In preparation for the Human Development Report every year, the HDRO commissions a number of experts to write papers on issues related to the theme of the Report. The following is a compilation of selected Occasional Papers written since 1992. Individually, each paper brings to light a key facet of human development in different parts of the world. Together, they help establish a framework of tools, concept and action to address the issue of human development worldwide.

Occasional Paper 32

Globalisation and liberalisation: Implications for poverty, distribution and inequality

Kevin Watkins
1997 

Introduction

Section One - Globalisation, poverty, and development

Part 1 - Elements of globalisation

• Trade, foreign investment, and speculation
  Globalisation and distribution
  Regionalism
  Broad distributional questions

Part 2 - The Uruguay Round: an assessment

• Tariffs
  Non-tariff barriers
  Special and differential treatment
  Textiles and clothing
  Agriculture and food security
  The distribution of benefits
  Policy considerations: a post-Uruguay Round agenda
  Box - The European Union

Part 3 - The international trade agenda: some missing issues

• Commodity markets
  Debt and trade
  Trade and Funding Development Assistance

Part 4 - Trade liberalisation, poverty, and distribution

• Trade liberalisation under NAFTA: the case of Mexican agriculture
  Liberalisation and poverty in Mexico
  Flexible labour
  Regional factors
  Trade and poverty: some concluding comments

Part 5 - Governments and markets

• Building domestic capacity: first - and second - generation NICs
  Quality and quantity in foreign investment

Section Two - Trade, poverty and employment: issues for the industrialised world

Part 1 - Unemployment and inequality

• Political background
  The US experience
  The European Union
  Wealth and poverty

Part 2 - North-South linkages: the evidence

• The skills gap
  Distributional considerations
  The technology thesis
  Monetarist orthodoxy as a barrier to employment
  The limits to flexible labour
  Policy options

Part 3 - Social clauses and trade agreements

• The case for a social clause
  An alternative framework
  Box - The NAFTA 'side-accord'
  The dangers of inaction

Section Three - New policies for new challenges

• Looking to the future
  The national level
  The international level
   

References

Introduction: The 'new order'

Some concepts come to define entire economic policy eras. For the 1990s, 'globalisation' will be recorded as the dominant theme. States are in retreat in the face of powerful international economic forces which, we are constantly told, are circumscribing their sphere of action. The resurgence of laissez faire economic theory celebrates the fact. While carrying different connotations for different people, globalisation encapsulates both a description of changing patterns of world trade and finance, and an overwhelming conviction that deregulated markets will achieve optimal outcomes for growth and human welfare. Seldom since the heyday of free trade in the nineteenth century has economic theory inspired such certainty - and never has it been so far removed from reality.

To the detached observer, noting the contrast between the presumed benefits of globalisation and developments in the real world, the international economy displays a number of worrying trends. Most obviously, poverty, mass unemployment, and inequality have grown alongside the expansion of trade and foreign investment associated with globalisation. In the developing world, poverty continues to increase in absolute terms, and the gap between 'successful' and 'unsuccessful' countries is widening. In the industrialised world, unemployment has reached levels not witnessed since the 1930s and, in some countries, income inequalities are wider than at any time this century. In a world of disturbing contrasts, the gap between rich and poor countries, and between rich and poor people, continues to widen. It is increasingly apparent that this reality will not be changed through growth alone. As the Pakistani economist Mahbub ul Haq once wrote:

In country after country, economic growth is being accompanied by rising disparities, in personal as well as in regional incomes. In country after country, the masses are complaining that development has not touched their ordinary lives. Very often, economic growth has meant little social justice. It has been accompanied by rising unemployment, worsening social services and increasing absolute and relative poverty (ul Haq 1976).

Written two decades ago, these words apply with equal force to one of the central myths surrounding globalisation: namely, the conviction that the expansion of the global economy is synonymous with an improvement in human welfare. Economists often explain the benefits of growth through the familiar parable of the boat, insisting that all vessels rise on a rising tide. The problem is that we are not all in the same boat. Some boats, containing perhaps four-fifths of the population in the industrialised world, around two thirds of the populations in the more successful East Asian countries, and the richest tenth of the populations in Africa, Latin America, and other parts of Asia, are floating steadily. Drinks are being served on the main deck. In the water around these vessels, other boats, containing the majority of the world's people, are either barely afloat or sinking. Some have already sunk, with their occupants floating in the water clutching flotsam for their survival. The challenge for governments is to create a fleet capable of carrying all of the world's people on a rising tide of shared prosperity and stability into the next millennium.

Some aspects of globalisation have also offered advantages which are more illusory than real. For instance, the extension of free trade and the integration of capital markets was expected to generate massive foreign investment flows which would support economic development, reduce poverty and enhance stability. In the event, investment flows have been highly concentrated in a small number of countries. Moreover, optimism about stability has been eroded by the chronic instability of unregulated financial markets. The volatility of these markets has revealed the inability of existing international financial institutions - notably the IMF - to create the conditions needed for a stable trade environment (Stewart F and Fitzgerald E, 1996).

National governments have been unable to respond to the power of global financial markets, which are now able to mount a direct challenge to the monetary sovereignty of nations. The main reason is the power of foreign-exchange markets. Each week the daily turnover on these markets outstrips the annual value of world trade flows. So far, governments have not developed a clear strategy for dealing with this phenomenon. As a result, their monetary policies are dictated not by employment and output objectives, but by a concern to appease corporate pension-fund managers and currency speculators. The full power of financial markets was highlighted during 1992 and 1993, when European currencies were tossed about on a sea of speculative activity to which central banks were unable to respond.

None of this is to deny that globalisation - or, more accurately, the deepening of economic integration between countries - has contributed enormously to the creation of wealth in some developing countries. The average income gap between south-east Asia and the industrialised world is continuing to narrow, albeit from extremely wide levels, with export growth driving economic expansion. In most countries in South-East Asia, the benefits of economic growth are being widely distributed, as witnessed by sustained reductions in poverty and improving human-welfare indicators. The problem, as we suggest in this paper, is that the policies which have unleashed this potential in the success stories of globalisation bear little resemblance to the neo-liberal idyll which now dominates economic policy thinking. In particular, they were based upon a dynamic interaction between states and markets, which many of the most vocal celebrators of globalisation in the World Bank, the IMF and national governments now reject (Streeten P, 1993).

Learning from history

Internationally and nationally then, the world economy of the mid-1990s is characterised by persistent poverty and widening inequalities. Current growth patterns appear more likely to exacerbate than to alleviate these problems, creating the risk of systemic collapse. This is a cause for international concern. If history offers one lesson, it is that poverty and inequality do not create a fertile soil for political stability, either at a national level or on the world stage. Recognition of this simple fact dominated the vision of policy makers in the post-war period. With the poverty, economic collapse, and international tensions caused by the Great Depression still a recent memory, the Bretton Woods conference in 1944 marked a concern on the part of governments to prevent a recurrence of those events. This implied a rejection not of markets, but of the notion that unregulated markets could create the stability needed for shared prosperity, or distribute the benefits of growth equitably. The spirit of the time was captured by one of the architects of the Bretton Woods system:

All of us have seen the great economic tragedy of our time. We saw the worldwide depression of the 1930s. We saw currency disorders develop and spread from land to land, destroying the basis for international trade and international investment and even international faith. In their wake, we saw unemployment and wretchedness - idle tools, wasted wealth. We saw their victims fall prey, in places, to demagogues and dictators. we saw bewilderment and bitterness become the breeders of fascism and finally of war. ( Dan Morgantheau, cited in Van Dormeal)

When the history of the 1990s is written, the words 'idle tools, wasted wealth' will be fitting accompaniments to 'globalisation'. Perhaps at no time since the 1930s has the need for concerted international action to regulate global economic forces been so much in evidence - and at no time since then has the vision of political leaders been so myopic. Nowhere is this more evident than in international financial markets. One of the main concerns of the founding fathers of the Bretton Woods system was to avoid a return to the unfettered capital movements which caused such havoc in the inter-war period (Marquand D, 1996). That is why Keynes stressed the need institutional structures which would reduce uncertainty and promote productive investment, while at the same time curbing speculative activity. Such structures seem more relevant today than ever before. Yet the IMF, which was created to fulfil the role of global financial regulator, has lacked the resources to act as a lender of last resort to countries facing problems, and it has failed to address the task of regulating international capital markets. Instead, it has used its policy influence to promote the very deflationary responses to balance of payments pressures which the architects of the Bretton Woods system sought to consign to history.

What is needed today is a vision as broad and ambitious as that which guided the Bretton Woods meeting. Unfortunately, policy making in the mid-1990s appears to be led either by the received wisdom of free-market economists, or by opinion polls. Governments have collectively failed to address the challenges created by globalisation. Instead of seeking to develop institutions capable of distributing global wealth more equitably, they have allowed public-policy choices to be circumscribed by the presumed diktats of the global market. Economic forces are running ahead of political responses. The same is true at the national level. No government, so the argument runs, can seek to maintain welfare states, pursue full employment, or protect basic social rights in a global economy where capital is free to roam the world in pursuit of the largest profit margin. This approach is all-pervasive - and it is misplaced. While it may be true that governments are less powerful today than in the past, such defeatism is not justified by the facts of globalisation. States are not rudderless boats driven by powerful currents of private capital flows, and they retain the power to shape policies in the public interest.

Winners and losers

For developing countries, globalisation and greater openness to trade is widely perceived as being doubly blessed. As for the developed countries, improved access to markets and competition from imports are seen as sources of improved efficiency. The additional benefit is that enhanced trade is supposed to open the door to increased specialisation in labour-intensive goods. Liberalisation is thus not only good for growth, but a means of reducing inequality through increased demand for labour, the main asset of the poor. So deeply rooted is the conviction that globalisation is beneficial for human development that the World Bank has developed a set of 'integration indicators'. These purport to confirm a close and mutually reinforcing correlation between growth and integration into world markets. To cite the 1996 Global Economic Prospects report: "policies that are good for growth are also apt to be good for integration" (World Bank, 1996).

It follows from this that the central role for governments is not to regulate markets, but to facilitate their relentless expansion by removing barriers to trade and investment (Ghai D and Alcantara C, 1994). National governments have been adopting this approach with enthusiasm, especially in the developing world. At an international level, deregulation is being pursued under the auspices of the World Trade Organisation (WTO). The Uruguay Round agreement, now being implemented under the WTO, was widely regarded as a triumph for multilateralism - and a decisive step towards the creation of a globalised international economy, in which the benefits of free trade could be distributed to all countries. Developing countries have been identified as among the prime beneficiaries, with important gains for export earnings and poverty reduction widely anticipated (Safadi and Laird, 1996).

