Progressives often cite median income data when trying to show that average Americans do not benefit from rising productivity. Real median income (by some measures) has been fairly flat since 1973.
I’ve done numerous posts pointing out that this income data is extremely misleading, indeed almost useless. In material terms, living standards are obviously much higher than in 1973 (a period I can recall.) Scott Winship has also exposed a number of flaws in the stagnating income hypothesis.
But let’s say I’m wrong and the median income data is meaningful. In that case this data would also be highly relevant:
The stereotypical embodiment of America’s high university costs, much loved by journalists, is the part-time barista with a liberal-arts degree and a six-digit debt. Such luckless espresso-pullers undoubtedly exist, but they are far from typical. The average recipient of a bachelor’s degree in America graduated with $16,800 in outstanding debt. Though this is 24% higher than it was in 2003, it seems unlikely to trigger the kind of indentured servitude so often imagined.
Keep in mind that this is for grads with a bachelor’s degree, who are the creme de la creme of our labor market. Many Americans don’t go to college, and many who do get only 2-year degrees. And yet even for this elite group the typical student debt is only $16,800. Surely this is not a problem that progressives need to be concerned about!
Progressives probably view the $16,800 figure is being misleading, as they advocate that taxpayers (including those who didn’t go to college, or just went to community colleges) need to pay off this debt. Perhaps the figure is misleading, just as the income data that progressives use in their articles on the plight of average Americans is misleading. But you can’t have it both ways. Either median income and median debt data are both useful, or neither are.
PS. The Economist article is unclear as to whether they are describing the median debt level or the average debt level. In context, it seems more like the median. (They refer to the “average recipient”, meaning “typical”.) If it were the average debt level then my point would be even stronger, as the median debt is certainly well below the average.
READER COMMENTS
Brandon Berg
Jan 10 2020 at 5:36am
“Keep in mind that this is for grads with a bachelor’s degree, who are the creme de la creme of our labor market.”
Not really. About a third of Americans age 25-29 have at least a bachelor’s degree. The top third arguably doesn’t even qualify as the creme, given that milk is well under 33% cream by volume. If we relax the metaphor a bit and call 4-year graduates the creme, professional and postdoc degree holders would be the creme de la creme.
But yes, the Student Loan Debt Crisis is pure media hysteria, and the narrative crumbles in the face of even the most cursory examination of the data. The figures I’ve been seeing are that about 30% of recent graduates graduate debt free; excluding those, the remaining 70% have a mean debt of a bit under $30,000. The college wage premium is about $25,000 per year [i.e. the median bachelor’s degree (only) holder makes about $25,000 per year more than the median high school (only) graduate].
Default rate is negatively correlated with loan balance, suggesting that the main cause of default is having no income at all, not having too much debt.
Tuition at state schools has grown dramatically in percentage terms, but only because it was such a screaming deal to begin with. Even today, tuition is a small part of the cost of attending a state university; the bulk of the cost is the opportunity cost of four years out of the labor market. Taking opportunity cost into account, the total cost of attending a state university has likely risen only slightly above the rate of inflation, if that.
nobody.really
Jan 10 2020 at 6:57am
So can we really be certain that the median debt would be below the average?
I can’t see the whole article ‘cuz it’s behind a paywall, but does it cite any source for this $16,800 figure? It seems lower than other figures I’d read.
In any event, it seems odd that a publication called The Economist would fail to distinguish between the concept of median and average.
Matthias Goergens
Jan 10 2020 at 8:14am
Not absolutely certain, but for most natural occuring distributions that are bounded by zero from below (we only count debt, not offsetting assets) and open above, it is a safe assumption.
For a counter example: the median human has two hands. The average is probably something like 1.999 due to a few amputees.
robc
Jan 10 2020 at 8:34am
Another example in Scott’s favor is that median credit card debt is zero, while the average is obviously much higher. With 30% of students having zero debt, the median is almost sure to be lower than the average.
BB
Jan 10 2020 at 9:29am
LOL. Love it.
