kind of like your non-falsifiable belief that regulation must cause higher prices.
Not at all
@Bob Noel . My beliefs about this are completely falsifiable. I can think of lots of evidence and analysis that would persuade me that belief is false.
By contrast, I have not heard mention of any evidence or analysis, which could exist in principle, which would falsify the claim that regulation does not cause an increase in prices (and I have asked in the V8 thread and received no answer). That means that belief is non-falsifiable for the person who can’t name some. And there really is little point in discussing such beliefs to determine the truth - since by definition they can never be falsified. I am discussing it further with you here because I always assume people are rational and their beliefs falsifiable until demonstrated otherwise.
Frankly it strikes me that the fact that the same equipment (literally) costs $2000 more for a certified version is a rather powerful argument that exactly that happens.
When I think about it more, I think there is also a good intuitive argument that the price of compliance being zero (or even not statistically significantly different from zero) is extremely unlikely. Consider the work that has to go into assuring compliance. The testing and paperwork at a minimum, if not design changes. It seems extremely unlikely that those cost nothing.
Years ago there was a NASA/FAA investigation into software certification costs (Streamlining Software Aspects of Certification). People running it soon realized that companies that knew how to get "178B certification" efficiently weren't going to tell other companies how. There were a lot of people not familar with DO-178B that had this view that getting "178B certification" would increase software costs by an order of magnitude. But there were companies that knew how to do it without causing massive cost and schedule problems.
That’s a good datapoint and certainly shows that compliance can be done more cheaply and efficiently. But that isn’t really evidence that the cost is zero.
I think our best estimate, based on this data point, would be that for products which are primarily software, the cost of compliance is about 13% of the total.