Highlights from the DOJ Antitrust Filing Against RealPage

And the devil's advocate list of rebuttals

It’s not every day that you get to read a 115-page court filing against one of the largest real estate software companies in our industry. The DOJ seemingly left no stone uncovered, going as far back as 2017 to support their stances. So what exactly did the DOJ argue? And how will RealPage respond?

I read interesting summaries, as you probably have, but they were all high-level. That’s why I went to the source. Below is a summary of the key points from the DOJ filing by alleged violation. I highlight where RealPage will likely rebut, because who doesn’t enjoy playing devil’s advocate? This girl 🙋🏻‍♀️ LOVES it. I’m all for a good debate. Opinions below are expressed as if I’m speaking on behalf of RealPage’s revenue management software. I’m not saying that I agree with all of them, but I did enjoy the thought experiment!

First Claim for Relief: Violation of Section 1 of the Sherman Act by Unlawfully Sharing Information for Use in Competitors’ Pricing

In layman’s terms, the DOJ argues that RealPage shared non-public data to enable price-fixing of multifamily rents. We’ll cover more on the price-fixing in the second claim. For now, let’s focus on the sharing of non-public data.

RealPage, by the nature of their services, has access to non-public lease transaction data. Notably, this include renewal prices. Here’s what one section of the lawsuit argued:

31. The sharing of nonpublic, competitively sensitive data with RealPage, and its use in AIRM and YieldStar, is anticompetitive. It harms or is likely to harm the competitive process and results, or is likely to result, in harm to renters and prospective renters in at least the relevant antitrust markets identified in this complaint.

The suit goes on to state that RealPage even acknowledges that it has to avoid potential pricing collusion…

130. Performance Analytics with Benchmarking “tracks transactional information therefore due [to] the potential pricing collusion, it’s anonymize[d] by RealPage.”

And then in contracts with landlords, RealPage ensures that both parties agree that RealPage can use that data to recommend prices.

23. Each landlord using RealPage’s OneSite, Business Intelligence, and Performance Analytics with Benchmarking products agrees to share its proprietary data with RealPage and agrees that RealPage’s revenue management software can use the data to generate pricing recommendations.

Step back for a moment with me. It’s standard practice in every software agreement that the software vendor can aggregate customer data. Sometimes, the agreement will spell out use cases. Often, the agreement is silent on specific use cases but will rely on the reasonableness of the vendor on how to use aggregated data.

Take Uber.

Pretend we live in a world where aggregated data and automatically set prices were illegal. Riders would have to haggle with each driver. Yes, Uber could still solve the problem of hailing drivers from anywhere. But would drivers be as willing to drive, knowing they’d have to do extra work to set their own prices? Would riders want to haggle each time? Or would they prefer the set rates that taxi drivers (usually) are required to offer? Uber probably wouldn’t exist without the opportunity to aggregate data and automatically set prices.

The same argument could be made for the airline industry, hospitality, or even Amazon with their dynamic pricing, which is also now in court. RealPage could argue that a court ruling on aggregated data would have far-reaching implications beyond conventional multifamily real estate. Will jurors want to get rid of Prime Day deals? Or Uber? AirBnB? I could go on, but I think you get the point.

Real estate is one of the few industries where local, individual pieces of data are at times more valuable than aggregated data. Managers and owners know this and may share information with each other. Here’s where the suit addresses it, specifically talking about LRO (which RealPage will sunset this year). Emphasis is mine.

48. Instead, each week, LRO users manually input competitor information into the system that they have obtained from public websites or more questionable means, such as communicating directly with their competitors.

If I was RealPage, I’d be like, “C’mon, really? You expect our staff not to stay in touch with their colleagues, even if they go to competitors?” It wouldn’t be the strongest argument in court, but this is a weak statement nonetheless. Frankly, it’s common practice for individuals to share best practices, lessons learned, etc. at conferences even outside of real estate. Of course they will share some of these insights with competitors at these events.

Plus, if you’re really going to be a stickler on direct communication among competitors, you’d be better off honing in on hospitality or student housing. There, talk amongst competitors is common practice. Student housing also has College House, which is a software vendor who’s entire value prop is that they collect your company’s student housing data. In exchange and for a price, you get aggregated student housing data. (Now, I agree, the argument that “everybody’s doing it” is not valid. That said, I do think that courts have to be careful and thoughtful about setting precedent.) This shows up the suit many times. Here’s another excerpt.

