 Yn mod i'r byw pwzled, mae'r gymhwysgfa yma am yn cyflaenau. Ond wrth gydain yn tu'r gweithio Lethertyn gan hwnnw deall gwneud drwyddiad o'r hoddiadau rhef i gyd o weithio gwneud yma. Rwy'n mynd i'n disgynfaint i chi, ac rydyn ni'n ymddych chi'n meddiadau. Rwy'n mynd i'n mynd i'n meddwl dafodol, ac roedd i'n mynd i'n meddwl armigion am eu cyflaenau. I don't try to convince you that these kind of people are quite important for growth. So you might know this is a fictional manager of the American version of the office, Michael Scott, who is his world's best boss here, is actually the world's worst boss. So I'm going to argue in this when we think about growth and productivity, we should actually be thinking a lot about managers and management. And I'm going to tell you about why I think that as we go through and what it means for government policy. So it's very easy to motivate this. We care about growth a lot. The world, you know, Ireland, Britain, the rest of the developed world is still suffering from a serious hangover from the global financial crisis in the EU growth is particularly weak. And the question is what is to be done about it? I mean broadly, in my view, I think there has to be a combination of three things. The three pillars that both Mario Draghi and Abe have talked about, which is a combination of monetary policy, fiscal policy, but also, and this is the focus today, structural reforms. I'm going to really talk most about those today. So there's four plans that I'm going to talk about. First was the economic challenge that we face. Then I'm going to talk a bit about the research we've been doing over measuring management, then talk about the relationship with performance and finally come back to what this means for policy and the drivers of growth. So just to give you some sense about the differences and the challenges we face. So this is a comparison between the 15 largest countries in the EU and the United States from 2014. So Europe has about on the measure of GDP per capita something like 30% lower income than the United States. That's quite a big gap to think about. When people see these numbers, then one view as well, this is just because unemployment is a lot worse in Europe than it is in the United States. So if you correct that and look at a measure of productivity, which is GDP per worker, then we still see a very significant gap. So somewhere like 22%. So through to some of the gap, maybe 8 percentage points, but certainly not all of it. The other story is of course that maybe it's all to do with taking longer holidays and in fact that's not the case. So if you correct for hours, you still see there's a very significant gap still remaining. So most of the gap isn't simply due to the labour market, hours and jobs. It's also to do with the fact that those workers are less productive when they are in those jobs. So this problem, you might think, well this is only a reflection of the financial crisis, but in fact this actually preceded the financial crisis. So if you look at productivity, output per hour worked over a period since 1980, comparing, say, the EU and the US, you can see that for a long period of time, this is the levels, the US was more productive than Europe, but Europe was catching up. So Europe is the red line there. You can see, and this has been true since the Second World War, there was this gradual convergence until in the mid-90s in Europe it had almost got as productive as the United States was. The gap started to go in the opposite direction. There was divergence between the European Union and the US even prior to the financial crisis. So this has been exasperated by what's happened in the last few years, but it actually preceded what had happened before. So again, just to give you a sense about the impact of the crisis. So this takes the ten years before the crisis versus the seven years after the crisis until last year. So here's the US, so productivity growth, genie pupil work of growth was going at about 2% points per year prior to the crisis, and then after the crisis, and the US has been slowed down to around 1.5%. If we compare that with Europe, this is what I showed you in the previous graph, Europe was already had slower productivity growth before the crisis, but after the crisis, there's been a disaster where productivity growth has fallen to about half a percentage point. Of course that's even worse in some countries than in other countries. So that shows you that the crisis has exasperated a problem which was already before the crisis. If you get interested in Japan as a third comparison, Japan was actually doing worse than Europe pre-crisis in terms of productivity growth, and during the crisis, after the crisis, or during the crisis it's done slightly better, but still there's been a slow down. So although all countries have been affected by the crisis, it's happened to a greater extent in some parts of the world, particularly Europe than others. So that's the kind of macro, the big picture. Now it's useful to look at the big picture, this is where Dan has come from, but I like to go underneath that to look at the firms and plants and stores underlying those numbers. So the interesting thing is that there's these big differences across countries. You see also huge differences in productivity between firms within countries. So in a typical narrow industry, say take US concrete, but this is an average, and you can look at any industry. If you compare the plants at the top 10% versus the bottom 10%, labour productivity output for workers is something like four times higher in the typical industry in the top 10% to the bottom 10% of plants. Of course you might say that's all due to different capital or different types of workers. If you control for all of those things, economists have this ugly name, total factor of productivity, which basically means efficiency. If you look at the inputs and look at the outputs, how much of the difference of output is simply due to the way you use those inputs. So that TFP difference is still absolutely enormous, a kind of two to one difference. Even in the single industry, in a country like the US where things are very competitive, if you look at other countries, those gaps are even bigger. So very large amounts of variation and heterogeneity within industries, and these are kind of persistence. And as I said, they're even bigger in the European Union and developing countries. Acro trends I described before, those increases of productivity. It's not true to have this kind of view that somehow every firm is kind of getting more productive. A lot of this productivity increase is coming from what you know, Schumpeter, that Austrian economist called the creative destruction, which is the phenomena by which more productive firms kind of grow unless productive firms shrink and exit. So at least half