The Rise and Fall of American Growth

welcome uh to the London School of Economics all of you is that I'm uh V danan I'm a professor here at the LSC it is my privilege to introduce uh today's speaker Professor uh Bob Gordon uh before I do that a couple announcements So the plan is is that Professor Gordon will start by introducing his new book the rise and fall of American Grove so after that there will be the opportunity to ask questions and it's always hard to predict these things but we expect this great event to be over quarter to 8 uh 8:00 uh please turn off your mobile phone I ask it every time and still it happens um sometimes to me but um not this time uh for those of you who want to tweet the hashtag is Aly America um they're not very creative in coming up with these things um the event is being recorded as you can see and hopefully it will be made available uh on the LSC events page uh and after the lecture there's going to be the possibility to buy Bob's book and to get it side and the way that's going to work is you can buy it outside and then you come back to the stage to uh to get I'll just go out there I'll just go out there where the books are usually the stewards get really angry at me if it's not okay well whatever you want to do um that would have been a revolutionary move on okay on our part anyway so let me turn to the the more important bit of of my job so Professor Gordon is the Stanley G Harris professor at Northwestern University uh whereas many macroeconomists like myself focus on either business Cycles or growth is that Professor Gordon actually has uh covered both areas and uh the the other aspect that's impressive about is research although it's often you know on really topical questions is that you know solid knowledge of the past and the lessons uh we can learn from it you know are part of his research I not surprisingly has had you know numerous uh articles published in top economic journals couple books uh before yet this one including a very popular textbook and not surprisingly had many awards both you know from within the academic Community as without uh as outside the academic Community an example of the L is that he was named in 2013 as one of Bloomberg's top 10 most influential thinkers so please join me in welcoming this evening's [Applause] speaker delighted to be here and talk about the current economic growth slowdown from the perspective of 50 years and in particular from the perspective of the special Century which is my name for the 100 years between 1870 and 1970 when we talk today about slowing economic growth in the United States and certainly in Britain you have exactly the same uh situation um we're combining the effect both of productivity that of uh output per hour and slowing growth in potential uh the potential labor force we've had a decline in the growth rate of the working age population um in the United States and we've had a substantial decline in the labor force participation rate that has survived the uh recovery of the economy we've had our unemployment rate fall from 10 to 5% um but along the way the labor force participation rate has declined by enough to take about eight million people uh out of the labor force only half of whom are elderly retiring and the others are prime AG people and youth um who are no longer in the labor force hopefully some of them will come back as the economy continues to recover but when you put it all together in the 30 years from 18 1974 to 2004 the US economy grew at 3.12% per year and in the 11 years since 2004 the growth rates only been exactly half of that half of that that almost to the second decimal point of 1.56 now the uh famous phrase secular stagnation was coined by Alvin Hansen the president of the American economic Association in his presidential address of 1938 uh and he was most concerned about declining population growth and its effect on net investment if we didn't have the population growing uh so rapidly if population growth had fallen by 2/3 said it as it had appeared to him at the time then we wouldn't need so much Capital to equip the new workers since there wouldn't be so many new workers and here's the population growth History of the United States since 1875 we had a very rapid growth in population up to World War I fueled by immigration uh a Revival of immigration right after World War uh one and then Draconian anti-immigrant legislation in the 1920s and from Hansen's point of view shown by the first vertical line in 1938 um he was observing a catastrophic decline in population growth little did he know that very soon population growth would revive and here you have the famous American baby boom generation with the big hump in population growth that occurred in the 1950s and 1960s we're now back down to population growth that is not that different than faced by Alvin Hansen through a much more gradual decline and the US Census Bureau predicts uh that population growth will continue to decline down to a rate even lower than Hansen faced in 1938 but one thing that Hansen did not have to worry about he had to worry about slowing potential labor force growth and its effect on investment he did not have to worry about productivity growth now they didn't have very good figures on that in those days but in retrospect using the numbers in my book the growth in productivity in the last four years of the 1930s was almost 4% a year and the growth rate in the five years ending in 2014 was only 0.8 actually if we extend that out to the beginning of 2016 it's more like 0.5 so Alvin Hansen had a population problem and there was a big aggregate demand problem because unemployment was still 15% in 1938 1939 there was no productivity growth problem 1930s was one of the most uh fruitful uh periods for the growth of efficiency U and productivity now in order to understand what I'm about to show you which are the three big eras of slow then fast then slow productivity growth that gives its name to my book the rise and fall of American American growth it's good to keep in mind the chronology of what economic historians call the three industrial revolutions well everybody knows about the first Industrial Revolution that started in Britain with the steam engine their primary offshoots railroads and steamships cotton spinning and weaving and the transition from wood to steal those inventions of the 1770s through 1820s and 30s uh persisted in their effects throughout the 19th century U and in the United States for instance fully five six of the railroad mileage was built after uh 1870 the big Mighty Industrial Revolution was the Second Industrial Revolution and here I want to emphasize its enormous scope how many different aspects of human life were affected by the Second Industrial Revolution and here are the different pieces electricity which made possible light elevators portable and fixed um Machine Tools in manufacturing and later on air conditioning the internal combustion engine invented only 10 weeks after Edison's first successful electric light experiment which made possible motor vehicles and Air transport the end of isolation with the telephone the phonograph radio movies and television the role of running water uh an invention of the Romans but brought into every household During the period between 1870 and for urban households the transition to running water waste disposal and indoor bathrooms was complete uh by 1940 uh we didn't even need the full special Century to see that and after the war uh the remaining uh diffusion of running water came to uh Rural America we had a whole set of inventions uh many of them taking place in Germany in the realm of chemicals and Plastics and then antibiotics in the roots of modern medicine which were uh well underway uh by 1970 and then finally we had an utter change in working conditions and think of working conditions for the male bread winner 50% of employment in 1870 was on the farm farmers were subject to the vicissitudes of the weather to plagues of insect to insecurity and Housewives who did the work at the home had to carry the water into the house there was no running water coming to the pipes they did the laundry on a scrub board in one of the anecdotes in my book a survey uh carried out in North Carolina uh indicated that in 1885 the average North Carolina housewife walked 148 mil a year carrying 35 tons of water and all that water brought into the house had to be brought out of the house uh think of all the different kinds of advances in this broad spectrum of advance that were uh one-time only inventions you could only have this happen once the transition from kerosene lamps with the pollution with the dirt and staining on the lamp Globe to the instant onoff switch of electric light the transition from Factory power with steam power generating uh propulsion