Q3/2020, Redfin, Realogy, eXp: Tide is High

Q3 was by all measures a bangin’ quarter, as my wife Sunny might say, for most of the companies in residential real estate. I already wrote about Zillow’s amazing Q3 results, and I thought I would tackle the three major brokerage companies (sorry, Fathom, I’ll get to you as well eventually) that are publicly traded. I’ll talk about RE/MAX separately as they are a franchisor, not a brokerage.

All three had great results in Q3, but for different reasons, and I think it’s the differences underneath the results that are interesting both for investors and for the industry. As I said in Q2, I think it’s more useful to look at some of the details rather than pore over the numbers since anybody interested in poring over the numbers have already done so or can do so elsewhere.

The long and short of it is that they all benefited from the hottest real estate market in recent (and not so recent) memory. It’s a bit like a participation trophy, but the way the three of them went about it highlights some interesting takeaways. Underneath the surface, however, their respective strengths and weaknesses, advantages and flaws, and strategic direction all remain unchanged. And none of them are anywhere near ready to take on the apex predators in the ecosystem today.

Let’s get into it.

The Agent Team: Redfin

As I mentioned in the Q2 analysis, Redfin absolutely killed it in Q2. Then they went and slaughtered even more in Q3, as the rising market buoyed everything (other than iBuying):

From the press release:

Revenue decreased 1% year-over-year to $237 million during the third quarter. Gross profit was $93 million, an increase of 74% from $53 million in the third quarter of 2019. Real estate services gross profit was $92 million, an increase of 70% from $54 million in the third quarter of 2019. Real estate services gross margin was 44%, compared to 35% in the third quarter of 2019. Operating expenses were $56 million, an increase of 22% from $46 million in the third quarter of 2019. Operating expenses were 24% of revenue, up from 19% in the third quarter of 2019.

Net income was $34.2 million, compared to net income of $6.8 million in the third quarter of 2019. Dividend on our convertible preferred stock was $1.5 million in the third quarter. Net income attributable to common stock was $32.0 million. Stock-based compensation was $11.3 million, up from $7.5 million in the third quarter of 2019. Depreciation and amortization was $3.7 million, up from $2.6 million in the third quarter of 2019. Interest income was $0.3 million and interest expense was $2.5 million, compared to $1.6 million and $2.3 million, respectively, in the third quarter of 2019. Direct and incremental costs related to COVID-19 were $0.3 million and are included in general and administrative expenses.

• Revenue: Down 1% YOY to $237 million, but most of that was due to RedfinNow, the iBuying operations
• Brokerage Revenue: $210 million, up 36.2% ($194.4 million in brokerage, $15.5 million in referrals to partner agents)
• Gross Profit from real estate: $92 million, up 70% YOY
• Real estate gross margin: 44% vs 35% in 2019, and up from 34% in Q2

If we ignore RedfinNow (we’ll come back to this later), it’s hard to imagine how Q3 could have gone better. Except that Glenn Kelman not only can, but did: things would have been even better had Redfin been able to staff up to take advantage of the insane housing market. From the earnings call:

The gain would have been larger had we recruited more agents as Redfin employees or as Redfin partners. In September, we gained 14 basis points year-over-year, our greatest monthly share gain since December of 2019. Keeping pace with demand is Redfin’s number one challenge. Year-over-year growth in customer inquiries for Redfin agents or our partner agents increased from 38% in July to 58% in October.

If October’s year-over-year demand growth remains at the same level to start 2021, we’ll have to increase our agent capacity 13%. In our last call, we said it would take until the end of 2020 to hire enough employees and partner agents to serve all our customers, but it’s now likely that we won’t be able to match supply with demand until the second quarter of 2021. This will limit share gains through the first half of 2021. From July one through October 31, the team recruiting employees and partners as agents is more than doubled.

