> All policies which affect the wealth distribution do so by moderating the gini coef, ie., the specific shape of the power-law -- they do not make it normal, uniform, etc. or however the ideologue would wish it to be.
Again, you are assuming your conclusion, and you do not even do us the favour of stating premises, which is hardly conducive to a productive discussion (or, any, really). On the other hand, it is quite fundamental to welfare economics that under a few assumptions, some unrealistic and some less so (namely: perfect information, fully internalised externalities, absence of market power, local nonsatiation and convexity of preferences and production) it is possible to reach any point on the Pareto frontier. The policies required to reach a desired point in such a toy model is left as an exercise for the reader. Now, the actual juicy questions here are how things like market power, externalities and imperfect information affect this? Is the model still a good approximation? To what extent are we able to correct things, and do we even want to? If we do, how best to do it?
> The "natural mechanism" of "fair procedures" in the case of systems wherein parts attach to other parts based on their current size
Again, you assert that it's "natural" and "fair" as if the result is a foregone conclusion. If it is inevitable that dx be a f(x) ~ x then yes, that's an exponential. Why would it be unnatural or unfair for dx to be instead a function of y, some other factor that has no proportionality to x? You are stating as a conclusion and premise that financial wealth experiences preferential attachment unconditionally, that is in essence, circular reasoning. What is the actual mechanism, the force, and how irresistible is it really? Would resisting it simply divert its effects to something else, for example, concentration of political rather than economic power?
Quite frankly, the dismissal of everything outside of a frankly ridiculously narrow space, where your model with an unspecified mechanism (which does achieve the results that you say it does) clearly must be an accurate reflection of the real world in all circumstances... Well, it gets old quite fast. If you really must use that model, describe and justify the accuracy of your proposed mechanism at least, if you please.
In some sense i'm assuming wealth is a pref-A phenomenon, yes; not that "capitalism makes it so". There need be no formal argument here: all research into all wealth distributions in history are power-law; and a vast array of social phenomena are pref-A. And pref-A produces power-law.
As an empirical matter, the burden of proof lies with those who say "there is some other process" to show it. Since the PA process is demonstrated without exception.
Formally, why should we expect wealth to be a PA process?
Well, formal arguments will always be toy and disappointing, but get someway to explaining what's observed empirically. But:
Economic transactions are, roughly, exchanges of time. We have a fixed quantity of time each, and Alice offers some time to Bob, who offers some time to Eve... and so on.
Wealth is your net ability to draw down on the time of others, as a function of time. By spending X amount, I can obtain Y time from P people.
Assume, at t=0, everyone has an equal ability to draw down on the time of others, and an equal preference for exchanging their time (1hr for whatever it takes me to do, for 1 hr of whatever it takes you to do...).
Now, since we have different rates of production and produce different things it becomes wasteful for me to spend 1hr for your 1hr when you dont make what i want; or someone else makes more of it faster.
Our desire not to expend our own time needlessly, to waste our finite quantity of it, means we allocate it to where the ratio from ours/others is most efficient.
That is to those who can draw down on the time of others most, we give some of our own; and to those who can the least, we give none. (To, roughly, quote Matthew).
Simply in the pursuit of not having our 100 years wasted egregiously. And there are many compounding effects, not least this. For as soon as any inequality enters those who then have more can do more, and so become ever more effective at what they do.
We will not work a month to buy a t-shirt when a t-shirt can be made in a factory for a minute of our time. It makes no difference that the factory is owned big Capital (omg!">?!?). And if we were to that, there could be no "factories" of this kind -- since allocating resouces according to what minises wasting our time gives us more time, without which we cannot develop anything new.
All this is somewhat irrelevant however. As above, the burden is on those who say something else is conceivable to show (1) what their conception is; and (2) it is physically possible.
If all alternatives amount to global coordinaton across the species such that we are each programmed Marxists acting like drones -- or, likewise, non-agression libertarians -- you're then talking about a different species.
Wealth amongst the bees might be a little more normally distributed, for sure.
No, you've made the claim that that it's impossible to alter, I'm fairly sure it's up to you to show that one of imperfect information, externalities, market power, ~LNS or non-convexity of preferences or production result in the impossibility in the real world of producing a single even slightly different distribution when you're making a universal quantification on what is certainly a non-null hypothesis (never mind the abuse of mathematics that is drawing a straight line on a log-log plot and calling it a day, I know it's a good approximation for almost everything but so do you that it often fails to hold for the whole domain if you've ever took a single maths class) which otherwise clearly contradicts a well established fundamental result of economics. This is not a new theorem, if you're making the case that it cannot possibly hold in the real world in any way under any conceivable circumstances then you really ought to have a rock solid case already prepared, one that stands up to slightly more scrutiny than looking at any graph, literally any single one of them, and seeing that there is clearly a non-linear region there. You know, where all the people actually are. Maybe if you had the time to take a second math class you'd know that being able to fit a power law to the upper tail of a distribution is utterly meaningless considering you can fit it to the upper tail of basically any empirical distribution. You can go for sub- and superlinear PA schemes in your third one. The questions posed continue to be uninteresting, and they don't even have the honour of being accurate.
I've already conceded that a small tribe could do it. It's not entirely clear that they have 'wealth' at all, since i'd define it to be that thing which accumulates 'alike pref-A' (ie., I will concede that processes which disrupt/regulate social networks produce, at best, power-laws in a theoretical limit).
But the issue is that small tribes arent stable. The primary objective of tribes is foremost survival, and in the pursuit of every more gauretees of survival, they will pref attach to other stronger tribes (often by concequest, occasionally by consent, see eg., the formation of the UK vs. the formation of the british empire).
