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

Section One - Globalisation, poverty, and development

Part 3 - The international trade agenda: some missing issues

Part 4 - Trade liberalisation, poverty, and distribution

Part 5 - Governments and markets

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

Part 1 - Unemployment and inequality

Part 2 - North-South linkages: the evidence

Part 3: The international trade agenda: some missing issues

Under the Bretton Woods system, the management of international trade was seen as one element in an integrated framework designed to distribute the benefits of global growth more equitably, and to prevent a return to the instability and systemic collapse of the inter-war period. The provision of finance to prevent temporary balance-of-payments problems being translated into deflationary pressure and beggar-your-neighbour protectionism was one element in the post-war settlement designed to promote shared growth. Another was support for reconstruction and infrastructural development under the auspices of the World Bank. Other structural issues relevant to developing countries also figured prominently. Under the original Bretton Woods blueprint, trade management was to include the regulation of international commodity markets, the instability of which was regarded as a prime cause of the crisis in the 1920s (Whalley et al, 1986).

This is not the place to review the origins, successes and failures of the Bretton Woods system (see Commonwealth Secretariat, 1989). But it is difficult not to be struck by the contrast between the vision of the post-war period and dominant orthodoxies of today. The post-war settlement was premised upon a conviction that governments shared a collective responsibility to regulate international markets in the public interest. That conviction derived from a recognition that unregulated markets in trade and finance had been responsible for the crisis which led to the Great Depression. Today, the assumption is that markets should reign supreme, with governments responsible only for removing barriers to the free flow of goods, services, and finance.

Derived from the familiar, and generally implausible, assumptions about the operation of global markets, the dominant view is that trade liberalisation and market deregulation will produce optimal outcomes in terms of welfare enhancement (World Bank, 1996). As a result, some of the most important structural constraints facing developing countries in the global trading system have been conspicuous by their absence from the international agenda. Problems in commodity markets and debt are perhaps the two most obvious cases in point.

Commodity markets

The share of primary commodities in the exports of developing countries has been declining at an accelerating rate for over four decades. Today, they account for 15 per cent of merchandise export earnings, compared with 50 per cent in 1970. Such averages conceal wide regional variations. For sub-Saharan Africa, non-fuel primary commodities account for over one third of export earnings. For 27 countries in the region, the share rises to over one half. This is roughly the same proportion as for the 48 least-developed countries. In addition, many of the world's poorest countries are acutely dependent upon the export of a small group of primary commodities. For instance, coffee accounts for over half of the export earnings of Uganda and Tanzania, as does copper for Zambia and tea for Malawi. It follows that, for many of the poorest countries, trade prospects are dictated in large measure by trends in international commodity markets.

Financial losses

For several decades, price trends in commodity markets have been unfavourable to exporters, and catastrophically so since the early 1980s. During the 1980s, short periods of price recovery were superimposed upon a relentless downward trend. By 1993, real non-oil commodity prices had fallen to less than half their 1980 level (World Bank, 1993a). In the case of tropical beverages such as coffee and cocoa, the decline was even more severe than the average, with prices falling by almost 70 per cent. The impact of this deterioration is summarised by Maizels, who points out that "the general level of real commodity prices had fallen by 1986 to below the nadir reached in 1932 during the Great Depression of the inter-war era" (Maizels 1994).

Translated into financial terms, the cumulative losses suffered by developing countries amounted to $290bn between 1980 and 1991 (Maizels, 1994). For sub-Saharan Africa, the most seriously affected region, the loss was equivalent to 5 per cent of GDP. In a region where around half of the population lives on or below the poverty line, the human costs were enormous. Adverse trends in commodity markets contributed to the evolution of Africa's debt crisis, the disintegration of social and economic infrastructure, and the collapse in investment - all of which have in turn contributed to the widening gap between that region and other parts of the developing world (Singer H and Edstron J 1995).

It is sometimes argued that, contrary to the 'pessimistic' Prebisch-Singer thesis which holds that real prices for primary commodities are in secular decline, recovery is around the corner. During the 1980s, the World Bank consistently forecast rises in commodity prices which failed to materialise. More recently, the Bank's terminal optimism was reflected in a 1990 study which concluded that "world coffee prices are projected to be on an upward trend throughout the 1990s" (World Bank, 1990). Similar predictions were made for cocoa. In the event, after a brief price hike in 1994-1995, normal service has resumed. Since the price peak of 1994, international coffee prices have fallen by half. Cocoa prices have also fallen, albeit less dramatically (International Cocoa Organisation 1996).

For individual countries, the losses associated with declining and volatile commodity markets can be highly destabilising. Consider the case of Uganda, where coffee accounts for between 60 and 80 per cent of export earnings. In 1995-1996, coffee exports increased by one third, only for the foreign exchange generated to decline, as world prices fell. As a result, the national trade deficit widened by 20 per cent, undermining the country's capacity to sustain the imports needed to maintain its economic recovery; and reinforcing dependence on aid. International coffee markets also have a major bearing on Uganda's capacity to service its massive external debt, which absorbs between one quarter and one half of the foreign exchange generated by coffee exports, depending on world prices. Once again, this leaves the country in a highly unstable position with regard to import capacity and currency stability.

The domination of global markets by powerful - and largely unaccountable trading houses compounds the problems facing developing countries. In early 1996, international prices for copper were around $2,600 per tonne. Evidence that an individual trader in the Sumimoto Corporation, one of Japan's big five soga shosha, had been attempting to drive up prices through futures trading on the London Metal Exchange led to a spate of panic selling which drove prices down to $2,000 per tonne in less than one month (The Economist 1996a). For Sumimoto, the episode translated into a loss of $1.8bn. That sum is equivalent to over half of the GDP of Zambia, which derives more than 80 per cent of its export earnings from copper. For Zambia, the collapse of copper prices translated into foreign-exchange losses of around $150m (more than the level of government spending on health and education combined), a widening of the current-account deficit and reduced import-capacity (The Economist 1996b).

The factors behind the long-term decline in commodity prices have been extensively documented (for example, Maizels, 1987; ODI, 1995). They include slower growth in the industrialised countries, low income-elasticities of demand, technological developments - including substitution by synthetic and manufactured goods - and competition from subsidised agricultural production in the OECD countries. In the specific cases of cocoa and coffee, IMF-World Bank structural adjustment programmes have probably contributed to the problem by expanding output during a period of chronic over-supply. During the 1980s, cocoa production in West Africa increased from 1.6m tonnes to 2.3m tonnes, while prices fell from $2.6 to $1.2 per kg. As a result, increased output was rewarded with dramatically reduced export earnings - a model of the 'fallacy of composition' (World Bank 1993a).

Learning from the past

Whatever the precise range of forces influencing particular commodity markets, the underlying trend towards lower prices seems unlikely to be reversed. This is a harsh fact which poses policy makers important questions. Without concerted international action, a large group of countries will become increasingly marginalised in the world trading system, with their citizens unable to escape poverty. The implications for global stability of excluding a large share of the world's population being excluded from shared prosperity merits serious consideration.

There are lessons here from the past. Prior to the Bretton Woods conference, Keynes had developed plans for an international agency for stabilising commodity prices under an International Commodity Control Programme which would establish buffer stocks in defence of remunerative prices (Kaldor, 1987). The objective, as Keynes saw it, was "to provide [producers] with proper nutritional and other standards in the conditions in which they live" (Keynes 1980). It was not until the emergence of International Commodity Agreements (ICAs) as one of the main planks of the 'new international economic order' that this approach was operationalised, only to collapse in the 1980s in the face of producer rivalries, and hostility on the part of importing countries. While there is scope for debate over the most appropriate policy instruments, the objective identified by Keynes is as relevant today as it was half a century ago.

Improved risk management would help to stabilise foreign-exchange earnings. However, some form of supply-management is probably essential. Action by coffee and cocoa producers to abandon buffer stocks in favour of export-retention schemes, and to restrict production until at least 1998-1999, could be a step in the right direction (Africa Recovery, 1996). Unfortunately, evidence on implementation offers little cause for optimism. In 1996, the export-retention scheme introduced by Association of Coffee Producing Countries (ACPC) - an arrangement based on voluntary export quotas - collapsed, as producers started to sell domestic stocks. A new programme, which aims only at stabilising and not raising prices, has been agreed, but the large number of non-participating countries (which include Mexico and Guatemala), allied to differences between members over quota sizes and market share, have prompted the Economist Intelligence Unit to "foresee very little impact, if any, from the ACPC's new export programme" (EIU Third Quarter 1996).

Looking to the future, supply management is not a panacea. There could, however, be a link established between supply management and diversification, with some proportion of the gain in export revenue being made available to help to finance processing operations. Ultimately, vertical diversification (through increased processing) and horizontal diversification (through an expansion in the range of goods produced) holds the key to resolving the crisis facing commodity exporters. The UN's Common Fund for Commodities provides a framework for addressing these challenges in the context of the African Plan of Action for Commodities adopted in 1996 (Africa Recovery 1993).

But although primary-commodity exporters themselves can do much to meet the challenge of resolving their shared problems, wider international action is also required. That action has failed to materialise. Appeals from the UN Secretary General for a $75m fund to support commodity diversification have not met with a constructive response from donors. Meanwhile, the industrialised countries have shown little inclination to support the export-retention scheme proposed by African governments. During the Uruguay Round, there was virtually no discussion of the deeper problems facing primary-commodity exporters, underlining again the demise of the integrated trade agenda of the post-war era.

All of this has disturbing implications. The world commodity economy remains in deep crisis. Yet international policy has been in a state of paralysis since the early 1980s. The problems will not be resolved by a piecemeal approach to trade, with tariff liberalisation as its overarching objective. What is required is a new attempt to evolve a coherent strategy. In this context, it is interesting to contrast the laissez faire approach governing commodity markets with the co-operation and co-ordination of policies between the G7 countries with regard to exchange-rate management. What is missing in relation to the commodities equation is the key ingredient of political will, and a sense of shared interest in distributing the benefits of trade more equitably.

Debt and trade (this section is based on Oxfam 1995; Oxfam 1996; and Killick 1995)

During the 1990s, the 'international debt crisis' has drifted imperceptibly out of the headlines. Commercial debt relief under the Brady Plan, an increase in private capital flows and a recovery in exports have improved the situation of debtor countries in Latin America and reduced the level of risk facing the international financial system. However, for many of the poorest countries unsustainable debt has remained a serious problem, diverting resources needed for investment in social and economic recovery, undermining investment, and contributing to financial instability. Each of the consequences has had damaging implications for trade prospects, although the issue of debt relief has not figured in trade negotiations.

There are thirty-two countries classified by the World Bank as severely-indebted low-income countries (SILICs). The debt ratios for these countries are considerably worse than they were for the middle-income debtors, even at the height of the debt crisis. For instance, the ratio of debt-to-GNP averages over 110 per cent, while the debt-to-export ratio, measured in net present value terms, is over 400 per cent. According to the World Bank, 200 per cent is the upper ceiling for sustainability.

Debt repayments constitute a considerable burden on the limited foreign-exchange resources of the poorest countries, typically absorbing around one fifth of export earnings. These repayments seriously undermine the capacity of highly indebted countries to engage in international trade on reasonable terms. Thus the import volumes of the SILICs are still around the level of 1980s, reflecting acute foreign-exchange shortages. What this means is that import compression is depriving local manufacturing and agriculture of access to imports which could increase productivity and generate employment. By undermining the capacity of governments to develop infrastructure, debt also acts as a deterrent to foreign investment and wider export competitiveness. Debt servicing has also contributed to a fiscal crisis in many countries, absorbing huge amounts of domestic revenue and eroding the capacity of governments to maintain basic social and economic infrastructure.

