Look at the data for Austria presented in the link in the article. Supermarkets raised and lowered prices on the same day to the exact same amount and lots of other pretty damning evidence.
If supermarkets are getting together and forming handshake agreements like “we will not change prices between November and February”, like Canadian supermarkets apparently did, then that could constitute collusion.
I'd imagine that even for low-income people, it's the cost/benefit of comparison shopping has been squeezed out by how much prices change on a day-to-day basis. Like sure, if you're buying a lot of something all at once it might make sense to do on that occasion, but once you do that a few times you either learn:
1. Which store usually has the lowest price (and if they have coupons for store X, they might just only go there, because... they can't use the coupon at the competitor's)
2. The difference in price doesn't offset the time-cost of going to multiple stores (and the consolidation of stores means that going to two different stores will take even more travel time).
So, if most people aren't really doing comparison shopping anyway, then you make more money by matching your prices to your competition.
Of course the time it takes to do this makes it all but impossible to actually save money, so it has to be a small portion of their customers.
I have heard that only something like 10% of customers actually give a shit about prices at all, and it is them who keep the prices in check for everyone else. i.e. with razor margins, the retailers can’t afford to lose that 10% customer base
Yes, the vast majority of people don't comparison shop. But people do decide to not buy something if it seems it is too expensive and vice versa, and the effects are seen on the statistical level.
This will be even more popular in situations like supermarkets, where a significant part of the stock has an expiration date that isn't so far from today. Turning your inventory too fast is just as bad as turning it too slowly, so there can be immediate reactions to make sure things are being consumed at just the right speed. And the more uniform the models of consumption the supermarkets are running, the more similar their decisions will be anyway.
So I wonder if we even need to focus on needing damning evidence, or on whether there is collusion, and instead aim for what we want: Dynamics that put negative pressures on prices. If we aren't seeing that, I don't care much about how much is collusion, and how much is models that have tacit agreements because, as market players optimize for what is best for them, there are solutions where high prices across the board makes all sellers win.