I believe one common argument is that if the overcurrent protection doesn't serve as earth leakage protection disconnection times can be much longer, even going into the thermal overload range of the fuse/MCB rather than the magnetic trip. That's often cited with laypeople plugging in multiple extension leads, occasionally exceeding 50 m of 1.5 mm2. Besides, most wiring regs worldwide only apply to fixed wiring and flexes are governed by different standards (that obviously need to take fixed wiring designs into consideration).
C16 is fairly common for commercial and even domestic socket circuits in Austria where sockets and lights are on separate circuits. If there are mixed circuits it's usually B or sometimes C13, mostly out of habit. Around 1980 the regs were updated with reduced ampacity tables, banning the use of 1.5 mm2 and 16 A L/U trip curve MCBs and Diazed fuses under most circumstances (except cables directly buried in masonry walls, concrete or run underground). That update was followed by the introduction of L/U 12 A MCBs that quickly became the de-facto standard for domestic general-purpose circuits and small commercial lighting circuits. The introduction of B/C curve MCBs in the early 1990s saw slightly higher ratings due to the faster overload trip (1.45 times rated current rather than 1.6-2.1 depending on the size) but most electricians had grown accustomed to the smaller size and went with B13. C13 are rare, although sometimes helpful because even a regular home PC can trip a B-curve MCB if plugged into a switched power strip along with a lot of peripherals and then switched on at the power strip. That's even true for B16.
The German regs only had a more moderate ampacity reduction and manufacturers eventually offered huge discounts on B16 MCBs due to the quantities they're sold in - a B16 single pole is usually 1.99 while a B10 is between 4.99 and 9.99 there.
Yeah, there's only so much we can do... (Though I would like if the standards considered fixed wiring and flexible cords in conjunction, rather than separately.)
I wish product engineers would include better inrush limiting, too. (Generally, they seem to have been concerned mostly just with protecting the product itself from damage.) Anyway, if we want to discuss that further I'd suggest going to another thread...