I had to shake my head in amazement when I heard that; Kelman is quite literally admitting that Redfin left a lot of money on the table simply because they couldn’t take orders fast enough. As someone who worked in retail, I know the feeling: Christmas shopping season at Bergdorf Goodman is less about selling skills and more about getting to the register fast enough to ring up the purchases. Customers often leave when that happens.

If Redfin’s lead agent numbers in Q3 had remained the same as it had been in Q2, we’re looking at an astonishing 13.6 transactions per agent for Redfin. I think that’s a fair analysis, since new agents (Redfin added 278 of them in the quarter) typically don’t get productive in Redfin’s unique system for a few months. Yes, they added almost 300 agents, but how many of them were actually producing in Q3, versus the existing agents getting way, way more productive? I’m leaning towards the latter.

This points to the flipside of Redfin’s strength: the W-2 employee agent. On the one hand, they are far more productive than the independent contractor agent and Redfin is able to control the consumer experience far more. On the other hand, it takes longer to hire and train an employee as opposed to just recruiting them, throwing them into the market, and taking a cut of their commissions.

There is little doubt that Redfin left a lot of money on the table; they were on the other side of the spillover effect that eXp talked about. So let’s turn to eXp.

The Recruiter: eXp

On the other side of the spectrum, we have eXp, which has shocked the industry with its growth. Q3 was just stellar for eXp by every metric. From the press release:

• Revenue increased by 100% to a record $564.0 million in the third quarter of 2020, compared to $282.2 million in the third quarter of 2019
• Gross profit increased 103% to $46.8 million in the third quarter of 2020, compared to $23.0 million in the third quarter of 2019
• Net income was $14.9 million, or $0.20 per diluted share, in the third quarter of 2020, compared to a net loss of $1.8 million, or $(0.03) per diluted share, in the third quarter of 2019. The third quarter of 2020 represents the most profitable quarter in the Company’s history
• Adjusted EBITDA (a non-GAAP financial measure) increased 512% to $21.8 million in the third quarter of 2020, compared to $3.6 million in the third quarter of 2019
• Cash flow from operating activities increased 188% to $43.2 million in the third quarter of 2020, compared to $15.0 million in the third quarter of 2019
• Agents and brokers on the eXp Realty platform increased 56% to 35,877 at the end of the third quarter of 2020, compared to 23,034 in the same year-ago quarter.
• Residential transaction sides closed in the third quarter of 2020 increased 95% to 75,392, compared to 38,567 in the same year-ago quarter.
• Residential transaction volume closed in the third quarter of 2020 increased 112% to $23.6 billion, compared to $11.1 billion in the same year-ago quarter.

eXp has every reason to celebrate, and they will. Wall Street has already rewarded them — eXp’s market cap is over $3 billion and their stock performance looks like this:

And since a huge part of their value proposition to recruiting agents is stock ownership, I suspect eXp will see even more growth in agent count in the months ahead. Just as a point of reference, eXp’s market cap as of this writing is $3.07 billion; Realogy’s market cap is $1.24 billion. RE/MAX’s market cap is $579 million. At what point do the shareholders of the latter two traditional stalwarts think about swapping their shares for eXp shares?

But that’s too much speculation for today. Let’s move on.

Given that eXp’s model is the exact opposite of Redfin’s, the enormous gains in revenue, sales volume, and profitability came at the expense of… well… not Redfin specifically but of top agent teams. (I have long maintained and continue to maintain that Redfin is best thought of as a large agent team, rather than a real estate brokerage.)

From Glenn Sanford during the “earnings call”:

And as it really comes down to the idea that right now you have got so many people out looking to purchase homes that — the fact is, your busy agent who was already busy before this uptick and so there is a lot of spillover to other agents who aren’t as busy. And so it really is across the agent base and it’s actually one of the reasons why in Q3 we had so many of our agents capped. And that one of the reasons why our gross margin goes up in slower seasons versus higher — versus busier seasons is because when agents cap, we don’t earn as much for the balance of their year. [Emphasis added]

If we think about it, what he’s saying is that the good agents were in the same boat that Redfin found itself in: not enough staff to meet the demand. So they just let some of the business go, which then flowed to less productive, less great agents. That’s good for eXp and for other headcount-based brokerages… but just like Redfin is hiring like crazy and ramping up like mad, I have to imagine that the other agent teams are also hiring like crazy and ramping up like mad to service the demand. In the longer run, the weakness remains the same. This is not a structural change in the business model.

