 I'd like to welcome everybody here today to the sixth Innis Christi lecture in Labor and Employment Law. And I'd like to extend a particular welcome to members of Innis Christi's family who are joining us here today. Before we introduce this year's guest speaker, I'd like to begin by saying a few words about my former colleague, Innis Christi. Innis was born and raised in Nova Scotia and started his academic law career at Queens University in 1964. In 1971 he took a position at Dalhousie Law School where he taught until 2007. He also served as dean during my time as student at the Law School and that would have been in the late 80s and early 90s. Innis was one of this country's great legal scholars in the area of Labor and Employment Law. His seminal book Employment Law in Canada was on the desk of every Labor lawyer in the country at the time. He's also had a leading role in law reform. He drafted the Nova Scotia Trade Union Act in 1973 with former Dean Reed and the Nova Scotia Labor Standards Code in 1972. In the 1970s he served as a member of the Canadian Anti-Inflation Appeal Tribunal. He was counsel to the Nova Scotia Labor Standards Tribunal and was chair of the Nova Scotia Labor Relations Board. Innis was also deputy minister in the Nova Scotia Department of Labor and he served as chair of the Nova Scotia Workers Compensation Board and served on the Canadian Human Rights Commission Tribunal. Innis taught and inspired generations of students. His influence lives on through this biennial symposium which was initiated through donations and gifts from the many friends and admirers of Innis from across the country. I'd like to extend a note of thanks to Bruce Archibald, our resident Labor law expert who took the lead on arranging our eminent speaker today, and to Elizabeth Sanford for organizing the events associated with the Symposium and Associated Short Course. Finally, I'd just like to thank you all for coming here tonight to celebrate Innis' legacy. I'm going to turn things over now to Professor Archibald who will introduce you to this year's guest speaker. So, good evening to Sean, Mike, and members of the Christie family I stand with, to my colleagues and to Chief Justice and all you important labor and employer lawyers and students in the room. It's a great pleasure for me to have the privilege of introducing this year's visiting lecturer in the Innis' Christie series. Simon Deacon, I first began to read his stuff. I don't know, I was a latecomer to teaching Labor law, as you know. But it's been a huge influence on me personally but it's influenced people around the world. So that Simon's an... Oh, sorry, I don't know, I just do. You've got a sneak preview as a result of my technology and competence. Simon is a professor of law at Cambridge University in the UK but he's been a part of the Judge Business School in Cambridge and he's currently the director of the Centre for Business Research in that institution. He graduated from Cambridge, he's a Cambridge guy, he has a BA with honors first class and a PhD. Despite that, he's been asked to lecture at Oxford on occasion, strangely, as well as in Budapest, in various cities in China, Cape Town, Nigeria, Moscow, Luva, New Delhi, Frankfurt, Nantes, Kyoto, Melbourne, New York, Florence, Antwerp. He's a sought after speaker. Because his research product is so extraordinarily interesting on the one hand, forward thinking and balance. This is a man who thinks outside the box but does so in a very careful fashion. I could list the huge numbers of articles that he's written. I think, I don't know, I hate to make these kinds of statements Simon but it seems to me that your book with Frank Wilkinson, The Law of the Labor Market, Industrialization, Employment and Legal Evolution, has changed the way people think about labor law around the world and it's an extraordinary thing that you too were able to do and you continue to enhance and build upon that seminal work and it's an extraordinary thing that you're here. Those of you who've been involved with the lecture series in the past know that you kind of have two formats which we alternate. We have always have the lecture of this event but one year it's associated with a symposium and the next year it's associated with a short course. Simon is doing the short course so we have people who are involved in this year's current issues and labor and employment law is what the curriculum says but Simon's focus is labor regulation and economic development past, present and future. So over the next three days including this people are going to be reading some fascinating stuff I've had a chance to look at it and I think it's really important in the context of the Ivony Commission and the challenges that we face that we be able to address the kinds of questions that Simon is working on and today's lecture really is the first lecture for the course of being something which all of you are attending forging a new consensus, labor law and economic development this is an important topic and so I look forward to hearing from Simon and I'm going to sit in on his classes over the next couple of days so it's a real privilege. So Simon please come and address him. Thank you very much Bruce for those words of introduction and also Michael. It's a tremendous pleasure and a privilege for me to be here to honour Ines Christie, one of the greatest of labor lawyers and a great pleasure for me to be here at Dalhousie Law School. Thank you very much indeed for this invitation. What I'd like to talk about tonight is the relationship between labor law and economic development and to explore theoretically and empirically how they might be linked and to address these two linked questions what are the effects of labor laws on poverty, inequality and employment worldwide and then in particular do labor laws help or hinder economic growth in the developing world and when I mean labor laws I don't just mean collective labor law although that's a very important part of what I want to talk about I'm also referring to what we in Europe sometimes call individual labor law or what in Canada might be referred to as employment law, the law of employment protection but also social security law law providing a basis for social insurance all these areas of law I think are linked and in the history of the development of the global north or the west they've played I would say a pivotal part in industrialisation not just responding to the rise of an industrial economy not simply tempering the effect of the market economy tempering the effects of capitalism I think these labor laws, employment laws social security laws helped to create a labor market that was a subject of my book with Frank Wilkerson that Bruce kindly referred to the labor market, labor law, social security law helped to create the conditions for capitalism in west and is still today I think helping to foster economic development in emerging markets so the basic message I want to develop labor law isn't a hindrance on the economy but is a fundamental part of capitalism and I want to address therefore how labor law can constructively contribute to continued economic growth and to sustainable economic growth, socially cohesive economic growth in the global north and also to development in the global south theory, what does theory have to say about labor law quite a lot and development economists address these themes from the 1950s onwards almost at the same time as many of the great contributions were being made in the field of labor law by our foreigners but so-called structural transformation theory associated with the Nobel Prize winners Lewis and Kuznets argued somewhat optimistically but nevertheless powerfully at the time that industrialisation would create the conditions for labor law, for social insurance and for the welfare state and also for a virtuous cycle of social progress and economic growth this is a linear notion of development which argued that economies as they grew as they moved from being subsistence economies in which the vast part of the population was dependent upon the land and upon the family for subsistence as economies developed there would be a transfer of resources from low value added to high value added activities and that capitalism the replacement of subsistence economies by the market would create the conditions for a counter movement that would create the welfare states because capitalism