Why did growth slow after 1973?

My grandmother died at age 79 on the very week they landed on the moon.  I believe that when she was young she lived in a small town or farm in Wisconsin.  There was probably no indoor plumbing, car, home appliances, TV, radio, electric lights, telephone, etc.  Her life saw more change than any other generation in world history, before or since.  I’m already almost 55, and by comparison have seen only trivial changes during my life.   That’s not to say I haven’t seen significant changes, but relative to my grandma, my life has been fairly static.  Even when I was a small boy we had a car, indoor plumbing, appliances, telephone, TV, modern medicine, and occasional trips in airplanes.

In response to my previous post, a commenter asked why the neoliberal revolution hadn’t produced a more impressive increase in living standards, after all, he argued, there have been huge increases in computer technology, cell phones, etc.  You’d think these are impressive, but they are actually in a rather narrow segment of everyday life.  Here is an example I use with my class:

When Lindbergh crossed the Atlantic in 1927, he had a fairly small and primitive airplane.  By 1969 Boeing had produced the 747.  Now another 41 years have gone by and commercial aircraft really haven’t advanced far beyond the 747.  Sure the internal avionics are improved, but the overall improvement is nothing like the 1927-68 period.  You could do the same for many other products.

Here’s what I think happened.  There were a few underlying technological developments in the late 19th century that dramatically affected living standards in the 1920-70 period, when they were widely adopted in advanced economies.  I would certainly include electric power and the internal combustion engine.  I also think indoor plumbing is underrated.  Imagine having to rely on outhouses in cold climates.  And also recall the health advantage of safe drinking water.  And I suppose modern chemistry should be included—something I know little about.  Many key products were first invented in the late 1800s or early 1900s (electric lights, home appliances, cars, airplanes, etc) and were widely adopted by about 1973.  No matter how rich people get, they really don’t need 10 washing machines.  One will usually do the job.  So as consumer demand became saturated for many of these products, we had to push the technological frontier in different directions.  And that has proved surprisingly difficult to do.

You can’t really make airplanes 10 times bigger than the 747, at least in a cost effective way.  We went to the moon in 1969, but still don’t have much better propulsion technology.  The Titanic was much bigger than earlier ships, but 1000 feet is long is long enough.  The new ships are wider, but no one is building 3000 foot liners.  Improvements in cars are now incremental.

I suppose there are three technologies that may drive the next wave of growth; computer chips, biotech and nanotech.  But only one of those has had a major effect so far, and although improvements in computers and cell phones have been awesome, it is just not enough to drive GDP figures higher at the rate my grandma saw.  Isn’t amazing enough that my grandma saw all those improvements in her lifetime?  Are we so spoiled that getting to live our entire life with all these goodies isn’t enough, we want another growth miracle as impressive as the 1920-70 period?  Unfortunately, it doesn’t seem like it is going to happen.

[I’ll defer to people like Robin Hanson on the prospects for the “out decades.”  Personally, I fail to understand why progress in biotech is such a great thing, if it means the ability of a Unibomber-type who thinks the world is overpopulated to build deadly viruses that spread like the common cold advances faster than our ability to develop cures for those threats.  And I’m not normally a Luddite.]

To summarize, I think people misunderstood my previous post.  I absolutely agree with Krugman that growth in living standards is now slower than in the 1920-73 golden age, which saw technological change affect almost all aspects of our life, not just the images we watch on illuminated boxes.    I was simply trying to compare neoliberal reformers, to those countries who reformed less aggressively, to get the marginal impact of those reforms.  I still think that the marginal impact of free market reforms and lower MTRs was positive.

So don’t tell me about 90% tax rates (which actually collected little revenue, BTW), even the inept Soviets churned out lots of goods in those decades.  You can run a steel mill or washing machine factory like you run the army, if you don’t care about quality.  Trying to recapture the 1950s with a statist economy is fool’s gold.  We’ll never again see that sort of growth–unless Robin Hanson is right.

PS.  I thought of another example.  In the 1960s we built lots of nukes.  What’s the newest thing in electric power generation?  Windmills!



60 Responses to “Why did growth slow after 1973?”

  1. Gravatar of marcus nunes marcus nunes
    23. May 2010 at 14:21

    “So don´t tell me about 90% MTR´s”. With 90% and with 33% (or 39%), Federal revenues stayed the same relative to GDP (always a bit less than 20%). Ceratinly efficiency was much lower with 90%

  2. Gravatar of William William
    23. May 2010 at 19:06

    A chemistry related example for you – the Haber-Bosch process. Basically this fixes nitrogen in the air and converts it into ammonia, and then fertilizer. First used on an industrial scale in 1913, currently supporting one third of the world’s population. And after nearly 100 years, it’s still probably the most important agricultural innovation yet. (Haber and Bosch both won Nobel Prizes for it.)

  3. Gravatar of back40 back40
    23. May 2010 at 19:07

    As Robin notes, it is AI that will bring another discontinuity in the step function.

    “I fail to understand why progress in biotech is such a great thing”

    Biotech, nanotech, much the same. . . but slower than AI. If the thunder don’t get you then the lightning will.

  4. Gravatar of Why did growth slow after 1973? « Economics Info Why did growth slow after 1973? « Economics Info
    23. May 2010 at 22:02

    […] Source […]

  5. Gravatar of Doc Merlin Doc Merlin
    23. May 2010 at 22:07

    ‘PS. I thought of another example. In the 1960s we built lots of nukes. What’s the newest thing in electric power generation? Windmills!’

    No, actually the newest thing is 4th gen nukes. They are just passing us by, however. Due to regulation in the US, we cannot build the newest styles of nuclear plants.

    In the 1970 we established a regulatory structure that required pre approval before certain types of technologies could be made. Since approval was almost impossible to get, this, in turn, meant that we simply stopped developing those technologies. (e.g.: We had a roughly 30 year period with no new refineries or nuclear plants.)

  6. Gravatar of Tim Worstall Tim Worstall
    24. May 2010 at 02:16

    Something I’d love to see (and don’t have the technical skills to do myself).

    We hear a great deal from the likes of Krugman about how post war growth was so wonderful. Then we get that linked to high unionisation, high MTRs and so on.

    However, as I recall, the period 1929-1945 wasn’t really all that great for either economic growth or growth in household income nor even growth in individual incomes.

    What I’d really like to see is how much of that great growth in those three things in 45-70/75 was in fact just a catching up with trend growth over the entire 1929-1975 period?

    To put it another way, over the long term rising incomes (however we define them) are a result of increasing productivity which is itself driven by technological advances (using a wide definition of technology, management systems are technology here).

    Both technology and productivity advanced in 20-45….incomes not so much. Incomes did increase in 45-75 as did productivity. But how much of the higher incomes in 45-75 comes from the catching up with the productivity improvements of 29-45?

    Would Mr. Sumner care to have a quick stab at charting this out for us/me?

    Maybe to make it really clear, it should be something like 1890-1990 as trend….

  7. Gravatar of Mattias Mattias
    24. May 2010 at 03:16

    Hi Scott,

    Mike Mandel has made a similar argument, that innovation seems slower today contrary to what many people think. He wrote several articles on this subject while he was at Business Week and has continued as a blogger. See http://innovationandgrowth.wordpress.com/

    I think the really interesting question is why innovation has slowed down. And is it just temporary?

  8. Gravatar of Indy Indy
    24. May 2010 at 04:43

    Stop, stop! I’m starting a book closely related to this topic and getting the ingredients of a premise circulating in the blogosphere will just end up crowd-sourcing all the insights before I can finish it. I promise you, we can revisit this conversation after publication.