in manufacturing through leather and rubber belts trans translating uh later into electricity with individual machines at every workstation um we made a transition from temperature alternating between cold and hot through central heating and air conditioning we became uh through the uh early and middle part of the 20th century into a world of uniform uh temperatures initially in business offices and then uh throughout the home we went from horses to motor travel and air travel and we went from a mainly rural Society uh as I mentioned 50% of employment on the farm in 1870 to a mainly Urban Society after 1950 and a lot of the Improvement in productivity that you're going to see uh came about through the transition of people moving from the farm uh to the City what about the third Industrial Revolution some people call that the EIC Revolution standing for entertainment information and communication uh technology well with entertainment we've had the evolution of television from black and white to color to cable to time shifting to streaming Information Technology we've gone from mainframes to personal computers much of the real productivity impact uh of the digital age came from the arrival of personal computers with spreadsheet and word processing software uh making unnecessary repetitive retyping and linotype operators uh resetting type for Publications all that was pretty much gone before the arrival of the web in the mid 1990s we've gone in communication from landline to uh cell phones and mobile phones and smartphones and we've had uh a discret series of productivity enhancers the ATM machine that almost made bank tellers obsolete the barcode scanning at the retail store credit cards with instant credit card authorization so here's the background we had a first Industrial Revolution we had the second the big Industrial Revolution with its huge wide scope of change and then the third Industrial Revolution since about 1960 um and if you look at the rate of Labor productivity growth in the three eras the first era is 1890 to 1920 the middle era the real big one is 1920 to 1970 it took a long time for the inventions of the Second Industrial Revolution to have their main impact on productivity and then the period since uh 1970 here is labor productivity growth we all know about growth Accounting in which we decompose labor productivity growth by stripping out the impact of Education stripping out the impact of capital deepening the rising uh ratio of capital to labor and when we do that subtraction we're left with the black areas which are called total Factor productivity the average productivity of Labor and capital put together and as you see we're taking out subtracting about the same percent growth from education and capital deepening in all three periods which means that the superiority of that middle period for total Factor productivity is even more pronounced since 1970 we've only managed to achieve about one-third the growth rate of total Factor productivity than we did in the middle five Decades of the 20th century or putting it another way in all the period since 1890 fully 2third of the total Factor productivity growth was achieved in the middle five Decades of the 20th century now let's bring this up to date here's total Factor productivity a 5-year moving average going from 1952 right up to the uh most recent quarter the end of uh 2015 and I'd delineate four eras of productivity growth first we have the end of that big burst of productivity that characterized the middle of the 20th century if you look at the vertical axis you'll see that we managed more or less 2% uh tfp growth uh for the 1950s and 19 60s and then it tailed off radically during the 1970s uh revived a little bit um but then experienced a major Revival in uh the late 1990s and early 2000s people call this the dotcom Revolution but in my view the um Revival of productivity growth in the late 1990s brought together all of those contributions of personal computers and the software that was written for them the transition to electronic cataloges and much else um and then what we have is the pathetic performance of total Factor productivity growth over the last six or seven years and remember each of these is a fiveyear moving average so the fiveyear moving average has really stayed down around half a percent per year uh for most of the last seven or eight years now the US economy has recovered from the Great Financial collapse of 2008 and 2009 our unemployment rate is now back down to 5% so we can't blame this shortfall of productivity on a weak economy so in my view the third Industrial Revolution failed the total Factor productivity test on average since 1970 during the years of the Mainframe and the personal computer coming on board we had only onethird the pro the tfp growth that we did before 1970 and then when the productivity impact of the digital age really arrived in the late 1990s we had only uh a 10-year period of Revival unlike that 50-year period of the housian days of the middle of the 20th century now I challenge audiences by asking both on the basis of that temporary Revival of productivity growth um and also some introspection which I'm going to guide you through whether the major impact of the third industrial revolution has already happened and in fact was pretty much over uh by 10 years ago it's a startling idea it's very much in conflict with the Techno Optimus that we hear about uh today so let's remind ourselves of the contributions of the third Industrial Revolution in stages and let's think first of the the all the advances that were made before the web arrived before the web browsers first uh came in around 1995 we started off in 1970 with mechanical calculators where to do a multiplication you go went clunk clunk clunk waiting for the result of a single multiplication to come out we had repetitive T retyping we had lines and lines of secretaries whose only job was to retype law briefs in legal offices or retype uh bank statements in um Bank offices uh we had filing cabinets to store all the paper then we had a transition in the 1970s toward memory typewriters electronic calculators mini computers then the 1980s the real productivity driver were the personal computers with their word processing and spreadsheet software that made obsolete completely the slide rule the calculator and the typewriter and then finally by the late 1980s we were already into the world of email well before the web arrived we were doing web email in universities on uh T so-called T1 lines with proprietary uh software uh and uh many of the productivity enhancements of the digital age were already um upon us then we had the web browser search engines and e-commerce that came in in the late 1990s followed by flat screens and another advance of the early part of the 2000s between 2001 and 2005 was the main period when we saw the rolling out of automatic check-in kiosks at airports which were like an ATM machine uh in terms of the technology that they used but by 2005 the revolution in business practices was almost over and we look as I go around and look at people's offices and see them depicted in movies and other uh ways I go around and uh visit newspapers and they interview interview me about some of these growth issues and I see newsrooms that look very much like they did 20 years ago with the exception that we now have flat screens instead of big bulbous long uh cathode ray monitors um newspaper publishing since 1994 has hardly changed all the main uh uh enhancements had taken place by then and the whole layer of secretaries and typ Setters uh had been removed by 20 years ago U I like to think a lot about cataloges because this is another example that came in before the web we had libraries that had wooden boxes with card catalogs that originally were manually typed later on they were printed but the catalog cards didn't tell you whether the book was on the shelves there was no inventory management so then we got in the late 1980s early 1990s no matter where you were if you were uh around at the time in any kind of University or Public Library setting we had that transition to electronic cataloges which was not dependent on the web which took place before 1995 and all of a sudden we could tell whether the books were on the shelves or not um we had a revolution in retailing in the United States we call that the big box Revolution associated with the names of stores like Walmart Home Depot Target very large uh rectangular stores often out in suburbs near Expressway interchanges um and uh coming along with them was a revolution in inventory and Supply Chain management I mentioned before the barcode scanners and the revolution in uh checking out um and is isn't it amazing how minimal is the impact so far of self checkout I don't know if