Yes and no but those are secondary to a common root. ttrpgs fall into lots of groupings but the important & relevant one here is pointbuy systems & class based systems. Pointbuy usually tries to value the cost to buy a given ability so that it's comparable to those with equal cost & better/worse than those with lower/higher cost & that works great but (in a nutshell) has problems with a higher bar to get started & a need for more oversight to prevent unchecked munchkinism that goes outside the spirit of expectations. Class based systems are usually easier to get into & provide some rails in the form of classes that allow characters to be a little more front loaded & not all equally balanced at every level. That's fine & good for both types as long as they have rules that work within & support their mix of strengths & weaknesses.Couldn't you reframe this as a problem with higher level class features? If level 1-2 abilities are more compelling and interesting than level 8-10 ones, then why shouldn't players prefer them? That feels more like you really need better mid-high level class abilities.
The only real stand out abilities (barring the coffeelock stuff that's mostly just poorly written short rest mechanics in the warlock than anything) are things like paladin smite that live outside the action economy. You could just move those back in (maybe introduce a 4th action type, like 3e's swift or 4e's minor actions) to give them a real cost.
The 5e style multiclassing tries to give players the strength of both without acknowledging the weaknesses of either. To a degree 3.x style did as well but at least tried to do things that raise the bar a little bit like oddball prerequisites all over the place & tools the gm could use to dial back on Bob's excess without crippling Alice's utter averageness on top of ways the GM could enhance Alice while being fairly confident the effort wouldn't be given to Bob to further widen the gap. (DR/SR/body slot conflicts/magic item churn/various forms of weapon specialization*/etc)
* the feat, crit fish many attacks with high crit range high crit multiplier weapons vrs fewer attacks with a freight train/etc
That's a good solution in pf2 but that solution depends on the syste,'s structural framework to make it work. I think that would just result in a quagmire of design incentives that ultimately avoids being a solution. It works in pathfinder2 because there are a lot more feats & taking something like bard dedication or the 4 feats that for of it & the advanced versions that fork of those you do so at the cost of things from your own class. 5e & so far oned&d don't have anything like that so it creates incentive to say "well this is a feat so it should give something cool" leaving you with something cool on top of an overly good front loaded class dip and all the stuff from their primary class.Still a bad idea as the base option. Make it a different option: medium multiclass: you need to take the 4th level multiclass bard feat to take levels in the bard class. No new system needed.