I’m trying to get labor force growth rate and productivity, by year, for the 1900-1920 time period. And thank you for the debate, sir!
I see population growth at a bit over 2%, but my hypothesis is that there were disproportionate numbers of women, immigrants entering the labor force. I had 4% or thereabouts in my head.
If you’ve got any data to help here, I’d appreciate it.
Dude, you are right.
I’m staring at the tables in the back of Maddison’s book. Which is awesome, by the way. This is a great reference. Please don’t spoil it by telling me he’s under attack from some left-wing or right-wing group for extreme bias due to some political reason or another. That would ruin my day.
I always had in my head 5% real per capita GDP growth for boom years in rapidly developing economies…the ‘double in 15 years’ rate. And it’s there for post-war South Korea, Hong Kong, etc. All of the usual suspects. I had it in my head for turn-of-the-century America as well.
But I’m staring right at the 1870 to 1900 to 1913 p.c. GDP for the US and it clocks in right at 2%. Which is hard for me to believe. I’ve researched electric power production, steel production and oil output for those years and that may be where I got the 5% rates from. But I’m sure Maddison has factored that into his analysis. If I got the numbers from somewhere, I’m sure he did as well.
I never would have guessed that. Even if you give me 4% labor force growth, which I haven’t found yet, and throw in 1-2% for undercounting quality and an inability to measure services, you’re only at 7-8%. And I haven’t even found sources to prove that yet.
Point conceded. 15-love, M4M. Well played. Well played, indeed.
Thanks again for Maddison’s name. I’ve been browsing through that for the past few hours. Great stuff.
There is something that really bugs me about that 2% figure from the ‘boom years’ of the American economy. It just doesn’t make intuitive sense to me. I’m not saying Maddison’s figures are bogus. But something just doesn’t compute upstairs.
I have a hypothesis. I’m going to go do a little research. But in the meantime, MeasureForMeasure, if you could spare a few minutes and read below, and try to poke holes in my argument or suggest some analysis, I would appreciate it. You’ve shown an affinity for the subject so I’d like your feedback.
5% real growth is a doubling of output in 15 years. 7% is a doubling in 10 years. 3% is a doubling in about 22 years.
Those are timelines that fit within a human generation and are visible to the layman’s eye. And when I hear ‘doubling of output’ I immediately think ‘doubling of living standards’.
Which I think is the core of the issue. The two are usually pretty close, but as we’ve mentioned before there are quality differences in output that aren’t captured in GDP metrics very well. Usually they are small in mature, diversified economies. More on that in a moment.
You can see 5% growth, 7% growth or more with your naked eyes. I’ve lived and worked in countries with that kind of growth, and maybe so have you.
Even if you haven’t, you can get a feel for it just by watching a newsreel or reading a newspaper from 15-20 years ago and then looking at where things are today. Look at films of China or India from the mid-80s and then look today. That’s 5-7% annual growth happening right in front of your eyes. Doubling of living standards. Or more.
There are construction cranes everywhere. Train stations and airports are jammed. People are moving off substinence farming and into cities to take up wage-paying labor. Kids are going to school longer, or maybe even to college. People stop riding bicycles and start riding scooters or driving cars. Families move out of tenements with shared outhouses and get their own apartments with air conditioning.
And even if you’re not an economist or government statistician, you could hazard a guess at 5+% or 7+% growth in societies like that and not be too far off. South Korea was at 7% a year for almost 30 years, according to Maddison’s data. Estonia, where I recently travelled, is at 10%. Hong Kong is a classic case, of course.
I would have bet $50 that the US from 1900-1920 experienced 5+% real pc GDP growth. We’ve seen from the figures that it didn’t.
I would have bet $50 not because I had statistics at the ready, but because I’ve done research on the explosive growth in electric power, steel production, automobile production and a bunch of other things that occurred during that time. Plus, I love reading about American history and have a better-than-average-Joe idea of what sorts of things transpired over that timeline.
I was shocked when I saw that from 1870 to 1913 Maddison’s figures showed 2%real per capita growth. 2% growth takes 35 years for a doubling of living standards. 35 years! And Maddison’s span is only 43 years. About the same timeframe.
That just doesn’t make any sense to me. But I think it’s because I’m conflating ‘living standards’ with ‘GDP growth’ and I need to research what major differences might exist between the two.
1870 was just after the Civil War. A lot of the country was in ruins. The telephone had not been invented. We were still only 10 years removed from the Pony Express, for God’s sake. Many sea-going vessels were still under sail.
1870 to 1913 is 43 years, only slightly more than the ‘doubling of living standards’ timeframe for 2% growth. In that time,
There’s a delightful little paper at
http://www.iea.org/textbase/work/2004/eewp/Ayres-paper3.pdf
that has that information
And a lot of other things. Things that I’m going to go research. It was a great age of American invention and commerce.
Does that sound like only a doubling of living standards during that timeframe? Could that possibly be right? That doesn’t sound like it could possibly be right. That sounds like a tripling, or quadrupling, or some other large factor increase in living standards. 5+% growth or more in ‘living standards’ maybe, if not precisely measured GDP growth.