created a working class and this working class would mobilise politically for labor law reforms which would give rise to better housing, better wages, higher wages collective bargaining was a very important part of this so social citizenship associated with the welfare state and over time this process which had been begun in western Europe and North America in the 19th century would spread to the developing world so developing economies, emerging markets of the 1950s of course had an advantage over the west in terms of low labour costs but this advantage wouldn't be permanent because over time this would begin to converge with those of the global north so Louis and Kuznets argued that capitalism produced growth eventually for all both within countries and globally and that there would eventually be some kind of convergence between developed and developing countries now that maybe hasn't happened but we need to treat carefully the evidence for and against the so called Kuznets curve Kuznets argued that economies begin the process of industrialisation in a way that very often requires cheap labour requires minimal welfare state provision so in the early years of industrialisation we do see increasing inequality inequality in the earlier stages of industrialisation we might associate this with the immiseration associated with the early industrial revolution in England lakes, dark satanic mills and though economies all reach Louis and Kuznets argued a tipping point or a turning point after which a more equal distribution of earnings and incomes is established and there's gradual inequality reduction now when Kuznets was writing in the 1950s this was convincing at the level of the empirical evidence today of course and recently controversially Thomas Piketty has argued and maybe demonstrated that this process of equalisation which was occurring up until the 1970s began to go into reverse in the United States and Western Europe in the late 1970s early 1980s now that Piketty hypothesis is certainly well made out in relation to earnings more controversial is the claim that we're seeing increasing wealth inequalities and more controversial still is the idea that the period between 1920 or so and 1979 or so was itself the exception, was itself the blip and that capitalism actually produces greater inequality those claims by Piketty have yet to be made out we don't really have enough historical data to be absolutely sure that the Kuznets curve has been refuted if we look at the cross-sectional data for where countries are today placed on the Kuznets curve we actually find that the period of growing inequality or at least the association between inequality and development is actually quite weak and even economies like China as they grow are beginning to reach what we might think of as a tipping point but the process isn't straightforward and why is that one reason is of course that Kuznets and Lewis were too optimistic about the conditions under which emerging markets were able to grow. The big difference between Western Europe and North America in 1830 or so and countries like India and most of Africa today is that they're developing of course in a context where Europe and North America have already developed. America and Western Europe didn't have to contend with other nations which were more developed than they were but today's developing nations do not operate on a basis of symmetry or equality with the already industrialized world. This was the observation made by Ralph Prebbish and Hans Singer in the early 1960s and late 1950s they demonstrated empirically that in an economy in a global economy where some countries are more developed than others the terms of trade between countries systematically disfavor emerging markets and low income countries. So dependency theory in today's emerging markets must overcome a disadvantage which yesterday's developing economies the now industrialized economies of the global north never had to face. Developing economies are not simply primitive versions of more developed ones. Now what's the answer to that? Prebbish and Singer argued one should try to move labour in developing economies into high income elasticity occupations. That has proved very difficult to bring about. Developing economies continued to be highly dependent upon natural resources and highly vulnerable therefore to fluctuations in the terms of trade which might disfavor them. Now of course they're not the only economies today which depend heavily upon resources for economic competitiveness but it's particularly a problem in economies which have yet to develop advanced manufacturing sectors. So dependency theory was far less optimistic about the chances for converges developed in developing economies. Then in the 1980's and 2000's gradually they developed the so-called Washington Consensus which had a different view again on this question. So now there was no longer the optimistic belief that development itself would create the conditions for labour law. Now there was the belief that the advanced industrial economies of the west had welfare state systems, had collective bargaining and social insurance, had job protection. But this mode of labour market regulation was harmful to them but would in particular be harmful to the developing nations. So what was needed in the developing world in particular was deregulation privatization certainly not Western European or North American style collective bargaining or social protection. So this was the idea of the World Bank, very clearly expressed in a doing business report in 2008. But going back to the early 1990's the so-called Washington Consensus, laws created to protect workers often hurt them. So in North America and Western Europe too the idea to hold collective bargaining would price workers out of jobs employment protection laws would bring into effect rigidities in the labour market which would undermine job creation, push up the price of labour and distort the way the market worked. This is also associated with so-called legal origin theory, the idea that the civil law countries, the French and German influence countries interfered more in the market and these pro-labour policies had a negative influence on employment and productivity. So a famous paper with World Bank support by Batero et al. in the early 2000's apparently found evidence, empirical evidence, that a high level of labour protection in individual employment law, unfair dismissal legislation, working time protection and also in collective labour law, freedom of association rights collective bargaining rights, protection of the right to strike all these laws were correlated with high unemployment and low productivity in the west in the north and in the developing world these laws were correlated with a high informal economy and with the inability of an economy to reach the Lewisian turning point where formal employment began to replace informal employment. Now this old consensus, I call it old because I think it's fading away, has nevertheless been very powerful and I want to discuss this consensus this evening. That's the old consensus I think, even the World Bank has begun recently to change its mind about this consensus. So they're doing business report for 2015 in an appendix tucked away at the back, maybe not much red but I read it, some of the people did too so they'll actually, labour laws do help as well as hinder, they may help economic growth and not having enough regulation can be as big a problem as having too much it's tucked away almost in a footnote but the World Bank itself is starting to recognise maybe the story pushed in the 2000's wasn't quite as simple and the evidence certainly isn't as simple as it was originally believed. In my earlier work I've argued that in fact labour laws are not just a response to industrialisation some are but not all and many of the institutions for the regulation of the labour market predated industrialisation in western Europe. So we had wage determination in the late middle ages we had also the beginnings of a social insurance system of course called the poor law and the poor law we associate with the workhouse with discipline and coercion of the poor but before the workhouse in England for many centuries almost there was a poor law in place which provided a type of proto-social insurance to wage earners and the term poor initially meant in legal terms it was a juridical term