A specific problem facing the SILICs has been the structure of their debt. Multilateral creditors, mainly the IMF and the World Bank, account for over one quarter of stock and one half of payments. Traditionally, these institutions have refused to reduce or reschedule debts, claiming that such action would jeopardise their preferred-creditor status. During 1996, this approach was abandoned with the adoption of a new debt framework - the Highly Indebted Poor Countries initiative. This initiative has the potential to generate positive benefits for many countries. Two of its core principles are especially important. First, the objective it sets is the reduction of debt to levels below a sustainability ceiling (defined in terms of debt stock and debt-service ratios). This marks a welcome departure from the previous principle, which dictated that the aim of debt renegotiation was to maximise repayments to creditors. Second, the initiative will deal with all debt - including multilateral debt - in a single, comprehensive operation.

Effectively implemented, this initiative could remove the formidable obstacle to trade recovery posed by debt. The problem is that the time-frame is hopelessly inadequate. Countries such as Ethiopia, Tanzania, and Mozambique will not qualify until after 2000 (because of the requirement that they comply with two consecutive IMF programmes). Other countries - such as Honduras and Mali - are likely to fall foul of the eligibility criteria, and may not qualify for debt relief at all. Viewed from a global perspective, debt relief is vital to reducing poverty and inequality, and to expanding the opportunities for welfare improvement offered through international trade. Highly indebted poor countries need debt relief now - and not at some indeterminate point in the future. Moreover, debt relief is eminently affordable. Providing effective debt relief to the twenty countries deemed potentially eligible would cost between $5.5 and $7.7bn - less than the cost of one Stealth bomber, and roughly equivalent to the costs of establishing the Euro Disney theme park in Paris. As in the past, the one ingredient missing in the existing debt relief framework is the most important: namely, political will. (xxxx)

Trade and Funding Development Assistance

Beyond more narrowly defined trade-policy issues, the poorest countries will need additional financial support if they are to take advantage of the opportunities for trade and poverty reduction provided by trade. Debt-relief measures (see below) would help to release scarce foreign-exchange resources for investment in infrastructure and diversification. But donor support will also be required to enhance the competitiveness of the poorest countries. Facilitating access to new technologies, enhancing technological capability, investment in education, and support for diversification are among the areas in which support is most urgently needed.

The Final Act of the Uruguay Round acknowledges the need for increased technical and financial assistance to the poorest countries, but it is silent on the question of how that assistance is to be generated. In view of the stagnation in flows of concessional assistance and the effective exclusion of the poorest countries from private capital flows, this is an important omission. To date, there is little evidence of donors considering concrete measures aimed at helping aid recipients to benefit from the Uruguay Round - and less evidence still of them addressing the question of how to generate new sources of development finance.

Globalisation could help to provide part of the answer. Taxes on foreign-exchange transactions, along the lines proposed by James Tobin some two decades ago, could generate a significant new source of development finance, while at the same time stabilising increasingly volatile financial markets. In this sense, it offers a win-win scenario for developed and developing countries.

One of the virtues of such a tax is its simplicity (for an account of the tax proposal see ul Haq 1996 ed). By imposing a small uniform tax on foreign-exchange transactions, the relative incidence would fall heaviest on yields from short 'round trips', where speculators send a portfolio of finance around the world's financial centres, stopping off to take speculative profits at various sites on route. Currency speculation and high-risk portfolio investments, which helped to create the Mexican crisis, would face the highest incidence, while long-term financial flows linked to trade and investment would be affected only marginally. Northern governments would have much to gain. Monetary and fiscal policy in the industrialised countries is increasingly driven by a concern to appease financial markets, often by keeping exchange rates high, at the expense of output and employment. Moreover, the risk of systemic collapse remains high. Currency instability in Europe during 1994-1995 and the Mexican financial collapse have provided powerful evidence of the destructive power of globalised financial markets. The risk that a future crisis may exceed the capacity of central banks and multilateral financial institutions to contain the effects of speculative transfers is rising, and with it the threat of mutually reinforcing cycles of exchange-rate instability and global financial surges.

Revenues from a tax on foreign-exchange transactions would be sizeable. According to one recent estimate (Felix, UNCTAD 1996), a phased-in 0.25 per cent transaction tax would reduce global foreign-exchange transactions by up to 30 per cent, while generating tax revenues of around $200bn. Large as this sum is, it understates the scale of benefits available. In the developed countries, revenue from a such a tax would reduce the deflationary pressures associated with fiscal deficits, and provide resources for investment and a return to full employment. Developing countries would benefit from faster-growing demand for their exports; and all countries would benefit from the restored confidence in the multilateral trading system which would accompany the end of mass unemployment.

The most direct benefit to developing countries would derive from the use of the tax to finance an increase in development assistance, which would in turn help to facilitate their more equitable participation in the international trading system. Once again, this would not be a one-way street, since increased prosperity in the poorest countries would create demand and employment in the industrialised countries. Finally, the tax would provide a means to address pressing global environmental problems. It could also enable developing countries to gain access to 'greener' technologies geared towards higher environmental standards, thereby reducing the ecological costs associated with trade.

If the potential benefits of a tax on foreign-exchange transactions are so apparent, why are governments so unenthusiastic about adopting it? Partly because of the belief that its implementation would require the creation of a vast international administrative body; and partly because of a conviction that it would be unenforceable (The Economist 1996c). This assessment is based upon a flawed understanding of global financial markets. The seven leading financial centres account for 80 per cent of global foreign-exchange transactions. Financial institutions in these centres are already subject to scrutiny by government authorities. International cooperation by these authorities to prevent evasion and ensure uniform implementation ought not to be beyond the realms of possibility. Such co-operation is likely to be considerably less costly and less difficult in the long run than responding to the effects of exchange-rate volatility caused by currency speculation. As in other areas of international co-operation, citizens are being denied the benefits of international co-operation in adopting a tax on foreign exchange transactions, not because of technical difficulties, but because of the absence of vision and political will.

Part 4: Trade liberalisation, poverty, and distribution

As suggested earlier, there are obvious benefits to be derived from specialisation and trade. At least in theory, the gains to individuals and countries which can emerge from producing in areas of comparative advantage are self-evident. The problem is that, in the real world, people do not enter the market place, locally or globally, as equals - and they do not leave it with equal rewards. Differences in income distribution reflect differences in access to land, credit, marketing infrastructure, information, skills, political power, and a host of other factors. The impact of trade on these differences is not neutral. Where trade increases returns to goods or to factors of production - such as land - controlled by upper-income groups, it can widen inequalities. Similarly, foreign-investment flows and production for export can exacerbate differences in income between richer and poorer regions. Earlier, we outlined some of the broad mechanisms through which globalisation transmits market signals to different social groups. Here we focus on the specific experience of Mexico. As a country which has gone further than most in adopting policies designed to accelerate integration into the world economy, Mexico is an important case-study. Moreover, under the umbrella of the North American Free Trade Association (NAFTA), Mexico is developing particularly close ties with the USA. As such, it provides an insight into the problems of economic integration between countries at widely divergent levels of development. This section also looks at the implications of 'flexible' labour markets for the distribution of benefits from trade.

Trade liberalisation under NAFTA: the case of Mexican agriculture

Research into the economic and distributional impact of NAFTA provides an instructive cautionary tale. Econometric analysis suggests that Mexico will gain in net income terms as a result of further specialisation and trade with the US (Ros, 1995). But, behind this overall projected income gain is a complex picture of winners and losers. In the agricultural sector, the winners will be concentrated in the fruit and vegetable producing areas. These account for around 40 per cent of the total value of Mexico's agricultural exports (Appendini, 1995), but only 6 per cent of arable land area, with production concentrated on predominantly large-scale, irrigated farms along the North Pacific coast, the irrigated valleys of El Bajo, and lowland areas such as Michoacan, Guerrero, and the coastline along the Gulf of Mexico. NAFTA has already contributed to a dramatic increase in investment in these areas, with large farms or firms leasing land.

The losers from free trade will be concentrated among producers of maize, the country's staple food. Maize accounts for around half of Mexico's agricultural land area - and maize production occupies a pivotal position in maintaining rural livelihoods, generating income, and ensuring food supplies. Most smallholders are net-deficit households, selling maize in the post-harvest period and carrying out wage work to buy it after household supplies have run out. The vast majority of Mexico's maize farmers are operating on poor land, with limited access to credit, inputs, and equipment. Yields in rain-fed areas average around one fifth of the average for the US Mid-West, against which the maize sector will be required to compete as trade restrictions are withdrawn. Estimates of the number of livelihoods which will be lost in the maize sector as a consequence of trade liberalisation vary. According to one study, between 700,000 and 800,000 livelihoods will be lost as maize prices fall, representing 15 per cent of the economically active population in agriculture (Levy and van Wijnbergen, 1993).

If this assessment is accurate, it has profound implications for rural poverty, and for regional inequality. According to the World Bank, the depth and scale of poverty in Mexico has been worsening in rural areas. Over 30 per cent of the rural population live below the income poverty line, and the poorest rural areas are characterised by significantly lower access to water, electricity, and housing (Deninger and Heinegg, 1995). With real wages in rural labour markets declining and unemployment rising (Appendini, 1995) it is unlikely that an increase in off-farm employment will compensate for income losses from maize. As a result, households will be forced into increasingly desperate survival strategies, including labour migration to commercial farm areas, to urban centres, and to the US. There are important consequences for women. Research on the Tarascan plateau of Michoacan has shown how male labour migration increases the workload on women and children. The withdrawal of children from schools in response to these pressures is one of the prime mechanisms for transmitting poverty across generations (Moser, 1996). At the same time, there has been a sharp increase in the frequency with which women are forced to migrate in search of work as day labourers. It has been estimated that women now comprise about one third of all the day labourers working in the Mexican countryside (de Alcantara, 1992; Lourdes, 1989; and Oswald, 1990). To the extent that liberalisation accelerates these trends, it will exacerbate problems of inequality and rural poverty.

Liberalisation and poverty in Mexico

The Mexican case is particularly instructive for the insights which it provides into the accuracy of broader claims about the benefits of liberalisation and structural adjustment. Since the mid-1980s, Mexico has been an international pace-setter in pursuing policies conducive to globalisation. Financial markets have been deregulated, the agricultural and manufacturing sectors have been exposed to increased competition through the reduction of trade barriers, and public assets - including most of the commercial banking system - have been privatised on a large scale. By the early 1990s, almost 90 per cent of imports fell into the non-controlled category. All of this marks a profound departure from the period up to 1980, when Mexico remained a highly regulated economy (on the transition see Coorey, 1992; Apse, 1994). Indeed, in many respects, the shift has been as revolutionary as in the former communist countries.

Charting trends in inequality and poverty across this period is difficult, because of problems with the comparability of data. It is also impossible to separate the effects of liberalisation from other factors, including the debt crisis of the 1980s and the financial crisis of 1994. With these provisos in mind, some disturbing trends emerge. For example, between 1989 and 1992, the richest 5 per cent of the population increased their share of income from 24 per cent to 29 per cent of the total. The income of the poorest 5 per cent fell over the same period from 0.6 per cent to 0.5 per cent of the total (Panuco-Laguette and Szekely 1996). Capturing the wider trends in distribution, Mexico's Gini Coefficient increased from 0.47 in 1984 to 0.53 in 1992. Since 1992, the poorest half of the population has suffered a further decline in its share of national income (Castaneda 1996).