The good news for eXp, I think, is that based on what we’ve seen, they have had and are continuing to have some success in recruiting those high-level agents and their teams to eXp. I predicted this in Q2 based on interviews with some of the top teams in the country, as COVID made them think about virtual operations, which led to eXp’s virtual brokerage model.

On the flipside, since every strength has a weakness, eXp’s twin weaknesses remain: productivity and margins. Per-agent productivity improved significantly for eXp in Q3, but it’s still 2.1; yes, that’s a helluva lot better than the 1.7 posted in 2019 (which was a hot market) but it’s still just 2.1 transactions per agent in 3 months.

And on margins, the spillover effect really helped eXp post an amazing quarter. But note that eXp doubled transaction sides, and since home prices went up by 8.3%, the Sales Volume more than doubled. But its Company Dollar margins went from 8.2% to 8.3%, a 10bps improvement. Jeff Whiteside, CFO, touched on the reason in eXp’s “earnings call” (put in quotes as it was just an interview conducted by eXp itself, with no analysts asking questions):

If we look at gross margin, our gross margin was $46.8 million versus $23 million last year and that was 8.3% versus 8.2%. And Q3 margin was lower than the first half of 2020 and basically driven by higher cap agents, and that was consistent and expected with what we see from a seasonality standpoint. So that will change in the fourth quarter and go up slightly higher. [Emphasis added]

eXp has always been, and remains, what looks fundamentally like a 90/10 split brokerage with a cap. As more productive agents do more business, they hit the cap and the company dollar goes down fairly dramatically. (As far as I know, eXp charges a per-transaction fee after the cap.) Whiteside thinks that will change in Q4, and I’m not entirely sure why it would, unless eXp is counting on the less productive agents who haven’t hit the cap yet this late in the year coming in with deals in Q4. That could happen, of course, but it goes counter to the trends we’ve seen to date.

So on the whole, amazing performance in Q3. Seriously impressive all around. But how much of that is the crazy housing market that resulted in spillover effects that particularly advantage headcount shops like eXp and how much of it is something fundamental to eXp? Time will tell, but I don’t see the strategic change.

The key insight here is that it isn’t eXp’s “virtual brokerage” model that made the difference, but the recruitment of top producing teams that made the difference. The virtual model appealed to those teams, especially in COVID times, but it isn’t really built for them. And I don’t think you can claim today that eXp as a whole is built for teams, because it isn’t. That could change, which then changes eXp’s future prospects as well. But the peril for eXp and other high-growth models is that the top producing teams will eventually hire enough people to meet the demand, and the spillover effect will be minimized. Then what?

The Traditionalist: Realogy

In between the two extremes, we have Realogy, which has long been the standard-bearer for traditional brokerages. They too had a fantastic Q3, which is great news for them since their Q2 was so horrifying. From the press release:

• Generated revenue of $1.9 billion, an increase of 20% or $307 million year-over-year.
• Reported Net income of $145 million from continuing operations and Net income of $98 million including discontinued operations.
• Generated Operating EBITDA from continuing operations of $309 million, an increase of $103 million year-over-year, driven by higher transaction volume and strong performance at the GRA mortgage JV.
• Title and mortgage continued to contribute meaningfully to our business results, generating approximately $95 million in third quarter Operating EBITDA.
• Combined closed transaction volume increased 28% year-over-year in the third quarter with improving transaction volume through the quarter. Unit growth contributed 12% to this volume improvement across both Brokerage and Franchise businesses.
• Generated Free Cash Flow from continuing operations of $344 million vs. $116 million for the corresponding quarter last year and $395 million including discontinued operations vs. $174 million for the corresponding quarter last year.
• Strengthened the balance sheet reducing net debt by $276 million vs. third quarter 2019 with the Net Debt Leverage Ratio declining to 4.2x.
• Grew Brokerage agents 2% year-over-year with continued improving retention.