to refer to people who were dependent upon wages or subsistence. People who sold their labour power their capacity to work in return for wages were the poor the propertyless and there couldn't have been economic historians now believe an industrial revolution in Britain had there not first of all been mechanisms to deal with labour market risk a labour market is a market where we sell our labour power under conditions of propertylessness we as workers as employees are excluded from ownership of the means of production as Mark's pointed out we sell our labour power that's our most vital property in a capitalist economy and we are uniquely vulnerable as a result given our exclusion from the land from the extended family we're vulnerable to labour market risks. Risks initially which were physical health and safety risks risks which were economic the risk of unemployment the risk of loss of income through old age psychological risks risks of stress at work these are the risks which labour and social insurance law is there to deal with and the labour market can't exist without institutions for diffusing and controlling those risks and the poor law in England predated industrialization now a Pallani in perspective Car Pallani would say industrialization triggers labour law triggers a response that's partly true there's been a double movement capitalism a disembedding of institutions in the period roughly from 1760 to 1830 in England was followed by a period of re-regulation eventually leading to the modern welfare state but of course Pallani's point is that capitalism requires these institutions capitalism requires the state and it's not just human beings he said and natural resources only which are threatened by the unorganized or disorganized market but even capitalism itself needs to be protected from the self-regulating market and what he had in mind there was of course the role of the central bank Pallani writing in the 1940s after the Great Depression and in the middle of a war triggered largely by that Great Depression pointed out that capitalism and the market are not the same thing capitalism to survive he said firms need to be protected ultimately finance itself needs to be protected by the state by the institution of the welfare bank of the central bank and he compared the welfare state to the central bank these are critical institutions which enable capitalism to survive and we've seen that today we've seen for the past seven years since the crisis of 2008 central banks all around the world engaging in emergency operations to save the private sector to save the private banking sector and ultimately to save national economies so what we're seeing at every stage of the development of capitalism is a tension between regulation and the market the market is never completely free the market is legally constituted the labor market in which capitalists own the means of production and workers sell their labor power turns upon a legal juridical resolution of the fundamental tension between employer and employee under which the worker sells their labor power and the property rights and the residue from production investing the employer not in the worker but it's always been the case that labor law or its predecessors has protected the worker against the risks which inevitably arise from that highly unequal asymmetrical work relationship now when we think today about emerging markets and developing countries what can we say some labor law rules some labor law rules which have taken 200 years to emerge in the west or the global north may not be ideal for today's low income countries and middle income countries but I want to suggest that the argument that labor laws in principle are antithetical to growth in those countries is not well grounded and that not just employees but firms in those emerging markets need institutions for the diffusion of risk to provide insurance labor market risks assist market access and overcome information asymmetries and collective action problems firms employers need these laws as much as workers do well with Shelley Marshall and others I've tried to develop criteria by which we might assess how labor laws can contribute to economic development I fully accept that in many contexts it's inappropriate to apply the labor laws of an advanced economy to those of a middle income or low income country and I also accept that when thinking about labor law reform we can't ignore the lessons of economics about incentive compatibility for example but very often the lack of labor law creates perverse incentives both for employers and for workers and when I say labor law I include laws which very clearly protect the poverty rights and interests of the employer and protect employers from predation by the government as well so the generic criteria that we developed meant to provide a template for understanding how to fit labor laws into countries at certain levels of development so as I've said in England in the 18th century there already was a poor law, a kind of proto-social insurance and I think in many emerging markets today social insurance should be the first thing to consider before collective bargaining and certainly before employment protection when thinking about labor law reform because first and foremost the transition from an informal labor market where most workers are dependent upon either the family or own account work maybe in a family setting or where they still have access to the land the transition from an informal labor market to a formal labor market where workers do not have access to the land and generally do not work in a family or own income or own account setting that transition won't take place without insurance against labor market risks including unemployment and sickness so social insurance almost certainly comes first in the process of labor law assisting economic development but labor law doesn't just do that labor law is not just about risk distribution or about economic coordination although that's part of it but also about demand management the Keynesian argument that high wages are compatible with high growth and also with local growth indigenous growth is still important but labor laws are also about rights at work they're about democratization and labor laws can critically help emerging markets move from authoritarian regimes to becoming more established democracies because labor law is both about solving collective action problems but also about empowering workers both in the workplace and also more broadly in society so if you want to see democracy developing we should very strongly support labor law reforms in emerging markets now a case study let me now think about the empirical evidence to support what I've been saying a good case study to begin with is India and this shows I think just how difficult it can be to get an empirical grip on some of these issues but also how empirical research can help us to answer some of them in 2004 two prominent English economists Tim Besley and Robin Burgess published a paper about Indian labor law and they developed an index of state level labor laws an arithmetical measure to spend your disbelief you're nearly all lawyers but you can measure law an arithmetical measure of the degree of protection for employees and also trade unions in different Indian states so a lot of state level law determines these rights in India and they found that pro-worker laws were associated with reduction in investment and a reduction in employment in firms in the organised sector so the logic of this is quite clear protect workers and the result is less investment and also an inability to move from the informal to the formal because employers have very strong incentives to employ people on a casual or informal basis if they're going to have heavy labour law protections imposed upon them so they claim to show that pro-worker laws are correlated with under development and this paper has been cited thousands of times and has been used by the World Bank to develop its policy there's a lot wrong with this paper and a lot of what's wrong with it goes right back to the data which they produced which was then embodied in their statistical analysis because the legal coding data were full of biases and omissions which probably any labour lawyer could have spotted and indeed many labour lawyers complained about the data that had been used so this is a good example I think of what