Turning to the incidence of poverty, we see a marked deterioration in the extent, depth and severity of poverty between 1984 and 1989. This is reflected in increases in the headcount ratio and the widening poverty gap. For example, the proportion of the population categorised as extreme poor increased from 19 per cent in 1984 to 24 per cent in 1989. After 1989, the number of individuals living in extreme poverty in urban areas decreased slightly until 1992. In rural areas, where over 80 per cent of those living in extreme poverty are located (Deninger and Heinegg 1995), the absolute numbers of poor people increased throughout this period, rising from 6.7 million to 8.8 million (McKiley and Alarcon 1996).

While evidence for the period since 1992 is sparse, it is likely that poverty levels have worsened in the wake of the 1994 financial collapse. More than one million Mexicans lost their jobs during the crisis as the economy was thrown into reverse gear. Financial adjustment pressures were transmitted to the poor through deteriorating public sector provision, in addition to lost employment and lower wages. As the government cut public spending to meet its debt payments and comply with restrictions imposed by the United States and international financial institutions, the threadbare social safety-net on which vulnerable Mexicans depend is disintegrating.

Even before 1995, the growth strategy pursued in Mexico was contributing to structural problems of poverty and inequality. Central to that strategy were the twin (and inherently contradictory) objectives of maintaining the value of the peso at levels agreed with the US Treasury and the IMF, while boosting export competitiveness. By 1994, the current-account deficit had grown to 8 per cent of GDP (higher than in the 1992 debt crisis), forcing the government to attract speculative foreign capital by issuing government bonds at increasingly high interest rates, and for diminishing maturities. It was this lethal cocktail which led to the financial collapse. Less well publicised was the crippling effect on local industry of high interest rates, which deterred investment and undermined employment creation. Employment was also lost through the surge in imports sustained by currency over-valuation. Between 1988 and 1995, 6.7 million Mexicans joined the economically active population, while fewer than one million new jobs were created (Herdia and Purcell 1995).

Since 1995, there has been a strong export-led economic recovery, especially in the maquiladora sector, where imported components are assembled for export back to the US. While exports grew by over 25 per cent a year in 1995 and 1996, production for the domestic market and domestic investment has collapsed. Domestic demand fell by 70 per cent in 1995 (Casteneda 1996). The problem is that the export sector remains, in many respects, an enclave economy. Just 750 companies account for over 80 per cent of total exports. Moreover, almost all of the inputs into the maquiladora zone are imported, limiting the employment-creation and income-distributing effects of export-led growth. This mirrors the situation in the agricultural sector, described above.

The upshot is that liberalisation and deregulation in Mexico have provided widely divergent sets of opportunities and threats to different regions and social groups. For owners of capital, the privatisation of State industries and the 1992 land reform, which allows investors to purchase smallholder land, have created new sources of wealth accumulation. In the midst of one of the most severe economic crises which the country has ever faced, the number of billionaires increased from ten to fifteen. In 1996, their combined wealth was equivalent to 9 per cent of Mexico's GDP. Large-scale commercial farms and private industries geared towards the North American market have also benefited. New jobs are being created in the northern states which span the maquiladora zone and in commercial farm areas along the Pacific coast and irrigated valley of El Bajo. But these are not the states in which the social dislocation and loss of livelihoods are occurring on the largest scale.

While the contrast between a poor south and a more prosperous north is an over-simplification, this divide is likely to widen alongside a wider urban-rural divide, as traditional agriculture declines in states such as Chiapas, Guerrero and Yucutan, where poverty is most intense and pervasive, while industrial expansion occurs around industrial conurbations in the states of Mexico, Nuevo Leon, and Sonora. These conurbations, which lack basic social infrastructure and health and education provision, will act as a magnet for impoverished displaced rural labour. Yet on current trends, insufficient jobs will be created to absorb the victims of market deregulation in agriculture. For these victims, migration across the Rio Grande will remain the only option. More generally, as the revolt in Chiapas and the mounting level of political violence in Mexico testify, serious questions have to be asked about the wider social implications of continuing along the current economic policy path.

Some parallels with the corn sector in the Philippines

Mexico is not an isolated example of the problems associated with trade liberalisation. In the Philippines, the government is liberalising the corn market as part of its obligations under the Uruguay Round. Tariffs on imports will be halved over the next eight years and minimum import quantities expanded (the following is based on Oxfam 1996). According to the Philippines government, the resulting competitive pressures will lead to major efficiency gains in domestic agriculture. In more concrete terms, the upshot is that, depending on world price trends, corn imported from the USA could be available at prices 30 per cent below current market prices by the end of the decade. Domestic farmgate prices will decline towards this import-parity level.

What does this mean for producers in the Philippines? Corn is the second most important crop in the country after rice, with around 1.2 million households involved in production. Poverty levels among these households are intense and pervasive. On the island of Mindanao, the main corn-producing area, over half the population lives below the poverty line, with incomes insufficient to meet basic needs for nutrition, shelter, and clothing. Human welfare indicators in corn-producing areas are more reminiscent of sub-Saharan Africa than south-east Asia, with one third of all children below the age of five suffering from malnutrition. Against this backdrop, any decline in household income will have potentially disastrous effects. Many of the poorest households derive over three quarters of their income from corn sales. As their income declines with falling prices, resources available for food, health, and education will be diminished, with large-scale migration a serious prospect. According to Oxfam, up to half a million livelihoods could be lost. Set against these social costs, the view of the Philippines government is that, in the long run, trade liberalisation will increase average incomes. Viewed from Mindanao, Keynes' observation that "in the long run we are all dead" carries rather more resonance.

Flexible labour

Even the most committed free-traders acknowledge that liberalisation will impose costs on some social groups, especially in the agricultural sector. However, it is widely claimed that these costs will be offset by increased opportunities for export-oriented commercial agriculture, with the income gains outweighing the losses to generate a positive net welfare effect. Up to a point, this argument would appear to be justified. In much of Latin America and in the Philippines, commercial agricultural production is rapidly expanding, creating demand for labour in the process. Once again, however, it is important to know how income generated by commercial agricultural exports is being distributed.

It is not possible to address this question in the general terms favoured by free-trade theory. Other things being equal, income flows will tend to reflect the distribution of assets used in production. Thus in countries such as Brazil and Zimbabwe, where export production is dominated by large-scale farms, the bulk of income gains generated by trade will accrue to large landowners, whether in the form of foreign-exchange revenues or increased rent. For lower-income groups, the income benefits will be derived from increased employment, so the terms and conditions of employment are of crucial significance in estimating the impact of trade on human development.

Unfortunately, employment conditions in export-oriented agriculture are often highly exploitative and insecure, with temporary, often seasonal, wage employment dominating. In Chile, which has witnessed a spectacular growth in fruit and vegetable exports, approximately two thirds of wage labour is employed on a temporary basis, double the level at the start of the 1980s. The expansion of temporary employment has in turn been accompanied by the 'casualisation' of labour. Temporary workers are generally paid by piece rates, are not usually entitled to social security protection, have no sickness or maternity rights, and enjoy limited trade union rights (Kay, 1995). Women figure with increasing prominence in casual labour forces, from the flower sector in Colombia to the grape pickers of Chile (Stephen, 1991).

These are specific cases in which trade liberalisation and globalisation offer, at best, a mixed blessing. But there is mounting body of empirical research to suggest that the wider results are less benign than is often assumed. One recent detailed empirical study evaluating the effects of trade liberalisation in Latin America has concluded that "the new economic model has done little to improve poverty and has a tendency to harm income distribution" (Bulmer-Thomas ed., 1996). Real minimum wages fell substantially in almost all countries as local industries adapted to increased competition from imports (Fitzgerald, 1996).

Even the experience of Chile, widely cited by the World Bank as the liberalising success story of Latin America, raises serious questions about the linkages between liberalisation, poverty reduction, and inequality. During the 1980s, the proportion of the population categorised as "extremely poor" increased, from 12 per cent to 15 per cent of the population. After 1990, poverty rates began to fall as economic growth accelerated and average income rose. However, between 1990 and 1992, the share of national income allocated to wages actually fell, while the share of national wealth enjoyed by the wealthiest 10 per cent remained at around 45 per cent. Thus while the number of people living in poverty is declining, the degree of inequality is not. Another disturbing feature of the Chilean experience is that around two thirds of the poor are in employment. This suggests that labour-market 'flexibility' has become a euphemism for the creation of jobs at sub-poverty-level wages.

After over two decades of intensive economic reform, hailed around the world as a model of its kind, the Chilean example - like that of Mexico - raises wider questions. In particular, it is unlikely that other countries in the region, starting from even more intensive levels of poverty and inequality, will be able to implement the model with such little concern for its impact on equity (Whitehead 1996). Despite this, there has been little effort to integrate policy reforms capable of achieving more balanced distributional outcomes.

Regional factors

There are important regional dimensions to the impact of trade and investment flows on poverty and distribution. In much of the literature, globalisation is treated as a force integrating national economies within the international system. Often, however, the locus of integration is geographically concentrated. In Vietnam, over two thirds of the $5.5bn in foreign investment which has poured into the country since 1988 is concentrated around the centres of Ho Chi Minh, Hanoi, and Dong Nai, exacerbating the already wide income gaps between these areas and the poorer mountainous regions. In China, investment resources are heavily concentrated in Guangdong and Fujian provinces, where average incomes are rising rapidly. The income gap between coastal and inland regions is widening dramatically, with potentially damaging implications for social stability. By 1992, per capita urban incomes had risen to 2.3 times rural incomes, and there are big regional differences in education, health, and social security provision (Knight D and Song C 1995).

Such trends are likely to continue in accelerated form. This is because globalisation is increasingly being driven by localised growth points. In south-east Asia, the growth triangle of Southern China, Hong Kong, and Taipei China is one of the most dynamic regional economic entities. In 1990, total trade within this triangle was estimated at $60bn, with Guangdong province receiving about 40 per cent of incoming investment from Hong Kong and Taipei. Other growth-triangle arrangements are being formalised. One of the most prominent is that centred on Singapore, the southern part of the Malaysian state of Johore, and the islands of Riau province in Indonesia (Asia Development Bank 1996). The shift of labour-intensive industries from Singapore to Johore and the Riau islands is being accelerated, as companies capitalise on the availability of low-cost land and labour in the Riau islands, and semi-skilled labour in Johore. Inevitably, the surge of foreign investment into these localities will exacerbate intra-regional differences.

At a national level, government strategies are fuelling the drive towards a concentration of investment resources. For instance, the Philippines government's medium-term development plan is premised upon the creation of a number of growth corridors, with tax incentives being provided to foreign investors. Apart from exacerbating regional inequality, such strategies can pose powerful threats. In the Philippines one of the most conspicuous effects of the growth-corridor strategy has been an explosion in land values, with foreign investment and speculative activity driving up prices. One of the uglier consequences in Manila was the brutal removal of Smoky Mountain, a sprawling urban slum built upon a garbage tip, to make way for South Korean companies to build warehouses. Similarly in Batangas province, communities without established tenancy rights have been served with eviction notices by landowners seeking to capitalise on high land prices.

