If the usual thing sufficed to inform people, well, it’s already out there, so there’s no point in doing yet another website. The new site should show something new. Also, 50%-40%-10% might be an observed pattern but is it fundamental? Who knows? I don’t study this topic but I’m used to the idea of a power law distribution so I’d like to see whether the relevant graphs look that way. Mostly though I’d like to see an axis that’s linear by percentile, to get a sense of how wealth is distrbuted across the population, rather than in arbitrary buckets of differing sizes.
Use separate graphs or colored lines for different time periods.
The slider is to let people look at counterfactual situations where the distribution was different (or became what it is) due to some (imagined) intervention in the past. If we say there’s no point to doing that, i.e. the data in the graphs is an unavoidable consequence of a “financial field theory” where dollar bills stick to other dollar bills and end up concentrating in a few places like charged particles, then why is either economics or history even interesting? I’d certainly want to see what happens if you can move a slider, but of course that’s just me.
That said, if gini doesn’t capture the top peak (I gather gini is a different parametrization of a power law curve, though I haven’t gotten around to checking into that) then I wonder if there’s some simple model that better fits reality. And again I’d still want a slider, because I don’t know what else to do with the info in the graphs.
~10% wealthy is stable. Middle class existing is recent. Who knows: pretty well explained in Piketty’s capital and ideology (there are interviews on youtube summarizing the book if you want details.) Basically it’s more or less stable, and the given reason for it differs with each society, so it is a social choice.
The rest is mostly you having issues reading the graphs and wanting them to show something else than what their purpose is.
Edit:
Searched but couldn’t find a simple overview of why a single parameter analysis like the gini fails. It is mostly a debate amongst scholars (so harder to read), which led to the whole we are the 99% movement when Piketty first raised the idea to the general public.
Easiest to understand I found was this, showing 2 very different model societies with the same gini, but the blog goes further with an analysis in don’t really find useful of adequate other than this graph.
The rest is mostly you having issues reading the graphs and wanting them to show something else than what their purpose is.
Well, the graphs have their purpose, and readers also have purposes which might not coincide with the graphs’. I can’t speak for other readers but I found the graphs to be unenlightening about whatever points were being made. The blog post about gini looks good and indicates that I need to get a clearer understanding of how gini works. I’ll see if I can read the wikipedia article about gini. Thanks.
Thinking about this a little more, that particular example doesn’t seem important. I’d like to know more about the shape change at the high income part of the curve you mention. That would be more significant. It still sounds like you’d get a curve that you can describe with two parameters instead of a single parameter, still not that big a deal.
Pikkety’s observation was that invested wealth tended to grow faster than the overall economy, if I have it right. So that would give some exponential growth at the high end that’s not present in other sectors. It’s not mathematically obvious to me that this makes the distribution stop following a power law. Maybe it does, either necessarily or as an observational matter, I just don’t see a quick proof either way. It would at minimum change the shape of the Lorenz curve (i.e. change the exponent aka shape parameter of the curve) but that’s still a power law. So I’d be interested to know what economists are saying about this.
I still haven’t read up on GINI enough to see if it means something different from power law.
If the usual thing sufficed to inform people, well, it’s already out there, so there’s no point in doing yet another website. The new site should show something new. Also, 50%-40%-10% might be an observed pattern but is it fundamental? Who knows? I don’t study this topic but I’m used to the idea of a power law distribution so I’d like to see whether the relevant graphs look that way. Mostly though I’d like to see an axis that’s linear by percentile, to get a sense of how wealth is distrbuted across the population, rather than in arbitrary buckets of differing sizes.
Use separate graphs or colored lines for different time periods.
The slider is to let people look at counterfactual situations where the distribution was different (or became what it is) due to some (imagined) intervention in the past. If we say there’s no point to doing that, i.e. the data in the graphs is an unavoidable consequence of a “financial field theory” where dollar bills stick to other dollar bills and end up concentrating in a few places like charged particles, then why is either economics or history even interesting? I’d certainly want to see what happens if you can move a slider, but of course that’s just me.
That said, if gini doesn’t capture the top peak (I gather gini is a different parametrization of a power law curve, though I haven’t gotten around to checking into that) then I wonder if there’s some simple model that better fits reality. And again I’d still want a slider, because I don’t know what else to do with the info in the graphs.
The rest is mostly you having issues reading the graphs and wanting them to show something else than what their purpose is.
Edit:
Searched but couldn’t find a simple overview of why a single parameter analysis like the gini fails. It is mostly a debate amongst scholars (so harder to read), which led to the whole we are the 99% movement when Piketty first raised the idea to the general public.
Easiest to understand I found was this, showing 2 very different model societies with the same gini, but the blog goes further with an analysis in don’t really find useful of adequate other than this graph.
Well, the graphs have their purpose, and readers also have purposes which might not coincide with the graphs’. I can’t speak for other readers but I found the graphs to be unenlightening about whatever points were being made. The blog post about gini looks good and indicates that I need to get a clearer understanding of how gini works. I’ll see if I can read the wikipedia article about gini. Thanks.
Thinking about this a little more, that particular example doesn’t seem important. I’d like to know more about the shape change at the high income part of the curve you mention. That would be more significant. It still sounds like you’d get a curve that you can describe with two parameters instead of a single parameter, still not that big a deal.
Pikkety’s observation was that invested wealth tended to grow faster than the overall economy, if I have it right. So that would give some exponential growth at the high end that’s not present in other sectors. It’s not mathematically obvious to me that this makes the distribution stop following a power law. Maybe it does, either necessarily or as an observational matter, I just don’t see a quick proof either way. It would at minimum change the shape of the Lorenz curve (i.e. change the exponent aka shape parameter of the curve) but that’s still a power law. So I’d be interested to know what economists are saying about this.
I still haven’t read up on GINI enough to see if it means something different from power law.