can happen when poor data are used in econometric analysis when the sources being used are not properly reported and also over-claiming from limited econometric evidence is I think a really huge problem in this literature because this study was essentially cross-sectional they weren't able to capture historical dynamics or change over time so when for example my former colleague Sonia Faganas entered into their regressions something which should really have been obvious all along which is that most labour laws in India exist on paper only and are not properly enforced when she put in data to relate to enforcement of labour laws many of the effects in the Bezlie Burgess study went away now Sonia's work is very important but of course maybe only takes us so far because what her work appears to show is that labour laws don't matter because they're very poorly enforced which is not exactly a great message for labour lawyers but I think this field has been moving on now a field of research which has been developing in Cambridge and elsewhere for the past 15 years or so has been given the name by one of my colleagues Matthias Seems of Leximetrics so this is quantitative approaches to the study of law we've been trying to get better measurements of legal data than those initially used by Bezlie Burgess in their study and also by the World Bank in its studies the patera at our paper I referred to earlier which was so influential I think as lawyers when we were faced with this new data we could have done one of two things we could have said there are many fundamental flaws in the data it's really impossible to do this to measure law accurately and then to put these legal data into an economic model this is fundamentally flawed that's one type of response which many lawyers adopted the other type of response is to try to do better than they did at coming up with some legal data so what I want to show you is what we did and how we went about doing it and try to defend what you may regard as the indefensible but I'll try all the same to explain how we did this we developed a series of indicators which attempted to capture the degree to which laws in various jurisdictions protected workers so we've constructed an index of labour laws which is built about of individual indicators so these are some of them there are 40 indicators in our index and they each try to capture on a 0 to 1 scale the degree to which laws protect workers by one means or another so some of these indicators as you can see here are about the way the law defines the employment relationship and some of them are about employee representation altogether we have 40 indicators some of them refer to working time some of them refer to dismissal protection some refer to employee representation and some refer to the right to strike each of these indicators tries to capture arithmetically the extent to which the law protects the worker so a score of 0 would indicate no protection for the worker and a score of 1 indicates the best possible type of protection for the worker but equally for the employer maybe these laws indicate high costs we'll explore that these indicators look like this or something like this when we put them into an Excel spreadsheet so what we have here are 40 or so indicators or variables and the data set is longitudinal so we have data for 45 years or so beginning in the 1970s now for 117 countries so the data I'm going to report from a slightly earlier study with 60 countries but we now have over 100 countries coded for every year for every indicator there's a score so as you can see there are literally thousands of data points in a very very wide and large data set this is telling us that in France for example variable 1 says how easy is it in France for an employer to get out of labour law by trying to present a wage dependent worker as if he or she were self-employed how strictly does the law define the employment relationship and is it a matter of contract or a matter of status which the law defines in France it's very very difficult to contract out of employee status variable 2 tells us whether part-time workers have the right to equal treatment with full-time workers and there's a certain score for that and so on and so on and so on this data set mounts up to provide data for over 100 countries for 45 years over 40 individual indicators describing different legal rules now those data can then be aggregated and weighted to produce an overall score for a country so a colleague of mine said this is rather like the question that battled everybody in the hitchhiker's guide to the galaxy you may remember that the answer to the question was 42 but nobody knew what the question was so you may feel that this is rather on the same level that we now know that labor law in France is something like 0.76 over a 45 year period well yes ok it's a bit like that but let me try to persuade you if you're not yet convinced that this is at least interesting right ok now many people say that you can't add up all apples and oranges you can't combine the data from all these different indices this may be an example of the pitfalls of our mathematical analysis well ok so we have to be very very careful when comparing the law on employee representation to the law on the definition of the employment contract rather as the inhabitants of Camp Nelson here adding a population of feed above sea level and so on and so forth obviously there's a real danger with this sort of analysis that we're mixing together different things that the weightings may be very artificial why have 40 indicators and so on and so on well of course there are practical issues in coding all these laws just think about what went into the creation of our data set so alongside the data set there's an explanations document which is 800 pages long which explains where these scores come from and the people who did the coding and there were only about half a dozen of us really did have to read all these laws well at least read about them and we had to read a lot of languages and be proficient in them or get help there were some very difficult languages ok finish is very difficult but most of that's translated into English right ok it isn't easy to create a data set like this and of course to certain point there are diminishing returns from trying to get more data and indeed the legal data may be a bit superficial now I think that's undeniably true however of any econometric analysis like this superficial not deep but superficial but broad picture a broad picture for over 100 countries for over 40 years but it can't be deep to go deep we need to do something else we need to do face to face interviews fieldwork all those things so this is one type of social science others it is at least interesting it'll tell us something about cross country comparisons but it isn't very deep but there's always a trade off I think of empirical legal research now whether Alfred Einstein actually said this is again a matter of speculation not everything that counts can be counted and not everything that can be counted counts so yes it's perfectly possible that what we've been counting isn't what matters and that indeed we can't count what matters that what we've been counting is irrelevant these are all possibilities that we have to take very seriously but I think that lawyers and social scientists do need to take quantitative analysis seriously I think there's undeniably a need for this type of analysis if only because in the past there's been so little of it so let me now then tell you something about the results that we're getting from our analysis so we do find and here we agree with the World Bank that civil law countries protect the worker somewhat more than common law countries and we do find that developed labour law developed country labour laws are somewhat more protective than those in developed countries which by the way was not what the World Bank found they appear to have found something else but they only looked at one year the famous material paper is for one year's worth of data only we at least have longitudinal historical evidence and when you put our data into charts you can then begin to visualise trends which are otherwise invisible so rather imagine a law library reduced to a chart okay that's what this is in a sense that very complex data lying in law books word after word line after line has been put into a chart which can