One important aspect is that we may be embedded in a vicious circle of discontinuity events and smoothing. A broad expectation that smoothing will be applied af

One important aspect is that we may be embedded in a vicious circle of discontinuity events and smoothing. A broad expectation that smoothing will be applied after a discontinuity event will tend to make people more willing to accept the future risk of discontinuity events in their decision making. Less risk-avoiding behavior regarding discontinuity events makes discontinuity events more likely. Then, when the next discontinuity event happens, if smoothing is again applied, it reinforces this expectation, which reinforces the probability of future discontinuities, etc. For this reason, swearing off active smoothing could actually result in a smoother system, eventually (after the pain of transition, of course).