That is very well thought out.
A few things, What will be the incentive for a POS worker if the return is only 1-4% a year. If there is ether volitilty you could easily lose money for the year, all while having it tied up to service ethereum. Also, you can guarantee there willl be hacks on the POS pools, so there is another risk . I would think the return would need to compete with the stock market and real estate return percentages, 6-10%.
Great analysis , just some things to review.
Force
I was somewhat active in the Peercoin forum and there we have discussed that topic.
The crucial number to know would be the block reward for the Proof of Stake block. In the case you describe that we have 1 PoS block per 99 PoW blocks, the typical block reward a "PoS miner" could mine per timeframe (day, month, year) must also be divided by 100.
In Peercoin's model the PoS reward is so low (1% of your staking holdings per year) that the incentive to buy in is relatively low because you can't call this 1%/year a real "profit". Volatility in this case is comparable to a PoW coin like Bitcoin.
If the PoS reward is very high, then - according to my limited economic and game theory knowledge - the incentive to buy in is higher, but the incentive to sell early to take profits is high too, because of the higher coin supply inflation that would drive down the price in the long term. That would lead to a high volatility and a high probability of "pump and dump" games. That's why I never was interested in "high-PoS-reward" altcoins.
My conclusion would be: the higher the PoS block reward, the higher the volatility. In the model you describe, because of the low number of PoS blocks, I would say that the incidence of PoS to volatility because of a "spiralling demand" is low.