The world of wealth is undergoing a massive change due to one, the rise of AI and LLMs. Within a year or two, the S&P 493 companies start telling you that they're beating earnings, and the reason why is productivity, and the productivity is stemming from the use of AI. And two, a generational wealth transfer: $90 trillion flowing from Boomers to Millennials and Gen Z. So on number one, AI. If you look at the previous tech platform shifts like the rise of internet, cloud, and mobile, they did not have a particularly large impact on the world of wealth and RIA. That is not going to be true of AI and LLMs because the unique affordances of this new model and new technology will impact how financial advice is delivered. And then two, if you look at the generational shift, about $90 trillion of wealth is going to start transitioning from Boomers to Millennials and Gen X over the next 20 years. Those new clients have different expectations about how wealth advice should be delivered and what financial services should look like. The way that their lives are structured, they don't have as much time to read. One of the things that you've seen us do is pivot to multimedia. So we're at the cusp of two very big trends changing how the world of wealth will operate. To unpack those trends, I have joining us today Josh Brown, the founder and CEO of Ritholtz Wealth Management. Ritholtz serves over 4,000 families and has $4 billion in AUM. They're seen as an industry leader and an early adopter of technology, and even do angel investing in and around the ecosystem. Josh himself is a popular financial commentator, serves as a CNBC contributor, is an author, blogger, and has over a million Twitter followers. This is going to be a great conversation. So let's dive right in. Fun opener: ChatGPT, how are you using it at work?
Here's something that's really interesting, and I think it'll be solved. There are a lot of announcements from large firms that have built their own, I guess you'd call it an instance of ChatGPT, and they're kind of like these gate-kept pools of information that the AI knows it can utilize, and then it can't go outside the fence. And I think that first iteration on the street makes sense. It's really dangerous to tell. So if you're Morgan Stanley, you need to take what's going on. Morgan Stanley did this specifically; they built their own version with OpenAI. We actually recorded a podcast with OpenAI folks about how they did this with Morgan Stanley. It's crazy. They rolled it out to all of their advisors in a year, which even if it doesn't work very well for a big firm to adopt a new tech platform like that quickly, yeah, even if it's not amazing, that's pretty impressive. That didn't happen with mobile, cloud, internet. Well, when you realize that Morgan Stanley could probably keep itself from hiring an additional 3,000 paraplanners to support their advisors, you realize it's like, it's every bit as much forward-looking as it is being very realistic about the present, like how can we drop more money to the bottom line. So I think what they did is smart, though. You can't keep... All right, this is what I've learned. I'm doing this 25 years. I swear to God, this is a true story. I sat in a meeting in 1998 at the brokerage firm I worked at, or '99, might be '97, and the head of the firm came to the front of the room and he said, 'A few of you have been asking me about the firm's official policy on email.' I kid you not. Here's our policy: no email. You are not to write an email, receive an email, talk about email. You're not going to have an email address. We do not do email.' And you could understand why. It was FINRA, which at that time I think was still called the NASD, and they had these very stringent recordkeeping rules. And the last thing the firm and its compliance department wanted was to have another avenue through which people could commit, put things in writing, or clients could complain about their advisor, or anything that represents a new thing that needs to be policed. And then there was another meeting like a month later. 'Okay, due to client demand, we have decided there will be a firm email address, and all of your clients can send emails to that address. And if they say it's meant for you, we will print it out and we will bring a copy to your desk.' And that was the policy. And then another month went by. They said, 'All right, you know what? Actually, we're going to let you guys have email addresses, but you cannot send an email until you print out what you want to say, bring it to compliance, there will be a wet ink signature, we will put a copy in a manila folder, and then you could press send on the email.' And this went on and on and on. And eventually there was a surrender. It was, 'All right, you know what, guys, you can send emails, but also we hired an archiving firm through our IT outsourcer that's going to be scanning every email and storing it. And then actually we're going to read these emails with computers and we're going to know exactly what you're saying real time.' So that was the evolution. It always wrecks in my business in Wall Street. It always starts with no, and it always eventually gets to a place where it's like, we can't ignore the outside world. We have clients that are very comfortable with this technology, therefore we have to enable. Okay, so that's that. Now eventually you get to yes. It takes a while. So this was 1999. First major email service with MSN Mail was like in 1989, 1990. So we're talking 10 years to get to yes. But what I want to say is that for the most part, the adoption of technology on Wall Street is always in response to what clients want, and then secondarily to what employees are comfortable with in their personal lives. So it's a com—like Blackberries, like this is very common. Morgan Stanley kind of broke the mold. Instead of waiting 10 years and saying no, they said, 'All right, we acknowledge this is coming.' And on the underwriting side of our bank, we're raising a lot of money for companies that are doing this. On the research side of the firm, we're covering Nvidia. We can't ignore that this is going to happen. So rather than spend 10 years fighting it, what if we get out ahead of it and we engage with the leading technology provider and we build our own version?' And it was really smart.