Such projections rest on highly exaggerated claims about globalisation and its presumed benefits. Important changes are under way, but the global economy is not a construct of the late twentieth century. Nor are its benefits as self-evident as is often assumed. In this paper, we question some of the assumptions underlying the unbridled optimism about the capacity of unregulated markets to sustain growth and address problems of poverty and inequality. The aim is not to challenge the argument that the economic forces associated with globalisation have the potential to enhance human development. There are enough success stories to establish this fact beyond reasonable dispute. Our argument is rather that, under existing international trade and finance rules, globalisation is marginalising some countries, and actively threatening the livelihoods and welfare of vulnerable communities. This was true of the 1980s, when market liberalisation emerged as a dominant political and economic force under the auspices of IMF-World Bank adjustment programmes (Berry 1996; Stewart 1993) - and it remains true of the 1990s.

Part of the evidence is to be found in the declining share in world trade and investment suffered by many of the world's poorest countries. The forty-eight least-developed countries now account for less than 0.3 per cent of world trade - half the level two decades ago. Marginalised in trade these countries are also being bypassed by private capital transfers, which have now displaced aid as the main conduit for North-South financial flows. This pattern of distribution is not accidental. Inequality in wealth mirrors a deep inequity in the rules governing world trade and finance, which have been structured around the interests of the most developed countries. The Uruguay Round agreement has not changed this picture. Issues of vital concern to many of the world's poorest countries - notably the management of primary commodity markets and debt - were conspicuous by their absence from the Uruguay Round agenda. Other problems were inadequately addressed, including the problem of industrial and agricultural protectionism in the developed countries. The distribution of benefits from the Uruguay Round agreement reflects these realities, as we argue below. Meanwhile, traditional North-South divides are being reinforced by growing inequalities within the developing world, with sub-Saharan Africa falling further and further behind.

There is a parallel problem which is seldom discussed in the literature on globalisation. For much of the post-war era there has been a political consensus in the industrialised world that international trade expansion has positive outcomes for employment and income levels. The slowdown in international trade growth in the 1970s and a parallel rise in unemployment reinforced that view. Today, the consensus is breaking down. Open unemployment in the OECD countries now affects 34 million people - ten times the average for the 1960s. At the same time, income inequalities have widened in some countries to levels not witnessed since the 1930s. Increasingly, globalisation is regarded as a causal factor behind these trends, with competition from low-wage labour in developing countries cited as the main culprit (Marquand, 1996).

While such concerns are as old as trade itself, the difference today is that powerful political alliances have emerged, linking mass unemployment, income inequality, and poverty in the industrial world to trade with developing countries. These alliances are growing partly because of the failure of mainstream political parties to offer compelling alternatives to their populist message; and partly because they reflect deeply-rooted concerns. For poor communities in Europe and North America who are suffering the effects of unemployment, rising wage inequality, and increasing poverty, sermons on the long-term benefits of globalisation offer little comfort. By contrast, the overtly protectionist alternative offered by 'anti-free trade' alliances (see Goldsmith, 1995), ill conceived and potentially damaging to human development in the poorest countries as they may be, have an obvious appeal.

This paper is structured as follows. Section One outlines the forces associated with globalisation and considers their implications for poverty, inequality, and development. Section Two examines the debate about the impact in the industrialised world of trade with developing countries. This section concludes with a review of the cases for and against a social clause in international trade agreements. Section Three examines some of the policy options needed to underpin more equitable patterns of globalisation.
 

Section One: Globalisation, poverty, and development

Part 1: Elements of globalisation

According to one school of thought, the globalisation of economic life marks a watershed in history. Nation states are seen as an anachronistic left-over from a bygone era, their sovereignty eroded by vast flows of goods and finance (Opmae 1994). The prime movers of the new order are transnational companies, operating in a borderless world linked by global production and consumption systems (Oman 1996). In the words of the former US Labor Secretary, Robert Reich, "each nation's primary political task (is) to cope with the centrifugal forces of the global economy" (Reich 1993). Such assessments raise obvious questions. Most obviously: what is new about globalisation? After all, the salient processes identified with globalisation pre-date the 1990s. Each decade since the 1940s has been marked by the evolution of stronger interdependence. Trade has been expanding faster than output since the 1950s, foreign investment has been growing since the 1960s, and international financial markets started their dramatic expansion in the 1970s. Each of these trends had contributed to the emergence of a global economy long before the term 'globalisation' became fashionable.

In many respects, the idea that the globalised world economy is a recent product is a conceit of the late twentieth century. From the early days of the industrial revolution, manufacturers were concerned to create global markets. The company Dombey and Son in Dickens' novel of the same name memorably believed that "rivers and seas were formed to float their ships", and the Utilitarians who Dickens derided saw trade as part of a "civilising mission". On a more self-interested level, the textile mills of Lancashire depended on foreign markets to absorb over one third of their output by the end of the nineteenth century. The establishment of the great trading companies formed in the seventeenth century - such as the East India Company and the Royal Africa Company - colonialism, and the opening of the Suez Canal, and the Union Pacific railroad, were all defining moments in the creation of a global economy (Krugman P, 1995). By the 1920s, steamships and railroads had created markets for standardised and globally traded goods such as wheat, wool, and textiles. While it is true that the informatics revolution and improved communications are playing an increasingly important role in reducing the economic space between nations, the differences between steamship and aeroplanes, and between telegraphs and computers, is arguably of more quantitative than qualitative importance.

Even the most basic economic indicators used to gauge globalisation point to the need for caution in evaluating the present one. Measured as a proportion of GDP, trade in merchandise is no greater than it was before 1914 (Hirst and Thompson 1996). During the inter-war years, the share of world output that entered into international trade declined; it did not recover until after 1950. Developments since the mid-1980s mark a continuation of the trend which resumed at that time, rather than a distinctive break with the past. Much the same applies to capital flows. As a proportion of national income, capital flows from Britain at the turn of the century represented a larger share of GDP than for any major industrialised country in the 1990s. Such facts do not mean that there are no new forces driving globalisation. They do, however, point to the need for the more specific features of these forces to be defined.

Trade, foreign investment, and speculation

Although globalisation is not a new phenomenon, over the past decade it has been given a new impetus, and taken new forms. Most obviously, trade has re-emerged as a dynamic force for economic growth. In 1960 the share of trade (measured as the average of imports and exports of goods and services) in the GDP of the industrialised countries averaged 12 per cent; it is now over 20 per cent. The growing economic importance of trade is especially evident in the US, where the share of trade in GDP has almost tripled to 12 per cent since 1960 (Krugman 1995). In parts of the developing world, trade growth has been even more impressive. For East Asia it now represents around half of GDP. Even China, virtually isolated from the world economy a quarter of a century ago, now exports 25 per cent of GDP.

Recent interest in trade as an engine of globalisation reflects its recovery since the mid-1980s. For a decade after 1974, the rate of increase of world trade growth fell to 3 per cent, from an average of over 5 per cent for the previous twenty years. The ratio of world trade growth to output growth also fell, from 1.6 to 1.2. In the decade since the mid-1980s, that ratio has climbed to 2.8. (World Bank, 1995). Even though trade growth remains substantially below the 1964-1974 average, during the 1990s merchandise exports have expanded at three times the rate of output. From a post-war perspective, the rate of increase in the ratio of trade to GDP marks a return to the rising trend of the quarter of a century up to 1974, during which it climbed from 7 per cent to 15 per cent (WTO 1996) of global output. It remains to be seen whether this trend will be maintained. But if it does, it will deepen the economic interdependence of all countries.

While the rising ratio of trade to GDP marks a return to the trend interrupted between 1974 and 1984, private capital flows have been expanding at a rate which is unparalleled in the post-war period. Flows of foreign direct investment in 1994 exceeded $220bn, a four-fold increase over the nominal level for 1981-1985, compared with an increase of a little more than one half in the value of trade in goods and services. Between 1991 and 1993, the world stock of foreign direct investment grew about twice as fast as world-wide exports (World Bank 1994).

Transnational companies

As the principal agents linking cutting-edge technology to low-cost labour, transnational companies (TNCs) are one of the driving forces of globalisation. There are an estimated 37,000 TNCs, controlling four times as many affiliated companies. Over 90 per cent of these companies are based in the developed world. Collectively, they generated sales of $4.8m in 1993, an estimated one third of which was conducted on an intra-company basis. Foreign- investment activity is dominated by a core group of around 100 TNCs, with the largest one hundred accounting for one-third of the $2.4 trillion in global investment stock (UNCTAD 1995b).

The recent boom in foreign-investment activity reflects a process of corporate restructuring. Increasingly, companies are re-ordering their production on a global basis, establishing a presence in fast-growing markets, and shifting production from high-wage to low-wage economies. The resulting trade and investment flows have accelerated the movement of goods, services, and investment across national frontiers, reducing the economic distance between nations - an important aspect of globalisation (Oman C 1996). In many respects, however, this is also more of a quantitative than qualitative change, since increasing economic interdependence has been a feature of the post-war economy.

If there is a single defining feature of globalisation in the late twentieth century, it is the increasing ease with which technology can accompany capital across borders. This shift threatens to break irrevocably the link between high productivity, high technology, and high wages. It is now possible for transnational companies to combine through their investment activity high productivity, high technology, and low wages (Harvard Business Review 1993). Once again, however, it is important to set this development in a broader context. Comparisons of wage levels unadjusted for productivity differences are particularly misleading. Thus average hourly labour costs vary between $12-25 dollars in the industrialised world, with Britain at the lower end of the spectrum and Germany at the higher end. But if real wages are measured against productivity, wage costs for the US (where hourly rates are around $16) are lower than for the Philippines (average hourly rate less than $1) (The Economist 1995). This productivity gap reflects a wide array of factors, ranging from skill levels to infrastructure and access to capital. But while differences remain, there are already signs that the productivity gap is narrowing. For instance, Mexico's productivity per worker has risen from one fifth to one third of the level in the USA between 1989 and 1993, in part as a consequence of increased foreign investment geared towards production for the US market. Meanwhile, the average wage gap has narrowed far more slowly, with the Mexican wage still only one sixth of the US level (The Economist 1994a).

The changing structure of labour market competition between North and South is part of a broader picture. Four decades ago, product markets in international trade were largely segmented. Broadly, developing countries produced unprocessed goods, while the developed countries monopolised the export of manufacturing. Labour markets were similarly segmented, with the most productive technologies being utilised in the high-wage economies of the North (Stewart, 1994). This model began to break down in the 1960s, as developing countries emerged as competitors in labour-intensive manufacturing markets. During the 1980s, however, more profound shifts occurred, as the informatics revolution made technologies increasingly transferable between countries, and as barriers to investment were removed. It is these changes which have made it possible to link the most productive technologies with low-cost labour. To take one illustration, Ford's plant at Hermosillo in Mexico (see below) has productivity levels which are comparable with those of the most modern plants in Detroit (Carillo, 1995).

