The Delusion behind Opendoor’s Virtuous Cycle

Is Opendoor a Virtuous Cycle, or a Vicious One?

For those of you who don’t know, an iBuyer is a ground-breaking tech-enabled home-flipping product. The cost of using an iBuyer is incredibly high. Selling a home to an iBuyer is one of the most expensive ways to transfer title ownership of real property in the United States.

There are several reasons for this: (1) higher than usual risk (2) double holding costs (3) high costs of operations and (4) expectations of higher than usual returns.

Opendoor is probably one of the most well-known iBuyers in the market, mainly due to backing from SoftBank throughout its existence. Here are three things SoftBank got wrong on this one:

Price Fixing via Opendoor Brokerage

One of the biggest and largely misunderstood problems with Opendoor is their failed attempt to deliver savings to consumers.

In 2017, Opendoor has purchased a brokerage called Open Listings. Open Listings is a savings broker in California (and few other states) that offers consumers a 50% refund from a buyer’s agent commission as its main selling point. Sounds great, doesn’t it? Well, in theory, yes. In actuality, Open Listings decided that it cannot itself scale this promotion fast enough, so it began developing a network of brokers who do not work for Open Listings.

Open Listings Referral Network (Partner Agents) is a referral process that connects buyers with third-party real estate agents in exchange for an undisclosed referral fee.

Once Open Listings refers a customer to a Partner Agent, that agent, not Open Listings, represents the customer from the initial meeting through closing. Open Listings dictates that Partner Agent rebates 50% of their commission to receive a referral, while Open Listings takes a commission cut after the transaction is complete. This is called price-fixing.

Opendoor Brokerage is, effectively, an extension Open Listings’ consumer brokering and price-fixing policy.

According to Opendoor, “In certain of our service areas, if you use an Opendoor Partner Agent to purchase a home (an Opendoor Home or any other home listed on the market), you can save up to 1% off the purchase price of the home in the form of a commission refund at closing. The amount is subject to a minimum commission to your agent of $3,000, which means it is calculated as the lesser of either (a) 1% of the price of the property you buy, or (b) your agent’s commission minus $3,000. It may also be reduced on the basis of purchase type (e.g., short sale), loan restrictions, seller contributions, or law.”

“Agent Partners only pay a referral fee to our brokerage if they close on a transaction with a referred seller or buyer. The fee is a percentage of the agent’s commission, and averages 1% of property sale price unless otherwise noted in agent partner agreement. Opendoor Agent Partners are eligible to receive both buyer and seller referrals. Actual volume by referral type may vary over time. The Opendoor Agent Partner program is a broker-to-broker client referral partnership and is supplemental to your existing brokerage affiliation.”

The biggest difference between Open Listings and Opendoor Brokerage is the amount of a price fixed rebate (decided randomly, it seems, in either case as a nice round number) and the fact that while Open Listings sometimes represents consumers, Opendoor Brokerage never represents consumers at all — it merely enters into blanket referral agreements with a network of brokers.

Even bigger and more exposed use of such blanket agreements is a partnership between Opendoor and Redfin — an obvious case where direct competitors willingly enter into consumer allocation pact between one another.

Blanket referral agreements between brokers run as a violation of the Sherman Act because they always restrain free trade.

My objection to this practice, specifically calling out Opendoor Brokerage, is now on file with the Department of Justice and can be found here

Buying Itself Bad UX

The fact that Opendoor consistently must underprice consumers out of their hard-earned equity is a massive negative UX factor. Most consumers do not outright own homes in the United States, meaning, most homes are financed. Banks receive the same amount of the remaining mortgage sum regardless of how any given home is sold, whereas only homeowners’ net equity is lost in transaction fees paid to Opendoor.

This means that a loss of equity of 20% easily translates into a 50% to 70% loss of homeowners’ net equity. This is a massive problem to overcome since people do not typically like being taken for a ride. Small-time home flippers can avoid bad scaled exposure because these experiences are contained and not readily publicized, but this is not the case with massively built products.

Opendoor aggregates bad UX all in one place attached to their massive operations. In fact, in 2018 Yelp has removed (or rather hid the page using a deceptive “robots” “noindex” here a massive volume of negative reviews for Opendoor. A lot of reviews, such as this one, all come together to address a single unsolvable problem with Opendoor — unreasonably high loss of equity.

