Corroboration!

A while ago, I made a post which argued for reverse causality in the search for the source of the US’ widening income inequality. In that post, I made this point:

What is it that European countries do? Massive income redistribution. That may seem superficial, but it’s the answer that I’m most happy with. It has long been known to network theorists that competitive networks (with a return to scale to node connection) are characterized by power law distributions. This is a natural phenomenon; it happens in the blogosphere, the financial markets, in sports, and it happens in economies as a whole. Left to its own devices, it is inevitable that such networks will evolve an shockingly large disparity between the best-performing actors, and the mean actor.

Lane Kenworthy, who writes a lot about inequality issues, has a recent post on his blog (and a subsequent link to a longer article in a U of Arizona journal) which seems to corroborate the story I told above:

What about in an absolute sense? Would the incomes of low-end households have grown more rapidly in the absence of the top-heavy rise in inequality? If we look across the rich nations, it turns out that there is no relationship between changes in income inequality and changes in the absolute incomes of low-end households. The reason is that income growth for poor households has come almost entirely via increases in net government transfers, and the degree to which governments have increased transfers seems to have been unaffected by changes in income inequality.

I was probably subconsciously channeling my inner Kenworthy when I wrote the previous post, as I read his blog sporadically. If I had it my way, the discussion would break along three lines:

1. Wealth Inequality.
2. Income Inequality.
3. Consumption Inequality

And on three lines, we would be able to analyze poverty. What would be the value in this? We would more easily be able to define public policy on narrower grounds. For example, wealth distribution is always going to be sharply unequal. People with higher incomes relative to consumption will always be able to amass more wealth than lower income groups (wealth meaning savings + assets, wealth is something I think many people conflate with income). However, what is the best policy for addressing this aspect of the problem? Simple income transfers won’t work alone, we need to incentivize intergenerational savings among the poor. In that light, Social Security is not the most optimal program, as it discourages savings.

I tend to not really focus on income inequality, but it does feed into the more important aspect (in my view); consumption inequality. Are poor peoples’ consumption patterns keeping up with what we as a society would consider some measure of a “quality” standard of living? Are they able to afford necessities like electricity, HVAC, food, and medical care? If not, this is where we get to fiddle with income in my most preferred way (and the way which allowed other countries’ poor to keep up with growth): simple transfers. Look for efficiency on the supply side, but on the demand side, just augment income. This could be in the form of a universal deduction for income under a certain level, or cash or voucher transfers. It is relatively cheap and easy to structure these transfers in a way that incentivizes future-time orientation…indeed, that is what the Mexican organization Opportunidad does.

I realize that democracy is much messier than this, but it is something to work toward. The welfare state need not be cumbersome…indeed, the US is unique in the world in the inefficient ways it implements a patchwork system. Canadian Philosopher and pop-economics writer Joseph Heath has said that the US government gets away with it because we don’t redistribute a whole heckuva lot of income…but as that enterprise grows, we’ll inevitably have to address our abhorrent inefficiency issues.

Addendum: Lane Kenworthy has also done some good work on the question of measuring poverty and material wellbeing, so check that out as well. Seems the real barrier to constructing a time series from this data is that it doesn’t go back far enough to create a reliable measure of wellbeing over time.