• Google
    This Blog Web

October 2011

Sun Mon Tue Wed Thu Fri Sat
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          

RSS Feed

Bookmark and Share

Email Feed

  • Powered by FeedBlitz

« Self-Replicating Fab Lab | Main | Quarterly Report »

September 03, 2004


Feed You can follow this conversation by subscribing to the comment feed for this post.

John Michelsen

How about a global Bill of Rights? The development goals would take care of themselves if people were free of dictators and tyrannies.

1) Freedom of Religion
2) The right to bear arms
3) Consent to house soldiers
4) No unreasonable searches or seizures
5) No self incrimination, due process
6) Jury trial for all, public defense
7) In common law, right of jury trial
8) No excessive bail or cruel and unusual punishment
9) Constitution cannot deny rights of others
10) Governmental power default to the states

Janessa Ravenwood

I agree. The US (and it's Constitution) has certainly done more to help the world than that dictator's debating society called the UN ever has (or ever will). Every so often the UN can talk a good game, but they never actually DO anything beyond empty saber rattling - see Sudan.

Chris Phoenix, CRN

John, the bill of rights is a good start, but it's not enough. If people are hungry enough they tend to get uncivilized. Then they attack the rich, each other, or both. In any case, people hurt people, and the bill of rights is meaningless.

How to avoid people going hungry? For that, you have to design your economy and your government. There are a few basic principles, but I see three main variables: Wealth distribution; productivity; and population. The goal is to keep people at the tail of the wealth distribution from being hungry.

Caste and class systems tend to produce multiple wealth distributions and I would expect at least the lowest ones to be relatively flat.

Communism doesn't work. It may be viewed as an attempt to make a class system with one class (and relatively flat wealth distribution). But that's unstable, and tends to generate a kleptocracy or thugocracy; in addition, productivity suffers markedly in the absence of markets.

Markets tend to maximize productivity. Capitalism improves productivity where economies of scale are involved. The combination, where people can "pick the winners" (associate with richer entities), produces power-law wealth distributions.

In a power-law distribution, the poor really do get poorer as the number of people in the distribution increases. If the Nth person has 1/N of the resources (a hyperbola, exponent = 1), and each person brings one unit of resources, then a power-law curve of 10 people will have the head owning 3.4 and the tail 0.34. But with 100 people, the head owns 19.2 and the tail owns 0.19. With 10,000 people the tail is 0.10. With 10,0000,000 it's 0.06. This means that if two similar populations merge, the majority of people in them will be worse off. This appears to be true regardless of the exponent--but the effect is stronger for smaller exponents (e.g. 1/N^0.75); ironically, these are the ones that give more equitable distributions (with fatter tails).

There's something else interesting about power laws. Suppose there's a product that people will buy one of if they can afford it no matter what the price. Something like a life extension pill. If their resources are distributed according to hyperbolic power law, gross sales will be the same for any price inside the wealth distribution: the increase in sales exactly compensates for the decrease in price. (In a hyperbola, XY = constant.) With smaller exponents, you get more gross sales dolalrs by pricing the thing low enough that everyone can afford it. But with exponents greater than 1, you get more sales dollars the higher you price it. Ideally, you would make exactly one, and sell it to Bill Gates. Some Googling finds that tax-return data says that the exponent for U.S. and Japanese income distribution hovers around 1.5-2.5.

If the tail of the system's wealth distribution is too far into poverty, then the system needs a safety net or people will go hungry. That safety net can be regulating of necessary resources/infrastructures, welfare, charitable organizations, ... there are a lot of ways to do it. But I don't think it's safe to assume that a "successful" economy will lift the extreme tail of the distribution out of poverty.


John Michelsen

>John, the bill of rights is a good start, but it's not
>enough. If people are hungry enough they tend to get
>uncivilized. Then they attack the rich, each other, or both.
>In any case, people hurt people, and the bill of rights is

First I do not mean by "bill of rights" a few slips of paper
that can be ignored at will. I mean the system of government
which backs them up, that has the physical force and moral
authority to carry them out, to compel people to obey them.
2 allows for people to defend themselves, 5, 6, 7, and 8
are all about justice for those who hurt people.

In modern democratic systems of government such as the US,
hungry people generally do get help. There are plenty of
resources to spare, food is cheap, and people are generous.

