 Thank you very much and good morning everybody. We're changing theme now, talking about employment and innovation and growth. Innovation, as we know, and the adoption of new technologies are essential in order to maintain the country's competitiveness in an open world economy. And as we just heard from Olivier, in fact, when there is loss of competitiveness and then we have the fixed exchange rates, free trade, free capital mobility, the equilibrium is unsustainable without huge reductions in wages. And all the problems associated with those reductions. And that's why it's essential that in the countries of the eurozone which do have those characteristics of fixed exchange rates and free capital mobility, that it's essential that we innovate enough and we adopt enough innovation so as to converge to common levels of high competitiveness, otherwise there is no future to the union. It's absolutely essential. Now the important questions that will occupy me in this presentation are what market conditions are conducive to more innovation and growth and how does more innovation influence the dynamics of employment. I will argue the obvious that innovation and growth can benefit all sections of society, but policies often needed to give extra incentives and help reintegrate any losers in the new economy. My focus will be the structural changes that are caused by innovation and growth and the implications that they have for employment. Now innovation can raise the productivity of all factors of production so both the owners of capital and the workers can have more income but how this income is allocated between them depends on their relative bargaining strengths and on market forces. One of the policy concerns in the last two to three decades is that capital owners have been able to secure more of the gains from growth than they were in the long period of industrialization that preceded them and this has raised income inequality across different labour market groups with potentially difficult policy challenges. In general, wealth creation comes from growth and although in the long run innovation is the only driver of growth there can be lengthy episodes of growth, some of which can stretch to two to three decades without innovation. This mainly comes from above normal capital accumulation but also from other sources. The equilibrium implications of each growth episode depend on the reasons for growth. Growth can come from a reorganization of production a shaking out of inefficiently employed labour and an improvement in the organization of companies and markets. The impact of this development is very similar to that of innovation in the sense that it increases the productivity of labour which is what we want to achieve even if it's at the cost of some jobs. It's an innovation in a general purpose technology office and production line reorganization as opposed to the more easy to understand innovation activity that impacts mainly on production directly. Now temporary growth episodes can also come from imitation and capital accumulation that rises above the trends of the past. Growth in Europe after the destructive activities of the Second World War in Japan a little later in the Asian Tigers which include China here in the 1980s and 90s was mostly of this kind. This kind of growth can be very fast creating large amounts of wealth but it cannot last beyond 20 to 30 years because of diminishing returns to capital. Eventually capital will need to find new and more productive ways of producing in order to maintain its high rates of return. Now the only sustainable form of growth over long periods of time is new innovation. New innovation requires research and the incentive to apply the new ideas in the marketplace. The factors that are needed to make it happen are the three that you see in the slide that I just put up. They are a well-trained labour force to generate and apply the discoveries sufficient investment to enable the replacement of old and unproductive capital equipment and the market environment that is flexible enough to accommodate the required changes in production before competitors get in first and take the benefits. Europe as a whole is not doing well along these dimensions although there are obviously some exceptions and I'm going to take them one at a time. It has a well-trained labour force but when judged by the growth of modern technology companies it does not perform very well. The modern technology giants are mostly American or Asian, in particular Korean and Japanese. The way that higher education and research are organized is important in this respect talking about the highly trained labour force. The best way to achieve results is to combine university research with industrial R&D as is most successfully done in the United States in places like Silicon Valley which benefits from its proximity to the high level universities of Stanford and Berkeley. It's not easy to achieve high standards in research in most countries to some extent due to educational policies that do not give top level researchers the incentive to work in those countries. At the level of top research workers are very mobile and they need strong incentives to stay in countries that offer less good facilities through their universities than the very top ones. The incentives needed by top researchers are not just the high salary although that would be nice that rewards their input into research. In order to achieve high standards in university research universities need to be well funded and be independent. In most countries and most universities funding is primarily provided by government so it may be difficult for politicians to accept that they should have limited say over salary levels, appointments, promotions and other types of spending when they provide the money from tax revenues but it's necessary if research is to be free of political interference. This is why there are large advantages to obtaining funding directly from private industry or distributing state funding through independent bodies like the National Science