Description

I recommend purchasing Opendoor’s unsecured 2026 notes (par amount $978 million) at a price of $56 with a yield to maturity of 16.6%. Assuming that in 3 years the notes have a yield to maturity of 7% given they will become current liabilities, the 3-year IRR would be 19.5%. The conversion price of the notes is $19.23 and as a result they are relatively unlikely to convert to equity. This opportunity exists because these are busted converts and there is skepticism around Opendoor’s business model. The key benefit to investing in the debt over the equity is that while the debt has an equity-like return, it has a much lower impairment risk, especially given the low price relative to par.  

Several previous writeups have described the Opendoor equity investment (pcm983 in September 2020, lightspot in April 2021, and StaminaVIC in August 2022), so we will not go in depth into the business model but rather highlight the key factors for the notes. 

At a high level, Opendoor buys homes from sellers at roughly fair current value less the agent fee, and then keeps the fee, as well as home appreciation as gross profit. In an environment of rising home prices, Opendoor can bid more aggressively as home price appreciation (HPA) will likely be a bigger number, while in a weak HPA environment OPEN will price homes more conservatively and be paid more for the value it is providing to a seller in an uncertain environment. Over the last twelve months, it had revenues of $15.4 billion, adjusted EBITDA of $429 million, and $123 million of Adjusted Net Income. 

-

OPEN’s capital structure is relatively simple, as are those of its VIEs. De-consolidating its VIEs and ignoring goodwill / intangibles, it has $2,465 million of cash, $605 million of VIE equity, $174 million of Real Estate Inventory, $160 million of other assets, and $54 million of PP&E. These are funded by $2,289 million of tangible equity, $213 million of various liabilities and $956 million balance sheet value of convertible notes. OPEN’s bonds are well covered by its holding company cash: it has 2.6x of its converts in cash and marketable securities. 

The VIE’s capital structure is fairly levered but does not have recourse to the parent company. It has $605 million of equity and $6,538 million of debt, which back $6,454 million of real estate inventory and $606 million of restricted cash:

As of June 30th, the interest rates were 3.0% for the current revolving credit facilities, 3.6% for the non-current asset-backed senior term debt facilities, and 10% on the mezzanine term facilities. The final maturities of the current facilities range from June 2023 to April 2025, and those of the non-current facilities range from April 2025 to July 2027. 

As of June 30, 2022, the Company had fully committed borrowing capacity with respect to the Company’s non-recourse asset-backed debt of $8.7 billion; this is comprised of $4.3 billion for senior revolving credit facilities, $1.9 billion for senior term debt facilities, and $2.5 billion for mezzanine term debt facilities.

The senior revolving credit facilities are typically structured with an initial revolving period of up to 24 months during which time amounts can be borrowed, repaid and borrowed again.

OPEN’s asset-backed senior debt facilities generally provide for advance rates of 80% to 90% against the cost basis in the underlying properties upon acquisition and the mezzanine term facilities will finance up to 100% of the cost basis in the underlying properties upon acquisition. The maximum initial advance rates for a given financed property vary by facility and generally decrease on a fixed timeline that varies by facility based on the length of time the property has been financed and any other facility-specific adjustments.

-

Given the company’s strong current capital structure, the main risk for the bond’s impairment would be continual losses that erode Opendoor’s asset base. These losses would come from Opendoor’s operations. 

Opendoor’s operations’ economic model is to primarily generate gross profits from its seller fee, pay for direct selling & holding costs resulting in a contribution profit, and use that to fund the required interest expense and corporate overhead. GAAP net income would also subtract SBC but that is not a cost that takes away assets from creditors. Understanding Opendoor’s potential for losses requires making estimates for its various margins. 

Gross margins: As is well known, we are currently in one of the sharpest residential home price declines in recent memory. It Is important to note that even in this environment, where very much has gone wrong for Opendoor, including a very sharp market turn, it will likely achieve adjusted gross margins in Q3 very close to 5%. This is a reasonable lower bound for gross margins, with 6% a more reasonable near term target given that (1) Opendoor is unlikely to repeat its pricing mistakes from Q2 as it now clearly knows we are in a difficult HPA environment, (2) that is below its 2019 adjusted gross margin of 6.3% – a period that is more representative of what normal for the company should be, and (3) it achieved a 7.3% margin in Q4-21 when it was selling homes it acquired during a period of significant overpayments by Zillow. More specifically, Opendoor substantially increased the spreads it charged clients beginning in May, and those higher spreads will apply going forward, but Q3’s and some of Q4’s sales will still reflect purchases at lower spreads made in early Q2. 

There are already signs of significant margin improvement. Based on third party data, from its trough at just under breakeven for homes purchased in May 2022, embedded gross margins (based on listing prices) improved 230 bps in June, 350 bps in July, and 360 bps in August.  