Outward investment

Recent trends in US investment illustrate the profound impact of globalisation. In 1994, the US outward stock of foreign investment reached a record 9 per cent of GDP, as American TNCs established a growing presence in foreign markets. The North American Free Trade Association (NAFTA) has been a focal point of this investment, with the level of US investment stock in Mexico increasing from $8bn in 1989 to $16bn in 1995. The linkages between trade and investment flows are much in evidence, with one quarter of US-Mexican trade now being conducted on a intra-firm basis. Restructuring through direct foreign investment (DFI) has gone further in the US than in the other OECD countries. World-wide sales by the foreign affiliates of American TNCs amount to 250 per cent of US exports, well over double the industrial-country average (UNCTAD 1996a). This suggests the scope for an accelerated drive towards globalisation in the medium term as Japanese and European TNCs 'catch up'.

The process has already started. At present the total flow of DFI from Japan is only slightly more than US flows to Mexico. But many Japanese firms see shifting production overseas, notably to South-East Asia, as a strategy for penetrating new regional markets and expanding exports to Europe and the US. In 1994, Japanese investment in overseas plant reached the equivalent of almost one quarter of domestic investment - and that ratio will rise. Similarly, over half of all companies in Germany, which has a ratio of overseas sales by foreign affiliates to domestic exports less than one quarter of that of the US, are reported to be planning a transfer of production to other countries (Financial Times, 8 December 1996).

The phenomenal concentration of corporate power within the global economy, and the economic exchanges which underpin it, provide an important pointer to the second decisive feature of globalisation in the mid-1990s: namely, the production of goods in a growing number of stages, adding a small amount of value at each stage (Krugman P 1995). In the classic model of heavy industry during the inter-war period, the components making up a Ford motor car were assembled in one integrated factory in Detroit. Today, the components in a typical Ford gear box will have passed through three or four countries. The same applies to everything from fridges to computers and garments. Fifty years ago, exported consumer good typically would be transferred between countries only once. Today, it is exported many times. Goods produced in one country may be assembled from components produced in other countries, which in turn comprise sub-components produced in yet other countries. This segmentation of production is both a cause and effect of the increasing mobility of capital and technology.

Currency speculation

Foreign-investment flows have overtaken foreign trade as an engine of world growth. Through such investment, TNCs are exerting an ever-more powerful and visible influence on the future of countries and their citizens. Less visible, but infinitely more powerful, are the world's financial markets. The deregulation of capital markets, the development of a wide range of financial products, cheap telecommunications, and computer equipment have fundamentally shifted the balance of power between governments and financial speculators. In the mid-1970s, the daily turnover of foreign exchange in the world's money markets amounted to around $1bn. Official currency reserves were equivalent to around 15 per cent of this total, giving governments considerable power to counter speculative activity. Today, daily turnover on foreign-exchange markets has reached $1.2 trillion, having doubled since 1989 (Financial Times 1996). Official currency reserves now amount to less than 1 per cent of this total. This profound change is another defining feature of globalisation.

In the past few years, there have been some striking demonstrations of the inability of governments to withstand speculative onslaughts. The breaching of the Exchange Rate Mechanism in 1992 and the collapse of the Mexican peso in 1994 are the two most powerful examples. Both episodes demonstrated that markets can change their views with astonishing speed, and that even the most powerful central banks are unable to resist concerted attacks. During 1992-1993 the Bundesbank spent some $130bn in defence of the currencies of six of its EU partners. The combined IMF quotas of these countries was $23bn, underlining the increasingly marginal role of the IMF in stabilising markets. So far, governments have failed to develop policies capable of controlling these markets. Instead, they have resorted to increasingly restrictive monetary and inflation targets, with interest rates geared towards their attainment. The criteria for economic and monetary convergence in the EU reflect this approach. In consequence, the expansion of output and demand has been relegated to the policy back-burner, with disastrous consequences for employment in the industrialised world (UNCTAD 1996c). Developing countries have also suffered from the depressed state of Northern demand for their exports.

Globalisation and distribution

Globalisation is revolutionising economic relations between countries. But it is a revolution built upon powerful elements of continuity, as well as change. Developed countries, representing 20 per cent of the world's population, typically account for between three quarters and four fifths of foreign investment, and for a similar share of world GDP and exports. The Group of Seven countries alone account for half of world trade flows. These shares have changed only marginally over the past two decades - and there is no indication that new patterns of globalisation will erode this concentration of power.

Within the developing world, there has been a widening inequality in the distribution of benefits from international trade. Overall ratios of trade to GDP have fallen in 44 countries over the past decade (World Bank, 1996). Sub-Saharan Africa has the lowest and fastest- declining ratio. A further 17 countries experienced only modest rises. At the other end of the spectrum, the ratio of trade to GDP has increased by over 1 per cent a year for over two decades in East Asia. These changes matter, because trade is becoming an increasingly important engine of growth - and because over one billion people live in countries which are being left behind by that engine, decoupled from the rise in average incomes.

Export growth-rates in excess of 9 per cent per annum have widened the already large gap between East Asia and the rest of the developing world. In the first three years of the 1990s, per capita incomes in East Asia rose at three times the average for all developing countries, and seven times the average for sub-Saharan Africa. Thus while trade expansion has enabled a significant number of developing countries to narrow the gap between themselves and the developed world, albeit from a low starting point, others are falling behind. For instance, the 48 least-developed countries (LDCs) have suffered a steady decline in their share of world trade since the 1970s (UNCTAD 1996a). In 1993, these countries accounted for a mere 0.4 per cent of world exports - almost half the level in 1980. This helps to explain why the poorest fifth of the world's population have seen their share of world income decline by almost one quarter since 1960, to 3.6 per cent of the total. Within this broad picture of East Asian success and LDC failure, there is a wide range of variations. Even so, trade is becoming an increasingly important engine of inequality as well as of growth. Globalisation is unlikely to reverse this trend towards marginalisation, and it may accentuate it.

In the case of manufacturing trade, the share of industrialised countries in global value-added has fallen since the 1960s, but only to 80 per cent. Moreover, almost all of the shift in manufacturing activity has occurred in east and south-east Asia, whose market share has doubled from just over 4 per cent in 1970 to 11 per cent in 1995. After growing strongly in the 1970s, Latin America's share has declined over the past 25 years, and sub-Saharan Africa's already tiny 0.6 per cent share in 1970 has been halved. According to UNIDO's projections, these trends will continue into the twenty-first century, with East Asia and China almost doubling their world market-share over the decade to 2005, and Latin America and sub-Saharan Africa continuing to stagnate.

Foreign-investment gaps

In part, trends in manufacturing-export performance can be traced to direct foreign-investment activity. As a group, developing countries have shared in the rapid growth of foreign investment, accounting for 37 per cent of total direct foreign investment in 1994. Between 1990 and 1995, total private capital flows to developing countries almost quadrupled to $167bn (UNCTAD, 1996). The largest element in these flows has been direct foreign investment, which reached $90bn in 1995 and is now the single largest source for financial transfers to developing countries. However, just ten countries received over three quarters of all transfers, with China alone accounting for more than one third. Thus foreign-investment resources are being concentrated on those countries - such as Thailand, Indonesia, Colombia, Malaysia, Taiwan - which are performing most strongly in international trade. Eight countries that account for 30 per cent of developing-country GDP absorb around two thirds of total DFI flows. At the other extreme, the 48 LDCs received around $800m in foreign investment in 1993 - roughly the same size as flows into Brazil, and less than 1 per cent of total transfers to developing countries (UNCTAD 1995).

One recent study has attempted to estimate the spread of foreign investment between the world's citizens, while correcting for the size of the Chinese population (Hirst and Thompson 1996). Taking together the industrialised 'triad' of North America, Europe, and Japan and adding the eight Chinese coastal provinces and Beijing, the study conducted that around 28 per cent of the world's population receives over 90 per cent of foreign direct investment. In other words, two thirds of the world's population is virtually written off the map as far as foreign investment is concerned. As investment activity becomes increasingly concentrated in this core group of countries, the idea that benefits will eventually trickle down through the economic global economic system appears at best far-fetched - and at worst an exercise in delusion.

This has important distributional implications. While foreign direct investment represents a relatively small proportion of total investment and national income, accounting for 3 per cent of GDP in East Asia and 1 per cent in Latin America, it is an important conduit for the transfer of new technologies. As international trade becomes more and more knowledge- intensive, access to these technologies becomes increasingly important to future competitiveness - and the difficulties faced by LDCs in attracting foreign investment threaten to exclude them from the major source of technological innovation, exacerbating their technological weakness in the process (UNCTAD, 1996). Foreign investment in the poorest countries is deterred by a variety of forces, including weak infrastructure, the small size of domestic and regional markets, shortages of skilled workers, low levels of education, and political instability. Once again, globalisation is at least as likely to exacerbate as to resolve these problems, as investment resources are concentrated on stronger economies.

Economic growth, trade expansion, and access to foreign investment tend to be mutually reinforcing, offering the potential for countries to enter virtuous cycles of rising average incomes; and posing the threat of a vicious circle of decline. To take one dimension of international finance, new commitments of export credits to developing countries doubled between 1990 and 1995 to $80bn. These flows play a potentially important role in financing the imports upon which export-competitiveness depends. Yet sub-Saharan Africa, the region in which access to imports is most constrained, is being bypassed by the most important source: financial transfers, including export credits. In 1994, sub-Saharan Africa received FDI flows worth $1.8bn, or the equivalent of flows to New Zealand. As a result, the region is becoming increasingly dependent upon concessional aid flows which, in contrast to private investment flows, are in decline. In nominal terms, aid flows have changed little since 1993, but they have fallen by 3 per cent per annum in real terms. In 1995, net flows of overseas development assistance (ODA) fell to their lowest level as a proportion of donor GNP since 1973, and budgetary pressures, allied to the unwillingness of governments to defend aid budgets, make it unlikely that this trend will be reversed (OECD 1996).

Regionalism

In parallel with the movement towards globalisation, regionalism has acquired a new lease of life since the early 1990s. Out of 98 preferential trade arrangements reported to the GATT up to 1995, one third were established during the previous five years (WTO 1995).

During the early 1990s, the revival of regionalism was widely perceived as a threat to the multilateral trading system. Developing countries were concerned at the prospect of a 'fortress Europe' emerging from Single European Market. Meanwhile, pessimism about the prospects of the Uruguay Round accelerated the emergence of new trade groupings committed to regional and bilateral liberalisation. The creation of the Asia Pacific Economic Forum (APEC) and the conclusion in 1993 of the North American Free Trade Agreement (NAFTA) between the USA, Canada, and Mexico reflected these fears, as did the US drive to extend NAFTA into a hemispheric free-trade area in Latin America during 1994. The spectre of three mutually hostile trade blocs centred on Europe, North America, and Japan, with a proliferation of discriminatory arrangements, loomed large behind the Uruguay Round, and probably helped to prevent its collapse (de Melo and Pagariya, 1992).