of the overall productivity growth you see is due to that churning, that kind of creative destruction. And that makes you immediately think, well, understanding why firms' productivity is so different is a really important path of thinking about these macroeconomic productivity differences. So are there reasons for these differences? Well, as I said, technology matters a lot, but it's not anything like the whole story because you can control for any measure of technology under the sun that you want to, measures of search and development or patterns or these type of things, you still end up with a very large unexplained residual. And furthermore, if you look at the impact of new technologies, so I've done some work with colleagues at Harvard and Stanford looking at the impact of, say, new computers, new information communication technologies, the impact that they have on firms' productivity depends fundamentally on the management and organisation in place. You can spend huge amounts of money on computers and they can actually do nothing for your profits or products or in fact they can reduce them sometimes. So thinking about management actually may be as important or more important. And certainly business case studies and the history of economic thought has placed these as an important thing, but until recently it's been impossible to really gauge that in any kind of systematic way. So what I'm going to try and do now is convince you we have some way of actually strengthening some light on this kind of issue in a kind of quantifiable way. So how on earth do you quantify management? Isn't this impossible? Well, so Nick and I, Nick Bloom and I for the last ten years or so, I've been trying to make some progress on this working with many other people. So there's three kind of steps, I won't bore you too much with the methodology of this, but there's basically three steps trying to measure management. So I want to sort of think about different practices which may be important for improving productivity. So we work with lots of people in the area and in consultants and coming up with 18 of these. So not this is rocket science, this is all about how you do what you do more efficiently. So things like how you set sensible targets, you set things which are stretching what's impossible to reach, do you have a mixture of financial and natural targets, things about monitoring, so how you collect and use data, how you do lean manufacturing type of techniques and people. How do you reward people? How do you promote people? Are you based on effort and ability or are you basically just promoting people who are connected, who are based on tenure rather than their actual effort or ability? So I'll give you a couple of examples on the second, but none of them are kind of politically complicated things. So we have a set of questions, we then interview firms, so we have a basically, usually between about a 45 minute interview of plant managers, manufacturing plant managers co is a great to talk to, but they often don't know what's going on in their firm, so it's good, we'll target the people in the middle. So we have these set of questions, and the question is how do we get the truth out of people, don't people just lie to us when we interview them. So we have a bunch of tricks that we try and do, we call this the double blind technique, so the interviewers doing the interviews don't do anything about the performance of the company, and on the other side the managers who we interview are not told they're being scored, so we're going to score these managers from a low score to a high score, from 1 to 5, they don't know that they're being scored, and the reason that we keep that blind on their side is because the well-known psychological bias that if I'm interviewing somebody they're going to say what they think I want to hear. So they think they're having an open-ended conversation, typically these were students, often MBA students from all over the world, in the meantime these people have been trained to score them rigorously across these different terms. We can talk about the ethics of this later. It passed the Stanford's Ethics Committee under the category of what's called Necessary Deception. Finally, why does anybody participate in these interviews? We've interviewed about 15,000 firms in 34 countries. Why does anybody participate? Well, we have very tricks to get people to participate, so we introduce as a lean manufacturing interview, we get financial information from company accounts and Mr Data, we get official endorsements from respected institutions like the Bundesbank, Bank of England, the Federal Reserve and so on, different things work in different countries. So in Germany, we send the German managers before we interview them an email which has the double-headed Prussian crest of the eagle and the Germans are very respectful of authority as many of you know, so the managers respond well to that and answer our surveys. In America we had left up on the Federal Reserve and we had things like Liberal Communist shouted us down the phone. We found that Americans respond very badly to anything which smacks to the governments, but they respond much better to the fact that these are run by MBA students doing that stuff. So it's horses with horses. But the bottom line is we get about a 45% response rate which is very high for a voluntary survey. Because we know the non-responders well as the responders, we can show that it's not as if it's the more productive or better profit firms are responding, it looks very balanced on the characteristics. Okay, here's the kind of questions we have. So one question is how you track performance in your firm. So a low score here would be companies who are not tracking anything or the things that are tracking are unrelated to their business objectives. A high score would be firms who are actually tracking what goes on frequently and communicating that to the staff. There's no good collecting information on our portfolio inventories or how people are doing. It was locked away in your office drawer. So the important thing is actually communicating this. Sometimes when interviewing don't see this we'd have an open-ended conversation with them. I'd say things to them. I'd say well Dan, tell me about some of the information you collect here and he'll talk to me and from that ulcer that would be how I would score him. So we've done things like sending in students into these factories afterwards with clip balls to see what happens in the variety of things. This type of open-ended questioning seems to be somewhere between a case study approach and a more standard type of survey based approach which economists are used to. So as I said, we now have in the world, to all this data is freely available. You can download it from the websites. People are using it all over the world