And what do you think changed? Like, why with AI did Morgan Stanley say yes within, call it a year or even less from the ChatGPT moment?
So they don't have to hire everyone's nephew and everyone's son-in-law to be PowerPoint, to be like, yeah. So it's a CEX for senior advisors. The thing about AI, what I'm most excited about with AI, I know it leads to revenue growth for the hyperscalers, I know it leads to revenue growth for Nvidia, I also know it leads to a lot of cost and a lot of spending. But what I'm optimistic about is within a year or two, if it really has the amount of promise that people seem to believe it does, within a year or two you're going to start seeing the S&P 493 companies, not the FAANGs, start telling you that they're beating earnings. And the reason why is productivity, and the productivity is stemming from the use of AI. So we heard Domino's Pizza CEO on a conference call last quarter say they just made a $45 million investment in AI. Does that make the drivers more efficient? Does it better space out the time between when the product is ready? Does it enable the stores to do more local marketing? Does it answer the phone instead of a person? Whatever the reason is, I have a suspicion—not a suspicion, I have a hope—that Domino's will tell us a year from now how much more productive they are and more profitable, and it will be because of the upfront investment they made in AI. Replicate that across the 3,000 publicly traded companies in the United States, and you got something. So I totally—2025 is going to be amazing to actually see if the meets-the-road productivity gains—that's what.
So let me tell you a funny thing that happened last quarter. This is also the opposite of every other tech shift, which is you never saw the productivity number go up even though we had 30, 40 years of internet. This is, you know, the whole rise and fall of American growth—we actually had a slower last three decades of productivity growth, and this might be the first platform. I want to back up though. So I think we are probably going to see that, but why did they decide to lean in? And I've heard a couple different theories, so I want to hear your thoughts here. One is, well, they just learned there were three other shifts and they're like, okay, we get how technology works now. Two is the demo was amazing. It was this visceral demo you could do with ChatGPT in a way like an email, it's like I can do a phone call. Cloud is so abstract, it's like that back-office room that I never went into is now in the cloud, like I don't even know what that means. Whereas this is like, oh, I can see it, I can feel it, I can touch it, even if it's not perfect. So I'm curious, why did they lean in this time?