I’m going to see if any academic or respected economist has done any research on this. I think the GDP numbers from that age may suffer from the ‘Moore’s Law effect’. That is, recent computers benefitting from Moore’s Law are probably 1,000x, maybe even 10,000x more powerful than the first IBM PCs and Apple IIs that rolled out 30 years ago. But I don’t think worker productivity in the computer industry shows 1,000x or 10,000x factor gains during that timeframe, because it’s a leading-edge product and the stats don’t adjust very well for dramatic changes in quality on the front edge of introduction.
I think because from 1870 to 1913, America was on the leading edge of electric power, automobile production, air travel, and a lot of other things, the prices dived so quickly at the front edge relative to quality improvements that they don’t show up in GDP. Because I think GDP uses prices in it’s calculation. Need to check that, too.
So I need to spend some time checking that out. But we can still mix it up a bit, if you want. This has been enlightening to say the least. Thanks again.
Here’s the problem, IdahoMauleMan. Generally speaking, real GDP growth is quoted - which adjusts for inflation. But real per capita GDP growth adjusts for inflation and population growth. I took that figure because… well it was the first one I found and it’s good for getting a handle on changes in living standards.
You have to make some mental adjustments though: we know that the US had a terrific run from 1950-1973, so we should consider 2.45% real per cap to be a gold standard.
Flipping through the Maddison’s work again I see from Box 3-1 (p. 138) that there have been large revisions to the 1929-50 US growth figures. Huh.
But anyway, from 1950-73 real GDP growth (not adjusted for population) was 3.96% (as opposed to 2.45% per cap): from 1973-98 it was 2.90% (vs. 1.99% per cap). That’s the sort of scaling that we’re more familiar with.
How do these figures map to the common man’s experience? I would guess that the number of skyscrapers might track real GDP growth, but that living standards would match up better if you divided by population.
Incidentally, the 1870-1913 period was awesome. Doubling per person output every 39 years, rather than every 53 years as was the case earlier (1820-1870) is something to celebrate. It’s just that the 1950-1973 era was off the charts, and 1973-1998 wasn’t too shabby either.
By way of contrast though, here are the per capita GDP growth rates from 1973-98 for some of the Asian tigers (p 216):
China 5.39
Hong Kong 4.27
Malaysia 4.16
Taiwan 5.31
Thailand 4.91
Extraordinary.
Agreed.
And my not-confirmed-by-any-data throw-out of 10+% real GDP earlier was not adjusted for population. It was a combination of p.c. * population. Sorry if I didn’t make that clearer, earlier.
My confusion is coming from this conflation of ‘living standards’ thing with the GDP thing. My gut tells me that the increase in per-capita living standards in the US from 1870-1913 was just as fast, if not faster, than the ramps in the recent Asian tiger success stories.
I have no way of proving this, other than eyeballing quality-of-life metrics like life expectancy, availability of electric power, order-of-magnitude increases in communications effectiveness like those from telegraph to telephone to radio, etc.
I’d love to see some sort of correlation between an index of ‘living standards’ (health, communications, power, travel, leisure time, indoors air temperature control, etc.) and GDP. That’s what I’m digging into. If you’ve got something there, please shoot it my way.
But here’s my question to you. Measuring real capita GDP of recent Asian tigers feels like measuring prices of ‘trailing’ or ‘mature’ product categories. Things like cars, steel, electric power etc. have prices that have settled down a lot since their introduction around the turn-of-the-century. Therefore, when the Asian tigers started churning out those quantities due to rapid growth, we could measure their GDP using prices from goods that have been around a while and have reasonably well-established markets.
But when those things started taking off in the US in the 1870-1913 timeframe, they were brand new. They were sort of like personal computers taking off in the very late 1970’s and early 1980’s. And I would contend that productivity measurements (which are related to GDP measurements) in the computer industry are probably way too low due to the quality effect. Feel free to challenge me on that one too, if you’ve got any more information on the subject.
So even though the US was roaring through a number of front-end innovation curves during the 1870-1913 timeframe, real per capita GDP growth (in the 2%-ish range) doesn’t match that of the recent Asian tigers (in the 5%-ish range). That just doesn’t make sense to me. And is why I’m trying to cross-compare it with this whole ‘living standards’ thing for a sanity check.
Does that make sense? Sorry for the long-winded explanation.
You pose an interesting question. Keep an open mind though, and you might get an even better answer. (eg US advancement in living standards rocketed upwards in this way, but not in that. Or: compare the gains from city public health measures in the 1800s to the advances of the typical peasant in (pick your Asian tiger)). I might be wrong, but I understand that simple water treatment was a huge material advance.
There was a compilation of older Statistics put out in 1976: Historical Statistics of The United States, Colonial Times to 1970. Hey, I see that it’s public domain:
http://www.census.gov/prod/www/abs/statab.html
Check out page 56 of the 1st volume, “Expectation of Life at Specified Ages, by Sex, for Massachusetts”: 1850 to 1949-51. There’s also a long MA infant mortality series.
If you have access to a university library of course, you might see if you can pull a treatment of US economic history, demography or public health from that era.
I free associate: http://delong.typepad.com/sdj/2008/02/night-thoughts.html
And here’s another one: Six Families Budget Their Money, 1884
This looks like good stuff.
Where do you find this? On this newfangled Internet Web-thing, that the kids fool around with as they listen to rock and roll music?