Scott Sumner
Jan 11 2020 at 3:37am
The relevant group is not 25-29, but all taxpayers. And among that group the share with bachelor degrees is far less than 1/3.
nobody.really
Jan 10 2020 at 7:20am
Not wowed by the thesis. Why should people’s opinion about the rate of change in median wage growth bear any relationship to their opinion about the absolute value of median (or average) school debt?
The article states that college debt has increased 24% since 2003. According to FRED, real median household income grew by less than 7% over that period–and presumably the discretionary portion of that income was reduced by the growing burden of school debt. Would this provide some stronger basis for evaluation?
Alan Goldhammer
Jan 10 2020 at 7:36am
There are two things to parse out here: median income and student debt. Looking at the latter, it is something that is easily avoided. One can go to a local community college for the first two years and then on to a state college/university to finish the degree. This is very low cost and should entail very low amount of student debt, if any.
With respect to median income, I think it’s the wrong thing to look at. Income, net of living expenses, is more appropriate. $70K/year income in the the SF Bay area is not equivalent to $70K/year in a large midwest metro area, say Indianapolis. Paying off a student loan does affect living expenses but clearly is region dependent in terms of bottom line impact.
BB
Jan 10 2020 at 9:45am
Scott,
I think you cherry pick on this topic. It’s true that measuring income inequality has serious flaws, but that doesn’t mean that inequality is not an issue. You recently had a post arguing that progress, which I think translated into higher living standards in this post, doesn’t make people happier. And I agree. However, there is significant data pointing to inequality making people less happy. And I think inequality has risen, even if income is not the best way to measure it. I think how do we reduce inequality is a much more interesting question than why income inequality or median debt are bad metrics.
The twitter left’s obsession with student debt is goofy, and I can’t wait until the primaries are over and we don’t need to hear about student debt for four more years, but inequality is real. And it is a much more interesting question, than why do journalists pick the wrong metrics for measuring it.
Mark Z
Jan 10 2020 at 2:54pm
But how we measure it is important because that’s how we know how much of it there really is. I’m also not sure in what sense inequality makes people unhappy. Is it that they dislike their neighbors making 4x their income a lot more than merely 2x? Or that they get upset when pundits tell them how unequally wealth is distributed in society in general?
An example to illustrate my concern; the extent that measured inequality has increased due to things like growing inter-generational inequality and growing inter-regional variation in cost of living, rather than growing inter-class inequality, I think measured inequality deceives the public about how ‘bad’ it is and may make them more upset than they otherwise would be.
I’m also not sure why the principle of hedonic adaptation doesn’t apply to inequality as well. At the margin, people want to be better off, and more equal. But if once they’re better off they return to equilibrium, I’d expect once they’re more equal, they’ll ultimately settle into regarding the guy making 2x as much as them with as much displeasure as they used to regard the guy making 4x as much.
Thomas Hutcheson
Jan 11 2020 at 6:25am
I think people are aware, whatever the statistics can be made to say, that someone graduating from college today has less prospect of income earning trajectory than someone in 1965-70. That feels like increasing inequality, less opportunity.
Michael Rulle
Jan 11 2020 at 11:16am
Hi Tom. Since I do not know your age, I cannot tell if your comment is based on your perception of “1965-1970” because you lived it, or if it’s just an opinion based on your view of history. Anyway, I certainly did not believe back then that my life would be obviously better off than my parents. All I remember was standard anxiety about becoming self sufficient. To me, that seems the same as today. Also, we really do have a tendency to “want more than we have”——regardless of age or time. As a friend of mine once said to me “freedom is living within your means”——-which many of us do not.
Scott Sumner
Jan 11 2020 at 3:40am
BB, You said:
“It’s true that measuring income inequality has serious flaws, but that doesn’t mean that inequality is not an issue.”
I’ve never claimed that inequality is not an issue.
bb
Jan 13 2020 at 9:35am
Scott,
My question is why do you devote more posts to the flaws in income inequality data as opposed to actual inequality? It gives me the impression that you view overstating inequality to be a bigger problem than inequality.
BTW: I think you are wrong that consumption is the ideal metric, because it doesn’t account for utility value of wealth and the security that comes with wealth. Warren Buffet’s low consumption combined with incredible wealth and security is not comparable to to a family struggling to achieve a similar level of consumption through debt.