So let me give you an idea of what a typical day for a financial planning firm looks like. You've got—so Morgan Stanley is obviously, it's like, you know, 20,000 advisors, but just a typical financial planning firm, you have a calendar. Your calendar is filled with client meetings. If you're going to get on the phone with a $10 million household, you better be prepared with notes on everything that happened in the prior meeting and what actions have been taken since then. So that's 30 minutes to an hour prep. Then you get on the meeting, it's an hour. Then you come out of the meeting, it's an hour of debriefing other team members, sending instructions to your administrative assistant, sending instructions through Salesforce for different tasks: we need to open an IRA, we need to fund a beneficiary, we need to do this, we need to do that, philanthropic, we're missing this piece of paperwork. So that's an hour just entering that notes and the structured data in your CRM, and then just keeping everyone abreast of what happened and making sure that you're in the CRM saying we have this call, these are the 10 things we discussed, these are the actions that are being taken. So that's an hour of face time with the client, which is the only thing the client's really paying for, by the way, and then an hour on the front end of prep, and then an hour on the back end. It's three hours. What if there's an AI notetaker riding shotgun in that meeting, whether it's Zoom, which is very general, or even before the meeting pulls up the contextual information, or it's a purpose-built AI, which we have 10 of them now. There's Jump AI, there's all sorts of programs now being written specifically for wealth management, and there'll be a couple of winners there. So now I just took the typical financial planner, who is a very highly paid, high-value employee at a wealth management firm, and I just gave him or her back three hours of their day that they can use to provide more face-to-face contact with clients, which is what the clients are really paying for. And not only that, I probably eliminate the need for the 23, 24, 25-year-old hire who's not going to be profitable for 10 years, who's basically there to do all of this prepping and instructing and follow-up, if a lot of that can be automated. And by the way, there's a compliance component to this too, which we could talk about.
Yeah, and there's so many unack—like if you don't hire that, you know, 20-something to do grunt work, where does the next generation of advisors come from?
We'll talk about that second. I mean, there's so many other threads to pull on here. But so from a compliance perspective though, Rex, like if you own the firm or if you run compliance at a firm and you see that right after the call ends, the AI automatically summarizes the most important things that happened in the call and sends out that blast, then you know the client is fully aware what happened on the call. It matches up with what the advisor says. The firm knows what happened. The follow-up tasks can then be checked by AI to see if they actually happened. And we're talking to a startup right now, I can't reveal the name, but they're able using AI to pinpoint the moment the mood changes on a call, the moment the client veers from optimistic to nervous. And armed with that information, a skilled financial planner can literally work wonders for that family, knowing exactly when there's a tone shift in the conversation, what made them nervous. This is something that very skilled financial planners can pick up on naturally, but most can't. And that's where we're going. And so it's very exciting. I think the note-taking piece probably gets commodified pretty quickly.
Yeah, let's go there. This is one of the things I wanted to talk about. Is there's a bunch of ways you could just for the note-taking piece, which sounds small but is the tip of the spear in terms of connecting with everything else. You can take the Morgan Stanley approach, you can build it in-house on top of OpenAI. I think the big benefit there is you get access to your walled garden of data without having to permission to a third party, but you have to build it. The second is you can buy a general-purpose note-taking solution as a smaller RIA, but maybe not customized for wealth workflows, and maybe it doesn't have direct API connectivity to the fragmented wealth advisor tech. Or third, you could buy a wealth-specific version of this note-taker that does have that connectivity and is kind of trained. So like where do you see—I mean, Morgan Stanley's clearly chosen the first, there's kind of this middle ground that maybe some others, and then there's the third, which is customized tools. So like where do you see people starting to build? And I know you're in this too, right? Like as a firm evaluating these tools, as an individual playing around with ChatGPT.