demonstrate a time trend and I think this is interesting so we see that shareholder protection around the world we have a data set on that as well there's been more or less continuously increasing everywhere but especially in civil law countries catching up with the common law since the 1990s whereas in the case of labour law there's less of an obvious trend but protection has been increasing since the 1990s again in this field too and the civil law systems are consistently more protective than the common law system so that's interesting we also see that the gap between the developed and the developing world is not changing all that much so over this period in shareholder law shareholder rights has been an increase in all countries for shareholder protection largely driven by the need to attract investment but in labour law terms the developed world leapt ahead of the developing world in the 1970s and the state ahead transition systems is surprising but it seems to be the case are often the most protective and that's not just because EU systems have been exceeding to EU rules is also because some emerging markets including China as they have industrialized have been adopting pro worker labour laws briefly how does our data set differ from other widely used data sets the OECD Employment Protection Index is possibly the most used in economics and econometrics of all the data sets which are available what's interesting about this is that their data set is apparently based not on laws as ours is not on the coding of actual statutes and common law rules judge made law but to a large extent it appears on returns made by civil servants and others in ministries across the OECD now initially it appears it was all basically ministerial returns well recently the OECD says it has been using original texts and collective agreements but they do not disclose the source of these texts and there's no explanations document you can refer to to find out exactly how they've coded their labour laws so as you can see from this diagram we find the same sort of difference between Germany, Britain and the United States but the gap is a lot less in our data set between Britain and Germany than it is in the OECD case now Tony Blair the British Prime Minister of the 1990s and 2000s said around about 1998 that Britain had among the least regulated labour markets in the developed world and it seems to me highly likely that the low British score on this chart reflects more a governmental view than the actual state of British labour law which I can assure you on matters relating to employment protection is much closer to Germany than it is to the United States so I think we are getting more accurate coding and that's just one example now getting these data is of course only the first step the next step is to do some statistical column metric analysis and this is also difficult so here I want to invoke the figure an important intellectual figure of our time of Donald Rumsfeld who you may remember made an important remark about unknown unknowns in the context actually of the Iraq war modern statistics has the same approach to dealing with unknown unknowns a lot of time series econometrics statistical analysis of time series is about trying to deal with the fact that we just don't know what's out there most of the time the data I've been describing shine a very bright light it seems to me on one tiny part of a complex structure that's the big danger with this sort of research and a lot of what we're looking at a lot of the structure we can't see properly it's too dark we just can't observe the differences between countries there are many issues many questions we can't answer using this analysis and we have to assume that we don't know a lot of the time things that we'd like to know in order to carry out a statistical analysis of the sort I'm proposing if we want to correlate data on labour law with data on unemployment and employment and productivity we need to control for lots and lots of things at country level that we can't observe we don't know how quickly those things change over time we don't know the exact extent of cross-country difference or heterogeneity these are all fundamental problems so what panel data and time series econometrics do is they try to control for the unknown that's what they do now this you may say is also very unsatisfactory but I think the problem is it will never have enough data to carry out a completely thorough convincing analysis of the sort that would definitively resolve these questions it's a matter of technique and of using whatever economics can provide to minimise the risk of false results so in our work we try to use developments in time series econometrics many of these are new and also controversial and so therefore the results I'm going to present must be regarded as somewhat conditional but I think they're interesting so this is an analysis and sorry for those lawyers present who don't like dealing much with numbers and I can assure you that I have felt the same from time to time this is an analysis which demonstrates the effects of stronger labour laws in the area of flexible forms of employment different forms of employment on various economic outcome variables so what this shows in the top line is the correlation between strengthening labour law and its effect upon the employment to population ratio, labour force participation self employment, productivity and unemployment and also the labour share the labour share is the share of national wealth going to wages as opposed to profits or dividends essentially so what this is showing is whether an increase in labour regulation leads to an increase or a decrease in these other things and so a negative sign shows there's a decrease in the other variable and the asterisks show whether the result is statistically significant whether it's a false result or not so we can see from this as well that we put various controls into the regression there are controls for the level of GDP so we strip out the level of GDP and GDP growth in a given country so we ignore that so we say given how fast the economy is growing does labour law cause unemployment or not we also put in a variable we call freedom house freedom house is a conservative libertarian think tank which produces data on the rule of law in various countries so it gives us some indication of how well the legal system works in a given country and what do we find the top line shows when labour laws get stronger they're more protective for workers labour force participation goes up employment to population ratio goes up so more people are employees than self-employed self-employment goes down productivity goes up but it's not significant there are no stars next to that number labour share goes up significantly labour share goes slightly down so what this is saying is that when labour law defines the rights of part-time fixed-term and temporary workers when labour law makes it difficult to contract out of employment rights more people are in employment productivity goes up unemployment goes slightly down self-employment goes down when we do the same thing with unfair dismissal law employment protection law we find that these laws again are by as much and productivity goes down now what's the intuition here the intuition here is that when workers are protected they're likely to work more productively we get similar results for employee representation law also when employment is protected versus self-employment fewer people wish to be self-employed people prefer to have a job, a regular wage and some security than be self-employed so these are very challenging results because they appear to show that these laws don't have the negative effects often found for them and we're gradually rolling out this analysis to our larger data set the results are preliminary but they're completely consistent with earlier results we've had from the same set of studies with smaller samples of countries and we have similar results for employee representation law not just for employment protection law so in other words we get similar results for collective bargaining than we do here for as we do here for employment legislation now this of course doesn't mean that we as I say that we can straightforwardly begin transplanting western style labour laws into developing countries and it may mean that we've got a very superficial wide account of what's going