New start-up flips homes and cars

When Luke Dalien and his family needed to quickly sell their Phoenix-area house, they didn’t turn to a real-estate broker. Instead, they sold the home within two weeks to an Internet startup eager to pay cash.

The startup, OpenDoor Labs Inc., resold the house a month later, pocketing an estimated $20,000. The San Francisco company has repeated this profitable flip scores of times in the past year, catching the eye of Silicon Valley insiders.

OpenDoor is among a small group of startups armed with data scientists and software that believe they can identify the right prices on homes and cars and simplify the sales process online. Shift Technologies Inc., for example, guarantees it can sell people’s cars for a certain price within 60 days, and pays the difference if they sell for less.

But OpenDoor is unique in that it owns all the homes it lists for sale, countering the prevailing trend in Silicon Valley of solely running a marketplace that matches buyers and sellers. As a result, OpenDoor’s strategy is steeped with risk, potentially backfiring if the economy tailspins.

“We’re introducing liquidity to a marketplace that doesn’t have any,” said the company’s co-founder, Keith Rabois, a venture capitalist who was an early executive at payments firms Square Inc. and PayPal Holdings Inc.

That is sort of good news, and you can think of that as a partial response to the housing finance overregulation in some parts of Dodd-Frank.  A second and less obvious lesson is that an economy with higher income inequality is more dependent on leverage in some of its corners, but also can resort to corners and pockets which don’t require much leverage at all.  That creates new potential solutions to liquidity problems, requiring only a minimum of coordination.  That all said, this is also a potential longer-term source of reemergent systemic risk.  No matter how sound this start-up may be, you can see the potential for less well-capitalized and less well-run successor firms.

The full story is here, via the excellent Samir Varma.


Excellent post, deserves more than one comment. Just returned from HK sightseeing: double decker buses are awesome here and better than London's.

Tyler, you should have mentioned that one of the founding team members is a 12 year MR reader! This was awesome to see Opendoor on here.

This article had some inaccuracies and as someone who started reading MR as a college newspaper editor, it was frustrating. We don't flip cars, for instance. Also the reporter's analysis was wrong on our profitability (it's not that high). The reporter wouldn't be corrected on either fact.

Liquidity in the residential real estate market may be a good thing or it may be a bad thing. For most people, the home is by far their most valuable asset. If it is liquid, will owners respond with irrational decisions to take the cash and buy the dream vacation or the dream car. For most people, a home with a mortgage is a form of forced savings. If it is liquid, will fewer owners of them accumulate the "savings" they would have under the forced savings plan. On the other hand, liquidity is the key to avoiding a financial crisis, and that's true at the micro level as well as the macro level. At the macro level, regulators worry about over-extended lenders who can't perform (i.e., their creditors/depositors can't access their funds) in the event of a crisis, causing the entire economy to freeze-up. Economists who are critical of such regulation (Dodd-Frank) believe markets provide a better solution; what they seldom admit, however, is that the market's solution, like old age, ain't for sissies. Cowen, like my church (Episcopal), occupies the middle way, supporting markets on the one hand but acknowledging the pitfalls for us (yes, me) sissies. As for unregulated home liquidity providers (we are all "providers" now), there are a couple of risks: they won't be able to perform in the event of a downturn in the market or, more likely, the "fine print" will shift much of the risk to the home owner.

"If it is liquid, will owners respond with irrational decisions to take the cash and buy the dream vacation or the dream car."

To them, that would be a rational decision. If a society purports to be free why should there be regulation limiting liquidity for the supposedly irrational?

Houses are useless as savings/investment because you can't part with the house. You also realize it conflicts with the policy goal of making housing affordable, right?

People part with the house very often. If they need to access the capital saved up they sell the house and move somewhere smaller or to a cheaper location. Where I live a whole generation reached retirement age and moved out of four bedroom detached houses and into one or two bedroom apartments. They then have lots of cash to spend on doing all the things they'd wanted to do for decades. And they get to live in a low maintenance home after the kids have left.

It's true that homes are a pretty illiquid form of savings - moving house is an enormous hassle. But on the other hand the increase in value of the house has given me a far better return than I'd get from a savings product. And my investment is a comfortable place to live as well.

