[HN Gopher] Price fixing by algorithm is still price fixing
___________________________________________________________________
Price fixing by algorithm is still price fixing
Author : nabla9
Score : 142 points
Date : 2024-03-02 21:10 UTC (1 hours ago)
(HTM) web link (www.ftc.gov)
(TXT) w3m dump (www.ftc.gov)
| csa wrote:
| I would like to see that this action will lead to meaningful
| change in (imho) an unhealthy rental market.
|
| Unfortunately, I think the genie is out of the bottle, and the
| actions by the ftc will be reduced to whack-a-mole attempts to
| bust transgressors who make the effort to have plausible
| deniability while still using price-fixing algos.
| nurtbo wrote:
| Adding to that: once you've shown that keeping rates high while
| allowing greater vacancies works, no one is going to lower
| rates afterwards. So we are stuck with this higher rent
| environment.
| josh-sematic wrote:
| IIUC it's still valid to use an algorithm to recommend prices,
| but you're not allowed to make an agreement with a competitor
| that you will both use the same alg? Or is this saying that any
| algorithm which takes competitor prices into account is out-of-
| bounds?
| spamizbad wrote:
| The former (you can't collude with a competitor to use the same
| algorithm to set prices)
| bobthepanda wrote:
| There were antitrust suits filed against RealPage after a
| ProPublica piece
|
| > In one news release, Realpage offered its property management
| clients the ability to outsource daily rent-setting and revenue
| oversight. "We believe in overseeing properties as though we
| own them ourselves," the company said in a presentation that
| plaintiffs' lawyers referenced in the lawsuit.
|
| > The lawsuit quoted one unnamed witness, a RealPage pricing
| advisor, saying that some pricing advisors told property
| management employees that they had to follow the software's
| recommendations. A leasing manager at a RealPage client said,
| "I knew [RealPage's prices] were way too high, but [RealPage]
| barely budged" when the manager asked to deviate from the
| suggested rent.
|
| > An update to the software tracked not only clients'
| acceptance rate, but also the identity of the landlords' staff
| members who had requested a deviation from RealPage's price,
| the lawsuit said. Compensation for some property management
| personnel was even tied to compliance with the company's
| recommendations, it said.
|
| So if this is true, this also means that managers are being
| compelled to adopt the recommendations more than as mere
| suggestions.
|
| https://www.propublica.org/article/doj-backs-tenants-price-f...
| __s wrote:
| Reading linked doc, sounds like first
|
| But yeah, your latter case sounds easy to fall into gray
| territory, since it seems illegal to agree on prices even if
| intermediary makes it unclear with whom exactly you're agreeing
| with
| advisedwang wrote:
| The law doesn't work on hard and fast rules like that. It's
| more like "If the effect is that you are working together to
| set prices, that's illegal", regardless of whether you use
| algorithm A or B, regardless of whether you share prices in
| advance or watch how other companies set prices, regardless of
| any details of the mechanism.
|
| If you use the same algorithm it's not necessarily illegal, but
| it will be if it results in price fixing. If you take into
| account competitors prices its not necessarily illegal, but it
| will be if it results in price fixing.
| kirse wrote:
| After pricing auto insurance recently it's pretty obvious this is
| happening in that industry as well. While shopping around
| multiple quotes across 7-8 providers and calling at least 5
| separate insurance agents to try to gather quotes, all of these
| companies are providing similar quotes within a few cents/dollars
| of each other.
|
| I vent my frustration to a few agents about the yearly rate
| increase insanity and they all shrug, give their non-empathetic
| "I understand" telephone script and blame it on the "system"
| calculating the prices and make some useless excuse about
| inflation.
|
| I've got a clean driving record, a fully paid-off cheap vehicle,
| in a reasonably responsible age bracket, and the cost of decent
| auto insurance these days is essentially another car payment.
| Within 5 years I'll have paid back the insurance company 60-70%
| the value of the vehicle. The Gov/FTC needs to take a look at
| these companies, especially if they're forcing us to hold the
| insurance to reasonably participate in society.
