There are several factors at work here.
First, it's not easy to make even an f/1.2 at a mass-market price. Go faster, and the price REALLY goes up.
Second, demand is tiny. It's all very well for a few dozen people here to say that they'd really like one, but it's still only a few dozen people, and how many of those would put their money where their mouth is, even at $2000 (a thousand pounds)?
Third, the weak dollar has already been discussed at vast length. A lot of people in the USA would have severe sticker shock, even if the lens were affordable elsewhere.
Fourth, any such lens would attract a lot of attention, i.e. it would be a reputation-builder (or breaker). The manufacturer therefore needs to make sure that it's a bloody good lens.
Fifth, unless you particularly want very shallow depth of field, ultra-fast lenses are a lot less necessary than they used to be. When the Canon f/0.95 came out, the fastest colour film was High Speed Ektachrome at 160 ASA (the Ansco 200 and 500 came out much later) and the fastest B+W film in 35mm (Kodak's ASA 1250 Royal-X Pan wasn't made in 35mm) was the short-lived Ilford HPS (800 ASA). Today, 160 is the slowest ISO equivalent on a Leica M8 and Ilford Delta 3200 (ISO 1250) and Kodak TMZ (ISO 1000) are amazingly good at EI 2500-3200 and can be pushed to 10,000 and beyond.
Sixth, if you do like very shallow depth of field, e.g. for portraiture, it is often feasible (and possibly even desirable) to use a longer lens, on a reflex: 85/1.2 or 85/1.4, or even one of a number of cheapo 135/1.8.
Finally, although I have f/1.5, f/1.2 and f/1 in the cupboard, I find that I almost never need the f/1.
None of this is to say that a new super-speed lens won't appear -- in fact, I rather suspect it will, at faster than f/1 -- but it is quite likely to be an expensive, landmark, prestige lens, not a fast lens for the masses, because there probably isn't a big enough market.
Cheers,
Roger