Hacker News Reader

Comment by mertd | Hacker News Reader

The idea was not house flipping at scale. They wanted to "market make" for housing. Housing market is illiquid. People would leave decent money on the table if they could just sell their houses within hours without having to deal with realtors, open houses, listing and closing process. Of course Zillow would have made some repairs and improvements but I don't believe that they thought this was their upside.

View on HN

Exactly. People sell cars to CarMax for less than they could otherwise get because it's easy, fast, and certain. People buy cars from Carvana over the lowest price they could pay elsewhere because it's easy, fast, and certain.

I think there's a lot of money to be earned by turning a slow, tense process into a fast, lower stress process.

What market making opportunities do you see that aren’t currently aggressively competed for?

I'm no expert, but as a homeowner and someone who as a result has to employ contractors to fix things, I'd love to see someone make home repairs easier, more certain, and lower aggravation.

I think there are probably thousands of other industries/segments where an aggregate hundreds of millions or billions of dollars are spent annually and where people would spend 1 to 2.5% extra if friction were noticeably reduced. That used to be flights, hotels, and cabs that have been made easier in the last 2 decades.

> I'm no expert, but as a homeowner and someone who as a result has to employ contractors to fix things, I'd love to see someone make home repairs easier, more certain, and lower aggravation.

While it's definitely possible to improve the experience around that, I don't think it really qualifies as market making. Specifially, look at wikipedia's definition:

>A market maker or liquidity provider is a company or an individual that quotes both a buy and a sell price in a tradable asset held in inventory, hoping to make a profit on the bid–ask spread, or turn.

You can't really hold "renovations to be done" as an "asset in inventory". It'll be something like uber, ie. a marketplace, not a market maker.

That's fair and I agree that it's friction-reducing but not market-making.

Except for really basic stuff like snaking a toilet, home repairs and remodeling are all different. It's not a standardized commodity service like flights, hotels, and cabs where the majority of customers all want pretty much the same thing.

All home repairs - it's a damn nightmare. Home building - complex, slow and expensive RE agencies, seem to compete... but it's an oligopoly market. (just to stay in the RE segment)

And all three have severe government monopoly powers and competition killing regulations.

Some would say the real estate brokerage market isn't that competitve. The national realtors association keeps prices artificially high, there is room for ibuyers or market makers to offer a better service, and charge less than 6%

Except, it never works that way for houses. Banks are reluctant to give mortgages when the roof is EOL, or the siding has dry rot or there is some black mold basically anywhere, etc. Giant Corp may be able to buy quickly and efficiently but Joe Homeowner can't get his loan approved until A, B, and C are fixed. Good luck getting 3 different contractors lined up quickly and under budget while in escrow.

> Banks are reluctant to give mortgages when the roof is EOL, or the siding has dry rot or there is some black mold basically anywhere, etc.

but used cars seem to do fine, even though they can be similarly damaged?

Used cars loans tend to have higher interest rates though.

Maybe because used cars cost/risk about 1% as much as houses?

Also, every car has scrap value. Houses don't have that.

The equivalent for houses is the land they sit on. In metros that could easily be worth more than the house, and in most cases it's worth more (in % terms) than the scrap value of a car.

Why? because the unit price is lower? Can't zillow just... buy less houses? I don't see anything different between zillow buying a $500k house and one hundred used cars worth $5000 each.

Easier to sell 50 cars than 1/2 of a house tho.

While I don't doubt this effect exists, I seriously doubt it's a significant factor when you're dealing with 7000 houses.

7000 figure from: https://news.ycombinator.com/item?id=29081118

Used cars are more fungible and have a higher volume of comparable sales constantly underway.

Repair costs are also better known and less daunting.

Do banks even look at these things; I mean, beyond the appraisal?

I don't think my mortgage lender has ever inquired about the age of the roof on the purchase or during any refinances. For my first house, I bought it with a roof that needed work, had some small leaks, and was well beyond its design life. I got a mortgage without the bank bringing that issue up (albeit in 1996).

Home inspectors look at these things and buyers might choose to walk away based on the inspection report (which the bank never saw on either of my purchases).

Like seemingly everything in the modern world, it depends on your credit worthiness. Some mortgages also come partnered with homeowners insurance, and those policies are what often lead to sales being blocked by even minor issues.

In my experience, the level of underwriting is a function of how levered the purchase is. The LTV and your Debt to Income are both major factors in how thoroughly the underwriter is going to review your mortgage and how much they will require to prevent a loss.

