A few things>Only recently became usable to the point where they can be realistically sold at profit>Major patents covering them only very recently expired so previously there were only really a few OEM's manufacturing them at inflated production cost due to the patents
But the main things are>Limited use case and market
E-ink is a relatively niche technology that basically does one thing, giving a paper like display, really, really well and does a lot of other things absolutely terribly, which dramatically limits the products its suited to, which limits its market reach and profitability
This is further compounded by the fact that the single largest market for them by a massive majority, e-books, have been flagging in sales for years, as ebooks are the kind of thing that get bought and used until they fall apart, unlike phones and televisions which are upgraded continually as they become outdated, an almost decade old first generation kindle still gives an excellent reading experience and will not have noticeably degraded in performance from the day it was purchased, displaying text is all they do so there's no need to upgrade from those old, now cheap devices to something new unless you need really niche features like displaying manga, which means ebooks are an eternally low volume product purchased by relatively select demographics, especially when you can get a cheap android tablet that displays books OK and does everything else immensely better for the same or less money
These lead to>Limited financial incentive and difficult swallowing cost of painful R&D
E-ink is a niche technology, with reliable, but low marketability, mostly used in single purpose devices, to break out of that mold, it would need a lot of work done with it to bring its feature set up, being able to accurately display colour is I'm sure something lots of Ebook enthusiasts would die for, as is a higher refresh rate that would make E-ink panels at least useable for things other than displaying and occasionally changing static text, however, low volume item with massive R&D costs to improve t;o the point where it can break out of its silo and work towards being a higher volume item means a long period of companies swallowing thousands and thousands of spent man hours and millions of dollars in research money for a middling transitional product that costs too much they can sell right now as they go, or even better, eating all those costs for at least a few years then releasing a finished product with a dramatically inflated prices due to the dev costs that will sell like hot aids for at least a few product cycles
This is in a symbiotic relationship with the last and main reason, the two feed off of each other
>E-Ink simply doesn't benefit from the massive economies of scale that LCD and now OLED does
Rather than ask why E-INK is so costly, look at why regular panels are so cheap, how can amazon sell you a fire tablet for $50 when their most basic ereader is almost double that and does about 1% as much stuff?
The answer is massive production runs, other market incentives and being able to use already paid for manufacturing techniques and facilities
E-INK is basically in the stage that stuff like 8k tvs were 5 years ago, primitive proofs of concept packaged up and sold at a seemingly absurd price to relatively few people, the difference is that 8k panels etc benefit from existing technology, their development will have massive application outside of just tv's and they have an obvious mass appeal across multiple market sectors, so they get money and scientists thrown at them until they're really good, and the R&D cost passed onto each individual consumer is kept low by very long production runs and large sales volumes.
E-INK, by its nature, cannot really benefit from those