you think about your average day of going from one retail store to another how many times you interact and check out just with a machine instead of having an interaction with the human uh but I'll bet it's a fairly small percentage it certainly is for me finance and banking in addition to the Advent of the ATM machine we had the transition in London as well as in New York and Chicago from transactions that were measured in millions of shares per day to transactions that were made measured in billions of shares per day and that transition took place between the 1970s and the 1990s um and think about how long it's been since some of the major players of the digital age were founded Amazon in 1994 Google in 1998 Wikipedia and iTunes in 2001 and so a lot of things that we take for granted now have been around for a long time the big Innovations of the last decade have been social networks and smartphones and if there is time I'll get back to them in the issue of how much we're uh Missing the measurement of the benefits of these new Innovations of the last decade everywhere I look I see things standing still I see offices with the same software and Hardware as they had 10 years ago I see shelves and retail stores shop stocked by humans I see meat and cheese being sliced behind counters by humans I see humans checking out uh groceries and other uh market items with barcode scanners in medicine I see an electronic recordkeeping Revolution but I see doctors and nurses doing pretty much the the same thing as they did 10 years ago with one exception now when you go to see the doctor instead of looking at you the doctor looks at the screen of his electronic medical records and higher education uh you all know about tuition inflation in American Education I don't know enough about lsse to to uh uh to know how much inflation and tuition has occurred at at lsse but this is not representing an improvement in the quality of instruction it's a result of increased administrative overhead a lot of fancy new buildings uh and a uh General uh inflationary cost pressure that has not made the college students learn any better or any faster so as I look at that same chart this is identical to what I showed you before I see this stasis uh of retailing office work medicine and higher education coming out and those are the big service industries that dominate the value added in our economy uh I see those helping to provide the background to explain those rather mysterious looking numbers uh that we see now we could always ask well aren't we on the verge of a new technological acceleration uh isn't the next technological Revolution right around the corner and to answer that question I take a look at some of the other dimensions of what was special about the 1990s that does not seem to be recurring now we had a whole bunch of things happening together in the 1990s and uh things are not looking from today's perspective as if the 1990s are going to be repeated now one lament that's getting more and more attention in American economics is the so-called decline in business dynamism and business dynamism is often represented by the share of new firms as a percentage of all existing firms so the red line shows you the percentage of all business firms that are newly established the Blue Line shows you the percentage of business firms that are exiting that are going out of business and you see the lines are crossing in recent years and we've had not just we there was no great Revival in the 1990s for this series unlike the others that you're going to see in a minute we've had a steady decline in the importance of newly fored business firms this is net investment as a share of the Capital stock the black horizontal line is the post-war average as you can see before 1987 almost every year was above average since 1987 almost every year is below average except for a succession of years in the late 1990s so the late 1990s were unusual in this Dimension that we had a big burst bur of innovation and that burst of of sorry of investment and that investment was heavily skewed to ICT information and communication technology investment the share of value added going to computer investment in the US economy was almost double in the late 1990s what it's been in the last three or four years so special in the 1990s business investment even more special was the growth in manufacturing capacity the red line is a annualized one-year change in the capacity to produce of American manufacturing and after bubbling along at around 2 to 3% in the 1980s and early 1990s we had this takeoff in American productive capacity that peaked around 2000 just about the same time as the investment boom peaked and then tailed off and has been theand ing between 0 and 2% in the last few years you may have heard a lot particularly if you read The Economist about the demise of Moore's Law the prediction made by Gordon Moore in 1965 that every two years the number of transistors on a computer chip would double the uh top graph shows the rate of change of the prices of computer equipment which as you see reached a maximum rate of decline that is things getting better faster in the late 1990s and the bottom line shows the behavior of Moore's Law the dash black line is the prediction of a doubling of transistors per computer chip every two years you see the blue line which is the actual outcome went along with the Gordon Moore prediction more or less until 1995 then things accelerated it took between 1995 and 2005 it took less than two years for the number of transistors on a computer to chip to double and then the whole thing falls apart after 2005 and we don't any longer have the increase in the clock speed the density of computer chips the way we had for so long uh and this is something that has received widespread attention U very recently now I'm not saying Innovation is over I'm just saying Innovation is slowing down and even though there's frenetic activity in Silicone Valley even though we see that uh enormous sums are being uh capitalized on the stock market eight of the top 10 stock market capitalization firms in the world are American and the number of those are specifically uh products of the internet age like Microsoft Facebook and Google um but uh the issue is what is the technological Revolution achieving and we've got got some problems this is not to say we're not observing uh Innovation and Technical change it's that its impact is smaller and we have to start out with Medical Care the Techno Optimist point to uh decoding The genome to DNA to the uh effects of genetic research and making possible uh new drugs that will cure all sorts of diseases actually the uh rate of discovery of new drugs is slowing down the cost of discovering new drugs is rising rapidly uh the new drugs are highly expensive and often targeted to very specific forms of cancer and meanwhile in in the United States we've got problems life expectancy actually fell in 2015 compared to 2014 uh life expectancy is three or four years lower in the US than it is in Canada Britain Europe and Japan we have stunning new evidence popularized by Nobel Prize winner Angus Deon just last fall showing that mortality rates for American white males and females in their 40s and 50s are going up that is mortality getting worse not getting better we have enormous Gap in life expectancy between the top 10% and the bottom 10% if you take the life expectancy of American males white males uh age 50 if you're in the top 10% of the income distribution your life expectancy as an American male is 87 years if you're in the bottom 10% of the income distribution your life expectancy is 73 years that's a 14-year Gap and it raises questions about the frequently suggested solution to our long run fiscal problems of raising the retirement age because raising the retirement age for the bottom income people Cuts them out of the benefits of pension plans that they otherwise would have achieved and the Improvement of Medical Care in physical well-being is colliding with the intractability of Alzheimer's and Dementia uh and so we're having a growing burden to Society of taking care of people with mental illness even as their physical longevity improves um we've got robots of course robots are a delightful topic everybody likes to talk about them industrial robots were first introduced by uh General Motors in 1961 and uh 20 years ago in touring uh Auto Plant in uh Ohio I found that robots were already 20 years ago had taken over the welding together of the body parts had taken over the paint shop greatly improving the air quality for the workers uh that were uh working there uh so why haven't robots taken over why do I go through my daily life and I never see a robot I love to play this game find the robot I don't see them in the supermarket I don't