in teaching and other places. We have it in five major waves. We cover 34 countries including Republic of Ireland and it's about 15,000 firms. I'm mainly going to show you the results in manufacturing. The typical firm here has about 250 workers. It's a medium size, not a small or large firm. The basic methods we use here can be used in other sectors. So we've done this now in hospitals and schools and retail in a variety of things. The question needs to be changed to some degree but a lot of the questions around people are basically very similar, no matter which sector you work in. So what do the schools and the doors look like? So what we do is we take all this data at the firm level and we calculate to begin with the average management scores. This is a measure of management quality across different countries. You can see that the ranking of countries looks similar to what you might have expected from the relative wealth level. The US gets the highest average management score. You have Japan, Germany, Sweden coming next. European countries like Britain Ireland is down here as well coming further down and then southern European countries lower and then countries at the bottom are African countries and Latin American countries. So if you correlated that with GDP per capita you see as you expect that there's nothing causal here, this is just saying as you expect that richer countries tend to have higher management than the scores in the poorer countries. So you can see that lines up reasonably well. Okay. More interesting to me though is not just the averages across countries but looking at within countries. So what we can do with our data is we can say well let's look at every single firm that scores by firms. So these people down here on the left between one and two are the really badly managed firms. They're not collecting information about anything. Their targets are really easy to reach. They're not promoting people based on merit. They're paying people irrespective of their ability. It's a wonder to an economist how these firms even exist but we know from experience they do. In most countries this shape is like a bell curve but the thing which strikes you about this and I put an island typical of many other countries there are really bad firms or really good firms. It's an amazing variation of the quality of management just as you would see for firms' products. So it's not like when you compare the United States with India every American firm is awesome and every Indian firm is terrible. There's good and bad firms in both countries. It just happened in America for example. The future is there's just a lot fewer of these really badly managed firms. You can see America in the top left hand corner there. Okay. So that's fine. There's some data. What about, does this matter in terms of the things we might care about in terms of productivity? So the first thing you can do is you can say well here's the measure of productivity on the vertical axis here's the measure of management on the right and these are all our firms that tend to have better management scores. So positive relationship is very strong. So that's good in one sense. It shows our data is correlated with something we care about which is productivity. You can see the same thing for profits or survival or growth. But of course, is this causal? Maybe there's something else we haven't controlled or maybe the better managed the more productive firms can afford to pay well pay consultants to come in to help and talk up the management practices. Causal. So what we've been doing in recent work has been running experiments. So one of these experiments for example is in India. So in India we did experiments on textile firms outside Mumbai supported by the World Bank and we marketed a program whereby firms could apply and if they were successful they were given free management consultancy by Accenture over a five month period to turn around the management practices. So the randomized treatment plants who applied and were successful got these heavy management consultancy to improve the scores that I've just shown you. The control plants got a four week very light consultancy basically just enough to get their data and then we looked over the next few years what happened to their performance. So one thing which happened is our measured management practices improved but more importantly their productivity improved a lot. So the management index improved by about two standard deviations and productivity increased by about 20% in kind of dollars, this is US dollars that meant that the treatment plants the ones who got this intervention made about $300,000 more per annum than the other plants and that's compared to if they'd have bought this on the free market this would be about $200,000 so very very large increase of profitability and productivity and just to give you some idea this is after the kind of weight loss diet plan this is the kind of prior this is what these kind of places look like prior to the intervention so very disorganised machine tools left lying around instruments not removed oil leaking from machines and afterwards we introduced this is a basic kind of management change program that you'd see in many consultancy interventions organising the factories more clearly tagging things when they went wrong collecting information to try and improve that changing the kind of weight that many people have managed so if you look at say the treatment versus control plants you can see there's a very big improvement productivity which seems to persisted even when the consultants left after the end of the intervention so these it does seem to be this is like a clinical trial of management if you like it does seem to be that the dose had a response and if you compare this to the evidence from the 15,000 firms we get something like a standard deviation increase of management improves productivity by about 10% so ok so it matters at the firm level but doesn't matter at the macroeconomic level so just to give you some sense of that we looked at the differences of management across countries taking into account different sizes of firms and then we said well how much of the overall productivity gap can this explain if any so here's one example for one country which is quite badly managed so compared to the US Britain has about three quarters of a standard deviation worse management than the US has about 20% less productivity so about 27% so just over a quarter of Britain's gap of productivity in the US is due to worse management so you know I showed that picture of Michael Scott from the office the office came from Britain that was a successful British export about bad bosses you can see that bad bosses have managed to knock over a day a week of productivity of the UK so you know it differs in different countries but overall it's about a