Trade and poverty: some concluding comments

It must be stressed that trends towards income inequality are not inherent in globalisation. The commercialisation of production for domestic markets can have similar effects, as can domestic investment activity. However, to the extent that liberalisation exposes domestic producers to more volatile global markets and flows of capital which are large in relation to the local economy, integration in global markets is a policy choice which carries high risks, as well as the potential for delivering considerable benefits. Viewed from a poverty-reduction perspective, the challenge is to identify policies which enable poor people to participate more equitably in markets, at both the national and the global levels.

With appropriate policies, the poor can participate in trade and growth, and contribute to both. Marked declines in poverty and inequality are consistent with trade-led growth, as witnessed by the experience of south-east Asia. By contrast, where growth processes are grafted on to highly unequal social structures and inequitable international trading arrangements, the distribution of benefits will reflect this. What is needed is a new paradigm for measuring the value of trade, for example by monitoring the distribution of export earnings and increments to growth. At the risk of stating the obvious, citizens in South Korea derive a considerably higher benefit from each dollar of export earnings than their counterparts in Brazil. Including these distributional effects in an internationally recognised index - such as the Human Development Index - would go some way towards countering the received wisdom which holds that all trade is good trade.

Part 5: Governments and markets

The globalisation of economic relations has been accompanied by the growing dominance of neo-classical development economics and the resurgence of a laissez faire development paradigm. With some variations, that paradigm is characterised by a number of central themes. Prominent among them is the view that unregulated markets will optimise resource allocation, that government interventions in markets are inherently inefficient, and that the most appropriate development policy is to leave it to the market. Trade liberalisation is one of the keystones of this paradigm. The claim is that trade liberalisation will enhance efficiency and international competitiveness, enabling a higher rate of growth to be achieved than would be the case under protected conditions. For the World Bank the presumed benefits of openness to trade and investment flows are self-evident - as are the costs of attempting to defy markets:

Trade liberalisation constitutes perhaps the most important opportunity for raising the welfare of both developing and industrial countries in the longterm...Globalisation comes with liberalisation, deregulation, and more mobile and potentially volatile cross-border flows...Penalties for policy errors arise. Globalisation thus requires closer monitoring and quicker policy responses at the country, regional and global levels (World Bank 1995b).

In the World Bank view, good development policies are those which accelerate integration into global markets by removing barriers to the free flows of goods, services, and investment. Apart from optimising growth, it is assumed that trade liberalisation will have positive outcomes for poverty reduction and income distribution. This is because of a presumption that market deregulation, including the deregulation of labour markets, will increase returns to the assets of the poor, notably labour. The assumption here is that labour and natural resources will be used more intensively, and the price of agricultural goods produced by the poor rise, as exports and imports adjust to comparative advantage, as the cost of previously subsidised capital rises to market levels, and as taxation on agricultural output (both direct and in the form of exchange-rate over-valuation) declines (Corden, 1993). Once again, the incentives towards increased efficiency created by competition from imports and improved access to export markets are central parts of the equation.

The WTO is emerging as a central force in implementing the paradigm for deregulated globalisation outlined, especially with regard to investment. During the Uruguay Round, the industrialised countries lobbied vigorously, but with limited success, for a new legal framework which would enhance the rights of foreign investors in host countries. They are now pressing for the integration into the WTO of an OECD foreign-investment treaty. One version, drafted by the EU, decries the "host of barriers that prevent foreign investors entering host countries freely", citing restrictions on ownership (such as the requirement that foreign investors engage in joint ventures, or include domestic investors in share equity), profit repatriation; local contents requirements (under which foreign investors are required to source from local suppliers); and requirements that foreign investors provide the foreign exchange to meet their import costs for balance-of-payments reasons (Khor, 1996b). It concludes: "In general, the host country should treat the foreign investor and his investment /sic/ operating in its territory in the same way as the domestic investor." All sectors would be covered except for defence. If adopted by the WTO, the investment code would be enforced by the threat of trade sanctions - in effect, reinforcing deregulation through international trade law.

The strength of the dominant development paradigm, with its focus on trade and investment liberalisation, can be traced in part to the failures of import-substituting industrialisation (ISI) models, even though the picture here is more mixed than the World Bank or WTO would concede. But whatever the failures of ISI, it cannot be taken as axiomatic that unregulated markets will either optimise efficiency, or distribute the benefits of growth equitably. This is not to suggest that 'protectionist' alternatives are inherently more likely to succeed. If the economic history of the post-war period teaches anything, it is that universally applied and ideologically motivated blueprints for development, of both the free-trade and protectionist variety, are a prescription for social and economic disaster. The lurch from State planning to unregulated capitalism in eastern Europe provides the most recent evidence in support of this proposition.

For trade growth to be translated into human development, two broad conditions would appear to be essential. First, countries need to climb the value-added chain. In a world economy dominated increasingly by the production and exchange of knowledge-intensive goods, dependence upon primary commodities and labour-intensive goods is unlikely to provide a lasting foundation for prosperity. Second, concrete policies are needed to distribute the benefits of trade equitably. In this context, public-spending policies and the distribution of income and productive assets have a crucial role to play in creating an enabling environment in which poor people can participate in markets on beneficial terms.

Building domestic capacity: first and second generation NICs

One of the lessons of the East Asian 'model' is that there is no model. Instead, there are a variety of forms of government interventions aimed at promoting trade, and at distributing the benefits across society. If there is a single lesson to be learnt, it is that, contrary to the claims of the dominant development paradigm, government intervention can play a crucial role in facilitating the development of the technologies needed for successful participation in global markets (Singh A 1995). Regulation of foreign investment in the interests of enhancing domestic capacity and employment has also been much in evidence. Considered collectively, government interventions in south-east Asia have not been geared solely towards the creation of a dynamic comparative advantage, rather than the exploitation of static advantage. Without these interventions, it can be said with some certainty that South Korea would not be a major manufacturing power; and that its citizens would not enjoy living standards approximately those of the poorer OECD countries.

The closest approximation to the globalisation model favoured by the World Bank in Asia was Hong Kong, which combined free trade with an open door to foreign investment. As Lall (1994a) has pointed out, however, Hong Kong was a very specific case, with a huge population of entrepreneurs and skilled engineers, well-established entrep™t links, and a large contingent of foreign transnational companies. After the mid-1980s, the island's manufacturing sector went into decline as labour and land prices rose. In the absence of a coherent strategy for promoting technological innovation into higher value-added products, the bulk of the manufacturing sector was relocated to China. While prosperity has continued to rise on the back of the financial-services sector, "the lack of industrial deepening and de-industrialisation, a direct result of the absence of industrial policy, would be very undesirable in other developing countries" (Lall 1994a).

At the other end of the spectrum, South Korea adopted highly interventionist policies. Foreign direct investment was restricted, with the government keeping control over industries in local hands. Foreign investment in the 1980s was equivalent to 0.5 per cent of domestic investment, compared with over 10 per cent for Hong Kong and 17 per cent for Singapore. Majority ownership was not permitted, except where it was tied to the transfer of technologies or the promotion of exports (policies which the industrialised countries now want the WTO to prohibit). In place of foreign investment, South Korea relied upon the adaptation of imported technologies, using a variety of incentives to persuade firms to obtain the most advanced equipment. One of the pillars of Korea's development strategy was the promotion of technological deepening. The deliberate creation of large private conglomerates, the chaebol, was part of this strategy. These conglomerates were protected from competition from foreign TNCs in return for investing in technological upgrading, geared towards export markets (UNCTAD 1995b).

Another part of the South Korean strategy was the use of tax incentives and/or lower import duties to promote investment in industrial research, human-capability upgrading, and the adoption of new technologies. Direct subsidies for private research have also figured prominently. The Designated Research and Development Programme, established in 1982, and the Industry Technology Development programme, created in 1987, are two example. More recently, the Highly Advanced National Project was created to support investment in high-technology products, with $350m being invested in the first two years. Direct government intervention was also used to create a massive research and development infrastructure, and to promote technical skills. Today, South Korea has the highest level of educational attainment of any developing country, and it produces more qualified engineers than the whole of India (Lall 1994b).

Like South Korea, Singapore invested heavily to upgrade its technical skills base. Unlike South Korea, however, it relied upon foreign investment as the vehicle for transferring technology, rather than attempting to develop an indigenous capacity. Instead, the government has used incentives to direct foreign investment into specialised high value-added areas such as fibre optics, precision instruments, aircraft servicing, banking, and insurance. There was no systematic policy for encouraging local-content use, and the technological depth of domestic enterprises remains limited. However, the high skills-base of the economy, which has been further increased through training programmes by foreign investors, has facilitated a sustained increase in real wages and export earnings (Lall 1993).

The second generation of 'tiger economies' have also been highly interventionist. In the early 1970s, the government of Malaysia actively promoted the country as a site for firms relocating their more labour-intensive operations from Singapore and the industrialised countries. The combination of low wages, good infrastructure, and its location at the hub of a dynamic region provided an incentive. Manufacturing exports boomed, leading to a structural transformation of the economy. The share of manufacturing in GDP increased from 14 per cent in 1970 to 30 per cent in 1993, when the country was exporting some $34bn in manufactured goods. The 'skill intensity' of these exports was even higher than for South Korea (Lall, 1994 a), setting Malaysia apart from Thailand and Indonesia, where foreign investment has been more concentrated in low-technology assembly activity (UNCTAD 1995b).

In contrast to South Korea, the technological base in Malaysia has remained small and underdeveloped. While the government has pursued an active industrial policy, this has applied principally to creating public enterprises, promoting bumiputra, and assisting small and medium-sized firms. Until recently, however, there was little effort to promote linkages between domestic firms and foreign investors operating in the export sector. Transnational companies used Malaysia primarily as a site for the assembly and re-export of imported parts. Few linkages were established with domestic firms; local content in the export sector is minimal; and foreign investors have done little to upgrade the technological skills-base of the economy. Thus while Malaysia's export structure became more skill-intensive, it has not attained the levels of local research, design, and technological capacity of countries such as South Korea and Taiwan. The export sector has remained isolated from the rest of the economy, and Malaysia has remained at the assembly stage of production.

These limitations act as a constraint on upgrading production facilities and increasing productivity. As a consequence, rising real wages have become a deterrent to foreign investment. Recorded approvals of foreign-investment projects have fallen sharply since 1993, and many companies are relocating to lower-wage economies such as the Philippines and Vietnam (Far Eastern Economic Review various). It remains to be seen whether Malaysia will sustain its export growth path. However, its experience does highlight some of the risks faced by countries which fail to establish dynamic linkages between foreign investors and the domestic economy. In a situation where investors are able to transfer operations relatively easily from higher-to lower-wage economies, cheap labour is likely to prove a transitory incentive. The danger, faced in acute form by Malaysia and Thailand, is that the transfer of foreign investment will increase unemployment and reduce real wages, with damaging implications for poverty.

Quality and quantity in foreign investment

This points to a wider issue in relation to foreign investment. Too often, the presumed benefits of foreign investment are measured in quantitative rather than qualitative terms. Investment geared towards the pursuit of short-term profit, with a view to repatriating the bulk of it, is unlikely to generate long-term benefits. In some cases it may generate considerable costs. Foreign investors operating with a short-term time horizon are more likely to engage in socially exploitative labour practices and to treat the local environment as an expendable resource, than are those who develop a long-term stake.