see them in the doctor's office um I don't see them in hotel room service there's still a human bringing me whatever I might order uh so here's a uh indication of where we stand with robots a story from The New York Times uh last October an interesting reality check took place in June when two dozen teams of leading robotics engineers gathered in Pomona California for a competition sponsored by the pentagon's research agency the robots had to do simple tasks walk up steps turn a valve operate a power drill the chores took humans five or 10 minutes at most the winning robot took 45 minutes most struggled bad ly falling down steps and taking long pauses to figure things out even with remote controlled assistance turning a knob to open a door proved daunting to many one young man in the audience observed if you're worried about the Terminator just keep your door closed so that is some indication what I think about uh robots I think actually uh people also bring up a 3D printing uh 3D printing is uh rapidly expanding in use for prototypes and models it's not uh mass production we're not producing 17 a. half million uh cars in the United States with um 3D printing it's the oldfashioned assembly line aided by those robots that as I mentioned have been there for the last 20 years I think the real progress is being made in artificial intellig um we're seeing increasing use of masses of Big Data qualification much of it's being used for marketing much of it's been used being used by firms studying how to switch customers from their Rivals over to themselves and that does not add to society's well-being that's a zero sum game but we do have evolutionary change that's associated with artificial intelligence we've had a uh great advance in the ability of computers to do legal searches and that's created unemployment for some types of young lawyers we have computers helping to read Radiology reports voice recognition is amazingly uh developed my doctor dictates after each patient visit and is Amazed by the accuracy of the printouts achieved by electronic voice recognition including the most esoteric medical terms and we have language translation which is gradually making language translators uh obsolete this is all evolutionary it's not happening just between today and tomorrow it's a process underway for 10 or 15 years that's going to continue to be slowly evolving over the next 10 or 15 years then we have driverless trucks and cars I'll get to driverless cars in a minute I have an important point to make about driverless trucks you think that would be a big enhancer of productivity if the trucks were producing the same output of Transportation but there was no driver so the denominator of productivity uh was Zero Labor uh you think that would be a really important Improvement that would be true if it weren't for the fact that most truck drivers are in dual roles they drive to the destination and then whether they're a federal express or a UPS uh driver or uh delivering bread beer or Coca-Cola to the supermarket mark their job includes getting out of the truck unloading it Wheeling the stuff in and often arranging it by hand on the shelves in terms of the Practical use of driverless cars uh the challenges are daunting I only like to predict technology out for 20 or 25 years I have no doubt that some of these obstacles will be overcome if we wait long enough remember we've got a huge Fleet in the United States of more than 200 million cars that require drivers that are going to have to be integrated with the driverless cars as they emerge um but here's an evaluation of where we stand from the current May 2016 issue of consumer reports which is a long-established US Consumer Testing organization experts agree it'll be a very long time before autonomous vehicles will have free reain over every road in America before that happens every road Highway byway bridge and obstacle needs to be mapped and we're not just talking about AR Maps uh currently on your car's navigation system according to google autonomous cars need detailed 3D maps that capture all features of the road including Lane markers and traffic signs it goes on to talk about how weather rain and snow obscuring Lane markers are a serious obstacle for driverless cars and a self-driving car software has to be ready for even the most bizarre circumstances and be ready to temporarily violate traffic laws say if a police officer or traffic worker waves cars into an oncoming traffic lane uh to avoid an officer an obstacle just imagine the billions of lines of software code that will be needed to make cars truly driverless and free humans from any need to worry about uh unnatural uh unplanned events well so far I've just talked about Innovation Innovation isn't stopping it's slowing down in its impact but in growth for the United States we Face an additional problem that I call the headwinds I'm going to list four headwinds they're really five because I'm going to go on and talk a little bit about some of the social consequences uh that are dimming the future for the Next Generation uh but let me first of all talk about the forehead winds as they're uh enumera in my book The slowing contribution of Education to economic growth the demographic headwind the Aging of the population Rising inequality which everybody knows about which an enormous amount has been written about and the fiscal headwind now what these things do is in the case of Education reduce productivity growth below what it would be if we just paid attention to the pace of innovation demography as we'll see reduces the growth in income per capita relative to productivity inequality reduces median income growth compared to average income growth well in the United States a major contribution to that enormous rise in productivity in the middle of the 20th century was the advance of educational attainment from eight years of educational attainment in 1900 to virtually everyone going to high school by 1970 since then we've had uh College completion gradually increase but the Dropout rates are enormous uh only about 50 or 60% of American four-year college students uh complete their college degrees after uh six years and uh an even greater problem is that of college graduates fully 40% are unable to find jobs requiring a college education after they graduate we have uniquely in the United States more than anywhere else the High Cost of college education and growing indebtedness that is changing the lifestyle of people in their 1920s in their in their 20-year-old bracket having to delay household formation delay having children and contributing to the slowing of population growth if we look at American Education it doesn't look good on an international basis the international Pisa test given to 15 year olds by the oecd if we look at 34 oecd countries the US ranks 17th in Reading 20th in science and 27th in math in college completion the US has fallen way down in the international league tables and uh the experts who parse out the impact of education on productivity growth think that in the next decade compared to the last 20 years or so there's going to be fully a minus 0.3 percentage Point drop in in the growth of Labor productivity coming as a result of this set of unsatisfactory outcomes of the American Education process the demographic headwind is common all over the developed world and in China especially as well Japan South Korea Europe uh and that's the Aging of the population the consequence of the declining uh fertility rate and in the United States we have not only this enormous bulge of the baby boom generation which is in the process right now of retiring uh but we also have had over the last seven or eight years a decline in labor force participation of prime age men prime age women and youth there's an increasing share of Youth who are neither in school nor in um in jobs and then we have inequality um over the last 20 years the difference between the average growth rate of the total income and the growth rate of the bottom 99% is fully half a point per year so if I told you that future growth is going to be 1% a year per capita after we look at the bottom 99% it's only going to be half a percent per year that's an enormous difference and you can ask as I fully expect you to later on are we sure that this trend toward Rising inequality is going to continue well I think it is continuing let us count the ways the Silicone Valley billionaires the new unicorn companies formed with more than a billion of market capitalization the continuing push back in at least the American setting of firms uh pushing employees into part-time work cutting benefits cutting pensions switching pensions from defined benefit uh to defined contribution uh pushing employ into part-time work instead of full-time work still even with