You have to be very wary of the fact that the Gini coefficient is a unique characteristic of a distribution, not of an economy. You could calculate a Gini coefficient of the number of hairs on people’s heads. Very commonly people confuse the Gini coefficient of the distribution of income with the Gini coefficient of the distribution of wealth.
Gini coefficients of income get into differences between the distribution of income before the effect of taxes and transfers and after taxes and transfers, also, the difference between the distribution over individuals and the distribution over households. Then there is the effect of transfers-in-kind such as public education and free health care, and public services such as effective policing that obviate expenses rather than increasing what is recognised as income.
As for Gini coefficients of wealth, they are fraught with problems with valuations and exclusions. Home mortgages are customarily included but home values are not (which stupidly counts middle-class people with millions in home equity as being poorer than the destitute homeless. Student debt is counted in countries that have it, but the value of professional skills and oligopoly licences is not.
And then, of course, there is the issue of wealth and income that don’t officially below to any one person, but which may have very different effects on real material inequality, depending on what benefits they produce for whom.
The idea of making up a scale with limited discrimination and qualitative definitions is probably compelling. I must get around to it some day. Perhaps using star and half-star emoji.
Well, the idea of “total wealth” is inherently problematic, because you could assess the value of any one item of wealth, tangible or intangible, by asking how much people would take to give it up or would pay to keep it, but you cannot do that for all the assets in a society at once, so it’s not operationalizable. Even at the level of one country, I suppose you could imagine selling all the assets of Australia and its people to some other country, but after you did so, Australia would not be a going concern, so it would be sort of like a bankruptcy sale, I think. You can go on the basis that every asset was at some time bought, sold, or produced, but that only works if you assume that value is conserved over time. . . .
You get a robust and informative result by totalling the sums that would hypothetically be realised if each asset were sold alone, because in the usual ideal circumstances the return to each asset is the value of its marginal product
Do you? You can’t sell all those assets alone AT THE SAME TIME. If you only sell small samples at each time, you have to assume that no meaningful change in value occurs over the span required to sell them all; alternative, if you keep selling till everything is sold, the marginal utility of the later assets will have been altered by the very fact that you’ve sold the earlier ones.
It seems like what my old math professor Errett Bishop would have called a “nonconstructive proof.”
No, but then you never will and don’t want to, either. So what would happen if you tried isn’t of much interest.
The income (possibly imputed) that the owners get from having the assets, the power that each of them has to liquidate it with the economy as a going concern is related by way of the discount rate to the value of its marginal product.
The marginal value of selling just one of the assets is, I think, what you want, because that’s what an individual actor is likely to do. Sure, if George Soros sells all his shares and buys gold, that’ll have a huge effect on the markets and he won’t realise anything like the potential value of selling a small stake multiplied by the number of stakes; but that doesn’t make him not a very rich person.
Yes, but if, to oversimplify ruthlessly, George Soros has N assets, for a very large N, and each single asset is worth $x if sold by itself, it seems meaningless to say that his total wealth is $xN. If he starts selling assets, the value of the remaining assets will change. When he’s sold a substantial number, the value of the rest will change A LOT.
We have this thing with illiquid assets all the time. You can’t realise their whole value quickly, but they still produce an income (perhaps in kind, or imputed), which has a certain net present value at the discount rate determined by the market. The owners enjoy the income of an asset with the stated value. They won’t (unless irrational, distressed, or under duress) sell them for less than the full value. Assets have uses other than to be liquidated, and they have values in those uses, and the total of those values is informative.
I’m not happy with this design of Navabharata’s culture or this formulation of its values and taboos. It keeps turning into a caricature of India, which I don’t want. The social units based on occupation have to go. So far I have the following, but I’m groping ineffectually with the “taboos”.
Law & enforcement
source
customary
enforcement
bailiffs & investigating magistrate
powers & resources
3/10
judgement
magistrate
procedure
investigative
protections
1/10
penalties
death
for major felonies
flogging, hard labour
for lesser felonies
caning, fines
for misdemeanours
peculiar laws
death penalty for “blasphemy” and “sacrilege”
telecoms strictly licensed
high-tech vehicles restricted
serfdom
retaliation against “intolerable provocation” is permitted