Have you given thought to opening up your pricing model/algorithm to the public for every home in a market (a la Zestimate/Redfin estimate) as a means of eventually driving seller inventory even if someone is not ready to sell today?
If someone could look up what their place is worth and, unlike Zillow and Redfin, actually be told explicitly why it's worth that as opposed to being told a "black box" value, you'll open the top of the funnel by driving folks who aren't yet ready to sell (today's owner is tomorrow's seller). And since those values change frequently the not-ready-to-sell owner could track the changes to value and the underlying data driving that value, which would keep you on the tops of their minds when they're ready to sell down the road.
If you think the pricing model/algorithm is too proprietary to share beyond the "I'm ready to sell today" market, I'll stop by for a coffee to try and convince you otherwise.
Good luck. I'm stoked to see someone trying to fundamentally change the residential market.
Edit: oh, btw, not advocating sharing the actual algorithm, but the results of it (e.g. your house is worth 5 psf less than the comp from 3 blocks over bc your place is directly under the airport's flight path).
This isn't game changing, this is just good marketing. The real game changers are the ones over at OpenListings (YC F15) who are giving back commission checks and doing their best to eliminate the real estate agent entirely.
He's not trying to fundamentally change the market. Just like someone else mentioned, this basically a bonafide residential REIT.
They're introducing MORE costs to the transaction, creating a closed marketplace where another closed marketplace already exists, and framing it as some revolutionary startup. They're only doing this because they're big name valley guys who raised a ton of money.
I believe they are trying to fundamentally change the market.
For a fee (I guess it's 1-6% on top of the typical 4-6% commission?), Opendoor removes the pricing (list price), appraisal and financing risk and let's the seller unlock the value of the house in a fraction of the time. On top of that, and unlike the current process, if I'm understanding correctly, they also make it explicit to the seller why their house is worth what it is. And I can't belabor that point enough: the reason the residential real estate market is so opaque is because of the uncertainty/inaccuracy around pricing. Realtors do their best to determine value, but there is a specific reason properties sell in 2 days (priced low) or 6 months (priced high) and it's strictly because of erroneous pricing (btw, this isn't true in SF, for instance, where agents intentionally price low to get bidding wars, but that is not the norm nationwide). We've reached this point where the data is largely available so that the historical method of picking a few "comps," licking your finger and seeing which way the wind's blowing will soon be a thing of the past and that's the crux of what OD is doing. If that means they're not fundamentally changing the market, I'd ask you to look into how property gets priced today by the two parties responsible for making sure the deal gets done (the broker/agent and the appraiser).
As for how the seller perceives things, if I tell a seller your home is worth X, and here's all the data supporting that--actually quantifying why your friend John's house down the block sold for 5% above what we're offering you--the seller is informed and can decide if they're willing to assume those risks I mentioned above to get a higher price than OD is offering. Assuming I have a house at the average US home price ($190k or so?), if you tell me that I can take OD's $186k "no risk" offer or wait it out and hopefully get the full 190k given all those risks mentioned above, I suspect many people would just take the 186 and move on. And either JD's in his own little reality distortion field (probably somewhat true) or he and their investors are actually seeing this work with the unit economics to back it up.
One last note in this horribly long comment: there is so much opportunity for economies of scale here with the pricing of items like marketing costs, inspections, repairs, appraisal costs--heck, even financing rates--that I suspect and probably hope in vain that over time OD will accept even lower fees. In some fantasy land I can imagine OD getting confident enough in their pricing that transactions end up costing the seller 2% (total, completely wiping away the need for a real estate commission on the sell side and once they merge/acquire Openlistings they'll drive down the buy-side commission rate as well). And of course if this ends up working at scale for OD every deep-pocketed institution will create divisions that do this as well, which will just mean that OD is the first of many new-age residential real estate broker (call them a market maker or REIT if you like that label better, but everyone will sell in this more liquid, transparent way in the future... how long will that take? Well, if there's another housing bubble right now, all bets are off, but otherwise this could happen relatively quickly).
what's the dream business model of every VC? a marketplace.
the ideal outcome, though perhaps improbable, is likely to become the ebay of real estate. what they lack is critical mass.
today, they're focused on seeding the marketplace with sellers, that is filling the supply side, which is the natural place to start.
this is a long way of saying that it seems not at all unreasonable to hope they will emulate amazon's AWS strategy of reducing fees as costs fall, not only to deepen their competitive advantage but also to drive them closer to critical mass. but given the high costs and uncontrollable risks, they need to, in the early stages, compromise growth for unit economics and cash flow.
fundamentally, they're using machine learning and automation to remove guesswork and fat from real estate transactions, so this company will be interesting to watch.
what many commentators miss is that OD is selling a service to home owners. it would be helpful if knowledgeable real-estate people focused on (a) estimating how many sellers want this service; (b) if 2% seems reasonable for OD's service; and (c) if this initial market is large enough to jumpstart their marketplace ambitions (assuming this is the end goal).
I agree with so much of what you say here (hope for the AWS strategy; the fact that this is a service business, which is what the existing brokerage business is as well of course; using data/ml to remove (massive) fat), but I'm torn about whether their end goal is actually, or even should be a marketplace in the traditional ebay/zillow/trulia sense. There are countless residential marketplaces with good reach already that as long as their pricing model works really well they can focus on supply side exclusively (i.e., buying for 98 cents on the dollar) and then using the existing marketplaces to market. They have to find buyers, of course, but that doesn't mean they need to directly source the buyer. In this model, as the old real estate adage goes, you make your money when you buy the property (that's not meant literally of course).
you could be right. i only started thinking about OD with this thread and have no inside information. that said, a marketplace seems like the ultimate ambition for two reasons.
1. one co-founder is a VC who previously led a marketplace business, paypal. VCs are already conditioned to aim for marketplaces. tasting marketplace success as an operator probably only deepens this reflex. it's harder to eat ground beef when you're used to wagyu.