In more backward anarchical and dictatorial societies such
as exist in Africa and North Korea, famine occurs because
of the lack of just government. Much food is shipped to
these places, but the powerful simply divert it to their
own ends.

How does it help a hungry person to give them food, just
so someone else can take it from them? Even teaching them
a profitable trade doesnt help if they are enslaved and
all their effort is siphoned off by someone else.

To say that the bill of rights is meaningless is quite a

>If the tail of the system's wealth distribution is too far
>into poverty, then the system needs a safety net or people
>will go hungry. That safety net can be regulating of
>necessary resources/infrastructures, welfare, charitable
>organizations, ... there are a lot of ways to do it. But I
>don't think it's safe to assume that a "successful" economy
>will lift the extreme tail of the distribution out of

Democracies are prone to progressive taxation. Because votes
are apportioned regardless of wealth, poor people tend to
vote to take money from rich people and give themselves a
safety net.


Chris Phoenix, CRN

Argh. I didn't say that our Bill of Rights is meaningless. I said that in a country with rampant poverty, any bill of rights they have will be meaningless. Even if the country has a few rich people, even if the average wealth per person is high, too much poverty leads to a social breakdown. Then the bill of rights will not protect them from being hurt by other people, in or out of the government (often their neighbors).

As for your "take money from rich people" argument, note that rich people manipulate the government to take money from poor people. In the interests of "keeping the economy strong," the rich write themselves all sorts of policy, from bailouts of failed funds, to slaps on the wrist for people who steal billions from stockholders, to porkbarrel spending that recycles that tax money right back to corporations. Given that this happens, is it better that the poor tax the rich somewhat, or that the wealth only flows one way?


John Michelsen

I understood your intent Chris - I was commenting on the case of countries with rampant poverty. Lack of just and enforced law leads to crime, which is very effective in keeping people in poverty. Give everyone in a crime ridden society a million dollars, and you will soon have a a few with billions and most with none. A bill of rights is essential to lifting people out of poverty for this reason.

I am aware of the extent to which rich people will bend the system into pretzels to get what they want. But the poor and middle class do it too. And the poor barely pay any taxes in the first place. 48% of goverment spending goes to Social Security, Medicare, and Income Security.

In this document from the CBO, they divide the share of total federal tax liabilities into 5 groups by income level. The lowest quintile of earners pay 1.1% of federal taxes. The second pay 5%. The middle pays 10%. The fourth pay 18.5% and the highest earners pay 65.3%. The top 10% of earners paid 50% of all federal taxes. The top 5% paid 38.5%. The top 1% paid 22.7%. Looks like a progressive tax structure to me.

Tom Craver

Chris: Your comments on "power laws" overlook one obvious thing - the total value available to "distribute" is not a simple constant multiple of population - it increases as humanity progresses. Nanotech will greatly further that trend.

In 2040, a homeless person might have cast-off clothes that keep them warm, dry and clean; a cheap bracelet that monitors their health; a canteen that purifies water just by shaking; and likely a cheap electronic entertainment unit. The merely poor might have virtual homes that seem like mansions, even though they really live in a closet sized space. They'll be poorer (smaller fraction of net wealth of a society) but richer (larger absolute well-being than today's poor).

And that's assuming that private individuals don't get direct access to nanotech (no free wish boxes), which would pretty much flatten the distribution curve as previously poor people poured out of the cities to squat on a patch of "junk" land or out on lakes and oceans, and live very well.

Tom Craver

John: WRT taxes - don't forget that people pay hidden taxes embedded in the cost of goods. I.e. corporations ultimately end up passing along taxes to consumers in higher prices.

From the stats at the IRS, it looks like the combination of employment taxes and corporate taxes is about 40% of the total. If 90% of that is ultimately paid by the bottom 80%, that's 36%, on top of the 35% they pay of the 60% individual income taxes (21%) - or about 57% of taxes coming from the bottom 80%. So it's not quite so skewed - and certainly not as skewed as the distribution of wealth.

Of course, the vast majority of the wealth of the rich is tied up in productive capital assets to the ultimate benefit of everyone, and the majority of income gets poured back into productive uses rather than into consumption or tying up resources (e.g. large mansions and estates). So no, I'm not advocating "soak the rich".