    But let’s imagine a world where future technological innovation and progress in livings standards becomes much more difficult. Not a Dystopia, but just a kind of leveling-out of the path of improvements. I call it Plateaupia. Imagine what that world might be like. It’s an entertaining idea, especially the Economics of it, for me at least. I invite you to speculate about it. Call it “Social Science-Fiction” or maybe just “Economics Fiction”.

    But one of the things that might change in the idea that one can provide for one’s retirement with ease because of one’s ability to “free-ride” on the innovation-derived pace of real-growth in one’s investments and the simultaneous real-deflation in the price of certain technological products (think TVs, Microwave Ovens, and Computers, though I would imagine most technologies went through a similar price-path).

    In Plateaupia though, investments wouldn’t be expected to grow much in real terms – most markets saturated and most technologies matured. Prices wouldn’t change much easier, most product margins having been cut to the feasible bone.

    In short, the miracle of compound interest – the future progress of which we seem to have organized ourselves to be entirely reliant upon to get us out of our various jams (like housing-price appreciation was supposed to do for sub-prime borrowers) – would lose a lot of its magic.

    One final thing. Don’t expect to be able to convince almost anybody this is a real phenomenon (which it is). It’s kind of an unspoken religion – the faith in the endlessness of progress, that there is no natural limit to the frontier of knowledge or efficiencies in applied technologies. People implicitly believe there is no end of ideas in every domain – but in some fields we may have already reached it.

    In the few conversations I’ve had with people about this topic people ask me to give instances, but when I provide (what I consider to be) plausible examples (i.e. interstellar travel), I’m met with not just skepticism but a very puzzling and outright angry resistance, like I’m some kind of dangerous heretic for speaking my blasphemy.

    Perhaps Space-Sci-Fi fulfills some fundamental entertaining-hope function in modern people’s psychologies, and the very broaching of the subject that it may turn out that human-movement into deep-space may be permanently impractical deeply threatens that need for hope of change.

    If we are all surprised and “The End Of Ideas” arrives – and much sooner than any of us currently think – it seems that our culture will not be ready to quickly accommodate itself to an nature-imposed stagnation.

  9. Gravatar of scott sumner scott sumner
    24. May 2010 at 05:15

    marcus, That’s right, and we actually collect a higher share of taxes from the rich under the lower top rate. So the 90% rates even failed to make the rich pay a lot of taxes, which was supposedly the purpose. Any egalitarian should oppose those high rates.

    William, Good point. I knew chemistry was important, but I didn’t know how. But by 1973 all the low hanging fruit were taken. That’s why rocket technology stalled after the 1960s.

    back40, I’m not convinced that sophisticated AI will be here anytime soon. But then I don’t know much about the field.

    Doc Merlin, But even if you are right, that supports my point. In the 1960s we move from coal plants to nukes. Fifty years later the advance is – – – one more generation of nukes? That’s pretty slow progress.

    Tim, Good point. Alexander Field argues that the 1930s were the most technologically advanced decade in the 20th century. Obviously not much of this was implemented in the civilian sector until after 1945. So there was a lot of built up growth potential. There is a reason stocks were high in 1929–we were on the edge of a Golden Age. Unfortunately bad monetary policy and war deferred the Golden Age for 16 years.

    Mattias, I don’t think we should be surprised that growth has slowed down. If it didn’t then at the rate we were progressing in my grandma’s life in a few 1000 years we would have God-like powers. Does that seem likely? We picked a lot of low-hanging fruit once the basics of modern science were put in place during the 19th century. Now for the hard part.

    Indy, Interesting idea. But I don’t see why a steady state technological society can’t do fine. Just raise the retirement age as people live longer. Don’t worry, I won’t steal your ideas–I still think technological progress will continue, but at a slower rate.

  10. Gravatar of StatsGuy StatsGuy
    24. May 2010 at 05:57


    “So as consumer demand became saturated for many of these products, we had to push the technological frontier in different directions. And that has proved surprisingly difficult to do.”

    First, I agree with this post, but doesn’t this sort of put the kabosh on the notion that long term supply drives long term demand (as a one way road)? Doesn’t it look a little more two way, based on the arguments you presented above?

    It depends what you mean by supply of course – supply of existing demand, or supply of inherent demand that requires major new technologies (e.g. life extension). In the absence of the latter, productivity improvements in the former can actually cause oversupply, depress the price level, and contribute to structural unemployment _IF_ there is a minimum wage floor caused either by govt. policy _OR_ by a cost of survival in a society where some survival goods are bid up by higher income echelons (e.g. gasoline, healthy food, housing).

    To the extent the technology frontier is not bendable through additional resource allocation – and indeed many neoclassicals believe technology is a purely exogenous factor – then woudn’t the above serve as an underlying argument for socially agreed supply restriction (e.g. leisure preservation)?

    BTW, I disagree that the technology rate is exogenous; One of the key failings of govt. policy in the the past 20 years has been poor investment in basic research, at a time when industry has focused more on taking-to-market tech from universities than developing its own tech (and when universities have effectively developed into marketing agencies that sell brand names to students in exchange for funding professors to publish papers to improve their brand name).

    Good marketers, btw, believe that much demand is not inherent, but rather created. This would be supported by the correlation between inequality in advanced societies and happiness (if we can even measure it). Likewise with health and life expectancy.

    Has happiness improved in the US since 1973???

  11. Gravatar of Mark A. Sadowski Mark A. Sadowski
    24. May 2010 at 06:10

    Referring back to your previous post, Krugman’s central point, which was by extension Richard Green’s central point, was that Peter Wallison is a data free ideologue. After all Wallison is quoted as saying the following:

    “Indeed, the modern era of rapid economic growth commenced after both Democratic and Republican presidents undertook to lift costly and stultifying New Deal regulations.”

    We both know that this is simply not true. Growth was much more rapid prior to the neoliberal reforms (particularily median income growth). That being said I favor most neoliberal reforms and I think you have made a reasonably good case for them. People like Wallison on the other hand are a disgrace.

    P.S. I have an old book called “The Great Leap” by John Brooks (1966) that very vividly describes the economic progress achieved in the US in the quarter century or so between 1939 and 1965. It should be required reading for people like Wallison.

  12. Gravatar of Jeff Jeff
    24. May 2010 at 06:11

    Has happiness improved in the US since 1973???

    Of course it has. We didn’t have blogs back then.

  13. Gravatar of q q
    24. May 2010 at 06:59

    not to mention that nobody has written a poem as good as this one since 1973.


  14. Gravatar of Morgan Warstler Morgan Warstler
    24. May 2010 at 07:00

    Let’s go the other way…

    The rate of technical innovation today is much faster – who cares if you can fly a 747 to Europe, what matters is I can video conference and I don’t have to go to Europe.

    The moving of the atomic is brutal and becoming meaningless. And putting emphasis on it out weighs the body when the mind is just getting started. From the aqueduct to the toilet took how long? From teepee to skyscraper took how long?

    And there’s the rub, we’re smarter now. Far smarter now, than in the 1950’s caring about things lower on Maselow’s hierarchy is silly.

    Would a revolution in Napsterizing Education – driving down the cost of knowledge transfer to 1% of what it is now impress?

    Would we all stand back and say “that’s much BIGGER than the airplane?”

    Because if not we’re not doing the math right.

  15. Gravatar of Rafael Rafael
    24. May 2010 at 07:39

    I remember a post where Krugman has made exactly the same arguments!!!

    Here: http://krugman.blogs.nytimes.com/2009/11/08/technology-in-daily-life/

  16. Gravatar of spencer spencer
    24. May 2010 at 07:48

    Is your basic premise correct?

