As the Department of Labor's fiduciary rule proposal was being discussed last year, robo-advisors were hailed as a potential solution to bring low-cost fiduciary advice to consumers. Labor Secretary Perez explicitly touted companies like Wealthfront as being a key part of the fiduciary future. However, it turns out that the final fiduciary rule has created a significant problem for several robo-advisors like Schwab and Blackrock's FutureAdvisor, that use the platform to distribute their own proprietary products.
In this week’s #OfficeHours with @MichaelKitces, my Tuesday 1PM EST broadcast via Periscope, we look at what the final DoL fiduciary rule really said about robo-advisors, how robo-advice must be delivered to satisfy the Level Fee Fiduciary requirements, and why the use of proprietary products by companies like Schwab and Blackrock - especially when combined with having investment discretion - may present a serious fiduciary problem when the rules take effect next April.
Which means ultimately, some robo-advisors may need to substantially change their business practices and how they operate in the coming months, to ensure they fit within the DoL's fiduciary rules. The most likely outcome may be for the companies to simply raise their fees, and eliminate their proprietary-product-based profitability, to ensure they're operating as Level Fee Fiduciaries. But doing so will throw a significant wrench in the growth of robo-advisors that were being cultivated specifically as a distribution channel for ETFs. On the plus side, though, it means the DoL's fiduciary rule may be more effective at limiting proprietary products than was initially believed!
(Michael’s Note: The video below was recorded using Periscope, and announced via Twitter. If you want to participate in the next #OfficeHours live, please download the Periscope app on your mobile device, and follow @MichaelKitces on Twitter, so you get the announcement when the broadcast is starting, at/around 1PM EST every Tuesday! You can also submit your question in advance through our Contact page!)
#OfficeHours with @MichaelKitces Video Transcript
Welcome, everyone! Welcome to Office Hours with Michael Kitces!
As many of you know, I wrote last week about the robo-advisor movement and how I think the B2C robo-advisor movement is effectively dying. Their growth rates are crashing, as the incoming dollars have flat-lined.
Robo-Advisors Pivot Towards A Distribution Model
I should note that I think there's no question that the robo technology is great. The technology is going to be here for the long run, and clearly commoditizes elements of the investment process. As many of you know, I'm a technophile, not a technophobe. I love the technology. But the whole idea of a direct-to-consumer robo-advisor seems to be dying off. That idea of being a viable intermediary and being able to market and not go bankrupt trying to get consumers to come onto your robo-platform.
And of course, that challenge is further undermined by a number of companies now that are competing by coming in and viewing robo-advisors as a distribution channel. I had actually predicted this as a trend a year and a half ago, saying one of the shifts you would see in robo-advisors is that the price of the pure technology would converge towards zero because companies that can distribute ETFs through them, or advisors who can package it onto some other value add, would eventually take the price of pure robo down to nothing, because they make money in other ways instead.
The Department of Labor threw an interesting wild card into this as well. As a number of you I'm sure noticed, when the DoL fiduciary rules were being discussed robo-advisors were actually cited by Labor Secretary Perez as an example of fiduciary advice through technology. But since then, a lot of questions have cropped up around robo-advisors, and how they're working, and some of the potential conflicts of interest. For instance, while Secretary Perez held up robo-advisors as a solution for fiduciary advice for small accounts, Massachusetts a month ago released basically a position paper that said they were very concerned about the conflicts of interest in robo-advisors.
How DoL Fiduciary Will Regulate Retirement Advice
So now we have this new DoL fiduciary rule. And I want to talk about the DoL fiduciary rule, and what it actually says about robo-advice in the final rules.
At a high level though, just to make sure we understand what DoL fiduciary said, the basic framework is: a DoL fiduciary cannot give conflicted advice and receive variable compensation where their compensation will vary depending on what advice they give. That's a fundamental conflict of interest and it's banned for fiduciaries.