Foundation in the United States and the European Research Council in Europe which unfortunately is having its budget cut next year. Interference with the university administrative structures and their internal procedures is less likely when the funding bodies are independent of national politics. In the United States university budgets are over 3% of GDP but in Europe they are less than half about 1.3%. Americans also give more independence to their universities and public donations are more generous. According to influential observers this is an important reason why Europe lacks behind the United States in top university performance and top innovations. This is reflected in expenditure R&D which is also lagging behind the United States and Japan and you can see this here. Europe is still ahead of China but this is not likely to last for long as Chinese R&D spending has been an upward trend and it closed the large gap vis-à-vis Europe in the last 10 years but there are obviously exceptions. For example, I listed Germany along with the other averages and you can see that the success of German industry in producing top-level export manufacturers is reflected in its R&D spending which is even better than that of the United States more like Japan's. Now, moving now to my second requirement successful innovation and growth also require investment because most innovations need new capital. The disembodiment assumptions of traditional economic theory which is the solo growth model are convenient simplifications not depictions of reality. In particular, new investment by private companies is needed to embody the new technology and new investment by the public sector is needed to build the infrastructure that is a necessary support for private companies. Now, in total volumes, capital formation in Europe is on par with that in the United States at about 18 to 20% of GDP. It's well below the Asian countries like China but that's not surprising and you see this here private fixed capital formation. But the dynamics of investment in the recent crisis do not bode well for the structural recovery of the heavily indebted countries. In figure two that you see here I show private investment in the United States, European Union and the four program countries whereas investment fell everywhere between the peak year of 2007 and 2012 the end of recession year for most countries in 2012. In the heavily indebted countries private investment fell by a lot more than in the other countries but it's in those countries that we need most of the investment currently. Now public investment is even more worrying because public investment is which is more important for the infrastructure that supports industrial growth. It tells a similar more horrifying story than private investment and you can see that in figure three in proportional terms public investment fell more than private at the time when budgets had to be cut to check the explosion of debt. Public investment is the easiest to cut or postpone as governments say amongst government spending programs and this shows up especially in the indebted countries of the Eurozone. Whereas in the United States public investment fell by a mere 6% in the European Union as a whole it fell by 11% and in the foreign indebted countries it fell between 40 and 60%. You might recall Larry Summers complaining yesterday about the very extremely low level of public investment in the United States. I wonder if he was aware of these figures for European Union as a whole which is well below that of the United States something below 2.5 whereas in the United States it's still 3.5 GDP but then the program countries which are all below 2%. And about the program countries and this level of public investment obviously cannot be good for their recovery and even for their debt reduction programs at the time when GDP levels fell due to recession a robust way of cutting the debt is to focus the government spending programs on items that yield a high return in the future and government fixed capital formation is foremost amongst these. The final analysis I think default for this fall in infrastructure spending lies with the rescue programs which did not discriminate between different kinds of government spending. Although I should say that when I pointed this out in a conversation strictly under Chatham House rules to someone who belonged to a Troika or one of these countries he said to me that even if they tried governments wouldn't have allowed them to tell them what they should cut. They said just give us the overall budget and we'll work on it. And you can see why they are doing it because politically it's a lot easier than social transfers for example which I think is a mistake from the innovation point of view. Now innovation and productivity growth however also needed well functioning labor market my third pillar in this. Labor markets need to be flexible because innovation carries risks. Startups and older firms introducing new products need to know that they can recruit the workers quickly and terminate contracts when things don't work out. In flexible labor markets make firms too cautious and introduce too many administrative burdens that work as disincentives to innovation. A flexible labor market will not by itself lead to more innovations but it will remove impediments that could slow down the innovation potential of the country. It helps in the speed implementation of innovations and the faster adaptation to new technological conditions and the ability to do this makes countries more competitive in work markets so it gives them additional incentives for new cost cutting in new cost cutting techniques. The same applies for product market regulation and I applaud what Cathy of the OECD just said about product market regulation including the enterprise environment. High startup costs either as direct payments or administrative procedures act as disincentives to the formation of companies into the dissolution of unsuccessful companies. High administrative operating costs have similar effects and discouraged development