| tpmoney wrote:
| Aren't most insurance rates / rate ranges set by the states?
| And any further variance is down to the actuarial tables? It
| seems pretty reasonable that most of your quotes would be
| within a few dollars of each other because they're all insuring
| the same risk, in the same location, under the same legal
| framework.
| kirse wrote:
| _Aren 't most insurance rates / rate ranges set by the
| states? And any further variance is down to the actuarial
| tables?_
|
| If that's true, I certainly couldn't find a table on
| allowable rate ranges when I did some basic research on
| pricing and what factors influence it. Certainly open to
| being schooled on how auto insurance works.
| iudusuux wrote:
| Not quite right. Auto insurance is a commodity. They don't have
| price fixing per the same way housing does. Also they are
| heavily regulated and their prices probably have to be approved
| by your state.
| Ekaros wrote:
| I wonder if they actually have that much margin to give
| discounts... Or are the prices already at or near lowest
| possible level... After all outgoings must be lower than
| incoming in on sufficiently long term. Has something pushed the
| pay outs too high, compared to what is being paid.
| jethro_tell wrote:
| I'd expect the cost of cars, which is has become tethered to
| 'monthly payment' instead of actual cost.
| sroussey wrote:
| Insurance is based on the cars you may hit that are not your
| own. Prices have been going up. So not surprised.
| seanmcdirmid wrote:
| Liability insurance is based on other people's cars.
| Comprehensive and collision is based on your own car.
| xapata wrote:
| Liability is more about medical costs, no? If it were
| bounded to vehicle cost, I'd self-insure.
| seanmcdirmid wrote:
| I'm sure car insurers pay out significantly more for
| vehicle damage than person damage. Most accidents don't
| involve harm, but body shops are getting expensive. For
| WA:
|
| > In 2019, there were 45,524 reported car accidents in
| Washington State. Although 32,106 resulted in no
| injuries, 325 were fatal and 973 resulted in serious
| injuries.
|
| https://www.weierlaw.com/dealing-with-an-auto-accident-
| in-wa....
| Scoundreller wrote:
| > Liability insurance is based on other people's cars.
|
| This is what I dislike about insurance. If someone hits my
| (hypothetical) $100k car and ruins it, I win $100k, but if
| someone hits my $1k car, I only win $1k.
|
| Yet the person that hit me did precisely the same
| action/error.
| dylan604 wrote:
| The price to repair cars are also going up which is making
| the payout claims cutting into profit margins. So the
| solution is obvious
| akira2501 wrote:
| Insurance is based on the coverage I purchase. Are you saying
| the insurance company is instead selling me an unlimited
| amount of liability based on the price of _other_ cars and
| not the $50k of coverage that I selected when I bought the
| policy?
|
| How is that justified? $50k is $50k.
| ensignavenger wrote:
| As another anecdote, I get very different quotes for auto
| insurance from different companies.
| mschuster91 wrote:
| > I vent my frustration to a few agents about the yearly rate
| increase insanity and they all shrug, give their non-empathetic
| "I understand" telephone script and blame it on the "system"
| calculating the prices and make some useless excuse about
| inflation.
|
| Thing is, they're not wrong. The cost of accident coverage has
| gone up, actually way beyond inflation - assume you hit a Tesla
| and it sits around 9 months until Tesla can be arsed to get
| spare parts, your insurance will be billed for the damage
| itself as well as a loaner car for the counterparty. And damage
| repairs themselves have gotten more expensive as well: what
| used to be a simple bend that your everyday farmer neighbour
| could fix with the basic tools in his garage all while being
| drunk out of his mind isn't even possible with modern cars made
| from aluminium or carbon-fiber composite, not to mention all
| the tech like distance sensors that go into modern fenders
| which has to be replaced and carefully recalibrated.
