 |
||
        |
 |
|
 |
| GLOBALIZATION AND EMPLOYMENT |
| A Public Lecture by Jeffrey Sachs |
Globalization & Employment
Jeffrey Sachs
Director, Harvard Institute
for International Development
I have the honour, the privilege, and the very hard task, of discussing the implications of the emerging global economy for a number of issues that are central to the concerns of both the International Labour Organization (ILO) and the World Trade Organization (WTO). I do not believe that there is a single issue of greater importance in the world today than that of understanding the links between globalization, economic growth and employment.
In my brief remarks I shall try not to fall prey to the usual problem of academic economists which is well exemplified by the story of the two men who decide to spend a Sunday afternoon in a hot-air balloon. They take off and are flying over the lovely Swiss countryside when an unexpected storm breaks out. Caught in the storm winds they are blown kilometres off course. Finally, the storm ceases and they find themselves over a vast field. Seeing a man down in the field, one of the men in the balloon calls down, "Hey, mister, where are we?". The man on the ground looks up and says "You're in a balloon". At which point one of the men in the balloon says to the other, "that man must be an economist". "Why do you say so?" asks his friend. "Because only an economist could have given an answer of such precision but of such little usefulness." replies the first. My story is meant to warn you that probably nothing I shall say will be precise, given the scope and nature of the theme of my discourse, the limits of our knowledge, the paucity of the data, and our lack of experience with the global economy. However, I hope that, unlike those of the man in the field, my comments will be useful in providing some indications about where the ILO, the WTO, and the academic community currently are in trying to understand how to adjust to, and take advantage of the enormous potential benefits of this dramatic process of globalization.
My presentation has three parts. First, I will look at the issue of globalization and world economic growth. In particular, I will address the basic question of what globalization means for developed and developing countries in terms of aggregate economic performance and will examine what the evidence says about the economic policies that countries should undertake to take best advantage of the international economy. We have considerable knowledge about what works and what does not work in the international economy; about what produces growth and what does not. I will give you my own biased perspective on this and look forward to hearing your equally biased views since this is an area where each and everyone of us has his or her own point of view. Nevertheless, I think we do know something about how individual countries can best adjust in terms of overall economic growth.
The second topic I will address is what globalization means for income distribution, both within countries and internationally. Most academic economists think it means very little, which is odd since most of the rest of the world think it means a great deal. I find myself comfortably somewhere between those two views, believing that globalization has a significant effect on income distribution in our countries without, however, exaggerating the magnitude of that impact. I think that globalization is a key factor in lowering the relative wage of unskilled labour in the advanced industrial economies and that it has important implications for the division of income between capital and labour. This is also an issue which I intend to address.
Moving logically from issues with which I am really familiar to those I shall be grasping at, the third topic I intend to address is one which concerns the ILO most nearly, namely that of globalization and labour standards. I will examine what the international experience has been in this area and what data on growth, employment and income distribution suggest for this incredibly knotty question of labour standards. I will outline my own perceptions of this issue and will look forward to hearing your views and to learning from you.
Globalization and economic growth
Let me begin with the question of globalization and economic growth and tell you why I am basically optimistic about the implications of globalization for world economic growth and, even more importantly, for growth in developing countries, where I feel it will be of positive significance for the development prospects of the vast majority of the world's population that still lives at very low-income standards.
To do so let me first ask why we are experiencing this massive globalization. Why do so many countries proclaim policies of radical economic liberalization and international integration and enter the world market through policy change thereby affecting the lives of perhaps as many as 3.5 billion people all over the world? While technology certainly plays a role in this process and while the information superhighway, and improvements in communications and transport are undoubtedly important, the most fundamental factor underpinning globalization is policy change rather than technological change. What we are observing is a deliberate policy change reversing decades of fundamental policy error in much of the developing world. Developing countries are integrating into world markets which they had largely cut themselves out of for three decades, with dire consequences. It is a policy reversal which ends decisions taken in the immediate post-colonial era, at the end of the Second World War, or at the beginning of the period of independence; choices that led countries to the infamous import-substitution strategies, or autarkic development strategies, or, in extreme cases, to socialist strategies, all of which failed to produce sustained increases in living standards.