Currently the sentiment around Opendoor outlook is fairly bad given articles like this (Home-Flipper Opendoor Hit With Losses in Echo of Zillow Collapse, Company lost money on 42% of its August resales after it failed to anticipate slide in housing demand) https://www.bloomberg.com/news/articles/2022-09-19/home-flipper-opendoor-hit-with-losses-in-echo-of-zillow-collapse which ignores the revenue contribution from the spread charged to home sellers. Including that spread results in a Q3 adjusted gross margin of ~4.6% according to third party data. There are also fears that home prices could drop very quickly but as Black Knight’s Home Price Index chart shows, we are already in an environment of significant weakening home prices and Opendoor is faring reasonably well given the circumstances. By the end of Q3, Opendoor will have already withstood 2-3 months of reasonably negative HPA (July and August, and probably September) that are large by historical standards and still done ok (namely: roughly breakeven unit economics after interest):

There are several avenues for gross margins to increase longer-term, with the most significant being a more normalized market environment, increased learnings in newer markets to more accurately price homes, figuring out how to attach mortgages (could be a 50 bps opportunity if Opendoor executes long-term), and additional ancillary services. For example, by Q2-21, Opendoor Backed Offers crossed a $1 billion GMV run rate. This product allows Opendoor to earn mortgage origination economics. Alternatively, when Opendoor helps sellers purchase their next home, it earns buy-side commissions. Longer-term, the company is likely to expand into home warranty, upgrade & remodel, home insurance, and moving services which will potentially add $7,500 in profits per home. Finally, Opendoor has the ability to increase spreads without losing significant volume (based on 2019 data):

Selling and holding costs: The company currently incurs approximately 3.2% of selling and holding costs split between selling (2.4%) and holding (0.8%). While it is difficult to reduce holding costs, since those comprise home maintenance, taxes, utilities, etc. it is possible to reduce selling costs as Opendoor sells more homes directly to buyers and pressures agent commissions. I estimate that Opendoor could lower its selling costs by 50 basis points if it was able to keep growing its mix of homes sold without agents or by forcibly reducing buy-side agent commissions. 

The above savings are not only theoretical as the company has a track record of improving operations. As the image below shows, from 2018 to 2021 Opendoor structurally improved gross profits by 99 bps, and holding & resale costs by 141 bps. 

Interest: Interest expense is currently running at about 2% and should increase to 2.4% as OPEN’s borrowings reprice to a higher fed funds rate. 

Overhead: Opendoor guided to $190 million of adjusted opex in Q3, down from $204 million in Q2 representing approximately a 7% decrease, and it expected exiting Q3 at an even lower rate. These cost cuts are accomplished by flexing down third-party capacity, focusing marketing spend on the most efficient channels, and reevaluating all corporate costs. The company has demonstrated a strong trend of leveraging opex, with its percentage of revenues declining from 12.4% in 2017 to 4.2% over the LTM period. Additionally, Opendoor’s partnership with Zillow will likely reduce marketing costs. I assume medium term that opex will come down to 4% of revenues, but there should be room for further downside given OPEN’s original model called for it to be approximately 3%, assuming a 7% contribution margin. 

Assuming a medium term 6% adjusted gross margin on $3.5 billion of quarterly sales (roughly consistent with its Q3 purchasing rate), 3% selling and holding costs, 4% in adjusted opex, and an interest expense corresponding to 2.4% of revenues results in approximately $100 million of quarterly equity reduction. Note that this is somewhat conservative given it assumes that Opendoor achieves no operational improvements or market upside. Under these assumptions, Opendoor still has $571 million of tangible equity in Q2-26 before the bonds come due and is therefore likely to be able to pay them out. 

Assuming for the sake of conservatism that gross margins are 5%, Opendoor will have approximately no equity by Q2-26. Given that in this scenario, Opendoor would have assets that roughly match liabilities, it would still be able to pay off the bonds. Since the bonds currently trade at 56 cents on the dollar, a pay off at par would represent an attractive return. At a payoff of 80 cents, the IRR would still be approximately 10%. On the other hand, if gross margins are 50 bps higher, estimated cash at Q2-26 increases by $286 million vs. the 6% gross margin scenario highlighted above. 

The main reason Opendoor is unlikely to suffer large losses on its homes is that home prices, especially nationally, move very slowly. Furthermore, Opendoor will adjust spreads higher to compensate for a declining price environment. Outside of the subprime crisis, there were only 6 quarters of 1% or less HPA declines. The worst quarterly price drop during the GFC was ~3%. Some of these quarters included the highly inflationary early 80s where the drops were between 1-2%. Even an unexpected 4% loss (for example 2 cohorts that underperform by 200 bps), on $3.5 billion of home sales would have a one-time reduction in equity of just $140 million. 

To the extent home price declines are the result of increasing mortgage rates, Opendoor has already had the mortgage rates reset meaningfully. While 30-year mortgage rates averaged 3.1% in December 2021, through the end of Q3 they had already increased to 6.7% and over that time frame OPEN had two extremely profitable quarters (Q1 and Q2) and will likely manage ok in Q3.  

Additionally, much better credit standards and strong income growth has resulted in record low mortgage delinquencies materially reducing the risk of forced selling:

Unlike mortgage lenders and “fix & flip” operators, Opendoor intends to hold the homes for only several months which does not give homes enough time to lose substantial value. Even in slower markets, existing home sales almost always exceed a 4 million per year rate allowing Opendoor to continually operate at meaningful scale:

-

There are several paths for faster upside, including repurchases of the debt, any improvement in operations, a sale of the company to a well-capitalized entity, or an investment by a strategic. For example, Jon Jaffe, the Co-CEO of Lennar is on OPEN’s board and Lennar used to have an investment in the company. 

The management team is likely to be reasonably aligned with the debt holders. The company’s founder and CEO, Eric Wu, owns over 31 million shares (5.03% of the common stock), while the remaining board and executive officers own over 38 million shares (6.2% of the common stock). Additionally, if the company ever restructures, it will likely liquidate rather than reorganize. These facts incentivize the management team to avoid bankruptcy risk – unlike a management team that has no shareholdings and/or hopes to gain an equity stake through the bankruptcy process. Outside of the recent difficult patch, management has executed well. 

Catalyst

A normalization of the home price environment, demonstrating higher gross margins on new home acquisition cohorts, cost cuts, debt repurchases, strategic actions by the company.