Screen readers, the computer replacement for Braille, are also stupidly expensive (one I looked up was $900 for the home edition).
Both. The one doesn't exclude the other. I'm a heavy user of text-to-speech at high rates to digest content easily, but prefer braille when coding or doing some work on a shell/terminal.
You'll see much more braille displays in some European countries because they get subsidized by health insurance or a government entity that provides assistive technology for the workplace. This workstation has a 88 cell braille display, which means it can display 88 braille characters at a time. That unit costs $10k or so, depending on market and EUR/USD rate etc, but to give you an idea of the price tag. I also have a mobile unit with 40 cells, I guess that one is about $6k or so.
For someone who is able to read text at close distances with significant optical assistance, it is far more beneficial for them to learn a written language than to learn braille characters. Until recently, many legally blind US children (and their parents) had the choice to either learn braille or written English. Those who learned braille ended up severely disadvantaged because of the utter lack of braille information produced in our society, and the difficulty of using any screen-based technology.
Equally important for HN-type readers to know is that Braille literacy is highly correlated with employment among the blind, and rates of Braille literacy are falling rapidly in the US, from a peak of around 50% in the 1950's to more like 12% today [1].
Smart phone apps are great and important but from what I've read they don't seem to be having the impact on blind employment that Braille has had.
Hopefully someone here will know of a survey reporting measurable economic gains for the new generation of Braille-illiterate technologically assisted blind, thus far what I've heard anecdotally on that front has not been encouraging.
[1] http://www.nbp.org/ic/nbp/braille/needforbraille.html