Why Local Factors Matter

Kindred Winecoff (via Brad DeLong) has some thoughts on inequality:

1. Global inequality, which was expanding from 1800-1975, and receding since (due primarily to China and India).

2. US inequality, which was declining from 1925-1980, but increasing since, in two ways:

2.a. The separation of the top 20% from the bottom 80%, partially as a function of college wage premium.

2.b. The separation of the top 0.1% from the rest of the top 20%, for reasons unexplained.

3. The rise of the top 1% globally — the development of a new international plutocracy — since… 1980 or so?

For reasons that escape me he does not link the three together. It seems an obvious thing to do. It’s what Oscar Landerretche (the other participant in the discussion) does, in the context of explaining outcomes in Chile. But he doesn’t do it.

This is something that has disappointed me about the conversation regarding American inequality over the past few years more generally. (I’ve written about it before.) Folks like DeLong, Krugman, Cowen, and others think in global terms quite frequently. But when they seek to explain the Great Stagnation and the rise of inequality they concern themselves almost entirely with local explanations. I think such analyses are very likely to suffer from omitted variable bias.

Around the 43rd minute, DeLong takes a direct question about this from the audience. He answers it fairly well, but still downplays the role of the global economy in influencing outcomes in the US, favoring cultural explanations (e.g. explosion of CEO pay) and changes to marginal tax rates. He does not consider that these might be related to global dynamics either. If global forces eroded the bargaining power of American Labor, then maybe that’s why society has tolerated the enormous increase in executive pay. If global forces are pressuring American corporations through new competition, then maybe we’ve cut top-end taxation in an attempt to gain back some competitiveness.

This takes a highly simplistic view of inequality in the world. The reason you can’t just “link” the three together are captured in this one, little chart from Conceição and Ferreira:

The GINI Index is intuitive but ultimately handcuffed in how much it can tell us about the economy. Especially when we’re talking about the world. A far more robust measure is the so-called Theil Index. The math behind the measure (between 0 and 1) requires a fair understanding of information theory but the idea is lower index implies a higher economic “entropy”.

Your physics teacher might tell you that this is a bad thing but, economically, it’s a little more complex. As Boltzmann showed, entropy increases as predictability of an event decreases. This means the entropy of a fair coin is higher than a biased one. Similarly, in a very equal economy it is very difficult to distinguish between two earners based only on their income. Indeed in a perfectly equal society this is impossible. However, as society stratifies itself, knowledge of ones income conveys far more information (redundancy), thereby decreasing entropy.

Within a system, Theil makes it easy for econometricians to understand the amount of total inequality due to within-group inequality and across-group inequality. If this is a little hard to grasp, think about it this way. If the total differences in economic output remained constant between countries (that is, India is still poor and Norway rich) but income was equally distributed within each country the residual inequality would be the “across-country” inequality. The residual from the converse, where all countries remain as unequal as they were, but world economic output is distributed equally to countries (not people), represents the “within-country” inequality.

So what’s my point? While global inequality has been on the decline (because of, as DeLong notes, China and India) the inequality within Asia is about 27% more significant than it was in 1970. This means DeLong’s answer is pretty accurate. While the GINI shows a worrisome increase in American inequality over the past thirty years, its contribution to global inequality hovers around 9%.

There are strictly local factors that can be attributed to increases in American inequality. Sure, global competition puts a downward pressure on American wages but this comes from basic trade and, theory tells us, that this increases output for both parties. This means the American pie is growing (and, clearly, it has) and bad policies are inhibiting the necessary redistribution.

Winecoff also conveniently ignores that the median American worker has become exponentially more productive which would, in normal circumstances, imply higher wages. Alas, local factors have prevented this from happening.

DeLong is very smart in not extrapolating meaningless inferences from his simple observations. If what Winecoff asserts is correct, American worker bargaining power is undermined by a smaller between-country inequality but, relatively, it is the within-country inequality that has grown and, most significantly, in Asia.

American CEOs are earning 300x their median worker is nothing more than the fraying separation between Chief Executive and Chairman resulting in the extraction of severe rents. Lower marginal tax rates on the rent-earning minority have no effect on global competitiveness but explain pretty well the growing importance of within-country inequality.

Edit: Winecoff comments below:

So first of all, trade is not a “strictly local factor”. That’s my point. And what economic theory tells us is NOT that trade is redistributed equally; far from it. Comparatively advantaged sectors/factors benefit from trade, while comparatively disadvantaged sectors/factors suffer. What I’m suggesting is that the American comparative advantage changed post-1970 or so. Previously, it was advantaged in manufacturing, and post-GATT trade liberalization helped it exploit those advantages. Post-1970s or so, the collapse of Bretton Woods capital controls and pursuit of export-oriented industrialization efforts in Asia and elsewhere, the US comparative advantage shifted to finance and other high-capital output (e.g. information tech, pharma, etc.). These require smaller inputs of labor, so the share of national income going to labor declined. Hence: increasing domestic inequality in the US, without a concomitant rise in global inequality (since trade benefits all societies in aggregate).

I’m not denying any of this, in fact I pretty clearly noted above that trade in the last decade or so has accrued in the hands of a relative few. America’s competitive advantage doesn’t benefit the middle man, anymore. But this is not cause to throw our hands up and call it a day, because trade is “global”. There are plenty of things we can do at home. If America’s competitive advantage is exporting math/science know-how, then we can make it a lot easier for people from poorer families to get a STEM degree. If the Ivy League is our competitive advantage, we can take a lesson from this BPEA report and encourage low-income students to apply to top colleges.

In fact, over the past thirty years, as our competitive advantage shifted, we did precisely the opposite of what we ought to have done. It is hard to fight against a falling wage share of GDP. In fact, economically, this is probably a good thing. However, we should have fought hard against the growing inequality within wages themselves.

A side note: I have not “conveniently ignored” the fact that median productivity has increased exponentially. That is not an established fact. As far as I know there is no evidence that it has. Productivity statistics like those issued by the BLS are calculated on a per-worker (i.e. mean, not median) basis. Mean compensation has tracked productivity fairly well. It’s the median that stagnated, which is one definition of rising inequality: a separation of the mean from the median.

I don’t buy this. While the BLS statistics are average productivity the way to explain Winecoff’s argument would be that only the top 1-5% of American workers got better at their jobs. The median worker, with better capital, has gotten a lot more productive. The last 40 years also brought many women into the workforce. With a 100% increase in productive workers for most households, one would not expect such a stagnant increase in overall income.