Dull Men's Club

companies are purposely designing them wrong to shorten their service life

This 100%. And specifically for readers unfamiliar with how product R&D works, the malice doesn't even have to metastasize throughout a whole company in order to design inferior products. The following summarized, fictional exchange should depict the problem:

Management: we see Competitor X released a new light bulb that lasts 800 hours and costs $1. We need our own light bulb product, with at least 40% gross margin.

Marketing: OK, we can be competitive if we make a 1000 hour light bulb and consumers are willing to buy it for $1.10. We can maintain 40% gross margin if our cost per unit is less than 25 cents.

Engineering: OK, we'll go work on that

[3 months later]

Engineering: right, we've built this light bulb that lasts 1500 hours avg (std dev of 100 hours) and only uses bog-standard tungsten from our long-term supplier, so the cost is 20 cents per unit

Marketing: nice, but we don't need 1500 hours. Can we reduce the cost per unit further?

Engineering: What? But we're already below 25 cents.

Marketing: No, you see, management wants at least 40% gross margin. More margin, more profit, more better.

Engineering: No, you don't see. We're already using the thinnest tungsten possible. We can't change an element's melting point, we can't draw it any narrower, we can't do .... [insert ten other reasons why pursuing further savings is of diminishing return]

Marketing: This is an ultimatum: we cannot accept this product into production unless it meets exactly the specification we wrote. We will cancel the project and outsource R&D if you cannot achieve this.

Engineering: WTF??

[a month later]

Engineering: Per your ~~insane~~ request, we have produced this light bulb with 1000 hours (std dev of 400 hours), by taking the earlier design and knicking the tungsten filament every few millimeters, so that those thin points will eventually fatigue and break. The cost to do this knicking is an extra 1 cent, but the material savings is 2 cents, so we are now at 19 cents per unit. But the std deviation shows wildly varying behavior for when any particular bulb will fail. /exasperated sigh

Marketing: Excellent! We'll ship it!

Engineering: .....

When the incentive structure is bad, all sorts of perverse results will occur, even for well-meaning participants. I leave it to you, dear reader, whether R&D's complicity with such perverse, capitalistic goals is morally damning, but the fact remains that if a company does develop a superior product, it might just never see the light of day, or will be intentionally delayed/deferred until needed for "competitive" reasons. This is one possible mode, but the other would be to chronically underfund R&D, so that superior products cannot possibly be developed, barring a spontaneous and unplanned stroke of genius.

To that end, whichever the cause, the result is all the same: the guise of "competition" is actually a local minima, where the thread-bare minimum is accepted as a maximum, where maximum profit can be extracted. Why put money into innovation if no one else is? Why compete when all the other competitors know the game as well: all will save costs by maintaining the mediocrity. It's like monopolies and trusts from a hundred years ago, but they don't even need to say a word to each other in order to collude.