Behind Zillow's Shiny Facade
It has been a little more than a month since the last issue. I spent most of November traveling - part work, part leisure. But now I am back and I have planned more frequent updates in the coming months. Keep your fingers crossed.
Sadly, the great Charlie Munger passed away. His wisdom and witticisms will be missed. Luckily, we have a large body of his wisdom conserved in the form of books, videos, and podcasts. Munger was fortunate to live in a time like this, and we too as his audience. Think about the countless great people of whom we have no record left - either because none were created or because they were destroyed.
Student of Value is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
His life advice:
It's so simple to spend less than you earn, avoid toxic people, toxic activities, keep learning all your life, defer gratification because you prefer it that way, and if you do it all this way you will succeed. If not, you will need a lot of unusual luck.
Plus, how he led by example and stayed an inspiration to the end.
Before we dig into today's company, I would like to express my deepest gratitude to all of you. Thank you for reading and subscribing. I would do this even if no one was reading it, but it is a lot more fun when it reaches people and elicits reactions. The more, the merrier. Besides, sharing ideas and opinions is what makes us better.
Our vision of a “housing super app” is to help customers across all their real estate needs serving as one ecosystem of connected solutions for all the tasks and services related to moving.
With mortgage rates close to 8%, a housing inventory shortage, and the biggest gap between the cost-to-buy and the cost-to-rent in the US in over 50 years, some of you might be consciously or subconsciously filtering out the whole real estate sector and its satellites. But you have to keep in mind that tough times, like these, create the conditions for future outperformance - just like bubbles carry the seeds of their own demise.
Zillow is one of those companies whose name has become synonymous with the industry. It is the most visited real estate site in the US and the 18th most visited site overall, currently. News Corp owned realtor.com is a distant second with 40% of zillow.com's traffic - 125M vs 307M monthly visits. And this is only the zillow.com domain. Zillow, the company, also owns Trulia, StreetEasy, and HotPads. They rank 4, 16, and 13 among real estate site in the US by traffic, adding another 50M monthly visits to the total. Mind you, these numbers are for October 2023. They are 2-3 times lower than what they used to be when the market was still hot.
Today more people search for "zillow" than for "real estate."
It is important to note that almost 60% of the traffic is direct to the site and another nearly 40% is via Google, where the most popular searches were: zillow, zillow rentals, zillow homes for sale, zillow.com. This is mind-share. People looking for real estate go straight to Zillow. Paid search was only 2.6%, according to the latest numbers.
For those of you not familiar with the names:
Zillow: The core platform where users can search for real estate properties. It also provides tools for homeowners to estimate the value of their properties.
Trulia: Acquired by Zillow in 2015, Trulia is another online residential real estate platform, offering similar services to Zillow.
StreetEasy: Focused on the New York City real estate market, StreetEasy provides local real estate insights, property listings, and other related information.
HotPads: Acquired by Zillow in 2012, HotPads is a rental-focused platform, helping users find apartments, houses, and other rental properties.
Next to its real estate sites and complimentary to them, Zillow is known for its Zestimate - a proprietary algorithm to estimate the market value of any home based on various data points, such as location, size, features, and recent sales data. Zestimates are widely used by homeowners, potential buyers, and real estate professionals to assess the value of homes and make informed decisions about buying, selling, or financing real estate.
At the core of Zillow is our living database of approximately 140 million U.S. homes and our differentiated content, most notably the Zestimate, our patented proprietary automated valuation model through which we provide home value estimates.
During 2022, our Zestimate had a median error rate of 2.7% for homes listed for sale and 7.6% for off-market homes. We believe our data and content has helped the Zillow brand become synonymous with residential real estate.
The Zestimate is part of the trend towards democratizing finance and real estate, making real estate valuations much more affordable and readily available. The issue of accuracy has come up and is likely to continue coming up, with some embarrassing failures (more about these later) calling into question the reliability of automated valuations. The real question is how well they are performing compared to real estate appraisers. I haven't seen a paper studying this, but if you know any, send them my way. I would love to read them.
In my simplified view, there are two pivotal points in the human vs algo real estate appraisals competition:
As of today, human appraisers, the good ones among them at least, are better able to spot emerging patterns and inflection points than state-of-the-art valuation algos. It was precisely the inability of the Zestimate to capture the slowing down of home price appreciation that led to the dismantling of Zillow Offers, the company's house flipping business. Undoubtedly, others got burnt as well. However, if the choice is between a good human appraiser and the best valuation algo, I would go with the former. Of course, at the pace these technologies are improving lately, I reserve the right to change my opinion.
