Is it not a bit tautological to say the income inequality causes lower wage growth at the median? Whatever measure of inequality you use, Gini coefficient or otherwise, is a function of median income. Maybe you address this with a lag, but this is not particularly convincing given that these things are autocorrelated.
Yglesias seems to be saying, we should focus on median and lower incomes as important outcomes, rather than inequality itself. One reason is that a country can become more unequal if top incomes grow, all else equal. This is a Pareto-improvement!
Now, given Yglesias’ point, he still misses what you point out here, that median income growth has slowed down. But the reason for this stagnation is more important. Is it some distortion that we can fix? Or, a popular hypothesis among economists, does it reflect drastically increasing returns to high end human capital?
I'm sorry for the late reply! Been de-escalating computer and internet work during the summer.
Really appreciate your comments, and you have a sophistication to your thinking that is, well, way beyond what mine was at your age!
Re. tautology: mostly, but not always. Milanovic and a few other historical economists show that GDP (or rough proxies of median earnings) fluctuate in their relationship to inequality - positive, negative, null associations over long stretches of time.
I read Yglesias as pretty explicitly saying that it would be bad if inequality affected the typical person, which would be read as the median. Maybe I'm being too un-generous to him!
Pareto-improvement: perhaps. Maybe I'm too Robert Frank-pilled. But once you pass a certain threshold, lots of goods become positional goods (what's a big house? A house that's bigger than your neighbors!). Inequality can distort things in weird ways, if you're a Frank-er like I tend to be.
Human capital: maybe! But the way that human capital and earnings are hyper-linked in the US isn't the only necessary pattern (I wrote about a great NBER paper a little while ago focusing on this topic in Nordic countries).
Thanks for the thought provoking comment! Really pushed my thinking.
> “Milanovic and a few other historical economists show that GDP (or rough proxies of median earnings) fluctuate in their relationship to inequality - positive, negative, null associations over long stretches of time.”
Doesn’t this contradict the claim in your post? Maybe I’m misunderstanding some details here. *Edit: I later realized I misunderstood you here. We were originally talking about the growth in the median income, so a cross-sectionally null relationship between median income and inequality does not contradict your claim. Apologies for the mistake.
You also raise an interesting point regarding positional goods. I guess an important point here is: How much do people care about their absolute standard of living, compared to relative income? Maybe people are willing to give up some absolute wealth to be closer to their neighbors in status. If this is true, maybe we ought to shift towards a greater attention to wealth equality as policy outcome. Still, I’m tempted to reject this preference even if people do hold it, as it implies that we should tradeoff against total growth/wealth because people are jealous.
I do not mean to caricature your views; I know that much of why people are concerned with inequality is because they want to prevent poverty and suffering (rather than people having swimming pools as big as their neighbors’). That goal I am all for!
Yes - I think some of the non-material aspects of inequality may be the more important ones. Losing a boat is bad, but losing your sense of community and shared purpose may be worse. And these may have larger long-term effects on well-being. These include: political capture, degrading communal values and willingness to contribute/sacrifice for communal good, distribution of opportunity, sense of community (writ-large).
Yes, we're very much on the same wavelength. I've been thinking about those issues in terms of sports these days, particularly Jay Caspian Kang's new article about intergenerational NBA players.
You look at aggregate sports statistics and you'd probably just notice pro leagues humming along with mild to moderate improvements in performance. But there seem to be massive bottlenecks forming that really distort the whole pipeline. I know that my 2nd grader's now entering into a world where you need to decide whether to stay in local leagues and, well, likely lose opportunities to play in HS leagues or devote oneself entirely, time and financially, to high-octane leagues. Just one of those fundamental distortions that I think follows in the wake of massive rising inequality.
Is it not a bit tautological to say the income inequality causes lower wage growth at the median? Whatever measure of inequality you use, Gini coefficient or otherwise, is a function of median income. Maybe you address this with a lag, but this is not particularly convincing given that these things are autocorrelated.
Yglesias seems to be saying, we should focus on median and lower incomes as important outcomes, rather than inequality itself. One reason is that a country can become more unequal if top incomes grow, all else equal. This is a Pareto-improvement!
Now, given Yglesias’ point, he still misses what you point out here, that median income growth has slowed down. But the reason for this stagnation is more important. Is it some distortion that we can fix? Or, a popular hypothesis among economists, does it reflect drastically increasing returns to high end human capital?
Hi Liam,
I'm sorry for the late reply! Been de-escalating computer and internet work during the summer.
Really appreciate your comments, and you have a sophistication to your thinking that is, well, way beyond what mine was at your age!
Re. tautology: mostly, but not always. Milanovic and a few other historical economists show that GDP (or rough proxies of median earnings) fluctuate in their relationship to inequality - positive, negative, null associations over long stretches of time.
I read Yglesias as pretty explicitly saying that it would be bad if inequality affected the typical person, which would be read as the median. Maybe I'm being too un-generous to him!
Pareto-improvement: perhaps. Maybe I'm too Robert Frank-pilled. But once you pass a certain threshold, lots of goods become positional goods (what's a big house? A house that's bigger than your neighbors!). Inequality can distort things in weird ways, if you're a Frank-er like I tend to be.
Human capital: maybe! But the way that human capital and earnings are hyper-linked in the US isn't the only necessary pattern (I wrote about a great NBER paper a little while ago focusing on this topic in Nordic countries).
Thanks for the thought provoking comment! Really pushed my thinking.
Thanks for the kind comment!
> “Milanovic and a few other historical economists show that GDP (or rough proxies of median earnings) fluctuate in their relationship to inequality - positive, negative, null associations over long stretches of time.”
Doesn’t this contradict the claim in your post? Maybe I’m misunderstanding some details here. *Edit: I later realized I misunderstood you here. We were originally talking about the growth in the median income, so a cross-sectionally null relationship between median income and inequality does not contradict your claim. Apologies for the mistake.
You also raise an interesting point regarding positional goods. I guess an important point here is: How much do people care about their absolute standard of living, compared to relative income? Maybe people are willing to give up some absolute wealth to be closer to their neighbors in status. If this is true, maybe we ought to shift towards a greater attention to wealth equality as policy outcome. Still, I’m tempted to reject this preference even if people do hold it, as it implies that we should tradeoff against total growth/wealth because people are jealous.
I do not mean to caricature your views; I know that much of why people are concerned with inequality is because they want to prevent poverty and suffering (rather than people having swimming pools as big as their neighbors’). That goal I am all for!
Yes - I think some of the non-material aspects of inequality may be the more important ones. Losing a boat is bad, but losing your sense of community and shared purpose may be worse. And these may have larger long-term effects on well-being. These include: political capture, degrading communal values and willingness to contribute/sacrifice for communal good, distribution of opportunity, sense of community (writ-large).
Good to hear from you, Gawain!
Yes, we're very much on the same wavelength. I've been thinking about those issues in terms of sports these days, particularly Jay Caspian Kang's new article about intergenerational NBA players.
https://www.newyorker.com/magazine/2025/06/30/heir-ball-how-the-cost-of-youth-sports-is-changing-the-nba
You look at aggregate sports statistics and you'd probably just notice pro leagues humming along with mild to moderate improvements in performance. But there seem to be massive bottlenecks forming that really distort the whole pipeline. I know that my 2nd grader's now entering into a world where you need to decide whether to stay in local leagues and, well, likely lose opportunities to play in HS leagues or devote oneself entirely, time and financially, to high-octane leagues. Just one of those fundamental distortions that I think follows in the wake of massive rising inequality.