With this perspective as my point of departure, I believe that, for the first time in decades, 60 or 70 countries are availing themselves of sensible economic policies. By sensible I mean policies that are oriented towards the integration of poor countries into the larger global market-place through markets. This combination of global integration and market structure is, I feel, almost a guarantee of the rapid economic growth of developing countries assuming, of course, that the overall framework holds - that is, that political difficulties do not undermine this newly created global system.
At the end of the Second World War, the world really did divide into the First, Second and Third Worlds. At one point these terms were considered a cliche. Now, looking back analytically, I think that the concept of the so-called Third World must be understood as a description, not of living standards, but of the choice of economic systems, based on state-led growth supporting autarkic trade policies that prevailed in most of the developing world. Such systems were pursued in Indonesia under Sukarno, in India under Nehru, in Ghana under Nkruma, in Tanzania, in Egypt after the 1952 revolution, in Argentina under Perón, in Brazil under Getulio Vargas. They also prevailed in many other developing countries up to the 1980s. In the past two years the Harvard Institute for International Development has tried to separate out and enumerate developing countries on the basis of those which adopted this kind of state-led economic strategy, those which used Second World socialist strategies and those which pursued a First World, market-based, open trade strategy. We found that the vast majority of developing countries (which, depending on timing, were as many as around 70 of the 90 developing countries studied) pursued either the Second or Third World type of economic strategy after the Second World War and upon independence. Only 15 to 20 chose an open trading strategy. As late as the mid-1980s, one-third of the world's population was still in socialist economies and about 50 per cent was still in countries that had closed trading policies as measured by the objective criteria of tariff levels, inconvertible currencies, quota and licensing rates or official control over exports. For example, this was common throughout sub-Saharan Africa. What this analysis shows is that almost all of the handful of countries which were quick to adopt an open-trade strategy (again as measured by such objective indicators as tariff and quota levels, convertibility of currency and low export taxation) achieved sustained economic growth between 1970 and 1990 of above 2 percentage points per capita per year. In fact, all but two achieved 3 percentage points per capita growth on average between 1970 and 1990. The most telling fact is that only one of the open economy developing countries fell prey to an extreme macroeconomic crisis in the 1980s and this was Jordan in 1988, after the collapse of oil prices. On the other hand, of the developing countries which chose closed trading strategies virtually every single one fell into macroeconomic crisis during the 1980s, as measured objectively by either high inflation (that is, an inflation rate of over 100 per cent per year) or by a debt crisis as measured by default on the servicing of, or the need for an emergency rescheduling of foreign debt. Indeed, of the 73 developing countries that could be characterized as being closed in the 1970's, 59 fell prey to an extreme macroeconomic crisis in the 1980s and most of the rest did so by the early 1990s. What we are now observing is reform that has been forced by events because the corollary of the current globalization is that virtually not a single country in the world opened up before it went over the macroeconomic cliff of high inflation or debt crisis. Neither the exhortations of economists nor their own experience with low growth induced a single sub-Saharan African country or a single Latin American country to liberalize on its own initiative, but the debt crisis, fiscal bankruptcy and hyper-inflation were very persuasive arguments. Today we are witnessing the end of one era and the beginning of another because these countries which are in desperate shape, and which have reached a cataclysmic financial crisis have moved in the direction of internationalization, both because it was a prerequisite for aid and because it actually looked like the right thing to do.
All in all, if one were to compare the respective average growth performance of the closed and open groups of countries in this study, one would find that the open economies out-performed the closed by 3 percentage points growth per year on average for the 20 years between 1970 and 1990. The implication is that one cannot find an open developing country that failed and can hardly find a closed developing country that succeeded during this period. I think the reason is clear. In part it lies in the efficiency that derives from trade competition, from the gains from specialization and the like. Even more importantly, countries which cut themselves off from the outside world also cut themselves off from the world of ideas and of science, from technological flows, and from the organizational improvements taking place in other parts of the world. Their stand appears, with hindsight, to have been an almost wilful rejection of the opportunities available to poor countries to choose the best of what they could find in the rest of the world. I do not, therefore, find it surprising that the countries such as Singapore, Hong Kong, Mauritius, Cyprus and others which were always open to trade did vastly better, than those that were not. Being totally engaged in the international economic scene through trade, capital flows and the like, they were able to acquire the world's best ideas at no cost. This, I feel, is the fundamental gain to be derived from globalization - the fact that poor countries can take advantage of all the world's knowledge, both in its disembodied form and as it is embodied in capital technology. It gives them an unique opportunity to catch up with the more advanced countries without having to reinvent the proverbial wheel - or fax machine, cellular telephone, biotechnology or whatever area of knowledge is embodied in the goods that are engaged in international trade and financial flows.
