Imagine if the punishment for stealing was to be fined for 10% of the amount you stole. Like, you get to keep 90% of it.
You'd rob a bank and then drop by the police station to drop off their cut just to avoid the hassle of "getting investigated" later.
So obviously the punishment has to be worse than the value of doing it. But not 1:1 either, if you know you'll only catch 50% of robbers, then the fine would have to be something like ${amountStolen}*3 so that the equation makes the Expected return of robbery sharply negative. I.e. if you expect to get caught 50% of the time, and only had to return the money you stole when caught... then robbery would be very profitable. If you had to pay twice what you stole, then the Expected return is actually a wash, and wouldn't necessarily deter potential robbers. However if you have to pay a fine of 3x what you stole and you think you'll get caught 50% of the time then it's a money-losing venture.
This is all ignoring jail time, of course, because no one can send Google to jail... so from the corporations point of view it just comes down to money.
At $1.5B, it's about collecting revenue to fund EU activities. It's far beyond the amount of revenue the infraction generated by order(s?) of magnitude, and was a gray area to boot.
Keeping the taxes as "fines" ensures they can give EU companies favorable treatment. It's the definition of corrupt: "a willingness to act dishonestly in return for money"
In European law, jail times and fines are multiplied by consecutive infringements to be punished. Meaning if you stole three times in a row and get caught after the third, you get sentenced three times the theft sentence.
You'd rob a bank and then drop by the police station to drop off their cut just to avoid the hassle of "getting investigated" later.
So obviously the punishment has to be worse than the value of doing it. But not 1:1 either, if you know you'll only catch 50% of robbers, then the fine would have to be something like ${amountStolen}*3 so that the equation makes the Expected return of robbery sharply negative. I.e. if you expect to get caught 50% of the time, and only had to return the money you stole when caught... then robbery would be very profitable. If you had to pay twice what you stole, then the Expected return is actually a wash, and wouldn't necessarily deter potential robbers. However if you have to pay a fine of 3x what you stole and you think you'll get caught 50% of the time then it's a money-losing venture.
This is all ignoring jail time, of course, because no one can send Google to jail... so from the corporations point of view it just comes down to money.