The number of ways that your solution can fail are innumerable:
- Your solution may not be possible to implement at all.
- Your solution may violate human nature. [Gender Neutrality programs]
- Your solution may only be implementable in pilot state. [Education reforms?]
- Your solution may require more long-term will/discipline than is available [Democratic War]
- Your solution may not actually impact the problem. [ Race-disparity in education]
- Your solution may not solve the problem even though it impacts it.
- Your solution may solve the problem for a short period of time, until reality routes around your solution. [ Internet Censorship]
- Your solution may make the problem worse instead of solving it.
- Your solution may solve the problem at too great a cost (to other values, including $).
- Your solution may improve the situation now, but lock in a solution that prevents even better solutions, thus creating a net worse result over time. [Seat Belt Law]
- Your solution may do exactly what you thought it would do (in predicted terms), and you may still prefer the old state.
- Your solution may solve the problem you wanted to solve, and create a different (worse) problem. [ AFDC ].
- Your solution may solve the problem for a very narrow subset of the relevant group, and screw everyone else.
- You may almost solve the problem (with nearly zero net welfare gain), but not be able to find the control levers that would actually solve the problem.
The only game in town is how to arrange the system to allow/encourage trial, error, and observation...and to improve the feedback loop.