Yeah, we met all these firms. We know what people are working on. We've invested in a bunch of things, but we are not utilizing any of it. And I'm going to explain to you why. I'm an early-adopting idiot. Every time something new comes along, I have to blog about it, I have to tweet about it, I have to try it out, I have to evangelize what's good. It never ever works in your favor in my business to be early because the early version is never the best version. And this is the thing that no one will tell you, but I will because I can't help it. When you early adopt and you're public about it, you get an SEC examination. There's going to be an SEC examination of every firm that has in their literature or in their Form ADV Part 2 or anywhere in writing that they're using AI. The SEC is going to say, 'Oh, are you? Interesting. Tell us more. Here's a 100-question questionnaire. You have 30 days to respond in writing. Tell us the nature of the AI. What are the safeguards?' So I've been through this twice already. When the robo-advice thing happened in 2012, we were very early. I think by 2014 we had launched our own robo-advice platform for small clients who wanted to work with us. We used a third-party software provider, which has since been acquired and orphaned. Now we're working with Betterment. But like in those early days, we were the idiots writing blog posts about how we're doing robo-advice, and the SEC said, 'Oh, are you? We'll see you soon.' And then they do a sweep. What I'm describing is what's called a sweep, and they look for every firm that's publicly talking about robo, and then they come in with a questionnaire, and then they follow up with an examination. Nothing wrong with that. You're going to get examined anyway. It's not something that you want to do all the time. So you answer. By the way, we had $2 million on this robo platform. We spent two months talking to them about it. So that's what you get for early adoption. And as if that weren't enough, when crypto came along, idiot that I am, yeah, you blogged about crypto, did the same thing. We launched a crypto index, basically like an SMA with WisdomTree. It was probably not perfectly timed. We did it like right at the end of the crypto rally of '21, right before the Sam Bankman-Fried stuff came out. Crypto is higher now, and the clients are fine. But I'm saying like we had to be the idiots who went out with a press release that we're doing crypto. And guess what? Yeah, right on cue, sweep. We get caught up in the sweep, and boom, we're in another five-month examination. There's again nothing wrong with being an examination, but the toll that that takes on your compliance people is incredible. They're working till 10:00 at night responding to endless questions. We made zero money on robo the year it generated an examination. We made less than zero money on crypto. It was so tiny, I don't even want to tell you the number of the amount of clients who invested in it. Both just complete and utter waste of time. Waste of time. Now we still have clients invested in crypto to this day. They're fine. We have now 50 million, 40 or 50 million in assets on the robo platform. Great, great situation. But my point is there's no award for being first. Pioneers get the arrows in the back. We sit back, and especially in financial services with adopting new technology, you have to be an idiot to run around screaming about how you're on the bleeding edge of technology publicly as an RIA. Nobody in Washington claps for you.
Yeah, and by the way, they're doing the right thing. That's exactly what the regulators should be. When there's an emerging technology, they should get their arms around, well, how...
I don't know if I would say exactly the right thing. There are elements they're punishing. So what they're not doing is punishing. What they're doing is they're making sure that these new technologies are not putting investors at risk or putting the financial system at risk. And you know, you could disagree with their stance on Ripple and all—I don't get into that stuff. My point is, yeah, we will adopt the right methodology of incorporating AI, both for advisor use and for client use, within our practice. But there is no race, and we're going to watch a whole lot of car accidents, and we're going to drive by those car accidents contented to just say, ultimately this will be ready for prime time, and then we will make big decisions. But we ain't going to be first ever again.
Yeah, I think it's one of the hard things for founders building anywhere in technology around financial services is you face what is true for everyone building in tech, which is you need to find early adopters, and early adopters are even more shut off by fear about regulatory recrimination.
Well, I try to balance. We're seen as a cutting-edge firm, and I think in many ways we are. I try to balance that ethos of the willingness to try new things. So we have an angel fund where we've made investments in some of these early AI startups. I think to some extent we can play kingmaker. We throw the biggest wealth management festival in the world basically once a year. We have demo drops. We have the ability to get a lot of eyeballs on a product. There are companies like Riskalyze and Vestwell where I was an advisor and a shareholder, and the firm was a user, and these companies have had unbelievable success. I hope that I played some role in that. So we're not Luddites, but we also recognize, no, there is no financial advisory firm that ever wins anything, market share or anything, for being the guinea pig for a tech company. It's just not a thing. So in some industries it might be. In my industry, it's the opposite. You win by implementing the best solution, and it will not be apparent what the best solution is in inning one.
Yeah. How—what do you think this does to the wealth, the RIA market from a fragmentation perspective? Right, we've got the big wirehouses like Morgan Stanley with a ton of advisors. We've also got, you know, decent-sized independent RIAs, but we have a new trend of there being more breakaway RIAs than ever. Does this create an opportunity for like more breakaway, like individual advisors starting new practices? Yeah. Do you think that this accelerates consolidation, fragmentation? Too early to tell, maybe?