2. if a marketplace is possible, why not? it creates a larger moat to keep out competitors and assuming they don't abuse monopolistic power, a single marketplace can make pricing fairer (more buyers, more sellers) and more efficient (lower fees). instead of 6%, what if OD enables a world where transaction fees are 0.5% or 1%? the classic disruption playbook (hi, telsa!) is to aim for a dominant market position by starting high, then using technology to steadily reduce prices, expand use cases, and gain market share.
I come from a marketplace background myself and went through two bubbles so you were spot on when you said it was easy for me to say "bets are off in a bubble"--it was easy! And it would have been even easier for me to say marketplace = no brainer. But I just think there are some benefits to leveraging existing demand-side players.
At the risk of agreeing with you too much, though, "why not?" is true, but I think the numbers could be pretty attractive regardless if they ever go/get there.
No inside info here either, btw, though I like to think I get some of the nuances in this market. Sounds like you do as well.
This whole system is so bizarre to me. In Australia, standard practice is that a seller pays ~2% to an agent to have them sell their house. They then pay a sales tax to the state government, that scales weirdly, but call it ~0.5 to 0.75%.
Buyers pay nothing to anybody, except ~$1000 to a conveyancer/lawyer to sort out paperwork when the deal is done (and of course the price of the house, to the seller or mortgagee...)
It is bizarre. The difference in numbers between what you're saying exists in Australia and the US is staggering. I'm clueless about the market there, but even if there are laws on the books that better protect buyers/sellers, and so there's some justification for the "risk premium" that US buyers and sellers pay for protection throughout and after the transaction, I'd wager a princely sum that the magnitude of that premium is in no way justified.
Btw, this wasn't always true. Some of the protections/standardizations introduced post-recession actually makes the US residential market more ripe for an OD-type model. I could go on and on about why that may be the case, but I'm guessing others are sick of me writing long-winded comments in here even though you and I are probably the only people still interested in this thread.
As someone else said, though, sure is interesting to watch.
It is possible to hire a buyer's agent, but I don't know anyone who has. Everyone is on an equal footing, sellers just want to sell to the highest paying buyer.
The most common way of selling is via public auction. Each weekend for 6-8 weeks beforehand there's an hour long "open house" where the sales agent shows people through. Then on auction day, interested parties bid in public. I don't see how a buyer's agent would help change an auction's outcome.
Sometimes sellers elect to sell by other methods; best offer, best offer by set date, sales agent to negotiate with interested parties, etc. I suppose in those flavours a buyer's agent might be of some assistance.
I can only assume in the US that there's more risk of something going wrong - do buyers drop out frequently after making an offer or something like that? How else are the fees anywhere near justified for a simple transfer of property?
In the US sometimes auctions are used, but it is not the norm for most markets. Usually the seller will hire a real estate agent, who helps the seller determine what to ask for the home. Then, the sellers agent will be responsible for listing and marketing the home. The cost of marketing is usually paid by the real estate agent, so that can be part of why the commission would be so high. Back in the day when there was no Internet and listings were primarily advertised in newspapers and special directories, marketing was probably a much higher cost.
As far as the buyer goes, they will usually have a buyers agent to help with showings, negotiations and to coordinate the process. An offer is usually contingent on several things- the main ones being 1) home/pest inspections; 2) Appraisal, 3) Financing.
So real estate deals do fall through often enough. Maybe somthing is noted on the inspection- the buyer might ask the seller to pay for a repair, and the seller decides not to. FOr example, when I was shopping for my first home we found a nice little place, I was little concerned about a few things with the roof, but I made an offer contingent on an inspection (as well as the other things noted above). My offer was accepted, so I had it inspected. The roof was really the only concern, the inspector felt that it should be replaced. I told the seller about the problem, and they decided that they were not going to pay anything towards a new roof, so we mutually agreed to nullify the contract.
Later I found another home- it passed all inspections, but the appraisal came in $900 less than what I had offered. The seller was going through a bankruptcy and had to have the sell approved by the bankruptcy judge, and since they had already approved it at my asking price, it was not worth delaying the closing over getting the price reduced $900, so the bank and I agreed that I would just pay an additional $900 down at closing to bring the debt/value ratio to where they wanted it, and we closed. (If it had been a more significant proce difference, that may not have worked out, resulting in another voided contract.)
In other cases, financing may not come through- the lender will pre-approve the buyer, but that is not a guarantee- the lender can decide they don't want to lend on a property due to the condition or something else, they may discover something during underwriting they weren't aware of about the buyer, the buyers situation may have changed, or the market conditions for loans may have changed (interest rates increased); all of which may result in a denied loan, and thus a voided contract.
Real estate agent are supposed to make this all as smooth as possible and help inform buyers and sellers to minimize the risk of any of these things happening.
I agree that with modern technology the current commission structure is probably not optimal, though.
on the surface, it's easy to say "all bets are off" in a crash/recession, but would you say the same for retailers like gap and tiffany?
the business models are conceptually similar: buy a product from suppliers (clothing manufacturers for gap, home owners for OD) and hold in inventory until a seller is found.
the key difference, of course, is cost of goods. each product costs OD hundreds of thousands of dollars (and eventually millions) while each gap product only costs a few dollars.
like gap, OD can directly control how much it buys from suppliers while indirectly controlling how much it sells via discounts.
predicting home demand is obviously much more difficult, because both the buying psychology is different and the sheer volume of apparel transactions makes forecasting demand more reliable.
nonetheless, the principles of managing cash flow and inventory risk remain the same. provided OD isn't carrying excessive inventory, the business model can survive downturns in different geographies. (what they can't survive is an environment where homes everywhere perpetually depreciate.)
put another way, if people can earn money in the stock market during a recession, why can't OD do the same for the housing market? OD can treat homes like equities in a stock portfolio where they try to balance losses in one area with gains in another.
in summary, OD's fatal flaw won't be a housing bubble (unless it's global and permanent), the fatal flaws will be the risk, pricing, and inventory models. and until you have insight into those, it's impossible to properly assess OD's prospects.
it's also important to note that inventory risk could decrease over time, as once the pricing and demand models are honed, they could allow third-parties to finance some transactions (e.g., home X has a 80% chance of turning over in 30 days).
