Compass’ Cash Crisis Closes

After 15 months of cost cutting, Compass is free cash flow positive, making more money than it spent in Q2 2023.

Why it matters: With a high cost base and dwindling cash reserves, Compass was forced to cut operating expenses as it pivoted to a profitable, sustainable operation – which it has done.

Dig deeper: I first wrote about Compass' cash burn problem in May 2022, and it's been a busy 15 months.

• Over the past year, the company has cut expenses by about 35 percent – or $500 million – through three rounds of layoffs.
   

• Compass’ annual operating expenses have dropped from $1.45 billion to $950 million, with a goal of getting down to $900 million by the end of the year.

A declining market is an especially challenging time to achieve profitability.

• As of Q2 2023, Compass’ trailing 12 month revenue was $5 billion – down from $6.7 billion a year ago.
   

• Which means cost control is the only realistic option available to get cash flow positive.

Compass’ cash balance has stabilized at $335 million – and, in fact, has increased for the first time in years.

• Management clearly has confidence in the business: in July, it repaid the outstanding $150 million draw from its revolving credit facility.

What to watch: Compass isn’t entirely out of the woods yet, but it’s on a much more solid foundation.

• This past quarter is the high-water mark for revenue; from here seasonality kicks in with progressively lower revenue for the next nine months.
   

• It’s likely that cash flow will remain relatively flat for the rest of the year – the question is whether it will limit the company’s ability to invest for future growth (M&A and organic).

The bottom line: Compass’ turnaround has been an instructive case study in managing a business through a turbulent market. 

• Like many businesses, the company was caught flat-footed last year when the market changed, but it executed a necessary turnaround to sustainability.
   

• The broader lesson is around adaptability – it matters less how you got to where you are, and more how quickly you can adapt to a rapidly changing environment.

Brokerage Profitability

A common response to my previous analysis, Agent Compensation at the Top U.S. Brokerages, is that the brokerages paying the most out to agents couldn’t be profitable or sustainable – but, perhaps counterintuitively, the evidence suggests otherwise.

Why it matters: In a shifting market, the low-fee brokerage models are structurally designed to thrive, and are operating much more profitably than legacy brokerages.

• To recap, the low-fee models are paying out a significantly higher percentage of their revenue to agents than legacy brokerages. 

Those same low-fee brokerages are also profitable or closest to profitability: eXp Realty, United, Real, and Fathom (RealtyONE declined to share this information with me).

• The only companies that were profitable in Q1 2023 were eXp Realty and United Real Estate, and the largest legacy brokerages were really unprofitable.
   

• Note: This analysis uses Adjusted EBITDA (think of it as Adjusted Earnings) as the metric of profitability – it allows a company to portray its earnings in the best possible light by backing out expenses like stock-based compensation, one-off legal or restructuring charges, and other non-cash expenses.

To account for variations in brokerage scale, we can look at Adjusted EBITDA per transaction, which yields similarly revealing results.

• It’s worth directly comparing the two fastest-growing models of the past five years, Compass and eXp: in Q1 2023, Compass lost $1,900 per transaction, while eXp generated a profit of $130 per transaction – quite the difference.
   

• The outlier is Douglas Elliman, which has a much smaller transaction volume (4,600 in Q1 '23, compared to 36,000 at Compass), so its loss per transaction is much higher than its peers.

The next most common response to this analysis asserts that the low-fee models don’t provide as much support to their agents. 

• Therefore, while agents are able to “make more money,” they’re on their own and, without brokerage support, are less productive.
   

• Once again, the evidence suggests otherwise.

Across the nine brokerages in this analysis, the average production was one transaction per agent in Q1 2023.

• The average for the legacy brokerages (Compass, Anywhere, Keller Williams, and Douglas Elliman) was 1.03 transactions per agent, while the average for the low-fee brokerages (eXp, Real, RealtyONE, United, and Fathom) was 0.98 transactions per agent – effectively the same.
   

• In aggregate, agents at low-fee brokerages, with “less support,” were just as productive as agents at the legacy brokerages with “lots of support.”