of an underground economy. Sorry, they encouraged the development of an underground economy with high administrative costs which is a strong disincentive to new innovation when a job is not secure because it is not registered and regulated. Neither the employee nor the employer has an incentive to engage in research or training that improves the productivity of the workforce. In other words, I'm completely in agreement with what President Draghi was saying yesterday that I should only only one qualifier. I do think which also relates to the discussion that Tito Boeri had with his discussion earlier this morning. I do think the structural reforms in the labor market are more effective when there is more investment taking place because we do need the investment to support new companies especially if we are grandfathered in the old structural reforms but investment will not by itself come on if we don't first reform the enterprise environment and the product markets. You won't by itself come in response to labor market reforms so I strongly supporting this case prioritizing the reforms that product market reforms should come first encourage the business, open up the business environment, make it easier to set up businesses. The listing that Mario Draghi showed yesterday was the bottom and Italy next bottom open up those markets those will encourage investment and once you have the investment it will be a lot easier to implement labor market reforms. In other words, sequencing I think is quite important rather than throw everything at governments when investment is so low. Now the reward structure moving on with my presentation the reward structure within the company and the labor market should be such that it rewards the investors more generously than the rest even if the input is similar and the incentives to innovate is bigger when success brings big payoffs but here one needs to be concerned about inequality as larger inequality due to larger rewards to successful innovators will create tensions and potential conflict. The balancing of financial incentives of top innovators with sufficient payoffs for lower level employees is a very fine act and may vary from country to country and even from company to company it's a challenge for every policy maker in the developed world and I don't really know the answer to tell you the truth about this one. Finally, a labor market needs to be competitive to attract new innovation even in situations where a large player dominates the market there should be no barriers to entry that can give protection to the incumbent in market environments where there is either actual or potential competition from new entrants the incentive to innovate is greater because the incumbent needs to innovate to keep ahead of the competition when monopolies are protected as in so many state controlled enterprises the incentive to innovate disappears and the company stagnates and since there was a lot of mention of Greece earlier on I do think that this last point is the one that is holding innovation back in Greece is the most heavily protected and most heavily concentrated economy in the Eurozone almost all major sectors of the economy are dominated by two companies which are one or two companies which are protected by others and of course that gives an incentive for innovation now let me now move on to innovation and job creation and job destruction which is really the theme here employment the structure perspective for employment now innovation brings wealth creation and more prosperity but now but how does it influence employment the best way to share the rewards from new growth is to employment not transfers namely it is to ensure that growth is inclusive in the labor market the important question to ask them is about the ways in which innovation and growth influence new job creation whether any displaced employees are absorbed quickly back into employment now many many commonly heard answers about the relation between innovation and jobs suffer I believe from what one might call an aggregation fallacy there is no doubt that a company that is successful innovator creates more jobs but usually this is at the cost of jobs at its competitors even if the company is a massive monopoly this is still the case for example Google displaced a lot of jobs in non-digital advertising and selling even though a newspaper may not be regarded as a direct competitor Amazon displays many jobs in high street bookstores but if we aggregate over all companies in a sector we find that innovation usually destroys jobs in the sector as a whole the jobs created by the innovator are not enough to replace all the jobs destroyed in the sector that fail to innovate it's very likely that fewer personas are today employed to sell a euros worth of books than before Amazon was created but at the level of the company as a whole sorry at the level of the economy as a whole innovation creates jobs through the wealth creation and higher income of those working for the successful innovators the jobs are created elsewhere usually in labor intensive services put it in terms of supply and demand when a sector innovates it increases its own supply demand for the sector's output does not usually increase by a similar amount and so the increase in the sector supply spills over to other sectors creating demand elsewhere in the economy employment shifts from the innovating sector to the less innovating labor intensive sectors such shifts spread the benefits from sector specific innovations more widely and make growth more inclusive in the structural perspective of employment that is the topic of this session now historically innovation has been uneven across sectors of the economy and before looking more closely into the implications of this for employment I discuss more generally the type of jobs that are created in the economy as a whole when there is new innovation in some sectors usually the jobs that are created to absorb the workers who are displaced by innovation are in service sectors where the possibilities of saving technology