It's too early to tell. I don't think so. I think there's a universe in which these tools become so common and the usage so widespread that it enables a smaller firm to hang on longer and hold off on hiring employees they can't afford. Maybe there's a world where that exists. I think though, I think it's probably a separate issue from whether or not there's consolidation. Because here's another thing about ADV wealth tech that nobody will tell you, but I will. Once something is good and somebody has it, everybody else has it. There's no exclusivity. And when you hear about an advisory firm bragging that they're building their own in-house tech, what a nightmare. That means they have to absolutely—that means that they have to hire full-stack engineers in-house, pay those salaries, and somehow manage to keep their tech up to speed with the APIs at Fidelity and Schwab, who can turn on or off a feature.
You think the Fidelity and the Schwabs of the world can do that successfully, or do you think even for them—well, no, they're the back—they're the custodian that all of the advisors rely on. Oh, you're talking about from an API integration level. I meant more because some of them also have advisors.
You have to make sure that Fidelity is not going to flip a switch and knock you offline, and that's all year. And that's not just Fidelity and Schwab. Or if you're building your own tech, ostensibly you're trying to connect your portfolio accounting software, which is Addepar or Orion or Black Diamond. So you have to keep that connection. Salesforce, don't get me started. The trading technology that you use to allocate accounts and rebalance—there is so many systems that have to all work in sync, and any company at any time can make a change to their own product that sends a ripple effect throughout your stack. So when you hear somebody saying, 'Oh yeah, we're building tech in-house,' no, you're not. LOL. Good luck maintaining it. So our industry is very, very reliant on wealth tech and very reliant on gigantic platforms, and it always will be. So I think the way to get this right, Rex, is to wait for a winner to emerge and then use the winner, because the winner is—we're in a business of compounding advantages—the winner is going to keep their product in sync with everyone else's because there'll be a two-way motivation in order for that to happen.
Yep, and I think I'm aligned with you. One of the questions I wear as a venture investor, and you do too because you do some venture investing in this category, is where will the value accrue in the stack? And I agree it's very hard to tell in early innings for all these different AI note-takers and different tools who is going to win. I mean, you've said it yourself, you're like, 'I'm a buyer, I don't know who to buy, and I don't want to buy anything because I'm scared of the SEC.' But there's another question—I'm going to talk my own book a little bit here, but I'm curious to hear your perspective on this. You mentioned custody. There's really only one new custodian who's come to market in the last few years, and that's Altruist. And you know, they're building modern APIs, they're probably easier to integrate with. How do you view someone like an Altruist? And maybe you don't even have a formed opinion, that's okay too. In terms of, you know, they're probably a lot of breakaway advisors, they have a modern custodian platform with modern APIs that you can plug into, and so at least your custodian becomes less of a blocker to potentially integrating and building some of these new services.
Altruist is winning because they have the flexibility to do things that the larger platforms don't. The larger platforms have a lot of tech debt. They've spent a lot of money already on systems that they won't just rip out and replace every time it seems convenient to do that. And so Altruist doesn't have that baggage. Can they maintain that? Call me in 10 years. When you're a scrappy startup and you have a client who asks you to build something specific, you're like, 'Yeah, we'll do it.' When you have 10,000 clients and all of them want something different, like how do you make everybody happy? So it's really exciting to watch Jason just run the table.
And are you an investor in Altruist, any?
I am. I want to disclose that. Okay. No, but we think the world—we think the world of the ethos and the speed with which he moves. And we have a lot of friends in the advisor community who are smaller firms, and they're happy. They're happy customers. Talk to them, and they say, 'I'm not in business without this.' Right. And now that I am in business, it makes my life way easier.