Variability in the number of transactions per agent over time provides further evidence: between Q4 2022 and Q1 2023, the average number of transactions per agent dropped 10 percent across the same nine brokerages.

• Four low-fee brokerages (eXp, Real, Fathom, and RealtyONE) were at or below that average – meaning that their agents saw less of a decline in transaction volume than agents at legacy brokerages.
   

• Support or not, agents at low-fee brokerages were more resilient and saw less variability in production during a changing market.

Yes, but: These are averages, and as with all averages, there will be overs, unders, and outliers.

• Not all low-fee brokerages and legacy brokerages perform similarly.
   

• Furthermore, inside of each organization, there is significant variability in individual agent performance and compensation. 

The bottom line: The dual hypotheses that low-fee brokerages aren’t sustainable, and that their agents are less productive due to less support, are false.

• Low-fee brokerages are in fact more profitable than the legacy brokerages, even after paying out a significantly higher proportion of their revenue out to agents.
   

• This is classic Innovator’s Dilemma: while the legacy brokerages are racing to cut costs, the low-fee models – built from the ground up with a lower cost operating model – are taking market share and competing where they can win.

Agent Compensation at the Top U.S. Brokerages

Agent compensation structures at U.S. real estate brokerages vary, with some firms paying out a significantly higher percentage of their revenue to agents than others. 

Why it matters: Commission and fee structure is foundational when it comes to agent loyalty and recruitment; as agents seek to maximize their earnings, they're naturally drawn to brokerages offering higher splits and lower fees.

Dig deeper: Brokerages generate revenue from the commission on the sale of a house. 

• The brokerage and agent then effectively split the commission, through a varied combination of commission splits, fees, and revenue sharing.
   

• Once added up, the result is a percentage of total brokerage revenue that is paid out to agents – ranging from 77 to 96 percent in this analysis.

The low-fee / high-split brokerages like eXp Realty, Real, RealtyONE, Fathom, and United end up paying significantly more of the commission out to agents.

• Compass, which famously offered agents high commission splits as a recruitment incentive, is, on aggregate, closer to traditional industry stalwart Anywhere (home of Coldwell Banker and Sotheby's) than the others.

Compass is paying its agents less over time, through a combination of reducing existing agent splits, recruiting new agents with lower splits, and higher fees.

• The one percent reduction in revenue paid out to agents over the last two years represents about $9.5 million retained by Compass and not paid out to agents in Q1 2023.

Compass highlights this achievement, which it is clearly proud of, in its quarterly earnings, declaring that its “commissions expense as a percentage of revenue” is “improving.” 

• In fact, Compass is so pleased with this achievement that it is the first data point in its earnings press release and the second financial highlight, behind total revenue, in its investor presentation.
   

• Enjoy it while you can; after this article, I doubt Compass will be highlighting these numbers so strongly in the future – and it certainly won’t kick off its agent gatherings with the same metrics.

Agents like to make money, so it’s unsurprising that as natural entrepreneurs they flock to brokerages where they can maximize their earning potential.

• It’s no coincidence that the top 5 brokerages that grew agent count over the past quarter are the exact same brokerages that paid the most out to agents: eXp, Real, Fathom, RealtyONE, and United.

The bottom line: It’s a simple equation: More money attracts more agents, and more agents sell more houses.

• Brokerages paying a smaller proportion of revenue to agents is an effective strategy to improve short-term financials, but the long-term implications around agent recruitment and retention are significant.
   

• In a period of fewer transactions and overall belt-tightening, it's more important than ever for agents to carefully consider commission splits when deciding where to work.

Agent Migratory Patterns

The first quarter of 2023 saw a variety of winners and losers in terms of brokerage agent count – the numerical manifestation of agent recruitment and retention.

Why it matters: As I’ve previously asserted, agents are central to the real estate transaction; growth in transaction volumes goes hand in hand with growth in agent count, because agents sell houses.

Dig deeper: Disruptive, low-fee brokerages like eXp Realty, Real Brokerage, United Real Estate, and RealtyOne all added agents during the first three months of the year.

• Meanwhile, some of the largest incumbent brokerages and franchises – Keller Williams, RE/MAX, and Anywhere – all lost agents.