'Open regionalism'

Since the conclusion of the Uruguay Round, the political impetus behind regional trade has continued. In some cases, pre-existing preferential arrangements have been revived. The Central American Common Market (CACM), the Andean Group, and the Caribbean Community (CARICOM) were given new leases of life, as were regional integration strategies in Africa under the Common Market for Eastern and Southern Africa (COMESA), and the Southern African Development Community (SADC) in Africa. So far, the revival of these arrangements has been more cosmetic than real (Bouzas, 1995). In other cases - such as MERCOSUR - dynamic new alliances have emerged. In general, however, the new order is founded upon 'open regionalism' and liberalisation, rather than the principle of increased protection against non-associate members (Mistry P 1995).

This is an important - and almost universal - departure from the regionalism of the 1960s, when preferential tariff systems were seen an integral part of import-substituting industrialisation strategies. During the second half of the 1990s, regionalism is likely to evolve in a manner which reinforces globalisation, accelerating the removal of trade and investment restrictions in a complex patchwork of trading arrangements integrated into a process of multilateral liberalisation under the WTO. It is certainly difficult to see regionalism offering an alternative to globalisation. On tariffs, the 10-20 per cent average levels agreed under the Uruguay Round for developing countries leaves little scope for regional initiative. The major possible exceptions are textiles and agriculture, although regional groupings such as ASEAN and MERCOSUR have so far failed to overcome deeply entrenched internal differences in these areas. With regard to investment, the development of a multilateral investment code under the WTO is likely to go much further than any regional initiative. Moreover, the WTO sets limits on the preferences which regional groups can provide (Page S 1995). The Uruguay Round agreement demands that regional groupings must aim at completely free trade among their members within "a reasonable length of time", without raising barriers against other countries.

MERCOSUR

Among the most significant of the new groupings to emerge is the Southern Cone Common Market (MERCOSUR), linking Brazil, Argentina, Paraguay, and Uruguay, is a customs union, with a common external tariff, in January 1995. These countries form a market of 200 million people, and account for around one half of Latin America's GDP. The potential for the creation of a dynamic integrated trade space is clearly enormous. It remains to be seen whether MERCOSUR realises that potential, but the liberalisation measures taken so far are more than cosmetic. Trade among the MERCOSUR countries has increased by over 250 per cent since 1990, led by a boom in trade between Brazil and Argentina. The mutually reinforcing tendencies of regionalism and globalisation are much in evidence. For instance, Japanese car companies have invested heavily in Argentinean plant to produce for export to Brazil (Financial Times, 25 January 1995). At the same time, MERCOSUR has boosted foreign investment from countries in the region, in many cases through partnerships with foreign TNCs. The scope for MERCOSUR to act as a force for liberalisation in the future is underlined by the participation of Chile, which joined as an associate member in 1995 and is now an important source of foreign investment within the group.

South-east Asia

In regional terms, south-east Asia has remained the most dynamic site for trade and investment. Foreign-investment flows into the region have more than doubled since the late 1980s, rising to $44bn in 1994, and exports have grown at over 10 per cent a year. Intra-regional flows have become increasingly important. Trade between developing countries in the region has risen from 25 per cent to 40 per cent of their total trade since 1980 (Asian Development Bank, 1996). Foreign-direct investment originating in the region is also growing in importance. It now accounts for over one third of total stock. Regional trade initiatives and the liberalisation of import and investment regimes make it likely that the above trends will continue. For instance, in 1992 the seven ASEAN countries formed the Asian Free Trade Agreement, which aims to create a free trade zone by the year 2008.

Investment flows are becoming an increasingly powerful force for regional integration in Asia, as companies respond to changes in relative labour costs and opportunities for production and export. Over one half of South Korea's foreign investment is now directed towards Asia, with China, Indonesia and Vietnam the main growth points. Taiwanese and Singapore-based companies are also relocating. Between them, these two countries and South Korea account for more than twice as much of the foreign investment going into Vietnam, one of the region's fastest-growing sites for investment, as do the USA and Japan. Second- generational newly industrialised countries (NICs) are also restructuring. For instance, Malaysian companies are emerging as major investors in Vietnam and the Philippines, relocating in response to rising wages in the domestic economy (Business Review 1996).

Growth triangles and cross-regional initiatives

Sub-regional initiatives are fuelling the drive towards more open trade, shaping the local experience of globalisation in the process. Examples are the East Asean Growth Area (Brunei, Indonesia, Malaysia, and the Philippines), and 'growth triangle' arrangements such as the Singapore-Johor-Riau triangle (involving Indonesia, Malaysia, and Singapore) and the Indonesia, Malaysia, Thailand triangle (Asia Development Bank 1992). At varying speeds, these triangles are creating important economic linkages, as witnessed by the transfer of electronic assembly operations from Singapore and Malaysia to Indonesia, the Philippines, and Vietnam.

Superimposed on this regional jigsaw is an equally complex system of linkages between regions. In 1995, MERCOSUR and the EU signed a landmark agreement which commits the two groups to the gradual establishment of a free-trade zone. That agreement also links Europe to the web of arrangements between MERCOSUR and other Latin American countries and groupings. More significant still has been the emergence of the Asian Pacific Economic Cooperation (APEC) forum, which links three of the world's largest economies (the USA, Japan, and Canada) to Asia and Latin America. The fourth APEC summit was held in Manila, the Philippines, in 1996, with the 18 countries represented collectively accounting for over half the world's GDP, 40 per cent of its total population, and half of world merchandise trade. It many respects, APEC remains an embryonic consultative group, even though its ultimate goal of full trade and investment liberalisation by 2020 is an ambitious one. So far, practical action has taken the form of highly publicised tariff-reduction measures, most of which have been within the parameters set by the Uruguay Round. What is important about APEC is less its concrete achievements to date than its potential for linking MERCOSUR, NAFTA, and AFTA in a process of global market liberalisation.

It will be apparent from the above account that one region is conspicuous by its absence from the more dynamic regional alliances: namely, sub-Saharan Africa. This raises the danger that the world's poorest region, with the highest concentration of poverty, will be excluded from the trade and investment processes underpinning global and economic prosperity into the next century. During the first four years of the 1990s, East Asia's GDP grew at five times the rate for sub-Saharan Africa, and Latin America's at twice the rate. The resulting divergence in average incomes will be further widened if sub-Saharan Africa remains a marginalised participant in international trade.

Broad distributional questions

It has become an article of faith that liberalisation and deeper integration into the global economy are good for growth and for poverty reduction. For developing countries, globalisation is seen as the door to new opportunities - wider markets for trade, private capital inflows, improved access to technology, and greater efficiency. It is taken as axiomatic that 'outward-looking' reforms - a euphemism for liberalisation and deregulation - are the key to improved economic prospects. East Asia is cited as evidence in support of this view. Allied to the belief that globalisation accelerates growth is the parallel conviction that poor people will be the main beneficiaries. This is because trade is presumed to increase returns to labour, the most abundant asset of poor people.

Reality is more prosaic. Globalisation does create opportunities for wealth enhancement, but those opportunities are not equitably distributed among countries - or among people. At the national level, as the evidence presented earlier suggests, some countries have been unable to exploit the opportunities provided by trade and foreign investment. For sub-Saharan Africa, the ratio of trade to GDP is lower than it was twenty years ago. Meanwhile, the region's declining share of world markets has cost it the equivalent of $60bn per annum in current dollar terms over the past fifteen years. To put this figure in context, it represents around three times the flow of development assistance received by African governments.

What, then, are the conditions for successful participation in the global economy? This question is examined in more detail in Part 5, where the experience of south-east Asia is considered. In broad terms, however, seven interlocking conditions are important.

•  
• Sensible policies. Blanket protectionism, over-valued exchange rates, excessive taxation of producers, and over-regulation are policies which have had disastrous consequences for economic growth in Africa, and have contributed to the region's deteriorating trade performance. The south-east Asian model illustrates the importance of allowing markets to work within a viable regulatory framework, including selective and time-bound protection and investment controls.
• Avoiding 'big bangs'. Radical across-the-board trade liberalisation is unlikely to have the desired economic results, and highly likely to lead to unnecessary social costs. Experience suggests that imports are likely to increase rapidly, while exports increase more slowly. While old-style import-substitution strategies may have been misplaced, it is important not to throw out the baby with the bath water. The problem with these strategies was that they granted protection indiscriminately and failed to provide appropriate incentive structures. Selective intervention in response to market failures and the need for longer adjustment periods should not be ruled out. There are two broad strands to a viable alternative. The first involves liberalising imports of capital goods needed to generate increased exports and employment more rapidly than the liberalisation of imports which compete with labour-intensive local industries. The second strand would include the use of time-bound and performance-related protection for potentially viable industries.
• The development of diversified export structures with progress up the value-added chain. South-east Asia's success was based upon diversification and policies which established a dynamic comparative advantage in higher value-added sectors. Central to these policies were coherent industrial-development strategies, in which import controls and investment regulation were geared towards raising productivity, expanding employment, and competitiveness in world markets.
• Access to imports. Export competitiveness depends upon local industries absorbing new technologies. Limited foreign-exchange cover, chronic balance-of-payments deficits, and a large debt overhang impose obvious constraints upon import capacity. So too, does dependence upon volatile primary-commodity markets, because of the resulting exchange-rate instability. Access to finance is of crucial importance to trade expansion. In this context, the stagnation of bilateral and multilateral assistance to the poorest countries poses acute problems.
• Access to markets. Trade links high-income countries to low-income countries, expanding the market in which producers operate. In order to derive maximum benefits from trade, developing countries need access to markets in the industrialised world. This remains limited across a wide range of sectors, especially in the labour-intensive sectors which are of greatest relevance to developing countries.
• Access to infrastructure. Production for world markets requires a marketing infrastructure, including roads, ports, and telecommunications. Movement up the value-added chain typically requires an increasingly sophisticated infrastructure. Public investment in this area is therefore vital.
• Skills upgrading. As trade becomes increasingly knowledge-intensive, so the demands upon workforces will increase. Investment in education and the attainment of universal primary education would appear to be of crucial significance, as would investment in training, and research and development.

