The answer was something called marginalism. Well worth reading about, and I can't do justice to it in a few sentences.
Of course marginalism itself requires you to think about value in a certain way, often a very technical way (eg von Neuman-Morgenstern) that makes you forget that there are sometimes contradictions in how we think about value.
Perhaps the concept of "social value" is only useful in the context of societies small enough so that all participants can agree on values?
WRT. doctors vs. drug/diagnostic equipment manufacturers, a good point, but here I specifically picked my example - public health and safety vs. entertainment - to maximize the contrast.
> Perhaps the concept of "social value" is only useful in the context of societies small enough so that all participants can agree on values?
I think there might be a correlation between community size and the ease of agreeing on what is social value, but I wouldn't discount it for larger societies or even humanity as a whole. Humans do have a lot in common, and I feel you can get far in ranking jobs by social value by going along the gradient of reduction of suffering. And I say "jobs", not "professions", because IMO, professions are almost orthogonal to social value. For example, in terms of the social value, I'd rank a programmer working on diagnostic equipment > programmer working on a videogame > programmer working on advertising tech. The skillset and productivity may be similar or even the same, but these jobs occupy three very different point on the "making the world a better place" scale.
Not necessarily. I agree that having values dictated by fiat is a recipe for disaster - such arrangements are bound to get corrupted over time. But on the other end of the spectrum, there is some sense of social value - a sense in which e.g. garbage collectors, firemen and doctors do more socially valuable jobs than athletes and entertainers. A sense that's easy to convey - imagine if suddenly, any one of the former three occupations disappeared; the death toll and suffering would quickly overshadow the largest of wars. Meanwhile, even if both of the latter occupations disappeared, the culture would be a bit poorer, but the society would function just fine.
Unfortunately, it seems that this concept of social value is very hard to robustly define. And if we can't define it, effectively optimizing for it is pretty much out of the question.