are limited such jobs are in both business and personal services as businesses become larger and more complicated with new and more specialized technology they hire more specialist service providers these could be hired by the firm to provide the services internally in which case they are classified as workers in the sector of the firm or the firm buys the required services from specialist providers that is outsourcing in which case the workers providing them are classified as business service employees distinguishing between the two in official statistics is very difficult although it's generally agreed that there has been net growth in business services everywhere in the industrialized world the claim has been made that much of the growth in business services that we see in official statistics is due to outsourcing job creation in the personal services sector is much more straightforward to understand as wealth grows, households travel more, consume more and require better service that saves them time a resource that becomes more valuable with rising living standards this is reflected in the growth of jobs in retailing, catering, healthcare childcare, education and domestic service despite many technological improvements that benefit these types of jobs especially retailing in the final analysis there are jobs that provide services directly to the public relying on person to person contact for example in retailing technology improvements in stocking and reordering may accelerate the time needed to deliver products but when visiting a store the consumer evaluates the quality of service from the personal contact that she has with the retail assistants similarly domestic service has been revolutionized with the invention of consumer durables and gadgets but although I might safely argue that the number of domestic employees is not as big as it was the beginning of the 20th century the numbers of employees in domestic jobs recorded in official statistics is growing as incomes grow jobs in both business and personal services could be well paid jobs but there is no guarantee that they will be especially the ones in the personal sector in contrast the profits that go to the innovators that displace the workers in the first place are usually very high otherwise the incentive for innovation would not be present this may sound like reverse causation and indeed there is an element of it but do not offer rewards to innovators may end up with less inequality between the innovators and the service providers but they will also be characterized by less innovation countries that permit the growth of large inequalities attract more innovation activity and the final outcome is the one that I have described high profits go to the innovators and whether wages in the jobs created elsewhere are also high depends on other factors we see the risk of increasing inequality in innovating countries and balancing the two as I pointed out before is one of the biggest challenges that societies face today I now look at the dynamics of employment and I'm going to overshoot by five minutes because you took about half an hour of my time by the way I look at the dynamics of employment now during periods of employment and the structure had changed beginning with a long term look at the United States a brief one but a long term I choose the United States for this purpose because it's the leading nation innovation in its implementation it also has reliable data going back to more than a hundred years similar stories can be told about other industrialized countries and that will refer to some important cases in 1900 the United States in the United States when industrialization was well underway the shares of employees in agriculture, industry and services was about the same it was about the third each it had about 35% in industry 35% in in agriculture and 13% in services agriculture was mainly based on family concerns and was inefficient services were also inefficient large numbers of new immigrants were working in domestic service looking after the children and homes of rich industrialized John Leeds writing in 1917 in one of the very first house to house service of working conditions in Philadelphia found that middle class homes employed large number of people in early 20th century working as cleaners, washers clothesmakers and maintenance workers then the big innovation started reaching the general public cars, refrigerators, washing machines, vacuum cleaners and the many other household appliances that today we take for granted there's never been a more intense period of innovation and there will probably never be one as Gordon, as Bob Gordon many times pointed out and I do have to say agree with him I don't see him here today but even in his absence he's been praised oh he's here, oh he is I'm sorry, oh it's too far to see now people became wealthier and demand for all kinds of goods services rapidly but the share of employment and this is what I want to emphasize here the share of employment in industry where all this growth started increased by just a little from 35% to 40% to over all employment it remained at 40% until 1970 when the big innovations came to an end and growth slowed down it subsequently declined to 20% in the face of competition from Europe and Asia which copied US technology the domestic service that John Leeds found in 1900 had virtually disappeared the work has been displaced by machines agricultural employment also virtually disappeared falling from 35% of total employment at the beginning to 3% at the end the main beneficiary from all this growth was services which now account for more than 70% of employment this story illustrates that innovation is needed to increase a country's wealth and competitiveness but the jobs created to employ the citizens of the innovating country are not in the sectors that experience most productivity gains the jobs are created elsewhere where productivity gains are much smaller because wealthy citizens are also big spenders and want professionals to look after them when they are sick to educate