the decline in unemployment to 5% which is more or less historically normal we still have six million people in the United States who are working part-time and tell the surveyors that they would prefer to have full-time work and then there's the future of the debt to GDP ratio you see how much our debt to GDP ratio in the United States increased as a result of the financial crisis um this graph projects out from a couple of years ago what's going to happen to the debt GDP ratio as we go out into the future and uh an updated version of this graph which I didn't have time to get into this set of slides shows just the same thing that we're going uh right up toward 100% US debt to GDP ratio up there in competition with Italy or Japan if nothing is done to turn around the impact of the Aging population on fiscal uh outcomes now what about these sociological changes that I mentioned might be considered a fifth headwin well uh a lot of it has to do with the decline in marriage and the decline in marriage has occurred in many uh developed countries uh but uh in the United States I can just give you a few of the statistics for white high school graduates the percentage of children born out of wedlock increased over the 30 or so years since 1982 from 4% to 34% for black high school graduates the percent born out of wedlock from 48 to 74% uh in a very controversial book which is credible in its statistical presentations uh Charles Murray who was formerly considered to be a racist uh because of some of his comments on genetic inferiority of the black population he wrote a book in 2012 called coming apart and it was just about the white population so he didn't have to be accused in that book of doing anything to do with racial differences and Murray divides up the white population into one group to study that's the top 20% and the other bottom the other uh portion that he analyzes is the bottom one-third um and what for my mind is his most startling statistic is that in this bottom one-third of the white population if you take 40 30y old women and look at their biological children the percentage that are growing up with two parent with their two biological parents declined over the 50-year period since 1960 from 95 to 34% that's an enormous change and it predicts for the future uh more children growing up in poverty because single mothers are much more likely to be in poverty than married couples a greater likelihood of dropping out of high school G proclivity to uh become engaged in crime and a lower percentage likely to complete College there are additional uh signs of social deterioration that affect the outcome of social Mobility what is the chances of low-income people moving up into the middle and the higher tax brackets in the United States if you're a married couple who both gone to college your chances that your children are going to go to college are 90% plus if you're a single black mother raising children the chances that your children will complete a four-year College are 10% or less here's an amazing statistic 40 46% of 20 to 24 year old black males in Chicago are neither at work nor in school now I realize being in Europe that that might sound not sound quite so shocking as it does in the United States because we know that youth unemployment in Spain is 50% but this is the equivalent figure for the United States Chicago is an unusually bad situation the equivalent figures in New York and Los Angeles are 32% um so the percentage of a white high school dropouts with prison records over the last 30 years has gone up from 4 to 28% more than 2third of blacks who have not completed High School have prison records and the in and the associated difficulty in finding employment after they come out of college so uh once again what are the significance of these headwinds the education headwind reduces future productivity growth the demographic headwind uh with the Aging of the population reduces the number of hours per person and thus reduces the growth and income per capita relative to Output per hour the inequality headwin if it continues will reduce growth in median income compared to average income and the fiscal headwind means that at some point in the future there'll have to be a downward adjustment in old age benefits or an increase in taxes which will further reduce the growth of disposable income the way this is Quantified at the end of the book compares the gray bars with which are the actual outcomes for uh the near Century from 1920 to 2014 with the prediction that's made for the next 25 years shown in black and the the productivity story uh shows productivity uh dropping from about 2.2 to 1.2 you can't quite read that on the on the screen um and I might add that one .2% is about twice as fast as our productivity has been growing in the US over the last six years so this is actually an optimistic outlook for productivity growth and forecast that we're on the verge of having a Revival compared to the actual recent outcomes output per person grew at more than 2% a year over the past Century uh because of declining work hours per person in the retirement uh issue uh we're down from 1.2 to 0.8 for output per person if inequality continues to grow we're down to 0.4 for median output per person and a little bit of an adjustment for a fiscal retrenchment would bring our median disposable income per person down to um 0.3% so one way to characterize this outcome is that uh growing at 2% a year for generation after generation in American history each generation could expect to have roughly double the standard of living in terms of measured GDP that the previous generation had the current generation of young people is going to have a increase in its standard of living of maybe not double maybe 20% higher over the span of the 30 years that separate the generations now that doesn't mean that we're not going to have inventions that like always throughout history have not been included in GDP running water wasn't included in GDP the invention of Television wasn't included in GDP the benefits of social networks are not included in GDP um so it's not true that the future generation isn't going to be better off at all the benefits of future in Innovation that we can't measure are going to be there for future Generations just as they were for the past but this is the story in terms of measured uh Innovation so I hope I brighten up today uh and just to summarize some of the conclusions as I mentioned before 70% of measured us total Factor productivity growth since 1890 took place in the five central Decades of the 20th century the big impacts on total Factor productivity of the third industrial or digital Revolution were largely completed by 2005 Innovation continues but it's less important it's just less sweeping across the whole span of human existence than it was back in the days of running water and the uh Improvement in uh infant mortality and life expectancy much of the future slowdown has nothing to do with Innovation but due to the headwinds the headwinds affect each country differently some European countries have much more of a demographic problem some European countries have much less of an inequality problem um and a moderate pace of innovation means that I can be optimistic as my very last remark the more skeptical I am that we're on the verge of a technological Revolution that's going to destroy all the jobs the more optimistic I can be about the future of employment thank you all right so now there's the uh the opportunity to ask questions there's only two rules first wait until one of the stewards hands you the microphone and the other rule is try to be concise don't embarrass yourself by giving a little speech okay we start out here gentleman well dressed with a tie hi thank you very much for their for their talk um I wondered if you could set out some policy prescriptions that would be able to combat some of these headwinds that you've been talking about well that was a very concise question um and of course has a very long answer uh what are your policy prescriptions um I'll tell you what my policy prescription is not and this is in the American context I don't think the government should get into the uh role of subsidizing innovation or steering Innovation from one area to another uh I think the American innovation machine is doing very well it's the Envy of many other countries in the world we have a uh amazing Venture Capital industry that stands ready to finance um any kind of uh Innovation that has half a promise of success uh so I don't think that there's a role for policy on Innovation itself where I would go is on education I think we need two different forms of re reform um actually three if you start from the top about the escalation of college debt I think we need to move toward a British or Australian