on we need to get a deep account as well of what's going on in particular countries so we sacrifice width we sacrifice comparison and maybe we sacrifice replicability of results when we do case studies using face to face interviewing and field work in countries speaking to managers, speaking to workers face to face may often just tell us what those people think it's not really giving us data which other people can replicate because nobody can do again the interviews we have done we may be accused of bias in our sample or having a non representative sample of firms but on the other hand if we're doing field work especially in emerging markets we're getting a deep picture because the people we speak to can give us their account of a narrative a historical process involving transition involving the application of labour laws that we just can't get from this superficial econometric analysis so China I think is a critical case and we've been interviewing in China for the past three years but China has undergone of course a great deal of deregulation and privatisation since the 1980s China has become a market economy and for much of that time China functioned with a minimal labour law system one has to remember that China was moving from a semi-marketised system the so called iron rice bowl system provided workers as state owned enterprises with secure work but not with a meaningful wage in any sense and their work was directed there wasn't really a labour market because they were allocated to jobs by the state so at its height the socialist command system couldn't be described as a market system and China's transition is not I think a transition from stable employment to precarious employment because there wasn't really a waged employment system in place before we do see of course in this period enormous precarisation of work and growing inequality associated with the closure of state owned enterprises and we also see agency work and fixed time employment becoming more widespread but in 2008 China enacted a labour contracts act which attempted to reinvent the employment contract for China's emerging market economy so this law which was passed after a huge internal debate between labour lawyers and economists and others and was very strongly fought over ideologically and empirically but did in the end result in this formally the passage of the law which significantly realigned China with international labour standards relating to the individual employment contract there's still no freedom of association in China there is still one dominant state run trade union but there is certainly collective bargaining in China affecting hundreds of millions of workers and tens of millions possibly hundreds of millions of people have been drawn into collective bargaining in the past five years and perhaps many millions tens of millions have had their employment relationships formalised in the past five years so it's not true that the employment contract is disappearing globally and being replaced by casual work or by self employment actually formal employment is increasing in East Asia largely thanks to China's development is also increasing but more slowly in India so it's not the case that globalisation is leading to the replacement of the employment model by something else so these were very significant reforms reforms also to dispute resolution system so from 2008 China adopted a labour arbitration system which was a little bit like a labour court system with an arbitration element and we are now seeing hundreds of thousands of claims going to this hybrid arbitration labour court system from individual workers in China every year the high minimum wage enforced in the coastal provinces in particular in Guangdong now you may say what's going on in China is essentially a formal development what's actually happening on the ground our experience of interviewing is that certainly in the coastal regions in Guangdong and also near Shanghai in the Yangtze River Delta multinational firms and larger Chinese firms do find that in this effort labour laws are being enforced and they will complain about it they will say we can no longer compete as effectively with other emerging markets because it is state policy in these areas to enforce strong labour laws for workers protection it is state policy in Guangdong to raise the minimum wage and yes hundreds of firms in Guangdong did go out of business when the minimum wage was being raised in the past 10 years and despite the controversy around the Labour Contracts Act my assessment would be that this won't be reversed China will not reverse the pro worker reforms that were brought in in 2008 and in fact just the opposite there's a commitment in China to a wage led employment growth strategy now that progress may not be linear and it can easily be reversed and China isn't a democracy China isn't a place a country which recognises freedom of association for trade unions in the sense that we recognise it in the global north but nor is China still now a low wage economy China is becoming a middle income country and is becoming a country dependent upon wage and employment led growth I would like to suggest that development is a process which affects all countries has affected the global north has affected the west and maybe is still going on and simply something which is happening in China or India all countries undergoing the transition to capitalism are still in a sense in an early stage we're in an early stage of capitalism and labour law protection for the worker health and safety law, social insurance collective bargaining strike law employment protection law have been a critical part of capitalist dynamics and that's true of the global north and of course there have been periods when this linear progress has been halted and we've been living through such a period certainly in north America in parts of western Europe absolutely in the United Kingdom since the 1980s and I'm not suggesting that this process is seamless the process by which labour laws are established is political because labour law is about distribution is about power and labour laws are not self-enacting they have to be fought for by working class movements and political parties that fought for those laws by lawyers, by human rights activists who fought for them so labour laws have to be fought for and argued for but tonight I've tried to make the argument for labour law in economic terms not because I think that economics is more important than human rights but economics is an important part of this argument and winning the economic argument it seems to me is an important part of the story of labour and employment law which is yet to be written thank you very much I'll answer questions and perhaps we can take 10 or 15 minutes to do that and see how it goes Don McDoodle Professor, thank you very much for that in my labour practice I tried as a negotiator for capitalism and the employer side to get both parties interested in profit sharing and the sort of model that we thought was coming out of Sweden and perhaps some other countries so that the interests of labour and the interests of the capital simply the management were brought close together what if anything can you tell us about what's happened in that coming together if it's ever come together are there any jurisdictions even in the so-called western world where profit sharing will occur because I certainly wasn't successful in doing it here thank you very much I think collective bargaining itself is a form of profit sharing and collective bargaining can be associated with other forms of worker voice co-determination in Germany, in Sweden but not just there in other northern European countries that's the basic model and it has worked up to a point so if we if we're looking at the countries which are most productive where labour productivity is highest they are the countries with the strongest labour laws those which promote worker representation and which promote voice in the workplace and which also give the worker some guarantee protection against wrongful or unfair dismissal certainly in Europe they are now the debate about whether labour law is compatible with economic development has to confront the big exception to this which is the United States the United States is a high productivity country but doesn't have strong labour laws by international standards in fact they appear to be very weak