Steve
Jan 10 2020 at 10:49am
We see this kind of thing all the time. Remember when the individual insurance mandate was a penalty during the bill’s debate…but the second somebody sued them in court it was definitely a tax, and had always been a tax! Words mean nothing anymore, it’s all just [Tyler Cowen voice] mood affiliation.
rovingbroker
Jan 10 2020 at 6:08pm
On page 35 of the January 6, 2020 New Yorker, Sheelah Kolhatkar wrote without qualification and without naming the “one estimate” …
“In the U.S., executive compensation has increased, on average, by nine hundred and forty per cent since 1978, according to one estimate; during the same period, worker pay has risen twelve per cent. Income inequality hasn’t been this extreme since the nineteen-twenties.”
https://www.newyorker.com/magazine/2020/01/06/the-ultra-wealthy-who-argue-that-they-should-be-paying-higher-taxes
Mike Sandifer
Jan 11 2020 at 12:47pm
This is not just a liberal perspective. Greg Mankiw also talks about income inequality, somewhat differently than those on the left. He also thinks its a problem, and he’s not alone among moderates on the right.
Scott Sumner
Jan 11 2020 at 1:47pm
Mankiw’s views are quite close to my views. He also believes that income is a misleading indicator, and prefers looking at consumption.
Mike Sandifer
Jan 12 2020 at 2:23pm
Mankiw does prefer a VAT, and opposes wealth taxes, but he does use the term “income inequality”, not “consumption inequality”, though he does use the same Warren Buffett example you do, saying that Buffett shouldn’t pay much in the way of taxes considering his extremely low consumption rate and high charitable giving. Perhaps it’s just more of a word choice difference, but Mankiw does actually discuss income data.
Jeff
Jan 12 2020 at 8:38pm
“In material terms, living standards are obviously much higher than in 1973 (a period I can recall.)”
I can’t think of a more important term than by which to measure things like this than materially. Indeed, every other measure can be – at least – misleading.
Chris
Jan 13 2020 at 2:22pm
We can see both of those statistics and still believe both issues are relevant, though obviously at much different scales. Additionally, both statistics are very simplistic views of fairly complex discussions, so it would be misleading for either side to make use of only these to make a conclusive argument.
In material terms, we should always be better off than the previous generation, due to the continual advancement of technology. Historically, the only times that hasn’t been the case were during large scale catastrophe, such as resource depletion, societal failure, or invasion. Better measurements would be cost of living, financial security and well-being. Compared to a generation ago, Americans now work more, are less financially secure and are more stressed. All the while, productivity has gone up and executive earnings have skyrocketed. That is what makes people upset. The average American can afford a 60” TV yet is one hospital visit away from financial ruin, because TVs are cheap and the cost of housing and healthcare are astronomically expensive compared to average wages.
while educational debt is not nearly as large scale of a problem as the issues above, it does very much impact a number of people in very real ways. There are a large number of people with substantial college debt and jobs that do not pay enough to pay that debt without putting off other important investments or savings. As the economy has improved this has become less of an issue, but for those of us that graduated at the height of the recession, it has had a long lasting impact on our financial security. Possibly more important is the impact rising college costs are having on those starting to plan their future. For this group, the cost of a good college education is becoming less and less of a financial reward due to future wages. Those that argue that community college is adequate are ignoring the rising costs at these institutions, the failing state and local support for them, and the comparatively limited prospects available with these degrees.
robc
Jan 14 2020 at 11:52am
One good question is why have TVs gotten cheaper but health care more expensive?
Also, if the college debt is difficult to pay off with the job received by having that degree, then a mistake was made somewhere along the line. The problem is that most mistakes can be recovered from, if you buy too much house, you can go into foreclosure, credit card debt to bankruptcy, etc. But because of the government involvement, college debt isnt dischargeable. Before the government guarantee, college debt required a cosigner, which was a much better system. Going back to that is the right answer, but that doesn’t help the people in the current system.
But, you know, https://cdn.shopify.com/s/files/1/0535/6917/products/mistakesdemotivator.jpeg?v=1554328460.
Comments are closed.