“Received my initial offer and the numbers DO NOT make sense, especially in a thriving market like Charlotte. Initial offer was at least 20K below market value. They pulled 3 old comps. My property is in a dense, cookie-cutter subdivision in a sellers market. There’s no shortage of extremely close, extremely recent comps. That they went back 6+ months is an indicator of deliberate misinformation that a less informed seller may fall for. Also — — Their comparison of traditional selling fees versus their fees is a joke- especially in a sellers market. The number are majorly trumped up. Their comparisons MAY resemble reality if it were a buyers market (most markets aren’t). But if it were a buyers market this company wouldn’t exist. All of that said- this business model would work for people who want out fast IF THEY OFFERED MARKET VALUE as they say they do. They don’t. So you unless you’re cool w leaving tens of thousands on the table- go traditional.”

Enough said with this single hidden Yelp review.

Reliance on Hard Cash in a Debt-driven Market

Opendoor, which operates in 21 markets, and says that it sold more than 18,000 homes last year. This number sounds like a lot, but it isn’t. There are about 5 to 6 million homes sold in the United States each year. This means that Opendoor has a market share of exactly 0.3% in its heydays: 2018–2019 housing market.

The problem here is the reliance on hard cash for operations. Opendoor is highly limited by the reliance on hard cash to make offers in what is effectively a debt-driven market. United States housing market is an asset class worth about $31 trillion in total. This number is only made possible with a mass adoption of mortgage-funded home purchases.

Opendoor cannot possibly make any difference in such a large market because it is cash-strapped, so it will always revolve around the same number of sales — there is no mega-scaling here to speak of. Of course, the only way to make up for such low volume and high costs of operations is higher fees, which is, in effect, what company’s founders and investors aim to sell to consumers as a “one stop virtuous cycle.”

In reality, the Opendoor cycle is not so much virtuous, but rather a vicious one.

Downturn Profiteering

When the housing market collapsed in 2008, many US homes went into foreclosure, and securities backing residential mortgages imploded. Due to the economic reality of the COVD-19 pandemic, another housing takeover can begin a transfer of massive amounts of wealth from homeowners to massive corporations, such as Blackstone.

In Q1 and Q2 of 2020, the Federal Reserve thus far has offered vast liquidity to the market. In the last several months, the Federal Reserve has been lowering interest rates to what is now nearly zero and buying hundreds of billions of dollars of mortgage-backed securities, but all this may not be enough in the long term.

Just like in 2008, foreclosed homes could be soon be bundled together by banks and sold to a single buyer, such as Invitation Homes that successfully convert them into rentals. In the event foreclosed homes would have to be sold individually to bidders (something that Blackstone aims to avoid at all costs by lobbying the current Trump administration) companies such as Opendoor are likely to become massive channels for such distressed properties transfers with an existing framework and a “clean premise” of buying homes from consumers at marked down prices.

As of September 2020, there have been no major bank failures, and banks have yet to begin unloading heavily discounted real estate assets to private equity firms. However, the Opendoor SPAC merger allows it to obtain massive amounts of new cash only available in public markets.

It is quite plausible that Opendoor’s recent move to be acquired by Social Capital could be aimed at the next downturn profiteering opportunity — buying and renovating single-family homes and turning them into rentals.

In my personal opinion, Opendoor has never offered anything new to consumers, but rather more of the same — fees, hidden kickbacks from “Partner Agents,” unlawful price fixing, and a loss of equity… no matter what.

Opendoor SPAC Merger by Numbers

SPAC merger is, by far, the most non-transparent pathway for a private company to go public. It is no wonder that Opendoor has decided to take this route, as one of the least transparent companies in the US real estate market.

The following links may be as close as one can get to full disclosures, but there are some interesting numbers to look at.

Opendoor offers very rough non-GAAP figures in these slides. I could not find any direct mentions of break-out between Opendoor Brokerage vs Opendoor revenue or any other auxiliary revenue, such as loan originations. Opendoor (iBuyer)and Opendoor Brokerage (Price Fixing Scheme/Referral Fee Network) are two very different business models, and they seem to just bundle all of it together. Obviously, a company going IPO route would never be able to get away with such poor financial disclosures.