So there are different 'social logics' taking place at each level: the individual wants to minimize the amount of their life wasted; the group wants to maximise its chances of survival etc.
It's not that I do not see 'local violations' of pref-A-type expectations; they're quite common indeed.
It's more that I see the 'effect size' from pref-A processes to be, empirically, overwhelming. A force of nature at every scale.
People who are "anti-capitalist", ie., the person i was addressing, arent offering an analysis of the entire history of the human species in terms of these processes --- and how alternatives can be provided; rather they are profoundly ignorant of all this, have no interest in learning it, and are rather engaged in a religious crusade against their notion of sin.
The first step to help this ideologue out of their crusade, is if even possible, to show that it isnt Sin which makes the world Fallen -- it Falls even under extremely simple, extremely fair procedures. This is intolerable to the fantatic whose crusade is against a human psychic reality they thing they can attack -- tilting at windmills, of course.
If we're talking "two good-faith people who arent fantatics trying to model possible realitites" then the dialetic is different.
I'd say: i have sets of entirely plausible propositions each of which entail 'inevitable extreme inequality', each of which plausibly model reality, each of which have high credences from a range of experts (etc.).
I do not see the same on the other side; this may be my ignorance, but I have never encountered a systematic presentation of any explanatorily-plausible alternative.
How on earth do you propose to hault 2*2*2*2*2... vs. 0.001*1.001*1.0001* ... ?
By what means are most individuals on the planet coordinated so that the labours of one do not multiply benefit themselves, but "divisionally benefit" everyone? What process to the general benefit of society could be imagined where net wealth grows without inequality?
THe only "divisional processes" i can imagine not only reduce net wealth, they are horsemen of the apocalypse.
> concede that processes which disrupt/regulate social networks produce, at best, power-laws in a theoretical limit
I'm going to interpret that as acknowledging that the bulk of the distribution is not Pareto with Pareto fitting well only at the end of the right tail. This fits with a mostly multiplicative dNW/dt but in a random walk rather than monotonic increase, or a sublinear PA, but with increasing linearity at the very high end. Not that we actually need to model this blindly, dNW/dt is to a significant extent just Y, income, and we can and do measure income, at least. A lot more accurately than wealth, even. More importantly, there is enough data with both for the same individuals that we can examine the relationship between the two. Certainly, there is a positive correlation, but overall it seems R ~ 0.33 across many countries, which is fairly low, really, and if you want to eyeball the graph you can probably see the data points clustered tighter near the top.
Now, I can't tell you what distribution @linschn might target as desired or if there are things beyond mere distribution that they would wish to address. My reading of their comment would be that they wish to focus on ensuring non-NW effects >> NW effects, i.e. high mobility, not necessarily high equality (does raise an interesting question here, would a combination of mobility and inequality also tend to increase zero-sum vs positive-sum competition?) Still, given that the correlation, although low, does exist, any actions increase to mobility does seem likely to reduce final inequality as a side effect. This does not necessarily shift the shape of the distribution (rather than spread) significantly, but there is no reason why it should need to.
> How on earth do you propose to hault 22222... vs. 0.0011.0011.0001* ... ? By what means are most individuals on the planet coordinated so that the labours of one do not multiply benefit themselves, but "divisionally benefit" everyone? What process to the general benefit of society could be imagined where net wealth grows without inequality?
I don't see why anyone should care for growth of total NW over Y. Wealth isn't real, there is not literally a warehouse full of stuff and services that wealthy people can draw upon, so it can only be a claim on future Y plus or minus inflation. I entirely agree with @linschn in this respect, seeing growth in wealth being higher than growth in income as a good thing is complete and utter madness, and yet those are the scenarios where the PA or PA-like effects relatively dominate. If during relative stagnation it's a choice between infinite paper growth (plus increasing concentration) and the apocalypse, let's have the apocalypse. At least the apocalypse is honest about what it is. In a way, it's the Greek sense of the word.
Again, you are assuming your conclusion, and you do not even do us the favour of stating premises, which is hardly conducive to a productive discussion (or, any, really). On the other hand, it is quite fundamental to welfare economics that under a few assumptions, some unrealistic and some less so (namely: perfect information, fully internalised externalities, absence of market power, local nonsatiation and convexity of preferences and production) it is possible to reach any point on the Pareto frontier. The policies required to reach a desired point in such a toy model is left as an exercise for the reader. Now, the actual juicy questions here are how things like market power, externalities and imperfect information affect this? Is the model still a good approximation? To what extent are we able to correct things, and do we even want to? If we do, how best to do it?
> The "natural mechanism" of "fair procedures" in the case of systems wherein parts attach to other parts based on their current size
Again, you assert that it's "natural" and "fair" as if the result is a foregone conclusion. If it is inevitable that dx be a f(x) ~ x then yes, that's an exponential. Why would it be unnatural or unfair for dx to be instead a function of y, some other factor that has no proportionality to x? You are stating as a conclusion and premise that financial wealth experiences preferential attachment unconditionally, that is in essence, circular reasoning. What is the actual mechanism, the force, and how irresistible is it really? Would resisting it simply divert its effects to something else, for example, concentration of political rather than economic power?
Quite frankly, the dismissal of everything outside of a frankly ridiculously narrow space, where your model with an unspecified mechanism (which does achieve the results that you say it does) clearly must be an accurate reflection of the real world in all circumstances... Well, it gets old quite fast. If you really must use that model, describe and justify the accuracy of your proposed mechanism at least, if you please.