In Mexico, US car and electronic companies began establishing themselves on a large scale in the maquiladora zone near the northern border in the 1980s. The vast majority assemble components for re-export to the USA. Skills training is minimal and employment conditions are poor (Coote, 1994). Moreover, maquiladora plants are a major source of environmental pollution. The discharge of heavy metals and toxic emissions from factories have caused severe air and groundwater pollution on both sides of the border, with damaging consequences for public health. In one report, the American Medical Association described the maquiladora region as "a virtual cesspool and breeding ground for infectious disease".

At the other extreme is the Ford-Mazda plant in Hermosillo. Established in 1986 with an investment of $500m, the plant produces Tracer cars for export and for the domestic market. From the outset, important linkages were established with the domestic market. In 1990, plants located in Mexico accounted for one third of the value of components used in Hermosillo, compared with 17 per cent from the USA and 50 per cent from Japan. The plant utilises the same technology as Ford plants in the USA, and has productivity levels and quality standards to match (Carillo, 1995). Investment in training reflects the higher skill levels required, with all new workers receiving 700 intensive classroom hours (UNCTAD, 1994). The workforce in Hermosillo is unionised, and wage levels are above the average for the organised industrial sector. This picture contrasts with the typical plant operating in the maquiladora zone, including Ford plants. For example, the company's plant at Favesa is a low-skill assembly operation, with low levels of training, a high turnover of employees, and no union. Linkages with domestic suppliers are weak.

What can be learned from these various experiences? At one level, not a great deal. As the World Bank and it is fellow globalisers rightly claim, the South Korean example is not easily replicable. That said, South Korea did not 'replicate' the Japanese model; Taiwan's experience was not the same as South Korea; and Singapore's economic policy choices were different again. Each of these cases offers a variation on the theme that State intervention can play a vital role in creating the conditions for sustained trade growth; and in ensuring that trade expansion translates into poverty reduction. Efforts at rewriting history with a view to fitting the East Asian experience into the straight-jacket of neo-liberal economic theory have been far from convincing (Wade 1995; Lall 1995), with the World Bank's East Asian Miracle study (World Bank 1994) among the more abject failures.

The issues raised by south-east Asia's experience are of more than theoretical importance. Under the globalising trade regime of the WTO, many of the most important policy interventions behind East Asia's success would fall foul of international trade law and, in all likelihood, attract sanctions - a fact which the World Bank acknowledged in its East Asian Miracle study. The use of subsidies to promote capacity building and technological upgrading, restrictions on foreign investment, import protection, and the adaptation of imported technologies all fit into this category.

Set against the examples of successful State intervention, it would of course be possible to develop a long list of failures. From Eastern Europe to Africa and Latin America, the specific forms of protection adopted by governments have failed to provide a basis for sustained growth, or - in the case of Africa and Latin America - to reduce poverty and improve income distribution (Bulmer-Thomas, 1996). The protection of inefficient capital-intensive industries imposed a huge burden on public spending, with losses covered by public subsidies. In contrast to South Korea, where protection was applied selectively and for a limited period of time, after which companies were expected to operate in a competitive market, in much of Africa and Latin America protection was applied on a continual basis without regard to efficiency. The outcome was an industrial and manufacturing structure which was uncompetitive, and in which employers were provided with incentives to invest in capital rather than labour. Instead of fostering potentially competitive industries, protection often encouraged inefficiency and created windfall profits for those with access to import licences. Trade liberalisation can play an important role in addressing these problems. However, in the absence of a coherent economic strategy, it can also exacerbate underlying structural weaknesses, as the experience of trade liberalisation under structural adjustment in much of sub-Saharan Africa illustrates. Whatever the failures of state intervention, however, it is far from clear that they have been overcome through deregulation and liberalisation. For instance, under structural adjustment programmes, liberalisation has been seen as a prerequisite for the growth of non-traditional exports and for the manufacturing sector (Husain, 1993). It is also intended to increase the efficiency of domestic industry through the discipline of competition and improved access to imported technologies. In fact, import liberalisation has often failed to improve export performance and investment, undermining local industry in the process - an outcome which is not surprising, given that a swift removal of protection can be highly disruptive, even for countries with a strong industrial base (Stewart, 1993).

In many countries, the most immediate effect of wholesale import-liberalisation has been to widen trade deficits, in many cases following a sharp increase in imports of luxury consumer-goods. This in turn results in pressure to devalue, raising the costs of all imports, or to restrict demand through deflationary policies. The first option has the effect of restricting access to the essential imports needed to increase productivity; the second undermines investment through higher interest rates and a contraction of the domestic market. Across much of sub-Saharan Africa, the combination of a sudden loss of protection, devaluation, demand restraint and increases in real interest rates has resulted in a process of de-industrialisation (Stein, 1992), with local industries becoming decapitalised. Inadequate import capacity has been another factor behind the limited supply-response resulting from liberalisation, and one of the causes of sub-Saharan Africa's loss of market shares. Labour-intensive textile industries have collapsed in Ghana and Zambia, and come under severe pressure in Zimbabwe. In each of these cases, there has been a strong case for restructuring. But coherent restructuring requires an active role for the State, including a selective and carefully phased approach to liberalisation (see, for example, Riddel, 1990). It would appear to be especially important for the State to manage import and foreign-exchange regimes with a view to maximising the availability of the capital and intermediate goods needed to enhance competitiveness and employment, where necessary by controlling imports of luxury items.

It is not only in sub-Saharan Africa that the real effects of trade liberalisation have contradicted the theories underpinning globalisation. In Latin America, export volumes rose steadily in the 1980s, but with only a marginal effect on traded-sector employment (Fitzgerald, 1996). Moreover, the average compound-growth rate was insufficient to compensate for the loss of import-competing jobs. With the exception of Brazil, Mexico, and Colombia, no country succeeded in diversifying its exports on a significant scale, and real wages declined. As one of the more comprehensive reviews of the empirical evidence puts it: "the new trade regime as it currently operates has little potential for dynamic growth, as it is still largely based on limited primary export markets and cheap labour, with insufficient attention to industrial export expansion on the one hand and the sequencing of trade reform in order to sustain macro-economic stability and thus investment on the other" (Fitzgerald, 1996). By comparison with south-east Asia, the deregulatory regimes adopted in Latin America have achieved derisory results in terms of economic growth and poverty reduction. Between 1980 and 1993, average annual growth for Latin America was less than 2 per cent, compared with over 5 per cent in the 1970s; investment levels are falling; and the rate of export growth has fallen by one half since the 1970s. 

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

Part 1: Unemployment and inequality

The post-war settlement of the late 1940s was built upon a political consensus in the industrialised countries that an open and expanding international trading system was crucial to economic growth, employment, and political stability. That consensus was reinforced by the memory of the 1930s, with the mass unemployment of that period widely attributed to the collapse of international trade and beggar-your-neighbour protectionism. The liberal multilateral trading order enshrined in the GATT and the Bretton Woods agreement reflected a concern to avoid a return to this era - and a conviction that increased trade would act as the handmaiden of shared prosperity. From 1950 to the early 1970s, the so-called 'golden age' of prosperity and stability in the industrialised world, the consensus was virtually unchallenged. Low unemployment and the creation of welfare systems rendered the insecurity of the 1930s a distant memory. Critics such as J K Galbraith challenged the virtues of mass consumption and meaningless automated jobs, but few questioned the capacity of the global trading system to create employment and improve living standards; and nobody anticipated the crisis now facing the industrialised countries.

That crisis can be summarised in three words: employment, wages, and security. The post-war settlement was premised upon a conviction that deeper integration into the global trading system would provide all three, albeit against a backdrop of cyclical recession and recovery. Today, there is a deepening sense in Europe, the USA, and even in Japan that something has gone wrong.

In the US, the core of the problem is low wages (Freeman R and Kate L ed 1996). Unemployment remains low by comparison with the OECD average, but an increasing proportion of those in work receive poverty-level wages. Inequality has reached levels not witnessed in over half a century. In Europe, the wages of those at the bottom end of the labour market have fallen by less in relative terms than those of the US. But long-term structural unemployment has reached crisis proportions, and the security of those in work is under threat from the rise of casual employment practices. Some countries - notably Britain - have achieved the worst of both worlds, combining increased inequality and poverty with mass unemployment. In Europe, political debate is dominated by a discourse which poses trade-offs between wages and employment, with the US model of a deregulated, flexible labour market built upon low wages seen as the key to job creation (Krugman 1995). Welfare states are being dismantled and labour rights restricted to create the conditions for the emergence of 'flexible' labour markets. On both sides of the Atlantic there is deepening concern that a growing army of people are being economically disenfranchised, excluded from a share in the prosperity of what are still enormously wealthy countries.

Political background

The emergence of these disturbing trends in OECD labour markets has coincided with the latest phase of globalisation, including the rapid growth of manufactured exports from developing countries; and an increase in foreign investment. Not surprisingly, the conviction is gaining ground that globalisation is a cause of mass unemployment and inequality. While governments seek to accelerate globalisation by removing barriers to trade and investment, the consensus which underpinned earlier moves towards liberalisation is collapsing. At one end of the spectrum, self-made billionaires in Europe and North America have warned that increased trade with low-wage nations will impoverish workers in the North, destroying the social fabric of society in the process. While the evidence upon which this claim is based may be threadbare, the populist appeal of the argument is not in doubt. One of the billionaires in question, Sir James Goldsmith, has written a condemnation of free trade, The Trap, which has become a European best-seller. His US counterpart fought a campaign against NAFTA, claiming to hear "the giant sucking sound" of US jobs being transferred to Mexico. In both Europe and the USA, powerful new alliances have emerged behind a banner of protectionism, with developing countries the target. At the other end of the spectrum, the European Commission and a number of economists have identified competition from imports as one of the structural factors behind unemployment (Commission of the European Communities 1993). While most of this group have shunned protectionism, no coherent alternative has emerged.

For the most part, governments have responded either by denying the problem, insisting that there is no alternative to increased integration into the global economy, or by offering Utopian long-term solutions to immediate problems. Well-publicised 'jobs summits' reflect the depth of political concern, but have failed to go beyond rehearsing familiar arguments for and against the deregulation of labour markets. For the millions of disaffected people for whom globalisation has become synonymous with inequality, unemployment, insecurity, and endemic poverty, there is no political vision offering hope of a better future.

As we suggest below, protectionism is not the solution to the problems faced by marginal communities in the industrialised world. On the contrary, it could compound those problems by reducing economic growth and employment creation. But, while protectionism may not be the answer, trade with developing countries probably has contributed to unemployment and inequality, although the impact has been exaggerated. Denying the impact of imports on labour markets in order to counter protectionist arguments is a dangerous option - not least because audiences composed of anything other than free-market economists are unlikely to be persuaded. In the absence of broad-based programmes to address the underlying causes of mass unemployment, inequality, and job insecurity, populists offering protectionist and xenophobic alternatives to trade with developing countries will remain in the ascendant. If governments wish to maintain the multilateral trading system, the real fears and social problems faced by their citizens must be addressed, and they must face the challenge of combining trade liberalisation with social justice in labour markets. Failure to provide the vision needed to address problems of poverty and unemployment in an increasingly open global economy will carry a high price for developed and developing countries alike.