The five largest brokerages and franchises in this list – Anywhere, Keller Williams, RE/MAX, Compass, and eXp Realty – account for over 427,000 agents, down 1.1 percent from the previous quarter.

• Among them, only eXp Realty grew its agent count from the last quarter.
   

• Redfin, with less than 2,000 salaried principle agents, is the downside outlier after three rounds of layoffs.

Real Brokerage is the clear upside outlier, growing its agent count over 20 percent – to 10,000 agents – from the previous quarter.

• 10,000 agents is considerably fewer than Anywhere’s 58,000 or RE/MAX’s 82,000, but Real is catching up to Compass’ 28,000 agents. 

Speaking of Compass, Q1 2023 was the first time that the brokerage’s agent count decreased.

• Context is important: many brokerages lost agents!
   

• But Compass is no longer in the ranks of “fast growing brokerages,” a category it dominated over the past three years – that mantle now passes to eXp Realty and Real (and several other low-fee brokerage models).

What to watch: Agent migration patterns are a significant leading indicator of future brokerage growth.

• In a period of belt-tightening and fewer transactions, agents are moving away from traditional brands and flocking to relatively newer models where they're able to keep more of their commission.

The bottom line: Agents – and not AI, machine learning, a sophisticated CRM, a one-click transaction, nor any other tech buzzwords – sell houses.

• This period of market scarcity reveals the brokerage business models able to thrive in a downturn, as well as those facing more fundamental challenges.
   

• To identify the brokerage business models of the future, one simply needs to follow the agents.

Incentive Splits and Agent Retention

Brokerages are employing a variety of incentives in the ever-expanding battle to win over the hearts and minds of real estate agents, including an innovative golden handcuff known as “Incentive Splits.”

Why it matters: Incentive Splits create an ever-increasing bill that an agent must repay if they leave a brokerage – which can effectively lock an agent into staying even when they want to leave.

• This analysis focuses on Compass, where there is the most evidence, but the company is not alone in employing the practice.

Incentive Splits, a term included in some Compass agent agreements, are bonus commissions that range from 2–10 percent.

• These commissions accumulate over time, creating an ever-growing balance that an agent must repay if they want to leave Compass.

The contract language states that if an agent leaves within two years of receiving any incentive, including an Incentive Split from a home transaction, the agent must repay all Incentives back to Compass.

• In reviewing several contracts, this timeframe has ranged from one to three years.

The bill can add up quickly, especially for top performing agents that wish to leave: the invoice below (which may not be representative of the average Compass agent) includes over $72,000 in Incentive Splits, payable back to Compass.

Yes, but: It’s not clear how widespread this practice is – I’ve seen Incentive Splits included in the contracts of around 20 current and former Compass agents.

• A Compass spokesperson told me that since the company stopped offering cash incentives to agents in 2022, this only impacts a small group of agents.
   

• Compass is not alone: the invoice below from Corcoran includes a $43k bill for “Split Overpayments,” which is the same as an Incentive Split.

The incentive to agents is clear – extra cash in their pockets – but the mechanics are designed as golden handcuffs: financial incentives given to employees and contractors to discourage them from leaving a company.

• But unlike other golden handcuffs, Incentive Splits don’t expire or vest after a certain amount of time. 

The bottom line: Incentive Splits appear to be an inventive mechanism to incentivize agents to stay with their brokerage – by creating an ever-growing bill that must be repaid if they want to leave.

• Putting aside effectiveness and intention, the key takeaway for agents is clear: read the fine print.
   

• Perhaps at the “brokerage of the future,” agents will stay because they want to, not because they have to.

Market Shift Highlights Brokerage Fundamentals

2022 was a tumultuous year: the real estate market turned, transactions dropped, and the largest U.S. brokerages were all forced to chart new courses.

Why it matters: Agent count, transaction volumes, and the relative growth and decline of brokerages in 2022 are all powerful predictors of future performance.

Agent count continues to fuel the business of the major U.S. real estate brokerages.

• eXp Realty is the clear stand-out, with an exponential increase in the number of its U.S. agents – which now eclipses industry incumbents Anywhere and HomeServices of America (HSoA).
   

