Thursday, March 12, 2015

Some interesting graphs from the White House on Gender and Income

Here is a slide show from the White House with some interesting graphs (HT: UYSIT).

Here are a couple of interesting ones.  Several more are at the link.  Many of these raise as many questions as they answer.

I think these two charts point to the strong role of female empowerment in the shape of incomes over the past 40 years.  I linked to an interesting paper about that (here).

  • Since the early 1970's, most of the gains have gone to high income households.
  • Since the early 1970's, most of the gains have gone to women.

Both of these things are true.  In fact, they are both probably largely describing the same phenomenon.

Unfortunately, the first version is both more rhetorically exciting and  more likely to lead to poor policy choices than the second version.

There are a host of factors feeding these income trends.  The technology boom has fed the tremendous income gains at the very top.  (I've seen somewhere, but I don't have the link handy, that the entire climb in inequality over the past few decades goes away if you exclude a handful of counties around Seattle and San Francisco.) (Edit: if this is the paper I was remembering, then my counties are slightly off -NY instead of Seattle- and I have somewhat overstated the effect.  These counties explain most inequality during the dot Com boom of the late 1990s.) But, also, as referenced above, we can also explain all of the growth in inequality with the changing incomes of women within households.  We can also explain some of it with changing lifecycles and demographics (more low income students in early adulthood and more low income retired households).  We can also explain some of it with housing.  Imputed rent is not included in household income, and this is becoming a larger portion of the median household's true income.

So, it looks like inequality is explained many times over.  I don't think this is an error.  I think all of these things are legitimate explanations.  The counterbalance is that a rising tide lifts all boats.  These factors have all been balanced out by the natural equalizing tendency of capitalist markets and the tendency of economies with rising incomes to expand their social safety nets.

We can lower measured household inequality by supporting undistorted free markets, halting the technology boom, reversing the empowerment of women, replacing support for post-secondary education with support aimed at other forms of human capital development, reducing incentives for overinvestment in housing, or increasing the social safety net.  The low hanging fruit here includes the removal of barriers to entry (technology is doing some of this, reversing things like occupational licensing would help also), transformation of education subsidies, and housing policy.  These aren't necessarily political feasible, but they would be helpful.

The continuation of the technology boom and empowerment of women are clearly far more important than the issue of income inequality.  This is one of many reasons why poverty is a much more coherent issue to focus on than inequality.  Feminine and technological advances are clearly beneficial, even while making measures of inequality worse.  We wouldn't even think of reversing course on these issues in order to reduce inequality.  That is because inequality is not, legitimately, a concern, so it tends to conceptually collapse in the absence of strawmen and boogeymen.

If someone is concerned about inequality when thinking of CEO's or hedge fund managers, but he isn't so concerned about it when thinking about households of married physicians or married lawyers, then he is really just pretending that his bitter attitude toward CEO's and hedge fund managers is a concern about poverty.  This is especially egregious in moralistic complaints about globalization as a cause of income inequality, where corporations and developing economy workers have joined together in an equalizing revolution of global economic improvement.

How about the concern that the internet and pirating are preventing musicians from capturing income from their recorded music.  Is there a group of people with a more extreme level of income inequality than popular musicians?  Why do we bemoan their falling incomes, but cheer for moderation of CEO and financier incomes?

The inequality meme joins a long list of rhetorical emblems that have served to seemingly transform being against something into being for something.  These rarely serve the greater good.  If something is worth being against, then it's worth being against.  The use of the rhetorical device is, itself, a sign of weak moral backing.

7 comments:

  1. The difference between male & female wages, after controlling for things like education and experience, is statistically insignificant in U.S. labor markets. It differs by country, of course, with how women are perceived (e.g. countries with Sharia law). The main driver behind the observed wage gap, when one compares all men with all women, is educational choice. Far fewer women in Engineering, for example.

    To the extent wage gaps are a function of individuals sorting themselves into preferred work types, there's nothing to be said. But, in some countries, the benefits available to women represent a significant marginal cost to a hiring firm. In Canada, my native land, women are afforded a partial-paid leave post-child birth of up to one year, about twice as long as paternity leave. Thus a firm would be less likely to hire a woman in the first place because of this potential cost. Of course, there are anti-discrimination laws to address the imbalance caused by government interference, which just creates more potential costs. It's a wonder anyone works at all.

    ReplyDelete
    Replies
    1. It seems strange how the reaction is so frequently to regulate those who take their income as "profit" on these issues, even when this is unlikely to solve the problem, or is not a significant source of the problem. I frequently feel like Richard Dawkins, when he says that religious believers are atheists like him regarding 10,000 gods, and that it is simply they who are making one exception. In modern, liberal societies we accept that the existence of a problem is not sufficient for coercion. We will not demand that women enroll in high income studies. We will not institute arranged marriages or marriage patterns designed to create gender equality. We will not demand that household responsibilities meet some standard of equality. People have rights, after all, and they will do what they will do.
      But, we make an exception for those who work for profit. They are singled out for coercive public policies, not because these policies are particularly useful, but because these policies are available within the set of policies that we have come to consider acceptable.

      Delete
    2. That's an interesting point about monotheism. Never heard that before.

      You are quite right - most people are completely schizophrenic about their policy choices. I think it's because, at least partially, they haven't thought through their own values. What are your thoughts?

      Delete
    3. You are probably right. Robin Hanson (www.overcomingbias.com) really changed my perspective on this. One of his themes is that we are predisposed to be loyal, and so we naturally live like the observers in the experiment that don't see the gorilla on the basketball court.

      Delete
  2. Perhaps this was the paper you intended?

    http://utip.gov.utexas.edu/papers/utip_57.pdf

    I had heard Galbraith offer a variant of this argument on Russ Roberts' podcast as well.

    ReplyDelete
    Replies
    1. That could be where I heard it. The paper is interesting.

      If this was my source, I had the details slightly wrong and slightly overstated:
      "In particular, the same four counties that contributed most to the increase in between-county income inequality from 1994 to 2000 contributed most to the inequality decline from 2000 to 2003 – New York, NY; Santa Clara, CA; San Mateo, CA; and
      San Francisco, CA."
      After 2003, it is counties around Washington, D.C., and the housing boom centers.

      Browsing the paper, another interesting part:
      "Despite the close association in the annual movements of the between-state and between-county series, state per-capita incomes converged during the 1969 to 2006 period while county and household incomes grew further apart."

      Delete
  3. Arnold Kling made a good point recently, that when anyone wants to make a study of comparing wages over time, it is more accurate to use total compensation, rather than wages, as compensation includes all benefits like employer-paid medical plans. I'm not sure if that's important cross-sectionally, but certainly to do appropriate comparisons over time it is.

    ReplyDelete