You know, some of that—so let's put our cards on the table and be honest. Some of that is frankly because Altruist is willing to take on clients at an earlier stage in their development that the large firms aren't interested in. It's not quite a head-to-head competition. I think what will be interesting to see is—take a—that's true for other, like Chime and all the neobanks, even the robo-advisors. They took clients no one else wanted. They created a new category.
That's exactly right. That's exactly right. It'll be interesting to see if a firm like mine—I'm a $5 billion—
That was my next question. What about someone like you?
Yeah, yeah. Okay, so we've talked to them, and you know, we have relationships with Schwab and Fidelity, and we're happy. And we've built—
They take you golfing, I'm sure. They take you some great golf courses.
No, we don't play that. I don't golf. So we work with—
Me neither. One of the reasons I'm in California.
Yeah, we work with them because they are highly capable of meeting our needs at every level. And we're a firm that onboards $5,000 accounts and billionaires and everything in between. And so they are the right solution for us. If we were to open up a third relationship, the question is, is it with Pershing, which is kind of the old guard third platform, or is it with Altruist? And if it is with Altruist, what segment of our client base would we custody there? It's probably not going to be the ultra-high-net-worth. Then you've got companies—you've got Apex, who really wants to get into wealth management custody. You've got what you know. And then the next, the next thing.
The whole SI layer is interesting again, putting my venture hat on. One of the questions is where does value accrue? And if you look at just AI broadly, the platforms like Google, Amazon, Microsoft have seen like over a trillion, several trillion dollars of market cap accrue to them because it's going to be hyper-competitive at the foundation model level. It's going to be hyper-competitive at the layer on top of that of agents. But they kind of own and run and distribute the compute on top of, you know, the Nvidia, who's an even lower layer, and that's super. So one of the takeaways, and I want to hear your reaction, is like, look, if you have a new custodian who can actually work with and move faster towards where the puck is going, that can be a great layer to deliver not just new services but also deliver value, and that platform itself becomes value. So do you think that's a fair characterization, or are there some missing links there if you look at, you know, like the Nvidias and the Microsofts of the world?
Yeah, so my take on this is that the large tech platforms win no matter what. Of course, I think you're exactly right there. I know we're earlier in the innings of like, you know, what this means in wealth world because it just takes longer for it to adopt in a regulated industry. Here's a hard and fast rule about the financial advice business: whoever is closest to the client wins. Everyone else has to accept fee shrinkage, has to accept disintermediation. If you are the one that talks to the client, if you are the one that the client is most familiar with, if you are the face, you win because you decide what you're willing to pay all of the middle pieces. So you can kind of dictate terms.
One of the other marketplace speak, what venture will say is aggregate demand. If you aggregate demand, eventually you can figure out supply. It can still be hard, but that's where you can control.
Look at the asset management business. Every year the average fee that people are willing to pay for a fund drops. Do you know why? Because the advisors who are recommending the funds want that piece to be as small as possible so as to preserve their own piece. So if I'm a 1% AUM fee to my clients, I can't add another 1% in internal costs in the funds that I use. So I'm going to use Vanguard at three basis points or BlackRock at five basis points, and that way I can keep my 1%. If I start adding things that cost 150 basis points, then I have to reduce what I'm charging the client, and I'm not going to do that. So that's the incentive that has driven the low-cost passive revolution in the advice world. Think about that on the software side. Anything that I do to increase the cost of doing business—
