you lost me there, UPS meaning the one with a battery backup?
wall-draw can vary depending on load, you should've compared it with system-draw:wall-draw as a ratio from before vs after.
if the system-draw is the same while wall-draw dropped... well thats weird since your PSU suddenly became more efficient! thats a good thing by the way.
but if the system-draw dropped while wall-draw is the same, or system-draw is the same but wall-draw increased should mean that efficiency decreased.
as for powersupply max capacity declining, you can only monitor that if your PSU puts out a magic smoke when you near the declined capacity, or simply became unstable (BSOD or crash) when you near the declined capacity.
e.g. before PSU=1300watts, system-draw=1000watts. now PSU=900watts, system crashes at system-draw=901watts.
edit: this scenario is far from being common, if you ask about capacitor aging from an EE or ECE, capacitor aging only affects capacitance and ESR.
the former doesn't reach a value of zero, at worst they can only reach down to 50% for defective capacitors but on average they only drop down by 25% for quality-passed capacitors. now only the consistency of the power output gets influenced by this capacity decline, meaning it's ripple values will increase.
the latter does affect the PSU's max capacity the most, if the ESR rises too much outside of it's given specs it may cause a decrease in overall ampere output.
now if the manufacturer designed the PSU with quality and durability in mind, they will factor in these issues on their design and shouldn't cause much problem.
one proof of this is that "most" quality PSUs does have a marginal overhead outside of their max rating, notice that 400watt rated PSUs can still run at 450watt, or even 500watt for loose OCP.
edit2: now seriously, why do people get concerned with this mythical "capacitor aging", i've yet to hear of corsair or seasonic dying because of such...