> And I don't think you clarified what James is saying.  He said that more 
> real wealth to billionaires does directly hurt poor people.  I'd like to 
> hear the chain of causality he has in mind.

According to Graham, "wealth is not money. Money is just a convenient way 
of trading one form of wealth for another. Wealth is the underlying 
stuff---the goods and services we buy." The underlying stuff, the goods 
and services, are constrained resources. For example, using a wealth 
resource in one way often prevents its use in a different way; real estate 
that is used to build a library can't be used to build a sports stadium. 
Wealth is also constrained by the rate at which it can be produced. There 
are a finite number of automobiles that can be produced per month. There 
are a finite number of hours that doctors can spend treating patients. 
These figures may improve over time, but they will still be finite. This 
means that many types of wealth are scarce. Ergo, distribution matters. In 
particular, skewed wealth distributions directly hurt poor people because 
there is a finite amount of wealth for everyone to share, and giving a 
unit of wealth to one person is equivalent to taking it away from someone 
else. Thus, the Daddy Model of Wealth is not totally broken. Wealth is not 
money, but many types of wealth *are* constrained by natural limits.

A society's wealth can grow over time, but it will never be infinite. 
Thus, there will never be enough wealth to maximize everyone's utility. 
But given diminishing utility returns on wealth accumulation, sound public 
policy should ensure that wealth imbalances do not grow too large.

~j