And of course only a fool would cling to certainty on issues this complex. It's very possible my views are wrong. And if they are then that's a great outcome. If they're not? Well the cost of this experiment is not going to be that big, and it can be reversed relatively easily.
The one thing I think both sides often fail to do though is to set falsifiable metrics before running such an experiment. Take a group of qualified individuals against the program, and those for the program. And have them sit down and work out their expectations of the program. So for instance homelessness. The pro side probably expects this will cause it to decline, the neg side probably does not expect it to have much of an effect. Why not convert those views to numbers collected in a mutually agreed upon fashion, which can then be regularly published - alongside results? And let's see who's right, with accurate data. Of course there are confounders. If we hit a major recession, homelessness will increase regardless. These factors could be mentioned alongside such data.
Instead without doing things beforehand both sides are simply going to spin arguments entirely in their favor, such as with the case of minimum wage increases. Did minimum wage increases collapse the economies? No. Did they cause some economic damage? Yes. Did they bring about a great life for low earners? No. Did they improve the quality of life of some low earners? Yes. And so both sides just create disingenuous arguments painting themselves as 100% correct, which informs nobody and divides everybody, since both sides come back with 'told ya so!'
Citation needed. My admittedly cursory review of the latest results suggest no damage at all.
The reason I mention the Berkeley study at all is because this now led to arguments based on after-the-fact retrofitting. In particular increase supporters have now tried to argue that Seattle was in a unique boom phase and so the study was comparing against a local max, and thus there was actually no damage - but just Seattle 'regressing to the mean' if you will. But the Washington study also considered this possibility and contrasted it against other control cities within Washington that did not increase wages, and found that they did not experience the negative effects. And so on back and forth it goes.
And so who do you believe? Probably whoever you want to be right. This is why I'd suggest laying out metrics beforehand. If Seattle was in a unique boom phase before the minimum wage increases then this is something everybody could agree on beforehand. When you do after-the-fact analysis, you risk peoples biases getting in the way of things. And because of the complexity of systems such as these, it's always going to be pretty easy to prove X and also prove not X after the fact.
[1] - https://www.nytimes.com/2017/06/26/business/economy/seattle-...
