It's true that economics is not amenable to experiments as reliably informative as those of, say, physics. And it's true that, all else being equal, the less re

It's true that economics is not amenable to experiments as reliably informative as those of, say, physics. And it's true that, all else being equal, the less reliable your experiments the less weight their results should get relative to theoretical predictions. But it isn't only the experiments that are less reliable in economics than in physics. The theory is less reliable too. (For two reasons. Firstly, because it's so hard to do experiments whose results you can trust, the evidential basis for the theory is weaker. Secondly, because people are more complicated than, say, electrons.) How does this cash out in the case of the minimum wage? We have some empirical evidence suggesting that minimum wage increases don't increase unemployment much if at all, but we can't trust it too much for the usual reasons (not enough data, too many confounding factors, small effect sizes). And we have some simple theoretical arguments suggesting that minimum wage increases should increase unemployment quite a bit (supply and demand from Econ 101), but we can't trust those much either because the real people involved in the labour market are not the perfectly rational perfectly selfish perfectly spherical agents with which Econ 101 is concerned. It simply isn't true that increasing cost always reduces demand: there are multiple mechanisms by which in particular circumstances it can do the reverse. It simply isn't true that the labour market behaves exactly like the market for, say, corn. People are complicated. So I don't agree with what I take to be your argument about the minimum wage, namely that because we don't have great empirical evidence we must rely on theory and firmly believe that increasing the minimum wage will severely increase unemployment, and that this means that an honest supporter of minimum wage increases is obliged to accept that their preferred policy will increase unemployment substantially. Incidentally, I'm not even so sure about that pebble thrown into the ocean. The sea-level rise is a tiny effect, after all, so we can't just ignore other tiny effects. For instance: a slight variation in the shape of the earth, with less mass on land and more on the ocean bottom; a tiny change in ocean currents, which might change erosion patterns and change the volume of the oceans one way or another; maybe some funny effect via disturbing local ecosystems and ultimately affecting human food supplies a little. These are all really small, but so is the obvious effect of throwing the pebble in. For that matter, if you're on a boat when you throw the pebble in, the pebble goes from displacing its own mass of water (thank you, Archimedes) to displacing its own volume of water, which is smaller, so the immediate first-order effect is that the water level goes down.