I suppose the answer is 'it depends'. There have been advances in the past 10-15 years in terms of material science, computer modeling, and even in the scientific understanding of how we process sound and the measurements needed during loudspeaker development to optimize them for that.
Some companies embraced science in loudspeaker development far earlier than others. Harman, Paradigm, and PSB are great examples of companies that were science-forward before the crowd, and their older designs hold up better, at least IMO, than companies whose approach was 'throw a bunch of fancy drivers and high end components in a box' without regard to baffle interaction and diffraction, dispersion characteristics (either overall or between drivers at the crossover points), power step response, or smooth frequency response.
I'd use the Revel Salon2 as an example of a speaker that was designed right from the beginning. Even though they're 10 years old at this point Revel hasn't felt the need to redesign them because so far nothing has beaten them in Harman's blind listening tests in their sound lab, even speakers costing considerably more. In fact, they've held their own so well that they're about to undergo a 20% price increase with no substantive changes other than some new finishes becoming available. You can find them pre-owned for a nice savings, and I'd consider that a solid value.
At the same time, you could compare them to a new product like the Revel F228be which goes new for roughly what a pre-owned pair of Salon2 does, features a lot of trickle-down technology from the Ultima2 line such as the beryllium tweeter and improved crossover networks, and has things the Ultima2 line doesn't such as the new deep ceramic composite cones. Revel says the Salon2 still wins in listening tests, but the F228be closed the gap considerably from the F208, and the F228be comes brand-spankin'-new with a full warranty.
A great resource to see measurements of a number of speakers both new and older is Soundstage's list of speakers they've tested at the Canadian NRC's anechoic chamber. Since the research from Harman's Dr. Toole has shown that listener preference is very strongly correlated with not only a smooth on-axis response, but on a smooth off-axis response as well, even at extreme angles so that the first reflections and reverberant field don't create peaks or nulls, it's a great way to get an idea of well designed speakers that may be worth a closer look.
https://www.soundstagenetwork.com/index.php?option=com_content&view=article&id=16&Itemid...