I don't think there is any contention over the idea that miners that control a larger percentage of the hash rate, h / H, have a profitability advantage if you hold all the other variables of the miner's profit equation constant. I think this is important: it is a centralizing factor similar to other economies of scale.

However, that is outside the scope of the result that an individual miner's profit per block is always maximized at a finite block size Q* if Shannon Entropy about each transaction is communicated during the block solution announcement. This result is important because it explains how a minimum fee density exists and it shows how miners cannot create enormous spam blocks for "no cost," for example.

2) Whether it's truly possible for a miner's marginal profit per unit of hash to decrease with increasing hashrate in some parametric regime.This however directly contradicts the assumption that an optimal hashrate exists beyond which the revenue per unit of hash v' < v if h' > h.
Q.E.D

This theorem in turn implies the following corollary:

COROLLARY: The marginal profit curve is a monotonically increasing of miner hashrate.

This simple theorem, suggested implicitly by Gmaxwell disproves any and all conclusions of my work. Most importantly, centralization pressures will always be present.