If you pay spherical cows in a vacuum really high wages, then sure, you're completely right. But a lot of models have funny behavior when you get closer to the

If you pay spherical cows in a vacuum really high wages, then sure, you're completely right. But a lot of models have funny behavior when you get closer to the edges. I think you can intuit that if you pay people some amount they actually work more efficiently because they're happier and healthier and better fed and better educated. Of course if you pay them a very large amount, the efficiency gains per dollar will likely drop off again because there's only so much a person can improve themselves (as per your intuition before) . So it turns out you're really dealing with an optimization problem. I think a lot of people just assume the model behavior is linear and end up picking a point far too low down on the curve. (People have had some clue for a long time though: consider the saying "Penny wise, pound foolish")