Paul Krugman: Piketty’s big idea is that we are in the early stages of returning to a society dominated by great dynastic fortunes, by inherited wealth. … Imagine a wealthy family that has managed, somehow or other, to guarantee that a large fraction of its income is used to accumulate more wealth. Can this family thereby acquire a dominant position in society?
Why are we all implicitly assuming that "dynasties" is a static and homogeneous blob when in fact experience and data indicate that it is not just volatile at the aggregate level ("dynasties"), but extremely volatile at the individual level ("dynasty")?
Doesn't matter if it's families, corporations or countries we're talking about, the general rule is that what goes up, probably will come down.
If the issue is concentration of wealth in the hands of a few, does it matter if those few are related by blood and marriage, or contract, or geography?
r did outpace g most of the time since the invention of agriculture, but that was indeed only the case on average, periodic resets did happen. Still, inequality had grown to huge proportions in some civilizations like the Roman Empire and Tsarist Russia. If there are less wars and major catastrophes in the future then the resets may not happen often enough anymore to keep dynastic wealthfrom running off. In addition we may see something like a genetic divide increase r relative to g as well.
I think that problem is often circumvented by locking the wealth into companies and real estate that mostly don't get divided but all go to a single heir. A noticeable exception would be the Saudi royal family that keeps handing out money to more and more people.
I have problems with Piketty's graph. First, it is odd distorting the x-axis by having one unit represent 1000 years then 500 years, then 200 years, shrinking to 90 years and eventually 50 years. And does Piketty really have reliable data to show the slight increase in growth going back 2000 years? I know that Greg Clark has said Agnus Maddison's data prior to 1800 isn't worth anything.
That document discusses the rule requiring foundations to make "charitable expenditures" of at least 5% of their assets each year. The penalty is 30% of the additional amount that should have been spent, which is assessed against only the foundation and is not considered a crime by its members. Could the foundation simply pay the penalty, which would amount to a 1.5% yearly tax on the endowment, and continue to grow?
A technical note, r and g are actually random distributions that we're using shorthand to denote Mean(r) and Mean(g). It's possible that Mean(r) > Mean(g), yet dynastic wealth still not outpace growth. A simple example, assume g reliably was 50% every century. And r was either a doubling or halving with equal probability every century. Mean(r) = 75%/century > Mean(g) = 50%/century. Yet over the long run dynastic wealth will stay static and the economy will grow.
In financial parlance this is called volatility drag. And I don't think it's a trivial contribution to the problem. Dynastic wealth really probably is much more volatile than overall global economic growth. This is true for a multitude of reasons, including dynasties being overly aggressive in risk-taking (which is probably how they got rich in the first place), dynasties being more subject to local economic conditions and not globally diversified, as well as real returns to capital being much more volatile than economic growth rates.
Given the fact that rich dynastic families have considerably less power now than perhaps any other time since the dawn of civilization, it seems puzzling that this would be a concern.
Dear all,
Why are we all implicitly assuming that "dynasties" is a static and homogeneous blob when in fact experience and data indicate that it is not just volatile at the aggregate level ("dynasties"), but extremely volatile at the individual level ("dynasty")?
Doesn't matter if it's families, corporations or countries we're talking about, the general rule is that what goes up, probably will come down.
Just my 2c.
Whatever it takes to make sure that the accepted wisdom is correct...
Family may be too small a concept here.Sovereign Wealth Funds are the country-level equivalent of dynasties:http://www.swfinstitute.org...
As are multinational corporations:http://www.economist.com/bl...
If the issue is concentration of wealth in the hands of a few, does it matter if those few are related by blood and marriage, or contract, or geography?
Aren't you ignoring the demand side in your assumption that the correct policy will maximize investment?
Aka "No, you can't take over the world by compound interest. No matter how low your time preference."
That is one of the chief points of the book. R generally outpaces g, but, in the 20th century, events like you describe allowed g to dominate.
r did outpace g most of the time since the invention of agriculture, but that was indeed only the case on average, periodic resets did happen. Still, inequality had grown to huge proportions in some civilizations like the Roman Empire and Tsarist Russia. If there are less wars and major catastrophes in the future then the resets may not happen often enough anymore to keep dynastic wealthfrom running off. In addition we may see something like a genetic divide increase r relative to g as well.
I think that problem is often circumvented by locking the wealth into companies and real estate that mostly don't get divided but all go to a single heir. A noticeable exception would be the Saudi royal family that keeps handing out money to more and more people.
I think the idea behind the article acknowledges just what you said, but suggests that this is an anomaly and that we shouldn't get used to it.
I have problems with Piketty's graph. First, it is odd distorting the x-axis by having one unit represent 1000 years then 500 years, then 200 years, shrinking to 90 years and eventually 50 years. And does Piketty really have reliable data to show the slight increase in growth going back 2000 years? I know that Greg Clark has said Agnus Maddison's data prior to 1800 isn't worth anything.
I expect that if a foundation successfully grew via blatantly and consistently violating this rule, the penalty would be increased.
That document discusses the rule requiring foundations to make "charitable expenditures" of at least 5% of their assets each year. The penalty is 30% of the additional amount that should have been spent, which is assessed against only the foundation and is not considered a crime by its members. Could the foundation simply pay the penalty, which would amount to a 1.5% yearly tax on the endowment, and continue to grow?
it makes more sense to talk about ln(1+r) and ln(1+g). The long run outcomes do track the means of those quantities, without volatility drag.
A technical note, r and g are actually random distributions that we're using shorthand to denote Mean(r) and Mean(g). It's possible that Mean(r) > Mean(g), yet dynastic wealth still not outpace growth. A simple example, assume g reliably was 50% every century. And r was either a doubling or halving with equal probability every century. Mean(r) = 75%/century > Mean(g) = 50%/century. Yet over the long run dynastic wealth will stay static and the economy will grow.
In financial parlance this is called volatility drag. And I don't think it's a trivial contribution to the problem. Dynastic wealth really probably is much more volatile than overall global economic growth. This is true for a multitude of reasons, including dynasties being overly aggressive in risk-taking (which is probably how they got rich in the first place), dynasties being more subject to local economic conditions and not globally diversified, as well as real returns to capital being much more volatile than economic growth rates.
A trust that is instructed to keep investing and not spend is far more likely to actually do that than handing the money to a next of kin.
Given the fact that rich dynastic families have considerably less power now than perhaps any other time since the dawn of civilization, it seems puzzling that this would be a concern.