Nice to see someone still reading this. I again agree with almost everything you're saying here. Plenty of companies flourish in difficult climates, but the margin for error is much greater in a rising or static market. In a falling market there's just less room for failure (like there is for most things). Put another way, in reference to my previous comment, I'd be more comfortable betting that this new, more efficient residential real estate market will emerge sooner if there is not a housing bubble (or rather a bubble that bursts).
it seems we nearly agree 100%. the comment about a crash/recession was because it seemed like you were interested in joining the company but were concerned about the repercussions of a bubble.
if you have coffee with the founder, instead of focusing on macro risks, perhaps focus on OD's models for risk, inventory, and cash flow. because they can cherry pick geographies, it's actually easier for them to grow now during a recession than when they're much larger and need more transactions to move the needle.
if you believe the models are sound, then focus on whether you believe the team can adhere to these models, that is whether they can resist the temptation to grow at all costs -- and thus place riskier bets.
we also agree OD could stop short of a marketplace and remain an attractive company. we only diverge on their marketplace ambitions.
So it's a real estate speculation company. Fine. But let's not pretend like some tech wizardry bs is going to give OD an advantage over any other speculator
you're right. we don't know enough about OD's tech and models to conclude anything about OD.
however, most believe information confers an advantage in asset speculation. so if machine learning and data can provide someone with faster, more accurate pricing, that someone -- whether OD or another startup -- possesses a competitive advantage in terms of purchasing and selling real estate.
Co-founder here, I'll do my best to answer questions. The number one question we get is likely "what happens in a downturn/recession?", but that's more of a blog post than a HN answer. Anything more detailed/technical though, shoot!
I think I read in a separate article (maybe the Fortune one) that you're focused on mid-market homes in the $250k--$500k range, in a few metro regions across the country.
Notably absent in the Bay Area. In hindsight this looks like a great move since your value add to sellers is liquidity, and you collect an illiquidity premium, so to speak. Meanwhile, the Bay Area sellers are certainly not lacking in liquidity.
So 2 questions:
1) In hindsight staying away from the Bay Area looks great but I have to think there must have been a temptation to be a player in your "home market". What was the conversation like between the founding team to eschew the local market? (I think many companies default to addressing the Bay Area first without putting in much thought into it, and it is sometimes ill advised)
2) Blackstone was buying lower end homes from 2009--2014 or so. Is the transaction volume in this segment not robust enough for your business model (i.e. hard to find buyers)? Pricing model too uncertain/unpredictable?
I think we had coffee once back in 2013 when you were still at Addepar. Congrats on the move; always thought this could be a tremendous business since news first broke about the company.
1) We're building a business, so we do what's best for the company and customers and don't pay much attention to vanity-based things like winning our home market. SF is a hard market for this business, so we simple don't operate here :) We also haven't had much trouble hiring incredible people -- relevant to HN I think we're up to 8 former YC founders
2) We generally buy from $125K to $500K, so we do hit a lot of that low end. Below that, fixed transaction costs are quite high as a % of home value.
Thanks! We all really appreciate the support, the team's working really hard to redefine a space that hasn't seen enough innovation in the past half century.
> Second, Opendoor explicitly charges sellers for having replaced total uncertainty with a bank wire: not just the same 6% that typically goes for buyer and seller agent fees, but also an additional 0-6% for “market risk” — i.e. dealing with the uncertainty of actually showing and selling the house — along with the aforementioned repair costs.
So basically, the seller is guaranteed to lose 6-12% (on average 8%) of the value of their home even before receiving a paycheck. Why can't OpenDoor start low, instead of starting high and waiting for economies of scale? I just don't see the marginal benefits worth the extra costs, and am not sure how OpenDoor can stay competitive enough with those rates.
> To succeed, it has to price the homes it buys accurately, without seeing them, and it has to sell them quickly to minimize the costs of carrying them.
I feel like if the people behind OpenDoor can do this, then they can do day-trading or at least an index fund. (Yeah I know, stocks are more volatile and housing is pretty much guaranteed to rise.)
> for too long too much money and talent has been poured into low-risk digital-only businesses
What? Isn't there hardware startups?
Ok overall, I think I'm being a bit too critical on this company because all I see is house-flipping disguised as "theoretical arbitrage opportunity". I'd rather see a new company solve the housing crisis in San Francisco and New York.
Why would you say "lose 6-12%"? The first 6 is normal real estate fee. Above that it's a normal arbitrage of house flipping (buyer thinks can spiff it up and sell it for more a short time later).
There's a lot of grey area between engineering and data science for us. To somewhat simplify, we have 17 product engineers, 5 data engineers, and 19 data scientists.
We operate the company in "lanes" like Seller, Buyes, Homes, Accuracy. The "purest" technical challenges are on Accuracy where we're modeling home prices, days on market, pricing strategy, and various measures of risk. A lot of feature engineering, data viz, etc.
On the product engineering side, we start and end with the customer problems, not the technology. Projects range from building a mobile app for inspectors to catch all issues and normalize the data, to building custom security hardware with Raspberry Pi's that relies on mobile SIM cards because we don't own the homes long enough to justify setting up/turning off internet.
You mention 17 product engineers, 5 data engineers and 19 data scientists on board now...what was your launch team like? I assume you've had massive growth / hiring but curious what it looked like at the beginning (in boot strap mode/mvp mode)
Have you considered offering a "chain breaker" service?
What is a chain? (provided in case this is a UK specific term)
A common problem with buying & selling property.
In order to buy a new house you first want to know that your current one will be successfully sold.
As such, you may have a queue of buyers all waiting for someone down the line to get a buyer for their property; as if they fail to get a buyer they drop out causing the next person in the chain to have to find a new buyer before they can buy.
Equally sometimes people waiting in a chain are "gazumped" (another buyer makes a better offer to the current buyer), thus pushing out the current buyer and impacting the buyer's chain from the other end).
You could provide a service whereby buyer and seller have already met, but are stuck in a chain. You'd be able to buy the seller's property and hold it for the buyer until they're ready.