Perhaps the most striking way to bring the crisis facing industrialised countries into perspective is to recall the spirit which gave rise to the Bretton Woods system. Above all, post-war leaders had learned that unregulated markets had a vast capacity for social dislocation. The experience of the 1930s persuaded them that the State had a vital role to play in ensuring that equity and growth went hand in hand. In the US, the Truman Administration passed the Employment Act, which established full employment as the objective (Kapstein E 1996). Ambitious social-welfare programmes were enacted across Europe. Meanwhile, trade unionism was actively encouraged as part of a wider contract between States and citizens. Central to that contract, which has been labelled "embedded liberalism", was a commitment to protect, in the interests of political stability, vulnerable groups bypassed by the economic gains resulting from liberalisation (Ruggie 1994).

Over the two decades until the early 1970s, the dominant picture was one of economic and social success. While a hard core of poverty remained, average living standards doubled in the USA and Britain, tripled in France and quadrupled in Germany and Italy (Krugman P 1994). Open unemployment in the US fell to levels which were exceptionally low by historical standards. In Europe, unemployment levels were typically below 3 per cent. Poverty levels declined and social-welfare indicators improved in both Europe and the USA, in part because the benefits of growth were spread widely; and in part because welfare states provided a degree of protection for the unemployed, the sick, and the disadvantaged.

The contrast with the position today could hardly be more striking. In 1995, there were 34 million people out of work in the OECD countries, an average of 7.5 per cent of all workers (OECD) 1994. The figures for Europe have assumed particularly frightening dimensions. In France, unemployment affects 12 per cent of the economically active population - over four times the level in the early 1970s. In Germany unemployment averaged around 1 per cent in the 1960s; today it is over 10 per cent. To put this figure into historical perspective, it is the highest level since the early 1930s. For the EU as a whole, the number of registered jobless has doubled to 11 per cent since 1979. What is particularly disturbing about the situation in Europe is that around one half of the jobless have been without work for at least one year (OECD 1994). As many as one third have never worked at all, forming a class of permanently unemployed.

The US experience

In the US, employment has grown considerably faster than in Europe. Roughly half of America's unemployed find work within a month, compared with only 5 per cent in Europe (Freeman 1995). But a growing share of the labour force has suffered a reduction in pay, and has thereby been excluded from participating in national prosperity. The extent of this exclusion is summarised by one remarkable statistic: in the 1980s, two thirds of the increase in US national income accrued to the wealthiest one per cent of the population (Feenberg and Potera 1994). In the early 1970s, households in the top five per cent of the income bracket earned ten times more than those in the bottom five per cent; today they make almost fifteen times more (Kapstein E 1996).

At the other end of the social scale, the decline in real earnings among the low-paid has had important implications for poverty trends. During the 1980s, there was an increase from 12 per cent to 18 per cent in the number of workers whose earnings fell below the poverty line (Freeman R and Kate C 1996). Such facts help to explain why the number of North American children living in poverty has grown at such distressing rates. In 1992, a report by the National Centre for Children in Poverty estimated that the number of American children under the age of six living below the poverty line had increased by one million between 1987 and 1992, to a total of six million - or slightly over one quarter of the population in that age group (Washington Post 1995). Equally alarming was the finding that 58 per cent of these children came from families in which parents worked at least part time.

For all its success in limiting unemployment, the quality of the jobs created in the USA has been poor. Contrary to predictions confidently made by economists in the 1980s, those who have lost jobs in the manufacturing sector, which has shed 1.4 million jobs since the late 1970s, have not been absorbed by the services sector at higher pay levels (Luttwak E 1996). According to a New York Times survey carried out in 1995, two-thirds of all workers who found jobs after a lay-off did so at lower levels of pay. Thus while average wages have stagnated, real wages at the lower end of the labour market have fallen. The average real hourly wage of males with twelve years of education entering the jobs market today is almost one third lower than in 1979. Relative to average earnings, a North American in the lowest decile of the income distribution earns around one third less than his or her European counterpart.

The capacity of the North American economy to create jobs is viewed from Europe with a combination of envy and awe. Seen from a different perspective, the economic policy mystery of US job creation is the product of clear social and economic policy choices: driving down wages and enabling employers to hire and fire at will creates conditions conducive to job creation. As Freeman puts it: "the working poor in America are really poor" (Freeman R 1995a). Why, then, do poor Americans remain in work? Principally because their country's harsh benefits system threatens the unemployed with poverty, and then delivers on that threat. In Europe, unemployment benefits last longer and are followed by more generous welfare payments. Moreover, Americans who lose their jobs often lose their health-care coverage, whereas Europeans retain access to health services. Another problem in the USA is that labour-market turnover is more rapid, so that job loss is more likely. One recent survey found that over 40 per cent of those in employment expected to lose their jobs either temporarily or permanently over the next twelve months (The Economist 1996).

The European Union

Most of the large European economies have avoided the US experience of growing inequality, and in some cases - including France and Germany - income differences have narrowed. The most striking exception is the UK, which has succeeded in combining slow growth, a poor record on investment and high-albeit partially disguised - unemployment with rising inequity and poverty. After some three decades during which income inequality was falling slightly, the decade after 1979 saw an unprecedented widening in UK income distribution, with the income gap between rich and poor at its highest level this century (Johnson P 1995). In 1977 the income of the richest 20 per cent of British citizens was four times the income of the poorest 20 per cent; by 1991 the multiple had increased to seven. Over the same period, the Gini coefficient rose from 0.23 to 0.34, a bigger jump than for any other country (Jenkins S 1995). If we focus more narrowly on the gap between the highest-and lowest-paid workers, the changing patterns of income distribution are even more striking, with the gap between the highest-and lowest-paid workers the highest since records were first compiled in the 1880s (Glyn and Sutcliffe 1996). According to the OECD, the weekly earnings of a worker in the top decile compared with the bottom decile in Britain have increased from a ratio of 2.4 to 3.3 since 1979. Widening income inequalities have been accompanied by a deterioration in the real incomes of the lowest-paid sections of society. For instance, by 1992, hourly wages for the lowest-paid decile of male workers were lower than in 1975.

As in the US, the impact of income-distribution shifts on poverty in the UK has been considerable. Using the European Council's definition of poverty (income levels of less than half the national average), the number of individuals living in poverty increased from 5 million UK (9 per cent of the population) to 13.9 million (25 per cent) between 1979 and 1991. The number of children living in poverty tripled over the same period, from 1.4 million to 4.1 million (Dept of Social Security, HMSO 1994). Admittedly, these figures primarily capture the impact of relative income shifts. But they nonetheless point to the exclusion of the poorest sections of society from the benefits of economic growth. For the poorest tenth of the population, average household incomes fell by 17 after deduction housing costs (Low Pay Unit 1994). As one recent survey concluded: "the poorest 20-30 per cent of the population failed to benefit from economic growth, in contrast to the rest of the post-war period, when all income groups benefited during times of rising living standards" (Joseph Rowntree Foundation 1995).

Wealth and poverty

There is a paradox at the heart of current debates over globalisation. Industrialised countries are more productive and richer than ever before, even though growth slowed during the 1980s and has remained low by post-1945 standards in the 1990s. Yet economic misery in the form of inequality and mass unemployment is on the increase.

Why should this be the case? That question dominates a vast body of technical economic literature. However, the issues at stake are of considerably more than technical interest. Behind the facts and statistics which capture the extent of unemployment inequality, there is a story of human tragedy, isolation, and despair. In cities such as Hamburg, Rotterdam, Paris, and London, a facade of rising prosperity, property booms, and rampant consumerism bears testament to the sustained rise in average incomes. Yet that facade obscures a reality of maginalisation, polarisation, and deepening social tension. Some commentators have drawn parallels with the 'underclass' in the predominantly black ghettos of US cities. Whatever the accuracy of this assessment, the symptoms of social decay are plainly visible in rising crime rates, family breakdowns, drug abuse, and violence. As in the past, urban decay and insecurity have fostered political extremism. Openly racist political parties are in the ascendant in France, Germany, Italy, Belgium, and The Netherlands. The tried and tested technique of attributing economic problems to (non-white) foreigners and calling for their expulsion has been supplemented by another simple message: namely, withdraw from trade with the low-wage economies of the South.

In the USA, too, globalisation has been identified as the main culprit in explaining the trend towards wage inequality. Right-wing populism, tinged with a heavy stripe of xenophobia, has again been at the fore, notably in the shape of Congressman Pat Buchanan. But it would be wrong to attribute political concern solely to one part of the political spectrum. Both in Europe and the US there is a growing recognition that the post-1945 settlement is being torn up in the name of globalisation. In the past, the underlying rationale for integration into the global economy was that all countries would benefit, even though there would be winners and losers. Because the economic gains would outweigh the losses, the losers would be compensated and protected through a welfare state. Ultimately, public confidence in the model survived periodic setbacks for one simple reason: it appeared to work. The job creation and rising real incomes that went with robust economic growth created a sense of shared prosperity, backed up by institutions such as trade unions which reflected a sense of shared responsibility.

This is no longer the case. In most OECD countries, trade unions have been systematically weakened. The unionised proportion of the US labour force has declined by half - to 12 per cent of the total - since the 1970s (Bhagwati 1995). Collective bargaining rights were dramatically curtailed during the 1980s, notably in the US and Britain (OECD 1994). This helped to create the political conditions for more flexible labour markets, the consequences of which are reflected in the dramatic increase in wage inequality and the incidence of poverty among those in work. In the US, the minimum wage fell by one third during the 1980s (Freeman R 1995). In the UK, minimum-wage protection was eradicated, on the grounds that it hindered job creation. At the same time, part-time, temporary, and casual employment, as distinct from permanent contractual employment, has increased in all OECD countries, and the protection of workers against dismissal has been reduced. Allied to these labour-market changes, welfare reforms have weakened protection for the unemployed by shortening the duration of benefits (as in the USA) and reducing benefit levels (as in most of the EU). Recent welfare reforms in the USA have applied the whip of compulsion to force vulnerable groups back into the labour market, with an apparent disregard for the consequences in terms of child poverty.

The combined effect of these policy changes has been to erode the capacity of States to protect the losers from globalisation. Indeed, the creation of flexible labour markets and the contraction of social-welfare safety-nets are typically presented by governments as the inevitable corollary of globalisation, with higher wages and taxes seen as a potential deterrent to foreign investment and job creation. Britain's refusal to sign the EU's Social Chapter is an extreme manifestation of this tendency, with the government claiming, with some credibility, that the lure of cheap labour has attracted investment jobs from other EU member states. That view is certainly shared in the member states in question (notably in France), where Britain's deregulated labour market is perceived as a challenge to minimum wage provision. More broadly, almost all OECD governments now see the dismantling of welfare states and labour- market reforms as integral elements of policies geared towards greater integration into global markets. Such approaches derive from a blinkered world view which fails to address what is arguably the most important question of all: namely, why, in contrast to earlier periods, have trade and technological innovation created a vicious circle of increasing inequality and unemployment, rather than a virtuous one of economic growth, job creation, and shared prosperity?

Part 2: North-South linkages: the evidence

If the great problem facing OECD governments is how to counter the trend towards growing economic disparities by raising the income of the low-paid and drawing the long-term unemployed back into the jobs market, identifying the underlying causes of the problem is the first step towards its resolution. In recent years, increased trade with developing countries has emerged as one of the prime suspects, both in economic literature and in political discourse. With the onset of recession in most OECD countries in 1990, the associated loss of manufacturing jobs was widely perceived as a consequence of increased competition with developing countries. The rapid increase in direct foreign investment has further heightened concern about the transfer of jobs from North to South. Another view (Krugman P 1994) sees technological change, rather than international trade, as the main driving force towards wage inequality and unemployment. Debate between these two schools of thought has focused on the relative weight of trade and technology; and on the extent to which technological innovation is driven by competition from imports, as opposed to domestic market factors (Wood A 1995).