quarter of the productivity differences are due to management according to our kind of estimates Ireland has actually a similar amount of lower productivity because of management as the UK does ok so that's so what does this mean to policy well one of the things that we can do is actually look at some things which may be drivers of management practice so here's one, here's foreign direct investment so the grey line is kind of what I showed you before except here we just look at domestic firms and the average management scored domestic firms, the black lines the average management scores of foreign multinationals so you can see for foreign multinationals they have higher management scores than almost every country they go to was domestic firms they just mirror what the overall score was so what this says is that foreign multinationals are somehow able to transplant some of the better practices overseas and you know it does give you a sense in which well one of the benefits of having foreign direct investment might be bringing better management practice we were talking about this over launch in fact and you see this very clearly from the data the second thing is that so if you look at different ownership types these again average management scores some of the firms with the lowest management scores are firms which are family owns and have a family CEO it's not a family ownership per se which is a problem so if you're a family owned with a professional CEO, a non family CEO you have a pretty high management score there are firms with low management scores the ones which are second generation firms where the CEO is the son or the grandson or the great grandson of the founder so you know another way of saying this one of the surest ways to ruin your business is to give it to your older son we got rid of monarchy for a reason I'm showing these correlations these are robust and controlling for size for age for many other factors so a third thing is competition so I showed you in the US one of the reasons that the US is successful in having better management schools is they don't have some of these very badly managed firms if you look at a simple measure of competition which is like in this case a number of competitors against a management score the firms which seem to have very high management scores the ones which are facing more competition and that's true in the manufacturing sector the retail sector and even the other parts of the economy so a fourth thing that we've looked at a lot of is education so unsurprisingly you may be more educated managers you tend to get a better management score this is the proportion of managers who have a college degree but interestingly also for non managers for the firms which actually have more educated or trained workers non managers but also the firms where management practice tends to be better so it suggests that one of the ways to improve things is not just a managerial education but a general education of people in the workforce so just to wrap up so I've given you a flavour about some of these work and I've given you a rather unconventional view that in terms of thinking about long term structural reform policies things which policies which can aid growth also those which can improve management so in terms of policies one thing is to actually allow greater entry and greater exit of inefficient plants so removing protection for some of the dominant incumbent plants both small and large plants can be a way of allowing creative destruction to let the better management firms grow more easily competition seems to be particularly important so deepening the internal market especially around services being of the importance of the US-European free trade deal I think we can talk about Brexit in a minute but one of the big advances of being in the European Union I think for many countries has been actually strengthening competition which has had this very powerful effect I'd argue on productivity management so leaving the European Union is an extremely bad idea or failing to deepen the internal market is a bad idea third thing is foreign investment so I've shown you the importance of FDI so Ireland has been very successful in that I applaud that education and skills are very important I haven't talked about labour market reform one of the things that you also see especially in terms of the people management practices is very heavy labour market restrictions can also have a negative effect on managerial quality so thinking about having flexible labour markets is another way of improving management and productivity so overall Europe is at the moment there have been made a success as an integration in the European Union over the last 50 years it's created the largest single market in the world we've integrated ex-communist countries countries from form of fascist dictatorships in southern Europe but going forward we need to think about the structural reforms to improve our productivity I really want us as Europeans to think of ways in which we can improve management and reform otherwise we face another era of slow growth so just before I end I just want to mention that one of the great things about doing this kind of research for an economist is you do the radical methodology of talking to human beings which is something we don't do very much as an economist but actually is rather fun so you learn things that you weren't expecting to learn so you might think defining this foreign ownership variable was very easy in fact it turns out to have been rather more complicated so one production manager in Italy said we are owned by the mafia and then a rather nervous interviewer said well I think that's the other category although I guess I could put you down as an Italian multinational is how this firms code in our data if you have a look at it many will I ever present this in America so I was in America last year on Sebastian Harbour and America's got like this stuff because America does well in management as well in productivity America's not great at everything so my daughter was at school in America and there was some issues here with the schooling so one interviewer asked how many production sites do you have abroad and manage in Indiana well we have one in Texas Japan is the right answer Texas is a different country now these are some more plus being correct ones coming up so here's this is one of the most hilarious ones we had this is in Germany and the interviewer said hello hello are you still there the production manager said I'm sorry I got distracted by submarine servicing in front of my window it turns out this was a kind of defence plant and this is perhaps the most bizarre unbelievable one this is in Japan where a male manager said to a female manager I would like you to call me daddy when we talk and that was end of interview end of interview end of talk