The author's electricity bill went up and his cat got stolen in part because his colleagues working under the university incentive systems (i.e. don't publish stuff that pisses off the interests that fund your lab) created work that legitimized those policy decisions so that those decisions could be made and the funding interests, whatever they may be, could benefit from them.
One wonders if there are similar incentives in the university ranking, administration and consulting that legitimize the university's otherwise questionable decision to engage in these seemingly irresponsible ventures.
Nobody is waterboarding the money down their throat. They can say no. The actual question is: why don't they?
What is the downside to the school of a nicer student union or a public policy/international relations campus in the nation's capital?
So the researcher intentionally changes some of the ways the data is collected and poof, it looks like the policy works. Extra funding comes your way but now you have committed academic fraud. Not that anything will happen to you for this, but still, you know you did it. That's what the GP is talking about and it happens quite a bit in the humanities and economics. Its why private economists and public economists almost seem like different species.