That makes sense. Both times I was near the 25% front-end ratio that I considered to be a sensible limit, but was putting 20% down on a conventional conforming loan the first time and a conventional jumbo the second time.

From what I've seen it depends a lot on the kind of loan. I bought a house with a conventional loan earlier this year and I don't think the appraisal even looked at any of the potential issues (Despite at least a few thousand in repairs identified by the inspection). But I also know someone who used an FHA loan to buy a house and they couldn't get the loan unless the house passed a pretty detailed inspection.

FHA loans are 0-5% down - conventional are 20%. 20% covers a lot of repairs if they have to repossess and sell at a loss; 0% covers way less.

My mortgage was contingent on getting insurance and every place offering a decent insurance premium wanted to at least know the condition of the roof. I live in tornado alley though with lots of major hail events so roofing is kind of a big deal. The sellers of the house had a previous contract fail because the buyer failed to get a mortgage approved because of the bad roof, in the end we made a deal with the seller to adjust the sales price a bit and get a new roof on the house (a bit of give and take, we really loved the house otherwise).

I wonder if this is location dependent, because my bank barely cared about those things when they approved my refinance last year. The appraisal consisted of just driving by the house. They didn’t check for any condition of the house.

It probably helps that the entire house could just disappear and the price for the lot would only change 10% or so.

That's probably due to a favorable LTV. If they have to foreclose, not only would the bank not lose money it might actually be profitable!

Also the downsides are vastly different between buying a lemon of a car (although I’m those two cases I bet the companies offer a money back guarantee) that you might lose $1-2k on in repairs vs tens or hundreds of thousands of dollars in buying a bad house.

I think you're probably right. And in order to bootstrap the necessary volume (and starve their competitors), they were willing to actually pay more than market (similar to Uber's investor-subsidized rides in their early years). So the problem isn't so much that they lost money (which they probably expected for the first phase of the program), but that they lost more money than expected (and could tolerate). The additional cost and carrying time associated with labor and supply shortages in the construction industry may have contributed in that case, though I agree with the article that the adversarial market aspect was probably the biggest factor: in a competitive market, particularly bad offers were particularly likely to be accepted.

>The additional cost and carrying time associated with labor and supply shortages in the construction industry may have contributed in that case

And property taxes. And security. I do not know who ran the numbers on this program, but anyone with even a little bit of real estate experience was wondering what the hell they were doing. Blackstone has been around and killing it in real estate for 20+ years, and you would figure something obvious like what Zillow was planning to do would be on their radar if it made sense.

Market makers also typically make money on their spread, paying slightly under market rate for an asset and selling for slightly over. They can do this in, say, stock trading because people want to make trades right now to capitalize on moment-to-moment conditions, so that's the price of convenience and liquidity.

Zillow SAID they wanted to be a market-maker, but they ACTED more like a flipper or investor, expecting to make a profit on the houses themselves, not just on the spread. When they started making above-market offers on homes they pretty much left any pretense of market-making behind: at that point, you need to either sell at a loss or hold inventory and go (a little) long on your investment. Zillow realized they didn't want to hold inventory.

It doesn't help that houses aren't nearly as liquid or fungible as stocks: buying a house, prepping it for sale, accepting and offer, and closing takes months, and all that time can eat into your spread. Zillow would probably always have ended up in a situation where they were more concerned with the sale price 2-4 months down the line rather than their spread.

In the bay area there seems to be some companies that do this. Every time I turn on the radio there is an ad for "John buys bay area houses".

I think they do exactly what you say, give a quick cheap offer, then do some fixing and sell the normal way for a profit.

Basically, if you can convince someone to sell for x% less than the real price today, you can make x% - some costs. Of course this has huge variability, but the company I noticed do it in one market they probably know extremely well so they can make good bets.

I inquired about selling to a few in a different region. They all expected you to pay for repairs they identify. That's far too open ended of an agreement for me. Perhaps it was limited to structural issues, but that wasn't clearly outlined. I also assume they complete their own repairs using their own staff. I expect that gives them the opportunity to quote me at $1000 for a repair while their cost is lower.

In a red hot market, I opted to let the market decide the price and terms.

I think if you can it is almost always better to do it on the open market.

The exception being that there is a lot of middleman involved, loan agents and purchase agents. So cutting out these should save at least 5%.

Reply on news.ycombinator.com