Description

Summary

Compass is a real estate brokerage that has invested in building software that makes its agents more productive. Since going public at $18 per share, it has been shunned by internet investors for not being a real tech company, and shunned by real estate investors for not being profitable. The reality is, Compass is a brokerage that has gained massive market share from competing brokerages, and will continue sustainably doing so thanks to its ability to enhance productivity of its agents through software. Compass currently trades at a valuation cheaper than brokerage peers who are secular share donors, and I believe it’s a great long from $8.

Market overview

Compass operates in the US real estate brokerage market. In 2021, there were 6.1mm existing home sale transactions in the US. These homes were transacted at an average price of $370k, contributing to a total transaction value of $2.2tn. The vast majority of these transactions involved a real estate agent, an occupation taken up by over 1.5mm people as of 2021.

Historically, both US transaction count and average price grew over time, though the former is more cyclical than the latter. From the 1978 cycle peak to 2021, transaction count grew at a 1% CAGR. From 2006 to 2021, average price grew at a 2% CAGR, despite price first falling (20%) from 2006 to 2009. The number of registered real estate agents also grew at a CAGR of 1.5% from 1980 to 2021.

Despite 2021 being a hot year for the housing market, we finished well below the 2005 cycle peak of 7.1mm transactions, a year when the US population was 296mm vs 330mm today.

Where does Compass fit in?

Compass plays in the high-end segment of the real estate market, transacting homes at an average price of $1.1mm, well above the national average. It recruits the most productive agents in each market it enters, who tend to be agents transacting higher-end homes. It has thus been able to achieve 6% share of US transaction value with only 1.5% share of agents.

Per US Census Bureau’s American Community Survey and third party industry research, 20-25% of US home sales are priced above $500k. Given price, the transaction value mix of this subset will be a lot higher than the transaction count mix. Compass only has 2% share of transaction count today, so there is ample higher-end market share left to grab.

Driving agent productivity growth

Compass’ biggest competitive advantage is its ability to drive productivity growth in its agents. While its expansion strategy involves poaching the best agents in each new market with attractive split incentives, these incentives can only be temporary, and any competitor can replicate this strategy once Compass stops paying out high splits. This begs the question: will Compass’ market share gain be sustainable? My bet is it will indeed be sustainable because agents stay at Compass even after their splits normalize, and they stay because Compass with its integrated suite of software tools makes them more productive and efficient at their job.

From 2018 to 2021, the average agent at Compass went from transacting 5 homes a year to 10 homes a year. Meanwhile, the average agent at eXp or one of the Realogy brands made minimal productivity improvement. Compass is able to achieve above-market productivity gain for its agents not only because it has better software tools, but also because its agents actually use these tools. It is true that making CRM, marketing automation, market intelligence, and digital closing tools for agents isn’t rocket science. If you talk to a competing brokerage, they’ll likely point out that a collection of third party software tools are already offered by vendors like MoxiWorks and kvCORE. What they won’t say, however, is whether they can get their agents to actually use these tools. These tools are proven to increase efficiency and productivity, but agents often ignore them due to the nuisance of managing workflows across several disconnected tools. When your email marketing tool is provided by a separate vendor from your CRM, managing contacts across two separate databases becomes prohibitively painful. Compass spent the time and money to build a proprietary, integrated suite of tools, where agents can access every feature in a unified operating system, and it paid off in the form of superior adoption among agents. While 30% of agents in the country use a CRM, 85% of Compass agents do. And once an agent becomes accustomed to the easier, more efficient way of doing business, he is loath to return to the old, inefficient way.

Thus, principal agent retention is north of 90% at Compass, compared to an industry average of 70%. Agent productivity drives agent retention, and ultimately Compass can afford to spend more upfront to acquire agents because it can generate more lifetime commission income from them compared to competing brokerages.

Taking market share

Both Compass and eXp have been taking share from Realogy, though they don’t compete directly against each other. Compass transacts $1mm homes in the high-end market, whereas eXp transacts $300k homes and provides bare minimal support for agents who are just looking for a license and the highest split possible.

Notably, Compass continued to take share from Realogy in 2021, when its commission split improved vs. 2020. Realogy’s split continued its trend of decline in 2021, and that wasn’t enough to stem its share loss. As of 2021, the overall split difference between Compass and Realogy is quite small, which suggests there are other factors (such as productivity gain and agent lock-in) driving Compass’ continued share gain. 

Core brokerage profitability

A key debate on Compass, given its history of losses, is whether it can become profitable. Setting aside potential contribution from ancillary opportunities, which have higher margin than the core brokerage business, I believe the core brokerage business can achieve 6% EBITDA margin in the next few years.

