To me that is enough to qualify as an "economic driver".
If your argument is that Erlang is not contributing more to GNP than what disappear in rounding errors (or even that), then sure, companies using Erlang are contributing peanuts compared to what products from companies like Microsoft and Oracle are.
I was simply pondering on the whole discussion because I have yet to see a posting requiring usage or or experience with, Erlang.
Not to say they're not out there, just I haven't seen it yet.
All that is supposed to change, when Microsoft launches .NET 4 and C# 4.0 with it which is the first concurrent programming language from Microsoft and along with that Microsoft is supposed to raise awareness on multi core, and concerrency. And this is the problem!
The industry, economy, development and evolution of systems can not be held back pending a single company like Microsoft, Apple, etc.
Computer science and it's applications need to be utilized properly and timely.
Personally, when it comes to concurrent programming I'm more inclined to look at Clojure than Erlang but, when I surf the job sites - I don't see any job postings for either language.
That's not to say there aren't any places with a need for these two languages, I just think there's a long way before a critical mass level is hit to even match Ruby on Rails.