Thus, my first point, which I hold quite strongly, but which I would be happy to discuss later, is that globalization gives developing countries an unprecedented opportunity to catch up with the more advanced economies. If one were to take the subset of open economies in the postwar period, there is sharp empirical evidence of convergence which denotes faster growth in poor countries so that the gap between the developed and the developing tends to narrow. However, if one were to look at the universe of all countries there is no convergence. This difference does not occur because poor countries get caught in poverty traps as though they were a technological fact of life, or that the rich countries, willy nilly, get richer because they have more power and strength in the world system. Rather it occurs because poor countries have tended to choose their economic systems badly and so trapped themselves in poverty. Now, with the reversal of their economic policies, they have an opportunity to catch up. I do not think it is an accident that the developing world is growing more rapidly on an annual basis today than it had in at least 15 years. In addition, 15 years ago growth was fuelled by unsustainable foreign debt inflows. However, in 1995, the average growth rate of the developing world may have reached 5.5 per cent and may well continue to inch up to 6 per cent or more. The developing world is growing faster than the developed world, as indeed it should. It also appears to be growing at faster rates than in the past, and on a sustainable basis given that this growth is internationally integrated and occurs in a context of growing export capacity.
However, it is evident that openness alone cannot constitute an engine of growth and I would like to mention a number of other factors which contribute to the rapid growth of the global economy. My remarks are based on evidence generated by my colleague Robert Lawrence of Harvard University in the context of a number of very important cross-country comparisons of growth and on our own subsequent modifications of that work. The basic hypothesis of that work was that growth depended on various structural variables and initial conditions as well as on a number of policy variables. Cross-country statistical data covering the past 40 years were used to try to measure the respective roles of structure, initial conditions and policy. The conclusion was that in terms of initial conditions, all other things being equal, poor countries tend to grow faster than richer countries. This is the so-called "conditional convergence" idea, according to which, if everything else is equal, there is "an advantage to relative backwardness" which is that of being able to attract physical capital, technology and ideas. This is strongly supported by the data.
Structural variables also play a role. For example, where a country is located and its geography really matter. One structural variable appears to be of paramount importance today and this is whether a country is landlocked or not. Having said this in Switzerland, let me hasten to qualify my statement. For a country to be landlocked, as Switzerland is, in the middle of the world's richest market-place is probably not a disadvantage. However, the evidence suggests that to be landlocked in the Andes mountains of South America like Bolivia or Paraguay; in Africa like Mali, Chad, Burkina Faso, Central African Republic, Burundi, Rwanda, Zambia, Malawi and Zimbabwe; or in Central Asia like Kazakstan, Kirgyzstan, Nepal, Bhutan, and Mongolia would create real problems for growth. All else being equal, I would estimate that being landlocked subtracts between 1.5 and 2 percentage points per year of economic growth potential because if the name of the game is trade-related growth, the cost of being stuck on the inside is quite high. Possibly this cost will be reduced when neighbouring countries themselves open up to world trade and let goods through at lower cost. However, the experience of the past 25 years suggests that being landlocked is a very great disadvantage.
Another structural variable which, surprisingly, appears to be a disadvantage, is that countries with abundant natural resources tend to grow less rapidly than resource-poor ones. For example Switzerland has mountains but it is very fortunate in that it has no oil, gold or diamonds. This is why it is one of the richest countries in the world. Similarly, Korea, Japan and Taiwan are blessed to have no mineral wealth. This is the best of all advantages in that they must depend on their human skills and on the development of human capital, unlike countries like Russia, Venezuela, Nigeria and Zaire which are cursed with gold, diamonds, oil and natural gas. I do not know exactly why this occurs but being resource rich probably uses up one per cent per year of economic growth potential as compared to being resource poor. There are two theories which try to account for this. One suggests that resource wealth leads to a fight over rents, and the tendency of resource-rich countries to become rent-seeking countries is very powerful. In my view it explains why Venezuela went from being one of the richest countries, or perhaps the richest country, in Latin America in the 1960s to one of the most confused, lost and ailing countries today. Its oil resources hindered the adoption of sensible policies of openness, macro-stability and the like.
