I was surprised by the results of this IGM Forum poll. Agree or disagree with
The 9% cumulative increase in real US median household income since 1980 substantially understates how much better off people in the median American household are now economically, compared with 35 years ago.
The great majority agreed with the statement. My remarks.
1. When I first saw the question, I expected a sharp split. I figure there is a lot of “mood affiliation” with the view that the middle class is suffering.
2. It could be that economists are so well off that they have no sense of what life is like for the median household. They feel intuitively that median well-being is somewhere in between what the statistics show and their own well-being.
3. If you read through the response, the most common reasons for agreeing with the statement are improvements in health and in (the consumers’ surplus from) smart phones and such.
That raises the question of what these folks think would be a better measure of economic well-being. Marked-to-market overall consumption – or maybe expected lifetime consumption – including the value of different forms of social insurance?
A lot of economists harp on the the “Effective Marginal Tax Rate” problem, which is that for every extra dollar a poor person earns, they lose government benefits that are worth – either objectively in terms of market opportunity costs, or subjectively to them – a large fraction of that dollar.
And that’s certainly true. But another question is how much of a change in ‘lifestyle’ or ‘total consumption patterns’ does an extra, say, thousand dollars of income make to a poor who may already be receiving tens of thousands of dollars in government subsidies? Imagine someone living in a very poor underclass neighborhood making $10K and suddenly receiving a windfall net gain of a thousand bucks to spend over the year. Maybe on paper it looks like he’s got a 10% boost in income.
But really, the government is helping him to the tune of $20K a year with health care and subsidized food and housing, so the boost is really only 3%. He can’t really upgrade his neighborhood or health care or education much with that, doesn’t get any more ‘public goods’ which may form a large component of his welfare, and spread around transportation and food and clothes, it just doesn’t go that far.
So, we are used to hearing about already questionable concepts like the ‘marginal utility of a dollar’ and, assuming egalitarian commensurability, the ‘poor’ are supposed to get much more bang per buck.
But I think, just like ‘net income’ with high EMRTs, the curve is probably very flat, extending perhaps into the range of median earners, and only then does it start to decline.
It may even be the case that marginal utility measured in terms of lifestyle changes that can occur when one starts to pull away from the bottom even picks up slightly after you get out of ‘safety net’ range, if there is some kind of ‘phase-change’ hard-transition between, say, living in a safety-net-dominant neighborhood and one in which most people are about as well-off materially, but aren’t receiving benefits.
Looking around the neighborhoods close to where I live, there seems to be a big jump in price of the real estate between those two types of neighborhoods, even though objective quantitative criteria, like, say, square feet per person, age of structures, and health status or medical services consumption, don’t seem much different.
“It could be that economists are so well off that they have no sense of what life is like for the median household.”
That is suspiciously similar to what we know of the species known as politicians.
Economists need to heed the calling of fly-over country. The many mobile parks where your presence is desperately lacking. They too, need to engage in a scholarly and robust discussion on the velocity of money.
In light of these new revelations. I cast all my votes for Arnold Kling and Tyler Cowen. For what office, you might ask….?
Does it really matter?
If we are so much better off, why can’t we / dont we create higher incomes?
Higher incomes are generated by making other people’s lives better, not by making one’s own life better.
(Gratifying to see consumer surplus considered here.)
Which seems like what I’m saying. If we were making everyone’s lives better, wouldn’t higher incomes tend to go along with that. For a really long time as life wad truly improving, incomes were also rising.
We have no absolute yardstick, so we don’t know. The measures we do have are meaningless.
As someone who was in college back then, I think anyone who got into a hot tub time machine and went back to 1980 would feel like they were in a third world country.
A few discussed how well off the top is now.
But I wonder about that. I think the plight of the 50% is much, much better than the “plight” of someone at the 0.01% today compared with 1980.
Is median household income really the right measure? Marriage rates have fallen significantly since 1980, so when compared to something like gdp/person, it looks worse than it maybe should. I guess a policy of encouraging people to marry younger could boost median household incomes (by reducing the denominator), but that seems absurd as an economic policy.
What about de facto policies that discourage or delay marriage?
There’s some range there; Policies that wind up delaying marriage (and children) until people are established are probably good for growth; policies that discourage marriage outright, probably not so much. On the whole, I’d guess the effects are positive for growth, but negative for this particular metric. (I should probably find some actual data.) Mostly I meant that this particular metric (median household income) isn’t a perfect measure so that policies trying to affect that specifically are likely to have unintended consequences.
Since the descriptions of “households” change over time (1-earner, 2-married adults, 3.6 children becomes 2-earner, 2 living together adults, 1.9 children, or 5-earner, 5-married [allowing for polygamy here, because I have no doubt it’s coming] adults, 0 children, some such), how’s about calculating GDP per capita as GDP divided by the number of earning-age and older people? Get rid of household agglomeration altogether.
Well, it’s measuring a different thing. “median household income” is a proxy for “how is a typical citizen doing”, in a way that aggregated GDP/citizen isn’t. The problem with cross-time comparison is that the “typical” citizen now isn’t married, and that makes simple numeric comparisons not have the obvious meaning. I don’t have a good suggestion for a better statistic, though.
The most common American household today is one person living alone, without partner nor children. 35 years ago the household was younger and larger. It seems obvious that people, individually, is today much worse off than then. But, I for one, vacillate in casting my vote for Arnold against such distinguished consensus.
The poll results are more heartening than surprising. They show that economists have enough common sense to know when a statistic is at odds with even the most casual of observation, i.e., when it doesn’t pass the sniff test. People living in the median household today would not exchange the basket of goods resulting from the last 35 years of economic growth for a 9.1% raise.
The statistic shows how difficult it is to adjust for changes in household composition and to measure inflation, but mostly the former, since real GDP per capita growth statistics seem to have been pretty steady over long periods of time. By the way real GDP per capita increased 1.79x between 1980 and 2014 (1.73%/yr), more in line with historical trend than median household income and definitely more in line with observation of middle class American life.
Not surprising but quantification would be much better. How much more would they need for indifference between then and now? Hard for them them to answer though due to 2.