What I see are pretty consistent losses with growth, meaning the higher the number of homes company buys and sells, the higher losses it incurs.

2017 EBITA ($57M loss) = 3,100 homes

2018 EBITA ($131M loss) = 7,400 homes

2019 EBITA ($218M loss) = 18,800 homes

“If you just think about Phoenix, 80,000 homes are sold annually in that market. At $11,000 per home today for Opendoor, that’s over $800 million contribution in one single market that we’re going after. That’s $22 billion of GMV.”

This statement from Opendoor CFO, Carrie Wheeler, is clearly misleading. Even in its own projections Opendoor aims to “ibuy” the maximum of 37,000 homes projected total as the best case scenario in 2023. So, how would Opendoor be able to capture even a single market in Phoenix where 80,000 homes are sold? What Opendoor really going after is changing the 4% market share to maybe 5 or 6% which is a very far number from the quoted $22 billion of GMV.

Opendoor argues that it can make up for high costs of operations with higher contribution margin from each home sale, but the fact is, such approach is almost impossible because real estate is a highly competitive market.

So, yes, she did “just cherry-pick Phoenix” and in a bad way. With claims that $11,000 contribution is going to become a “$20,000 opportunity” from each home sale the question arises — why hasn’t it been the case in 2018–2019?

This “$20,000 opportunity” means, that on top of all her holding costs, costs of operations, buyer’s broker fees, transaction costs, property taxes, etc. the CFO wants to still make $20,000 in contribution margins from each home. This is coming from a company that claims to make buying and selling homes seamless.

The only way to archive this “$20,000 margin” is to rip off the home seller to the fullest and leave them without any equity whatsoever. Opendoor will have to make offers that are not just 80% (accounting for exigent fees) of the home value, but closer to 70% aligned with offers made by typical home flippers.

This, in essence, is the problem of all iBuying — it is a very expensive way to transfer ownership and the iBuyer has no responsibility to the seller of the home. iBuyers only answer to their shareholders. Unfortunately for Opendoor, poor UX and high costs of operations do not scale into savings.

Form S-4 SPAC Merger Numbers and Disclosures

On October 5, 2020, Social Capital Hedosophia Holdings Corp. II filed an updated S-4 disclosure with Securities and Exchange Commission. This is the first time we have some real numbers and disclosures on how Opendoor operates. Looking into this filing, everything I said before seems to hold, but there are few new revelations I personally found interesting.

Opendoor Operates on a Loss, and it may be permanent.

We have a history of losses, and we may not achieve or maintain profitability in the future. We have incurred net losses on an annual basis since we were founded. We incurred net losses of $240 million and $339 million for the years ended December 31, 2018 and 2019, respectively, and a net loss of $118 million for the six months ended June 30, 2020. We had an accumulated loss of $790 million and $909 million as of December 31, 2019 and June 30, 2020, respectively. We expect to continue to make future investments in developing and expanding our business, including technology, recruitment and training, marketing, and pursuing strategic opportunities. These investments may not result in increased revenue or growth in our business…

Opendoor is under an ongoing investigation by the FTC (Federal Trade Commission.)

In August 2019, the FTC sent a civil investigative demand (CID) to Opendoor seeking documents and information relating primarily to statements in the company’s advertising and website comparing Opendoor’s offers to purchase homes to selling in a traditional manner using an agent and statements pertaining to Opendoor’s offers reflecting or being based on market prices…

As a result of investigations of this nature, entities may face litigation or agree to settlements that can include monetary remedies and/or compliance requirements that may impose significant and material cost and resource burdens on the entity, require certain aspects of the company’s operations to be overseen by an independent monitor, and/or limit or eliminate the entity’s ability to make certain claims in its advertising collateral or on its website…

Opendoor refuses to call out how much revenue it generates with blanket referral fees, and/or commissions from Opendoor Brokerage and Open Listings operations.