- If they're able to sell their property they can then pay you the price of the new property plus the "holding fee" before moving in.
- If they're unable to sell their property they have the option to sell their current property to you and take the new property as described in the previous step; or to remain where they are simply paying you the holding fee for having held the original property for them, with the new property then entering your portfolio of properties to do up and sell on to the highest bidder.
- These holding fees could be charged monthly up front should that prove simpler than lump sum amounts at the point of decision.
Your service already avoids the need for a chain; but this alternate offering allows you to capture a part of the market of those people not interested in your full service, by having the potential buyers ask that the property be purchased by you instead of the sellers coming to you.
My favorite chain breaker service has to do with kidney transplants:
Some economists noticed a few cases where would-be kidney donors who weren't compatible with their family member who needed a kidney were able to find another incompatible donor-recipient pair, and were able to swap so that each donor was able to gave to the other donor's loved one.
Of course, this only works if the medical compatibility works out between the pairs. To facilitate more transplants, economists created a donor-matching network that could handle cases where three incompatible donor-recipient pairs (or even thirty!) could work out an exchange allowing all of the recipients to get a suitable kidney. Naturally, laws prevent money from changing hands for kidneys, everywhere on earth except in Iran.
That is taking a pretty big risk. Wonder what kind of price the market would pay?
I myself was able to chain break when I bought my current house, by doubling up on my mortgage (keeping old until after I had moved and it sold). Cost me maybe 5k in interest on the period of double mortgage? But because I could offer to buy my new house with a clean offer, I saved 5-10k in cost (as I would beat out a 10k offer that had a * on requiring a sale to go through first).
of course my strategy only worked because I buy cheap houses not the 40% of my income houses that the bank wants me to buy...
I have tried to use the site a handful of times and every time the offer has been uncompetitive, the site has been buggy, or the employee responses have been confused and wrong. It's actually laughable.
What gives? Or perhaps a better question is: Is this a common experience? Are these growing pains that you are aware of? I've rarely had a worse experience with a startup.
I'm really sorry, if you email me I'll look into it. Jd@opendoor.com
Can't say it's common, but pricing every home perfectly is hard. From what I can measure, our NPS is 70 and we track complaints pretty closely. For offer competitiveness, we track metrics like the final selling prices of homes we buy vs our offers and for homes that get offers and sell later on the market. I'm confident we're competitive in aggregate, but individual cases can vary.
We buy hundreds of homes per month, so not very small "n" anymore either
What immediately occurs to me (and I apologize if it's been asked already) is that you likely have to offer slightly below market prices to be able to make a decent profit and mitigate the risks. So someone in a neighborhood with a better than average home will not be enticed by the lower price and will likely sell through conventional means. Someone with a below average fixer-upper, will see the price as very attractive and sell through you. So anything that's just predicated on expecting homes listed on your service to be centered around the average value will not work. How do you account for this sort of thing if you don't see the house?
Also in a forclosure situation the owner needs to make a quick sale - your service would look very attractive (a killer niche of the market for you guys.)
Getting to the "range" for a home isn't too too hard, it's figuring out where in that range that matters. We use upgrade data and other smarter heuristics to accurately place a home vs its comps so we can try to avoid that adverse selection you're describing.
Our current pricing model & offer process is more accurate than the in-home appraisals we've done with third parties. The in-home inspection we do before close catches major repair issues and other sources of adverse selection (e.g. hoarders)
Targeted hold time varies by market, but optimal holding period is 0 days :)
We fund a small portion of each home with equity and the rest with debt-like instruments. I can't name names though.
As a former real estate appraiser, I am very Interested in your methodology and approach in valuing homes. Lenders require appraisals, do you find that the value you place on each listing come close to the appraised value?
Since you buy the house upfront for cash and hold 10% on the corporate balance sheet, isn't that a significant price & liquidity risk taken (in addition to the usual startup risks)? Because the time at which that house is eventually sold and at what price is somewhat unknowable, how do you mitigate that risk? Overall, I think what you guys are doing is very cool - wish you the best of luck.
What is the turnaround time you guys typically aim for? When considering what small improvements to make, does your algorithm factor in estimated time to complete various renovations, or is it purely an ROI calculation?
What is your agency relationship with the home seller? If the seller is working with you directly (i.e. not using a separate listing agent), is this essentially a FSBO transaction?
Full disclosure: I'm a licensed real estate agent, although I haven't practiced in years - I've handled around $200M in RE sales. FWIW, I was mostly a buyer's agent; the few listings that I took on were only for valued, returning clients and some REO's (mostly because it was a great way to meet potential buyers). (EDIT: I mention this because I'd never claim that everyone should use a listing agent, or that paying a listing agent 3% is necessarily good value for the service they perform)
I don't mean to sound overly negative here, but 6-12% seems like a pretty big cut. I understand the argument that the seller may be able to sell the house more quickly than via a traditional listing, but they are also taking on a fair bit of risk and headache as well (no guidance on making disclosures, no access to the standard forms, no unbiased opinion about true market value, etc).
It's not really an apples-to-apples comparison to say "you would have paid 6% anyways", because if they had paid 6% (or 4-5%) to a listing agent they'd get all the guidance/forms/advice I mentioned above, as well as someone actively marketing their property, MLS listing (thereby reaching a much larger market of potential buyers). From your website, it also seems like there is essentially no room for negotiations on adjustments for repairs.
Honestly, this seems like a variant of "we buy houses for cash", except "on the web" (admittedly, most of those folks try to buy properties at 30% below market value - but I'd wager that you're probably trying for around 15% below market in some cases). In most markets, if someone lists their house for 5% below true market value, they're likely to get several offers rather quickly. I'd also guess that you have your own contract, as opposed to standard Realtor contracts which limit your risk in terms of earnest money deposit, contingency removal time periods, etc. It really seems to me that the only people who would go for this are people who are rather ignorant about how to sell their real estate and therefore at a pretty big disadvantage when dealing with a very experienced buyer, such as Opendoor.