20. Landlords agree to provide this information for use by their competitors because they understand they will be able to leverage the sensitive information of their rivals in turn.

RealPage’s main argument should be this. They are a market maker. Their role in the market is to find the price at which renters agree to pay rent and landlords are willing to rent their units. Here’s what the suit said against this.

134. Revenue Protection Mode… Rather than lowering the price to stimulate demand, the algorithm reduces the target number of leases. AIRM and YieldStar then maximizes revenue for the reduced occupancy level, which tends to reduce price decreases or increase rental prices.

Of course RealPage focuses on increasing rental prices. They provide revenue management software. That’s like the literal definition of the service they provide. By the same margin though, they will decrease prices when the market (e.g. renters) requires it. While their main goal is to benefit owners because again, optimizing revenue is literally the goal of revenue management, they can only do so much. If demand declines, so will prices. RealPage will likely show many examples of this in their rebuttal.

146. In addition to discouraging discounts, RealPage discourages negotiating prices with renters. RealPage trains landlords that “YieldStar [or AIRM] is managing your Price,” so the landlord’s staff can focus on other things.

As RealPage, I’ll leave you with this. If jurors decree that managing prices are not allowable with AI, what kind of precendent will they set? What other actions will not be automatable with AI? Will this eliminate the ability of AI agents to make decisions for us, even simple ones that would unequivocably make our lives easier? Will the recent funding in AI have been for naught?! Venture firms beware…

Second Claim for Relief: Violation of Section 1 of the Sherman Act Through Vertical Agreements with Landlords to Align Pricing

The DOJ’s suit is rife with examples of landlords agreeing to set prices together. Here are several.

7. Sometimes RealPage is even more direct, acknowledging that its software is aimed at “driving every possible opportunity to increase price” or observing that among landlords, “there is a greater good in everybody succeeding versus essentially trying to compete against one another in a way that actually keeps the entire industry down.”

33. An executive of one landlord (which itself uses YieldStar and AIRM) said in a 2021 earnings call that more sophisticated, “high-quality competition” was better for that landlord when “they all use revenue management. They are all smart. They raised rents when they should.”

35. Another noted that without YieldStar “you’ll be pricing your renewals in the dark without insight into actual lease transaction data that YS uses to help you make pricing decisions. This is critical to price renewals right[,] especially in a downturn.”

81. One RealPage employee explained to his colleagues, reflecting on his former time working at a landlord, that these weekly inquiries “required cooperation among the comp[etitor]s but wasn’t hard to get that.” In June 2023, a senior director at Landlord 3 admitted that “this practice has been prevalent in our industry for a long time.”

111. As RealPage frequently trumpets to landlords, “a rising tide raises all ships.” AIRM and YieldStar ensure that the ‘tide’ flows primarily one way—higher rental prices. In a hot market, AIRM and YieldStar will recommend price increases to test what the market will bear, while in a down market AIRM and YieldStar will, to the extent possible, still increase or hold prices and minimize price decreases to reach the target occupancy rate.

The DOJ would argue that RealPage prevented the negative outcomes in the prisoner’s dilemma, thereby only incentivizing positive, cooperative behavior.

Their evidence is pretty damning. Look at how they describe RealPage’s auto acceptance of rent prices feature.

6. A significant number of landlords then effectively agree to outsource their pricing function to RealPage with auto acceptance or other settings such that RealPage as a middleman, and not the free market, determines the price that a renter will pay.

62. Instead, for every recommendation that she does not accept—whether overriding or keeping the previous day’s rent—the property manager must provide “specific business commentary” for diverging from the recommendation. This justification, RealPage instructs, should not be a mere preference for another price but must be based on a factor that the model cannot account for, such as local construction or renovations occurring in the building. It must be a “strong sound business minded approach.”

67. RealPage instructs its advisors on best practices: “[I]f a partner is not ready to use auto acceptance, are they ready to use revenue management?”

That begs the question, “Would RealPage have been as successful with their revenue management if they hadn’t automated prices?” Certainly not. Someone would have done it, and whoever had the most data would be the winner. Again, dynamic price setting is common practice in other industries. Some (not-so-lucky-now) company would have thought to try it in the multifamily market.

Last two comments on setting price floors and using proprietary data:

117. AIRM and YieldStar will not recommend a floor plan price that falls below the market minimum.

129. “I always liked this product [AIRM] because your algorithm uses proprietary data from other subscribers to suggest rents and term. That’s classic price fixing . . .”