What a contrast from Q2, when it was bad news all around! Congratulations are in order.

Underneath the surface, however, there are still a couple of troubling spots. And as it has always been, the issue for Realogy has been, is, and will be for the foreseeable future, agent splits. That worsened by almost 300bps YOY. From the earnings call:

Commission splits increased 293 basis points year-over-year, largely due to the significant volume growth in the quarter. This is higher than previous trends, but we believe this is a worthwhile trade as you can see in our third quarter results.

Yes, it is a worthwhile trade. Question remains, will it be a worthwhile trade going forward? We got an interesting back and forth with an analyst and Ryan Schneider, CEO, on that question. In response to whether the high production this year will result in worse commission splits next year, when things get renegotiated, Schneider says:

Yeah. That’s a great question. So, look, we’ve got 50,000 plus agents in our owned brokerage business and we don’t do 50,000 negotiations a year. There’s going to be a bit of upward pressure on that just because the way most agreements are written is – most of these things just kind of roll over naturally. But they typically roll over based in some way off the volume you had the year before. So in a really, really high volume year, that lets people often start at a little bit better position than maybe if it had been a low volume year. Though the other side of that, because I don’t want to you run away too much with that from a negative standpoint is, remember the mixed things Charlotte was talking about. We’ve seen a greater kind of concentration of production among our better agents and a lot of them will already be kind of at the top of their table. So there’s not like a higher place to roll to in many cases.

The problem with that statement, of course, is that there is a higher place to roll to in many cases: another brokerage. eXp above would be a real threat to siphon off these better agents who are already at the top of their commission table. And there are plenty of brokerages that will go as high as 100/0, so now what?

Furthermore, the per-agent productivity number for Realogy’s brokerage units are not great, using the 52,400 agent count number from the Q3 10-Q. It is only 1.9 per agent, which is lower than eXp’s 2.1 per agent. The implication is that the bulk of the revenue and profit gains in Q3 came disproportionately from these top producing agents who are already at the top of their commission schedules, plus whatever spillover effect happened as those top agents simply could not take on any more clients.

The data seems to support that concern. Consider these stats for the Brokerage division:

• Sales Volume went up 22.0% YOY
• GCI went up 21.4% YOY
• Commission Paid to Agent went up 26.3% YOY

Yes, Realogy had a great quarter, but that’s largely due to rising tides lifting all boats. The business is getting more and more concentrated at the top, which means a dark cloud still hangs over profitability. Spillovers will go away, as top producers ramp up, and then we’re right back in the think of the profit compression puzzle.

Ecosystem Economics

The final big picture thing to think about is how all three companies, all very successful, all with great Q3, are looking at ecosystem economics. Because all three addressed it in their earnings calls.

Redfin talked about Title Forward and Redfin Mortgage:

After losing money in 2020, Title Forward should be a reliable source of profits. Redfin Mortgage meanwhile, had a major breakthrough. It’s first quarter of gross profits. Third quarter gross margins were more than 30%. Mortgage revenues for the quarter nearly tripled year-over-year, lifted by a near doubling in revenue per loan. with rates below 3% and competition for mortgage talent fierce, Redfin has like many lenders, limited volume by raising loan prices.