system of income contingent repayment of college debt uh so that you don't have to repay debt if you're unemployed or if you're engaged in a lowp paying Social Service activity a uniquely American problem is we have Public Schools elementary and secondary schools financed by local property taxes which gives advantages to Rich suburbs over uh inner city ghettos uh and we need to have um the financing of Elementary and secondary education movement much higher up in the governmental structure and then what really concerns me is this enormous vocabulary Gap at age five between the poverty children and the middle inome children um as they arrive in kindergarten and this then determines the whole rest of Education with very poor test scores for children growing up in poverty I think we need a much more concerted effort um some European countries have gone a lot further than the us but I think we need to go further further still in using government resources such as they are for preschool education and preschool mentoring and tutoring for children growing up in poverty families the demographic headwind um there are two ways to go to cope with the Aging of the population one is to raise the retirement age and the problem with that is we have this enormous difference in life expectancy between the top bracket and low bracket uh income earners and so if we rais the retirement age for qualification for public pensions uh that cuts out the people at the bottom who have low life expectancy um a controversial but nevertheless I think uh better direction to go uh in confronting the Aging of the population is something that I think the United States is uniquely capable of doing so is Britain to almost the same extent and that is immigration um I think we need more high-skilled immigration we need more of a system like Canada has of uh awarding points for education and work experience um Canada now admits three times the share of population as new immigrants as the United States does and I think we're missing a big opportunity not to take advantage of people from all over the world who want to come and make the United States population younger um and in equality I don't think the government can tell people not to be successful but there are lots of uh levers that the government has to tax incomes of people who have already achieved success higher marginal tax brackets on income ending the preferential treatment of capital gains and dividend income getting rid of tax loopholes and uh what Martin feldstein a republican calls tax expenditures and uh using some of that uh saving and also uh there's Revenue to be gained from a carbon tax and all of that extra government revenue could be available not only to uh fund preschool education but also for a more generous earn income tax credit which is a kind of subsidy for work for lowincome people so those are just a few of the range of public policy interventions that I would uh push um hi one argument that um you haven't mentioned but I've seen pop popping up recently is the idea that there are essentially greater amount of Monopoly rent and a greater concentration um perhaps due to patents and things like this do you think this has any Merit in the context of the United States uh the question was uh about Monopoly rents um I'd like to link that to something that's related to it um we've had a very substantial rise in the share of profits in the US economy uh there's a substantial body of thought which think that the uh the balance of power has shifted from workers to corporations uh and uh this is one of the reasons why I think we need to reduce the corporate tax uh in the United States to keep companies from fleeing abroad uh we can't uh control the fact that the corporate tax rate in Ireland is less than a third of what it is in the United States so I think that the uh way to attack the Monopoly power issue is not through the corporate income tax it's through the individual income tax as I mentioned before one counter part of the low investment that we've got as a share of capital that I showed you in one of the charts uh is that corporations are spending trillions in buying back their outstanding shares um and we have corporate BuyBacks that are uh uh cutting into uh fixed investment and I'd like to see changes in the tax system which minimize the gains uh that'll uh partly be achieved by raising the tax rate on capital gains and dividends so here's one on the aisle right here sir you mentioned that the Piza scores of the American students is consistently less this has been happening for the past few decades then how come the American innovation machine continues to outpace the rest of the world that's exactly the opposite of the kind of uh question I normally get uh you're asking why American innovation is so successful uh in comparison with rest of the world I think um a lot of it has to do with uh the uh entrepreneurial tradition of the United States I mean why was Mark Zuckerberg a Dropout from Harvard instead of a Dropout from lsse I mean I'm going to ask you that uh as a way of turning that question around and why were apple and Microsoft uh founded uh in the United States um I'm not sure that I have the answer to that question other than the fact that there is traditionally an entrepreneurial culture and we have something new that's developed in the last 25 years which is this Venture Capital industry standing ready to finance the uh new ideas in a way that's not done by traditional banking in certainly in Continental Europe I think as in most things the UK is somewhere in between Continental Europe and the us but it's a very good question and my uh my view is that U the balance of innovation between the United States and the rest of the world is gradually shifting away from the United States as one concrete example we know that almost all of the technological innovation in solar power is being achieved in China uh and the China chines have succeeded in driving down quite rapidly the cost of solar power generation and we're going to see I think more Innovation taking place in India and China uh and less in the United States just as their level of development uh continues to rise wait in the back on oh back there yeah right hi um could you say a little bit more about the role of the financial sector in driving uh Innovation and productivity or otherwise it seemed to be a little bit marginal in your analysis and I wondered why perhaps I've misunderstood I tend to think of the financial sector is a bunch of parasites we you know there's a real dilemma in our national accounts you see these enormously High wages and high income that are earned by people in the financial sector uh and it's an arbitrary choice that the national accountants have to make between how much of this to it should be regarded as a pure Monopoly rent and how much of it should be regarded as a payoff for something substantial uh and I tend to be as you can tell from my introductory comment a skeptic on the uh necessity of having such a large share of our resources in the fin fincial sector I know it's even greater in Britain than it is in the United States um but I'm uh not a big fan of uh the financial sector and I'm a particularly adamant opponent of personal financial advisors they are the real parasites they're the people who take people's money invest it in high commission stocks that uh drain away the financial resources of innocent people through commissions and fees when these innocent people if they weren't so gullible should be investing in low-load mutual funds that don't have any expenses at all so don't get me started on finance uh yes right here in the aisle uh Professor can you elaborate more on the measure measurement errors on GD P because isn't it the case that the type of innovations that we are having with the digitalization makes the problem of measurement even worse than it was with other Innovations in the past and can it bu bias the create a bias on on on the results um let me uh take you back to the historical record here uh two very good research papers on exactly this topic of measurement error and to what extent this productivity growth slowdown is due to measurement error have come out just within the last two months in the United States and the conclusion is for measured GDP the error is that we understated the extent of this Big Revival uh in the late 1990s and early 2000s we had a much higher shap of computer investment than we've had since then and the computers are where the measurement errors took place the Improvement in the capability of computers is in reality much greater than it was in the national accounts so this measure of total Factor productivity based on the national accounts understates the height of that big boom in productivity in the late 1990s and early 