|
| On top of _that_ come all the issues with regular inflation
| (e.g. labor cost, real estate rental for shops) and the
| aftereffects of the covid pandemic and its supply chain shocks
| (there 's _still_ a massive number of car carcasses that couldn
| 't be completed and now get priority in parts delivery).
|
| [1] https://www.carscoops.com/2023/11/tesla-owners-stuck-
| waiting...
| dools wrote:
| I wouldn't be surprised if this was caused by concentration in
| the under writing market. You might find a lot of insurance
| companies at a thin retail layer on top of a very small
| wholesale layer.
| mint2 wrote:
| The cost is due to the increasing repair costs to other people
| vehicles, not mainly your own. You also don't have to get
| collision and comprehensive coverage, or at least not with a
| low deductible. Then you're really just paying for the damage
| you cause to others and in that case your vehicle cost doesn't
| mean anything at all.
|
| Also, not sure your state or credit score but if you're not in
| CA you'll need good credit to get good rates. The only way to
| change that is government regulation.
|
| Also if you're getting the same rates from different places, it
| sounds like you're being quoted the same company not different
| companies. If you aren't going directly to the actual insurance
| companies website, that's what's happening.
| standardUser wrote:
| > Rent is up nearly 20% since 2020, with the largest increases
| concentrated on lower- and middle-tier apartments rented by
| lower-income consumers. About half of renters now pay more than
| 30% of their income in rent and utilities, and rising shelter
| costs were responsible for over two-thirds of January inflation.
|
| We need better rent stabilization laws. Not the ones we have in
| many big cities that result in perpetually skewed markets and
| $200/mo rents. But a more comprehensive approach including
| requiring multi-year lease options, 12-month or greater rent
| increase notice requirements, and rent increase caps that prevent
| catastrophic rent hikes but still allow units to align with
| market rates over longer timeframes.
|
| Unfortunately, people prefer to strawman the entire concept by
| pretending that the only "rent control" laws that can exist are
| like those in NY or SF.
| 58028641 wrote:
| Doesn't rent control just reward people lucky enough to rent
| before prices increased? Wouldn't increasing supply be a better
| way to reduce prices?
| seanmcdirmid wrote:
| Ya, most rent control just rewards those that are there when
| it happens, if it applies to new renters, supply eventually
| seizes up.
|
| I'm not sure what model parent thinks will work. Maybe rent
| control light where price increases are limited but the limit
| is highish? I can see that working out, limiting increases to
| 1.5-2X inflation might keep the market fluid and provide some
| stability to renters without locking new tenants out.
| standardUser wrote:
| "Rent control" can do any numbers of thing - it really
| depends on the specific policies.
| Ekaros wrote:
| 12 months or more to increase rent seems counter-productive. As
| it would mean that what makes most sense for landlord is to
| just increase by maximum each and every time as they cannot
| forecast changes over such long period.
|
| Still, capping increases to something like inflation+2%, and
| making leases continuous with long termination periods would be
| entirely sensible fixes.
| lelandbatey wrote:
| If inflation+2% is an acceptable rent change, I don't
| understand how needing 12 months or more to increase rent is
| a problem. The only thing a shorter time period allows is a
| landlord to say "well, over this past 3 months we saw
| inflation change by +-0% but 0+2 is still 2%, so we're
| increase rent 2% in the next 3 months period".
|
| And that doesn't seem like a good possibility.
| Ekaros wrote:
| 12 months seems rather extreme range to me. So what would
| happen is that you rent a place, day after you move in you
| get notice that yes we are going to increase rent by
| maximum in a year. After all this is the most logical and
| effective time to do it. Now cost might not increase as
| much in year from that date, but landlord does not know.
| Come next year, they again also don't know, so again max
| allowed at time.
|
| But instead, if you can notice rent increase let's say 1
| month before year is full, they might look at what are
| current cost and think oh, I could do with lower increase.
| standardUser wrote:
| One option is to require multi-year lease options. Most
| commercial leases work that way. That creates a lot of
| stability and predictability for the family, just as it
| does for businesses.