    From 1850 — by this time most of the basic features of modern financial market capitalism were in place — to 1950 trend real per capita GDP growth averaged 1.6%. Moreover, the 1930s-1940s period does not change this trend as the 1850-1930 trend was also 1.6%.

    Since 1950 the trend growth of real per capita GDP has been 2.1% and we really did not fall below this trend until the most recent recession hit.

    However, if you want to use other data to argue their was a break in growth in the late 1970s it has to be associated with the 1974-1994 slow down in productivity growth. Since 1995 productivity growth has reverted to the pre-1974 trend, as did real per capita GDP growthdid in the late 1990s.
    So the question is why did productivity growth slow from 1974 to 1994. Several reasons have been advanced for this slowdown, but I believe that higher real energy prices played a major role because in the low productivity era firms focused capital spending on improving energy efficiency at the expense of labor efficiency– just the opposite of the long term trend.

    Maybe this is why we should be worried about slower growth.
    From the industrial revolution to modern times growth in standards of living stemmed from replacing muscle power with
    cheap energy. Even in food production when American farmers used animals rather than tractors to raise crops they had to allocate up to a quarter of their crop land to producing feed for their draft animals — that explains much of the collapse of real agricultural prices in the 1920-30s.. But now we are facing an environment where energy is becoming much more expensive and it will be much more difficult to replace muscle power with cheap energy.

  17. Gravatar of spencer spencer
    24. May 2010 at 08:11

    Tim Worstall — if you chart trend real per capita gdp from 1850 to 1950 you find that in 1950 actual real per capita GDP
    was significantly higher than trend real per capita growth.

    So on his basis is your basic premise that the 1950s growth was boosted strongly by a catch-up from from below trend growth from 1929 to 1945 actually correct.

    I’m going to email Scott an excel file with this data charted so if he cares to post it he can.

  18. Gravatar of Eric Morey Eric Morey
    24. May 2010 at 08:11

    1973 is an interesting inflection point since it coincides with the “Oil Shock”. I wonder if this actually creates statistical noise at the beginning of the longer term decline in growth rate be caused by other factors.

    I also wonder to what extent copyright and patent law changes from the 1970’s to present put a damper on growth in an increasingly information driven economy. As software patents in the form of business process patents have become more commonly accepted by the courts and companies have become more aggressive in enforcing increasingly strict copyright law over the last 40 years. Has there been a significant economic impact to the USA and the World?

  19. Gravatar of William Newman William Newman
    24. May 2010 at 08:19

    (on http://www.themoneyillusion.com/?p=5190#comments)

    scott sumner writes “Unfortunately bad monetary policy and war deferred the Golden Age for 16 years.”

    I recognize this as an honest paraphrase/simplification of the usual explanation of the Great Depression. Alas, I’ve never understood it. And in this context, when your’re discussing how the neoliberal policy changes 1975-1985 had a very large impact, it seems particularly hard to understand. The non-monetary-policy changes in US economic policy starting in 1929 look bigger to me than the more modern neoliberal economic policy changes under discussion. Hiking tariffs to over 50%, hiking marginal income tax rates to well over 50%, introducing a national minimum wage, various unionization and cartelization policies, new taxes targeted at employment, new subsidies for unemployment, and arguably a big jump in regime uncertainty making it difficult to calculate expected return on investments… If today someone highly-free-by-modern-standards economy with changes like that, I’d expect quite a shock even if their central bank acted brilliantly to maximize economic welfare subject to the constraints of the new non-monetary policy.

    Could you recommend a particularly well done summary analysis, ideally in article length and accessible to the interested but skeptical layman, explaining why we should be confident that monetary policy was the dominant cause of the Great Depression?

  20. Gravatar of Doc Merlin Doc Merlin
    24. May 2010 at 08:27

    “Doc Merlin, But even if you are right, that supports my point. In the 1960s we move from coal plants to nukes. Fifty years later the advance is – – – one more generation of nukes? That’s pretty slow progress.”

    Yes, agreed Scott, my point was that because Nukes in the US were so heavily regulated as new nukes to almost be outlawed, we didn’t have the advancement that we would have gotten otherwise.

  21. Gravatar of Doc Merlin Doc Merlin
    24. May 2010 at 08:35

    I guess what I am trying to say is, legislative/regulatory market restrictions create disincentive for technological advancement.

  22. Gravatar of sourcreamus sourcreamus
    24. May 2010 at 09:09

    The changes we are seeing now are more quality changes than new things. When I was a kid we had almost everything we have now, just everything sucked. Video games were two dashes playing ping pong with a dot. You had to adjust rabbit ears on you TV set so you could see which of the four shows available you wanted to watch. I could go on and on.
    I wonder if the regulatory state and the general zeitgeist of technology as something to be feared has something to do with it. The world of tomorrow was once about marvels like flying cars and microwave ovens, then it turned into Chuck Heston watching Edward G Robinson die. It seems like nuclear weapons made people much more aware of the dangers of technology and created a demand for the layers of red tape that slow everything down. The Empire State Building took 6 months to build and whatever replaces the Twin Towers will probably be built sometime in Marco Rubio’s second term.

  23. Gravatar of Lord Lord
    24. May 2010 at 09:36

    The risk is the return of the Malthusian age and the discovery that the last two centuries were an aberration. Life goes on, it just becomes less lively and attention is directed to both the more mundane and the more etheral.

  24. Gravatar of Jason Ruspini Jason Ruspini
    24. May 2010 at 14:11

    I’m surprised that no-one has mentioned demographic structure and population growth. Along with rate of technological change, this seems to be one of the most glaring differences between today and the first half of the post-war period.

  25. Gravatar of Scott Sumner Misses One Obvious Explanation Scott Sumner Misses One Obvious Explanation
    24. May 2010 at 15:00

    […] has a theory for why world economic growth suddenly slowed in the early 1970s: Here’s what I think happened. […]

  26. Gravatar of James Picerno James Picerno
    24. May 2010 at 16:16

    One could argue that the last several decades have been an information revolution on an unprecedented scale. Perhaps this type of revolution doesn’t show up in the standard economic statistics in the way that moving from candles to electric lights does or going from Lindbergh to 747s. But surely the explosion in information, from medicine to farming to industrial technology, represents an exponential change in the course of human knowledge. Think of the advances in pharmaceuticals today vs. even, say, 1980. Wow! Some of the improvement in knowledge is of questionable value–the rise of marketing intelligence, for instance. But on a grand scale, we know so much more today, for good or ill, compared with just a few generations previous. It’s debatable how much of this relevant for economics, but if there’s a short list of big changes, information is surely a leading candidate.

  27. Gravatar of Igor Igor
    25. May 2010 at 06:14

    One possible answer is that innovative spirit was more suppressed during 1930-1973 period then in the period before 1930 due to higher regulation and taxes. “Golden age” simply reaped the benefits of implementation of innovations and inventions of preceding era, while not creating a suitable environment for the same development in the future.

  28. Gravatar of Dolo Dolo
    25. May 2010 at 06:51

    The last famine in the US was the Dust Bowl during the Great Depression in the 1930s. Since then, the vast majority of people have had enough to eat. Today, people with jobs don’t try to work extra hard to get a promotion and feed their families, because the vast majority of people with jobs can feed their families.

    From lower middle class to upper class, most people have refrigerators, stoves, cell phones, cars, clothing, etc. These days, the marginal income that you get from extra hard work isn’t useful for creature comforts as much as for signalling and peacocking. This is important because working hard to buy nice things so that society respects you is in direct competition with NOT working hard and having a lot of leisure time so that society respects you. The ability to devote a lot of time to leisure is a powerful form of signalling, which means that sometimes NOT producing can result in more utility than producing.

