> You can't have fraud judged arbitrarily or allow anything regardless of how easy it makes it to commit fraud.
Why not, exactly?
Whatever rules you set up, you'll have false positives and false negatives. Some people will figure out how to do illegitimate things without being noticed by the rules. And some people will find the rules too burdensome and fail to do legitimate things aligned with the program's goals.
There is an old proverb in law: Better that ten guilty persons go free than that one innocent person is punished.
It might serve the program's goals better to decide, we're willing to allow some amount of fraud in order to make sure the program functions smoothly and people don't get trapped by rules, or we're willing to allow some amount of fraud because the cost of enforcement outweighs the fraud itself. (Major retailers work like this, for instance; they budget in some amount of expected shoplifting, because it's cheaper than both increasing security and potentially scaring off customers with increased and hostile security.) It might also not serve the program's goals better, and you want to take a hard line at the risk of blocking legitimate use. But it's not obvious the latter is the right approach.
Because we live in a democracy and everyone has access to the court system and you don't want to be in court all of the time. This program is funded by the taxpayer, after all.
And, yes, rules aren't perfect. But you can keep improving them as you learn more. And that leaves you with, yet again, a set of rules.
Why not, exactly?
Whatever rules you set up, you'll have false positives and false negatives. Some people will figure out how to do illegitimate things without being noticed by the rules. And some people will find the rules too burdensome and fail to do legitimate things aligned with the program's goals.
There is an old proverb in law: Better that ten guilty persons go free than that one innocent person is punished.
It might serve the program's goals better to decide, we're willing to allow some amount of fraud in order to make sure the program functions smoothly and people don't get trapped by rules, or we're willing to allow some amount of fraud because the cost of enforcement outweighs the fraud itself. (Major retailers work like this, for instance; they budget in some amount of expected shoplifting, because it's cheaper than both increasing security and potentially scaring off customers with increased and hostile security.) It might also not serve the program's goals better, and you want to take a hard line at the risk of blocking legitimate use. But it's not obvious the latter is the right approach.