| Ekaros wrote:
| My basis is continuous contract with let's say 3 or 6
| month clause of termination under limited reasons. Like
| purchase or taking unit to use by family. Or something
| like large renovation...
| standardUser wrote:
| I think families should have as much time as possible to
| handle the significant costs and disruptions associated with
| moving homes.
| justinclift wrote:
| Why "inflation+2%" rather than just inflation?
| Ekaros wrote:
| To allow some leeway for increase in property value and
| demand, in general lot of money looks that level of
| returns. Also inflation does not always full reflect the
| costs.
|
| Also something like 2% above inflation does not lead to
| unreasonable increases. Until market rates are reached.
| Workaccount2 wrote:
| The only fix for high housing costs is to build more housing or
| make your area shitty (so people want to leave). The latter
| isn't popular so that leaves only one option.
|
| You _cannot_ legislate away costs for an in demand product. It
| can 't be done. Put strict rent increase laws in place and all
| you do is make it so people sit in their apartment for years to
| lock in that low rent.
| standardUser wrote:
| Did you read my comment? It is literally about how "strict
| rent increase laws" are not the solution.
| arcticbull wrote:
| While true, if supply and demand are allowed to meet
| through new construction you won't need rent stabilization
| laws.
| NovemberWhiskey wrote:
| Everything is up 20% since 2020; but lower-income wages are up
| more, FWIW.
| patrickmay wrote:
| Rent control has been repeatedly demonstrated to increase
| prices and lower the quality of the units under control. The
| solution to high housing prices is reducing zoning
| restrictions, not increasing government intervention.
| ilikehurdles wrote:
| > an agreement to use shared pricing recommendations, lists,
| calculations, or algorithms can still be unlawful even where co-
| conspirators retain some pricing discretion or cheat on the
| agreement.
|
| Someone explain how this is different from pricing vehicles based
| on KBB or other data? Genuinely curious because I don't know what
| these agreements look like.
|
| If multiple seller/landlord parties all agree to have a single
| algorithm set prices I understand the FTC's point. If however
| they all reference an algorithmic data point and freely choose to
| set prices I don't see how that's collusion.
| rahimnathwani wrote:
| If multiple seller/landlord parties all agree to have a single
| algorithm set prices I understand the FTC's point. If however
| they all reference an algorithmic data point and freely choose
| to set prices I don't see how that's collusion.
|
| You've hit the nail on the head. It's the former, which is why
| it's collusion.
| tyingq wrote:
| I imagine it's tricky to really prove. Airlines, for example,
| have been doing somewhat similar things for decades, but
| there's no simplistic path to showing collusion.
| ilikehurdles wrote:
| But based on the post, this really seems to not be the case?
|
| "Price deviations don't immunize conspirators. Some things in
| life might require perfection, but price-fixing arrangements
| aren't one of them. Just because a software recommends rather
| than determines a price doesn't mean it's legal. Setting
| initial starting prices or recommending initial starting
| prices can be illegal, even if conspirators deviate from
| recommended prices."
|
| Would be better to see an argument from the FTC grounded in
| evidence and analysis, because I don't get what this is based
| on. Software recommends a price, sure, but so does any
| appraisal in any industry. For how long can I use an
| appraisal service before the FTC says I'm committing a crime?
| I just feel like the lines are not drawn very clearly with
| this argument.
|
| Aside, if anyone from the FTC is here right now, for God's
| sake do not use "IRL" in official writing.
| usui wrote:
| > Such software can allow landlords to collude on pricing by
| using an algorithm--something the law doesn't allow IRL.
|
| There are probably other things I should comment on first
| instead, but I must say this usage of "IRL" made me laugh. I grew
| up with the initialism IRL while chatting online so when I
| emailed a teacher where I accidentally used "IRL", he quipped,
| "What? As opposed to 'in fake life'?"
