That 10% faster blocks can be "used" only once, so if this week miners buy 10% more hardware, it's already included in the hashrate 2 weeks from now. Of course, in reality the hashrate is changing gradually, but a higher Bitcoin price makes more difference than the (current) transaction fees.
Just to be clear, I was talking about the short-term, as, within the current individual difficulty epoch, the number of transactions should have the same impact on the hashrate as the price does, anything that transfers to more profit = more hashrate = faster blocks.
Also, I am not talking about miners buying new mining gears and adding them to the network, because that can't really affect how things are in the short-term, but I am talking about the automated process of turning mining gears off and on as well as the profit switching which many if not the majority of miners use when fees on bitcoin are super high compared to the other SHA256 coins, a large amount of that hashrate will switch to mining bitcoin in a few minutes, that could potentially help with speeding the process of finding blocks for a certain period of time.
Long-term, the hashrate needs exponential growth to sustain 10% faster blocks.
True for long-term, the hashrate growth needs to beat the difficulty adjustment which isn't feasble.