Establishing the effects of globalisation on poverty reduction and income distribution in respect of people, as distinct from countries, is more difficult. But crude extrapolation based upon the conviction that trade and investment growth is inherently beneficial to poor people is not helpful. Reduced to its essentials, globalisation integrates countries and people into a wider market. However, the relative weakness of the local economies, especially in the developing world, and the strength of global market-forces means that adjustment to world market pressures poses tough challenges. In the domestic market, local producers have to adapt to competition from imports produced in countries which have access to more sophisticated technologies, more capital resources, and a more skilled workforce. Foreign-investment flows can re-shape relations between social groups and regions. The resulting distribution of costs and benefits will reflect factors such as income distribution, the distribution of assets, levels of education, and intra-household gender relations.

Problems of distribution: access to assets

Other things being equal, it is not difficult to show that countries benefit from trade and specialisation. The problem is that other things, and notably the distribution of economic and political power, are not equal. Poverty is partly about a lack of assets. In countries where wealth is closely tied to land, disadvantaged households are typically land-poor or landless, and they lack access to other productive assets such as capital, technology, and water. They also tend to be concentrated in areas which are ecologically degraded and geographically isolated, with limited access to markets. It follows that, in the absence of wider redistributive measures, increasing returns to export-crop production will have limited benefits for the poor.

Consider, for example, the current growth strategy in Zimbabwe (World Bank, 1995b). Under the country's structural adjustment programme, commercial farmers have been given extensive tax and foreign-exchange incentives to expand production for exports. Crops such as tobacco, flowers, and off-season vegetables have been identified as growth points, and public investment resources been concentrated in these sectors. The problem, from a poverty- reduction perspective, is that Zimbabwe has one of the world's most unequal patterns of land distribution. Some 4,400 farms and ranches occupy one third of the country's arable land, including the bulk of the land in areas of high rainfall. These farms, which average over 2,200 hectares in size, produce more than 85 per cent of the country's marketed output. In contrast, an estimated 2 million farms are concentrated in the densely populated, highly degraded communal-farm areas. The majority of these farms are less than two hectares in size.

Only a small minority of communal farmers are in position to exploit the opportunities presented by export markets. Most are unable to grow enough food to feed their family members. In the poorest areas, marketing infrastructure is non-existent. Even if it were available, soil erosion and inadequate access to water would rule out the production of export crops. The World Bank itself has conceded that most producers in these areas will not be able to participate in the export-growth drive. Yet the vast majority of the 2.6 million Zimbabweans who are unable to meet their basic needs live in the communal farm areas. Not only will poverty within this group intensify, but Zimbabwe will become an increasingly unequal society. Given that it is already the world's most unequal society as measured by the Gini coefficient (0.57 compared with 0.61 for South Africa), this is an obvious source of concern.

The Zimbabwean case graphically illustrates the importance of concentrating upon the distribution of increments to income associated with growth, rather than upon aggregate growth rates. In countries characterised by a high degree of inequality such as Brazil, the Philippines, and Mexico, the opportunities for poverty alleviation created by globalisation will be missed in the absence of redistributive reforms, even where some income trickles down to the poor. One recent World Bank study estimates that a deterioration in income distribution in Bangladesh during the 1980s resulted in the head-count index of poverty falling by 0.3 per cent, rather than the 1.9 per cent which would have resulted had the income-distribution curve remained unchanged (World Bank 1996).

There are other reasons for questioning the simple extrapolation methods used by the World Bank and others. Encouraging export-crop production may imply a concentration of production on richer regions of the country (as is happening in Mexico and Zimbabwe). It may precipitate a land grab in which the rich and politically powerful dispossess the powerless (as in Brazil). In the Philippines, vulnerable urban squatter communities, such as those living on Manila's notorious Smokey Mountain, have been displaced to make way for the warehouses of foreign investors. Men may increase their cash incomes at the expense of women. Where export production is associated with mechanisation, it can reduce labour requirements and thus wage incomes.

On the other hand, export production can create jobs. This is clearly an important potential benefit. However, the quality of the jobs in question is important. In much of Latin America, export-led growth has been associated with low wages and, insecure employment, with women workers facing particularly exploitative conditions (Kay C 1995). Not only is the insecurity and vulnerability associated with such employment bad for poverty reduction, it is also bad for income inequality.

It is true that all of these processes relate as much to commercialisation in general as to globalisation in particular. One cannot draw general conclusions based on the presumed benefits or disbenefits of trade in the absence of specific information about particular situations. In broad terms, however, four conditions are of paramount importance in ensuring that globalisation contributes to poverty alleviation.

• Access to assets. Policies to redistribute land and improve tenancy rights are critical to ensuring that agricultural export-growth benefits the poor. Credit can also help the poor to accumulate assets and engage in markets. Developing financial institutions for the poor to mobilise savings and investment is thus crucial.
• Improving access to infrastructure and technology. Public investment in technology and infrastructure is critical in raising incomes and reducing poverty. Removing biases against small farmers demands the development of indigenous capacity to do research built on genuine participation and investment in training extension workers. Poor farmers invariably have less access to roads, electricity, and water than richer farmers, which restricts their ability to grasp market opportunities. Public investment in poor people and marginal areas is needed to correct this imbalance.
• Investing in people. There is overwhelming evidence that human capital is one of the keys to reducing poverty. It is also one of the keys to successful and equitable participation in global markets. Poorly educated people suffering from ill health do not provide a foundation for building prosperity in a competitive global economy. In most of the countries failing in world markets, there is too little investment in human capital, and this increases the probability that the country and the next generation of its citizens will remain poor.
• Protecting labour. For people who participate in global markets by selling their labour, there are two determinants of the benefits which result: incomes and security. Employment practices which drive down wages to the point where they barely meet basic subsistence needs are economically inefficient, and they are socially inequitable. Similarly, 'flexible' labour practices which disregard the most basic employment rights will diminish the potential benefits of trade and investment flows.

Part 2: The Uruguay Round: an assessment

The Uruguay Round of multilateral trade negotiations, which was concluded with the signing of the Final Act in 1994, was the longest of the eight rounds held under the auspices of the GATT. It was also the most ambitious. Even before its conclusion, the Uruguay Round was widely celebrated as the dawn of a new era, with enormous gains for all countries confidently predicted, including the poorest. The ministerial declaration which launched the Round included in its first objective the extension of benefits from international trade to developing countries as one of the major aims of the negotiations. Tariff reductions, the lowering of non-tariff barriers, the integration of textiles into a new set of trade rules, the liberalisation of agriculture, and the maintenance of special and differential treatment for developing countries are commonly cited as evidence that this objective was achieved and that the Final Act has created a framework for the more equitable distribution of benefits from world trade.

In fact, the benefits accruing to developing countries from the Uruguay Round are considerably more limited than such accounts suggest. Under the old GATT regime, trade liberalisation was geared towards policies and sectors (mainly tariffs on manufactured goods) of interest to the industrialised countries (Williams, 1994). Under the new regime, the benefits of tariff reduction have again been weighted in favour of the industrialised world. Moreover, the move towards full liberalisation in textiles and agriculture is considerably more restricted in practice than the principles enshrined in the WTO framework might indicate. An additional problem for the poorest countries is that many of the structural problems which hamper their capacity to benefit from more liberal trade did not figure on the Uruguay Round agenda.

Tariffs

Despite its lack of teeth, the old GATT system played a central role in reducing tariffs, the main form of protectionism during the inter-war period. In 1947, the average tariff on manufactured trade was 47 per cent; by 1980, following successive rounds of trade negotiations, it had fallen to 6 per cent. Full implementation of the Uruguay Round agreement will further reduce the average industrial-country tariff to 3.9 per cent (GATT 1994).

Disaggregating this average figure makes it clear that the benefits will be biased towards the developed countries and South Asia. Thus the average tariff reduction for trade among industrialised countries is higher (45 per cent) than the overall average (38 per cent) for developed-country tariff cuts (Woodward, 1996). For imports from Asia, the average tariff reduction is about one third, compared with 20-25 per cent for other developing regions. Tariff concessions between the developed countries are also more extensive with regard to tariff peaks. Thus the share of imports from all sources paying tariffs in excess of 10 per cent will fall from 15 per cent to 10 per cent for all countries; but only from 21 per cent to 15 per cent for developing countries (Weston, 1994).

Paradoxically, the smallest tariff reductions have been adopted for the poorest countries. With the full implementation of the Uruguay Round, the average tariff on imports from the least-developed countries into the developed countries will be 30 per cent higher than the overall average. For developing countries as a group, it will be 10 per cent higher. This reflects the lower reductions applied to products of greatest interest to the world's poorest countries (Safadi and Yeats, 1996). Tariff rates in the four major developed-country markets after the Uruguay Round tariff reductions are implemented will remain considerably above the average for agricultural goods, textiles, leather, and footwear - all areas in which developing countries have a major interest.

Another problem is tariff escalation: the practice of setting higher tariffs on processed goods than on raw materials (see Safadi and Yeats 1993). This obstructs an obvious way for developing countries to add value to their exports: namely, by processing raw materials before they sell them. The impact of tariff escalation is to deter investment, undermine employment, and lock developing countries into volatile primary-commodity markets, where real prices are in secular decline. In general terms, tariff escalation will be reduced under the Uruguay Round agreement. But it remains important in a number of key sectors. For some commodities of major significance to developing countries - such as leather, oilseeds, textile fibres and beverages - tariffs will continue to escalate by between 8 and 26 per cent on the final-stage product (World Bank, 1994b). While tariff escalation may not be the principal barrier to developing countries expanding the domestic value-added of their exports, this structure will inevitably discourage exports and restrict foreign-exchange earnings (UNCTAD 1994).

Non-tariff barriers

As tariff barriers declined under successive GATT agreements, industrial countries increasingly resorted to non-tariff barriers (NTBs) as the preferred means of restricting imports, especially from developing countries. The Multi-Fibre Arrangement (MFA), which is considered below, was one variant. Other measures included the euphemistically title 'voluntary export restraints' (VERs), under which countries were invited to limit exports or face retaliatory trade restriction, safeguards, and anti-dumping duties.

Under the Uruguay Round agreement, new rules have been adopted to reduce the incidence of NTBs. Developing countries stand to make significant gains, since NTBs have been applied disproportionately to products - such as steel, footwear, textiles, leather and rubber goods - in which they have a major interest. The coverage of developing-country exports is scheduled to decline from 18 per cent to 5.5 per cent (Woodward, 1994).

That said, the scope for evasion of the spirit, if not the letter, of the Final Act remains considerable. VERs are to be phased out over a four-year period. However, the picture with regard to safeguards is more confused. One of the reasons why developed countries have resorted so heavily to safeguard actions is that they can be applied selectively (unlike Article XIX actions) and without compensation. The EU had wanted this departure from GATT principles enshrined in the WTO. That outcome was avoided. However, selective safeguards may still be applied, albeit with a four-year time ceiling and with evidence of actual injury, against suppliers whose exports to a market grow 'disproportionately'. General safeguards may also be applied for longer periods, to prevent serious injury. Both provisions suggest that considerable scope for the harassment of exporters has been retained. Another danger is that safeguards will replace VERs as the main vehicle for discriminatory protectionism.