Also, it's kind of silly to give as much emphasis to the home warranty you give buyers - that's a pretty cheap & easy thing to obtain; most agents I know give them to their buyer as standard practice.
Again, I hope this doesn't come off as too negative - I think there's a lot of room for innovation in real estate; this just doesn't seem like it's actually advantageous for most sellers in my opinion, and therefore not particularly scalable once people start doing the math. Just my $.02. Please tell me if I'm missing something..I do wish you the best of luck with this!
There's a lot to unpack here, but I'll try to take it point by point. Macro-level, nobody builds a lasting company by screwing people. You have to provide a fair and loved service because all of your dirty laundry gets aired eventually. Our internal gut-check is to imagine everything we propose is placed on the front page of the NYT, how does it look? Real estate is full of sharks, our job is to not be one of them.
- Our cut is not 15%, it's 0-6% below market depending on home risk & projected holding costs. We make the fee pretty transparent. Without opening our books, you sort of have to trust me here for now unfortunately.
- We use the standard MLS contract in every market, along with those protections. We give sellers a large sum of earnest money when we make a mistake and need to withdraw from the contract. Long term, that's just good business.
- Imagine you're a contingent buyer, and need to sell before you can buy. The cost of selling, moving into a rental, buying, moving into your new home etc is very high! Most people can't afford two mortgages. We make it a single move, and we'll even let you do a "late checkout" from your home after you sell so it's even easier.
One thing not mentioned: Opendoor absolutely does not have to carry the balance sheet risk of owning the homes forever. They can chop up those assets and sell them to anyone (hedge funds, banks, individual investors, etc) at any point. At scale, they'd be able to do this and still maintain most of the upside.
At some point, Opendoor will be able to let businesses that simply manage financial risk for a living (i.e. banks) spread the downside risk out over the entire financial system, while Opendoor takes advantage of its customer acquisition and data advantages to capture the bulk upside. It's not a perpetually risk model if they don't want it to be.
Also, their position in the market gives them a very good insight into the likelihood of a crash. As demand for their services will increase whenever sellers are desperate to sell their houses.
If they track those indicators, they will be able to dynamically decrease their purchase price based on the likelihood of a crash and (in the worst case) limit their exposure to only the housing stock they had at the start.
In the best case, desperate sellers might accept the much lower price and Opendoor will make a profit off the crash.
There are two reasons why someone might sell their house to Opendoor...
A) They are willing to sacrifice some of the money for a much easier sales experience.
B) They really, really need to finish the sale now.
The size of the first group will follow a constant trend, but the second group will fluctuate based on various factors. So Opendoor can infer the desperation of the sellers in the market based on their numbers of sales.
Most other factors are publically available, but the desperation of sellers isn't publicly tracked. Regular real-estate agents will be exposed to the same desperation, but it's in their best interest to hide any desperation, to maximise the sale price.
Lenders will also know when some homeowners are falling behind on payments, but again they don't publish those numbers until a foreclosure actually happens.
It's not like Opendoor is going to predict the crash before it happens, but as long as they can spot the crash early, before it really picks up steam, they can minimise their exposure.
There's no such thing as reliably predicting a crash - that's why it's called a crash. There's bulls and bears but no one side is 'right' in aggregate.
real estate doesn't really crash, like the stock market. It kind of... slides as liquidity dried up for higher price points. All of this takes many months, with many early leading indicators if you know where to look and have the right data
The linked TechCrunch article mentions it's a debt instrument
> Norwest disclosed that OpenDoor has “hundreds of millions” of dollars of debt to its name. This debt is what the company uses to actually purchase properties.
Blackstone will definitely be eager buyers at opportune times (I believe they are currently divesting their considerable stakes in low to mid market real estate that they'd been accumulating since 2009).
There will be limitless buyers, and because Opendoor isn't competing on price, their ability to price securities to the financial markets will be absurd. I assume they'll have to keep a token % on balance sheet to show good faith but that's it.
Just remembered -- FirstREX does exactly this. Equity financing of homes, and selling off their stakes in collateralized fashion to institutional invsetors.
A market liquidity crisis would have to hit and have fully felt impacts over night for this to be a problem for Opendoor, since they flip inventory relatively quickly. Even a crisis situation would create only moderate medium term risk for them actually.
The issue a lot of people have with OpenDoor is their business model. Only some people in silicon valley can raise $320mm to wholesale homes using algorithms. The skepticism I have is that OpenDoor isn't disrupting anything because they're not Airbnb or Uber. They're not going after unused inventory and taking huge amounts of business from incumbents, they're simply amassing massive amounts of debt like no wholesaler could and when the economy crashes, and it will as nothing goes up forever, they're going to have this backfire on them. And then what?
When the economy goes down, Airbnb will do just fine. More people will want to do short term hospitality to earn cash and more people will want to drive for Uber. OpenDoor will have to liquidate their debt which will prove difficult. Also if I remember correctly, Keith Rabois said he was talking with Peter Thiel and Peter mentioned how real estate has rarely changed. Later Keith helped create, more like fund, OpenDoor. What I wonder is, why hasn't Peter invested in Keith's new company? It's not because he's too busy with Trump's transition team.
OpenDoor isn't a startup you should emulate. I wholly disagree with this article.
Without going into too much detail, investors aren't generally in the business of losing money and at the scale of investment we've taken we're far beyond "bet on the team" kind of diligence.
Managing inventory and pricing risk is a key piece of this business and has been since day 1.
We really owe the world some blog posts, but until then hackers, data scientists, and predictive modelers are always welcome to come by the office for lunch!
>>investors aren't generally in the business of losing money...we're far beyond "bet on the team" kind of diligence
Sure, in principle large investments should mean good DD. But Magic Leap fans were pretty surprised this week. Theranos also raised more than Opendoor.
It's not fair to put other company's problems on Opendoor, but the point it's hard to accept the large cap/DD argument with so many counter examples.
In any case, it would be great to see you succeed - good luck.
That's fair and those are embarrassing tales in Silicon Valley. One difference is they were/are bets on a technology that wasn't in the market yet.
