4 ms·
I think the argument of the comment you're replying to is that it was profitable and they did make the calculation. They're left with $7B, which is a lot. It's
by geofft 7y ago
I think the argument of the comment you're replying to is that it was profitable and they did make the calculation. They're left with $7B, which is a lot. It's hard to make $7B.
And the multiplier works in the other way. Because of the non-deterministic chance of whether you get caught and whether you get punished and whether all of this happens before you donate all your money to charitable causes and die, the expected value of your loss is a fraction of the fine. A truly effective penalty would fine them many times the illegal profit to account for that.
- loeg 7y agoThey also did not end up giving up any of their personal wealth in the settlement. $5B was just the starting bid. > A truly effective penalty would fine them many times the illegal profit to account for that. How much of the profit was illegal and how would you go about quantifying that?
- geofft 7y ago> How much of the profit was illegal and how would you go about quantifying that? I mean, I haven't done a criminal investigation of the Sacklers, so I'm the wrong person to ask. But on general principle - if the goal is to use fines to disincentivize future criminal behavior (which seems reasonable to me, but it's not the only factor at play in setting fines), you should treat all profit that was related to the criminal behavior as criminal and not try to reverse-engineer "How much money would they have made if they hadn't done this crime." Otherwise you're still incentivized to try 10 different legally-questionable strategies and keep the profit from the 9 that were found to be on the other side of the line. As a concrete example, if Google's copying of Java interfaces were ruled illegal (and to be clear I don't think it should be), then an appropriate fine is (some multiple of) all profit from Android, not some attempt at retconning how much money Google would have made if they had used something other than Java or waited for OpenJDK. (That's a fine, to be clear, not damages. Oracle's damages should be calculated by retconning how much they would have gained had Google kept the laws, and nothing more.)
- loeg 7y agoI mean, the first step is establishing a second crime. What's the crime on which to say any profit since 2001 is related to a crime? The company and some executives already plead guilty in 2007 to, essentially, false advertising, during the 1995-2001 period immediately after the drug was developed. During that time, "OxyContin brought in $2.8 billion in revenue for Purdue Pharma."[1] Double Jeopardy means that specific crime during that specific period is done and settled. The company paid out $600 million in fines and the execs another ~$35 million. So that's 22% of revenue in fines in that instance. (I guess the courts didn't agree with your notion that fines should be 100+% of revenue.) The New Yorker estimated in 2017 that OxyContin has brought in $35 billion in total revenue for Purdue[2]; if that's true, subtracting the 2.8 from earlier suggests that it produced $32 billion in revenue 2001-2017. Did Purdue Pharma commit another crime (or the same mislabeling / false advertising kind of crime) after 2001? [1]: https://www.nytimes.com/2007/05/10/business/11drug-web.html https://www.nytimes.com/2007/05/10/business/11drug-web.html [2]: https://www.newyorker.com/magazine/2017/10/30/the-family-that-built-an-empire-of-pain https://www.newyorker.com/magazine/2017/10/30/the-family-tha...