It is a bit more complex than that.
Assume that the offending behavior is worth $X to the offender. Also assume that the regulatory fine levied for that behavior is $Y. Further assume that $Z is the cost for the offender to achieve regulatory compliance. For the sake of argument, $W is the cost of resisting the regulator (including fines).
The benefit of passive non-compliance is $X - $Y.
The benefit of compliance is -$Z.
The benefit of active resistance is $X - $W.
A rational actor will choose the option with greatest benefit, so $Y > $X + $Z will encourage compliance over non-compliance. If $W > $X + $Z, that also encourages compliance over resistance. In either case, it is apparent that $Z amount is working against the regulator.
As long as the benefit achieved from continuing the behavior plus the benefit of not having to pay the cost to stop doing it exceeds the external costs of continuing, it will not stop.
But there's more. Assume $V to be the cost of resuming the behavior after it stops. Now, in order to achieve nonzero $X, $V must be paid.
Now the benefit of non-compliance is $X - $V - $Y.
The benefit of compliance is $0.
The benefit of resistance is $X - $V - $W.
That puts a higher cost on the other side of the equation. the ability to rip out, confiscate, and destroy the equipment forces the actor to discount the cost of just unplugging all of its devices, which is what allows the fine to be matched to the benefit from continuing the activity.
Thus, the FCC can look at the mean daily income realized from jamming and set that as the initial guess for a persuasive daily fine. But the FCC also has to make the total cost of fighting them in court higher than just paying the fines, otherwise everybody would fight.