We buy and sell thousands o homes, hopefully at the appropriate price for the market's current level of risk. Hard, deeply appreciated by customers, but not much handwaving on how it will go to market.
In negative market conditions, you can't stop gravity. Not with VC funding, engineers or technology. Your business is a double edge sword - the only way you grow is by purchasing more debt. You're ignoring reality with a delusional belief in consumer demand.
>investors aren't generally in the business of losing money and at the scale of investment we've taken we're far beyond "bet on the team" kind of diligence.
LOL. When an SV founder defends his business model by pointing to fundraising numbers, look out.
> OpenDoor isn't disrupting anything because they're not Airbnb or Uber
Prolonged sales cycles remove liquidity from the markets and generally restrict economic mobility of the population at large, as one needs to sell their current house in order to shop for a new one.
On the buying side having a no-haggle buying experience is also somewhat appealing to modern buyer https://www.opendoor.com/homes
The real estate is currently financed via debt instruments, and issuing debt for a real estate potfolio purchases is a fairly well-travelled path for many financial institutions and investors out there. OpenDoor is likely to spin that out as a short-term residential holding REIT, for better tax treatment, if nothing else.
Macroeconomic risk is always a factor with real estate investments, but the exit options are also well-known. Liquidate at any price, convert it to rentals, or wait it out.
If you can sell your house in 1 day vs 100 days that is definitely Disruption by any standard. I buy and sell used cars randomly and I would love to do this to the market.
Keith recounted the story here, a bit, but Peter didn't comment: www.forbes.com/sites/amyfeldman/2016/11/30/home-shopping-networkers-opendoor-is-upending-the-way-americans-buy-and-sell-homes/#6d08e3f5cff1
"A member of the so-called PayPal mafia, the founders and early senior execs at the payment platform who went on to start many other companies, Rabois, 47, came up with the idea for Opendoor back in 2003 while doing a stint at Peter Thiel's investment firm, Clarium Capital. Rabois recalls a time when he and a colleague presented an idea similar to Zillow, the online real-estate database, that Rabois says Thiel found so boring he kicked them out of the conference room. (Thiel did not respond to requests for comment.) Rabois then started thinking about how car owners can sell their used autos to dealers based on their Kelley Blue Book valuation and how he might create similar liquidity for homeowners."
The article talked about Opendoor being able to increase liquidity. However a lot of real estate illiquidity is due to people who refuse to sell at market value for irrational reasons.
For example, "I paid X so I won't sell below that", "I added a pool plus 100k in renovation so I should get that much of a premium", "I'll just wait a few months and see if prices get better".
A big function of realtors is to console and convince people to sell at a realistic price.
This behavior makes it harder for OD to improve liquidity, and in fact liquidity could take a hit with no realtor therapy for sellers.
I predict a major issue for Opendoor will be seller sticker shock. Many will balk at paying 8-12% in fees even if it's in their best interest.
Willingness to paying more for convenience or speed is not a constant. See mental accounting from University of Chicago, or ask any realtor about seller psychology.
Give it some time. Useful to look into why Redfin never really took off.
There's so much fat in real estate outside of the agent fee, too. Mortgage closing costs, title insurance, escrow fees... it's ripe for improvement full-stack.
The fact that real estate agents haven't suffered from the same negative publicity as banks for their part in the housing bubble tells you everything you need to know about their political power.
I thought similarly, too, but if time is money, then they really might be reducing the transaction costs of the seller, by allowing the owner to sell _now_ and without worries about the buyer's financing coming through, etc.
Also the cost of moving twice. First into a short term rental so you can get the money for your next down payment from your current home, then moving into your new home.
Interesting. So Opendoor removes that risk for a premium on fees it seems.
What is the path to reducing the fees overall? 6% seems a bit much in this day and age, particularly in the Bay Area where an agent sets up a few signs, posts the listing, does an open house for one weekend and that's it. And agents (both sides) are incentivized to close more deals, not get the best prices. Ultimately homeowners/buyers get screwed.
Market making is much closer to how we at Opendoor think of it internally. Flippers buy distressed assets at a huge discount and plow money into renovations and repairs. Not really our business.
It strikes me that maybe the best analogy for your financial model is old-school 19th-century shipping (as in; the reason London is London, more or less)? You're taking a trading risk on illiquid goods but your holding times are reasonably short (certainly under a year, presumably much less than that in hot markets).
> Or, in the next downturn, the entire company might go bust.
> To that end I hope Opendoor succeeds simply so it can be a role model for tech: taking on big risks for big rewards that create real value by solving real problems is the best possible way our industry can create benefits that extend beyond investors and shareholders;
The only reason (ok one of the big reasons, not the only) Opendoor exists is that the massive amount of capital deployed has the potential to make investors and shareholders a ton of money. It's an all or nothing proposition. Furthermore, I don't think the author has rightly assessed (or assessed at all) the negative externalities associated with Opendoor's model to the economy. Part of the whole reason the housing market revelation of "too big to fail" of banks was exposed was because too much of the real estate market was tied up in too few organizations (Fannie/Freddie, Wells, etc), which means when it crumbles down, our economy cannot sustain the burden. If OpenDoor's risk mitigation model is to basically "own more of the market" it means that if/when the market does go for a downturn and OpenDoor goes bust, then it's not just VC's who lose, but a whole lot of homeowners, or even worse, homeowners and taxpayers.
That's hardly a model I'd admire or try emulate, unless of course, I want to make a buttload of cash (or fail very hard trying).
> Opendoor is creating value as opposed to taxing a few bucks off the top of an existing market or simply trying to be cheap.
Opendoor's arbitrage is no different then an ad network, they're both exploiting inefficiencies. I'm really having a hard time understand why it's so much more "benevolent" as it's suggested here.
It's not that it is 'benevolent' - I think the article was making the point that what OD is trying to do is harder, and socially more beneficial, than just matching demand & supply using an online service/app (like Uber, Airbnb, RedFin. etc).