For rebuttals, RealPage already responded that their pricing strategy “yields lower advertised rents” that would otherwise be suggested without landlords agreeing to share their data. RealPage will likely continue their defense with examples of how their software does recommend decreasing prices.

The biggest question here is whether collusion can happen via aggregated data. If the DOJ succeeds in their case, most software agreements will need to be reconstructed. Because again, as I mentioned, most software agreements allow for the sharing and reselling of aggregated data by the vendor.

RealPage may try the self-deprecating approach. Their system is not perfect. I’ve seen lots of data inconsistencies, even within their own reporting. Is having access to private data really that much of a boon? Does it really enable these landlords to collude more easily, especially if the data is flawed? For example, when I worked in private equity real estate, we explicitly told our operators and managers to get off of Yieldstar (RealPage’s software). We were in the business of renovating multifamily properties, and Yieldstar wasn’t good at pricing renovated vs. unrenovated units. It was also too volatile for the types of markets we invested in. That said, I can see why they wanted feedback on overruling prices; they knew that their algorithm wasn’t accounting for every scenario, and feedback would give them ideas on what parameters to add to their model.

Another highlight:

I1. Renters are entitled to the benefits of vigorous competition among landlords. In prosperous times, that competition should limit rent hikes; in harder times, competition should bring down rent, making housing more affordable.

RealPage will likely say that rental rates have less to do with the overall economy at that point in time and more to do with supply and demand for that market along with access to capital. They may argue that high prices were actually a result of the recent, historically low interest rates, pushing blame back in the government’s direction. They could argue this resulted in essentially free money and a lot of new Class A developments or redevelopments. When owners spend capital, they then need to recoup their money through higher rents.

By arguing that multifamily demand is inelastic, the DOJ essentially argues against the morality of rent increases. These arguments may work well with the jurors. Who doesn’t want to pay less rent?! RealPage will likely rebut with examples of how they recommend price decreases. And I wouldn’t be surprised if they point to other industries where jurors can relate to dynamic prices being a good thing (e.g. Prime Day, cheap hotels).

Third & Fourth Claims for Relief: Violation of Section 2 of the Sherman Act Through Monopolization (or Attempted Monopolization) of the Commercial Revenue Management Software Market

253. There is a dangerous probability that, unless restrained, RealPage will succeed in monopolizing the commercial revenue management software market in violation of Section 2 of the Sherman Act.

Ouch.

The DOJ continues their suit against RealPage by arguing that the firm has a monoply—or at least an attempted monopoly—in revenue management. Here are their arguments for the monopoly.

14. RealPage dominates the market for commercial revenue management software that landlords use to price apartments, controlling at least 80 percent of that market, according to its own estimates.

209. “Revenue Management is experiencing strong growth driven by AIRM” due to its “PMS agnostic approach” which gives RealPage the ability to aggregate data from its clients resulting in “revenue management [that] has achieved a market share of 95% of the top 50 owners and operators.”

The suit also talks about how most markets are dominated by 5 or fewer landlords. This has more to do with the (alleged) oligopoly of property management companies than with software.

75. Demand for apartments is relatively inelastic. Rising rents have disproportionately affected low-income residents: The percentage of income spent on rent for Americans without a college degree increased from 30% in 2000 to 42% in 2017… For college graduates, the percentage of income spent on rent increased from 26% to 34% from 2000 to 2017.

I’d like to see this contrasted with the rate of Americans who have a college degree in 2017 vs. in 2000. How many more college graduates became renters instead of homeowners during this time?

The plaintiff needs to show harm. That’s what helps antitrust cases like these win. RealPage will likely push back with statistics of their own talking about how the middle class became wealthier. While upper income earners have grown at a faster clip, the middle class is still much wealthier today than it was decades ago.

Below are a slew of other arguments, with probable (or at least fun to think about) rebuttals after each.

124. RealPage has even manipulated competitor mappings to increase the likelihood that AIRM or YieldStar would recommend price increases. For example, a prominent client asked why a subject property had mapped peers located more than 100 miles away, in a different metropolitan area, when there were satisfactory mapped competitors within five miles.

RealPage to DOJ: This also has to do with changing renter preferences, as renters are more likely to move cross country than they were even a decade ago. Jobs are more flexible or even remote.

166. In June 2023 a landlord emailed RealPage and asked, “who are your competitors?” A RealPage sales executive responded, “Our revenue management solution does not have any true competitors, mainly because our data is based on real lease transaction data from all kinds of third-party property management systems . . . ”

RealPage to DOJ: Every good salesperson will say that they don't have true competitors. This is not evidence.

