Now to calculate damages correctly, we can apply several models.
Suppose somebody has put a song illegally onto the internet, and some other people might have downloaded it.
a) We calculate damages from the actual amount of people who downloaded it, and multiply it with the sales price. Yes, the people who had downloaded it might also put it onto the internet illegally, but then these people violate copyright, and not that somebody that put it initially there. So these need to be sued separately. This has a further problem that it's very hard to quantify if using peer2peer technology, but we could divide the outgoing peer2peer traffic of the file by the size of it, and get a number. Probably something like 20, multiplied by $1.
b) We deem the first person to put it up illegally to be responsible for all the infringement that might arise further down. Since we don't sue everyone else, we don't have any numbers who might have re-published it, and whether they've even republished it. With peer2peer, we can at least assume they've republished it illegally. So we could take the above figure and assume that everyone republished it 20 times. So this would end up at something around $400 for the above example. Now, since bandwidth is finite, sharing multiple songs and other materials would probably lead to smaller amounts of the work being sent out illegally, so a whole album might be sent out only 5 times at all, so this still wouldn't amount to a lot, maybe 5x5x$12 = $300 for the above example.
Obviously, it's rather difficult to even get exact traffic stats, and they also could be inflated artificially by the rights holder himself, this is probably not the right way to do it.
We could instead base the whole damages upon something else:
c) The amount of people that could have bought the song, if it hadn't been made available illegally. As it happens, the number is 0. Because those people spent their money somewhere else in the economy: The bought coffee, computer games, went to the movies, etc. So that's not really a good number, and we can't discern people who didn't buy the song but a cup of coffee instead, from the people who didn't buy it because they bought a weeks food instead.
d) The amount of people who would have bought it, if it hadn't been made available illegally. This is more interesting, but also hard to figure out. Obviously all fans would have bought it. You might get a ballpark figure of how many there are from homepages, social media and concerts. But it will probably be lower than the amount of songs already sold. You might use past sales, but then, you already have illegal copies in there, so these will be less than the possible amount. Probably the only way to find out is a survey and do statistics. Just ask a random set of people, worldwide, and ask them if they already bought, or would buy the song for $1. Which would be a huge effort.
But there still might be another way: Statistics on economy and spending.
e) Since the money not spent on paying for the song, the money was spent on something else. That "something else" is totally uninteresting if it's food or housing or other necessities. It's only interesting if money was spent on entertainment, and the amount happens to be known for a lot of countries. From this we can actually figure out the percentages for the different kinds of entertainment (in the US it's around $1000 a year for audio, but including equipment), so we need the number spent for audio recordings alone. If we divide that number by the average total amount of audio recordings acquired within a year (from sources published legally and illegaly both, as well as free/open content/commons/public domain; probably needs a survey), this will give us a price somebody will actually pay for a song. (And now I've gotten lost on the damages part, but that price is in itself interesting, because it's the price you want to sell your songs at). Ah yes, we need to know what percentage of the population, on average, has a certain song. This will be rather low. Multiply that with the population times the price somebody will pay, you've got your damages.
Needed for (e) are surveys on the amount of songs acquired in a year, on average; and on the chance a song has to be among them, on average. But this actually sounds doable, at least in certain countries.
When everyone that lands on some kind of "blacklist" isn't able to protect its own machine, then we might assume that their machines are soon part of some botnet, making everyone else less secure. Stupid gits.
Of course, that is assuming that AV products really work.
"Content ID has claimed millions of videos and is responsible for over a billion dollars in revenue so copyright holders love it."
Well, judging by the amount of people who had their self-made videos, of which they are the copyright holders, taken down, because of baseless accusations and misbehaving robots, I very much doubt most copyright holders love it.
Re: IP is all the US has left, that is why it is so important
The only thing we have left is IP.
No, you've got even more IP Protection.
Just about all works of art produced until 1978 should be in the public domain, but aren't. So you're inflating the amount of works artificially with extending law.
And all software patents are actually illegally granted, because software is math. So you're also inflating the amount of patents by granting patents on math, obvious things and 4000 year old technology (slide to unlock..).
And judging by TPP, TTIP etc., your biggest export also seems to be protectionist laws.