Such initial conditions and structural variables are givens which governments cannot change. However, there are actions that governments can take. Evidence suggests first, that by far the most important area of government action is the adoption of an open trade policy. I feel this is the sine qua non of good economic performance on the part of the developing countries. It creates various disciplines as well as the flow of knowledge which is a crucial factor. I can think of no consistently open developing country which has failed to grow rapidly in the past 25 years, and I am always looking for counter-examples of closed economies which have been equally successful. Second, it pays to have high saving rates and every 10 percentage points or so of saving in GNP tends to raise the annual growth rate of a country by about one percent. Saving rates are partly, or perhaps even mostly, a result of government policy. First, having saving surpluses in the public sector is very important to achieve a high savings rate. The East Asian miracle countries all run saving surpluses in the government sector while most of the rest of the world has deficits on the government account. Second, it would appear that having pay-as-you-go social insurance and retirement systems, such as those in Western Europe and the United States depress private savings because government transfers substitute for household savings. On the other hand, having mandated private savings schemes that are individualized, such as the provident funds of Singapore and Malaysia, or the privatized social security system in Chile, tend to raise saving rates. The issue is basically one of whether citizens save for themselves or whether governments save for them out of government transfers. In the advanced industrial countries the government saves for its citizens and private saving rates tend to be depressed. In a few developing countries the government forces citizens to save individually. In many other poor developing countries there is practically no state system at all and they probably fall somewhere between the other two groups in these equations. Yet savings count. Third, it really does count to have low marginal tax rates. I believe this is not just the neo-liberal stand but is actually supported by international evidence. For example, the fast-growing countries of East Asia tend to have low marginal tax rates. It is a fact that Hong Kong has very low tax rates, as do Singapore, Malaysia and Indonesia. In particular, they have either very low payroll tax rates, or none at all. For example, Hong Kong has no payroll tax at all. Similarly, Singapore has an obligatory individual savings account that is quite high but no payroll tax that goes into the general budget. People save for themselves in individual accounts. Korea has a payroll tax of around 5 per cent now. I believe that if one were to compare the experience of these countries to that of Western Europe, where payroll tax rates are 25 to 30 per cent, or to Central and Eastern Europe, where payroll tax rates are 40 to 50 per cent, one would find the evidence necessary to demonstrate that high payroll tax rates discourage labour input, create incentives for black market activity, force withdrawal from the labour force and generally give rise to conditions of inefficiency that slow overall growth. Thus, small government accounts and low marginal tax rates do show up in the evidence as favouring growth while high marginal tax rates and high levels of government spending appear to depress growth.
The rule of law is also important. Happily, the evidence suggests that it is not just good for the heart but good for the economy. Surveys of businessmen indicate that governments that appear to the business community to be more corrupt have lower growth rates. As an aside, Russia must be proving the case right now given that corruption is probably of an order of magnitude that is greater than almost anywhere else in the world. In general then, measures of the rule of law do correlate with positive economic performance.
Let me turn finally to one of the main issues of this house - labour market flexibility - which evidence suggests is very important. All of the fast-growing economies of the world have flexible labour markets. By fast growing economies I mean those eight developing countries which achieved 5 per cent or more per capita growth per year between 1986 and 1994: namely Hong Kong, Singapore, Korea, Taiwan, Malaysia, Chile, Mauritius and Thailand. In all these middle-income countries wage setting is at the enterprise level. This, I believe is of tremendous significance. None of these countries have industry-wide, region-wide or national-level negotiation. Many of them have active trade union negotiations but they are enterprise-by-enterprise negotiations rather than industry-wide negotiations. I believe that the evidence shows that this is very important for the start-up of new businesses and for creating the conditions, not so much for existing enterprises, but for the growth of new enterprises. While I know that this is a controversial statement and hope we may discuss it, I do believe that the evidence shows that problems arise when collective agreements are extended across the board to a sector or a region thereby preventing market forces from operating to facilitate the start-up of new enterprises. This is probably the key to the real flexibility of these economies which are characterized by enterprise-level negotiation and low labour market taxation - that is low rates of payroll taxation, value-added taxation and personal income taxation which together represent a gap between the cost of labour to the firm and the real take-home pay of employees.
One finds other evidence of flexibility in the labour markets of the fast-growing countries. This includes flexibility in wage negotiation, in hiring and firing, and in the low rates of taxation which would otherwise distort the gap between take-home pay and the cost of labour to the enterprise. I will return to this issue when I examine labour standards because the way in which one views the appropriate standards or the kinds of flexibility required by developing countries to encourage the rapid and dynamic creation of jobs and to participate effectively in the international economy is of crucial importance.
