I think you should look at point 3., slightly different from 1: everytime Google acts on something like this, they show that they can act. If the damage it cau

I think you should look at point 3., slightly different from 1: everytime Google acts on something like this, they show that they can act. If the damage it causes them, both in profit and image, is not very large then it might very well not be worth it to risk setting a precedent than Google is responsible for policing their ads. It is exactly the scenario they were caught in with YouTube, and becoming the police there didn't help them at all compared to the previous status quo. Of course that's the good scenario, the bad one would be that Google, being of a size and importance that can be called a monopoly in the Internet ad space, would be violating these "fraudsters" rights since technically what they do may not be breaking any laws, which would lead to this becoming a precedent and all the others who currently try to keep themselves mostly clean in their ads would be vindicated to go evil too. I'm not saying these are the real reasons, but again if the damage and risk they represent is insignificantly small relative to the whole, why would Google risk opening that can of worms.