Just like their brokerage operations, their ancillary services are limited by difficulty of hiring staff to service the demand. But they are growing, and it is evident that Kelman and his team place a huge emphasis on the profits from these operations.

eXp talked about it as well in their earnings call. This is Glenn Sanford, CEO:

Sure, yeah. The affiliates services in my opinion sort of represents an opportunity for — to smooth out the transaction whether it’d be on the mortgage or title escrow side. One of the things as right now agents and consumers are working with multiple entities throughout the closing transaction and it can be somewhat challenging when you have to go to — you go to a real estate agent and then you have to go talk to a entirely separate mortgage company and then you have to close with entirely separate title escrow company, which from a RESPA perspective may continue to be part of the backdrop.

But the more that we can create a one-stop shop with the best mortgage services, the best title on escrow services, the home warranty and other things that a consumer might need, we can streamline that, makes it easier for the agent but also makes it better for the consumer. And so from our perspective, that just makes sense. Every mature real estate brokerage has these services at scale. The difference is we’ve grown so fast that just layering these things in. It’s actually more work when the company is growing 50%, 100% year-over-year, then if the company is relatively static because then you’re only working on one thing. But here we’re working on multiple entities at the same time trying to get those layered in and then that sort of as part of the value mix.

It’s a little bit of delicate balance because we don’t want to tread on or hurt the agent, which we certainly do not. But at the same time agents are somewhat loyal to their current mortgage professionals, current title escrow professionals. And then until there’s pain in their business, they generally aren’t looking for other services and so we want to just be there when there’s an opportunity to help improve their business and their life.

So on the one hand, Sanford sounds a lot like Rich Barton when he talked about the consumer benefit of creating a one-stop shop. On the other hand, he has to do it gingerly so as not to upset the agents who have their existing mortgage, title and escrow relationships. And he’s just starting to try to layer that in, because eXp is growing so fast.

Realogy, on the other hand, doesn’t have such concerns because it has had those in place for years and years now. It’s part of the Realogy Value Circle. So we get this from Ryan Schneider:

So, much of the world is talking about someday capturing the integrated real estate transaction economics. We are doing it at scale and at unprecedented levels. You’re seeing this in the third quarter, given $95 million of our operating EBITDA was driven by title and mortgage. We’re benefiting from strategic initiatives like our new mortgage joint venture and our technology successes, digitizing both title and mortgage closings. We continue to invest and expand our efforts to capture an even greater share of real estate transaction economics in the future.

He ain’t lying, right? Everybody else is chasing ecosystem economics; Realogy has had ecosystem economics forever, and dropped $95 million to the bottom line from its title and mortgage operations.

So how to think about this? Who wins the ecosystem economics games?

Simply put, Redfin has the best chance of winning the ecosystem economics game because they have employee agents. Not only can Redfin instruct its agents to talk about Title Forward and recommend Redfin Mortgage, they can do so without running afoul of RESPA. There is an employee exception to RESPA’s rule against quid pro quo arrangements. If Redfin can figure out a way to get their partner agents to refer ancillary business, at the industry standard levels of 15% or so, they’re looking at a very nice arrangement there.

eXp and Realogy both suffer from having independent contractors whom they have to be careful not to upset. At least Glenn Sanford was forthright in admitting the challenge; Realogy continues to pretend that it somehow controls the pipeline of ancillary deals, when it really doesn’t. They’ll both be focused on ecosystem economics, of course, but then… so will everybody else.

It Ain’t iBuyer; It’s iFlipper

All three talked at some length about iBuyer. Kelman got into detail about RedfinNow, eXp launched a new program, and Realogy touted RealSure. Go read the earnings calls if you want the details, and I might have to write a full post dealing with this issue separately.

What I can say at this point is that none of these three companies are in the iBuyer business that Opendoor and Zillow are in. The word “iBuyer” has now been hopelessly corrupted, so I’ve been using the phrase “market maker” to describe Opendoor and Zillow’s operation. But maybe there’s a better way.

What eXp, Redfin, and Realogy are doing (and so many of the privately held brokerages in the industry are doing) can best be described as iFlipper. They’re not actually buying homes at market value and selling them at market value, collecting a fee for doing so. They’re dangling “cash for keys” from flippers and investors as a carrot to generate listing leads. They’re really in the iFlipping business, and whatever doesn’t get sold to a flipper/investor, they’ll list and sell the traditional way.