    Who is considered more wealthy in our society, the man who works hard to afford a new Mercedes, or the man who drives a Honda but has retired at age 40?

    Historically, there have always been idle elites who chose not to work hard to generate more income, but to enjoy what they have had. Today, we are seeing that level of prosperity extended to the majority of people. Suddenly more and more people are able to work less hard, and gain the respect of their peers not through hard work, but through the lack of it.

    Society has given up on the marginal production of many members of the middle class, and I believe that has caused economic growth to slow down.

  29. Gravatar of James Crimmins James Crimmins
    25. May 2010 at 07:11

    How much better can things get for the very well off in rich countries? Probably not much better. Because of reversion to the mean issues, we can expect great increases in growth for poor countries but slowing growth in rich countries. A related problem is the disappearance of the frontier. There are not vast areas to be settled/conquered on the planet any longer with attendant resource grabs. Biotech/nanotech may prove important in energy/synfuels and food areas. Again this benefit will (mostly) accrue to poorer countries and those with limited natural resources.

  30. Gravatar of Valuethinker Valuethinker
    25. May 2010 at 07:33


    Windmill is at least as technologically impressive a technology as nuclear power– perhaps more so.

    It is much simpler, it is deployable in relatively small units (no $10bn to build 1350 of GW power, you can build 1/1000th as much for less than $4m) and so is better attuned to the grid requirement, it uses more sophisticated technology (those gearboxes, the blade design etc. all come out of some very clever work on aerodynamics and mechanical engineering).

    And of course the cost is entirely frontloaded. Operations and Maintenance spending are less than 5% of cost, pa.

    Whereas with a nuclear plant you have back-ended the cost. The UK government is going to spend £70bn+ on nuclear decommissioning, and still has no long term plan for what to do with the radioactive waste (stick it somewhere, and let humans 5,000 years hence deal with the costs).

    That excludes the nuclear proliferation externality which is surely not small.

    Wind power and its widespread use is a measure of the pace of technological change in the power system, that we can build a power system that taps this resource, deals with its issues (intermittency) and produces electric power with little or no externalities (noise if you are within a few hundred feet, and a visual externality which is not worse than a power line or a tall chimney or a motorway spaghetti junction).

    Saying nuclear power is more technologically impressive than wind power is like saying that the SS Great Eastern is more impressive than a modern container ship, because it had more propulsion systems (sails, screws and paddle wheels) and because its piston engine was more complex than modern turbines.

    On radical technological changes of the 1940s period, antibiotics is probably far and away the most important, other than modern sewerage (which is much more important). The death rate from ordinary cuts as well as operations used to be far, far higher before the discovery of penicillin. One reason why penicillin was kept a military secret during WWII.

  31. Gravatar of Jon Jon
    25. May 2010 at 07:35

    Scott remarks:

    You can’t really make airplanes 10 times bigger than the 747, at least in a cost effective way. …

    But only one of those has had a major effect so far, and although improvements in computers and cell phones have been awesome, it is just not enough to drive GDP figures higher at the rate my grandma saw.

    ‘microchips’ are only possible because of a huge capital investment. Moore’s law is commonly cited as being about the bounty of technological improvement, but his original presentation what about something much different: exponential development cost. Figure two of the pdf gives the classically cited ‘Moores law’ chart, but is supporting evidence rather the message. The message is in figure five of his presentation, the vertical scale is ‘person-month development time’. http://lostdollars.org/static/moorekeynote_1979.pdf

    This was a talk given to research scientists at the university level calling for improvement in the tools used to build microchips.

    The capital investment necessary to bootstrap a semiconductor firm is regarded as around 250 million dollars. That’s a big number. The cellphone of today is the fruit of an incredibly deep capital structure. I suspect that if the Soviets had won, there would be compact smart-phone cell-phones. There simply is no equivalent semiconductor industry in Russia.

  32. Gravatar of scott sumner scott sumner
    25. May 2010 at 07:36

    Statsguy, Yeah, I shouldn’t have written that sentence, but I am not talking about demand, I think supply is the key. Thus there is demand for transport, light, etc, and the car and electric light dramatically reduce the cost of transportation and illumination compared to candles and horses (especially if you include the time cost of transport. That’s what I should have emphasized.

    That sentence was making another and different point, which I may expand upon in a post.

    You said;

    “BTW, I disagree that the technology rate is exogenous; One of the key failings of govt. policy in the the past 20 years has been poor investment in basic research,”

    Actually, this supports my point. The government has poured massive funding into medical science research, and we are getting less and less as they spend more and more. If you already have the smartest 10,000 scientists doing medical research, then doubling funding so another 10,000 more mediocre scientists start publishing mediocre papers in journals, doesn’t get you a cure for cancer any faster.

    Mark, yes, that statement was incredibly sloppy, I agree with Krugman on that, but I didn’t think it even worth addressing. But Krugman’s newer post makes it clear he opposes Reagan’s reforms, and that was the issue I addressed.

    Jeff. But Krugman’s blog makes me unhappy.

    q, Thanks for that.

    Morgan, Lots of people think this way, but I still think the gains from the 1890s to the 1950s dwarf what has happened since.

    Rafael, Well, at least we agree on that.

    Spenser, You said;

    “From 1850 “” by this time most of the basic features of modern financial market capitalism were in place “” to 1950 trend real per capita GDP growth averaged 1.6%. Moreover, the 1930s-1940s period does not change this trend as the 1850-1930 trend was also 1.6%.
    Since 1950 the trend growth of real per capita GDP has been 2.1% and we really did not fall below this trend until the most recent recession hit.”

    I’m confused. I didn’t argue that capitalism was the key, but rather the technologies of the late 19th century that got implemented in the 20th century.
    BTW, I think growth in the modern era (post 1994) is overstated by the GDP numbers, as they put too much weight on minor innovations in computers, cars, etc. So I think growth in actual living standards really did slow after 1973.

    You may be right about energy, (and I would add environmental regulations) but remember that energy got cheap again in the 1980s and 1990s.

    Eric. Yes, the oil shock played a role, but only short term. Our patent laws may be too restrictive, but I’m not an expert there.

    William, Actually I agree with you. Go back a few months and you see a bunch of consecutive posts on the Depression. In the first one I lay out my argument, which is that monetary policy caused the contraction, and the supply-side problems you mention explain the length of the Depression. Recovery was delayed by 6 or 7 years by those supply-side policies. But keep in mind that those bad supply side policies were a response to the initial stages of the Depression, which were caused by tight money. They wouldn’t have happened otherwise.

    Doc. Yes, but I’m not sure progress would have been that fast even w/o regs. The physics of energy is what it is, and basic science doesn’t yet give us a cheap alternative. We did lots of research on fusion, but it was a dead end.

    sourcreamus, Yes, regs have slowed things down. But as you say progress has gone from new things (what my grandma experienced) to better things (what I experience.)

    Lord, Yes, but at least we don’t have the high birth rates anymore, so at least the Malthusian age wouldn’t necessary bring famine back.

    Jason, It explains part of the difference between family incomes and per capita GDP.

    James, Yes, a lot more information, but life expectancy grew faster in my grandma’s lifetime. Don’t under-rate things like electric lights. The world was very, very dark before lights. Candles really aren’t that effective. When the power’s out and you light a candle, it doesn’t really help that much.

  33. Gravatar of Valuethinker Valuethinker
    25. May 2010 at 07:44

    Doc Merlin

    Nuclear energy had different regulatory regimes in other countries.

    Is the worldwide slowdown in nuclear power deployment in the UK, Germany, Russia, Canada and the rest of the world the result, then, of increased US regulation?