2000s since the middle of the 2000s since 2005 two different things have happened to reduce the importance of measurement error having to do with computer investment one is that the share of value added in computer investment has fallen by half the share of computer in investment in GDP and the other thing is really remarkable and that is in the last 10 years we've had almost a complete shift from domestic to imported produ uction of computer equipment uh and so the part of the economy where the most errors are happening in the modern computer uh ICT equipment is just not part of the US domestic economy anymore we're not producing the stuff anymore that has the uh measurement errors in fact it goes the opposite direction if you think of how we measure total Factor productivity we take labor productivity which Now does not have domestically produced computers in it because they're all all imported we subtract the capital labor ratio which does have the imported computers in it and they're growing fast so as a result our true tfp is growing even slower than these official statistics would suggest so that all has to do with measured uh GDP and total Factor productivity then there's the question of unmeasured consumer surplus unmeasured benefits that are produced by smart phones as compared to the previous generation of unmeasured computer benefits uh created by uh the invention of electronic cataloges the invention of free information through search engines the uh ability to buy your own airline tickets instead of having to deal with travel agents or Airline phone Representatives uh we have a steady history going all the way back to running water and infant mortality of benefits accruing to the consumer that are not included in GDP it's a relatively subjective and difficult exercise to try to figure out whether the benefits to Consumers of social networks and YouTube and the other fruits of the invention of the smartphone are more or less important than the previous episodes going back through the decades of things that we've omitted from GDP one of the things that makes the smartphone so diff difficult to interpret is that there's so much multitasking going on people are looking at their smartphones while they're listening to my lecture people are looking at their smartphones while they're watching TV so it used to be that we would omit from GDP the benefits of watching TV then we would omit from GDP the benefits of time shifting through VCRs and digital recording then we would omit the uh benefits of the increased choice in television made possible by cable and satellite now we're omitting the benefits of having uh the access to your social networks which you're uh catching up on while the TV is on uh in the corner of the room uh so I think we have to recognize these things are happening but I think we're a long way from being able to quantify them uh back here um yeah thanks for just think about the four headwinds you're talking about um and if you think there's any countries or regions of the world that are well placed for the headwinds so by the time Industrial Revolution 4 comes around do you think maybe sort of leaving me leading the world stage say in 20 years could you repeat the question because I what I got was different countries are are uh facing different kinds of headwinds there particular country that thinks if there's a particular country maybe the us maybe the UK maybe another that in 10 to 20 years from now uh will weather the headwinds better than others oh okay well that's um I've already identified some particular weaknesses in American Education that make our educational headwind uh I think more ominous and more threatening than in the UK from the to the extent that I gather from reading The Economist about the UK educational system I think the UK system is doing a much better job of allocating resources uh equally across income groups in the population um UK students do better on International tests than Americans um we don't have you don't have as much of a problem of student debt um and uh you have much more centralized finan ing so just to take that example of uh the education headwind I think it's less important in um the UK than it is in the US the demographic headwind I think both the US and the UK have uh equally important opportunities to dial the fraction of old people in the population by changing immigration policy I think immigration has benefited the UK a lot in the last 10 years with with the open immigration coming from the EU I know that's a big political issue at the moment but I'm all for more immigration in the UK just as I am in the US because primarily it's a way of keeping your population young and avoiding that fiscal uh headwind I think the problems of inequality in uh the US are more intractable than they are in the UK given there's a lot of wealth um in the UK but I think the um problem of poverty particularly the inner city U black ghetto issues uh are more severe in the United States qualitatively and quantitatively than they are in the UK you don't have the same history of racial discrimination of residential discrimination uh that characterizes the United States many of the much of the black population in the UK came from the West Indies voluntarily as opposed to uh coming to the United States uh initially as slaves uh so I think the sociological aspects of inequality in Britain are less daunting than they are in the US over here on here let's uh let's go for him and then in the back sir I'd like to uh ask a question about potential GDP versus actual GDP so if you take two countries say Japan where the population is so old is so aged the potential GDP say of their country could be minus 2% you know and they're actually growing at say 0% so it's really not bad and now suppose you take my country India we have a very young population and our potential GDP should be 12% and we grow at 7% so it's actually that we are not doing that well but the headline figure of 7% makes it sound exciting well you're just saying that potential GDP is not should not be our measure of the growth we care about it should be GDP per capita uh and uh if the Japanese population is shrinking so that potential GDP is falling uh we should deduct out the component of that that represents the uh decline in the population uh both because of fertility and also because of the movement of people from work into retirement uh so we all know that Japan almost above all other countries is confronting the demographic crisis earlier and to a greater extent that other countries are in the same way uh the Youth of the Indian population uh and the growth of population there should make us qualify the uh appearance of very rapid growth needs to be deducted for growth in the population as well now there was somebody back here on the aisle just one brief particular question to the business dynamism um graph you showed us um if I understand it right the absolute number of firms should be shrinking since 2008 and I'm just wondering how this uh goes along with the increasing um microenterprises we experience basically everywhere uh could you repeat that and make it a little bit clearer the question just wondering the number of micro Enterprises should be increasing um basically everywhere among the oecd countries and I'm wondering how this goes along with the graph you showed us about the number of firms which as I understand the graph about the business dynamism um should actually be shrinking but with with the huge with the increasing levels of um microenterprises I guess the number of firms should anyway increase uh I'll have to plead ignorance on the role of business dynamism in other countries I know this is something that's being studied a lot in the United States it seems to be a problem with young people not going into entrepreneurship in the way that they used to 10 and 20 years ago uh but otherwise I I can't tell you what's going on in that topic in the rest of the world um okay we have someone down here in the front row well how about you since you're closer to the lady with the microphone here keep your hand up thanks um so you've been speaking a lot about um the problems that America has been facing on a sort of national United States level do we know what the picture is like on a state level whether there are some states which are bucking the national Trends some which where productivity is falling even faster and if so do we know how this might change the economic geography of the United States uh the question is about the economic geography of the United States you have a um success and a failure story you have uh continuing um rapid population growth in the large metropolitan areas of the South that continues a trend of the last 40 years that was in Al made possible by the invention of air