Yea, the conversion barrier is a big one. The argument rested solely on the "expected value" of applying versus not, where E[applying] = E[marginal value] - E[m

Yea, the conversion barrier is a big one. The argument rested solely on the "expected value" of applying versus not, where E[applying] = E[marginal value] - E[marginal cost] = Pr(Will get a job and take it) * $4166 - $100. So if Pr(Will get a job and take it) >= 100/4166, then it would make sense to apply. Of course, most people would not think like that, which is the problem.