    The only country that really drove on with it was France, about as far from the free market paradigm as anyone could imagine. And France wrote off a huge amount when they privatized EDF (ie a state subsidy written off). In addition, even French reactor construction slowed to a near halt after the early 90s.

    In fact what happened is nuclear costs were far, far over budget and utilities overestimated demand (there was also a step change down in the cost of gas fired electricity production, arising from a technological revolution in the creation of Combined Cycle Gas Turbines– CCGTs now run over 50% efficiency, and have low capital cost, which meets the demands in a deregulated utility era).

    So the great nuclear building boom died a death. Three Mile Island surely did not help, but the nuclear boom was dying under its own weight.

    With deregulation, the next utility building boom did not take place until the 1990s, and that was met with CCGTs as above– where there was remarkable technological change.

  34. Gravatar of ssumner ssumner
    25. May 2010 at 07:50

    Igor, Maybe, but I think it was mostly luck. We invented the IC engine, which made cars possible. What’s the next step? Flying cars? That’s really really hard to do.

    Dolo, I think it is natural to enjoy more leisure as we get richer. Otherwise, what’s the point?

    James, Yes, we’d need scientific breakthroughs for the well off to get much better off.

    Value thinker, Yes, I should have mentioned antibiotics. I’m not against windmills, but they don’t produce electricity at a very low cost. Nukes aren’t that great either. We aren’t making much progress in energy.

    Jon, Thanks for that info.

  35. Gravatar of Valuethinker Valuethinker
    25. May 2010 at 07:56


    There’s a huge literature about the ‘step change’ down in global productivity growth post 1973.

    (see James Hamilton (UCSD) in particular for cites, also Robert Gordon (Northwestern))

    Best guess is that the oil shock caused a reallocation of capital resources. A huge rise in relative price of one factor input (energy) forced companies (and individuals) to scrap huge amounts of capital equipment and replace it with more energy efficient equipment.

    This process took years and fundamentally lowered the returns to capital for that period.

    There was a lot of ‘learning’ in this. We didn’t know how to make products in a more energy efficient way, using less energy, less plastic etc. We didn’t know how to build 35mpg cars, and so we had to figure out how.

    Given the subsequent oil shock in 1979-80, that process took over 10 years, and was accompanied by a significant restructuring of British and American manufacturing.

    So what might otherwise have changed labour productivity levels, went instead into compensating for a huge adverse change in the cost of making things (and creating services).

    Contrast to the mid 90s, when WalMart pioneered a distribution revolution, using computer technology. McKinsey estimated the main drivers of productivity growth were

    – WalMart and the Big Box stores
    – online stockbroking a la Schwab

  36. Gravatar of Valuethinker Valuethinker
    25. May 2010 at 08:11


    Wind is still on the downward curve, a -15% cost reduction per MW capacity with every doubling of world capacity (solar is on that curve as well, but at a much higher level).

    That’s usual for a new energy technology. Experience curve learning plus scale benefits are still at work.

    The result is wind-fired electricity is getting fairly cost competitive with new fossil fueled plant (it is so, already, in places like Ireland).

    Offshore will be the next step up. 2-10 times the capital cost but very significant performance improvements associated with more reliable wind and larger turbines (5MW+). So in that sense wind power is, if not in its infancy, having a robust early middle age. Incredible what has been achieved in 20 years.

    Solar of course there is more to go for (but you are starting from further behind).

    As I say, Combined Cycle Gas Turbines, by doubling efficiency between the late 70s and the mid 90s, staged a similar revolution in energy technology (from c. 35% efficiencies to over 50%).

    Energy shifts do not occur with any rapidity. Vaclav Smill has documented that fairly well. In 1930, the world generated electricity with water, and with coal.
    In 2010, the largest source of electricity is still coal (roughly speaking, with c. 2 times the thermal efficiency), then gas, then nuclear (about 16%), then water (about 8-10%).

    The optimal scale of coal-fired unit though peaked out at about 550MW (a really big station, like Nanticoke in Ontario, or Drax in the UK, or the Southern company one (Bowen) seems to basically have a series of c. 500-600mw units, giving total capacity of 3-5,000 MW). Until Intermediate Combined Cycle Gasification (ICCG) becomes a significant factor, thermal efficiency has topped out at around 40% (for the best in class European and Japanese plant- Americans having cheaper coal and weaker environmental regs, run somewhat less efficiently).

    Gas has gone from sub 2% in 1973 to around 15% of world electricity now, almost exactly offsetting the contribution of heavy fuel oil which has gone from something like 15-20% back down to sub 3%.

    The biggest productivity revolution of the 1950-1980 period was almost certainly the container ship and the multi-modal container (truck, train, ship). That really did alter world economics.

    You get another step change with WalMart and that in the mid 90s and distribution.

  37. Gravatar of Valuethinker Valuethinker
    25. May 2010 at 08:23


    ‘we invented the internal combustion engine, which made cars possible’.

    Oops. It’s axiomatic in the literature of path dependence that the steam car (an external combustion engine) had a number of advantages over a IC car (a Dobie, of the kind Jay Leno owns, can do 250k miles without a gear checkup) and, had Stanley and Dobie been run by better businessmen, might well have won the race to power the car. Like the QWERTY keyboard or the fact that we drive on the left (well, civilized countries do ;-)), it’s a function of path dependence.

    So cars were not made possible by Otto Diesel nor by the various inventors of the petrol driven engine.

    IC maybe made powered flight possible.

    The big revolutions in transport were the steam train (London to Manchester from 3 days down to half a day), the steam ship (Southampton to New York from 3 weeks+ to 1 week) and the telegraph (London to New York from 7 dads to less than 3 minutes).

    We haven’t, in the 20th century, really beaten those metrics. Maybe the internet and the mobile phone are getting close. The jet plane yes (7 days to 7 hours) but still not quite as dramatic as being able to sail from Southampton to New York in 7 days, and travel from New York to San Francisco in ?another 5? by rail.

  38. Gravatar of Chris T Chris T
    25. May 2010 at 09:58

    Much of the early to mid 20th century GDP growth was likely more due to the massive shifts in the population to a far more efficient economic distribution of skills and ability than by technology directly (technology allowed the shift though). Since the redistribution is more or less complete, GDP growth has slowed.

  39. Gravatar of Chris T Chris T
    25. May 2010 at 11:26

    A lot of people above are making observations based on an extremely flawed definition of technology. A car is not a technology. A modern car is made up of thousands of components, each based on an array of enabling technologies. Each based on decades of advancement in a variety of different fields.

    The only thing a modern cell phone and a rotary phone have in common is their basic function and name.

  40. Gravatar of Jon T Jon T
    25. May 2010 at 11:35

    Also, don’t forget that 1920-1970 saw WWII and much of the Cold War, both of which spawned significant technological innovation. First you had the establishment of a huge manufacturing base in the 40’s, then decades of technological rivalry during the space/arms race.

  41. Gravatar of Bob Roddis Bob Roddis
    25. May 2010 at 16:52

    Nixon announced closing the gold window simultaneously with WAGE AND PRICE CONTROLS in August 1971! And the Fed stepped on the gas. It got Tricky Dick reelected in 1972 in a landslide. Naturally, things started going south real fast in 1973 and by 1974 the economy was in the tank. Newsweek had a cover wondering “Are We Running Out of Everything?” Probably the real reason they got rid of Nixon that year.

    Lucky for me, I had discovered Murray Rothbard and the Austrian School in 1973 so I knew that the Fed money dilution and price controls were the culprit. The money dilution causes terrible distortions in the price, investment and capital structures and price controls CAUSE SHORTAGES!