their children and manage their businesses and properties European countries share this experience oops European countries share this experience considered three the UK, Germany and Sweden the data that we have here come from the sector of data set claims which goes back to 1970 and I report consistent data for the population age 15 and above for the United States in the three European countries the figure that you see here report overall employment rates for 1973 the year of the productivity slowdown 83 when productivity turned mostly up in 2003 Sweden is the only country that succeeded in increasing overall employment between 73 and 83 but it did it through a large expansion of its public welfare programs industrial employment declined everywhere and services employment increased comparing 73 and 2003 overall employment rates changed a little except in the United States where they grew faster due to the growth of female participation rates the more interesting data for our purposes is shown in the figure that you see here figure 5 the decline in industrial and agricultural employment was replaced with some overall gains as shown in the previous figure by business services and services catering for the individual including health and education in approximately equal measure the implications of this transition were the implications for employment in our clear cut giving incentives to R&D to speed up productivity growth especially in industry but at the same time liberalized services making it easier to start and run new businesses liberalize the business environment reform product markets to absorb the workers released by industry and draw more women into the labor market women have a comparative advantage in the performance of service jobs especially older women whose participation is weak have a big advantage in health care jobs which is a major growth sector Sweden has succeeded in drawing into the market women of all ages through the subsidization of jobs where female labor has the advantage especially in the health and education sectors and you can see this most clearly here where employment in health and education is rising in all countries but not at the levels of Sweden for example it provides childcare and health services to the market rather than directly to the home the United States and Britain have also been more successful in employment growth in these sectors mainly through the expansion of their higher education sectors and of the countries that you see here Italy is the country that has been least successful because of the way that they handle their welfare state I think given the time I'm going to stop here I had some few more things to say but the important things I wanted to make is about the structural change the structural transformation where R&D is absolutely essential but don't ignore liberalizing the service sector environment labor markets and product markets because those are the sectors that are going to attract the people that will be displaced by R&D activity Thank you very much It's an honor to be here the paper covers a lot of ground and a number of deep insights that I think are quite important now the paper doesn't present a formal model it doesn't estimate parameter values but what it does is convey a broad vision of how innovation, structural change and employment all fit together so to summarize a few of the main points of the paper innovation is crucial for raising living standards over time but structural change is important for getting benefits for innovation so innovation raises employment but it raises it outside of the innovating sector so putting those two things together leads to a lot of sensible policy recommendations in order to get innovation and ideas and in order to get the structural change to benefit from them so things like education and skills public and private investments and more flexible labor and product markets now I agree with all of these points so what I want to do is discuss some specific examples with a particular focus on comparing the US and Europe and that's going to let me provide some context for this broad and sensible vision so specifically the things I want to say is first I'll make three points so first is that innovation has both a global and a local dimension so there's a global frontier and ideas spill across borders and policies and institutions are going to shape the degree to which individual countries benefit from that the second is that whether a country implements reforms or not whether it has the right institutions there's going to be structural change and the degree of the institutions within the country largely just determine the degree to which those reforms give you benefits from that structural change and the third is innovation is highly uncertain what that means is that it raises the imperative to have sensible policies it makes them all the more important so the first point innovation has both a global and a local dimension so in this regard I actually think about think about it much the way that Jim Bullard did yesterday in his question that in many ways a framework of conditional convergence needs to fit pretty well for advanced economies so what this figure shows is GDP per hour relative to the US level for various economies in Europe and elsewhere now these are conference board data but the picture looks pretty similar in other cross country data sets whether it's looking at the OECD or data produced within the Bonk de France and so I just stuck with these and of course it's per hour so it's productivity, they're important issues about hours per capita that are not represented in this picture at least not directly now the US is 100 advanced economies that you see in this picture grew very fast and converged towards US levels over the post-war period so for example Greece and Japan run together until the 1970s and then Greece levels off at about half of the US level Japan keeps going until through the 80s and then it levels off at something like two thirds of the US level of GDP per hour look at France and Germany