Globalization and income distribution
Let me turn to my second topic, which is that of the effects of globalization on income distribution. Here there are three basic points that I would like to discuss. First, I would like to reiterate my belief that globalization will lead to higher overall growth rates for almost all economies and that there will not be a trade-off between faster growth for some and slower growth for others. Where distributional problems arise they are within income classes or between different skill levels but not between economies which grow more or less rapidly as a result of the international economy, with the obvious exception of countries that are disadvantaged by poor structures. For example, special measures would be needed to assist the poorest landlocked countries which do have an objectively difficult time keeping up with world economic growth. On the whole, however, the developing countries have a good chance of achieving convergent growth rates. In addition, if the developed countries follow the right policies, that is if they have flexibility, moderate rates of taxation and the like - something which is eluding most of Western Europe right now - they might also benefit from the global economy by being able to export their differentiated high technology products to a much larger world market. In sum, the issue of distribution centres, not on whether some countries gain and others lose, but rather on income distribution within societies. This is my first point.
My second point concerns the division of income between capital and labour. I would guess that the post-tax income of capital is privileged relative to the post-tax income of labour as a result of globalization and especially globalization that leads to openness of financial markets and not just of trade. For example, both evidence and theoretical logic make it quite clear that union wage premia are driven down by the openness of the world financial system and that the ability of capital to move offshore really does pose limits on the wage-setting or wage-bargaining strategies of trade unions which are restrained in their wage demands by the higher elasticity of labour demand. Similarly, I think that, over time, the evidence would show that the burden of taxation falls increasingly on labour and less and less on capital as a result of these changes given that taxation inevitably falls on the fixed factor and is, as inevitably, escaped by the highly mobile factor. At the end of the day, the fact that labour cannot move to the low capital income taxation countries suggests that we will find implicitly, both in terms of incidence, and in terms of choice of tax system, a movement towards a heavier burden on labour taxation and away from capital taxation in the taxation of factor incomes. Capital can still be taxed, not directly as a tax on capital, but indirectly through a tax on overall income or consumption. For example, movements towards progressive consumption taxation may constitute other mechanisms to tax capital income, if I am correct in assuming that the burden of corporate income taxation is likely to diminish given the increased ability of capital to escape taxation through international mobility. This is purely conjectural because the data has yet to be closely examined, but it is not inconsistent with existing evidence. It is also true, I hasten to add, that the direct evidence of income going to capital as against labour in the national accounts shows modest, rather than large, shifts in the direction of the share of labour falling and that of capital rising. My guess is that if one were to look at post-tax capital and labour income, one would find this trend even more strongly evident in the data.
The third distributional shift is within labour itself, between skilled and unskilled workers. Economic theory suggests that increased globalization will lower the relative wage of unskilled labour in the advanced countries and raise the relative wage of unskilled labour in the developing countries when these two groups begin to trade with each other after a period of autarky. This is the famous Stoker-Samuelson theorem, or rather an implication of it, or more correctly, of so-called factor price equalization. We now find ourselves in a very odd situation with respect to this most standard and central of all economic theories in that many of the leading theorists who propound it doubt that it is actually applicable to present circumstances. I have my doubts about their doubts. After studying international trade theory, including factor price equalization, with Professor Bhagwathi, I confess that I cannot just dismiss it. Although he contends that it does not apply at all to the international scene, my own feeling is that it does.
To begin, let me mention quickly the major caveat to the theory. If the developed and the developing countries have such unequal endowments - so much skilled labour in the advanced countries and so much unskilled labour in the developing countries - that they actually specialize, then factor price equalization cannot follow. Indeed, in such cases of specialization, being outside the cone of factor price equalization would mean that the increased export capacity of the developing countries would simply raise all incomes in the developed countries by bringing in the goods in question more cheaply. Thus, in terms of trade, the rich countries would enjoy an improvement that was a pure consumer gain for everyone.