outside the organised sector in the United States there's little protection and many of the most modern and innovative companies in the US especially in the high tech sector rely on this which is not at all compatible with the idea of collective worker voice and collective worker representation I accept all this and I think the American case is one which requires further analysis and maybe the Canadian case has something in common with the American one not completely but something but this North American model if I can call it that where there's often minimal individual employment protection and where collective bargaining is focused on the workplace not the industry or the sector but the exception worldwide in Europe collective bargaining tends to be sectoral so employers don't compete on wages as a floor of wages the state requires employers to invest in training in Germany there's legal obligation to train so the quick pro pro for high minimum wages in Germany is an obligation to train the state supports training in Sweden, in Denmark by active labour market measures all these things are of course of a social democratic system that's worked very well why doesn't it work in the United States that I think is a very interesting and important question but one country no matter how important an influential isn't the whole world and I think that we need to be very careful when extrapolating from the North American experience Sure My understanding is that Germany at one point had a law where it had to have so many workers on its board that still does but I thought that they eliminated some of that but let me do this lead the Germany part and thank you to the IT sector in the US where as you know the modern form of corporation there is pretty well leaving changing the role of shareholders all together the developers of the product of the IT are saying look at the international the technology are saying look we don't want to share our profits with all the shareholders that control the enterprise all the way my experience was that workers weren't prepared here to really take full risk that would not give them raises in bad time because they weren't directing the enterprise but nor did they want to direct the enterprise so my question to you I guess is do you see further development of the real profit of sharing ever coming closer to the point where capital and the capitalism and the workers interest will not eventually emerge but at least get a lot closer to that because capitalism is a system in which the workers don't own the means of production so workers are often opposed to proposals for share ownership because they know that share ownership doesn't really give them control because of the restrictions which have provided fiscal subsidies for share ownership the employees who get those shares don't have control of the enterprise and if they see their wages becoming completely dependent upon stock market movements they would justifiably say we're being exposed to economic insecurity without being given real power over the enterprise so the German co-determination model doesn't depend upon workers owning shares it has workers on the board with voice although not a veto and that's a co-determination model the executive board has both workers and shareholders but the shareholders always have the final controlling vote and the executive board is distinct from the supervisory board so the supervisory board can control up to a point only what the executive board does so the German model doesn't work because workers control the firm financially or economically it works because workers have voice at critical junctures in the decision making process both at board level and within co-determination of the firm itself so it's only one model but why don't we see worker ownership developing more generally within capitalism I think there are very good reasons for this I think worker owned firms often don't succeed or they operate only within particular sectors or they can't attract capital because they can't offer equity investors a stake in the company so I suspect the way forward is not to replace the employment relationship with some sort of capital type relationship for the worker to acknowledge what capitalism does it provides workers with access to employment and to a degree to stability but it's not the same thing as dissolving the capitalist firm and replacing it with some kind of worker owned firm the metric approach there on a state basis just because as a capital observer would say look at the difference between the round back company and the Kansas and this IP sector which developed in California the argument there in California is a much more regulated state than in Kansas there is a study which used our own data comparing the US to Germany Britain and France and a similar study by the same group of economists based at NYU that looked at California versus various American states so what they first of all found was that across country sample based on our data but they had data on patents they found that when employment protection rights were increased there was more patenting by high tech firms so the basic logic is when workers feel that they can't be summarily dismissed they're more willing to cooperate with the employer and to share knowledge and information which can be patentable now then the same team of economists looked at California versus other states now in California from time to time the law changes and becomes more strict and less strict even in California with its common law see periods when there's more or less protection against unjust or arbitrary dismissal so this study found and it's very interesting that when the law became more strict more worker protective patenting went up the number of startups of high tech firms went up and the number of patents issued by those firms went up and the number of people employed in those firms went up I think it's a very interesting study it's counterintuitive of course because people often just don't want to believe that employment protection laws can have this effect but this is a study by some very reputable financial economists and their own US study got to the same result as our well their own study using our international data now I can point to about a dozen other papers for various countries which show a link between employment protection and innovation so this will soon become a generally accepted stylized fact I think now it's a little good because you get high innovation, high productivity you may not get higher employment okay so you may be getting more unemployment because firms can do well with fewer workers because productivity goes up so this is a complex phenomenon but just in terms of the efficient use of labor I think it's going to be pretty clear that the optimal level of job protection is not zero my question has to do with the relationship between labour laws and development and I come from Ghana and I can see obvious examples of you know strong labour laws and investments on one side and then worker protection on the other because governments try to kind of reduce the strength of labour laws so that they encourage investments but I don't know whether generally industry is sensitive to strong labour laws because even in recent times in North America most industries are relocating to developing more countries because labour laws there are weak and so they can so many things like you mentioned environmental issues, safety issues and stuff like that so I don't know whether I don't know what I don't know where we're having because there's still an issue where industries here are moving out to like China, Mexico especially the auto industry because labour laws there are sort of weak so is there a time when this movement of industry because of labour law would result in I don't know it won't bring about kind of development but I'm just wondering where that is going and whether or not it is just a general issue that industry is always sensitive to strong labour laws rather than this dichotomy between developed countries and developing countries but I do understand that developing countries need to necessarily use the grip on labour laws to encourage some sort of I don't know going way back in time whether or not you can say the same thing for when say the West was developed what were the labour laws like at that time because there may not be a kind of strong comparison because