conditioning so you have places like Dallas Houston Atlanta and Austin Texas that are growing very rapidly uh there you have um minimal zoning so you don't have this inflation of real estate values uh so you have relatively reasonably priced residential real estate unlike San Francisco and New York uh at the other extremes where they're constrained both by geography and by their own uh not in my backyard kind of uh policy uh recommendations um if you look at per capita income the per capita income of the San Francisco area is as much as 20% higher than the national average um in uh uh parts of the Midwest um it's below the national average and the uh the traditional areas of slow growth continue uh to suffer from the post-industrial revolution uh the areas of Steel manufacturer and um heavy Reliance on uh other kinds of manufacturing ranging across the Northern Tier of States from upstate New York through to Missouri and Wisconsin are all growing uh if at all much slower than the national average but San Francisco really stand stand out as a Bastion of high income Prosperity Modified by extremely high real estate values a combination very familiar to those of you in London uh thank you for your presentation Professor Gordon it was Illuminating um I really like the uh productivity breakdown over the three industrial revolutions so I would like to ask the obvious structuralist question if that's that's okay would you say that the productivity increase during the Second Industrial Revolution uh is a one-off event of moving labor from the unproductive primary sector to more productive secondary or tertiary sector uh that cannot be replicated in already developed or urbanized countries yeah the the question is about whether the Second Industrial Revolution uh can be replicated and how it worked out in the primary and the secondary and the tertiary sector s um components of the Second Industrial Revolution were the enormous Improvement in manufacturing productivity made possible by the electric motor and the enormous increase in the productivity of retailing and the service sector partially made possible by uh motor transportation and the elimination of the horse uh so that uh both indirectly through transportation and uh then through the econom of scale that led from uh clerks handing you merchandise in small markets to the over to the big supermarkets and department stores uh we had a Revolution going on at the same time in the service sector uh what that middle part of the 20th century has as the basic feature goes back even further and that was a steady decline in the importance of the primary sector the U the farming uh and a lot of the productivity growth that we achieved back then uh in the movement from low productivity work on the farm where after all people were idle half the year uh either waiting for crops to grow or waiting to harvest there was a lot of time people spent in on farms not doing anything and the transition to the um Urban economy was a big factor behind productivity growth and that can't happen again we only have 1% of our labor force working in farming uh so just just like the speed of sound we can't go beyond the speed of sound we can't go below Beyond a certain fraction of uh transition from Farm to City we have one here another well-dressed gentleman with a tie thank you Professor Gordon um I just wanted to challenge you on the pace of innovation on two points so firstly um to what extent do you agree that uh there's a lot of Technology due to the increasing computing power which um you you perhaps didn't cover in your presentation things like artificial meat uh things like which could completely transform an agonic completely transform um agriculture things like uh energy um Innovation and things like Gene editing technology which could really quite substantially change our capacity to grow in the future and secondly also in terms of the numbers of people and the portion of global of the global population at the technology Frontier so at least it's something to which you alluded earlier but at least to to a certain extent um one of the rate one of the factors causing driving Innovation growth is people North Americans and Europeans but the Chinese and Indians and Eastern Europeans and others who will be increasingly well educated and having the scientific technological skills doesn't that also imply that in the future technological growth capacity and productivity growth capacity will be will rise as well well on the second the second point about India and China and the Emerging Markets uh my way of describing this is we have the developed countries up here growing slowly we have the Emerging Market starting down here looking at it from your angle uh and there is this enormous Gap uh and there is this uh a big uh period of catching up that is going to dominate uh the subject of worldwide economic growth over the next 10 to 20 years uh some of the developing countries are more successful than others in achieving this uh bridging of the gap between the emerging markets and the uh developed markets U and I think as I said before in reference to solar power we're going to be seeing more Innovation coming from the Emerging Markets as their Educational Systems produce more and more scientists and Engineers your first question about the growth in computing power I think has a real time Dimension to it uh we know that uh We've Got Big Data and artificial intelligence and we're moving toward the world in which computers can think for themselves um and I'm not going to stand here and tell you that that's not going to happen uh I think when computers do um learn to think for themselves they're going to be addressing different kinds of tasks than humans um uh man man AG routinely every day um I think the anecdote I read you about the robots uh suggests that robots doing human motions U as somebody once said the hardest thing you can teach a robot to do is to fold a pile of laundry but artificial intelligence computers learning to think we already know that computers won the American game of Jeopardy we have computers now beating South Korean experts at the game of Go uh we know that that kind of artificial intelligence is making great progress and so there are some kinds of human activity ryant on that kind of massive computer power uh that are indeed threatened I just think it's happening slowly and I never predict Beyond 25 years for all I know in 31 years the computers will have taken over but I don't need to predict that because I stop at 25 years thank you Professor Gordon um I was just wondering to what extent would the changing female labor force participation rates have had any impact on these uh growth rate changes I'm thinking particularly in the postc world war period um with that have helped with the very strong growth that was occurring then and and partly account for the Slowdown now as that's a onetime change that can't be replicated uh the question was about growing female participation earlier in the postwar era and think of the difference a lot of this slowdown and growth that slowdown from 3% to 1 and a half% in the actual growth of GDP that I mentioned before that happened before and after 10 years ago uh had to do with the rising hours per person achieved by the entry of women into the labor force followed by the completion of that the labor force participation of women topped out uh and has fallen slightly since the mid 1990s and then we have the retirement of the Baby Boomers which is reducing the labor force participation rate so you have the participation rate going up coming back down it means the potential growth in the labor force that contributed to potential GDP turned around by almost a full percentage point and that actually uh contributes more to this slowdown and overall potential growth than the productivity and Innovation story so demography is king or at least it's the Queen B I think uh we got to end it over here otherwise the stewards are going to get angry at me so before we thank our speaker one more time i' like to remind you if you want to buy Bob's book is that you can do it outside and you can come back in to get it signed but I want to thank you all for coming and please join me in thanking our speaker [Applause] [Applause]

Channel: LSE

Views: 32,961

Keywords: LSE, London School of Economics and Political Science, London School of Economics, University, College, history of economic growth, USA, Robert Gordon, American Economic Association, economics, National Bureau of Economic Research's Business Cycle Dating Committee, recession, United States, Wouter Den Haan, Centre for Macroeconomics, Department of Economics, American Growth, Rise, Fall, economic challenges, future

Id: B-Tia-LcLxI

Length: 87min 37sec (5257 seconds)

Published: Mon May 16 2016