As far as specialization and integration are concerned, it is probably true that there is more specialization than not. However, countries are not completely specialized. In the United States, for example, there is still a great deal of employment in labour-intensive, low-wage industry. Perhaps between a third and a quarter of all United States manufacturing employment is plausibly in direct competition with low-wage countries However, since this represents only about 18 per cent of the country's labour market, direct competition might involve no more than 4 per cent of American workers given that the vast majority of the United States labour force (more than 70 per cent) is in the service sector and another important segment is in manufacturing exports. As consumers and as export producers both groups stand to benefit from globalization. Thus, in thinking about the probable magnitude of this effect, my view is that increased trade has substantially eliminated jobs in textiles, apparel, footwear, assembly, toys and so forth, but that, surprisingly, it does not extend in any great extent to other industries. It was, therefore, somewhat of a surprise to hear Pat Buchanan bemoan the fate of American workers using textile and shoe workers as symbols of their plight. He was, in fact, really lamenting the disappearance of a very small proportion of United States manufacturing but not a great deal more.
However, this is the subject of considerable debate and, as I said earlier, I am somewhere in the middle on it. I believe that possibly about a third of the widening gap of around 15 to 20 per cent between skilled and unskilled labour is related to trade. This estimate is substantially lower than those put forward by Pat Buchanan, Ross Perot and James Goldsmith but it is considerably higher than those of Paul Krugman, Robert Lawrence, and a number of others. The data are imperfect, problematic, and difficult to evaluate. Many factors other than trade probably work against unskilled labour right now, the most important of these being technological change. Thus, with trade or without it, unskilled, poorly educated workers are disadvantaged in the labour market. However, trade theory and the evidence both suggest that globalization is adding to the effects of technology. While the negative effects of these factors show up in the United States as a widening of inequalities to an important, but not overwhelming extent, in Europe it shows up as greater unemployment because of the more rigid wage structures and labour market institutions which obtain and which do not allow trade competition to be met by lower wage jobs for unskilled workers. In Europe competition is met rather by firing unskilled workers and by the failure of the service sector to create new jobs for them, as happens in the United States. For example, in the United States, parking tickets are still handled by people; in Europe this job is performed by machines instead. The results of these two approaches are visible in the differences in relative factor prices in Europe and the United States and in the fact that the United States has created about 9,000,000 jobs in the past five years while Europe has created virtually none.
Ironically, as my colleague Don Davis of Harvard University has recently shown, Europe, because of its wage rigidities not only absorbs unemployment but actually boosts wages in the United States. One might even say that Europe is keeping American wage levels up to some extent because, in the face of competition from East Asia, Europe does the United States the favour of withdrawing its low-skilled labour from the labour market in the form of unemployment, reduced labour force participation and reduced labour hours. Thus, the world supply of low-skilled labour does not change as much as it might otherwise have done. This, in turn, indirectly boosts the living standards of low-skilled workers in the United States to the disadvantage of Europe. As I said earlier, this appears to be one of the ironic consequences of the differences in labour market institutions.
One of the most important and most difficult issues for analysis, not only for the ILO but also for the academic community is the question of whether changes in technology will mean that much more than just labour intensive manufacturing jobs will come into direct competition with lower wage workers. This is already taking place, for instance, in the software sector. For example, the back-office operations of American companies, such as accounting, are now being done off-shore in Jamaica; and Swissair vouchers are being processed in the Santa Cruz export processing zone in Mumbai, India instead of in Switzerland, as in the past. New technology and the ability to use dedicated satellite lines to transmit data is changing the division of labour. While, to date, its effect on aggregate employment patterns has been minute, it is really important that we understand whether its ultimate impact is going to be much bigger. For example, will Mongolia be able to use Internet to break out of its isolation and become a service economy for the rest of the world? Thinking of the longer term consequences of globalization we would need to know whether these technological possibilities are really going to open up new forms of sectoral competition that do not yet exist.
Globalization and Labour Standards
Finally, let me turn to the issue which I find the most difficult because I do not understand it completely. This is the issue of globalization and labour standards. On this I am going to speak from the point of view of a traditional free trade economist but will at least try to explain why I take the stand I do.
I feel that if we are to respond to the developments I described earlier we must look to better tax systems, or zero tax systems and other mechanisms, but not to (and here I know I use a loaded phrase) imposing minimum conditions of work or even institutional strategies for collective bargaining on developing countries. In my opinion the cost of such conditions and strategies could be quite substantial for the developing countries and bring modest, if any, gains to the advanced countries.