    Thankfully, I went straight from Marxist Democratic Socialism to the Austrian School without ever having to be dis-informed by the Keynesian Hoax.

    This post about a lack of technology is a joke, right? Trying to catch the sleepy-heads?

  42. Gravatar of James Picerno James Picerno
    25. May 2010 at 17:02

    Scott, yes, candles don’t help much when you’re in the dark, but maybe they don’t hurt either. James Madison wrote the Constitution by candlelight, and I can think of a few other rather well-received tomes penned in the dusky evenings of hte pre-electric age. This has nothing to do with economics, of course, but the topic comes to mind: technology usually helps, but not always. Or is the writing/art better in the 21st century? Hmmm…

  43. Gravatar of Larry Larry
    25. May 2010 at 21:48

    Definitely time to check out Kurzweil’s “Singularity”, in which he makes a bunch of ridiculously bold but well-supported claims about societal change rates. His thesis is that these rates, properly construed, are increasing exponentially in essentially every domain.

    He forecasts the most visible and utterly transforming changes in the GRIN-tech(genetics, robotics, IT (really AI) and nanotech.) I would add energy to the list, given that I believe that this century will finally bring cheap, sustainable energy from some combination of nuclear fusion, solar, and battery technologies.

  44. Gravatar of scott sumner scott sumner
    26. May 2010 at 08:46

    Valuethinker, Obviously you know a lot about these issues, but with all due respect I think you and others are too mesmerized by technological specifications. Economic growth is about consumption. Despite trains and steamships and other innovations, my grandma, like most people, lived a very primitive life in 1890. Again, no cars, no lights, no home appliances, no indoor plumbing, no airplanes, no TV and radio, not antibiotics, etc. Sure she could walk or take a horse to a train station, and go somewhere. And she could send messages by telegraph. But fast forward to when I was born in 1955. Everything was totally transformed. I was born into a recognizably modern world, with all the things I just described.

    Sure, things like windmills are incrementally getting better, and I favor them. But it isn’t going to transform my life. The internet is the only major change I noticed, although things like TVs are obviously much higher quality than before. I agree living standards are still rising, but the pace of change is slower. We don’t need to point to energy shocks in the 1970s, one technological revolution played out, and the new ones aren’t as fertile.

    Chris, I agree that you need to look at underlying technologies, and also functions. Going from no phone to phones, is much bigger than phones to cell phones.

    Jon T, You said;

    “Also, don’t forget that 1920-1970 saw WWII and much of the Cold War, both of which spawned significant technological innovation. First you had the establishment of a huge manufacturing base in the 40’s, then decades of technological rivalry during the space/arms race.”

    A common misconception. Almost all the key technologies for our post war lifestyle were invented earlier. Jet engines are one exception, but we already had airplanes. The nuclear stuff had little impact on living standards, ditto for rockets. The keys were electricity, the IC engine, modern chemistry, indoor plumbing, etc. And the jet engine would have been invented a few years later, even w/o the war; technological progress in civilian aviation was very fast prior to WWII.

    Maybe antibiotics? Was that from WWII?

    Bob, Thanks for warning people away from Austrian economics. 🙂

    Larry, I am no expert there, but am told his predictions have been over-optimistic so far.

  45. Gravatar of They’re coming to America! Some Thoughts on American Demographics and Growth « Left Outside They’re coming to America! Some Thoughts on American Demographics and Growth « Left Outside
    26. May 2010 at 13:01

    […] the collapse of the post war settlement and neoliberal response many states took to bolster growth. Scott Sumner argues that growth slowed because technological progress slowed and neoliberal reforms helped […]

  46. Gravatar of MixedRealities :: How rich can we get, creating our own economies? MixedRealities :: How rich can we get, creating our own economies?
    26. May 2010 at 13:54

    […] links to a post by the economist Scott Summer on the blog TheMoneyIllusion, about innovation. Summer says that the technological developments are relatively limited: When […]

  47. Gravatar of The Innovation Boom | The League of Ordinary Gentlemen The Innovation Boom | The League of Ordinary Gentlemen
    27. May 2010 at 07:38

    […] and technological innovation around these parts, you may be interested in Scott Sumner’s take on the 20th century’s innovation boom. The scope of technological change at mid-century really puts our own “digital […]

  48. Gravatar of Brett Brett
    27. May 2010 at 08:36

    You can’t really make airplanes 10 times bigger than the 747, at least in a cost effective way. We went to the moon in 1969, but still don’t have much better propulsion technology.

    I think the slow-down in external aviation technology has a lot to do with two events in the 1960s and 1970s:

    1)The hobbling of Supersonic Transport (SST). Had the politics not crippled it (including a US congressional ban on overland SST), it would have opened the door to a whole new arena of passenger transport planes and designs.

    2)The changes in US nuclear deterrent under Secretary of Defense Robert McNamara. Before McNamara, the US nuclear deterrent was based around bombers, and the trend was towards higher and faster bombers (culminating in the B-70, which was canceled). That’s a major potential source for innovation in aircraft design, that was more or less stunted when McNamara started pushing towards a deterrent based on ICBMs.

  49. Gravatar of ssumner ssumner
    28. May 2010 at 06:35

    Brett, I’m not so sure. We have the technology to go faster, I agree. But it is very expensive. The Concorde’s flights to Europe were too expensive for most people. The fuel costs are very high.

  50. Gravatar of valuethinker valuethinker
    5. June 2010 at 08:49


    I am surprised your grandmother did not have domestic appliances in 1890. Most rural Americans did at least outside of the South. The Sears catalog predates 1890 I am pretty sure.

    On domestic water closets I don’t know what the timing was in North America (except that until the 1950s the most popular joke in North America were ones about the dunny– a somewhat lost aspect of N Am humour). However in London, the domestic water closet is pretty universal by 1900– from the 1890s, houses are built with bathrooms included, rather than added on.

    The US was ahead of us in electric street lighting and electrified households, and way ahead of us in telephone service. (we had a big gaslight infrastructure, hence the famous line by the Foreign Secretary, Lord Grey, watching the gas lighters in Hyde Park ‘the lights are going out all over Europe, and we shall not see them lit again in our lifetimes’ re the onset of WW1).

    Your grandmother’s life (and mine, she was born about 1890) had transformed dramatically over the previous 30 years. Railway infrastructure had brought consumer goods and long distance travel. Telegraphs had brought near instant communication. Towns were installing elaborate streetcar networks, leading to the first commuter suburbs (London had earlier had horse drawn omnibuss). Typewriters had revolutionized office work.

    Over to the question of real consumption. AFAIK real consumption grew in the 19th century at c. 1% pa ie doubling roughly every 72 years.

    In the 20th Century, it has grown faster. From 1900-1913 (associated effect was a sharp rise in real wheat prices– which benefited the farmer and towndweller of the US midwest and Canada). Again from 1920-1929 (but not in the UK as much as the US).

    Then again from 1946 (US) and 1950 (Europe) until 1973. In that latter case rising by something like 3% pa.

    One problem is that we don’t measure real consumption correctly because our deflators don’t capture the sorts of quality improvements that have gone on. Consider medical care now vs. 40 years ago. Or the instantaneity of the internet or the mobile phone. Or the quality of the average US car vs. the average US car in 1968 (which wouldn’t be legal now given safety and pollution standards, let alone performance norms).

    If anything, I would say the pace of technological innovation has picked up in the last 20 years, as opposed to the 20 years before that. That lightbulb on my desk uses 1/8th the power of its predecessor to generate the same light.

