Because I've been so dismissive in this argument, and maybe this isn't entirely clear to the lay observer, I'm going to walk through this process step-by-step. First by means of historical comparison:
Congress cancelled the MBT-70 because its unit cost was too high ($6.43 million in 2014) and we built the M1 Abrams which costs $8.58 million per unit in 2012 dollars so it's way over budget.
Congress cancelled the Seawolf-class submarine because the unit cost was too high ($2.8 billion per unit in 1991 dollars, $4.88 billion in 2014 dollars) and we built the Virginia-class submarine which costs $2.7071 billion per unit in 2014 dollars. It was anticipated that it would cost $1.8 billion per unit in 1991 dollars ($3.14 billion 2014 dollars), so it's pretty well under budget.
Whether you get a better deal on starting over is kind of a toss-up. In the case of the MBT-70 and M1, we did need them, because the M-60 was essentially a jury-rigged WWII design and comparable-to-inferior to its Soviet rivals which vastly outnumbered it. In the case of the Seawolf, we didn't really need them, because the Cold War was over and there was no apparent pressing need for a new SSN right away. Costs are probably associated with this urgency.
The F-16 and F-15 will cease to be relevant over any kind of serious airspace in approximately 10 years or less, if they aren't really already. They're 36 and 38 years old, respectively, and you've conceded that building more of them isn't a good idea, which puts you ahead of most people on the opposite side of this argument. There's a need for an immediate new replacement. On this we agree.
The development time for the F-15 spanned from 1965 to its introduction into the force in 1976, or 11 years. The F-16's development time spanned from 1971 to its introduction in 1978, or 7 years. The F-22's development time spanned from 1986 to its introduction into the force in 2005, or 19 years. The JSF program began in 1993 and the F-35 will enter service next year at the earliest, or 22 years. We can see that the more complex the plane, the longer the time it takes to build and introduce, and if we were to look at some other examples, we'd see that this time has been increasing ever since WWII as they become more complex.
We can also see from the examples of cancelled programs like the MBT-70 and the Seawolf that coming up with the replacement for the thing that's been cancelled generally involves throwing away most or all of the work and reassessing from the new current state of affairs and takes about as long as the original program did to yield some sort of usable product from there.
Let's be charitable and say that if the F-35 was cancelled right now, it would only take 18 years to produce a replacement. That'd be 2032. USAF and USN are expecting a 6th Generation fighter to have an initial operational capability in
2030. So skipping the F-35 right now would mean functionally aborting 5th Generation aircraft development entirely because any replacement using the same technology would be obsolete by the time it was fielded. Even if we built a new 5th Generation plane (and really, based on what we already know, we'd be building
three, one for USAF, one for USN, and one for USMC) and managed it in a much shorter time and for less money, those hypothetical new 5th Generation aircraft would have a much shorter operational lifespan, so they would still be largely a stop-gap waste of money.
6th Generation aircraft development probably can't be sped up due to the high degree of new and untested technologies that will form the core of their capabilities, and that will probably not work exactly as anticipated given many of them don't even exist yet. Of course, factoring in technologies that don't exist yet means that you will run into unanticipated problems and probably spend a lot of money fixing those problems. You have to incorporate such technologies because delivering a product using wholly reliable and safe (and therefore likely wholly obsolete) technologies is unacceptable in the defense industry, as someone else will not share your caution and will have a huge advantage over you.
It takes approximately 5-10 years to produce the next generation of aircraft in a really usable quantity, during which time they're usually backed up by the previous generation and form the tip of the spear, with replacement taking place over the next 10 years.
Your solution is to cancel the F-35 today, and essentially start 6th Generation development now (which will be
just as buggy and almost certainly
more expensive, given we'll be skipping a generation and doing even more untested work with no attendant experience) while leaving America with effectively no fighter capability from 2020 to 2040, and undergunned all the way until 2050. So you want us to undertake a program
far riskier than the JSF/F-35 program, and you want us to do it with
no fallback systems, that we need functionally
right now. In the process you propose we cease having the ability to contest airspace for 30 years because you're worried about a program that isn't actually much more expensive or buggy than any program before it.
The
only vaguely politically and militarily acceptable solution in the wake of cancelling the F-35 would be to reactivate F-22 production in the
thousands, which would be
more expensive than the F-35 (all the production line tools are boxed up and sitting in a desert as we speak), and modifying the F-22 to be a multirole aircraft, given it was only ever intended for air superiority. This would also still leave both USMC and USN in the clutch, requiring yet more money (and piss off all the international partners to the F-35 and more or less destroy the American defense industry's credibility in international markets, in which it currently competes heavily and fairly successfully). In other words, it
isn't politically or militarily acceptable at all.
Your solution isn't
safe, it's
suicidal.
This is why your solution is 1. bad, 2. why I'm glad you aren't in any way associated with the decision-making chain regarding the F-35 or military acquisitions as a whole, and 3. have been extremely dismissive of this entire line of thinking the whole way through. The F-35 literally is too big to fail, which was a consequence of politicians demanding the services work on a joint fighter. Once again, this doesn't go back to some error in the defense industry or the military establishment. I continue to remain confident those two parties will find a way to make it work, because they have no choice but to do so. Of course, none of this is new. The Manhattan Project and B-29 are pretty good historical examples of putting all the eggs in one basket while under the gun and it paying off in spades. Our current situation is notably less dire.
In other words, yeah, if it doesn't pan out, we're totally screwed. That's still better than your solution, where we're totally screwed period and only
might not be totally screwed by the time all of us are middle-aged.