178. The decision to move from an apartment building to a single-family home is primarily a life-stage and lifestyle choice. For example, the decision by a household to have children may spur a move to a single-family home. In many areas, relatively few children live in conventional multifamily apartments.

RealPage to DOJ: This begs the topic of lower birth rates and later in life family formation. This leads to more multifamily demand (and higher prices as higher earners stay renting longer) because people don’t have the life changes requiring them to buy a home.

3. RealPage replaces competition with coordination. It substitutes unity for rivalry. It subverts competition and the competitive process. It does so openly and directly—and American renters are left paying the price.

196. Appendix B identifies submarkets by bedroom count that are relevant markets in which the agreements between RealPage and landlords, and agreements among landlords, to share nonpublic, competitively sensitive information for use in pricing conventional multifamily rentals have harmed, or are likely to harm, competition and thus renters.

RealPage to DOJ: The "or are likely to harm" statement is interesting. Most anti-trust cases require the plaintiff to prove that harm has been done. Jurors are more likely to be compelled if so. It sounds like you don’t buy that harm has been done. Let me go do some research to support you in this theory…

17. The combined troves of nonpublic, competitively sensitive data [from RealPage] are much more granular, sensitive, timely, and comprehensive than alternatives.

RealPage to DOJ: It’s not any better than CoStar in their arena. Why aren’t you going after them?

---

For the record, RealPage being broken up would be incredible for the multifamily and CRE tech startup industry. VC firms might actually start investing again if more proptechs stand a chance. But I don’t know that courts will agree to set legal precedent. If they do, it will be fought against for a long time. I expect to be writing about this topic many times over and eventually, when it’s a Supreme Court case.

---

Jen’s Reading Corner

This week, I recommend reading the DOJ’s filing in full. If you’d rather stick to the Cliff Notes, I excerpted a few other sections that didn’t make the summary above.

54. For example, if AIRM or YieldStar sees that a large number of units will likely be available in twelve months, it will increase the price recommendation for a twelvemonth lease relative to price recommendations for leases of other terms, such as 11 months or 13 months, in order to nudge potential renters to accept those terms. Expiration management can only raise prices—AIRM does not lower a unit’s price if the lease term would fall in an underexposed period.

73. Rental housing is a necessity for many Americans, meaning that demand is inelastic—that is, changes in rent produce relatively small changes in the number of renters.

79. RealPage takes no steps to avoid assigning the same pricing advisor to properties with different owners, even if those properties compete with each other or are RealPage-mapped competitors.

In practice, how would this work? To play devil’s advocate, a business wouldn't be able to grow without economies of scale in certain areas, and customer success-related roles are one of them.

82. Landlords not only knew of these so-called “market surveys,” but expected their property managers to participate. As a manager of Landlord 3’s revenue management department explained, “we have always expected our properties to continue doing a traditional market survey[,]” which “gives us insight into the very specific handful of competitors closest to the subject property.”

Market survey companies beware…

153. RealPage plans to remove LRO, a less restrictive alternative, from the market. LRO does not inherently contain the same competitive defects as AIRM and YieldStar. Unlike AIRM and YieldStar, LRO does not require the same type and quantity of nonpublic, transactional data pulled from competitors’ property management software or obtained from contacting competing landlords. RealPage has already stopped offering LRO to new clients and plans to discontinue LRO for legacy clients by the end of 2024.

177. In March 2023, a RealPage economist estimated an “entry premium” of $800 per month to home ownership over rentals. According to a 2021 RealPage strategic planning guide, the “myth” that people were abandoning multifamily properties for single-family homes is false, stating that “rising home sales do not hurt apartment demand.”

186. Defining relevant geographic markets help courts assess the potential anticompetitive impact of the agreements challenged... In delineating a geographic market for conventional multifamily housing, the focus is inherently local. Renters are typically tied to a particular location for work, family, or other needs.

Sidebar: Is this as true as it used to be with remote work becoming the norm?

191. A former RealPage chief economist explained that because “real estate is very local . . . you typically want to take a . . . more narrow view if you can on what’s going on in any given submarket.”

---

If you like All About CRE and want to support my work:

• 💎 Become a paid subscriber

• ☕ Buy me a coffee

• 📧 Forward this to a friend and invite them to subscribe at jentindle.substack.com

---

---