> 2- It's government regulation that makes the tests so expensive.
Yes and no. You'll see numbers saying that each approved drug costs several billion, with most of that money going to clinical testing. The reason for that is that something like 90-95% of candidate drugs never make it through clinical testing. So if you add up the clinical testing costs for each of these failed drugs, and amortize them over the approved drugs, you get these astronomical costs.
But I'm not sure if the takeaway is the same as what you're saying. Would drug development be cheaper if these 90-95% of drugs that currently fail clinical testing made it onto the market? If you replaced the regulatory regime with the usual recourse of letting people sue for drugs that ended up being ineffective or harmful, then it probably wouldn't be any cheaper for the drug companies. And if you blocked the legal remedy, then you've just externalized the cost of ineffective or harmful drugs onto the public.
It's like a balloon: you can push all the air out of one side, and crow about how you've shrunk it down, but the air just goes to the other side. The fundamental issue is that drug development is intrinsically expensive: there's an expense to proving drugs safe and effective through clinical trials, there's an expense to allowing consumers to sue to recover for unsafe and ineffective drugs, and there's an expense in allowing unsafe and ineffective drugs to enter the market.
The problem is that in a first-world country, taking years off of someone's life is incredibly expensive, whether it happens because you sell a harmful drug, or because you sell an ineffective drug that precludes someone from taking an effective one. Let's say the social value of a human life is $1 million per decade. One source puts the cost of regulatory approval for 12 leading pharma companies from 1997-2011 at $57 billion per year on average (http://www.forbes.com/sites/aroy/2012/04/24/how-the-fda-stif...). What is that measured in person-decades? Only 57,000. Out of not just the 300 million people in the U.S., but the hundreds of millions if not billions of people around the world who piggy-back off of U.S. drug safety protocols.