Concern over the consequences of trade between countries with highly unequal average incomes is nothing new. Since Ricardo, the idea that freer trade enhances the wealth of nations has been one of the most resilient ideas in economics. But Ricardo himself cautioned against rapid liberalisation in sectors of the economy marked by a high degree of poverty. Today, the concern is that the expansion of international trade has linked the labour markets of North and South, especially in the manufacturing sector. According to the ILO, around one half of all employment in manufacturing is now located in developing countries. At the same time, the pattern of North-South trade has undergone major transformation. In 1992, almost 60 per cent of developing-country exports to the industrialised world consisted of manufactured goods, compared with 5 per cent in the mid-1950s (World Bank 1995). The striking coincidence of declining manufacturing employment, the stagnation of real wages, and widening pay inequality in the US, allied to European unemployment levels not witnessed since the 1930s, has revived fears that cheap Third World labour is driving down wages and destroying jobs (The Economist 1994a).

These fears are especially pronounced in the US, where the increase in import penetration has been most marked. America's imports from developing countries accounted for 35 per cent of the total in 1990, compared with 14 per cent in 1970 (UNCTAD 1996). Imports of manufactured goods from developing countries now represent 11 per cent of the value of manufacturing output - double the level a decade ago. This has generated concern that imports from developing countries have caused a loss of well-paid manufacturing jobs, which have been replaced by poorly paid service-sector jobs. Developing-country exports to Europe and Japan have also risen, albeit more slowly.

The skills gap

It is not only the stagnation in average wages which is commonly attributed to the surge in imports from developing countries. The widening wage gap between skilled and unskilled workers is also widely considered to be a natural concomitant of increased North-South trade flows (Machin S 1994). The Stolper-Samuelson theory of factor price equalisation is widely cited as evidence. Briefly summarised, this holds that, in the absence of market-distorting barriers, trade flows will reduce the price of factors of production used intensively in imports; and that it will raise the income of factors of production used intensively in exports. Interpreting the Stolper-Samuelson theory in the light of North-South trade flows, if a developed country is exporting skill-intensive goods and importing labour-intensive goods, the price of skilled labour will rise, while that for unskilled labour will fall in relative (and perhaps even absolute) terms. Similarly, production and employment levels will rise in skill-intensive sectors and decline in low-skill sectors. Thus trade makes less skilled labour in developed countries and skilled labour in developing countries less scarce, thereby depressing its price in the labour market. By contrast, it makes the production of goods made by skilled labour in the developed world and by unskilled labour in the developing world more scarce. In equilibrium conditions, the long-term outcome is that factor prices will be equalised throughout the world, with less skilled workers in developed countries being paid the same as their competitors in the developing world.

Of course, this is a highly theoretical state of affairs. In practice, higher wages in the developed world reflect higher levels of productivity. But as trade barriers come down, transport and communications improve, and - crucially - the technology that supports higher productivity becomes more mobile, productivity differences will diminish. That is why the recent surge in direct foreign investment to developing countries has generated such concern. To the extent that this investment is linked to the transfer of technology and the deepening of industrial development in the South, and to a withdrawal of capital stock from the North, it might be expected to depress the productivity of labour there and to raise it in the developing world. This would in turn compound the impact of imports on wages and employment levels. Well-publicised cases of firms relocating from the US to Mexico, the dÆlocalisation of French industry, and the transfer of increasingly high-technology operations in to the first and second generation of NICs have reinforced the picture of wholesale industrial restructuring and jobs transfers.

Does the factor price equalisation theory explain the crisis of low wages, rising inequality, and mass unemployment facing Northern economies? At best, it does so only partially. As a causal factor behind unemployment and declining average wages, the impact of imports is probably vastly exaggerated. Imports from developing countries account for only around 2 per cent of GDP in the industrialised countries. Taking manufacturing production rather than GDP as the denominator gives a different picture, especially in the USA. But overall employment and wage effects cannot be captured in only one sector of the economy. Even so, in the OECD countries as a group, imports from developing countries accounted for only 3 per cent of the manufacturing market in 1992.

One of the most obvious ways in which Third World imports would reduce the demand for labour is through the trade balance. Here again, the evidence in support of sweeping assertions about linkages between North-South trade and unemployment is weak. The developed countries have maintained a positive balance in manufacturing trade with developing countries, equivalent to more than one per cent of GDP. As an explanation for Northern unemployment, the collapse in imports into developing countries associated with the debt crisis of the 1980s carries considerably more weight than an increase in the volume of exports from the South. Moreover, the OECD country that suffered the second-largest decline in its manufacturing trade balance with developing countries - namely Canada - achieved the largest increase in manufacturing employment. Such facts caution against over-simplified correlations between import levels and unemployment (UNCTAD 1996).

Distributional considerations

If the aggregate impact of imports from developing countries is more limited than is sometimes assumed, what of their distributional impact? This has probably been more significant, since Third World exports are concentrated on the labour-intensive sectors where unskilled workers are employed. Correspondingly, more labour-intensive jobs may have been lost in the North than were gained in more skill-intensive sectors. Such direct effects could also have been reinforced by the indirect effect of pressure from imports leading to rationalisation (Wood 1991). According to one account (Wood 1995), trade with developing countries reduced demand for unskilled workers by over 15 per cent in the 1980s, with import- penetration levels closely correlated with inequality.

Other accounts support this broad conclusion, while pointing to less significant overall effects. For instance, one detailed analysis of US trade flows with developing countries suggested that imports from developing countries may have caused a 6 per cent decline in demand for unskilled labour in manufacturing. Overall wage effects are more difficult to capture. However, these are probably more modest than is often assumed, since the manufacturing sector is not the main source of employment in most countries. In 1993, the manufacturing sector employed only 15 per cent of US workers. The vast majority of unskilled workers are located in the retail and service sectors. Capturing the transmission effect of lower wages from the traded to the non-traded sector is inherently difficult, leaving scope for widely divergent conclusions.

Some support for claim that imports from developing countries are having adverse distributional effect comes from evidence that prices for less skill-intensive goods have fallen relative to prices for higher-skilled goods (Sachs and Shatz 1994; Sachs and Warner 1995). However, it is difficult to see how wage pressure resulting from imports into the manufacturing sector can have had such pronounced economy-wide effects. This was the view reached by a major OECD study into growth and employment, which concluded that significant, though still quantitatively small, employment effects resulting from North-South trade were limited to "a small number of specific industries", such as textiles, clothing, footwear, computers, and electronics. Counter-intuitively, the OECD also concluded that more significant employment effects were to be found in industries employing a high proportion of skilled workers (OECD 1994), underlining the erosion of borders between below markets caused by the transfer of technologies. Recent evidence from UNCTAD points in a similar direction. This suggests that demand for skilled workers in the industrialised world has remained weak, and that unemployment has also increased among this group. There are also alternative explanations for income inequality trends. In the case of the US, according to UNCTAD, wage inequality trends have been a consequence not of surging average wages at the top end of the labour market, but of stagnation at the bottom end (UNCTAD 1996).

What such evidence suggests is that it is no longer accurate to depict import-competition from developing countries as being limited to the low-skill manufacturing goods in which Northern employment losses have been most substantial. The developing countries that have been most successful in sustaining exports to the developed world have experienced rapidly rising real wages, and have upgraded the technological and skill content of their exports. Some of these countries have been losing market shares in labour-intensive goods to other developing-country competitors - but most have not suffered rising unemployment or a slump in wages as a consequence. In the case of computers, which is one of the fastest-growing export sectors in these countries, employment losses in the industrialised world have been marginal. Meanwhile, not only has employment increased in some of the sectors (such as rubber and plastics) facing intense competition from Third World exporters, but exports of some labour-intensive goods from the developed to the developing world have also increased (UNCTAD 1996).

The technology thesis

If trade is at best a partial explanation for mass unemployment and rising inequality, what are the other factors at work? The main rival explanation is the technology 'shock' thesis. Briefly summarised, this holds that it is technological change in the form of the micro-chip, automation and robotics, rather than trade, which is driving down the wages of the low-skilled (Lawrence and Slaughter 1993). In addition, it is claimed that technological change has increased the productivity of skilled workers, inducing firms to bid up their price. To quote the most prominent exponent of the technology thesis: "the crisis of the West is largely a result of technological change that has, at least for the time being, evolved in a direction that is hostile to egalitarian ideals" (Krugman 1995).

In many respects, the gulf between the technology and trade explanations for income inequality trends is less wide than it appears at first sight. For instance, Wood claims that only one quarter of the reduction in demand for unskilled labour which he traces to competition from Third World imports was caused directly by imports: the rest results from labour-saving innovation spurred by trade. Ultimately, therefore, the difference between the two schools of thought hinges on what drives technological innovation, rather than its effects. However, advocates of the technological explanation point to one crucial empirical weakness in the arguments of those who claim that it is trade-induced shifts in the structure of production which are restructuring labour markets. What has changed, so the argument runs, is not the mix of industrial goods produced, but the movement towards skilled employment within each industry - including non-traded industries which account for the bulk of US employment (see, for example, Katz 1993; Lawrence and Slaughter 1994; Brooking Papers on Economic Activity 1993).

Powerful as this observation is, the evidence upon which the 'technology' shock thesis rests is equally open to doubt. For instance, increases in intra-group wage inequality might reflect increased returns to skill. But it might equally reflect institutional factors, such as the weakening of collective bargaining structures and the decline in trade union membership (ILO 1996b). There is another, more serious problem: namely, a lack of evidence to support the claim that the rate of technological change since the early 1980s has been any greater than it was in the 1970s (Mishel and Bernstein 1994). In the US, where the rise in wage inequality was most dramatic, the growth in investment per worker, and in research and development, appears to have slowed in the 1980s. It has also been pointed out that productivity growth in skill-intensive sectors did not begin to accelerate relative to other sectors until towards the end of the 1980s, while wage inequality increased most dramatically in the earlier part of the decade (ILO 1996) These caveats do not apply solely to the US. Since 1973, output growth in the OECD countries has halved, despite a small increase in the numbers in work (OECD 1994). This implies that labour-productivity growth has declined technology destroying jobs on an unprecedented scale (Glyn 1996). A fact which rests uneasily with the claim that the debate outlined above is informed by an analysis of the past. But what of the future? Some commentators who maintain that the past effects of trade have been negligible argue that, as vast countries such as India and China expand their exports of labour-intensive goods, the future position of unskilled labour will deteriorate dramatically as a consequence of trade flows (Sachs and Schatz 1995). Others, who argue that the past effects have been considerable, have a more optimistic assessment of the future. They point out that many of the goods in which developing countries have the greatest advantage are no longer produced on any scale in the developed world, so that the most painful adjustments may already have taken place. They also point out that poor consumers will benefit from cheaper clothes, textiles, and consumer durables. As direct foreign investment increases and the more advanced developing countries climb the skills and productivity ladder, it could be that the most intense competition will be felt in more skill-intensive industries, and in service sectors. This process is already under way. Computer giants such as Texas Instruments and IBM now maintain software development facilities in Bangalore; international airlines, banks, and insurance companies now employ armies of data processors in countries such as the Dominican Republic, India, and Thailand, all linked by computer; leading-edge German chemical firms such as Bayer have transferred industrial plastics plants to China; and AT&T has recently transferred its telephone-assembly plants from Singapore to Thailand in response to rising wage costs.

