Reporting on productivity change is a reasonable way to quantify the social impact of a phenomenon, which allows people to compare it in scale to other issues,

Reporting on productivity change is a reasonable way to quantify the social impact of a phenomenon, which allows people to compare it in scale to other issues, and prioritize scarce resources accordingly. Of course it's imperfect, and data is lost compared to a qualitative assessment of individual lives. But can you think of a consistently better quantitative measurement? Longevity? We've recently seen a lot of coverage of the forecast that climate change could cause a 10% reduction in GDP by 2100. Your objections apply to this measure equally. But by quantifying the magnitude we can compare it to things like the cost of remediation. Quantifying an effect also has the benefit of exposing the estimate to more effective refutation. The more numerate debate we subject policy preferences to, the stronger they'll be.