Anyway, this always occured to me as the "safe" way to counterfeit. The level of money generated stays below everyone's radar screen, the denominations are small enough and involve enough machines that they might not even be found to be counterfeit until they were so far removed from the transaction as to be impossible to trace without a level of effort that wouldn't pay off.
Interesting theory, give it a try and let us know how it worked out when you're released from custody.
The problem with this thinking is that enforcement is prioritized according to:
(1) the per-incident economic impact (e.g. a suitcase of $50 bills vs. a single bogus $10 bill)
(2) a discernible pattern to the incidence of counterfeit-bill complaints, particularly if the aggregate impact of the fake bills is significant
(3) the likelihood of an incident investigation leading to a source of counterfeit currency
That said, even if one were to optimize for each of the above, you run the very real risk of coming to the attention of an agent with time to pursue a lower-priority investigation.
Counterfeiting US currency, particularly in these times of heightened security awareness, is about one of the highest-risk paths to ill-gotten gains I can think of. Based on comments made to me by FBI agents with relevant experience, the risk/reward ratio for a single well-planned bank robbery is significantly better than any appreciably lucrative counterfeiting scheme anyone has come up with yet.
Entertainingly enough, most bank robbers are sufficiently deluded by their first success that they try to parlay that into a series of successes. This character weakness is generally how non-violent bank robbers get caught.
-Trevor