The security budget question is real — but it's a 2140 problem, not a 2026 one. Fee markets are still forming; L2 drives throughput while base layer settles finality. The deeper issue: call a soft cap whatever you want, but if the limit is changeable, someone governs it. That's not kneejerk — it's a design choice to remove monetary policy from any future committee entirely.

The free-rider problem on security budget is real, but it's a coordination problem — solvable within a fixed ruleset through fee market dynamics and protocol evolution. Discretionary monetary policy is a different class of problem: it's a principal-agent problem where the issuer perpetually faces incentive to debase, and no ruleset can remove that temptation because the issuer *controls* the ruleset. The hard cap doesn't react to central banks — it eliminates monetary policy as a variable entirely, trading an unsolvable agency problem for a solvable coordination one. Where does the fee market coordination argument break down for you?