True, but I maintain that boosting housing costs as a method of inducing forced savings is bad policy. This country's high housing costs just makes life more difficult than it needs to be.

The forced savings is from paying off the mortgage - an increase in the value of the house is just an added bonus. I know it's a foreign concept now, but that's what the last generation did.

If the housing were cheaper you wouldn't need such a big mortgage.

And you say "I know it’s a foreign concept now, but that’s what the last generation did." I disagree: I think people still have the idea that they will profit off of appreciation. But it's fundamentally unsustainable for housing to keep appreciating faster than wages. And total wages are bounded above by the total size of the economy. And a good investment can grow at about the rate the economy grows.

Think about it like this: suppose the homeowner could get the same house for a mortgage for 10% of their income, not 30%. Either way, when they repay the mortgage, they live rent-free. But with the lower costs, they additionally have the money they saved on the mortgage, and even if they didn't because you think they would waste it, they are no worse off.

There are other incentives to buy, regardless of cost. At least in my area of the country (prosperous suburb of major blue-state city), the level of housing quality you get by buying is higher than what you could rent for the same amount of per-month money, for almost any amount of money you can do both for. The fact that buying comes with a 30-year payment guarantee followed by a dropoff to zero (property taxes provide some variability, but less so than rent costs) is just a bonus.

It's a premium for lack of flexibility, because at many of those monthly payment levels you may be unlikely to do more than recoup your investment in real terms, let alone beat any kind of market return or even inflation.

If you want to force people to save it would be far less burdensome if you just require people to put money into retirement accounts (instead of home equity). As noted by Mullainathan, the interest penalty imposed for withdrawing from 401K accounts generally discourages the most reckless behavior and ultimately benefits the saver (rather than banks).

It's not that uncommon for the owner of owner-occupied commercial real estate to sell the real estate to an investment company and then lease it from the new owner.

I suppose something similar might be done with owner-occupied residential real estate, although I suspect transaction costs would make this a good deal less attractive.

Interesting business, though I suspect they have a very difficult business model. Previous attempts to "scale up" business models involving single family homes (such as acquiring a thousand at a time and renting them out) have mostly failed.

There is also the fact that others are already doing this on a local level. Ever seen those signs "We Buy Houses Fast - Any Condition, 7 day closing"? These signs target the desperate, but that's basically what the market is. Those who are not desperate don't mind taking 30-90 days to sell their home and probably getting a higher price. Making a $20,000 profit on one house sounds good, until you consider that not all homes sell this fast, and when they don't, expenses like property taxes, electricity, security, lawn care, etc. can add up fast.

I think a lot of people underestimate just how "local" the market is for single family homes. The small time landlord who knows how to move a permit through city hall and knows which tradesmen and handymen to call when there's a problem will always be more profitable than a big corporation which doesn't have this local knowledge.

The "knowing which handyman to call" factor explains variation in outcomes between small-time landlords, but I have trouble accepting it as a factor in the failure of large landlords. If you scale up to a few hundred units, there is enough maintenance work to support full time employees, at which point it's a personnel management question rather than a local knowledge question.

On the other hand, good vs bad tenants make a large difference in the profitability of a rental operation; and good tenants who know they are good tenants will want to internalize some of those benefits. For organizational and legal reasons, small time landlords have more pricing flexibility than large rental operations where more profitable tenants end up subsidizing less profitable tenants, and moving as they find better options.

Sure, you can scale up to a few hundred units and have full time maintenance employees -- IF all of those units are in one building or one development. Then you can standardize appliances and streamline preventative maintenance and take care of leasing from one central office.

It is a different question entirely if you are going to buy up hundreds of single family houses. In that case, you have a multitude of construction styles and building conditions. Nothing is standardized at all. Plus, you are almost certainly dealing across several municipal jurisdictions perhaps even states. LL/Tenant laws vary greatly from one state to another. Enforcement of building codes and ease of getting construction permits likewise varies widely from one town to the next. I would not recommend purchase single family homes across multiple jurisdictions as a wise investment strategy. You might have SOME luck flipping homes for a while. But a vacant house can just as easily become a magnet for trouble. And in a down market you could be holding onto dozens of vacant homes.

