3 ms·

You can get a large inequality effect just from differences in inheritance between children. If we call the child who gets the bulk of the inheritance the "firs

by mrow84 10y ago

You can get a large inequality effect just from differences in inheritance between children. If we call the child who gets the bulk of the inheritance the "first child", then the first child of the first child, etc., will end up being very wealthy. You can see it in a very simple model: 1. start with a fixed size population, each starting with the same amount of wealth 2. pair the members of the population randomly to form two "parents" 3. combine the wealth of the two "parents" 4. split the wealth into two unequal parts with some fixed ratio (say 3:1) 5. assign the two parts to two "children" of the "parents", resulting in a population of the same size as the original If you repeat the above procedure (from step 2) for only 5 or 6 generations then you get a distribution that is positively skewed, to a degree depending on the ratio (3:1 gives quite a large skew). It is a simple model, and I have only evaluated it numerically, but it shows how strong an effect you can get from only a single process (differential inheritance in this case). I am fairly certain that things like assortative mating would produce similar (compounding) effects.

mrow84 10y ago
So I just did an experiment with simple assortative mating, where the population is drawn into an ordered list, with the probability of selection being proportional to wealth, and then paired off according to their order in the resulting list (so the wealthy are more likely to mate with each other). The results surprised me somewhat, though as with most things are somewhat obvious in retrospect. If you take a (model) world with unbiased inheritance (a 1:1 split between the two children), then no "mating strategy" is able to increase wealth inequality - at worst it can keep it the same as it was. This can be seen by noting that neither of the children of the two wealthiest members of the population can have more than the wealthiest of their two parents. Over many generations, unless the mating selection is extremely carefully chosen, the wealth distribution will collapse to perfect equality, for any mating strategy. Assortative mating therefore can at worst exert a drag on the (natural) restoration to equality. In a (model) world with biased inheritance this leads to greater amounts of wealth inequality, because the expansion of inequality through the unequal inheritance is "resisted" less by the reduction in inequality caused by mating. In conclusion: try and treat your kids the same if you want a more equal world!

by mrow84 10y ago

mrow84 10y ago

So I just did an experiment with simple assortative mating, where the population is drawn into an ordered list, with the probability of selection being proportional to wealth, and then paired off according to their order in the resulting list (so the wealthy are more likely to mate with each other). The results surprised me somewhat, though as with most things are somewhat obvious in retrospect. If you take a (model) world with unbiased inheritance (a 1:1 split between the two children), then no "mating strategy" is able to increase wealth inequality - at worst it can keep it the same as it was. This can be seen by noting that neither of the children of the two wealthiest members of the population can have more than the wealthiest of their two parents. Over many generations, unless the mating selection is extremely carefully chosen, the wealth distribution will collapse to perfect equality, for any mating strategy. Assortative mating therefore can at worst exert a drag on the (natural) restoration to equality. In a (model) world with biased inheritance this leads to greater amounts of wealth inequality, because the expansion of inequality through the unequal inheritance is "resisted" less by the reduction in inequality caused by mating. In conclusion: try and treat your kids the same if you want a more equal world!