Still, it would be nice if some of the many barriers to wealth creation were eliminated - mostly regulations put in place nominally to protect the public, but mainly serving to disuade people from going into business for themselves in competition with established businesses. That's probably the #1 cause of sustained unequal wealth distribution, IMO.

Chris Phoenix, CRN

John, the U.S. Bill of Rights was not about creating an effective criminal law. It was about protecting citizens from their government. Both are important. My point was that simply protecting citizens from government is not enough to protect them from the consequences of poverty.

I'm not sure why you bring up progressive tax. I don't think this discussion is about that.

Tom, I know the rising-tide-lifts-all-boats argument. And I agree with it. But it is not the only factor. For example, if you have two populations with identical power-law distributions, and you merge the populations while keeping the same exponent, the tail of the distribution will get poorer.


Karl Gallagher

if you have two populations with identical power-law distributions, and you merge the populations while keeping the same exponent, the tail of the distribution will get poorer

In real life, a beggar doesn't get poorer if his country is merged with another one. Nor is a billionaire going to have double the wealth the moment the merger happens.

Tom Craver


Regarding your "product anyone will buy if they can afford it" scenario: You seem to be implying that only Bill Gates gets to buy the life-extension pill, while everyone else has to do without. But after Bill Gates buys his, the price can be dropped to attract more buyers. Eventually, the price will fall to where everyone can afford it.

Also, it usually makes sense to optimize short term income, since someone else will soon come out with a better life-extension pill. Unless Bill will pay more than N times as much as the next N buyers, you may want to price the pill to sell to Bill plus the next N buyers, even though you could theoretically sell to Bill at a higher price and later sell to the next N at the lower price.

Out-sourcing appears to be a fair example of your power-law effects of mixing populations. Advanced nation and developing nation working populations are merging. First for raw materials, then manufacturing, now for information workers. Average pay decreases, and wealth concentrates.

BUT! the merged populations still produce the same amount of value as before, so even though they have fewer dollars to spend (your power law effect), they will be able, on average, to afford to consume the same amount of goods and services as before the shift. By that measure, they aren't poorer.

By not wasting the talents of the workers in the developing nation, ultimately more will be produced and "consumable wealth per person" will increase. The workers of the developing nation will probably benefit the most from that, but that's a separate issue.

Chris Phoenix, CRN

Tom, you exactly missed my point. You said "Unless Bill will pay more than N times as much as the next N buyers, you may want to price the pill to sell to Bill plus the next N buyers..." But assuming that the amount people will pay is a fixed fraction of their total wealth (we're in spherical cow mode, OK) then the gross sales of selling to Bill alone will be *greater* than selling to Bill plus the next N buyers. This is true all along the curve. The higher the price, the smaller the N, but the greater the area under the rectangle. Counterintuitive, I know--but simple. An exponent of 1 is a hyperbola, in which XY = constant. With an exponent greater than 1, XY increases as X decreases.

I didn't understand your point about "the merged populations still produce the same amount of value as before, .... they will be able, on average, to afford to consume the same .... By that measure, they aren't poorer."

Your stipulation "on average" seems to change the topic. I am not worried about the average wealth of the population. I'm worried about the wealth of the lowest 80%. And I'm worried about strange pricing strategies that happen when the curve gets too steep, which imply that extreme artificial scarcity may be a strongly favored strategy--delaying the benefits of nano.


Tom Craver


No, I think I got your point - I just totally disagree that you are making the correct assumptions about the distribution of wealth. Take a look at www.forbes.com/lists/2003/02/26/billionaireland.html and click sort by wealth - Bill doesn't have 10x as much wealth as the next 10 people in line, which is what would have to be true for it to make sense to sell only to him, under your "fixed fraction of wealth" assumption. In fact, he has about 1/5th as much as them combined, so by selling only to him instead of the top 11 at half the price, you get about 1/5th as much income. I expect you could get 100M people to pay $100K - $10T vs maybe $25B from Bill. And as I pointed out, you COULD sell to Bill at $20B, and a little later to the next 10 at $10B, and so on, if you're willing to risk that no one else will come out with the same or better product.

As to my comments on consumption: The value of goods and services a person consumes is the best measure of that person's effective wealth - i.e. how well-off they are. So long as total production stays the same, the bottom 80% will consume about the same amount as before the out-sourcing - i.e. they are (on average for the 80% group) just about as well-off as before.