for my clients is coming from the money that the clients are paying me. I am, I am ultimately the arbiter as the client-touching adviser. I'm the arbiter of what I'm willing to pay for, and so the market of middleware has to fall in line. The other thing that we have is an epidemic of, I don't, I don't mean in a bad way, of feature companies that don't know they're feature companies. It's a, it's a Cambrian, it's a Cambrian explosion of companies that invented a feature and are raising capital around it and thinking that they have a company. And the first thing these, these, and this is a problem I face as investors, I think there's a lot of people doing important work, but the question is how big of a market is that work? If you're selling a feature to wealth advisers, because it's a big world when you look at assets, it's a small world when you look at how many different wealth advisory firms there actually are. In my experience, about half of them get it and their exit is to sell to a larger fintech, and then half of them really think that what they're going to do is get their, they're going to get their, uh, they're going to get their nose under the tent like with this feature, and then they're going to go horizontal and turn themselves into a platform. We really haven't seen that work. It's hard, it's so hard because you've got everyone else who's trying, or even if it's not everyone else, the 10% of the hundreds of everyone else are trying the same, same motion. Yeah, not everybody can be a platform, and not every feature is destined to become a, um, I understand you start it out as a, as a company because that's how you capitalize it and that's how you bring in investment dollars to build the product, but, um, this idea that RIAs are going to want to be platformed by 10 companies, uh, and, and each of them is going to do an individual trick and then that whole thing together is the bag of tricks and they're going to pay everybody separately, it's, it's, it's not great. But then we've seen the other extreme where you have Envestnet and they sell 10 buttons and you say to them as the client, I only want to push this one button that you have. I don't want to pay for the other nine buttons that I don't push, and they say, well, this is the platform. So I don't know what the right answer is. It's a moving target. I think the holy grail in our space is what we all want is single, single login. We want one layer that you log into and all of the features are interrelated and, and playing nicely together, and we never quite seem to be able to get there.
No, it's super hard. I do want to touch on business model. 1% fee is like the, the business model most of the wealth world is. This new layer of tech innovation going to put that under pressure? I think the, the bull case for why it would create fee compression is, oh, we're automating things. When you automate things, people will compete on price. The bear case would be, well, you're now doing more for your clients than you were before, potentially, right? So you can justify it, and, and maybe clients don't care. I think this is probably the biggest one is there's just a lot of inertia in the industry, even if you're seeing a lot of change. So where do you think about business model and fee compression shift? Obviously, it's happened at the asset management layer for ETFs, but what about for advisers?
We don't have fee, we don't have fee compression. We actually have people raising their fees in the, in the modern era. And what we, but what we have instead is margin compression. It's not enough to do asset allocation for clients. You have to do financial planning. You have to have CFPs. They cost money. Their time is very valuable. You also have to now be able to very specifically do, at a minimum, tax consulting. And a lot of clients are now, a lot of clients are now looking for firms that can actually file taxes for you. So now you need accountants in-house. Um, it's going to a place where you're going to have to be able to do trust and estate work if you want to compete for $5 million and up households. Right now you can outsource that. You can have, you can refer people to law firms, blah, blah, blah. Ultimately, where this is going is RIAs that have accountants on staff filing taxes and attorneys on staff literally going over legal documents. That's margin compression. The fee is the fee, not moving much. The amount of things you have to provide for that same fee is growing.
And that totally makes sense. And I think a lot of listeners can relate to that. It's like you don't want to have an asset management person and a tax person and then like a trust. It's just hard. It's hard to do. But the best business model is the 1% on assets, and so it makes sense for all those services to be aggregated under someone who can charge in that way.
