Hacker News Reader

Comment by currymj | Hacker News Reader

Algorithmic pricing that learns to tacitly collude is a hot area of study in computer science and economics. For example, if you train simple online learning algorithms to adjust prices, sometimes they can learn to keep prices high or to take turns winning customers, rather than just competing. People have found some empirical evidence of this on platforms like Amazon where a lot of small sellers use pricing bots.

However, it seems this is a more of a hybrid situation. A big part of the complaint is just all these incriminating emails and documents where RealPage appears to be coaching landlords to avoid lowering rents or giving concessions, independent of the software.

At the same time, there was an algorithmic component, which the customers appreciated: “I always liked this product because your algorithm uses proprietary data from other subscribers to suggest rents and term. That’s classic price fixing...”

View on HN

This gets into prisoner's dilemma though. If everyone but me is using the price fixing app, I have a strong incentive to undercut them, even by just a few dollars. So without a cartel-like enforcement mechanism, there is no reason it wouldn't just fall apart naturally in the long run.

In the complaint there was a mention of "compliance" which could get into that piece though.

>I have a strong incentive to undercut them

Not in a supply constrained market where your rental will be occupied either way.

If your rental is $2400, and it sits on the market for 2 months, you lost potentially lost $4800. It would take a huge rent increase to justify that.

It's one thing if the software is helping landlords jack up prices to the market rate. It's quite another to convince them to collude against their best interests.

It is a different unit every month though as renters are moving out all the time. This isn't about 1 unit that is empty or not. It is about 100+ units where it can be 89,90, or 91 empty - if the other units that are not rented pay enough higher rent because the empty is not on the market you are better off.

Sounds like removing the supply constraints would be a better strategy.

And supply constraints come from regulations on housing, zoning, and federal subsidies for "homeowners"

If all units are being rented there is no collusion as the market clearing price is being charged. Collusion requires taking supply off the market in order to raise prices.

> Collusion requires taking supply off the market in order to raise prices.

This is factually incorrect.

It is not possible to raise prices without lowering supply if the seller is already trying to maximize profit. Sure the seller could be undercutting the market but that has nothing to do with collusion. Apartments rent in an auction, each fills for the most anyone will rent it for. In order to raise prices landlords need to increase the competition for each unit.

I think you’re presuming elastic supply, where you can make up the lost profit from undercutting by selling more units.

Real estate supply isn’t very elastic, and demand already outpaces supply. Your units will likely get rented, as long as you aren’t in the top 0.1% of prices.

Therefore your incentive is to maximize profit per unit since you can’t move more units. Your incentive is to also join in on price fixing, because it’s the only way to make more money.

Supply outpacing demand would cause this to dissolve, as landlords actually have to compete for tenants and the highest priced landlords have empty apartments.

For the landlord units is elastic. If you have many units you can always not rent out a few, in fact you should always have a few units that you are remodeling and thus cannot rent out.

In general as a large landlord should have around 10% of units not rented, if you have more than that rented you should raise your rates until people go elsewhere thus bringing you down to 10%, while if you have less than 10% lower your rates until people start renting from you. Different landlords have different numbers, but 10% is a good starting place. 100 units at $900/month = 90,000, 90 units at $1000/month = $90,000, but you have a few units free in case someone desperate is willing to pay $1100/month (and of course you can remodel one of those empty units thus making it more desirable)

Supply was probably a poor term because it’s overloaded; I probably should have said stock. They cannot easily scale their stock of housing, so they can’t make up for lower profit with volume the way a grocery store or something might.

Just a note on vacancy targets, I worked for a MF REIT (~33k units) and pre-YieldStar their average target occupancy was around 97% for properties. After YieldStar was implemented the average dropped to more like 95%. As far as I'm aware, the other large managers also targeted the mid-to-high 90s.

90% is actually quite healthy. If all housing was always full, there is no slack in the system. We should always have some overcapacity in the housing market, just like we do for hospital beds, food, water, etc. If you're always buying the last loaf of bread at the grocery store, that means others don't get bread when they want to buy it.

Right, assuming there’s more supply than demand. In markets with low vacancy rates like Manhattan, landlords get away with murder (terrible maintenance, making renters pay for gatekeeping rental agents, high rent despite being shitholes, etc). In markets with higher vacancy rates, prices tend to be a lot more reasonable, and units tend to be somewhat better maintained, because they don’t have the same pricing power, and if they’re too bad, they won’t be rented unless they’re extremely cheap.

Interesting as airlines have been doing this for a really long time.

Airlines do this internally all the time. However they do not share this information with each other.

there is concern about tacit collusion if all the airlines start using the same pricing algorithms based on the same historical data. it doesn't even take a very smart pricing strategy for them to naturally learn to collude with each other.

However I would guess you will never have an email from an airline CEO that says "I really like this product...That's classic price fixing"

They don't share how many seats are left publicly, but fare changes are shared real time in a platform that distributes them. And there's quite a lot of website scraping that goes on that mostly skirts around "how many seats are left".

not everyone is affected by airlines' pricing though as not everyone travels. everyone needs a place to live though

Reply on news.ycombinator.com