The service OD is providing is liquidity - and the risk they are taking, which they are compensated for, is liquidity risk. This is what 'market making' is - you stand ready to buy & sell and create efficient market.
Beyond the 'data science' of pricing the property, and the tech of automated visits, - the main operational task would be to build an efficient market making business. Anyone who has spent any time in a bank trading desk would know how difficult this is. In a rising market - volumes and upward pricing and demand all tend to go hand in hand - the real test, will be when when the market turns - and maintaining liquidity (and a thin 'bid-ask' spread) in a falling market.
Someoned use the analogy of London trading houses... Another good analogy would be what Marc Rich (and others) did in the 1970s when the broke open the oil market - which had been dominated by long-term contracts - into a more liquid 'spot market'. OD is trying to create a liquid, 'spot market' for a previously illiquid market.
OF course - the market which need the most liquidity (ie the ones outside the 4-5 main metros) is where providing liquidity will also be the hardest. Unlike bonds, shares and oil - housing is not 'fungible' (for one, you can't move a house).
Sounds like a difficult but valuable enterprise - good luck to you and your team.
https://news.ycombinator.com/item?id=13164392 According to `jdross, purchases are funded with a small amount of equity and the rest with "debt-like instruments" so the taxpayer is still potentially on the hook. But, yes, homeowners don't seem to be at risk.
Possibly, but I think the simpler answer is that we have a product with...
- good unit economics in appreciating and depreciating sub-geographies
- customers who love us and great conversion rates
- a sane risk strategy
- a unique product with traditional economies of scale/economic moats that allow people who can't afford two mortgages to sell and buy at the same time, so they can move directly once instead of having to do the dance with short term rentals
- huge opportunities for growth from here, both within and across markets
I don't know why liquidity/trading in homes needs to be "all or nothing" to be a good business... it seems to work well enough for cars and other assets
> it seems to work well enough for cars and other assets
Definitely agree. Carmax being a great example.
> I don't know why liquidity/trading in homes needs to be "all or nothing" to be a good business
I didn't necessarily say that you couldn't build a good business, but the author is suggesting that the liquidity that your company creates will have a greater economic impact in that it will create a higher degree of mobility (which is then weighted against a supposed malevolence of investors cashing out big). If you're able to only operate in certain markets (like Carmax) then this theory starts to break down, because it becomes less of a universal truth. I'm simply dubious that the pronounced mobility will be as impactful to the market as suggested.
There are a few points you covered in your comment but the housing meltdown wasn't the result of purely a few large consolidated players. It was the fact that the debt should never have been issued in the first.
A person making $40k/yr is given a mortgage for $400k - that's upside down economics.
So loose debt created a lot of buyers, which then pushed prices up, with little regulation and oversight this created a large push in the housing prices to increase and the cycle lasted about as long as it took for the first massive wave of defaults.
That would have been just a normal reset, the problem was then how much debt everyone was carrying and that all of it was leveraged with not enough collateral to back it up.
If it was purely 1:1 debt to housing, instead of being wrapped up in CDOs that then had additional leverage placed on them several times over with various instruments we wouldn't have ended up in such horrid shape.
Now that's not to say there isn't risk with Opendoor, there is and everyone including the founder acknowledges that risk, but comparing Opendoor to the housing crash is a bit far fetched.
As long as they stay away from leveraged debt and they model things well they can stay a float in a down turn especially since as an organization it looks like they are trying to run a very lean ship by automating as much as possible.
Then when you consider the average transaction price as well as their cut, they don't have to move millions of homes in a calendar year to make serious revenue which means the real question is how long can they carry that debt and how is it serviced in the case of a downturn.
A lot of that also has to do with the markets chosen.
That would be my question. Sure the average buying/price selling price kind of dictates certain markets are out like prime-NYC and SF and secondary markets are in, but it seems that those bear the brunt of the downturn a bit more sharply than NYC.
NYC felt it at the high end, that's properties that were $5MM+, more like $10MM+, but in the low end there wasn't much of a dip. More like a bit of stall in housing price growth, and that was more related to the tightening of mortgages rather than the actual price dropping drastically because NYC still has very high demand. Especially in properties that are well below $1MM.
So there is something to be said about repeating this model in similar markets, but I think moving into a market that is dissimilar, while it won't be scalable, I think can provide some protection in a downturn.
> To succeed, it has to price the homes it buys accurately, without seeing them,
Which is why this will fail. No way you can get that right. Even in a block of identical flats there will be a range of prices: what floor, proximity to lift, aspect and a hundred other little things, not to mention the condition and style of internal fixtures and fittings. Having recently seen a lot of houses, every house is unique.
I once viewed two identical flats in a new block on the same floor right next door to each other, both facing the same way. One ended up selling for 30k more than the other. Why? because the view across town was partially blocked for one of the flats by a building across the street.
I don't think it matters that much. They will offer the lowest their algorithm thinks is correct. And then they send an inspector who where dictates which repairs the seller has to make out of pocket.
And if the home is worth more than what they paid and listed the home for, the market tends to adjust.
Trying to buy a home in NYC, everything I like is generally bid 10-40% above asking. And since open door already paid the seller, they likely pocket the difference, still collecting their 6-12% on top.
With the potential to make 30% on many million dollar homes, they will likely be fine.
Problem is no one in NYC or any other profitable market will use them, because no seller is going to accept 10-40% lower price AND a higher fee in a market where it's already easy to sell.
They already do. People paying all cash saves the seller at least three months. Sometimes the bank won't close for 8 months because of reason and then denies the application forcing the seller to start from the beginning.
People do accept all cash bids for less than asking. People know this, and there are tons of places that will pay all cash for your home right now for less. If you live in a place condo or a house, you will get letters asking to buy the house all cash.
Although I understand your point, the particular example you pick is somewhat easy to fix. Using image recognition it would be fairly straightforward to check if there is a park and what amount of that park could be viewed (maybe using geolocation to check for more images of the park to compare)
I could see OpenDoor working with large companies on relocation outsourcing. An employee needs to move, maybe on condition of hiring, and the guaranteed sale provides the removal of an obstacle to say no I can't move and sell my house... or the discussion really devolves not into schedule but money. If it's not the market rate, the company could pay the difference, or part of it.