Of course, if we banned all forms of getting paid to give advice, no one would give advice and we would squish out the entire existence of advice. So the solution to this in the Department of Labor rules is this thing called the Best Interest Contract Exemption. It says if you sign a Best Interests Contract with your clients that says you'll act in your clients' best interests for reasonable compensation and not have misleading sales communication - what are called the Impartial Conduct Standards - you will be able to give advice that is potentially conflicted, but still have it be permissible. That can even be advice with forms of commissions and other compensation attached to it, as long as you can demonstrate you're meeting the impartial conduct standards: best interest advice for reasonable compensation without misleading sales and communication.
In addition, the DoL provided a streamlined exemption that's called the Level Fee Fiduciary exemption. Level fee fiduciary essentially says if the only compensation you receive from your clients are level fees that do not vary based on the type of advice that you gave or where clients ended out putting their dollars, then you have a simplified form of fiduciary compliance. You still have to meet the impartial conduct standards. You have to give advice in the clients' best interest for reasonable compensation and not be misleading. But you don't have all the additional policies and procedures and additional disclosure requirements that will apply in the full best interest contracts. The idea is that if you are willing to operate as a level fee fiduciary, you get a simplified compliance environment because many of the worst conflicts of interest are eliminated by charging a level fee in the first place.
Can Robo-Advisors Meet The DoL Fiduciary Requirements?
In terms of robo-advice, the DoL fiduciary rule was interesting. It actually acknowledges robo-advice by name. They essentially stated "we know this thing called robo-advice exists." But then they said (paraphrased) "We're concerned that not all robo-advisors will necessarily meet the full fiduciary standards, but we're ok with any robo-advisor that meets the level fee fiduciary requirements."
Accordingly, then, the question becomes: Can robo-advisors actually be a level fee fiduciary?
Robo-Advisors As A Level Fee Fiduciary
This is where it gets interesting.
If we look at how robo-advisors can fit into the Level Fee Fiduciary rules, the answer is that some can fit easily, and some are going to have a serious problem with this. The classic robo-advisors, companies like Betterment and Wealthfront, should be fine in this environment. The reality is that they do appear to operate as a level fee fiduciary. They charge a flat AUM fee. It doesn't matter whether you get the aggressive portfolio or the conservative portfolio or which funds are in there. They charge you the same thing.
Notably, tiered pricing is okay. Wealthfront gives you the first chunk of dollars for free, then they charge you at the upper tiers. Betterment charges you more at the low end and then has a lower rate at the higher end; their fee schedule tiers down from 35 bps to 25 bps to 15 bps. That's fine. Level fee fiduciary doesn't actually say you have to have the same fee across the board regardless of assets. What it says is for a particular client the fees have to be level regardless of what you invest them in.
On the other hand, you would have a problem if, say, you charged more for the equities than for the fixed income. Now you have a conflict of interest where you can make more money by dialing clients' equity exposure up because you got paid more on the equity side than the fixed income side. Tiered pricing is okay, but the compensation has to be level no matter which investments you end out in within that robo-platform. Which fortunately is how many robo-advisors already operate.
Schwab Intelligent Portfolios: Not A Level Fee Fiduciary?
But here's the problem that Schwab Intelligent Portfolios has. When you look at the Schwab robo-advisor, about three-quarters of what they recommend are their own proprietary product, Schwab ETFs, and the infamous cash allocation to Schwab Bank that can be as much as a quarter of the portfolio in some scenarios. Schwab makes money on these. In fact, the reason that Schwab has a nominal price of 0% - it offers "free" robo-advice - is because they make money on the products inside the robo-advisor! In other words, they don't have to charge an intermediary fee because they're going to make money as the manufacturer of the products that go inside the robo in the first place!
Here's the problem, though: Schwab doesn't make the same amount of money on all these products. Some of the ETFs have higher expense ratios. Some have lower expense ratios. Schwab may make more money off the Schwab Bank cash allocation than it will off some of the Schwab ETFs. And Schwab makes no money off of the few non-Schwab ETFs in the allocation (or maybe they get a sub-TA fee or some back end fee but not the same fee that it gets on the products they manufacture).
