And "revealed preferences" are mostly bunk, because of what GP alludes to. It's evaluating people's choices out of options available on the market, not what they would buy if they could freely optimize the feature/price matrix. Another way to put it: "revealed preference" is just what one considers the least bad of the bad choices available.

Perhaps but that should be an unstable equilibrium, i.e. temporary.

As long as real competition exist, if it is possible to deliver what consumers really want then sooner or later someone will discover it, make a killing and prompt everyone in the market to follow suit or die.

It's easy to blame corporations but in fine I believe that the current situation has been driven by the choices of the consumers, i.e. all of us.

Sure we want better quality, repairability, longevity, but we also want cheaper, among other things. I think product ranges simply reflect where our priorities really lie.

It should be in theory, but in practice it's pretty stable on relevant time horizons. There are many factors fixing such equilibria. For instance, most consumers aren't rational actors carefully evaluating purchasing decisions. Or, when they pick, they may optimize for multiple targets simultaneously. There are economies of scale that create barriers to entry for new competitors. Trust takes time to develop. Etc.

To use two relevant examples:

Audio jack on phones: a lot of people want one, but companies seem to - one by one - eschew including it, and gain on the BOM savings and getting customers to purchase additional dongles/adapters. Phones are something you don't pick based on one particular feature. A new competitor that'd like to offer a good smartphone with an audio jack will have to first match many other features of competing smartphones. They won't be able to do all that and achieve similar price point, as incumbents benefit both from established supply chains and scale. Therefore, a new offering is going to be significantly worse in terms of features or price (or both), way beyond what established companies have saved on removing the jack.

Reliability of fridges: the three most important parameters that matter for customers are internal volume, price point, and overall aesthetics. The consumers would prefer a fridge that lasts a decade over one that last three years, but such decision is somewhat beyond the calculation horizon of most people (the reasoning is instead, "if and when it breaks, we'll deal with it") - and even if it was, it's very hard to verify. A company selling reliable fridges would have to first be able to reliably signal their reliability, which is hard to do. You could try offering very long (or lifetime) warranties, but unless you're an established brand, people ain't gonna believe it.

Then there's the general aspect of customer irrationality[0] wrt. price. People without much spending money prefer cheap products. There's a phenomenon in markets that probably has a name, where a market for a product bifurcates into race-to-the-bottom low-quality but cheap products, and high-end products (often for professional users) with low sales volume and high markup. The middle - quality products at moderate prices - disappears entirely. A new competitor, wanting to target the middle in such a market, will find themselves losing to both cheap and expensive ends.

--

[0] - Though I'm becoming less certain that this is truly irrational. The other day I saw some article highlighting that humans like to optimize for maximizing available options, in a way that comes close to mathematical optimum. Since in money-based economies money is, quite literally, unit of options available, this would suggest that this behavioral heuristic applies.

But in the context of this thread it seems that they do have (long lasting, repairable) Miele appliances available and they choose to buy other ones.

Perhaps they cannot price in reliability? How many people know that Miele offers long lasting and repairable appliances, instead of just expensive ones?

I for one "know" this by weighing HN anecdotes I've seen over the years - couple dozen saying that they sell solid stuff, and one or two saying that they used to sell solid stuff, but are now following the trend of replacing metal components with cheap plastics. How do I evaluate that? Brand trust isn't that strong a signal.

The abstract problem is that the market is imperfect due to insufficient customer knowledge. I know my last washing machine lasted X years, but how that will compare to the new one from the same manufacturer, I don't know. The manufacturer also won't (really) tell me, just the usual fluff about quality.

So as a consumer, I am unable to decide objectively, just using rumours if there isn't a hard guarantee or a trustworthy evaluation of the quality of all products I could pick. In that situation, I will price in the uncertainty, I will evaluate the rumours and the tendency to produce crappier stuff over time. And when in doubt, buy 4 cheap short-lived washing machines instead of 1 for 4 times the price.

Isn't the warranty pretty much the metric of "how long it'll last", as expected by the manufacturer?

A long warranty, while not a guarantee, is pretty correlated to the product lifespan.

Not so simple. Many manufacturers advertise a long warranty and then, when things break, drag their feet, blame improper use or wear and tear, offer workarounds instead of repairs, etc.

A long warranty is only good if the manufacturer is trustworthy. But untrustworthy manufacturers are one of the causes of the whole problem.

A fix might be warranty terms defined by law, where the manufacturer can pick the duration and nothing else. Disputes should be settled in court or by a neutral board of experts.

I have not found it to be the case, one of my displays lasted 7 years and still works, the other gave out in 3, both had 1 year warfanty. Dell and asus respectively.

Think of it the other way round. Long warranties are indicative of a durable, long lasting product. A short warranty, however, does not necessarily indicate a crap product.

Of course this is not a perfect measure, but I do think that a correlation would exist here.

Or, they may also be indicative of the company making its money elsewhere, and thus willing to eat the cost of regular warranty replacement just to keep you tied to their hardware, so that they can exploit you elsewhere.

Have you tried asking pe— sorry, “market research”? If you can make something that people want to pay for, that goes down a lot better than trying to make them buy what you're selling via an elaborate multi-agent optimisation algorithm.

My job is writing software, so I never have reason to do market research.

My strong impression is it's hard to do. People may say the want kale, but end up buying donuts.

They weren't lying about wanting to be the kind of person who buys kale. But at the moment of truth, they got donuts "just this once".

This suggests that they'd pay for the opportunity to be the kind of person who buys kale.