The same is true of anti-dumping actions. These have been widely applied by the US and the EU against developing-country exports of everything from steel to colour televisions and toys. Under the GATT, importers used a wide variety of anti-dumping formulas in adjudication procedures. However, they shared in common mechanisms designed to guarantee a high degree of success for the domestic industries bringing complaints. Under the old EU system, the level of success in prosecuting anti-dumping actions was unmatched in any other area of judicial action (Hindley 1989). Under the new system, more uniform rules have been adopted. The problem is that these rules retain some highly arbitrary criteria: for instance, in estimating reasonable profit against which to measure dumping margins. On the other side of the balance sheet, smaller developing countries will benefit from the provision exempting from anti-dumping action countries supplying less than 3 per cent of the market (UNCTAD, 1994). Beyond this there are no special provisions for developing countries. In view of the considerable costs and technical demands facing any country wishing to context an anti-dumping action, this is a serious shortcoming.

It is too early to evaluate the overall agreement on NTBs. However, close monitoring is required to ensure that the developed countries comply with the spirit of the Final Act. As in other areas of international trade, the only real defence against unfair recourse to NTBs is retaliatory action - and this is an area in which developing countries have unequal leverage.

Special and differential treatment

For practical purposes, the special and differential (S&D) provisions applied under Part IV of the GATT for developing countries brought few benefits. In theory, Part IV provided developing countries with an opportunity to benefit from global liberalisation without corresponding obligations. In practice, recourse to S&D, arguments resulted in developing countries being marginalised in successive multilateral trade rounds, with developed countries having no immediate self-interest in opening their markets on a non-reciprocal basis. Instead, developed countries were able to use their influence over the IMF and World Bank to secure the liberalisation measures which they would otherwise have been required to negotiate under the GATT. In the run-up to the Uruguay Round, developing countries themselves were less strident in their defence of S&D, which was widely perceived as a failed policy option. For their part, the developed countries saw the Round as an opportunity to further dilute Part IV provisions. The upshot is that, while the overall language of the Final Act uses the old language of S&D - i.e. developing countries are expected to make concessions consistent with their development, financial, and trade needs - the concept has been seriously eroded. What does this mean in practice?

One important implication is that developing countries have lost some flexibility in using trade-policy measures for balance-of-payments purposes. In particular, their ability to resort to quantitative restrictions on imports has been severely curtailed. Developing countries have also accepted tariff binding, albeit at levels significantly above currently applied rates. Ironically, developing countries now have less scope for resorting to quantitative restrictions than developed countries, which have retained the right to use them under the agreements for agriculture, textiles, and safeguards.

It is a similar situation with regard to subsidies, where the agreement is tailored to the requirements of developed countries. Thus subsidies are prohibited for product development (which is vital for many developing countries), but permitted for research and development and labour retraining, which are more relevant to developed countries. In the case of agriculture (see below), the subsidy rules have been contrived to allow the developed countries to maintain transfers to farmers through direct budgetary payments, while restricting market-based transfers. That approach may be viable in a situation where budget resources are large relative to the size of the agricultural population. But for developing countries, where the rural sector typically accounts for over half of all employment, and where governments face serious budgetary constraints, it is not an option.

In most areas of the Final Act, S&D treatment means that developing countries are given longer time-frames in which to implement more modest commitments to liberalisation. Thus in the case of agriculture, developing countries have ten years (rather than six) in which to reduce import tariffs by 24 per cent (rather than 36 per cent). The underlying logic of obliging developing countries to reduce restrictions on imports in markets dominated by the industrialised countries, where subsidised transfers to farmers represent half the value of agricultural output, is questioned below.

For developing countries, intellectual property rights are another significant aspect of the Uruguay Round agreement. Essentially, the Final Act requires the extension to twenty years of effective patent protection for all areas of technology recognised in the developed countries. The act also extends the scope of intellectual-property protection by limiting exclusions. Differential treatment, as it has been elaborated under the Final Act, will allow developing countries longer time-frames for implementation (with five-and ten-year grace periods respectively for developing and least-developed countries). One of the most important effects will be to increase the cost of imported technologies on which competitiveness in international markets depends. This will generate windfall gains for the owners of patents - notably US, European, and Japanese TNCs - while imposing new foreign-exchange demands on the poorest countries. Yet there is no provision to compensate developing countries for these additional costs.

More generally, intellectual property rights remain an area of international trade characterised by deeply rooted double standards. In the early phases of its industrial development, the USA adapted and developed European technologies without regard for patent rights. Japan followed this example after World War II. Today, however, developing countries are having their right to adapt technologies curtailed by these same countries. In contrast to other areas of international trade, where 'free trade' is the dominant ethos, in intellectual property the industrialised countries are adopting overtly trade-restricting practices as a means of enhancing rent transfers to transnational companies, which account for an estimated 90 per cent of patents (Gibb, 1988).

Textiles and clothing

Until the Uruguay Round, the textiles and clothing sector - along with agriculture - was the major exception in the movement towards freer trade. Managed trade in textiles began in 1961 and evolved through four successive versions of the Multifibre Arrangement (MFA). Initially, the move was justified as a temporary arrangement, under which industries in the developed world would be given time to adjust. The 'temporary' arrangement survived for three decades. Though posing as a multilateral system, the MFA was in reality a complex package of bilateral arrangements under which the industrialised countries fixed quotas on a country-by-country and product-by-product basis. In each succeeding phase, the MFA became either wider ranging (by covering more products) or more restrictive (by imposing tighter quotas) - or, more usually, both.

The costs of the MFA have been exceptionally high in the developing world. Textile and clothing figure prominently in the manufactured exports of all developing regions, accounting for 24 per cent of the total for Africa, 14 per cent for Asia, and 8 per cent for Latin America (Majumdar, 1995). For some of the poorest countries, textiles and clothing are the most important source of foreign-exchange earnings. For example, both Bangladesh and Sri Lanka depend on the sector for around half of their total export earnings. Estimates of the total foreign-exchange losses resulting from MFA quotas and tariffs range from $4bn to $15bn annually (the most widely used estimates are Trela and Whalley 1990; and Yang 1994). These losses are transmitted through the mechanisms of international trade to local industries in the form of lower wages, reduced employment, and lower investment, all of which have adverse implications for poverty reduction (Cable V 1990).

The delayed 'phase-out'

To what extent will the Uruguay Round Agreement on Textiles and Clothing (ATC) resolve the problems faced by developing-country exporters? In principle, the ATC provides a legal framework for integrating the sector into normal WTO disciplines over a ten-year transition period. Thus developed countries have granted themselves the same special and differential terms in textiles as they have extended to developing countries under the agricultural agreement! The phasing out of MFA restrictions comprises two strands: the elimination of restrictions in bilateral agreements negotiated under the MFA umbrella, and an increase in quotas according to a fixed growth rate. There are three steps in this process. In 1995, importing countries were required to remove from MFA restriction 16 per cent of textiles and clothing imports; in 1998, they are required to remove a further 17 per cent; and in 2002 another 18 per cent. In 2005, all remaining products are to be integrated. There are parallel arrangements for increasing growth rates for quotas (Trela I 1995).

There are two points to be made about this arrangement. First, tariffs will remain very high in the textiles and clothing sector. At 12 per cent, the average tariff will be three times the industrial-country average, implying a high degree of discrimination against developing- country suppliers. Second, the industrial countries appear to have adopted a policy of implementing the letter of the ATC, while violating its spirit. This applies per force to the US and the EU, despite wider differences in their policy approaches.

The USA

In contrast to the EU, the USA has published a full list of products to be phased out over the different stages to the year 2005 (Majumdar 1995). Most sensitive products will remain on the MFA list until the very end. This applies to 14 product categories exported by Bangladesh, 21 exported by Thailand, and 27 exported by the Philippines. The US has also exploited a wide range of loopholes to restrict import growth. Within a few months of the agreement being put into place, it had instituted 25 'safeguard' applications to the WTO, five of them against Honduras, and three against India (WTO 1996). Eleven of these resulted in restraint measures being adopted, five on a unilateral basis without approval by the WTO. In another worrying development, rules-of-origin provisions are being applied with increasing inflexibility, for instance by restricting imports of table cloths produced in the Philippines with fabrics imported from China or Thailand (Far Eastern Economic Review 1996). These measures have attracted criticism from Pakistan and the ASEAN countries, which are pressing for a review of the ATC.

The EU

The EU has adopted a different approach, linking implementation of the ATC to improved access to developing-country markets. Countries such as India have been invited to reduce their tariffs on EU clothing and textile imports, in return for concessions in the EU market. The issue has emerged as a source of tension between the EU and around a dozen of its Asian trade partners, point out that they are under no legal obligation to offer reciprocal tariff reductions. More broadly, there is something inherently questionable in principle about the developed countries demanding that the developing countries bargain for the withdrawal of trade restrictions which were themselves a violation of GATT principles.

The EU has also hit upon the ingenious idea of including in the first phase of 'liberalisation' items - such as parachutes and car-seat belts - not previously covered by the MFA. Almost 60 per cent of all the items offered for liberalisation by the EU fall into this category, thereby minimising the benefits to developing country exporters. Meanwhile, trade in clothing, which is the most important export for the poorest countries, has been the least integrated into the EU's liberalisation plans, accounting for just 2 per cent of the total product coverage. Items are also categorised by their 'sensitivity', with restrictions being retained for products likely to compete with domestic industries. One of the worst-affected countries is Vietnam, which has 27 clothing products under restriction, most of which are confined to growth rates of less than 2 per cent.

To be fair to the EU, it has not been alone in its efforts to flout the ATC. In October 1996, the Textiles Monitoring Body of the WTO evaluated the liberalisation measures undertaken by importing countries under the first phase of the agreement. It concluded that, with the exception of one product in one country (gloves in Canada), none of the items subject to liberalisation was subject to quantitative restrictions prior to 1994. The report also noted that the products in question were in the relatively lower value-added range, suggesting that the requirement that import volumes be increased by 16 per cent was being met in a way which secured far lower benefits expressed in value terms.

'End-loading'

It is difficult to escape the conclusion that the primary concern of the EU and the US has been to restrict imports in the interests of protecting domestic employment. The danger is that, by 'end-loading' the transfer of benefits to developing countries, both the US and the EU have, in the eyes of exporting countries, created a breathing space for protectionist lobbies to achieve a postponement of the final stage of the ATC. As the International Textile and Clothing Bureau has put it: "The notifications (by the EU and the USA) do not demonstrate political will...to integrate textiles and clothing into the liberalisation of trade" (DiDonato, 1995). Domestic political pressures - and social problems - provide part of the explanation for this lack of political will. In parts of the EU - notably Britain - and in the USA, the textile sector is characterised by low wages and declining employment levels (with the number of jobs declining by around 3 per cent a year for the past decade). That said, there is surely a case for governments to adopt more active labour-retraining and skills-enhancement programmes as an alternative to imposing adjustment costs on developing countries.