So Schwab has the fundamental problem that level fee fiduciary was meant to crack down on, which is that you can't make different compensation as a level fee fiduciary on some investment allocations versus other investment allocations. That's not a level fee fiduciary!
And in fact, the problem is even worse for robo-platforms like Schwab, because not only do the new DoL rules say you can't have this uneven compensation where you make more on some than others as a level fee fiduciary. You would have to go through the whole best interest contract. The problem is the best interest contract doesn't save you from variable compensation if you have discretion over those investments, too.
Instead, the Department of Labor said that if you have discretion over the money, you cannot have variable compensation directly tied to those dollars at all. Or else, in the logical extreme, you would have a situation where the manager could say "hey, you know what, we decided we're going to dial you 100% into our highest expense ratio ETF where we make the most money. Oh, it's a shame that it's 100% emerging markets and that we just took your risk through the stratosphere. But we're going to make the most money!" That's not even a problematic conflict of interest you have to disclose.
That's an untenable conflict of interest that's just flat out prohibited, not just for robo-advisors but all DoL fiduciary advisors. No one will be able to have variable compensation over portfolios in which they have discretion. If you can change the allocations, you can not benefit more from one versus the other using that discretion. It's a prohibited conflict of interest.
So Schwab now has a problem, because they fit the exact version of variable compensation with discretion that the DoL fiduciary rules ban! They're not only not level fee fiduciaries. They can't even qualify under a best interest contract. It just flat out won't work, once the new DoL fiduciary rules take effect.
Blackrock's FutureAdvisor May Not Be Level Fee Either
Notably, this problem isn't unique to Schwab, either. I suspect FutureAdvisor is now going to face this as well.
As many of you know, FutureAdvisor got bought by Blackrock last year. And Blackrock bought them because it can become a distribution channel for iShares' ETFs which Blackrock owns. In other words, the Blackrock FutureAdvisor platform is likely going to look a lot like the Schwab platform. Schwab fills their robo with Schwab ETFs. Blackrock will fill FutureAdvisor with Blackrock ETFs. Similarly, it's not surprisingly then Invesco bought Jemstep. Invesco makes PowerShares' ETFs. You can see where this is playing out. All the ETF providers are buying robo-advisors so that they can put their manufactured product into the robo as a distribution channel.
Yet again, this doesn't work in a DoL fiduciary environment. FutureAdvisor is likely going to have the same problem. It may be a little bit different, because it's not a B2C entity anymore, it's a B2B entity. They're going to work with advisors. But DoL fiduciary says that you have this problem either if you serve retirement investors, or the fiduciaries who serve them. So if the advisor is a fiduciary to the IRA under DoL rules, and then Blackrock's FutureAdvisor robo works with the human advisor, that is a fiduciary connection as well. For which, once again, having variable compensation with discretion is prohibited. Blackrock's FutureAdvisor is going to have the same problem that Schwab has.
DoL Fiduciary Is Ruining The Robo ETF Distribution Model
Basically what we're seeing is that DoL fiduciary has "screwed up" the whole strategy of using the robo-advisor as a proprietary ETF distribution channel, particularly when your distribution channel includes discretionary management of the ETFs inside. It's one thing when you just plug your products in. It's another when you plug your products in and you have discretion over the allocation amongst those products, which fundamentally is the asset allocation service that robo-advisors do.
Notably, this doesn't necessarily mean all robos are dead, and that DoL fiduciary will force the end of all robo-advisors. As I've written separately, I do think B2C robos are dying off because of client acquisition costs. And this DoL rule will create some wrinkles for some robo-advisors. But again, it's not like the unequivocal banning of all robo-advisors. As I said earlier, companies like Betterment and Wealthfront actually come out looking better in this DoL fiduciary environment because they do act as true intermediaries that operate for level fees. They do not have the conflict of being a product manufacturer putting their own product into their own robo-solution.
Schwab Intelligent and Blackrock FutureAdvisor May Have to Restructure?
