The researcher says
> this strange strategy will maximize your profit. āTo me, it was a complete surpriseā
It doesn't seem like such a surprise that algorithms that use information about rivals to optimising profit tend to price high.
Consider a small town with two gas stations, you own one. You can set the price (high or low) in the morning and can't change it until the next day. Your goal is to optimise profit for the next 1000 days. On day one you price high (hoping your rival will). But your rival prices low and wins lots of business. On day two, you price high again (hoping your rival will have seen your prices and cooperate). If your rival prices high, you both stay high for the most of the next 998 days (there's some incentive to 'cheat' and price low, but that is easily countered by the rival pricing low). If your rival priced low on day 2, you have to start pricing low too. But occasionally you'll price high to try to 'nudge' your rival to price high to avoid low-low. If they eventually understand, you can both price high for the rest of the 1000 days. Critically, even if stuck at the low-low equilibrium, you'll keep trying to 'nudge' high periodically. The frequency with which you try to 'nudge' will depend on the ratio of profit for high-high vs low-low. If you both make extreme profits when pricing high-high, you have more incentive to 'nudge', but if the difference isn't great, you won't nudge as often.
Seems obvious pricing high will be attempted in proportion to the reward relative to pricing low.
The researchers' conclusion seems reasonable:
> itās very hard for a regulator to come in and say, āThese prices feel wrongāā
and
> what can regulators do? Roth admits he doesnāt have an answer.
(i.e. in practical terms, there's no way regulators can police what algorithms sellers use - I can't think of exceptions to this, but perhaps there are some special cases)
The author cites the common CEPR paper [0], but missed its most interesting finding. It found that the algorithms definitively did show signs of collusive behaviour, but that their chosen equilibrium price point was far below the Nash equilibrium. That is, the researchers expected these algorithms to maximally extort the consumers, but they only modestly extorted the economy's consumers.
[0] - https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3304991
It's possible algorithms simply drive up prices because price isn't the main factor some people use in deciding what to buy. Algorithms are probably able to learn that raising the price outweighs the decrease in the number of people buying something, and if every algorithm does the same, then prices will continue to rise.
I never understand why people donāt pay more attention to ānā in these cases: number of other players.
If there are 2 suppliers in a market, they will collude without algos or private meetings: I can be pretty sure you will not cut your price if I donāt cut mine. The issue there is that there are only 2 suppliers, so trust is very easy.
If there are 100 other suppliers, I know ONE of them will cut their price. So I best cut mine first.
What I am trying to say here is that, algos or not, n is the major driver here imho.
Thatās kind of interesting since the US has been very relaxed about falling values of n as long as prices seem ok.
"Algorithmic collusion"... If using an algorithm leads to collusion, then choosing to use the algorithm should be considered regular collusion.
This was the Greystar situation that already happened with apartment rentals.
Are people not aware of this?
A switch to value based pricing for essentials (water,shelter,transport,utilities, etc.)is an extremely easy way to destroy disposable income and even make some areas impossible to live in for the existing members.
Austin, Texas in 2021 saw several of my friends who were renters see a 1 year price increase that more than doubled their rent, I had friends who we're doctors who we're forced to move out of one bedroom apartments, even if it wasn't the plan, it's still a great way to displace people like local musicians so hack comedian can move in.
I feel obliged to link to the Behind the Bastards podcast "Why is rent so damn high?" It's a lengthy 2 part piece, but quite enlightening. https://podcasts.apple.com/us/podcast/part-one-why-is-the-re...
How is this new compared to https://www.paecon.net/PAEReview/issue53/KeenStandish53.pdf ? (See the section "perfect competitors are not profit maximizers".)
>strikingly high probabilities to very high prices, along with lower probabilities for a wide range of lower prices
Isn't this describing the strategy of keeping ever high prices, then doing some temporary price cuts/sales/deals?
I mean we are in the age of digital pricing, even on the shelf. Modern price collusion is more apt to happen with A/B testing if prices at locations to see what the local market will bear.
I've seen Walmart do this in the past. Items that were not on sale could have significant differences in price, where in general the prices in more affluent areas are higher. We're talking 50 to 75 cents on common items, but sporting goods quite often had a different of 3 to 5 dollars.
Is this applicable to the stock market? Maybe this is why valuations are completely devoid of any intrinsic value.
https://news.ycombinator.com/item?id=45611361
On October 6, 2025, California Governor Gavin Newsom signed AB325, a law targeting the use and distribution of certain algorithmic pricing tools. This law is part of a larger legislative trend to try to reign in algorithmic pricing.. Californiaās bill targets pricing algorithms in all markets and will take effect in 2026. However, a violation of the new law requires a conspiracy or price coercion, so as a practical matter, it may not extend the range of violations already encompassed by the Cartwright Act."Algorithm" has got to be the least useful word in English today.
It isn't the software that's responsible driving up prices, it's the information.
> Yet a widely cited 2019 paper (opens a new tab) showed that algorithms could learn to collude tacitly, even when they werenāt programmed to do so. A team of researchers pitted two copies of a simple learning algorithm against each other in a simulated market, then let them explore different strategies for increasing their profits. Over time, each algorithm learned through trial and error to retaliate when the other cut prices ā dropping its own price by some huge, disproportionate amount. The end result was high prices, backed up by mutual threat of a price war.
This is nonsense. Those "algorithms" were programed to do that. I also notice they didn't add a third copy of the algorithm or a fourth. The summary of this research is that they built a novel algorithm (not one used in practice) and put it in a simulation. How this is representative of any real world scenario escapes me. They proved that software written to optimize profits optimizes profits. Shocking.
The researchers quoted in the article are, essentially, defining collusion as knowing what competitors prices are.
Very interesting. I looked at stability in learning agents in artificial markets back in the late 90s for my PhD and concluded that at least the systems I worked with weren't stable - they were prone to bubbles and crashes.
Very interesting to see that there is a class of stable systems that force high prices.
Would be interesting to understand if the no swap regret systems studied also give stable results when it is an N player game rather than a 2 player game
> hard it may be to regulate
Not hard at all. Outright ban fixes all.
Smells like Reddit in here
It's painfully simple. When vendors follow an algorithm, they are colluding.
Imagine if multiple vendors in a product or service area form consortium which launches an "independent price determination task force". If everyone follows the recommendations of the task force, they are colluding, even if they don't talk to each other to set prices.
Replacing "task force" by an algorithm changes nothing. The agreement to use the algorithm rather than to compete is the collusion.
Who wrote the algorithm?
"Everything is working as intended" but the real-world outcome feels dystopian
> Imagine a town with two widget merchants. Customers prefer cheaper widgets, so the merchants must compete to set the lowest price.
I always found this statement to be rather wishful. Individual lowering of prices makes sense if and only if your competitor is capable of saturating the market. Otherwise, demand elasticity becomes very relevant. Sure, your competitor may take the larger share of the market, but then you can compensate with higher per item profit.
The common wisdom is that in properly functional markets there's enough supply with n-1 market participants, therefore given a market signal of one participant lowering their prices the last one standing without lowering prices gets kicked out of the market, making maintaining prices the losing move. Yet, if the rest of the market does not react to the signal, the one lowering their prices hurts their profits and possibly kicks themselves out of the market. Making price maintenance, and depending on elasticity maybe even jacking of prices, the winning move in the presence of this signal.
Turns out the probability of either move being the winning move is dependent on probability of other market participants colluding/defecting. However, since lowering the prices hurts the profit a rational market participant would conclude that the rest of the market is inclined, even if a little bit, not to lower their prices in reaction given price cutting signal and similarly a bit more inclined to raise the prices given price hike signal.