Partial losers

Viewed from the perspective of some of the poorest countries, liberalisation of the textile and clothing sector poses problems of a different order. In a liberalised trade environment, the most competitive countries are likely to gain most, with China, Indonesia, Thailand, and South Asia figuring prominently (World Bank, 1996; Page and Davenport, 1994). By contrast, exporters with large quotas relative to their competitiveness will lose out, unless they are able to raise productivity, as will countries which are currently insulated from competition by quotas on more competitive suppliers. In this group, Nepal, Bangladesh, and Sri Lanka stand to lose out. According to one estimate, Bangladesh could suffer a decline of almost one fifth in export earnings (Page and Davenport, 1994). This would have disastrous implications for human welfare. Around one million people are employed in the Bangladeshi garment industry, 90 per cent of whom are women (Ahmad and Kabir, 1995). While conditions in the textile industry are characterised by low pay and poor conditions, there are few alternative sources of employment. Women workers in the textile industry are often the main household income earners, supporting large numbers of dependants. For many, the loss of livelihoods in the textile industry would be a one-way ticket to increased poverty and vulnerability.

Agriculture and food security

From the outset, agricultural trade was effectively excluded from the rules of the GATT. Initially, this was because the US refused to accept multilateral disciplines to regulate its right to restrict imports and subsidise exports. The EU was subsequently able to exploit GATT's weak agricultural provisions, which were sufficiently flexible to incorporate its Common Agricultural Policy (CAP). The Agreement on Agriculture adopted at the Uruguay Round marks the first step towards the incorporation of agriculture into the multilateral trading system. However, that step is a very tentative one, and the balance of responsibilities and obligations between developed and developing countries is highly uneven.

The impact of Northern subsidies

Agricultural subsidisation, over-production, and export dumping by the industrialised countries have been the most visible manifestations of the special status of agriculture in world trade. Developing countries have suffered foreign-exchange losses as a consequence of the lower world prices and lost market shares resulting from competition against subsidised exports. Such losses are inherently difficult to quantify. However, one estimate suggests that a 30 per cent reduction in subsidies and protection would increase developing-country export earnings by around $45bn (Goldin and van der Mensbrugghe, 1993). The major beneficiaries would be a core group of Asian (Thailand, Malaysia and Indonesia) and Latin American (Brazil, Argentina and Uruguay) exporters of temperate products such as cereals, oilseeds and meat and livestock. During the 1980s, these exporters suffered unprecedented losses as export dumping by the EU and the USA drove prices down to their lowest levels since the 1930s, with developing countries caught in the cross-fire of a subsidy war.

As a group, developing countries have progressively lost agricultural market-shares to the developed countries over the past four decades. In 1950, they accounted for around one half of world agricultural trade. Today, they account for one quarter (FAO 1996b). Paradoxically, the countries which depend least upon agricultural trade as a source of foreign exchange, employment, and income have been expanding their market domination. For instance, the EU was barely self-sufficient in sugar less than two decades ago, but is now the world's largest exporter of sugar. Conversely, the countries which are most dependent on agriculture for their social and economic development have been losing out.

It is not only agricultural exporters that have faced problems. Staple-food producers in many developing countries have been forced to compete in local markets against heavily subsidised imports. These imports have depressed domestic prices and created consumer demand for foodstuffs which are not produced locally. The displacement of cassava, sorghum, and millet by wheat-based bread and imported rice in West Africa is an example of this process (Reardon, 1994). Cereals imports now absorb around one quarter of sub-Saharan Africa's export earnings. In many of the poorest countries, dependence upon food imports is increasing as per capita food-production declines, partly as a consequence of the market effects of food-dumping. Even in Latin America and the Caribbean, per capita cereals production was lower in 1990 than in 1960 (FAO, 1996a).

Food-security issues

Cheap food imports have positive short-term income benefits for food-deficit countries and households. At a national level, subsidised dumping by the industrialised countries reduces the foreign-exchange costs of imports. For governments pursuing industrial development strategies based upon cheap food, the advantages are obvious. They help to explain why some developing countries attached such importance to securing compensation for the Uruguay Round agreement, claiming that it would increase their food-import costs. For the poorest households, which spend the largest proportion of their income on food, this is an important consideration. In many developing countries, urban populations are critically dependent upon imported foodstuffs. But the poorest rural households are also typically in a food-deficit situation, selling crops in the post-harvest period and purchasing staples after household supplies have run out (Mellor, 1995). In narrowly defined income terms, these households stand to benefit from any reduction in local prices caused by imports - and, by extension, from subsidised food-dumping.

The problem is that food-security problems cannot be viewed solely in narrowly-defined income terms. At a national level, many of the 88 countries categorised by the FAO as low-income food-deficit are not in a position to sustain imports (FAO 1996a). Nor can they rely upon food aid to cover shortfalls. The hike in world prices during 1995 added some $4bn to the import bills of these countries, while food shortages fell to their lowest levels since the mid-1970s. Food- deficit countries in sub-Saharan Africa, faced as they are with deep-rooted problems of debt and dependence upon volatile commodity markets, are not in the same position as South Korea when it comes to ensuring that food imports can be secured whenever, and in whatever quantities, they may be needed. That is why the Economic Commission for Africa has stressed that "Africa's viability resides, above all other considerations, in its being able to feed its own people from internal resources" (Economic Commission for Africa, 1991).

It is also questionable whether expenditure on food imports constitutes the most productive use of one of the scarcest resources of low-income countries namely, their foreign exchange. Collectively, the FAO's eighty-eight low-income food-deficit countries spend half of their foreign exchange on food imports. Yet in many of these countries, smallholder farmers are more than capable of feeding their countries. What is needed is the creation by governments of an enabling environment which facilitates the participation in markets of poor producers and more marginal areas. Public investment in infrastructure, marketing, and post-harvest facilities, allied to agrarian reform and tenancy legislation aimed at enhancing access to land, water and other productive assets, is vital to the creation of such an environment. Action in this area should not be viewed as part of a trade-off between growth and poverty reduction. Given an opportunity, smallholder producers are highly productive (IFAD, 1990), and increased rural prosperity can provide the foundation for dynamic urban-rural economic linkages. Unfortunately, governments in many developing countries are unlikely to undertake the investments needed in smallholder production while the industrialised countries are offering apparently limitless supplies of 'cheap food'.

All change and no change in OECD subsidisation

Against this backdrop, the Uruguay Round agreement needs to be evaluated against one central criterion: will it end subsidised over-production and food-dumping by the industrialised countries? The answer is 'no'.

Briefly summarised, the Agricultural Agreement envisages a 36 per cent reduction in spending on production and export subsidies (with the volume of subsidised exports falling by 21 per cent), and a similar level of tariff reduction. All non-tariff measures are to be 'tariffied'. While superficially impressive, the real effects of these measures will be limited. For instance, the subsidy reduction applies only to measures which 'distort' trade. Direct payments to farmers, as distinct from market-support measures, are not included in this category, largely as a result of a bilateral agreement between the US and the EU. These payments currently account for almost one quarter of total OECD subsidisation. Moreover, the proportion is rising, as governments restructure their subsidy programmes to bring them into line with the WTO regime. Another limiting factor is the choice of 1986-1988 as the reference period against which to measure budget-spending reductions. Subsidies during this period were the highest ever, thus minimising the need for liberalisation. The end result is that the $182bn in producer subsidies provided by the OECD countries in 1995 - a sum representing 40 per cent of the value of output - is unlikely to decline (OECD 1996). It is worth mentioning that the 1995 subsidy figure was some 15 per cent higher than that at the start of the Uruguay Round.

The position in relation to export subsidies is similar. The reference period from which reductions will be measured (either 1986-1990 or 1990-1991) has been chosen to increase the level of export subsidisation which is permissible, in the case of the EU by around by around 10m tons of cereal. Moreover, the 21 per cent reduction in the volume of subsidised exports which are acceptable will leave the other 79 per cent untouched - an obvious point, but one which has received surprisingly little attention (Fowler P 1996; Gardner B 1993). The upshot is that agriculture will remain the one area of international trade under which multilateral trade rules sanction and institutionalise export-dumping.

Turning to the question of tariff reductions, the 1986-1990 reference period again means that real reductions will be substantially lower than the headline figure suggests. Equivalent reductions from current levels would create double the level of welfare gains, according to the World Bank's estimate (World Bank 1995). Moreover, the industrialised countries appear to have used baseline figures even higher than the 1986-1990 average for a number of major commodities, with the result that actual tariffs are now higher than they were (Stevens, 1996).

It is difficult to escape the conclusion that the US and the EU have written the Agricultural Agreement to legitimise, under WTO auspices, their various subsidy operations. In any area of world trade, this subordination of multilateralism to the pursuit of self-interest ought to be regarded as unacceptable. In the specific case of agriculture, which is of such vital importance to the world's poorest countries, the case for a fundamental review is overwhelming. That case is reinforced by the fact that the Agricultural Agreement requires developing countries to reduce the level of import protection to their food-staple producers - in effect, exposing them to global markets distorted by US and EU subsidies (Oxfam, 1996). This would appear to be a violation of the very market principles upon which the WTO is founded. More importantly, unequal competition in global markets between smallholder producers in the South and the industrialised farming systems and treasuries of the North poses a major threat to rural livelihoods and human development.

The distribution of benefits

There has been a wide range of assessments of the overall income effects of the Uruguay Round. The increase in global income expected to result from liberalisation is variously estimated at between $212bn and $510bn (World Bank/OECD 1993; Nguyen et al, 1994; Francois et al, 1994). Developing countries are projected to gain between $86bn and $122bn. These wide variations reflect differences in methodology (for example, the higher estimates tend to reflect assumptions about efficiency-related productivity gains), the inclusion or otherwise of non-tariff barriers, elasticities in supply, and other factors.

In reality, these figures provide more of an insight into the underlying conviction shared by most economists, that freer trade is good for growth, than they do about likely outcomes in the real world. That said, the various econometric evaluations of the Uruguay Round raise some important questions about distribution. For instance, most studies suggest that developing countries, which collectively account for around three quarters of the world's population, will account for only between one quarter and one third of the income gains generated: hardly an equitable outcome.

