As an end note most of your modders here who tend to be fairly knowledgable and have done research have altered many of the games Manpower issues. If it was correct you would think they would just keep them the same as in vanilla...
I'm not entirely sure what your trying to say here, but what I can tell you is that the manpower distributions are somewhat 'arbitary' in my mind. It's not that minors are nerfed per se all the time, it's the fact that recruitment laws and national mobilisation in my opinion doesn't particularly well represent the historical mobilisation for war.
Writing without a source handy (so numbers might be slightly wrong). Britian in 1938 and 1939 (pre-war) only had an increase in milltary recruitment of about 30% then 45% on the years before. Overral, volenteerism came to about 60,000 IIRC. 60,000 men (assuming 1:1 ratio) is enough manpower for only 20 new divisions over an almost two year period. Factor in the years 1937 and 1936 at roughtly simmilar levels of low level volenterism, then we get an extra 30,000 souls. Therefore all pre-war MP gains in Manpower with historical mobilisation timescales, should only be enough for about 30 new divisions. I take this as an example because it was one that I roughtly can recall numbers for (although they may be somewhat off).
The big point is....it wasn't only Britian.
iWith the exception of Japan, Germany and to some degree the Soviet Union, most nations had very low manpower gains in the pre-war period, which then exploded in 1940 as the nations were forced into war, and then generally stayed steady, or declined in some nations, and others like the SU it expanded.
There is also the concept of women in industry that freed up a massive about of manpower in the UK, US and Soviet Union. Germany on the otherhand was very late to the game.
Then you have nation specific levels of war support. The Japanese could count on almost every single young man in its nation. In the US however Blacks were often prevented from serving or had to serve in special 'coloured units'. Other nations like the Soviet Union basically press ganged people into the armed forces at times, while in India Gandhis 'Quit India movement' sparked a massive resentment in the Indian population to actually take part in Britians war.
As such moddeling Manpower must be a function of many seperate issues, while in general the 'base population' should follow the historical populations of various regions;
1. Neutral Nations: Their populations generally didn't want to fight or mobilise ahead of actually being at war (Major MP nerf)
2. Strategic Doctrine: Nations that optted for national deterrance / general conscription had more manpower (Large MP buff)
3. Industrilisation: The Higher the Infra level and IC per MP province, the greater percentage of people can serve (at the best this is still a tiny fraction of total avalible manpower and this percentage should be in the 0-20% margin)
4. Recruitment Policies: Before war (large threat nations too) manpower gains are strongly linked to laws, in general the Vanilla laws give too big a pre-war bonus as one enters 'full mobilisation' and simmilar while not at war. While at war you should in general get a one time bonus to 'call in the reserves' while the 'trickle in' population by year should never exceed about 1.6% of total manpower avalible. This is because year on year cohorts entering the age to serve in the military are only a fraction of total population and even 1% is a gross overestimate.
5. Nationalism/Repression: The higher both of these are the more you can convince people to join up (again small numbers)
6. National Unity: This is a rather large gripe of mine, since the smaller nations of the war often had more per capita join up to defend their nations, and because they are smaller had more sense of group identity. Therefore minors should in general find it easier to keep high unity, and have a minor manpower boost. This could be easily done on number of provinces I believe.
These are in principle going to be the way I am changing my game to be more 'accurate' but I've yet to decide on speicifics as research needs to continue to be done.