Vernon Smith recommended this 1992 article [PDF] to me:

This analysis indicates that reference price formation does have significant effects on consumer behavior. Furthermore, these effects are asymmetric with consumers two and a half times more responsive to egg price increases that are in excess of the reference price than they are to comparable egg price decreases.

If you do not think that the response to egg price increases is important (demand falls more quickly when prices rise than it rises when prices fall), then you may want to apply this result to other market goods (share prices, houses, etc.) in your thinking of how we are going to respond to a change in current economic conditions — especially if (5%+) inflation returns to the US.

In relation to water prices, this result gives us reason to suspect that higher prices will lead to strong contractions in demand (not quite as strong as 2.5 times, since there are fewer substitutes for water than for eggs…)

Bottom Line: We respond more strongly to (real/nominal) price increases than decreases. Plan accordingly.