Pen testing is a viable alternative in the same way that driving a car is an alternative to designing an engine.
When your specialty is in finding novel exploits, there's not much of a market for you outside the government-o-sphere. In practical terms, pen tests are typically not focused on finding novel exploits.
Nevermind the vast difference in career expectations between salaried government work and consulting.
Ex military often have trouble finding jobs that match their skillsets as well. Not many civilian jobs encourage you to annihilate the denizens of under-developed, resource rich regions.
There's actually quite a lot of value in finding and fixing exploits. It's just that many companies prefer the illusion that $1k is a reasonable bounty for SQLi.