Because of the business potential, I doubt that Schwab and Blackrock will just walk away from their robo platforms in the face of DoL fiduciary. There are some other options. But what you'll see is a restructuring of the nature of the relationship. Perhaps they will become non-discretionary and let consumers build their own models, though that kind of defeats the robo purpose. I doubt they're going to go non-discretionary.
I think what you're going to see is that they will raise their fees, or at least they're going to raise their stated fee. What's likely to happen is that Schwab Intelligent Portfolios and Blackrock FutureAdvisor will mimic Wealthfront and Betterment in charging wrapper fees after all. The wrapper fee becomes a way to create a level fee fiduciary environment. For instance, they could try charging a 25 basis point wrap fee and then waive all the expense ratios of their proprietary products, claiming that if they just charge a flat basis point fee for everything and waive the ETFs expenses (making it up on the up front fee instead), they can operate in a level fee environment.
Though I'm actually not sure that will work. Level fee fiduciary looks not just at the platform itself, but the financial institution and affiliates and related parties. So in a world where the robo-advisor makes the ETF waive its fees, the ETF company still makes more money by having low fee funds waived then high fee funds waived. So you actually still have this conflict of discretion over variable compensation (or at least, variable profitability).
Thus, I suspect the necessary path to fix this may be what I call a "gross-up fee." They might say your total fee, your all in fee, will always be 30 basis points. If you buy a Schwab ETF that costs you 25 basis points, we're going to add 5 bps to that part of the allocation. If you buy an outside fund where Schwab makes no money, we're going to add 30 bps to it. We're going to gross up the cost so that no matter what your allocation is, Schwab's slice is a flat 30 bps, or 25 bps, or wherever they set the price point to be competitive. From the company perspective, they will be at a level fee. But it creates a huge mess of a fee structure, because clients will pay different out of pocket fees for different allocations to ensure the total fee is consistent.
But this is the challenge when trying to operate as a level fee fiduciary with your own manufactured product where you make different margins and different profits off of different products that you put in there! Particularly when they're managed accounts with discretion. It's the discretion plus variable compensation (from proprietary product) that create the problem. So either you have to give up the discretion, or you have to give up the variable fees and figure out how to level it across the entire structure.
Robo-Advisors For Only After-Tax Brokerage Accounts?
One other important issue to remember with DoL fiduciary is that at the end of the day, it is still only for retirement accounts and retirement investors.
Which means you can actually still have the conflicted version of the robo-advisor with proprietary products for brokerage accounts (i.e., taxable accounts), just not for IRAs. So we'll see what happens. We may find that suddenly a couple of robo-advisors start saying we're only doing brokerage accounts and we're not going to do retirement accounts anymore. They may say something like, "well we do it in brokerage accounts because that's where you get the value of the tax loss harvesting and all the rest". But in truth, the statement "we only do robo-advice for a taxable account" will be a red flag, a warning sign for, "we're so conflicted we can only do it in a brokerage account, we're not allowed to do it in a retirement account because of DoL fiduciary!"
So watch for this playing out in the coming year. The DoL fiduciary rules have just thrown a wrench in the works for this whole framework of robo-advisors as a distribution channel, even as companies like Blackrock spent $150 million trying to buy a robo-advisor to make it a distribution channel. Still, DoL fiduciary won't kill all robo-advisors. In fact, again, the early players are the ones that actually come out the cleanest in this because they already were functioning as level fee fiduciaries, at least from what we can see externally. It's the companies that were using this as a distribution channel for their proprietary ETF products where it gets messy.
So given that these rules kick in on April 10, 2017 - in a little less than a year - expect to see sometime over the next 11 months some announcements coming out. Schwab Intelligent Portfolios, FutureAdvisor, and perhaps some of the others - anyone that's got a robo-advisor attached to their own funds - will be trying to figure out how to restructure the relationship to be permissible in a DoL fiduciary future, where they have discretion and variable compensation. Because it just doesn't work under the existing rules the way that they've done it. Unless they avoid the DoL altogether by eschewing retirement accounts and just going after brokerage accounts.