Compass likely finished 2021 at EBITDA breakeven, which implies 6% incremental margins vs. 2020. This was achieved despite its entering 23 new markets in the first 3 quarters of the year, up from 44 at the end of 2020. These markets made negligible revenue contribution and were a drag to expenses.

Compass’ 2018 market launches, which consist of Dallas, Chicago, San Diego, Seattle, Atlanta, Austin, and Philadelphia, reached 4% market-level margin and 11% market share 3 years after launch. Two of these markets reached 6% market-level margin. The market share uptake is just as notable as the margin uptake because it confirms the viability of Compass’ strategy outside of gateway markets.

In fact, inland, mid-tier cities typically have 25-30% brokerage splits, which are more favorable for the brokerage than in San Francisco and LA (NYC is the outlier with 30%+ brokerage splits industry wide). As more of Compass’ future transactions come from mid-tier cities, its gross margin will improve from the mix shift.

Compass likely spent $250-300mm in R&D expense in 2021, up from $144mm in 2020. It employs 1,500 technology staff. While it continues to hire engineers from Big Tech and invest in new features, its core software suite has been built already, giving it a wide technology lead over competing brokerages. I do think Compass will continue to invest in R&D, but most of that future spend is discretionary in nature, and topline growth will outpace growth in R&D spend going forward. If management were to throttle all innovation and simply run the business for cash, it can find $300mm of EBITDA without impacting agent payout and marketing intensity, which is comforting in the context of Compass’ mere $3bn enterprise value.

Higher agent productivity and higher average transaction price help Compass generate superior unit economics. Despite taking a lower split than Realogy, Compass generates 2.3x as much net revenue per agent. Compass can afford structurally higher overhead per agent and still achieve 6% margins in its brokerage business.

Value from attaching title and mortgage

Compass has communicated its goal to achieve industry level attach rates in title & escrow and mortgage origination, but hasn’t disclosed the potential unit economics of these ancillary services. Title and mortgage have incredibly accretive unit economics to the brokerage business, and the eventual reflection of this in realized financial results will be a powerful catalyst for the stock.

At its core, the value provided by a brokerage in title and mortgage is its ownership of the customer relationship. Underwriting title insurance and providing mortgage loans are commodity business models. First American pays out 80% of gross premiums in its agent channel to the title agent because the hard part is bringing in the customer. In practice, most homebuyers see title & escrow as a necessary nuisance and don’t care which vendor is selected. Many homebuyers also lack the sophistication to rate shop their mortgage. The buyer agent ends up being the most influential force in the vendor selection process, and thus brokerages are better positioned than iBuyers or listing platforms to drive attach of in-house title and mortgage. While winning a brokerage side earns Compass $5,000 of net revenue, attaching title to that transaction will earn Compass $2,800 more, and attaching mortgage to that transaction will earn Compass an incremental $20,000.

Compass’ technology lead will also help them achieve above-industry attach rates. A digital experience that speeds up closing, eliminates offline interactions with multiple counterparties, and minimizes duplicative information upload is valuable to homebuyers and agents alike. Even Realogy’s CEO admits digital adoption drove improvements in their title and mortgage attach, and Compass with its single integrated dashboard provides an even better closing experience.

“We've invested in '18 and '19 in effectively, digital-only tools to make that customer experience better. So you can close on your house remotely on the title side, a digital-only mortgage product at our partner guaranteed rate and us using our Guaranteed Rate Affinity joint venture. And many of these things got accelerated by COVID and everybody being more accepting of digital. But what really happened is once people go through it once, an agent realizes, "Wow, that's a better experience than something before, what a competitive title company offers. I want to use that." Or a customer is like, "Wow, that mortgage experience is great. And now it's time to refi. I want that experience again." And so we've seen a real acceleration in the digital products we use to drive title and mortgage.”

– Ryan Schneider, Realogy CEO (8/11/2021)

As a base case, I assume Compass by 2025 achieves 35% title attach rate, in line with Realogy today, and 15% mortgage attach rate, reflecting greater homebuyer sophistication in the high-end market. These assumptions result in $290mm of economic EBITDA (accounts for 50% ownership in mortgage JV), compared to $740mm of EBITDA I expect the brokerage business to generate in 2025.

Valuation

Compass is the cheapest stock among both tech and non-tech peers in real estate. Despite taking steady market share from Realogy since 2018, Compass trades at a lower gross profit multiple than Realogy.

Realogy trades at 6x EBITDA, having been ex-growth since 2017. RE/MAX is a 5% grower and trades at 10x EBITDA. Using a 10x EBITDA exit multiple, Compass is a $22 stock in 3 years – $16 of value from core brokerage and $6 from ancillary services. From $8 today, that is pretty good.

Catalyst

Improved disclosure on unit economics

Re-rating driven by margin improvement