you a fraction in fees (usually around half a percent vs over 2%) for the service.

I used “Robotic Advice” in the title of this post because – costs aside – I consider bank mutual fund advice and robo advice equally robotic.

Likewise, risk assessment, the advice around risk, and the behavioral counseling that every new (and sometimes also experienced) investors needs is completely lacking in both models. The risk assessment process in particular, at the beginning of any long term investment strategy, can be the single-most important client-advisor (or client-machine) interaction in an investment lifetime. Think of the long-term cost resulting from a wrong assessment

In recent years I’ve taken on a number of young clients (30ish, give or take) who had their investment portfolio either at a bank or a robo advisor. In most cases, their risk assessments were botched, through lack of proper advice, and they’d come to me allocated 60% stocks to 40% bonds (the typical “balanced growth” mix). I don’t mean to sound like a risk-happy perma-bull here, but logically no young person (let’s define “young” as under age 40 so that I may still qualify) saving up for a retirement that’s 30+ years away, should have more than 20% of their portfolio in bonds. It just doesn’t make any logical sense to weigh down your portfolio with low-return bonds, simply to mitigate short to medium-term volatility, when said volatility has no effect on your long-term goal.

When someone is dollar-cost-averaging into a portfolio it makes even less sense.  30 years till you need the money… what you’ve invested so far is a tiny tiny fraction of what you’ll have in the end… and you could be averaging into wild swings in the market along the way. The answer isn’t a balanced portfolio for a 30 year investment strategy, just because you don’tlike volatility, the answer is proper education, advice, and counseling along the way!

The easy cop-out for a mutual fund salesperson at the bank, is going with exactly what the questionnaire tells you after having the customer provide their answers (ie: “I want high returns with no risk…”), or better yet put them into a market-linked GIC, that pays just as much sales revenue.

At a robo advisor, where they don’t employ licensed representatives to go through the risk questionnaire with customers or to make investment recommendations, the easy route is also to leave the entire interaction one-sided (human to computer) and whatever the outcome, it’s entirely on the customer, who filled out the risk questionnaire on their own and to the best of their abilities. There, a customer going through the form may also answer risk and objective questions in an inconsistent and nonsensical manner, ending up with a balanced portfolio for their 30 year time horizon.

Did I mention that historically, at no point did anything less than a 100% equity portfolio outperform during any 30 year time period? Even the absolute worst: if you invested the day before Black Thursday in 1929, a portfolio of 100% stocks, with dividends reinvested, would have been worth more by 1959 than any other mix of stocks and bonds. Even in more recent decades, I’m sure some of my former customers at the bank, whom I might have put into an all-equity portfolio at a time when the S&P 500 was approaching highs of around 1500 in October 2007 were upset when the index was plunging to 800 and below. However, even if they had the misfortune of investing at a high, by today they’re sitting on a decent return if they stayed invested (that would a big "if", seeing as they had no advisor to guide them through the worst bear market in 80 years).

Anyway, here’s the point I’m trying to make: bad advice or lack of advice can cost you a lot more than the cost of good advice. Let me quantify this with actual returns of different asset allocations. A caveat, first, that below numbers are based on indices (you cannot invest directly in an index, these are not actual investments) and are based on past returns, which are not guaranteed to repeat themselves.

Over the past 30 years (March 1989 to March 2019), we have the following compound annual growth rates on U.S. stocks and on an aggregate bond index. Returns are in nominal $USD terms:

S&P 500: 10.11% per year
Bloomberg Barclays US Aggregate Bond Total Return: 6.16% per year

A diversified stock or bond portfolio would have roughly tracked those numbers, with all dividends reinvested, subtracting for any management or transactional fees. We can thus calculate these theoretical portfolio returns through weighted averages (in reality, how and when these portfolios were rebalanced over the years plays into this calculation, but keeping it simple):

Conservative (20% equities, 80% income): 6.95%
Balanced (50e50i): 8.14%
Growth (80e20i): 9.32%

Using our Millionaire Calculator (this and more calculators at my Resources page), assuming someone was age 35 in 1989, with $20,000 in investments, saving $500/mo, at the above Growth rate of return, they would have over $1.2 million today. If they went with the Balanced rate above, that number drops to $937K. That’s over 20% lost, simply because they filled out a risk questionnaire wrong in 1989, or at least didn’t have a good advisor to guide them through that important process.

Again, I must reiterate that above are not realistic returns for the next 30 years, at least on the bond front, where it is virtually a mathematical impossibility to see over 6% annually going forward. Half that might be more realistic, and then (assuming similar long-term stock returns) we’re looking at dramatic differences between the three asset allocations.

Once you apply different types of fees to the above numbers, the outcomes change some more. The point I’m making though is that it’s not as simple as taking one of those rates of return and subtracting 0.5% when using a robo versus 2% when using a human advisor to compare potential outcomes. Most of the 1.5% savings can easily be eaten up simply by making the wrong asset allocation choice.

Add to that the many potential timing screw-ups along a lifetime of investing, which is the focus of behavioral finance, and probably a topic for another day, but really how likely is it that you’re going to start on a fixed path with your investments today and stay on it for 30 years? I’m not suggesting someone would screw up their portfolio through market-timing, but even little things, like what if the hypothetical investor chose to suspend their monthly contributions when the markets were bad? (very, very normal behavior, from my experience)

 The graphic design folks in my Marketing department helped me create the above graphic for my Our Fees page. What I’m trying to explain here is that there’s more to investing than just fixating on cost. Yes, clearly, dealing with a mutual fund salesperson at a bank, like me in 2006, is the absolute worst thing you can do with your money. The fees are high, and at best you’re getting equivalent value to a robo advisor, only you have someone entering numbers and risk questionnaire responses into a computer for you, at a cost of 2% annually, when you could be doing it yourself.

There is a lot of value to be gained from selecting and paying for a good advisor, from the initial planning stages, through major life events, to navigating the inevitably unexpected volatility of the markets.

If you've actually read this far, maybe you'd like to have your own wealth situation assessed, whether you currently invest on your own through a discount brokerage, have it all in bank mutual funds, or are working with an advisor; real or artificial. Click Here and get in touch with me to schedule a review consultation either in person or by phone.

Markus Muhs, CFP, CIM
Investment Advisor & Portfolio Manager

To view Markus' SeekAdvisor Profile please follow this link here!