we have developed countries today we have developed countries maybe then there was no that kind of comparison so I'm just going to rush into that I don't know if you'll take on that remember that Lewis argued that there would be a period of rising inequality as an economy moves away from subsistence to becoming a market economy the question is is that movement permanent or can it be reversed are we going to get to a tipping point where greater equality is established in part because of the political dynamic I was referring to capitalism creates a working class creates solidarity creates conditions for mobilisation for political change well only if clearly there's a democracy only if there's a certain critical mass which enables unions to form so it's not automatic that this takes place and of course today's developing countries and their multinational companies may have very strong reasons to suppress the growth of labour standards in developing countries there may not always be the strongest supporters of democracy in those countries that's right now how do we reverse that process the ILO used international labour organisation tried to play a role but the ILO is contending with a strong ideology from the World Bank and from the WTO for the past 20 or 30 years that labour standards impede growth and development but the ILO still does a lot of good work in this area and many labour standards are the result of the ILO's legislative and technical initiatives the ILO will also make the case as far as it can for the economic advantages of labour regulation and of formality in the labour market I'm not suggesting that there are not many situations in which high wage costs or high labour regulation may be cited by multinational companies as a factor driving investment we do know that there are many other factors driving investment multinational companies will benefit clearly from stable states with the capacity to ensure effective market exchange with a basic legal infrastructure that ensures protection of property rights and also with non-corrupt states and non-corrupt regimes in which those companies are not exposed to predation it seems to me that multinational companies need a strong state and need an effective legal system and a transparent and democratic one to function in the west and they take that protection for granted and they should be supporting the development of similar institutions in emerging markets because ultimately it's in their enlightened self-interest to support those initiatives and I think there are some enlightened companies and there are some enlightened governments there are many which are not but it's a complex picture Sir, as a social scientist and absolutely fascinated with his attempt to quantify the effects of laws I just had a minor methodological question for you and that is it was my impression that your data is based largely if not solely upon national legislation and yet in a number of instances especially for European cases you also have important labour laws on the supranational level for example work time directives so I'm just wondering do you utilise supranational legislation in corporate because it is important but if you do, how do you determine how do you keep the causality as being due to national legislation or something beyond national legislation Well the schools from each country are based upon the laws of that country and the law may be derived from a directive so we're not coding for the directive as such but we're coding for this implementation in a given country so of course you can see a pattern countries are adopting laws very frequently because a directive has been passed in the area for example of part time work or fixed term employment you can see most European countries coming into line with an EU standard but many of them of course had those laws as a directive so if you look at French and German law you can see before those directives were passed in about 1997 or so they were already moving towards equality between part time and full time or between fixed term and permanent workers so the simple answer to your question is the data set does consist of coded legal texts at national level now again you may say you're not coding for the law or for its implementation but my argument is that it's very very difficult to code for that on a national level partly because to enforce a law depends well because the way laws are enforced will differ according to what type of enterprise we're talking about and what conditions it faces and many firms will be paying more than or respecting more than the basic legal standard anyway so our argument is not that these laws are impute direct costs to employers our argument is that the data set indicates the weight of worker protective regulation in a given context and how that translates into economic outcomes is a complex process which would depend upon the interaction of many different forces not all of which we can observe in the regression model so the World Bank on the other hand tried to say we're measuring costs which I think is completely artificial for various reasons so there are there are real methodological difficulties which I don't want to underplay with this type of analysis and at the end of the day it can only be taken seriously if it's combined with other techniques I wouldn't like to put everything on this particular approach because I'm aware of its shortcomings but I do think it's an interesting attempt thank you you said that you think the US is really an anomaly it doesn't work within the theory or the conclusions and I'm just thinking about Canada but I know about Canada I'm wondering if it works in Canada either because in Canada most of our labor laws are provincial and when a company is coming to locate in Canada a big consideration of course is what are the labor and employment laws employment laws standards fairly consistent across Canada labor laws differ quite a bit so if a company comes and says where do I invest in the last 15 years and say well not in BC, not in Quebec maybe not in Manitoba or Saskatchewan probably Alberta or Ontario or whatever the case may be and the promise is where you say don't invest because of the labor laws or the provinces also that haven't been doing that well economically so I was really eager to see your data for Canada and how that fits in so can you comment on that? unfortunately the data for Canada are national and it's very very hard to code a federal country so what we've said is we will code for sometimes federal Canadian labor law sometimes for province level labor law particularly popular for provinces so most of the coding for the Canadian case it is complex maybe the most difficult country to code for this reason is a mix of federal Ontario and Quebec level labor laws and for the United States it's either federal collective labor law or it may be New York law for areas of law which are governed by the states for India again we code either federal law or the law of Maharashtra it is complicated so what I can't tell you is nothing about the intra Canada differences between the provinces and how that might affect investment again with regard to investment there are many many complex factors so we try to strip out all those factors and say is labor law the vital consideration driving economic outcomes it is difficult to control for all these other influences and I fully accept there may be many instances of investment being led by low wages but probably those investments are fairly opportunistic and may not last long there are certainly cases on the other hand of companies investing overseas in states they regard as stable and where there is a sizable domestic market for their products and those do tend to require wage led employment growth so why don't we draw things to a close and for those of you who have questions that you must ask so I'm going to we will stay for a bit but look thank you very much for coming out this has been a wonderful window on leximetrics and obviously it's a challenging but important field and I think that I'm in favor of evidence based policy making and this kind of information and this kind of thinking seems to me to be quite critical and it's been great to get this window on at simon and we're looking forward to the next couple of days for those hearty souls who are willing to go through this thank you very much thank you very much thank you very much