Nonetheless, it's an interesting wrinkle to hit the robo-advisor movement, and came out of left field from DoL fiduciary. I haven't seen any discussion of it so far in the industry media. But it is coming. It is a problem. Stay tuned to see how the companies handle it.
So I hope this is helpful as some food for thought around DoL fiduciary, where things are going in the future, how this impacts robo-advisors and some of the changes that you're going to likely see. It doesn't necessarily kill robo-advisors, but it is going to change how at least some structure their fees, in ways that I think ultimately are a positive, because you're going to see more transparency as robo-advisors try to make sure they get the same level fee regardless of where your allocations are in order to be Level Fee fiduciaries. This as a plus for consumers, but for companies that view this as a distribution channel it's a real wrench in the works.
Thanks for joining us. This is Michael Kitces Office Hours 1:00 p.m. East Coast time every Tuesday. Hope this has been helpful for everyone. Thank you very much. Have a great day.
So what do you think? How might robo-advisors adapt to the DoL fiduciary rule and its Level Fee fiduciary requirements? Is this a sign that the DoL's limitations on proprietary products are actually more effective than previously believed?
Kerry Pechter says
I have pointed out that with robo/digital/Internet commerce the “medium is the business model.” Digital favors a direct, transparent, low-cost, open-architecture business model. The publicly-held, proprietary-product financial services firms can play well with it… only as long as they use B2C robo as a sandbox for young/new/low-balance clients–the way MerrillLynch appears to using MerrillEdge.
But that’s not their main business, and they will probably focus their robo efforts on B2B, supporting the advisor who advises high net worth clients, rather than supporting direct communications with mass-affluent clients, as Mike has written. So I believe they can have it both ways, but in different segments of their market.
Robo will eventually hurt the Merrills and BlackRocks in the sense that it will continue to put downward pressure on all intermediation costs. The DOL’s purpose, after all, was to make sure less money ends up in the pockets of intermediaries and more in the pockets of investors. The companies will probably preserve profitability by cutting head count. Marginal, commission-supported employee-advisors will be released. The remaining advisors will have more clients, and robo will make them more “productive.”
The Internet itself is the game changer… it just a took a while for the tidal wave to threaten financial advisors. The publicly-owned model is vulnerable. When Jon Stein of Betterment told me he intends to go public, I urged him to be careful what he wishes for. Going public will elevate his shareholders to a position higher than his customers, and the Internet model doesn’t favor that because public-ownership, ironically, frequently abhors transparency. After the IPO, Stein’s white hat will inevitably begin to grey. On the other hand, he will be much richer than Jack Bogle ever wanted to be.
Michael Kitces says
Bear in mind that whether declining intermediary costs are ‘good’ or ‘bad’ depends on where the company is positioned in the food chain.
For the products that Blackrock or Merrill Lynch manufacture, declining intermediary costs implies lower distribution costs and is a PLUS for the manufacturer. It’s a cost savings and better margins. That’s why companies with manufactured product are seeking out their own robo distribution channels in the first place. It IMPROVES the profitability of the manufacturer (at the cost of the intermediary).
Obviously those companies (particularly Merrill) also have an intermediary role (at least in other contexts of the company), so there is some ‘conflict’ here. But I suspect for most of these companies, their view is still that in the end, if there IS going to be some disintermediation, they’d rather have it happen to one of their OWN alternative divisions. It’s better than being disintermediated by third-party competitors, at least. 🙂
Kerry Pechter says
Absolutely. Everybody will have a robo at least as a placeholder. Each business model will determine what proportion of productivity gains from automation go to owners, shareholders, employees or customers. It’s really hard to generalize. My big interest is retirement income planning, and I don’t see any robos tackling the complexities of it yet, and only a minority of advisors pursuing it as a specialty.
So could someone launch a robo at $10 a month and then allocate the funds to the lowest-cost ETF in each basket?
Michael Kitces says
The issue here is ETFs that are owned BY the robo-advisor, not just allocating to ETFs.
You could make a robo that at $10/month allocates to expensive ETFs, and it might be a crappy service, but it wouldn’t necessarily be a fiduciary violation, as long as the robo didn’t OWN the ETF provider and have a conflict of interest in recommending those funds.
