Thinking Zillow Flipped Homes Means You Aren't Thinking

[Bill F.]

A few active threads popped up on BP with the consensus being something like “hahahah Zillow lost a lot of money trying to flip houses…How dumb are they, lets all point and laugh’. Or lets make memes about them, because its 2021 and that’s what you do now.

That’s one explanation of some of the facts.

Before I get into how I look at it, we need to get on the same page with what flipping is. From Brandon Turner’s article “Flipping Face-Off: Is It Really Better Than Renting Out Properties?”

“Flipping houses means buying a property at a discounted price, improving it and then selling it for a profit.”

Now what was Zillow Offers? From their 2020 10-k:

We launched Zillow Offers in April 2018 to provide homeowners with the ability to receive cash offers to purchase their home, giving sellers peace of mind, control and convenience in one of the most stressful transactions of their lives. … When we buy a home from a seller, our title and escrow business, Zillow Closing Services, performs their due diligence for a clean title and a seamless close of the home transaction. Then, our renovation teams perform light, make-ready repairs to swiftly list the home…Home buyers are also able to purchase homes that are listed for resale through Zillow Offers.

It seems that ZO didn’t want to flip homes at, but what were they trying to do? Their CEO tells us in their Third Quarter earnings call:

…our aim was to become a market maker, not a market risk taker. And this was underpinned by the need to forecast the price of homes accurately three to six months into the future. We used historical data and countless simulations to test this belief. We set unit economic targets that required us to stay within plus or minus 200 basis points in breakeven…

Now that begs the question, what is a market maker? A market maker provides easy transfer of an asset for buyers and sells. They buy for a bit less than fair market value and sell at a bit more than fair market value and profit off the spread while the buyer and seller are willing to get less/ pay more for the service of someone always being there to buy/sell whenever they want. If you read at all about Robinhood a while ago the company that they sent all their orders to, Citadel, is one of the most well known market markers.

There are clearly some issue with doing a copy paste of how Citadel operates and applying to single family homes. A home costs hundreds of thousands of dollars, every home is different, you can’t raise your family inside a share of Tesla ect. But it happens in the bond market too, and bonds are more similar to homes than stocks, they trade in large blocks, some aren’t super liquid, but still, bonds don’t have leaky roofs and you can buy and sell them way faster than you can buy and sell a home. Citadel doesn’t have to care about predicting stock prices three months or three days from now, but if you want to do this with homes you have to have some ability to forecast home prices now, when you buy, or in the future, when you sell. They are kinda market making, but more like leveraged house trading.

That cuts to the core of how and why Zillow did what they did. If you want to connect a buyer and seller you have two main ways of doing it, principally or via an agent. Agent trading is what we have now, the seller contracts someone to sell the home and gives them a cut. Principal traders will buy the home from the seller and taking on the risk of owning the asset until they can find a buyer. For ZO, the risk is offset since they have their own title company, their own mortgage origination arm, plus they get the data, all of which generate revenue above the spread.

It also makes the process for the buyer and seller simpler and people will pay for simple. Instead of a buyer going to a bunch of open houses, putting in offers at different times, dealing with different listing agents, you go to Zillow and say, ‘I want a 3/2 single story home within 20 miles of here.’ Zillow spits out whatever they have in inventory. For a seller, you don’t have an open house three weekends in a row, where you can’t be home and have to clean all your stuff only to have the deal fall through because of a buyer who wanted to get a new ATV while they were under contract and the local Polaris dealer ran a hard hit on their credit and screwed up the mortgage underwriting process [that’s 100% a made up situation and hasn’t happened to me… oh wait]. You go to Zillow and say ‘ I want to move out three weeks from now’ and they say ‘great, what account do you want the money in?’

Not saying this is for everyone, but there we have to admit, there is some value in it for some people.

Since we know they didn’t want to flip, but rather do this leverage home trading, we have to look at 1, the validity of the idea, which we did above and found out, it probably works for some people and 2. How did Zillow do executing on the idea?

We see that they didn't close down bc they lost money this quarter. It was because they weren't hitting their targets, ie. +/- 2% of breakeven relative to some benchmark. That write down put them at like 7%. They shut down not because they lost money; they also made a bunch of $ in the first half of this year, but because the wide swings in earnings put the rest of their lines of business at risk. Their buying model wasn't accurate enough for their risk tolerances. At the end of the day Zillow can spend their money in areas with higher ROI, like mortgage origination or their Internet Media and Technology areas.

But does that mean this idea is dead? I don’t think so. Opendoor and Offerpads still offer basically the same service as ZO. Plus, they have two other advantages, 1. This is all they do. All their eggs are in this leverage house trading basket so they have no other distractions to worry about. If they mess this up, they go broke. That tends to focus the mind. That feeds into the second advantage: willingness to take more risk and their shareholders get this, since they only do this one thing, they don’t have to worry about messing up other parts of the business.

The question comes down to: can you make a model that, over a large enough sample size, be no lower than 2% of purchase price quarter? If you have enough of the right kind of capital [long term oriented and willing to ride out the ups and downs of a bad quarter or year] then you can't help but make money since home prices generally go up and to the right.

The first iteration of innovation is often clunky and worse than the status quo. The nature of evolution means that it doesn't stay like that for long and we know that it can only get better from here. I wouldn't call buying 7000 home in roughly a year and being within 7% of their fair market value an epic fail.

There is no reason for any individual to have a computer in their home. - Ken Olson