For bigger companies, Starbuck, HP(E), IBM, I could see Opendoor simplifying the HR conversations. Especially when companies reorganize and want to move people around.
>Opendoor is betting that there are hundreds of thousands of Americans who value the certainty of a sale over getting the highest price
How does that work in hot markets where sales are quick and certain? (a home in Seattle usually gets 6-10 offers). As a seller you're asking me to accept a lower price and pay a higher fee for little benefit.
Also, how does your model work if the housing market has a downturn? Suddenly all those arbitrage opportunities are underwater. Seems like what you are doing is basically real estate speculation.
I'd want to use OpenDoor if their pricing model came in above what I could reasonably get in the market, and would be more likely to wait it out if they came in low.
I think that most would say that about any kind of offer. In practice, most people want to sell their homes quickly and are nervous about turning away fair offers because they might not see a better one. You might as well be saying that you overvalue your home and never plan to sell, which is unlikely. :)
I don't know if OpenDoor is "the answer," but it bothers me that homeowners are often punished by having to work locally, where their skills are unlikely to remain ideal for the area over many years, or work remotely, which not every career allows and not every personality is suited. Being able to sell a home so quickly is even better than having a month-to-month lease. Presumably OpenDoor's higher fees are something employers could start to pay for as part of relocation (and large employers might negotiate volume rates.)
I'm an economist at Opendoor — yes, adverse selection is a big concern! A few things to note:
1. Every time we improve our valuation model, we limit the scope of adverse selection. A more accurate model means there is less tail in the error distribution, which means we are less like to acquire a home we have mistakenly overvalued. As we improve our model accuracy over time through feature modifications and training on more data, the adverse selection problem will partially self-correct.
2. Every time we reduce our fees, we limit the scope of adverse selection. Improvements to our operational efficiency mean we face lower costs, which over time will allow us to offer fees more on par with or below current market norms. When we offer high fees we run the risk that the only ones accepting those offers are those where we've missed something important about the home. Fee reductions help us limit the scope of this problem, and buy more typical homes.
3. Adverse selection is only possible in the context of asymmetric information. Anything we can do to reduce information asymmetries (i.e. better data collection and vetting up front) will reduce the possibility of adverse selection. We're working on this.
4. Adverse selection isn't a problem limited to institutional buyers like ourselves. Any time someone makes an offer on a home, and that offer is accepted, the buyer risks adverse selection. See for instance the "winner's curse." We can do better than the average buyer at solving this problem so long as we are better at information collection and valuation.
5. Lastly, the main economic prediction of adverse selection is that buyers will take adverse selection risk into account by bidding low, which means that a certain amount of otherwise mutually beneficial trade will fail to occur because buyers bid below their true valuations. To the extent we can reduce information asymmetries and help solve adverse selection, we can truly be a market maker by enabling additional mutual beneficial trade than would happen without us.
Across thousands of homes, we have a pretty good sense of the costs here. Since we do the full home inspection before close, it's not as bad as you might think. We've learned to ask about and look for the obvious things (airplane noise, busy streets, recent crime, terrible new neighbors)
The sellers are required to disclose any issues like this, ex. if the police were forced to be called to their neighbors, etc. Other neighbors in the area would know about "terrible neighbors" so if this came out, they could sue the sellers and more importantly the selling real estate broker.
Presumably whilst any properties are in stasis (i.e. awaiting buyers) OpenDoor could partner with AirBnB and rent out those properties to keep a cash flow from any properties. They'd presumably not do this when they're expecting buyers to look round, but could do it whilst minor maintenance is going on immediately after sale, or if there's a quite time when views are deemed unlikely.
Having the homes be vacant makes it hard, and we optimize heavily for turnaround time, but maybe there's a non-linear innovation we're missing! Open to ideas.
Maybe we could do "sleepover trials" for potential buyers!
Agreed that generally fast turnaround's a good idea were possible.
Would you guys hold on to property in some circumstances though; e.g. if prices are rapidly escalating (e.g. as occurred in the UK around 2000: http://www.housepricecrash.co.uk/indices-nationwide-national... where the expected benefit of holding out outweighs the fluidity and lower risk of the quick scale; or would that be too close to gambling?
I wonder how this can be viable. The closing and transaction fees (agent/title/escrow/closing/excise) make it tough for many home-flippers to overcome and make a profit if no major renovations are done. 2-5% for closing costs on the purchase and sale, as well as agent fees when selling mean they have to gross at least 7-13% on every house just to break even.
The Opendoor model doesn't seem much different than how relocation companies work apart from more directly marketing to homeowners rather than corporate HR departments.
They're a franchise with a network of flippers offering 65 cents on the dollar to people with major issues who are desperate.
We offer 98 cents on the dollar to people with normal homes to sell on their timeline so they can move once when they're ready. (Most people can't afford two mortgages and need to sell before they buy, usually renting in the middle). We're more like a trade-in for cars, or a market maker.
What if I apply to Opendoor, the #1 company I'd like to build software for, but Opendoor turns me down for not being interested, so I end up going with my second or third choice instead?
If someone could look up what their place is worth and, unlike Zillow and Redfin, actually be told explicitly why it's worth that as opposed to being told a "black box" value, you'll open the top of the funnel by driving folks who aren't yet ready to sell (today's owner is tomorrow's seller). And since those values change frequently the not-ready-to-sell owner could track the changes to value and the underlying data driving that value, which would keep you on the tops of their minds when they're ready to sell down the road.
If you think the pricing model/algorithm is too proprietary to share beyond the "I'm ready to sell today" market, I'll stop by for a coffee to try and convince you otherwise.
Good luck. I'm stoked to see someone trying to fundamentally change the residential market.
Edit: oh, btw, not advocating sharing the actual algorithm, but the results of it (e.g. your house is worth 5 psf less than the comp from 3 blocks over bc your place is directly under the airport's flight path).