|
| Now I feel vindicated as this is the first time I read a
| government website use IRL. It would seem that algorithms and
| cyberspace aren't part of real life to the government, even when
| it's about housing price collusion, one of the realest things I
| can imagine!
| SilasX wrote:
| Haha yeah that's often what happens, a useful slang expression
| becomes so widespread that eventually they accept it in formal
| context.
|
| I joke that, given enough time, the "English future tense" they
| teach will be "to be gonna".
| foooorsyth wrote:
| As a non-landlord, it seems quite odd that landlords and property
| managers can't "collude" on rent pricing.
|
| If I were to have a rental property and a management company for
| it to be hands off, I'd probably want to defer rent setting to
| the people that understand that local rental market. Both parties
| have aligned incentives - they want to maximize rent. I guess the
| government just wants the low-information party to set the
| pricing? Are there any near-monopoly property management
| companies in any major metro in the US? Seems like they're a dime
| a dozen with lots of competition -- doesn't seems very anti-
| trust-worthy to me, but what do I know?
| quasse wrote:
| It's not _understanding_ the local rental market that 's the
| problem. It's _controlling_ enough of it that you can influence
| prices globally upwards outside of economic competition.
|
| If 60% of a city's major property companies sign on with
| RealPage and *also* agree that they will not undercut the
| prices RealPage chooses (eliminating economic competition
| between the property companies), that's not just being a high
| information party, it's collusion.
|
| The problem is that housing in major cities has a relatively
| fixed supply, so the colluding landlords know that even if a
| competitor undercuts them eventually they will simply fill
| their units and cease to be competition for new renters.
| apendleton wrote:
| Yes, the companies targeted here aren't property managers
| themselves, they sell software to property managers, and have
| ended up providing services to many ostensibly-competing
| managers and building owners in some metro areas, who
| collectively control enough of the rental market in those areas
| that they can "maximize rent" by moving the whole market up in
| concert rather than having any of the owners actually compete
| with one another.
|
| From a ProPublica piece [1] about it, for example:
|
| > In one neighborhood in Seattle, ProPublica found, 70% of
| apartments were overseen by just 10 property managers, every
| single one of which used pricing software sold by RealPage.
|
| [1] https://www.propublica.org/article/yieldstar-rent-
| increase-r...
| justinclift wrote:
| Weirdly, the FTC site seems offline: NOTICE:
| The FTC website is currently unavailable. Thank you for your
| patience while we work to restore service.
| thelastgallon wrote:
| YieldStar software helps landlords set prices for apartments
| across the U.S: https://www.propublica.org/article/yieldstar-
| rent-increase-r...
|
| "To arrive at a recommended rent, the software deploys an
| algorithm -- a set of mathematical rules -- to analyze a trove of
| data RealPage gathers from clients, including private information
| on what nearby competitors charge.
|
| For tenants, the system upends the practice of negotiating with
| apartment building staff. RealPage discourages bargaining with
| renters and has even recommended that landlords in some cases
| accept a lower occupancy rate in order to raise rents and make
| more money.
|
| One of the algorithm's developers told ProPublica that leasing
| agents had "too much empathy" compared to computer generated
| pricing."
| lifeisstillgood wrote:
| Wait. If landlords gather _information_ about all other rents in
| the country, and then use that to set a price for _their_ rents,
| that, cannot surely, be price fixing.
|
| So is the problem the "RENTmaximizer" software and other services
| that basically gather price information ?
|
| Because if perfect price information is available, and yet there
| is no competition in the market driving prices down, well, that's
| rentier markets for you.
|
| There must be an economics PhD or two in that
| jhanschoo wrote:
| > Such software can allow landlords to collude on pricing by
| using an algorithm--something the law doesn't allow IRL. When you
| replace once-independent pricing decisions with a shared
| algorithm, expect trouble.
|
| Enjoyable to watch norms about language (IRL) change.
___________________________________________________________________
(page generated 2024-03-02 23:00 UTC)