This being said, I would like to stress that there should not be any argument about the so-called core human standards, or core labour standards, if we could agree exactly on what constitutes such a core. There must, indeed, be core Conventions that are very strong, closely monitored and taken quite seriously on such issues as slavery and other forms of involuntary servitude or coercive labour, especially coercive child labour, on freedom of association, freedom to bargain, and the like. However, I would draw the line at dictating deep institutional details in these standards, particularly on such issues as the nature of the bargaining structure, and so forth. Their basic strategies should be comfortable for all because they really are the underpinnings of free societies. I do not think these standards are ideologically loaded between the left and the right and, as believers in individual liberties and freedoms, trade economists should be able to accept them just as much as everyone else. Where many of them demur is on other kinds of labour standards such as minimum wages, conditions of work, hire and fire terms, industrial relations systems, the nature of collective bargaining, and the like.
In this regard, I have a request to make of the ILO to which I know it may not be able to respond at a time of strongly diminishing or squeezed budgets but which I feel is really necessary. My request is that far more empirical research be done in this area. The truth is that there is almost no empirical work, positive or negative, on the implications of labour standards for employment, and I feel that the ILO could usefully bring out an annual publication which goes beyond just listing the standards ratified and the countries which have ratified them and actually includes substantive descriptions of the labour standards procedures and systems for each country on a comparative basis. For example, it would be useful to the research community to know, by country, whether bargaining occurs at the enterprise, regional, or national level; whether it is tripartite or bipartite bargaining; whether the trade unions are craft or industrial unions. It should also contain information on trade union density by country or by major sector of the economy; on the unionization of white-collar as against blue collar workers; on whether civil service unionization is permitted or not; on the degree of freedom in hiring and firing; on the restrictions governing reductions of employment; on notification procedures and rights of appeal; on labour courts and other possible challenges. This would be a major contribution to the labour field and would undoubtedly generate considerable research in this area. It is vital that the ILO takes the initiative in this as economists are unlikely to do so. This area is outside their terms of reference and they would not even know how to set about collecting the data. The ILO, on the other hand is the repositories of data for more than 150 countries and is ideally placed to put together such a publication.
Returning to the issue of standards and trade, I must stress that, given the lack of hard data, I can only make some generalizations. This, incidentally, is one reason why I feel that much of the rhetoric on standards is misplaced. In preparing this lecture, I read the wonderful collection of essays on international labour standards and economic interdependence brought out last year by the International Institute for Labour Studies. The book was an extremely good one and makes fascinating reading, but there was not a single quantitative fact in it. In other words, there was no empirical evidence at all. Indeed, one of the chapters actually pointed to that fact, this being the closest the book came to a discussion of empirical evidence. This is a very serious problem. The ILO has had an ongoing, 75-year old debate on this issue unsupported by any real evidence, whereas there are vast amounts of cross-country data for growth rates, tax rates, government spending rates and the like. These data did not exist ten years ago, but do now and are generating thousands of academic papers. If the ILO were to put out some good cross-country data on labour standards it would generate a similar outpouring of academic research examining the pros and cons of labour standards rather than unsubstantiated assertions, on both sides of the fence, of what works and what does not.
Having said this, let me now address the present assertions. My concern is based on the fact that I see negative reactions to fast-growing countries in the developing world which have very flexible labour standards. They are viewed in some quarters as being abhorrent because unions are sometimes repressed or are confined to company level rather than operating at industry or national level. In my view, the greatest damage to growth is in across-the-board labour standards that dictate either minimum standards or minimum conditions for higher and fairer wages or, worse still, provide for the extension of wages across the economy; in short, the German system applied to South Africa or some other developing country. I do not think that this system works well in Germany, or in the other Western European countries where it is present in that it hampers job creation. It would be all the more pernicious in developing countries.