    You really notice the transformation that is going on if you visit Zambia say. 20 years ago bicycles were rare, cars even rarer, phones and air conditioners unknown. Now I would say 1 in 10 adults may have a bicycle, 1 in 4 a phone, there are serious traffic jams, and there is air conditioning on the homes of rich people and in offices. Note that Zambia is a particularly poorly performing emerging market in GDP terms (the worst, that have gone backwards, are those that have had wars).

    Over to the USA. 40 years ago, it was a sign of wealth to have 2 cars and more than 1 television. Dishwashers were not universal. Childhood leukemia was a death sentence. Houses (I’m talking NE & Canada here) had 1-2 room air conditioners.

    So in summary:

    – the pace of technological change probably really was the highest in the period 1850-1900 in terms of the drops in travel time, communications time etc. that took place

    Balanced against that the abiiility to fly transatlantic is important (and not completely dependent on the jet engine– prototyped in 1939 by the way)

    – major technological changes typically take 30-40 years to get embodied into capital. The canonical example being electric traction in industry, which did not become widespread until the 1920s and not universal until after WWII.

    We are seeing the same effect with the internet. Invented in the late 1960s, indispensable to do business around 2000, transformation of industries still not complete in 2010 (we still read paper newspapers).

    Note that none of the above probably show up in GDP numbers, because they change the quality of GDP (ie the deflator is overstated).

    – it may well be that the unusually strong per capita consumption figures of the ‘Golden Age’ of 1946-1973 had a couple of causes:

    – period of cheap and falling energy prices
    – post WWII ‘catch up’ – European living standards probably reached pre war in about 1955
    – uniquely favourable demographics led to steep real rises in head of family incomes, at the same time as consumer credit was vastly expanded (this latter factor was instrumental in the strong per capita consumption rises in the UK in the 1930s– the availability of hire purchase to finance consumer goods, and the widespread availability of Building Society mortgage finance. The latter, combined with the expansion of the streetcar and Tube networks, plus the first widespread ownership of motor cars, led to the creation of the vast 1930s suburbs of London, Birmginham etc.).

    – what happened in 1973 was the Oil Crisis (the first of 2)- -which would take oil from $3/bl to $40/bl in spring 1979. There were other factors (like the Food Crisis of the subsequent years, and the record bad weather of 1976). But for the next 10 years the economies of the western world were in an accelerated programme of investing capital to save energy, and discovering new ways of using less energy. An ongoing process which is still not yet complete (look at the McKinsey curves of NPV of energy saving innovations: a number of NPV positive innovations are still not universally adopted)

    – the microelectronics revolution may not have increased GDP much, but it has transformed life. It’s difficult to name a household item that has not been transformed by microelectronics in the last 40 years (just the control systems on your fridge, thermostat etc.– let alone your stereo, TV or home computer).

    – you and Brett are having a to and fro re transport. The revolution in transport is the low cost one. In the last 20 years, real air traffic out of Britain has gone up something like 4 times as a result of the growth of low cost airlines. That has been achieved by radical reductions in cost per seat (capital cost, fuel cost, service cost, aircraft turnaround time). No top speeds have not risen, bu the number of people flying and the price they pay per flight have changed dramatically. Twice as many people flying each week London to Toronto at half the cost doesn’t change GDP, but it sure as heck is growth in real consumption per capita.

    On energy itself:

    – the oil industry has had several technology revolutions in the last 40 years– 3D and 4D visualisation, horizontal drilling, offshore drilling (;-)) etc. Hence the Hotelling Curve has yet to come true

    – the natural gas industry has just had the technology revolution of all time (shale gas and coal bed methane) which has dropped the cost of c. 1/3rd of US energy consumption by about 60%. That revolution is only just beginning to play out (the geopolitical implications especially vis a vis Russia are titanic)

    – the power production industry has not radically changed (it is confronting Cornot’s Law) but it has made some pretty striking improvements– mostly around refashioning power production to be much more responsive to the demand curve (ie Combined Cycle Gas Turbines)

    My bet is the growth of real per capita consumption has declined in large part because of demographics (a shift from consumption to build cars, houses, schools, make baby food etc. for growing families to things like health care) and also perhaps because of rising income inequality– increased borrowing just has not been able to sustain consumption growth per capita against stagnant incomes per capita.

    But if we leave North America, we can see revolutions in consumption taking place. Bicycles, motorbikes, mobile phones, internet cafes, actual obesity in some countries such as India– hitherto unknown.

  51. Gravatar of valuethinker valuethinker
    5. June 2010 at 08:59


    When you count in the pollution externality of the SST (ozone layer etc.) then it wasn’t a very clever idea.

    Nor did the economics of the Concorde ever play out. A bit like nuclear power, it was only profitable after the government wrote off the development costs. Adn a friend of mine regularly flew Concorde (and was an aviation analyst)–the condition of those planes terrified him.

    Supersonic flight was probably a technological dead end. The time gains are only impressive when we talk really long distance (London to Sydney etc.). And of course supersonic flight was banned over land (pollution externality again).

    See my post above, the real revolution in flying in the last 20 years has been the low cost airline: very typical of a ‘system innovation’ which is not just about cheaper planes per seat, but a whole systemic innovation in how we fly (see WalMart)– Ryanair and Easyjet, Southwest etc. are the WalMarts of flight. Flying has now come into the reach of not just the professional classes, but down to the workign classes. In 1980 something like 20% of Brits had had a holiday abroad. In 2010 the number must be something like 80%.

    As to the strategic dimension of bombing, the B70 was obsolete because of improvements in Soviet Air Defence Missile (SAM) technology– flying at 100,000 feet no longer protected you. Same reason the British V Bomber force moved to low level attack. And the US did go on to build the B1 and the B2, so bombers didn’t end. There was never a good case for that, as the cruise missile and the ICBM both had a greater chance of penetrating Soviet Air Defence.

    There is of course a technological revolution going on out there in transport: the high speed train. In France, Germany, Spain the 100-500 mile city distance is seeing a halving of effective journey times. But that is missing the USA by which is why Americans don’t readily think of it. (it’s happening in China too).

  52. Gravatar of scott sumner scott sumner
    5. June 2010 at 10:03

    valuethinker, Don’t quote me on this, but:

    I once read that 40% of American lacked indoor plumbing as late as 1940.

    I believe that very, very few rural homes had electricity in 1890. I am more confident of that then the indoor plumbing statistic. I don’t thin even most urban homes did. Insull brought electrical power to large parts of Chicago in the 1920s.

    When you talk about places like London (or NYC) in 1890, that isn’t how most people lived–those cities were much richer and more sophisticated than average. People in rural America walked, or took horses, for most trips.

  53. Gravatar of valuethinker valuethinker
    6. June 2010 at 08:15


    OK on plumbing. It’s possible.

    On electricity, agreed. Domestic electricity was more the 1920s (it did arrive earlier in the US though).

    But there had been big revolutions in rural America. The iron plough, the regular train service to the nearbye town. The Sears catalogue was in existence, I believe, and that really was a revolution.

    But let’s loop back to our arguments about the speed of change (as I am losing track ;-)):

    – I think you are arguing (and many historians would agree with you) that the fastest pace of technological change in history was 1920s-30s. In which we saw the telephone, domestic electricity, widespread use of motorcars, the completion of the urban streetcar revolution, electric traction in factories, the birth of the aircraft, the radio (and even the beginnings of TV service in a few places around London).

    These were pretty much all the consequence of pre WW1 technological innovations, though, such as the internal combustion engine, the Model T etc. Takes a long time for a technology to get embodied in the capital of a society.

    – I would add to this that if you look at conquering time and distance, the 1850-1900 period saw an even faster transformation: the steamship, the transcontinental railway and the telegraph. The drop from months (transcontinental or intercontinental) to days or days to minutes (the telegraph) just has never been equalled again.