Factor in increased labor effectiveness ultimately creating more goods and services, and you understand why all the boats float a bit higher. This is EXACTLY what has actually happened - US and European quality of life has stagnated while that of Korea and China and India and elsewhere has skyrocketed. Not due to government policy or safety net policies - just "Heartless Capitalism" at work.

Another factor to take into account in thinking about wealth - most people have a major asset that conventional wealth comparisons ignore - themselves. If you consider a person's income as 'return on investment on human capital', the wealth distribution looks much different. A guy earning $20K a year with $1000 in the bank has a real worth of maybe $200K (figuring 10% ROI). A gal earning $100K a year, with $1M in hard assets, is worth about $2M - only about 10x as much, NOT 1000x as much.

Chris Phoenix, CRN

The Forbes distribution is for the world. If you sort out the US richest, you get a different story. I grabbed the top 300 richest people, sorted them into world and US, and for various indexes calculated the wealth at that index times the number of people with at least that much money (people who could pay X for, say, lifesaving medicine).

Rank 1 wealth 30.5 Sales 30.5
Rank 2 wealth 20.1 Sales 40.2
Rank 3 wealth 16.6 Sales 49.8
Rank 4 wealth 16.5 Sales 66.0
Rank 5 wealth 16.5 Sales 82.5
Rank 6 wealth 16.5 Sales 99.0
Rank 8 wealth 16.5 Sales 132.0
Rank 10 wealth 10.5 Sales 105.0
Rank 12 wealth 10.3 Sales 123.6
Rank 15 wealth 10 Sales 150
Rank 20 wealth 7.7 Sales 154.0
Rank 25 wealth 6 Sales 150
Rank 30 wealth 4.9 Sales 147.0
Rank 40 wealth 3.7 Sales 148.0
Rank 50 wealth 2.9 Sales 145.0
Rank 75 wealth 2.2 Sales 165.0
Rank 100 wealth 1.8 Sales 180.0
Rank 125 wealth 1.5 Sales 187.5

Rank 1 wealth 30.5 Sales 30.5
Rank 2 wealth 25.6 Sales 51.2
Rank 3 wealth 20.1 Sales 60.3
Rank 4 wealth 17.7 Sales 70.8
Rank 5 wealth 16.6 Sales 83.0
Rank 6 wealth 16.5 Sales 99.0
Rank 8 wealth 16.5 Sales 132.0
Rank 10 wealth 16.5 Sales 165.0
Rank 12 wealth 14 Sales 168
Rank 15 wealth 11.1 Sales 166.5
Rank 20 wealth 10 Sales 200
Rank 25 wealth 8 Sales 200
Rank 30 wealth 7.7 Sales 231.0
Rank 40 wealth 6.7 Sales 268.0
Rank 50 wealth 5.7 Sales 285.0
Rank 75 wealth 4.1 Sales 307.5
Rank 100 wealth 3.3 Sales 330.0
Rank 125 wealth 2.8 Sales 350.0
Rank 150 wealth 2.5 Sales 375.0
Rank 200 wealth 2 Sales 400
Rank 250 wealth 1.7 Sales 425.0

So the world seems to be distributed with an exponent less than 1; the gross sales increase as the cost decreases. In the US, it's inconclusive; note first that the sample size is small and at an extreme end of the graph, but note second that there is a bump in the middle of the curve.

I don't have time to think through the rest of your arguments right now; I'm busy launching a new website, and when that's done we'll be able to go there and hash it out in a better setting than a blog conversation. But I'd like to know whether the non-monotonicity of the US curve surprised you.


Tom Craver

I'm not surprised that there are variations from a simple curve. But Bill is 1 person with $40B or so. Even if the average assets of the rest of us are only $1000 apiece, there are 300M in the US - at $1000 a unit, it'd only take 15% of us to outweigh the maximum possible profit from Bill.

And again - it doesn't much matter - sell to Bill for $40B and to people like me for $1000, and double your net profits. That will apply even if nanotech results in one person owning more than half the world's wealth.

Chris Phoenix, CRN

Tom, that only works if the product can't easily be transferred--otherwise I'll sell my copy to Bill for $2,000.


The comments to this entry are closed.