• Overall growth slowed across the board in 2022, and Anywhere has seen a recent uptick in agent count, which has yet to translate into a corresponding uptick in transaction count.

Transaction volumes and brokerage performance are directly correlated to agent count (MDP: “Technology doesn’t sell houses, agents do.”)

• Among its peers, eXp Realty once again outperformed the market, growing its transactions by 12 percent in 2022, compared to an overall market decline of 18 percent.
   

• eXp’s large brokerage peers all transacted less in 2022 than the year before, while Compass, also buoyed by recent agent count gains, experienced less of a decline than the traditional incumbents (and Redfin).

The declining market has forced companies across the industry, brokerages included, to cut costs. 

• But those cuts are unequal among peers – cash flow and profitability dictate the speed and severity of cuts. 
   

• Throughout the year, Anywhere and Compass, which have the highest operating expenses, have been forced to cut the deepest, while eXp and Douglas Elliman have increased their operating expenses (from a smaller base).

What to watch: Agent count growth and transaction volumes will remain sluggish in 2023.

• A further slowing of the market will force some companies to cut even deeper.
   

• It’s a numbers game: less transactions will go to fewer agents – so brokerages with strong agent count growth will outperform the market (just look at eXp).

The bottom line: The market correction is a great equalizer – and it’s accentuating the differences between brokerage business models.

• In a market of musical chairs with fewer chairs available, companies with sustainable models, lower overheads, and strong recruiting are more likely to succeed.

Cash Burn Continues as Compass Navigates to Breakeven

The results are in and Compass burned $143 million in cash during Q4 2022, leaving the company with a cash balance of $362 million.

Why it matters: With a clearly articulated goal of reaching breakeven after billions in losses, 2023 is Compass’ seminal make-or-break year. 

• Despite another quarter of high cash burn, the company appears to be positioned to achieve breakeven in 2023 after massive cuts made over the previous 12 months.
   

• Even in a much softer market, Compass’ Q4 2022 cash burn is less than a year ago, a reflection of the cuts already made to the business.

Compass started 2023 with a cash balance of $362 million – which includes a $150 million drawdown from its revolving credit facility

• Without that additional loan, Compass’ cash balance would have been $212 million – dangerously low for a company that just burned $143 million in a quarter.

The key to breakeven is Compass’ ability to reduce its non-GAAP operating expenses (OpEx), primarily achieved through layoffs.

• OpEx has dropped significantly over the past six months, and is on track to drop further through the rest of the year.
   

• It appears that Compass is aiming for OpEx of around $950 million for 2023, down about 30 percent from $1.35 billion in 2022.

It’s fair to say that any brokerage, Compass included, falls apart if it’s unable to recruit or retain agents.

• For the first time, Compass’ agent count has gone flat, reflecting the challenging market and environment for agents (less transactions = less agents).
   

• Yes, but: It takes several data points to create a trend, and most brokerages are seeing a similar slowdown in agent count growth.

Analysis time: Assuming Compass’ national market share and revenue per transaction remain consistent with 2022, we can see what needs to be believed for the company to reach breakeven in 2023.

• Compass’ current non-GAAP operating expense target of $950 million suggests that 4.5 million transactions in the national market are necessary to give Compass the revenue and gross profit necessary to break even for the year.
   

• A bear case of 4 million transactions would require Compass to cut an additional $100 million of operating expenses to reach breakeven.
   

• Please note: these are back-of-the-envelope calculations that provide directionality, not certainty.

What to watch: Like other brokerages, Compass only has so many levers to pull to reach breakeven.

• On the revenue side, the company is considering franchising as a less-expensive growth strategy, and will be trying to accelerate its mortgage joint venture.
   

• If the market remains soft and revenue drops, the only option is to cut even more costs out of the business.

The bottom line: Compass is not alone in needing to cut costs during a significant market downturn – its future depends on it.

• The company dipped into its “emergency reserve” last quarter – $150 million from its revolving credit facility – for the home stretch to profitability.
   