I'll be watching them closely, but I'm not convinced this is a viable business model.

Re: lack of standardisation:

This is a big part of the problem for banks managing REO housing [foreclosures].

In addition to it not being their core business.

Surely NAR lobbyist will be all over this. They want their vig!

Minimum profit margins!

What is the real estate agent fee on a similar sale? Or what has it been historically? What is the rent for three months for such a property against what cost? What are the financing arrangements, duration, rate changes if the property doesn't sell? Who pays the costs of holding the property?

Oooh! $20,000 gross profit from one transaction. You could join us by investing. Call 1-888-xxx-xxxx.

Real estate fee is like 6%, so this is a pretty reasonable cost.

Something similar has been going on for some time in the hotel business. Online travel agencies or wholesalers buy up room-nights in bulk, guaranteeing the hotel a certain price. They then resell the room-nights through various sales channels.

What about adverse selection? The profitability of Opendoor would seem to depend on their ability to predict price. If owners posses information not available to the model, like the "feel" of the house, then we should expect that owners who think the model predicts a price too high will be more likely to sell. This might explain the large six month unsold inventory. .

My first thought was, Silicon Valley has run out of investment ideas.

They're sitting on 156 properties in Phoenix as of this minute on their website.

Prediction: bankrupt by spring.

They just raised $80M. I don't think 156 houses will take them down.

Let me add to the sentiment that this is a way to avoid the cumbersome and expensive traditional broker listing. Much like Uber tried to free cabbies from medallions, real estate is ripe.

The used car market is already fairly free. People list on eBay and Craigslist. Not sure how this belps an already liquid market.

I don't see the value to people selling their homes to these guys. They're still charging 7-12% in fees (which include 2% for 'market risk'!). The only thing they're changing is guaranteed sale, but of course that's only on houses they think are easy to sell, which sellers are likely to be aware of as well. It only really makes sense if you need to sell ASAP.

I suspect these guys are smart enough to internalize the extra-normal profits associated with the banks, title companies and mortgage taxes. I can only guess at how the business is structured but I don't imagine that they are going for a government-backed mortgage for capital. There are listings turning up that require a buy to use the sellers title company. Lower transactions costs go a long way to increasing liquidity and are rarely examined.

"[T]his is also a potential longer-term source of reemergent systemic risk"--risk of systemic *illiquidity,* that is. But isn't the Fed supposed to counter any such risk? Then the real risk is that the Fed won't do its job.

This is much like a hedge fund, in that their main value add is that they have better models for actual value. This seems to indicate that there could be some future for the business.

However, unlike hedge funds, their offer isn't anonymous. If, over time, it becomes apparent that OpenDoor does has a better model of price then people will get offer from them and then turn around and ask ~8% more from the market.

While I'm sure that they have thought this through, I just don't see how they can make it work.

Shiller maintains that the housing market is not efficient, making this model one which has risk.

Here is a piece on Shiller;s analysis of the housing market:

True, and I saw a graphic the other day that US housing as a function of income is back to where it was in 2007, just before the crash. Real estate has not been deleveraged in the USA.

“We’re introducing liquidity to a marketplace that doesn’t have any,” said the company’s co-founder, Keith Rabois,"

No they're not.

Home equity loans have provided liquidity for homeowners for years. And you can get a home equity line of credit without the need to move out of your house.

what could go wrong?

Highest ROI I ever had was a beach house in a crap town in florida I bought right after Obama's NFIP changes.

Nice little 3/2 a half a block from the water for 40k.

Put 10k into it and sold it 18 months later for 110k.

How much of the profit is due to:
1) Saving on at least half the broker's fees (2-3% of sales price per broker). I assume the initial sale has no brokers and the second one at most involves one.
2) Faster transactions through an open market?

Seems like these two of these could explain the entire profit without any need for "data science." It's not hard to infer the proper price of a home given its neighbors and the free data from Zillow...

OpenDoor is an interesting finance play, but not for the reasons listed. It is a group using their tech credibility to raise equity financing cheaply in an attempt to build a finance company.

That is a surprisingly good hypothesis.

Comments for this post are closed