The second point above, bias, favors algos. Algos are not affected by behavioral and cognitive biases in the way that people are. Volumes have been written, and forgotten by now, on the subject. It was the hottest topic at one point, culminating with Kahneman's Thinking, Fast and Slow. As any of you know, staying rational and considering issues in an even-handed way, without letting biases cloud our judgment is essential and this often gets in the way of great results. Algos have that going for them.
The Zestimate is an important part of the company's offering. Zillow is best at leveraging technology to provide consumers with transparent and easy-to-access real estate information. The company's website and mobile apps offer a wealth of data and tools that empower consumers to make informed decisions about their real estate needs. This way it builds trust with its customers and strengthens the Zillow brand, which is key to the company's success.
Now, let's take a look into how Zillow generates revenue.
Zillow generates revenue from a variety of sources, including:
Advertising: Zillow sells advertising space on its websites and mobile apps to real estate agents, brokers, lenders, and other businesses.
Subscriptions: Zillow offers subscription-based services to real estate professionals, such as Zillow Premier Agent and Zillow Closing Services. Real estate professionals pay Zillow to have their profiles and advertisements featured on the platform, allowing them to connect with potential clients.
Mortgage origination and servicing fees: Zillow earns fees when it originates and services mortgages for consumers.
Home sales: Zillow earns commissions on the sale of homes through its brokerage business.
Zillow makes most of its money selling advertising and leads to estate agents.
The company used to break down its business into 3 reportable segments:
Internet, Media & Technology (“IMT”)
Premier Agent - an advertising service that allows real estate agents and brokers to prominently promote themselves on Zillow property listings. The goal is to send qualified leads to those agents.
Rentals marketplaces - as the name suggests, this is the main marketplace(s) that Zillow offers, matching buyers and sellers.
Other IMT - the new construction marketplace and revenue from the sale of other advertising and business technology solutions for real estate professionals
mortgage originations through Zillow Home Loans
advertising sold to mortgage lenders
title and escrow services performed by Zillow Closing Services
After shutting down Zillow Offers, which started in Q4 2021 and finished in Q3 2022, all segments were collapsed into one. Partly because IMT dominates, with 94% of the total, and partly to obfuscate the performance hits the segments took over the past 2 years.
The Mortgages segment generated $246M in 2021, but only $119M in 2022, as loan originations decreased 62% due to risking interest rates, which put a damper on originations and refinancing alike. Premier Agent was also down 7% - from $1.4B to $1.3B. The addition of ShowingTime bumped Other IMT revenue from $226M to $274M. Gross profit dropped 12% to $1.6B.
In mid 2014, Zillow announced that it plans to acquire competitor Trulia - the second largest US real estate listing site. The two were quite complementary, with similar business models driven by advertising and making a marketplace. It was a $3.5B stock transaction, in which Zillow shareholders kept 2/3 of the company and Trulia - 1/3. This makes sense as Trulia was about 1/2 to 2/3 the size of Zillow at the time. In the quarter just before the acquisition, Zillow's expected EBITDA was $14-15M while Trulia's was $7-10M. The same proportions were valid on unique users basis as well - 54M vs 83M.
The premium paid was 25% above the closing price on July 25, 2014. Shares in Zillow rose 1 per cent to $160.32, while Trulia’s stock jumped 15 per cent to $65.04 by close of trading in New York.
More recently, in February 2021, Zillow announced the acquisition of ShowingTime, an industry-leading real estate showing software, in a $512M cash transaction. This seems like a steep price for a company that provides online tour scheduling and offer management tools. Management claims that it solves a major pain point in the house touring process. I see that it resulted in $389M goodwill. The lack of major impairments gives us some comfort that the things are going according to expectations.
The more exiting growth path that Zillow explored was iBuying. Based on the premise that the company could purchase properties straight from the sellers, leveraging the power of the Zestimate to never overpay, fix them up a little, then turn around and sell them to the troves of buyers. It sounded good on paper. The conditions were there:
tons of data gathered over the decades
cutting edge algos to turn that into insights
low interest rates
At the time, nobody seemed to mind that they were replacing the super asset-light ad model with the capital-intensive and much more volatile house flipping model. A lot could be said about the assumption that past results are predictive of future performance - both in the sense of what their models were predicting the fair price and time to sell to be and in the sense of low interest rates persisting. In any case, Zillow Offers didn't work out as planned.
In the fourth quarter of 2021, the Board of Directors of Zillow Group made the determination to wind down Zillow Offers, our iBuying business which purchased and sold homes directly in markets across the United States. The wind down was completed in the third quarter of 2022 and resulted in approximately a 25% reduction of Zillow Group’s workforce.
The most recent development, which could be a net positive for Zillow, was the federal court verdict that found the National Association of Realtors and large residential brokerages guilty of conspiring to keep home sale commissions artificially high. The damages are estimated at around $1.8B.
