It Matters How Rich the Rich Are

John Aziz at The Week argues that it does not matter how rich the rich are. I believe he is wrong for a number of reasons.

The Impoverished Do Suffer Because the Rich Are Rich

Aziz opts for the Econ 101 move to take on the idea that we should want to know how rich the rich are (not unlike the old "grow the pie" saw):

Focusing on how immeasurably rich the rich may be is putting the cart before the horse. The impoverished do not suffer because the rich are rich. Economics is not a zero sum game. There is not a predetermined amount of wealth in society to be divvied up and redistributed. More is constantly being created. The poor suffer because they lack wealth and access to economic opportunities. In other words, we need to focus on what the bottom doesn't have, not what the top does.

This is a clever bit of rhetoric, but nothing more than that. It is true, in some super-abstract sense, that the poor suffer because they are poor, not because the rich are rich. But it is also true that the poor can be made not-poor by reducing the wealth/income of the rich in order to increase the wealth/income of the poor. In that sense, then, the richness of the rich is a cause of the poorness of the poor.

Aziz's point is that we don't need to change the distribution in order to improve the plight of the poor because new wealth can be created that can do so. But creating new wealth doesn't improve the plight of the poor. Only the poor receiving more wealth/income improves the plight of the poor. And if the last half century has taught us anything, it is that wealth and income growth does not, by itself, translate into the poor receiving any of it. The uplift of the poor requires a specific distributive result that increasing production does not necessarily lead to. This is precisely why looking at the distribution (including how rich the rich are) is critical.

Even if you assumed that increasing production would automatically uplift the poor (which is wrong), it is pure ideology to say that means "the impoverished do not suffer because the rich are rich."

In that scenario, we would have to say that the suffering of the impoverished could be lessened by 1) increasing production, or 2) changing the distribution. Aziz fixates on the possibility of (1) in order to reject that lack of (2) is responsible for the suffering of the impoverished. But you could just as well make the exact opposite analysis! In this scenario, there are two independently sufficient mechanisms for uplifting the poor. Rejecting one altogether because the other one exists is not good analysis, though it is an excellent indicator of ideological bias.

Assessing Government Policy

In addition to its general weakness, Aziz's argument fails to take note of the most obvious reason why you'd want to know how rich the rich are: to assess government policy. Everyone seems to think its important (and I agree) to have good data on government programs. This is why we have robust administrative data on many such programs and a number of extensive Census surveys like SIPP and ASEC. We seem to think it's important to know how these programs are working and who is benefiting from them and to what degree.

But the economy is itself just a large government program. Through its laws and the police, the government has put in place a vast array of economic institutions that collectively determine who gets what at any moment in time. To assess the effectiveness of the institutional design choices the government has made, we need to know who is benefiting from them and to what degree. That means, among other things, knowing how rich the rich are.

This is not a purely philosophical point. Because we have decent data on who the richest are, we have observed, for instance, that a good chunk of them capture their wealth through intellectual property rents ("superstars" and the like). That's a useful thing to know when we go to assess whether the government program called "intellectual property" is working the way we want it to. We've also found that a good chunk of the rich are corporate executives, which has led many to suspect that the government program called "corporate law" (more specifically the way corporate governance is organized within corporate law) is enabling executives to capture rents. Data detailing the wealth of financiers has raised similar questions regarding the efficacy of finance and securities regimes.

In short, good data on the rich is necessary for us to evaluate the political-economic institutions of our society and whether they are doing what we want them to do.

Poverty Is Down 38 Percent

Finally, Aziz makes a technical error at the end of his piece. Citing to Census data from 1966 to 2011, he states that "America has not succeeded in reducing poverty over the last 50 years." This is true under the Official Poverty Metric, but this metric is widely criticized for failing to include tax credits and non-cash benefits, as well as utilizing a poverty line that doesn't make much sense. Under the more respected Supplemental Poverty Metric, the poverty rate has fallen by 38 percent (from 26% to 16%) since 1967, with all of the decline coming from non-market income programs.

This is an important corrective because Aziz uses his estimation that there has been no progress on poverty reduction in the U.S. to conclude that "the U.S. needs to seriously re-examine its approach to poverty." But it doesn't really need to do that. The prevailing approach of taxing and transferring has been very successful.

We need only to ramp it up further to get more poverty reduction. But to carry out any such expansion of transfers, we need good information regarding what resources can be taxed to finance them, which again brings us back to the importance of knowing how rich the rich are.