• It appears that Compass has made the deep cuts necessary to achieve breakeven this year, as long as nothing else unexpectedly goes wrong.

The Race to Cut Costs

Across the real estate industry, companies are racing to cut costs in the face of a significant market slowdown. 

Why it matters: With dropping revenues, cost control is one of the only levers in a company’s control – and is the key to a sustainable, profitable business.

• The need to cut costs – and the depth of those cuts – are a function of a company’s overall financial health and business model efficiency.

Some companies, like eXp, have the advantage of a more efficient business model with lower operating expenses (OpEx).

• Compared to its peers, eXp is servicing a disproportionately high number of transactions with relatively modest operating expenses.
   

• The more traditional industry behemoths, Anywhere and Compass, have a less efficient model with a much higher cost basis (and thus need to cut faster and deeper).

Dig deeper: Another measure of business model efficiency is the amount of revenue generated per $1 spent in operating expenses.

• Based on this metric, eXp was about three times more efficient in Q3 2022 than its publicly-listed brokerage peers (who are all in the $3–4 range).

Compass has been racing to cut its operating expenses as quickly as possible (it also recently announced a third round of layoffs).

• Compass is driving to cut its non-GAAP operating expenses by 40 percent, or around $600 million annually.

Layoffs are the most visible way that real estate tech companies are cutting costs.

• Since June of 2022, Compass has shed around 1,700 employees (40 percent), while Redfin has also cut deep with 2,000 fewer employees (26 percent).
   

• Many other real estate tech companies have also enacted significant layoffs to cut costs (and in some cases, in order to survive).

What to watch: Among the big brokerages, Anywhere, Compass, and Redfin have already made significant cost reductions, while eXp and Douglas Elliman are under less pressure to cut costs.

• Cost reductions limit a company’s ability to invest in future growth opportunities (ex: Compass has "paused all expansion into new markets” and Anywhere shut down its cash buying program).

The bottom line: The market downturn is forcing all real estate tech companies to cut their expenses in order to achieve, or maintain, profitability.

• Unprofitable companies with high cash burn and high fixed costs have no choice but to cut, and cut deep, to survive.
   

• While other companies operating more efficient, low-cost operating models are under less pressure to make big cuts – and may be better placed to invest in future growth.

Further Cuts on Compass' Path to Profitability

Compass continued to burn cash in the latest quarter, but also demonstrated a reduction in operating expenses as the company aims for breakeven in 2023.

Why it matters: With a historically high cash burn in a cooling real estate market, Compass needs to cut costs extensively to achieve profitability.

• Furthermore, the company signaled further cuts to bring expenses in line with revenue in 2023.

Behind the numbers: Compass burned $76 million in Q3 2022, continuing a run of cash flow negative quarters.

• Of that $76 million, there were one-time restructuring and litigation expenses of $40 million.

Compass ended the quarter with $355 million in cash.

• Historically, Q4 and Q1 is where cash burn is highest, due to the seasonal decline in revenue.
  
  

• Coupled with a cooling market, the next six months are going to be tough.

What they're saying: On its earnings call, Compass signaled that further cuts are coming:

• ”…we will be implementing additional cost reduction initiatives to get ahead of any future market declines.”
  
  

• ”We are committed to driving our non-GAAP operating expenses well below the low end of our range of $1.05 billion in 2023.”

Compass' initial cost reduction program was an effort to get annual operating expenses to a run rate of $1.05–$1.15 billion (down from $1.48 billion).

• In total, the proposed cuts would represent an overall reduction in operating expenses of around 33 percent -- a very significant change.

Dig deeper: If breakeven in 2023 is the goal, working backwards reveals the various revenue estimates needed to achieve that goal.

• Compass' first set of cuts suggested a five percent revenue decline in 2023, but the latest cuts suggest a deeper 14 percent drop in revenue.

The bottom line: All brokerages, Compass included, are entering the literal and figurative winter of real estate; the next two quarters are going to be tough.

• With a historically high cash burn, Compass needs to cut faster and deeper than most other brokerages -- and it is.
  
  

• The question for all brokerages making cuts remains the same: can it be done while still remaining attractive to, and providing the same value to, agents.