Comment by myrmidon
11 days ago
> Top marginal tax rate was surprisingly not too correlated to government revenue
Yes. Which is interesting, but also makes sense if you assume that a frequent goal is to shift the tax allocation between wealth classes (adjustments to top rate would be somewhat compensated by other changes).
I think it is always too easy to find arguments for almost any position in data like this, because the overall picture changes dramatically over just a few decades; wealth/income percentiles become qualitatively different as GDP grows ("workers class" pre WW2 is quite different from the same income percentile now) and the data is noisy too, so if you squint you can interpret almost anything in there.
In a perfect world, we would have twenty identical Americas with fixed tax policies, and be able to compare their development over decades; what we have is instead a bunch of different nations radically changing their behavior basically every time a different government comes into power, and many conclusions are inevitably just educated guesswork.
Totally in agreement that we always read a lot from the conclusions based on data. Data often obfuscates.
I do think, however, that empiricism is a better framework for grounding outcomes in reality than pure ideology. Pure ideology (either way) is usually just confirms biases by cherry picking data.