    Bubbling in the background is urban sanitation. Not universal by 1900, but most big cities had a sewer system and the beginnings of potable drinking water. However again that became universal 1900-1950.

    – postwar we have a few: antibiotics first and foremost, the computer revolution, the mobile phone. We now have the internet, but I would argue that, having been invented in the 1960s, that is only really beginning to have a significant impact now.

    Jet engines say on average halved travel times over intercontinental distances. Not as big a change as prop planes did over ships. Again the jet engine was invented in 1939, I cannot remember when the Comet airliner first flew (1955?).

    On power generation the step changes were again in the first half of the 20th century (including the electric grid, which is a big part of what made Minimum Efficient Scale electricity generation stations possible ie 500MW). Change since 1950 has been incremental (I suspect nuclear fission is a technological dead end– a transition technology from fossil fuel onto something else).

    The other big revolution in transportation was the multi-modal container ship. That really has transformed the postwar economy, and I think the first one of those was 1968?

    It’s not clear to me though that technological change has slowed down. I think, rather, with the diffusion of the internet and the mobile phone, that it has sped up– diffusion curves for new consumer innovations have much shorter time scales than they used to (the point where 90% of households had a DVD player, v a VHS player, for example). And we’ve only begun to glimpse the power of the biotechnological and materials revolutions.

    It’s not yet clear to me we are going to look at 1950-2000 or 1950-2020 say, and say that 1900-1950 was a period of greater change, nor that 1850-1900 was of slower. The world in 1900 really was a very different place, even for rural Americans, than the world of 1850– that business of a railway, and a telegraph, and an iron plough.

    That’s particularly true if we look at what is happening to the vast majority of humanity outside the developed world. Life has started to change unbelievably fast for a Kenyan or a Chinese, despite all the known problems.

    On consumption, (real consumption per capita), I’ll stick with my observation. Over the long run, we underestimate, potentially, growth in rcpc because we underestimate the impact of technological change on how we live our lives.

    The impact of a lot of technology (like the internet) is to fundamentally increase consumption per capita even if that doesn’t show in the short term in economic output. The internet lets us cut out the middleman, and that shows up as being able to buy more goods and services, not in more total spending in the economy.

    Because technology often comes into play in intermediate goods and services (Schlumberger’s introduction of 3D and 4D drilling only indirectly hits household consumption via lower gasoline prices/ more cars; WalMart’s distribution system genius comes through in higher standards of living (on the same income) for customers; ditto Dell’s just in time computer manufacture) the direct effect on rcpi is not easily observable (at least in the short run).

    I suspect we are mostly agreeing with each other. Just that you were arguing that technological innovation has slowed down? Given the 30-50 year lags between a technology being invented and it getting into near universal household use, I think that’s quite a hard thing to assert.

    On energy itself, the revolution is only just getting going, be it wind, solar, conservation technologies, fuel cells etc. But watch what has happened with shale gas, and how revolutionary that is. Changes in energy consumption grind ever slow.

  54. Gravatar of ssumner ssumner
    7. June 2010 at 08:38

    Valuethinker, Again, I just don’t see the improvement in my living standards since the 1950s being anything like what my grandma saw, but I suppose it is a highly subjective judgment.

    The 19th century developments like steam power transport and telegraphs were very important, but I think the next set of changes affected everyday life in a bigger way.

    Sorry I don’t have time to do justice to all your good points.

  55. Gravatar of valuethinker valuethinker
    8. June 2010 at 03:28


    Part of the issue may be that 1920-1973 change in living standards we saw (which was really about urbanization given that we agree places like London and NYC had those amenities in 1920 for at least middle class citizens) has now moved to the ‘third world’. Places like China and India are doing that now, and even in places like Zambia you are seeing some quite amazing changes in the material standard of living (phones, bicycles, cars etc.).

    1973 America v. now there have been some pretty big changes in standard of living: the average home has increased by 40% in size for example. Or the changes in healthcare technology.

    I generally don’t feel that the pace of technological innovation has slowed down. High speed rail is a classic example, but one a North American audience may just be blind to, as it is not happening in North America (nor much in Britain). A halving of 100-500 mile journey times. Add to that what has happened at WalMart or with Ryanair in terms of the cost of consumer goods and services, and I see some really quite fundamental changes in how we live, as measured on a consumption basis.

    Food in 1973 was 35% of the British household budget, and it’s 16% now. That’s a radical shift.

    The big slowdown 1973-mid 1980s I think a big part of it must have been the huge shift in factor cost ie energy prices. Measured productivity and per capita consumption stagnated because we were making huge shifts in how we use energy (particularly oil). We go from the 16mpg family sedan in 1978 (our Pontiac Lemans 8 cylinder) to a 32mpg Honda Accord in 1990, just as an example. Also demographic factors (and macro instability).

    Technological changes have a 30-40 year lead time before they become widespread in consumer lives (electric traction in factories) and so the changes now are often based on technologies invented in the 70s. Ditto the Internet.

    Thanks for taking the time to continue the conversation.


  56. Gravatar of ssumner ssumner
    8. June 2010 at 14:25

    Vt, Those are good points. Ironically a few months ago I had a post making similar arguments. I discussed many ways we are better off than 1973 to counteract the “wage stagnation” argument, which I think is overdone.

    One reason China and India are growing so fast is that they are in the easiest phase of growth–churning out ever more homes, appliances, cars, highways, rail etc. It gets trickier when you are already pretty developed.

    BTW, China’s new rail system will be mindboggling. Like NYC to Chicago (1300k) in 4 hours. Why even fly? (Sorry, don’t know the UK analogy.)

  57. Gravatar of azmyth azmyth
    14. June 2010 at 14:43

    Not only has the internet improved productivity, it has increase consumption by an order of magnitude in any good that can be reduced to 1s and 0s. Most people I know have downloaded hundreds of thousands of songs. If you buy the RIAA argument that they are worth $1 each, that’s incredible productivity from low skilled teenagers – they are producing hundreds of thousands of dollars worth of goods! Add in movies, games, etc and piracy is a serious industry – much bigger than most Fortune 500 companies. How much is it valued in official GDP? 0. There are millions of legal websites that people can enjoy for free as well. Google maps is far better than any physical map, email is better than any physical mail, etc. Once again, it adds nearly nothing to GDP, since only advertising counts. I easily value blogging at several hundred dollars a month, but since its free, I never have to reveal my preference to central planners. Internet goods are pure utility – no regrettables like locks on doors, the military industrial complex and conspicuous consumption arms races. If you don’t believe me, try living without the internet for a few days – it’s like trying not to breathe.

  58. Gravatar of ssumner ssumner
    15. June 2010 at 06:20

    azmyth, I still buy CDs, and don’t see any big advantage to downloading music. Sure you save a little money, but I just don’t see the internet as being a game changer like indoor plumbing, cars, airplanes, Tv, radio, stoves, refrigerators, etc.

    But it makes blogging possible. 🙂

  59. Gravatar of Are we coming to an end of economic grwoth? | Liberal Conspiracy Are we coming to an end of economic grwoth? | Liberal Conspiracy
    21. June 2010 at 03:16

    […] non-inflationary growth in the 50s and 60s – post-war rebuilding and the spread of some big technical advances – became […]

  60. Gravatar of Noget om udvikling og lidt miljø | Hedegaard Noget om udvikling og lidt miljø | Hedegaard
    23. August 2010 at 22:33

    […] Sumners indlæg om økonomisk udvikling indeholder en fantastisk satirisk […]

Leave a Reply