Well, let me tell, let me tell you another thing. Accountants are used to being paid hourly or on a per project basis. You can very easily afford that when you're getting the fees on the wealth management side, and then you're tacking on the tax filing fee as a separate cost, which you kind of have to do. Um, and we surveyed our clients a couple of years ago, and we said we had, we already had CPA-level consulting in-house, but we weren't filing people's taxes. And we said, if we were to offer to file taxes for you, would you be interested? And we got a 70% plus yes response. So immediately we started hiring accountants. Now we have a tax sub. I think historically taking on that burden is a bad idea. It's a lot of work. It's a lot of people. It's a lot of things that can get a client upset. That's margin compression, but it doesn't scale. And it doesn't scale. It doesn't scale like wealth management scales beautifully. The stock market goes up three out of four years. The portfolios go up. Your 1% is worth more, you know, uh, every 10 years. It's probably, allocation generally is pretty easy. Like you kind of pick the plan, you stick to it, and a lot of the tools automate it. Client management, making sure they don't screw themselves, is important. Um, to win that business, you need to have, now to be competitive, you need to also have tax expertise. Everything you do on the investment side for a client touches their tax situation. Everything. Yeah. Now the risk is, and if you look at investment returns, the most important lever you can usually touch is taxes. It's taxes, fees, diversification, and tax is the number one, and it's the, the stickiest wicket. So now here's, so now here's the risk. If you underperform the market or you get a trade wrong, like you, you know, you just had a bad investment idea or something, it's forgivable. If you up somebody's taxes, not only are they leaving as a tax client, they're leaving as a wealth management client. They are not happy with you. You can't, so you can't half-ass this decision. Are you going to do tax or are you not? That's, that's really what it boils down to. You can't sort of, um, so this is, I think, going to represent a dividing line, and not all RIAs will be able to get over that line. And I think you're going to see a bifurcation. You're going to see haves and have-nots, which is why scale becomes really important. It's really hard to hire a full-time CPA who's going to work with business owners and, you know, wealthy families, etc., etc., at, at $50 million under management. So I think you're going to see a separation now of the firms that could afford to invest and the firms that couldn't.
Yeah, and you know, I asked you earlier to prognosticate a little bit about what AI means for the future of the wealth management industry, and I think we actually touched on what probably the biggest thing. It's not about who's going to win, who's already in RIA. It's the idea that accountants and tax professionals are going to be subsumed within the RIA industry because they have the best business model. That's 1% on assets. And now for the first time, you're saying, could I do taxes for my clients without suffering from excessive margin compression and not upsetting them by screwing it up because the tools are getting better? And I think the answer to that question, it's, I don't, you're not doing it for all of your clients. You just surveyed them. It's not, yes, let's do it today, but it's, it's more on the table than ever before.
Okay, yeah, or not, yes, but probably not for all your tiers, right? Like there's certain layers. Did we, did 500 families last year. It's a lot of work. We've got, uh, got, got 4,000 as the public number.
I saw like number of families you've got going. I'm saying for the taxes alone. Yeah, so, so no, 10% of your families you're doing taxes for today.
Yeah, so in order for that to scale, we need more bodies. Right now it could be the case, to your point, that it's bodies and AI, and, and having an accountant be able to do this work for more families is the unlock.
Yeah, I'm with you on that. I think you're right. Yeah, and I think that, I mean, that alone, it doesn't sound that interesting because those people still exist. Um, they're just inside of a different firm than they used to be. Um, and I think that, that to me, I feel pretty comfortable about because I think best business model wins in terms of consolidation. And to your point, owning the client and the assets, it's probably the best. When, when you onboard a new client and part of your onboarding is not just here's your portfolio, here's your financial plan, but you, you have a call scheduled next week with our in-house tax group to see if you're a fit for, you know, consulting or for actual tax filing, tax prep. Yeah, uh, that's part of the business model now.
And we don't have clients that are like, no, I don't want that. It's a weird thing, and I can't believe it, but it turns out the majority of wealthy families do not particularly care about the current accounting solution they have. It's not meaningful to them. It doesn't inspire them. They're not like, I have a great guy. Some of our clients do. You know, I hear when I ask all of my friends, like, how do you feel about your tax person? They're like, I don't, I don't like them. Like, or like, hey, do you have a recommendation? I do not have a recommendation. But the right, and then the other thing is the, the degree to which it fits naturally. Like if you're doing the wealth management for a family, you already have 80 to 90% of what you need, yeah, in order to do their taxes. It's in, it's, it's intrinsic to what, to what you're doing in the first place. It's not like, hey, we do wealth management and also, uh, pizza. Like it's, it's, it's literally part of what we're doing. And so that's why I think the future of our industry is going to involve both. Um, the key, again, the key is going to be to make it really clear on the tax side we can't screw this stuff up because it's jeopardizing not just the $3,000 that we're getting paid for tax work, but it's jeopardizing like the wealth management fees that are really the whole reason that we exist.