Here’s Glenn Sanford of eXp talking about their new iBuyer platform, Express Offers:

And if you haven’t checked it out, you can certainly go to expressoffers.com, but that’s starting to really generate consistent growing lead flow for agents, for listing opportunities as well as partners who are looking to be buyers on the platform.

And one of the cool things but it’s — for us, we’re not in a position to lose money on every transaction. Some of the stats coming out on some of our competitors is that they are losing a lot of money on their transactions. We’re working directly with buyers who actually want to buy the homes and so they’re paying the value that they see or they’re going — turning into fully marketed listings to maximize the value that our — that sellers are getting through the eXp agents. So we’re really excited about the Express Offers platform.

This is typical of what all three companies said about their programs. Maybe iFlipping is the smarter play compared to market making/iBuying. I’ll have to address that in future posts.

But you know the one thing iFlipping cannot do? Change the world. And Opendoor and Zillow are both out to change the world. These guys are out to preserve the world as it is today. We’ll see which strategy wins out.

High Level Implications

What seems clear to me, looking across these three very different companies, is that they’re all kind of doing the same thing: get and keep high producers. The idea that you just need to recruit, because all agents are more or less the same, is gone, gone gone. Nobody believes that, if they ever did. All three companies talked about getting the best agents, even Redfin, which is a bit odd since what Redfin’s idea of a great agent is likely different from what eXp or Realogy’s idea of a great agent is.

So if they’re all going after the great producing agents, what are they offering?

In Redfin’s case, it’s simple: more money. Although Glenn Kelman did not really address this issue, fact is that Redfin’s agents are W-2; they are on salary plus benefits plus bonuses. Like any employer, Redfin will simply have to pay more for great talent.

In eXp’s case, it’s more money on the one hand, and stock in the company. There’s technology coming, but apart from Virbela, eXp has not positioned itself as some kind of a technology company. They’ll have iFlipping programs, and ancillary services, but so will everybody else.

In Realogy’s case, it’s a lot of talk about technology and growth initiatives, like affinity marketing and iFlipping, because it isn’t more money.

I guess we’ll see how these value propositions play in the marketplace.

This got way longer than I had imagined. And it was quite frankly a boring quarter. Everybody did great, but it’s largely due to rising tides lifting all boats. And because of COVID, the quarter was all kinds of weird, which led to fewer questions and discussions than normal. No one announced any major strategic initiatives, no one revealed much. And yet… I guess if you dig into the details, there are nearly five thousand words in there… sheesh, sorry but not sorry. You came here looking for this depth, didn’t you?

The big picture takeaways for me is that not much has changed for all three of these companies.

Yes, eXp made some real noise with its dramatic growth, especially in transactions and sales volume, but they were always the growth story, no? I’m not seeing how they turned the corner structurally on profitability concerns given their model. Spillover effects will fade, after all, since agent teams will grow to meet demand.

Yes, Redfin posted fantastic numbers, but they ran into a problem that is the flipside of its strength: W2 employee agents. As a giant agent team, Redfin was on the other side of the spillover effect that helped eXp and Realogy so much in Q3. And strategically, it is getting harder and harder to differentiate what Redfin is doing with what Realogy is doing.

Yes, Realogy had a fantastic quarter, but did not solve its core strategic problem: agent splits. Spillover effects will fade, top producers will find better offers from others, and Realogy is right back where they started.

All three should be congratulated for a great Q3, but rising tide lifts all boats. And underneath it all, their strengths and weaknesses and advantages and flaws remain exactly as they were before the tide came in. As Warren Buffett famously said, only when the tide goes out do you discover who’s been swimming naked. We’ll see who’s gonna be number one, because none of them are the kind of girl to give up just like that.

-rsh