Applying my selfish filter, today, I don’t need covenants. And I personally don’t expect to need them for a long time (if ever). BUT, they allow increased density at the UTXO layer, which (hopefully) means lower fees for me, whether I use them or not.

I think of it kind of like a chartered bus. If a bus line decides to operate, it makes denser use of the shared road than if the bus didn’t exist. So either more people get to travel, or there’s less congestion (than if those passengers were all in cars), or potentially both. The existence of busses makes the road more valuable and my drive less congested whether I ride the bus or not.

It’s not a perfect analogy (nothing ever is), but that’s how I justify it to myself.

The alternative to densification is either different tradeoffs (like Lightning) or, more likely, trust-based intermediaries. The value of Bitcoin derives from its censorship resistance, so staving off centralized third parties is in my interest (even if I don’t personally use those features).

Final note: arguments of this sort have been made about Drivechains—that they increase the value by making denser use of mainchain block space whether mainchainers use sidechains or not. But this argument doesn’t apply because of the increased incentives for miner centralization. Covenants do not have this changes-the-incentives drawback.