That's a idea I've suggested several times. I call it the MISS factor. It stands for Military Inefficiency Spending Syndrome. It's a factor that all prices are increased by to account for everything from bureaucracy to corruption. It can be a s low as 1.0 for militaries that have been engaged in serious fighting for a long time and goes up in increments of 0.1. Factors that increase MISS is size of organization (measured by size of budget) and decades since last serious conflict (brushfires don't count), and goverment type. I had a suggestion typed up many years ago, but I don't seem to have the file any more.
I've been implementing T4PE recently and having some thoughts along similar lines, although I've been favoring a reduction in usable (pocket) imperial income based on the log or Nth root (for an as-yet-undetermined N) of the number of worlds in the empire rather than a straight multiplier.
Upkeep and support is already factored in. And there's one kind of naval vessel we can't increase the cost of: patrol vessels about twice the size and armament of the usual PC ship.
It's easy enough to argue for leaving the cost of patrol vessels alone while inflating the prices of battlewagons: The patrol ships are civilian police vessels, not front-line military combat ships, so they don't need to be built to front-line-worthy military specs.
Only by decreasing the size of the population. The percentage of gross product that a population can sustain in the long and the short run is quite well known from Real Life.
Another option (which I'm also experimenting with) is to change the scaling factor for population. I don't have TCS, but T4PE scales GWP linearly with the actual planetary population, so +1 to the UWP pop code multiplies GWP by 10. Make it non-linear and it compresses the range in a way that I rather like. I don't know that I'd be able to do a very good job of defending it as realistic, but it brings the average down quite a bit as well as raising the income of lower-pop worlds somewhat. (My main motivation behind this was to keep a handful of high-pop worlds from completely dominating everything and making all other worlds irrelevant. The reduction in average GWPs was merely a happy side-effect.)