The long term trend in goods and government vs. everything else I think also shows part of what has happened to the American middle class.
Goods and Government are what we might have thought about as backbone jobs. These are police officers, fire fighters, school teachers, factory workers, construction workers. When you think of a stereotypical 1950s American, they are doing one of these jobs.
And, in the 1950s half of Americans were employed in these sectors. Yet, since then the labor market has radically shifted.
Notice these two are plotted on the same axis so from 1939 to 1965 – not including the war boom – goods and government was roughly half of the nonfarm labor force. Then a rapid falling off.
One thing I hadn’t considered but is probably true is that goods (via the magic of boxes) and government via transfers between jurisdictions, is not as dependent on urbanization as the rest of the economy.
This probably supported the stagnation in land rents that occurred over this period, which also supported real wages.
19 comments
Comments feed for this article
Sunday ~ March 11th, 2012 at 12:05 pm
Bruce Bartlett
If the government data include the military, that explains a lot of the divergence.
Monday ~ March 12th, 2012 at 12:49 pm
Stewart Rubenstein
USGOVT does not include the military. See http://www.bls.gov/opub/hom/pdf/homch2.pdf
Sunday ~ March 11th, 2012 at 12:16 pm
Alfred Askeljung (@alfredaskeljung)
Probably a retarded question but hasn’t fed.govt. spending as percentage of GDP increased to 24 % during the latest administration? If so, how is that correlated to this?
Monday ~ March 12th, 2012 at 8:18 am
addicted44
The percentage increase was largely because of the drop in the denominator (i.e. reduced GDP, hence, you know, the recession). A small increase was registered due to increasing unemployment payouts, once again, caused by the recession.
Sunday ~ March 11th, 2012 at 2:14 pm
Curt Doolittle
As Bruce suggests, does the data include the military through Vietnam?
More importantly, the chart doesn’t go back far enough to convey any meaning. You’d have to get back to 1910 before we could pull any meaning from it.
Why? Because we had a temporary period from 1914 to about 1980 where we were able to produce cheap, poorly made goods using the innovations of war technology, while the rest of the world recovered or decayed in the communist revolt against modernity. And during that time, we were able to suck up agrarian labor that had been displaced by harnessing fossile fuels.
We can’t do that any longer. Our ‘working’ population is too unskilled to maintain a relative position, and they have been ‘promised’ clerical jobs that are effectively unskilled labor, instead of production jobs that require technical competence.
And because of this, we’ve lost at least one half of generation during the recession, and probably permanently harmed the preceding one and a half generations.
Spending might put money into circulation but eventually this structural problem has got to be solved by fixing our antiquated education system.
Gingrich’s plan is excellent. But the progressives would never allow it.
Monday ~ March 12th, 2012 at 12:39 am
Daniel
Interesting, but I’d diverge on one point. It isn’t that goods and government are not as dependent on urbanization, is is that the correlation between urbanization and goods and government would be somewhat inverse. Cities follow Zipf’s law (among other scaling rules) and thus as they get bigger, become more and more efficient. This means the more people you pump in, the demand for government workers and goods will decline on a per capita basis. That would track on your graph – the number or people in cities keep going up, so the G&G line keep going roughly up, but it is flatter and flatter over time since the demand is less. Since the demand is less bu the population keeps increasing, you get the split observed.
Some references on City scaling and efficency
http://bit.ly/pdjYQt
http://nyti.ms/6JFSQs
Monday ~ March 12th, 2012 at 3:07 am
Economist's View: Links for 2012-03-12
[...] Where Middle Class America Has Gone – Modeled Behavior [...]
Monday ~ March 12th, 2012 at 3:22 am
Links for 2012-03-12 | FavStocks
[...] Where Middle Class America Has Gone – Modeled Behavior [...]
Monday ~ March 12th, 2012 at 3:22 am
The Great Decoupling of Income Growth for Middle-Class Households | FavStocks
[...] here [...]
Monday ~ March 12th, 2012 at 3:22 am
The Great Decoupling of Income Growth for Middle-Class Households | FavStocks
[...] here [...]
Monday ~ March 12th, 2012 at 4:14 am
Economist's View: The Great Decoupling of Income Growth for Middle-Class Households
[...] here [...]
Monday ~ March 12th, 2012 at 10:05 am
On Types of Jobs « Blog of Rivals
[...] Smith writes here, here, and here about the types of jobs that have grown since the late sixties. Specifically he compares [...]
Monday ~ March 12th, 2012 at 12:33 pm
Stewart Rubenstein
I pulled this data apart, splitting it into three parts – goods, govt, and other private – and looking at them as a fraction of total non-farm employment.
http://research.stlouisfed.org/fredgraph.png?g=5DN
This shows that on a percentage basis, goods employment has been on a steady decline since 1953. from about 40% to 13%. In fact, on an absolute basis, it was lower in 2010 than it had been since about 1950.
This was partly offset by a rise in govt employees from 12% to almost 20% in 1975, but that has since dropped to about 16%.
So, the story is that goods employment has been dropping, but until 1975, that was offset by growth in government. Since then, government employment has been shrinking, too (on a percentage basis).
By the way, active duty military peaked at about 3.55 million, 5.0% of total nonfarm employment in 1968, and is now about 1.45 million, 1.1%. One might think that explained the drop in the USGOVT series, but can someone explain why the peak in the FRED data series in 1975 didn’t match the peak in active duty military in 1968, or explain why the shedding of 1.5 million military jobs between 1968 and 1973 doesn’t appear as a blip in the FRED series?
Monday ~ March 12th, 2012 at 12:39 pm
Stewart Rubenstein
A bit more research answers the last question – USGOVT excludes military personnel, as well as civilians employed by the CIA, NSA, NIMA and DIA.
Tuesday ~ March 13th, 2012 at 8:34 pm
Brad DeLong: Sectoral Shifts and Income Inequality
[...] Where Middle Class America Has Gone « Modeled Behavior: The long term trend in goods and government vs. everything else I think also shows part of what has happened to the American middle class. [...]
Wednesday ~ March 14th, 2012 at 3:22 am
Sectoral Shifts and Income Inequality | FavStocks
[...] Where Middle Class America Has Gone « Modeled Behavior: The long term trend in goods and government vs. everything else I think also shows part of what has happened to the American middle class. Goods and Government are what we might have thought about as backbone jobs. These are police officers, fire fighters, school teachers, factory workers, construction workers. When you think of a stereotypical 1950s American, they are doing one of these jobs. And, in the 1950s half of Americans were employed in these sectors. Yet, since then the labor market has radically shifted. [...]
Wednesday ~ March 14th, 2012 at 9:24 am
Where the Middle Class has gone | Screw Cable
[...] modeledbehavior.com – Tagged: New Normal View on Counterparties.com → Amazon.com Widgets var [...]
Thursday ~ March 15th, 2012 at 10:34 pm
Curt Doolittle
Stewart,
Can you post that somewhere?
Wednesday ~ September 12th, 2012 at 6:22 pm
Counterparties: The tasks of the proletariat in the present recession | Felix Salmon
[...] that gloomy sense, America’s burgeoning lower class has something in common with its shrinking middle class. An earlier Pew survey found “85% of self-described middle-class adults say it is more [...]