they keep going up until the mid 90s when they get quite close to the US level and then they level off or as I'll discuss in a few moments even start to diverge a little bit so this conditional convergence framework sort of suggested by this picture is consistent with the ideas with the view that ideas don't respect borders so innovations in one country are in principle available to people anywhere so people in Japan or Germany or Greece can all benefit from iPhones or Android devices they don't have to produce them themselves now there is of course an important local dimension so I don't want to just talk about the global dimension because as you see at least a couple of points on this one is countries look like they level off before they reach the US level so they fall short of the US level and a second is that after the mid 1990s many continental economies stop converging or even being level and start to diverge slowly so they have lower growth rates after the mid 1990s so I want to talk about that now this picture took the US as the frontier and just set it to be 100 but what that means is that we can look at what's going on in the pace of innovation at the frontier by essentially looking at what was going on in the US so that's what I want to turn to now doing so is also going to help us to understand why Europe has slipped back a bit on this figure the main story that people will tell that I will tell is one that emphasizes ideas and institutions and their interaction to explain that divergence so to look at the frontier I'll start by showing US productivity growth and its long level since 1973 so you see the sharp pickup and growth rates in the mid 1990s I put a vertical line there but that pickup and growth rates didn't last formal statistical tests for breaks in labor productivity growth or in TFP growth total factor productivity growth both find breaks in the mid 90s that finds a speed up and then it finds a statistically and economically significant slow down prior to the great recession in the early to mid 1990s so indeed if you look at the four years prior to the great recession in this picture where I put the second vertical line the growth rate of labor productivity or TFP was almost identical to what we've seen since 2007 going up into early 2015 now a vast literature has explored the mid 90s speed up and found that it reflects innovations related to the production and use of information technology so things like computers and communications equipment software, the internet and the like but it's much more than just buying IT, getting the benefits really often came from substantial reorganizations to take advantage of how business is done, change how businesses are organized or even having new businesses that are better able to take advantage of the ability to, of changes in the ability to manage information and communications now only a small literature at this point has looked at the slow down after the early 2000s and of course with aggregate data like this you can tell many stories and obviously there were interesting things going on in the US economy and maybe the global economy in the 2000s but so I'll turn to looking at some industry data that will shed light on this and that's going to suggest that the slow down was the flip side of the speed up that it was an end to the exceptional gains from information technology in the US exceptional gains when it came to productivity go the other way so what I'm showing here is industry data on total factor productivity growth by different periods of time periods sort of suggested by the previous period, by the previous figure and so I'm mostly going to focus on the period prior to 2007 because obviously a great recession can cause all sorts of interesting issues with productivity dynamics as they have in the US and elsewhere and but also focusing before the great recession highlights the slow down in productivity predated at the recession so the bars here some to average TFP growth for the periods shown and I've divided the bars into four mutually exclusive pieces I keep going the wrong way so the first focuses on some of the unusual things going on in the mid-2000s the housing boom and bust excesses in the financial sector surging commodity prices and the like so the first slice of the data focuses on those kind of bubble sectors of construction finance natural resources so you see a slow down going from 2000 to 2004 to 2007 so their contribution becomes more negative but the other three quarters of the economy are more important as you can see from the bars that are actually above the zero axis and I've divided that into three mutually exclusive categories so IT producing intensive IT using and not intensive IT using so those latter two are based on estimates of IT capital payments or the share of IT capital in total value added and their measured the weight in GDP of those latter two of IT intensive is the same so what you see in the late 90s is this big burst in the contribution of direct IT production so things like semiconductors but that when you got into the 2000s that the contribution eased but what you saw instead is IT intensive industries just got a big explosion in what they were contributing so this is things like wholesale trade, business services utilities a few manufacturing industries but a lot of them in services well the story for that to a large degrees looks like reorganization so in distribution once Walmart was taking advantage of changing how it did things and their competitors were and then by the early 2000s your mom and pops who were competing with them largely disappeared so the low productivity firms exited so there was massive change in the organization of many of these industries that showed up as a IT boost or a TFP boost but it didn't last in terms of growth rates once you get to the period running up to the crisis and those are the 2004 to 2007 period you get back to something that looks more normal and perhaps incremental so lots of innovation going on but not at this transformative rate so this analysis is the reason