We generate revenue primarily from the sale of homes that we previously acquired from homeowners. In addition, we generate revenue from additional services we provide to both home sellers and buyers, which consists primarily of title insurance and escrow services, Buy with Opendoor, List with Opendoor and Opendoor Home Loans…

Opendoor is not a technology company; it is a real estate broker and a real estate investor.

Our existing and potential competitors include companies that operate, or could develop, national and/or local real estate businesses offering services, including real estate brokerage services, mortgage, and title insurance and escrow services, to home buyers or sellers…

Opendoor has identified a material weakness in their internal control over financial reporting.

We have identified a material weakness in our internal control over financial reporting. A material weakness is a deficiency, or a combination of deficiencies, in internal control over financial reporting such that there is a reasonable possibility that a material misstatement of our annual or interim financial statements will not be prevented or detected on a timely basis…

Opendoor thinks that it offers a an efficient and scalable business model.

Opendoor has a highly efficient platform to buy and sell real estate. Opendoor applies technology to reduce costs through centralization and automation and is able to achieve economies of scale not available to traditional agents, creating savings that can be passed along to Opendoor’s customers. The more transactions Opendoor completes, the more refined and cost-efficient its products become…

Opendoor derives it’s unicorn valuation from revenue, disregarding for the high cost of said revenue.

Opendoor Technologies will have an anticipated initial pre-transaction enterprise value of $5.0 billion (excluding unrestricted cash and marketable securities on Opendoor’s balance sheet as of June 30, 2020 and the proceeds from the proposed transaction), implying a 1.0x multiple of 2019 revenue and a 0.5x multiple of 2023 projected revenue…

Opendoor wants to redefine the residential real estate by flipping few thousand homes each year.

Our goal is to redefine residential real estate, the largest undisrupted category in the United States. In 2019 alone, more than 5.3 million existing homes were sold, representing more than $1.6 trillion in transactions. Additionally, with 68% of Americans living in a home they own, housing is the single largest consumer expenditure in the United States, ahead of transportation, food, insurance, and healthcare…

96.5% of all Opendoor instant offers are systematically declined.

In 2019, we sold almost 19,000 homes and generated $4.7 billion in revenue. In that year, more than 560,000 consumers requested an Opendoor offer on their home, averaging approximately one every minute…

Less than 1% of all transactions are processed by iBuyers.

In 2019, there were approximately 5.3 million sales of existing homes, totaling approximately $1.6 trillion of transaction volume with a median home price of $271,900. Online penetration represents less than 1% of home transactions, based on iBuyer volumes in 2019.

Opendoor hides the break-down for cost of revenue other than home sales revenue.

Cost of revenue includes the property purchase price, acquisition costs, direct costs to renovate or repair the home and real estate inventory valuation adjustments, if any. These costs are accumulated in real estate inventory during the property holding period and charged to cost of revenue under the specific identification method when the property is sold. Additionally, for our revenue other than home sales revenue, cost of revenue consists of any costs incurred in delivering the service, including associated headcount expenses such as salaries, benefits and stock-based compensation…

In 2019, Opendoor spent $384.4 million on Sales, Marketing and Operations.

Sales, marketing and operations expenses decreased by $64.0 million, or 33%, for the six months ended June 30, 2020 compared to the six months ended June 30, 2019. The decrease was primarily attributable to a $31.0 million decrease in advertising expense, as we largely suspended paid marketing spend in the second quarter of 2020 in response to COVID-19. In addition, property holding costs declined by $11.3 million due to lower inventory volumes and resale broker commissions and resale transaction costs declined by $9.7 million and $4.6 million, respectively, due to lower resale volumes. Personnel expenses decreased by $9.1 million due to headcount reductions, offset by an increase of $5.3 million for employee termination benefits incurred in conjunction with the April 2020 workforce restructuring.

Softbank will continue to hold the majority of stake in Opendoor.

The SCH board of directors considered that Opendoor’s existing equityholders would be receiving a significant amount of Opendoor Technologies’ common stock as its consideration and that 100% of the existing equityholders of Opendoor are “rolling over” their existing equity interests into equity interests in Opendoor Technologies which would represent approximately 82.4% of the pro forma ownership of the combined company after Closing…

Author: Litesand

Antitrust, real estate, e-commerce, fintech, proptech, bigtech

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: