fbpx

Focusing on Flat Incomes Obscures Fundamental Economic Changes

Economic growth concept

In General Sherman’s memoirs, he reports that in 1850 the U.S. Army reassigned him from San Francisco to the east coast of the United States. He mentions that the passage from San Francisco back to the east coast of the U.S. cost him $600. I priced a ticket from SFO to NYC for a flight next month. It came in from between $200 and $550 on AA.com. So today, we can get from one end of the country to the other end for about the same nominal cost, and for stratospherically less time — several hours instead of several months.

The qualitative improvements considered by themselves are astounding. But we shouldn’t compare nominal cost of transportation. Annual income in the U.S. in 1850 (in 2005 dollars) was around $2,500. So it took about a quarter of a year’s oncome (around 88 days at average wages) to pay for the travel from San Francisco to the east coast of the U.S. Today average income in the U.S. is around $45,000 (in 2005 dollars). It takes the average worker about three days of wages to pay for a ticket to cross the U.S. in a matter of hours.

To point out the obvious: the change for society and for economics is simply revolutionary.

The data got me thinking about comparing household wealth across time. To be sure, my example from Sherman’s memoirs is an extreme example, over a long period of time. But I do wonder about the accuracy of estimations showing that U.S. household income has been pretty constant since the 1970s. While median household income did bump up 5.2% between 2014 and 2015, it was still lower than it was in 1999, and not much higher than in the late 1960s.

I don’t desire to white wash bad, or at least mediocre, news. But there are two aspects of the data that make me wonder about the report. The first is that qualitative improvements are not captured by the data. Even assuming that transportation from the west coast of the United States to the east coast had the same real cost (in income) today as it did a century ago, the data would not pick up the astounding improvement in the quality of that transportation, today taking a few hours for a trip that took months.

Again, I have no desire to be pollyannishy. But individuals old enough can recall changes in the quality of consumer goods over the last generation: The quality of automobiles, shoot, simply consider the quality of automobile tires today relative to the 1970s, the quality of household appliances, the quality of clothes, the quality of houses (and the size!), and more. Plus, simply consider the qualitative improvement provided by goods that did not even exist in the 1970s. Personal computers, and easily portable cell phones which, annoying as they are, are used to do a lot more than simply phone other people.

The second change is the nature of “household.” Data show that while the median income of men in the U.S. has remained essentially static since the late 1960s, the median income for women almost doubled (although still less than men). Even with stagnating wages for men, if household composition in the U.S. stayed the same as in the late 1960s, median household income would have had to have increased simply from the increase in the median wages of women. That median household income has not changed, yet the median income for women increased as it did, must result in significant measure from the failure of households to form at the same rate in later years as they did in the 1960s, and the failure of formed households to sustain themselves.

Between these two changes, I’m not at all satisfied that the baseline isn’t moving concurrently with household incomes, and that we need to be careful about too-easily concluding an absence of economic advance over the last decade.

Reader Discussion

Law & Liberty welcomes civil and lively discussion of its articles. Abusive comments will not be tolerated. We reserve the right to delete comments - or ban users - without notification or explanation.

on April 24, 2017 at 11:30:11 am

Comparis cost in 1850 dollars with income in 2005 dollars is indeed "revolutionary economics".
To be expected: classical calculation gives more solid fondation.

read full comment
Image of Peter
Peter
on April 25, 2017 at 11:50:12 am

1. To evaluate the merits of a comparison, it helps to understand the purpose for which it is made. I sense that Rogers intends to challenge arguments that rest upon a belief that the living standards for median households have stagnated. But in this blog post we have neither the original argument, nor an entire rebuttal. So this post is really an exercise in shadow-boxing.

2. That said, McGinnis has regularly made this shadow-argument. And, for what it’s worth, I share the view that living standards for the median household have improved since, say, 1980, even if the real wage rates earned by members of that household have stagnated.

The examples are clear: Yes, today’s cars are qualitatively superior to cars of 1980. Yes, today’s houses tend to be superior than yesterday’s. Etc.

But also this: Wages don’t measure total compensation. Health insurance represents a larger share of many people’s compensation, even if it is not reflected in wages. Yes, the fact that heath care has grown more expensive might be understood as a loss to the median household. Or it might be understood as a reflection of the fact that health care is so much better than in 1980. I expect that the increased benefit is equal to or greater than the increased cost, but I imagine people could disagree.

True, college has grown more expensive. But I suspect that real returns on a college education have also grown (even if they haven't grown as fast as the price of education.) Last I'd heard, the return on education is still among the best investments around, the occasional college debt horror story notwithstanding.

3. Yes, it makes sense to compare today’s households to the households of 1980. For example, today’s households tend to be smaller than households in the 1980s, so per capital income might be increasing even if household income remains stagnant.

But just as the numbers don’t capture everything that has been gained over the decades, it doesn’t capture everything that has been lost. Yes, more women work, which has increased income into many households. But the flip-side of that argument is that 1980s households enjoyed the benefits of more domestic labor (“women’s work”) than today’s households do, but because that labor was generally unpaid it may not be reflected in the data. Likewise, I often read that workers are no longer taking advantage of all the vacation days they earn. So if median households in the 1980s took longer vacations than do median households today, that would also be a qualitative difference that might not be reflected in the data.

4. Here we encounter a fundamental question: What is the better measure of welfare—wealth or status? Even if we could demonstrate that today’s median households are materially better off than median households of a bygone day, they may still feel a sense of loss relative to the living standards of the wealthiest segments of society. The rich have grown so much VASTLY richer than the rich of the 1980s that median people feel poorer in comparison.

We can remonstrate that people should NOT feel envy, or at least should not stoop to acting on such feelings. But rightly or not, we social animals do and do—which lands pundits such as McGinnis and Rogers in deep do-do.

read full comment
Image of nobody.really
nobody.really
on April 25, 2017 at 11:57:51 am

"which lands pundits such as McGinnis and Rogers in deep do-do."

Perhaps, it lands them in deep "data-do" or as Mark Twain said: There are lies, damn lies and "income" statistics.

read full comment
Image of gabe
gabe
on April 25, 2017 at 13:44:00 pm

...which lands pundits such as McGinnis and Rogers in deep do-do.

Perhaps, it lands them in deep “data-do”....

Or in deep do-do.

And data-ta.

That's all I want to say to you.

[A]s Mark Twain said: There are lies, damn lies and “income” statistics.

He also said, "People commonly use statistics like a drunk uses a lamp post: for support rather than illumination."

But it was Aaron Levenstein who observed that statistics are like bikinis: what they reveal is suggestive but what they conceal is vital.

read full comment
Image of nobody.really
nobody.really
on April 25, 2017 at 15:22:27 pm

Luvv'd the quotes!

Now THAT last quote can get YOU in "deep *political* do-do"

read full comment
Image of gabe
gabe

Law & Liberty welcomes civil and lively discussion of its articles. Abusive comments will not be tolerated. We reserve the right to delete comments - or ban users - without notification or explanation.