Let me give you two examples. First that of India, which has wonderful labour standards and where firing is completely protected. Legally sanctioned employment reductions by enterprises needs local government permission, which is well-nigh impossible to obtain because the one rule of any politician is never to agree to lay-offs. This would be the most difficult decision for a politician to take in any country, whether the regime in place was authoritarian or not, but it is especially so in a democratic country. The result is that India has - and I do not mean to be glib in using the term - a labour aristocracy, a protected formal sector of perhaps 27 million people out of a labour force of 400 million, most of whom work for very low wages. Under such circumstances and especially without flexibility in reductions, the creation of new jobs is very difficult. In addition, it is equally difficult for a foreign investor to have confidence in such a system, to operate legally within it and hire new people. Thus hindrances to firing also hinder hiring and constitute major obstacles to the formalization of employment in India. Let me illustrate with a story which will bring home this point. The Harvard Institute for International Development is negotiating a joint research venture with the Delhi School of Economics. As part of this project the latter must hire someone to put macroeconomic data on the computer network once a month in India, as other parts of the network would do in their own countries. The research programme was easily agreed upon but the Delhi School of Economics contended that it could not hire the necessary research assistant for this time-bound job because the contractual obligation thus entered into would be permanently binding, even if the person in question was a graduate student. Their only option was to subcontract the job to some other institute with existing staff who were free to undertake the task! India is a country which needs to create tens of millions of jobs, and yet an offer, with the necessary funding, could not be accepted because of legal drawbacks. This is a vivid example of the impact of labour regulations, however noble, which raise the cost of labour beyond a market-clearing level.
Let me give you a second example which is very pertinent to tripartite bargaining. Right now, South Africa has an unemployment rate of 50 per cent among its black population. The wages of black unionized workers in the formal sector is roughly three times that of wages in the unorganized labour markets of Suweto, Johannesburg or Port Elizabeth. There is, therefore, a huge gap between the formal and informal sectors. The South African trade unions which were in the vanguard of democratization grew extremely powerful after liberalization and legalization in the early 1980s. They were able first to raise sharply the wages of their members and, subsequently, under the terms of the new Wages Act and the New Industrial Relations Act, they were also able to impose extension, geographically and by sector, to ensure that no other workers could be hired at below these rates which are three times the market-clearing rate. While strengthening extension through the new legislation makes political sense given the crucial role played by the unions in ending apartheid and winning democracy, it is, nevertheless, an invitation to disaster because South Africa has been without job-growth for decades now. In addition, such a step creates conditions which are untenable for the vast majority of the black population who will not be able to get first jobs and will be condemned to the informal sector.
This, in fact, is my major problem with tripartism in developing countries. The unemployed - that is the potential workers - are not at the bargaining table. It is very good to have the formal employers, the formal unions and the government at the bargaining table, but this does not hide the fact that the informal workers are not represented by anyone. The one thing the formal sector actors can agree on is that they do not want a competitive fringe in what is now the informal sector. This is the real conundrum faced by countries like India, South Africa and Argentina under their inherited labour codes. The same is true of Spain, where every lay-off is fought through the labour courts to the point that there is 23 per cent unemployment in that country. Yet, this does not mean that people are unemployed, it merely means that they are working illegally, again because of constraining legislation. In my view, it is crucial to allow competition and freedom of negotiation, but it is also crucial to allow a competitive fringe and to make sure that tripartism does not become a monopoly which rules out this competitive fringe to the comfort of the insiders. I fear that this is a strong feature of tripartism in many of the developing countries where I have seen it at work.
Five years ago this would have seemed an even more outlandish contention than it does now because people, including my colleague Richard Freeman, argued that Europe was doing all right, that the gaps were not so large, that the inflexibilities in Germany did not seem to be so severe and so on. I think we are seeing the end of that view right now. While I cannot speak for my colleague, I feel that, since Freeman published his work on labour flexibility, the evidence has suggested that Europe is paying a heavy price for its inflexibility in its inability to create new jobs. Given this evidence, the idea of imposing the same standards on others is much less attractive for two reasons. First, in general terms, it is inappropriate to go too far in imposing standards on other societies. Secondly, and more specifically, the debate should centre on the emerging doubt that these forms of labour standards actually work to engender improvements in human welfare, at least in the manner in which we would wish them to. Given such fundamental doubts, we should be far more reluctant to pursue the same standards in other societies.
Conclusion
To sum up, I feel that while there are difficulties with the process of globalization, winners vastly outweigh losers. I also believe that the issue of equity can best be addressed through government spending and taxation - that is, government spending to promote education, training, and benefits for people who desperately need them; and moderate levels of taxation which, nonetheless, have enough progressivity to allow proper taxation of high-income levels - and not through substantive labour standards or by viewing labour standards as the most efficient response, given the evidence that strong labour standards entrench insiders and de-privilege outsiders. This is apparent throughout the developing world where dualism is not an accident. While, to some extent, this dualism is inherent in development, it is greatly exacerbated by privileging and empowering insiders through negotiating systems which give them control over the overall structure of compensation and prevent the competitive fringe from forming and taking hold.
