Personally (which I'm probably extremely biased as this is my idea) I want a balancing act, and intentionally chose to design the concept with that at the core. As I've said, this isn't finalized (especially the numbers) but I think that the balancing act could be done realistically and add to gameplay as long as modifying these numbers can be done in a variety of ways. For example, say the occupation lowers the rations the population is getting. Now, the population's not as happy and their livelihood has gone down. If the happiness was getting too high, this could help you as the population is less content and more likely to join. If livelihood was getting low, this could mean the citizens start dying at higher rates, but also joining at higher rates with higher rates of malnutrition. In order to respond to this, the resistance could use its funds to smuggle in extra food and either keep its agents better fed to keep them healthy for operations (not too expensive to do, but won't help citizens) or they could start distributing food to the population, increasing both livelihood and hope, increasing recruitment rate as hope has increased and keeping soldiers at higher combat efficiency.
One important note:
The main stats will not be tied to eachother proportionally. If one changes, the others do not change. Only if an event/trend that has affected multiple had occurred will multiple change, and then each will be affected individually.
Edit: As the occupation will work separate from the population, their main stats may intermingle somewhat, especially as supplies are a huge part of the occupation's presence and ability to fight you. Namely, their livelyhood is what will affect the other 2. This will increase the effect that supplies have on the soldiers because supply shipments will be a huge part of espionage system I'm planning. It also makes the occupation more dynamic and volatile, hopefully increasing how much of a threat they are.