Monetarist orthodoxy as a barrier to employment

Competition from developing countries has contributed to the labour-market problems outlined earlier, although no consensus is likely to emerge on the quantitative dimensions. Protectionist responses might have addressed one aspect of this problem, by reducing the supply of imported goods. However, advocates of protectionist solutions have failed to devise a coherent strategy for achieving full employment by raising productivity, and by reversing the decline in investment activity which is at the heart of the crisis of low pay and mass unemployment. This points to a deeper problem. Viewed through the lens of history, the market has shown itself to be an unrivalled creator of wealth, and an equally unrivalled source of social dislocation and inequality. State interventions, by contrast, have often been guided by a concern to achieve more equitable social outcomes, but often at the cost of gross economic inefficiency, and disguised unemployment. The challenge for Northern governments - as it is for their Southern counterparts - is to harness the market to the cause of social justice through regulatory measures designed to generate and distribute wealth more equitably. We return to this issue below.

The case for growth

For the present, it is worth stressing an obvious but often neglected point: namely, increased economic growth is vital to the attainment of full employment, which is in turn one of the keys to poverty eradication. Despite the much-analysed phenomenon of 'jobless growth', the expansion of output is still the most important route to job creation in OECD countries. Over the period 1982-1993, every one per cent increase in OECD output increased employment by a similar level (Bolitho and Glyn 1995). Conversely, recessions bring major job losses, as witnessed by the recession of the early 1990s. In the two decades up to 1970, growth rates in the OECD countries averaged over 3 per cent. Since then, they have been a full percentage point lower, with catastrophic consequences for employment.

It is beyond the scope of this paper to consider the various policy issues raised by the decline in OECD growth rates (for a full discussion, see OECD 1994; Glyn 1996; UNCTAD 1996). However, three factors are of salient importance. First, restrictive monetary policies and financial deregulation have pushed up interest rates to historically high levels, and added to their volatility. In the UK real interest rates for the period 1981-1992 were double the average for the previous 25 years. For the USA, Canada, and Italy they were three times higher (UNCTAD). Second, and related to this trend, the deterioration in jobs and wages has been accompanied by a sharp decline in investment. In Europe, capital stock has grown at 3 per cent a year since the early 1980s - half the level in the 1960s (Rowthorn 1995). The third factor is a decline in government investment, which has risen at levels much less than GDP since the early 1970s (UNCTAD 1996). This is important because, while job creation may the primary responsibility of the private sector, private investment depends upon the availability of public goods such as transport, marketing, and communications infrastructure. On one estimate, had the stock of public capital in the USA grown at the same pace in the twenty years up to 1987 as it did in the twenty years after 1947, annual GDP would have been almost 2 per cent higher.

Reduced investment

To a greater or lesser extent, all industrialised countries have suffered the effects of reduced investment on employment growth. As a causal factor in explaining structural unemployment and low wages, the cumulative effect of lower rates of capital accumulation dwarf the effects of trade with developing countries. In turn, the policies behind this decline can be traced to the abandonment in the early 1970s of policies geared towards full employment and output expansion. During the 1980s, controlling inflation, fiscal retrenchment, and financial deregulation emerged as the new guiding principles. Reduced inflation, lower public-sector borrowing, tax reductions for potential investors in higher-income groups, and a concerted effort to undermine trade unions as a means of reducing labour costs became the hallmarks of a supply-side economic policies. Those policies failed because of their inherent contradictions. Thus profit margins increased, but the slowdown in demand and high interest rates deterred investment, as did the collapse of public investment. As UNCTAD puts it:

The generally restrictive monetary policies implemented in the past two decades have shunted economies into low growth paths in which low demand growth and low potential output growth have fed back into one another. Frequent under-utilisation of the existing productive capacity has made for slow capital formation. The slow growth of potential output has, in turn, provided the rationale for persisting in macro-economic policies that limit demand growth....These policies have thus created a weak economic growth dynamic, and generated an increasing imbalance between the labour force and the tools - i.e. capital - to employ labour productively (UNCTAD 1996c).

Unfortunately, governments continue to place their faith in fiscal discipline, regardless of the costs in terms of lost output, employment, and wages. In Europe, public spending is under intense pressure as governments attempt to meet the criteria for monetary and economic union set by the Maastricht criteria, which include a provision that budget deficits should be equivalent to no more than 3 per cent of GDP. In Germany, the equivalent of 0.8 per cent of GDP and in Britain 0.7 per cent of GDP will have to be trimmed from public spending over the next year to meet that target. Across the EU, social-welfare budgets have borne the brunt of the spending cuts, exposing marginal populations to the threat of increased vulnerability and poverty; and further eroding the post-war contract between States and citizens. The consequences in terms of social dislocation and polarisation appear to have been disregarded in the reckless pursuit of fiscal targets. In a situation where unemployment affects 11 per cent of the population, and where inflation levels average less than 4 per cent, a more balanced approach geared towards output expansion is desperately needed.

The limits to flexible labour

Other tenets of the monetarist faith also survive intact and are winning new converts. For instance, EU governments increasingly look to the USA and Britain as labour-market models to be emulated. Precisely why remains unclear. During the 1980s, Britain deregulated its labour market more than any other EU country (on this period see Barrel R ed 1994), yet Britain's output growth record since 1979 is the worst in the G7; its industrial base has shrunk more rapidly, with the manufacturing sector in persistent current-account deficit; and the bulk of the new jobs which have been created have been characterised by low pay and low productivity. In the services sector, Britain lags even further behind (highly regulated) France and Germany than it does in manufacturing. One recent survey comparative has also pointed out that France's supposedly over-manned transport sector is 40 per cent more efficient than its British counterpart - a crushing testament to the consequences of inadequate public investment (National Institute of Economic and Social Research 1996). All of this suggests that labour-market deregulation has achieved minimal economic gains, at an enormous social cost in terms of increased poverty and low wages.

If anything, official data probably exaggerate the benefits associated with labour-market flexibility. Increasingly casual terms of employment and low wages have proved a highly effective way of transforming open unemployment into disguised unemployment. One of the ways in which this happens is through the transfer of labour from the dole into low-paying, low-productivity service-sector jobs for which an increase in employment is not linked to increased demand (UNCTAD 1996). Not surprisingly, disguised unemployment has been increasing most rapidly in the USA and the UK, where wage inequalities have widened, and least in France and Germany, where there has been no trend towards increased inequality.

In the case of the USA disguised unemployment is not fully captured by employment statistics. The most extreme example of those excluded from the labour market are people incarcerated or on parole. In the USA 6.6 per cent of men fall into this category. Levels of imprisonment in Europe are one tenth of the US level, and the prison population is growing at 9 per cent a year. By the year 2000, the USA will have a larger share of its workforce imprisoned than the share of long-term unemployed in European economies (Freeman 1995).

For the most part, employers have used labour-market flexibility as a means of reducing labour costs and downsizing, rather than to expand productive capacity (Gregg P et al 1994). They have been assisted in this by the erosion of minimum-wage provisions, or their wholesale elimination. In the USA, minimum wage levels fell from 45 per cent of the average wage to 35 per cent in the 1980s. In Britain, legislation was introduced to withdraw minimum-wage protection on the grounds that it undermined employment. But evidence from France (Brazen and Martin 1993), the USA (Card and Krueger 1995) and the UK (Gregg 1994) suggests that the erosion of minimum wages had either a small effect or no effect on employment, and that in some cases employment actually rose in institutions forced to pay the minimum (Katz and Krueger 1994). In short, lower wages are not an effective route to higher employment (Oxfam 1996a). By increasing staff turnover rates and discouraging commitment and training, low wages reduce productivity - a fact which the OECD has now acknowledged (OECD 1995). Low wages are also expensive, since they have to be supplemented by welfare benefits. In Britain, the cost of these benefits to the Family Credit scheme has soared to over $2bn, raising the question of whether this money could be more productively spent in skills training.

Policy options

What, then, are the broad policies needed to address the underlying causes of high unemployment, low wages, and rising inequality? Six broad themes suggest themselves:

Macro-economic reform: Governments in the OECD are gripped by a collective fiscal paralysis, in which objectives such as full employment and growth are subordinated to the pursuit of low inflation. Without policies designed to bring about an increase in demand, higher public and private investment, and low and stable interest rates, little is likely to be achieved.

Training: Skills training has a vital role to play in enabling workers to adjust to the competitive pressures generated by trade, and in preventing long-term unemployment. In Sweden, successive governments have sought to prevent long-term unemployment by providing extensive skills training during the first twelve months out of work. That is a model which could be more extensively applied. While the short-term costs are relatively high, these have to be discounted against the long-term cost of supporting those in long-term unemployment.

Employment support: Policies are needed to increase the labour-intensity of growth. Real wage reductions are not an effective means to this end, for reasons explained above. Subsidies to employers to encourage them to hire more unskilled workers are part of the answer, especially in Europe. Work experience increases the chances of employment and, properly managed, can raise skill levels. Once again, the costs of such a scheme would need to be set against existing welfare-benefit payments to its participants. Another policy option is to shift the burden of taxation away from employment (i.e. through wages and employers' contributions), especially at the bottom end of the scale, and towards general taxation. Reform of this kind was proposed by the EU in 1993, when it argued for a reduction in social-security contributions financed by a carbon tax.

Social insurance: In the case of the USA and the UK, where wage levels are highly flexible and unequal, there is a need to support the incomes of the low-paid through minimum-wage protection, tax cuts, and public investment in skills training. Social insurance has a vital role to play, but it must be structured so that it helps the working poor, as well as the unemployed. Sharp reductions in benefits when minimum income threshold criteria - characteristics of social-insurance provision in both the USA and the UK - act as a disincentive to employment. Another important step is the provision of basic services, such as schools and health care, as universal entitlements, so that poor families are protected but are not faced with the loss of benefits if their economic situation marginally improves.

Protection of the welfare state: In both Europe and North America, welfare states are under unprecedented attack, with financial cuts falling in areas geared towards highly vulnerable groups. Single women, the long-term unemployed, the homeless, and immigrants have been particularly visible targets. During periods of social and economic crisis, such as that faced by the industrial countries in the mid-1990s, welfare services play a crucial role in maintaining access to basic services; and in limiting inequality. It follows that the protection of these services should be accorded a far higher political priority. The claim that States cannot afford these systems is curiously at variance with the fact that the German government is undertaking the biggest transfer of resources ever to an economically disadvantaged region, namely the former GDR.

Increased taxation: Achieving many of targets outlined above will require a reversal of the tax policies introduced since the early 1980s, with a particular emphasis on clawing back the transfers to higher-income groups for investment in public services. Society as a whole will benefit both from the services and jobs which can be created through increased public investment, in both the private and public sectors, and from avoidance of the social divisions associated with low pay and mass unemployment. While any increase in taxation carries the risk of reduced incentives, the adoption of more regressive taxation systems has conspicuously failed to create the investment response which was envisaged - and inadequate public-sector provision also implies high social and economic costs.

There is no single blueprint for moving towards full employment. But there is a wide range of policy options available to reduce mass unemployment, address the problems of low wages, and improve job security. What can be said with some certainty is that relying on unregulated market forces alone is more likely to compound than resolve the problem. The question is whether society has the sense of community and governments the political will to rise to the challenge.
  ^{NEXT SECTION}

   ^PREVIOUS

              UNDP Home

BACK TO TOP