This is about the conflict of allocating to ETFs your company owns, not just allocating to low-cost ETFs.
Conversely, if your robo cost $10/month and it had discretion between allocating to ETFs with a 0.01% expense ratio and one that had a 0.02% expense ratio (and was the owner/manufacturer of either or both), it would STILL be a DoL fiduciary violation, even if those are the two lowest cost ETFs on the planet.
You have to separate “allocating to third-party ETFs” and “allocating to your OWN proprietary ETFs”. Very different conflict of interest situation, and thus different rules apply.
Understood and many thanks for explaining. Really helpful. Let you know when we’ve got the service live 🙂
Really appreciate you taking the time to break this down and highlight concerns for those of us potentially looking to leverage the technology in our current practices. You point out that Blackrock’s FutureAdvisor and Schwab’s Institutional Intelligent Portfolios will have a problem if they also serve “the fiduciaries who serve [clients].”
We have been working with the prospect of using Schwab’s Institutional Intelligent Portfolios for an offering to clients with lower asset levels, and your post gives me pause in this regard: Even if we customize all of the portfolios with non-Schwab ETFs, all portfolios are mandated to have a small percentage allocated to the Schwab Bank fund. Do you believe that even if our firm selects the portfolios and their weights, and just has Schwab manage the day-to-day rebalancing, etc. that we would still have a fiduciary problem because Schwab stands to receive variable compensation on the amount of funds allocated to Schwab Bank?
I would love your thoughts on this as we are weighing the benefit of Schwab’s Intelligent Portfolios vs. Betterment Institutional.
Thank you for your time.
Michael Kitces says
The point here is not whether you’d have a fiduciary violation for using the Schwab robo. The issue is that the Schwab robo would be violating its fiduciary duty in serving YOU as the advisor. Because even if the end client is “your” client, YOU are a fiduciary to your client, which means Schwab is a fiduciary to YOU… and Schwab can’t be a fiduciary to you if it has discretion over its own proprietary ETFs for which it earns differential compensation. Schwab would ostensibly either have to levelize their own compensation, or cede discretion.
Steve Smith says
How does this reasoning apply to the standard 401k in which Fidelity or Vanguard is the platform and they have a good number of their own funds on the menu? In these cases, sometimes there is an outside advisor (choosing the menu) and sometimes there is no advisor. In Vanguard’s case, even if there is an outside advisor, half the funds have to be Vanguard’s.
Steve Smith says
I see, in one of FI360’s summaries, there is an exemption (non-fiduciary communication) from the fiduciary rule for platform providers.
Would product TAMPs which contain proprietary products/strategies also be covered here? Seems to be a similar issue assuming the TAMP provider has discretion over their accounts and have their own products embedded within the TAMP which inturn generate different fees levels.
Michael Kitces says
Yes if the TAMP has discretion and the potential investment choices include the TAMP’s proprietary products (or the proprietary products of a related/affiliated company), it would create a problematic DoL fiduciary situation of differential compensation plus discretion.
Where Fidelity, Vanguard, and others are providing 401k advice solutions in 401k plans that have their proprietary funds as investment options, they have the same problem as described above for the robos and IRAs.
Michael Kitces says
The rules are different for 401(k) plans that merely offer a platform OF funds (that happen to include some proprietary funds). The issue for the robos mentioned here is that they have discretion OF/across the funds of the account, including proprietary and non-proprietary funds. It’s discretion PLUS the variable compensation of the funds that creates an issue, not “just” the inclusion of proprietary funds on the platform.
Sounds like this would also impact dealers that are owned by manufacturers and their discretionary programs used in retirement plan accounts where their products are offered?
Could a company like Schwab (or Merrill) avoid the conflict of interest by hiring a third party to do the discretion (e.g. Morningstar/Ibbotson)? In this respect, they would then only be paid for the “technology platform” and the financial advice component would go to the third-party? It further clarifies what you are paying for.. and to whom you pay it