why when I think of Olivia and Larry's paper from yesterday I think in the US case it looks like there was a change in trend growth around the time of the great recession but in the US case it clearly looks to me like it predated it when I look at the industry or the aggregate data now even if it was a one time boost Europe never got that they just got the underlying load out so the leading hypothesis in the literature is one that's consistent with Chris's arguments product and labor market regulations get in the way so benefiting from managing information communications requires reorganizations of firms often requires new competitors and that's those reorganizations the whole process of creative destruction is more costly if product and labor markets are rigid so President Joggy yesterday noted some of the resulting differences in IT usage and efficiency between the US and Europe now Chris emphasized that innovation whether direct production of ideas or through its diffusion require structural change and I agree with that I keep going the wrong way but that's going to bring me to the second issue that structural change or a second point structural change is going to happen whether we wanted or not so I want to make this point in a way that's similar to what Chris does in his paper but it actually has a little bit more industry detail so in the past couple of decades it turns out the cross industry pattern of employment growth is almost identical between the US and Europe so if you look across the horizontal axis this is showing for an aggregate of major euro area economies shows industry growth rates or average industry annual industry growth rates by industry from the Clems data the vertical axis shows the same thing for the United States and what you see is the correlation is above 0.9 and actually the slope of this is close to one so the US and Europe got almost the same changes in industry employment growth despite very different institutions now if you can look at this for shorter time period what's going on in the UK and Japan it looks the same the correlations are extraordinarily high in terms of what industries were growing and which weren't now it's pretty clear that the industries in the upper right are mainly services the industries in the lower left are mainly goods so the pattern could reflect technology which is of course what Chris highlighted and innovation where TFP growth has been much higher as you move to the lower left than in the industries in the upper right but it could also reflect trade or preferences it doesn't actually matter the structural change has happened and it's happened the same way so an implication of this is Europe had greater labor and product market rigidities and barriers but that didn't stop the tide of structural change instead what it did is it reduced the necessary reallocations within and across establishments Europe got an inefficient allocation of capital and labor it got lower productivity growth lower income growth but it still got the structural change now let me bring a note of hope here European economies and others have been implementing structural change so this is a convergence plot for one measure of product market rigidities so the horizontal axis is an index is the level of the OECD's product market regulation as of 1998 so going to the right means greater regulation the vertical axis is a change from 1998 to 2013 so as you go down you see greater reforms sort of more improvement in regulatory barriers and so countries that were further from the frontier to the right in 1998 have implemented greater reform so that is a potential positive it's hard to know how it's going to play out obviously there's still much more to be done but Europe has been taking steps so the final point is that future innovation is highly uncertain and that's going to emphasize the fact that you need to be ready if you're going to be able to get the benefits from that so having markets that can adjust is important so of course one issue you could point to is issues like the technology of robots and driverless cars and artificial intelligence and the like so regardless of whether you think how you think these innovations are going to compare to the great pre-1970 innovations that Bob Gordon has highlighted they're going to lead to changes in structural in the structure of production and the types of jobs that are needed to predict and Europe can't escape that now a second source of uncertainty goes back to the fact that ideas flow across borders so this will be my final point so Chris compared R&D by country that matters for some issues but ideas flow across borders so having more people around the globe looking for new ideas has got to be good for generating ideas so this shows R&D as a share of opportunity for a selection of countries so 20 years ago if you were thinking about where is R&D research being done that's relevant for advanced economies you'd think the US Europe, Japan but what you see is places like China and Korea have been dramatically increasing their investments in R&D places like India as well so what this means is there's a larger pool of potential Thomas Edison's and Steve Jobs's who are out there making the chances for future major breakthroughs and those breakthroughs will spill across borders so to conclude this is a really nice paper my comments have focused on providing some examples and context for the thought-provoking thing so innovation has a global and a local dimension in the US it looks like frontier growth prior to the Great Recession but of course one of the issues of that is Europe never got the late 90s speed up in the first place there's an opportunity there which is something that President Draghi discussed yesterday now there's a lot of uncertainty about the future which means being ready for it is all the more important but regardless of whether you're ready for it or not structural change is going to happen so you want the reforms to be ready so that you can gain the benefits of whatever happens rather than just varying the costs thanks