What led to this was the democratization of real estate that I mentioned earlier. As online marketplaces, like Zillow, became the main meeting point of real estate buyers and sellers, the commission structure remained in the past century. Buyer and seller agents took 3% each, and consumers had no say in it, even though now they had the tools to discover and price homes on their own. Powered by Zillow.
This development may, just may, divert some of those cash flows towards Zillow. Ben Thompson has a great post about it. He is not a fan of Zillow and considers it just a marketing tool for real estate, which it is. I just happen to think it is a great tool. Anyway, I recommend reading the whole thing. Here is just the conclusion.
Before this decision, the real estate business was a franchise; going forward, assuming this decision leads to the sort of changes I expect, it is going to be a business. That’s good for consumers, and it’s good for Zillow, which, I will add, is not a surprise: remember that Aggregators derive their power from marshaling demand because they offer a product that consumers want. It follows that what is good for the Aggregator is good for consumers, and vice-versa; it’s the folks in the middle that get squeezed, whether that be publishers, TV studios, or, in this case, realtors.
Stock based compensation (SBC) is a serious problem that many investors have been decrying constantly to no avail. Zillow is a massive offender here. SBC has ballooned to nearly half a billion on $2B sales. The issue even came up on the latest earnings call. Management was extremely cavalier about it, passing it for investment in growth (not an expense) that leverages with growth. Yeah, right.
Here is then-CEO Spencer Rascoff on the topic:
And I just -- I think looking at SBC as a percent of revenue is an odd way to look at it. Looking at SBC on a dollar basis is an odd way to look at it because the stock that you grant to your employees at the beginning of the year or whatever -- at annual review season, that could end up costing the company a lot if the stock does one thing and a little of the stock does another. And so the dollar value of SBC, I think, is a weird accounting -- kind of accounting quirk.
- Q4 2016 Earnings Call
He is basically saying that SBC is Monopoly money and there is nothing to see here. Move along, people.
The chart below shows the performance of management compared to that of shareholders as measured by the cumulative change in SBC and stock price since 2016.
Shareholders made a meagre 56% profit, and that's only if they were not unfortunate enough to enter at the 2021 peak, in which case they will be deep, deep under water - nearly 80% deep. Meanwhile, management made out like bandits, compounding their wealth at over 20% per year. And unlike shareholders, they get to adjust the cost of their shares. The exercise price of stock awards was reduced to $38.78 on August 3, 2022. Moreover, the vesting period is only 2 years.
On August 3, 2022, upon the recommendation of the Compensation Committee, the Board approved adjustments to the exercise price of certain outstanding vested and unvested option awards for eligible employees. The exercise price of eligible option awards was reduced to $38.78, which was the closing market price of our Class C capital stock on August 8, 2022.
(...) The weighted-average total fair value of options repriced in August 2022 was $67.58.
SBC is mentioned 68 times in the 2022 annual report. It's a major driver of the company's profitability or rather the lack thereof. And it's true. Profit is mentioned only 28 times.
As you can see from the chart, if these funds went to shareholders, Zillow would be an amazing investment.
To management's credit, they did start repurchasing shares when the stock price sank down to $40. But it is nothing compared to the issuance that preceded it. 46% dilution between 2016 and 2021, compared to a 9% decrease since. And it looks like the trend is starting to revert.
Besides, there are currently around 30M shares under option awards, half of which have already vested, with an average strike price of $45. By the looks of it, these will be exercised and past repurchases will be undone. There are also 14M RSUs and on top of this, there are another over 30M shares issuable upon conversion of the convertible notes, maturing between now and 2026.
As of December 31, 2022, total unrecognized compensation costs were:
$409M related to unvested option awards (~2.5 years vesting period)
$470M related to restricted stock units (~2.5 years vesting period)
The potential dilution, if all those instruments are exercised, is no less than a third of the total number of shares outstanding.
This describes a sad dynamic that has become widespread among companies labelled "growth," and it makes them uninvestable, no matter how great the underlying business is.
No matter how much I like the business, it is not a good investment. Management failed once with the house flipping business. Then, they failed a second and a third time when they kept granting themselves options and repricing them on top of that to guarantee themselves a 75% upside (based on the fair value at the time of repricing).
When I started looking into Zillow, I knew it was a leader in real estate listings in the US and the stock had taken a beating. This combination is what drove me look deeper and see whether this is a great business at a fair price. I knew about the failed house flipping business and the broken growth story, but these were necessary for the price to come down. If it were not for the SBC excess, I would be happy to own the company.
In loving memory of Charlie Munger, I will end with another favorite quote of his.
When it comes to compensation consultants, prostitution would be a step up.
Student of Value does not offer compensation advisory services. Subscribe now.