I was also under the impression that PoS merge addressed the fees but is actually a later update for sharding what is going to improve that. ETH changed a lot in the 6 months since I reviewed it for feasibility for my project.
There is a big economy build around L2s and that is not going anywhere either. Just for small money transfer ETH is overkill. The community have shared plenty of good tips, but probably the easiest is to just hold until you really need it so it pays the transfer itself with valuation.
Is all about perspective. $10 fee might be a lot transferring $70, but a $70 fee is nothing transferring thousands, millions. People were paying hundreds in just fees to mint NFTs back in the original craze.
So what point it is then to keep holding a coin with such high networks problems?
Oh don't misunderstand me I'm not defending Ether, neither I'm going to endorse any alternative, I don't hold anything and a $50 is not too little from my POV, is higher than just doing an overseas SWIFT transfer + Taxes + bank tax. Even if the L2s were the real solution, it goes back to your question. Then what's the point of holding ETH?
I'm probably biased because I'm developing an idea, but seems like the pragmatic answer is: the EVM and ecosystem. I wish, but I'll not be using ETH due to, precisely, the fees and congestion. Even the fees and constrains of an L2 are too limiting for what I want to so, so I'll be doing a sidechain, and that will eventually need a bridge, and then we are back to step 1: you'll need to hold ETH and likely the token of wherever you want to bridge.
And free market kinda doesn't apply completely here, by having more L2 fees should go down, but since those are intrinsic to the L2 tech, then having more L2s will only make it harder to lower fees since there will be more pressure to keep the L1 fees as they are.
ETH fees may lower someday, but it will not go that down as to nuke the L2 ecosystem, not happening.