This may well understate the problem, since the most significant gains, amounting to 60 per cent, 34 per cent, and 20 per cent respectively (World Bank 1994b), are predicted for clothing (where liberalisation is likely to remain on the back burner for another decade), textiles (where current trends suggest a serious lag in implementation), and agriculture (where the benefits of the Uruguay Round agreement have been massively overstated). However, with all their limitations, the more optimistic figures mentioned above speak volumes about the unbalanced nature of the Uruguay Round agenda and it is bias towards the interests of industrialised countries. It also suggests that the Uruguay Round will further skew the distribution of global wealth in favour of the world's richest countries.

It is only North-South patterns of income distribution which will become more unequal as a result of the Uruguay Round. Within the developing world, the benefits of trade liberalisation will be unequally distributed between the larger, more advanced exporters of manufacturing goods in south-east Asia and Latin America and the poorer primary-commodity producers in Africa and elsewhere. Most studies suggest that the very poorest countries stand to lose from the Uruguay Round agreement, especially in the short run. This group includes countries which are beneficiaries of various preferential tariff schemes - such as the Lome Convention - which will face increased competition as a consequence of liberalisation. Preferences for the African, Caribbean, and Pacific (ACP) countries in the EU market will be eroded by some 30 per cent as a result of liberalisation under the Uruguay Round, and by 50 per cent for tropical products. One study (Page and Davenport 1994) estimates that Ethiopia, Malawi and Mozambique, three of the poorest countries in the world, will suffer export losses in excess of 4 per cent per annum as a consequence of preference erosion. Similarly, while developing countries as a group are projected to gain from liberalisation in clothing and textiles, the ACP and least-developed countries are projected to lose from the loss of preference margins. Bangladesh and Mauritius stand to be particularly affected.

Since Most Favoured Nation (MFN) tariffs are being reduced at twice the rate of Generalised System of Preference (GSP) tariffs, the wider preferences enjoyed by least-developed countries also stand to be eroded. While it is true that the GSP may have yielded limited results, its erosion could carry significant costs for some countries (Stevens and Kennan, 1994). According to UNCTAD, the least-developed countries stand to lose up to $600m annually - a sum equivalent to around 5 per cent of their export earnings (Weston, 1994).

The generalised conviction that trade liberalisation will ultimately benefit all countries has obscured a more complex balance-sheet of winners and losers. At a global level, the projected losses are small and heavily outweighed by overall income gains. Even so, the losses will be concentrated on a group of countries that can least afford them - and for some the costs will be significant. This scenario has disturbing implications for poverty and human welfare. Foreign-exchange losses will translate into pressure on incomes, a declining inability to sustain imports, and increased dependence upon aid. Revenue from trade will be lost, undermining the capacity of governments to develop the economic and social infrastructures upon which future prosperity depends. Thus, apart from widening the income gap between the world's poorest and the more dynamic developing countries in the short-term, the Uruguay Round agreement could initiate a longer-term trend towards increased inequality.

Policy considerations: a post-Uruguay Round agenda

While the Uruguay Round agreement poses a number of potential threats to some of the poorest countries, it also has the potential to generate income gains for others. Perhaps most importantly, a strengthened WTO may limit the recourse of the developed countries to the type of arbitrary and discriminatory trade policies which characterised the 1980s. As the weakest partners in the international trading system, developing countries as a group have the most limited retaliatory capacity, and are therefore most vulnerable to departures from a rules-based system. Against this, the industrialised countries have clearly not abandoned unilateralism. The Helms-Burton Act in the US not only reinforces sanctions against Cuba, it also extends sanctions to other countries which refuse to comply with US edicts. As such, the Act is a blatant departure from multilateral trade disciplines. So, too, is the continued recourse by the US to the threat of trade sanctions under Section 301 of the Trade Act. This entitles the US Trade Representative to impose sanctions against countries deemed to be pursuing 'unfair' trade practices. Significantly, it has been most heavily deployed in areas - such as the enforcement of patent claims and the pursuit of initiatives to open investment markets - in which the Uruguay Round agreement is regarded by the USA as inadequate, with developing countries among the most prominent targets. As the case of textiles illustrates, the scope for arbitrary protectionism also remains intact.

If multilateralism is to flourish, and if the poorest developing countries are to participate more fully and equitably in the international trading system, major reforms are needed. There have been some positive developments, both the US and the EU are considering reforms of their Generalised System of Preference schemes, reducing the per capita income threshold for eligibility and the upper ceiling on market shares after which preferences will be withdrawn (Weston, 1994). Reinforced preferences could help to redress losses suffered by the poorest countries as a consequence of Most Favoured Nation liberalisation under the Uruguay Round. However, the utilisation and benefits of GSP schemes for the poorest countries have been limited, and the share of preferential imports in dutiable imports has been declining (UNCTAD 1993). In 1994, UNCTAD's Special Committee on Preferences concluded that, apart from the weak supply-capacity of the poorest countries, GSP schemes were limited by factors such as their incomplete product-coverage, with the exclusion of 'sensitive' items in agriculture a special problem; the imposition of quota constraints on duty-free imports; complex rules-of-origin criteria which set upper limits on the proportion of value-added to exports by imported inputs; and by uncertainties surrounding the continuation of preferences, which limited an investment response. In each of these areas there is scope for accelerated liberalisation.

The WTO

One way in which developed countries could assist the poorest countries would be to conduct a review of the rules-of-origins arrangements with a view to simplifying the rules, and increasing - or, perhaps, removing - the ceilings for import-content. At the same time, GSP arrangements could be restructured, with exports produced by countries - such as the African, Caribbean and Pacific group under Lome - suffering preference losses as a result of liberalisation under the WTO being given special preferences. These options are considered in the Comprehensive and Integrated WTO Plan of Action for the LDCs, which calls for a comprehensive approach to "contribute to the expansion of trade, sustainable growth and development" of the poorest countries (WTO, 1996). But no detailed recommendations have been adopted or implemented.

More concrete and substantive action is required for LDCs (see, for example, UNCTAD 1995). This should include:

1. the elimination of tariff escalation, especially in GSP schemes for semi-processed tropical agricultural produce and natural resources;
2. deeper tariff cuts and duty elimination under GSP and other preferential schemes still subject to high tariff peaks;
3. more flexible and consistent rules of origin aimed at promoting labour-intensive exports;
4. exemptions from restrictions on textile imports for small suppliers, regardless of whether or not they are WTO members;
5. a prohibition on safeguard actions against products exported by LDCs.

With regard to imports, it is important that the special and differential provisions of the Uruguay Round are respected. 'Graduation' into full WTO obligations must take into account the specific circumstances of individual countries and economic sectors, with the objective of giving precedence to poverty reduction considerations over the claims of developed countries for access to markets. In this respect, the EU's approach to the implementation of the ATC is a major source of concern, since it appears that special and differential treatment is regarded as a trade barrier, rather than a legitimate claim. In the specific case of agriculture, there should be no obligation on developing countries to open up their food systems, because international markets are massively distorted by OECD subsidisation; and because the food security of people should take precedence over any requirement to liberalise.

Finally, there are important institutional questions which need to be addressed if the poorest countries are to participate in the WTO on more equitable terms. At present, there are 29 LDC members of the WTO. Only nine of these countries maintain a trade representative's office in Geneva. Even those countries which do maintain offices are massively over-stretched. While the US and the EU are able to mobilise and maintain armies of trade lawyers, technical specialists (often seconded from transnational companies), and official negotiators, most developing countries have one or two officials available. During the Uruguay Round, these officials were covering fifteen different negotiating areas, ranging from textiles to intellectual property and investment, while the developed countries were able to deploy entire teams for each working group. Such asymmetries in negotiating strength have obvious implications for the outcome of bargaining processes. Increased co-operation between developing countries, allied to shared specialisation, is part of the answer; but increased financial and technical support should also be made available through the WTO itself to create a more democratic structure.

The European Union

The European Union has special obligations to the poorest countries - and an important opportunity to meet them. Since the mid-1970s, the core of EU development policy has been the Lome Convention - an aid and trade preference arrangement linking it to 70 African, Caribbean and Pacific (ACP) countries. With the current convention due to expire in 2000, new policy directions are under consideration (European Union 1996). The outcome is a matter of vital concern to some of the world's poorest countries. Not only is the EU a major source of aid to the ACP countries, it also absorbs more than 40 per cent of their exports. Among these countries, there is a high concentration of those which have suffered an erosion of preferences because of the Uruguay Round agreement.

In many respects, the concrete achievements of the Lome Convention have been limited. Even with preferential access to the EU market, the ACP countries have failed to diversify, with over 80 per cent of their export earnings coming from primary commodities. They have also been unable to maintain their share of the EU market, which has fallen from 7 per cent in 1976 to 3 per cent today. Such facts have reinforced a growing conviction in the EU that existing Lome trade policies have failed - but the alternatives which are emerging are far from convincing.

One option under review is a modified status quo, with the EU maintaining a WTO 'waiver' for ACP preferences. Given past performance, this is not an attractive option. Another approach, favoured by some EU governments, is to focus upon reciprocal trade liberalisation, with the ACP countries taking on liberalisation obligations consistent with their development status. In effect, this would make the Lome Convention, at present a non-reciprocal contractual arrangement, an extension of the WTO, requiring the full integration of the ACP countries into the multilateral trade framework. A third option under review is the removal of the trade element from the Lome Convention (which would thus be reduced to an aid package), and its integration into the EU's GSP scheme for least-developed countries. The problem with this is that the EU's GSP scheme has been even less successful than the Lome Convention in improving the capacity of the poorest countries to take advantage of opportunities in the EU market.

What is needed is a more integrated approach, in which EU development cooperation policy is geared towards expanding market opportunities, and increasing the capacity of the poorest countries to take advantage of those opportunities. The first part of this equation requires a fundamental review of existing trade restrictions applied to the least-developed countries. Restrictions under the Common Agricultural Policy (CAP) and the enforcement of arcane rules of origin are two obvious areas for reform. The second part of the equation requires a closer co-ordination of aid and trade, with the EU gearing its aid policy more effectively towards the promotion of investment, production, and local processing in areas where it is providing trade advantages.

Another area in which the EU faces a major challenge is with regard to the special protocols attached to the Lome Convention. These protocols, which extend to sugar, beef, rum, and bananas, give individual ACP countries privileged access to the EU market in the form of quotas and higher prices. The banana protocol, upon which the survival of thousands of livelihoods in the Windward Islands and other Caribbean countries depends, has been successfully challenged at the WTO by the USA, acting on behalf of American transnational companies exporting from Latin America. Under the WTO ruling, the EU regime was found to be discriminatory and a barrier to free trade. At one level, that assessment was correct. At another level it has to be asked whether the livelihoods of entire communities should be sacrificed in the commercial interest of powerful TNCs.
 
  ^{NEXT SECTION}

              UNDP Home

BACK TO TOP