|Flying-geese and waddling-ducks: the different capabilities of East Asia and Latin America to
‘demand- adapt’ and ‘supply-upgrade’ their
export productive capacity
Faculty of Economics,
University of Cambridge,
Cambridge CB3 9DD,
“Wild-geese are said to come to Japan in autumn from Siberia and again back to north before spring, flying in inverse V shapes, […]”. "The less-advanced ‘wild geese’ are chasing those ahead of them, some gradually and others rapidly, following the course of industrial development in a wild-geese-flying pattern”.
“The world of foreign trade is one of change. It makes a great difference to the trade of different countries, and to the impact of trade on them, whether they are capable of changing with the world. [...] Capacity to transform is capacity to react to change, originating at home or abroad, by adapting the structure of foreign trade to the new situation in an economic fashion".
“[...] the obsession with competitiveness is not only wrong but dangerous, skewing domestic policies and threatening the international economic system”.
During the 1992 United States presidential campaign President Bush's main economic adviser was asked about the apparent decline in the technological content of his country's exports; he replied that he saw no difference in exporting potato-chips rather than micro-chips. This paper will investigate two related issues of this controversy for developing countries (LDCs). First, whether there are major economic differences -- in particular for the productivity growth potentials of the export-sector itself (and of the overall economy), for the welfare gains from trade (e.g., issues related to terms of trade), and for institution building -- between exporting one or the other of these two (types of) products. Second, whether regional dynamics -- and, in particular, the role of the regional power -- are an important component of the likelihood of LDCs exporting one or the other type of product.
Regarding the first issue, this paper will conclude that the economic differences in relation to which type of products an LDC exports are substantial from the points of view of both supply and demand dynamics and institution building. From the supply point of view, the fundamental difference is found not only in the different productivity growth potentials of each type of export-product, but also -- and at least as important, even in open economies with large export sectors -- in the different capacity of each type of product to induce medium- and long-term productivity growth in the economy as a whole (e.g., via cumulative causation-type processes). That is, some exports seem to have much greater capacity to generate -- and sustain -- overall ‘trade-induced’ GDP-growth than others. In turn, from the demand point of view, the main difference arises from the fact that international demand for some types of products (such as ‘high’- and some ‘medium-tech’ products and some resource-based products at an advanced stage in their value-added chain) has grown much faster -- and is likely that it will continue to grow much faster -- than for others (such as most ‘low-tech’ and resource-based products at an early stage of the value-added chain). Almost inevitably, important (and growing) asymmetries in international demand are bound to have a significant influence in the terms of trade and, therefore, in the welfare gains of international trade (via its effect on the rate of growth of ‘command-GDP’ -- or the purchasing power of the country’s GDP).
Regarding the second issue -- the likely role of regional dynamics and, in particular, that of the regional power in trade specialisation -- this paper will conclude that there is a significant amount of evidence pointing to the fact that regional dynamics can play a crucial role in the export profiles of LDCs, and in the welfare benefits that they are able to obtain from their international trade.
The general issue of ‘potato-chips versus micro-chips’ will be analysed through the study of the diverse export and growth performances of Latin America and East Asia since the 1960s. I will argue that the experience of these two regions supports at least two hypotheses. First, that GDP-growth, and the role played in it by exports, seems to be both a ‘product-specific’ and an institutional phenomenon; i.e., as far as exports are concerned, their capacity to generate and sustain (trade-induced) GDP-growth seems to be closely related both to the composition of exports and to the effectiveness with which they are ‘anchored’ in the domestic economy, and to the capacity of the state to implement appropriate trade and industrial policies. In particular, evidence will suggest that East Asia’s much superior growth performance is closely associated with their continuous effort both by the state and the corporate sector to supply-upgrade (along the so-called ‘learning curve’), and demand-adapt (to ever-changing international demand) its export productive capacity. Second, that regional dynamics, especially the specific type of leadership that Japan has exerted in East Asia, have played a significant positive role in this.
2.- Economic growth: regional diversities
Figure 1 shows that, with the major exception of China and India, in the era of globalisation there has been a significant slow-down in economic growth in both Latin America and East Asia -- in fact, a slow-down in most countries in the world --, and that there has been an increase in the instability of the modest growth that has actually taken place.
GDP and GDP per capita growth and their respective coefficients of variation, 1960-1980 and 1980-2000
● Growth rates at constant 1995 US$. G6 = Canada; France; Germany; Italy; Japan; and United Kingdom. US = United States. D = aggregate output of the 59 developing countries included in the WDI data-base. D* = D, but excluding China and India. N-1 = first-tier NICs (Newly Industrialising countries of East Asia: Hong Kong; Republic of Korea; Singapore; and Taiwan). N-2 = second-tier NICs (Indonesia; Malaysia; Philippines; and Thailand). LA = Latin American and Caribbean countries (Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Jamaica, Mexico, Peru, Uruguay and Venezuela). S-S A* = 28 Sub-Saharan countries (excluding South Africa).
● Source: World Bank, World Development Indicators, WDI (2003); as well as for the G7, this data-set gives information for only 59 LDCs for the whole period 1960-2000. Unless otherwise stated, data on growth throughout this paper will use this source. All coefficients of variations are shown multiplied by ten. When data refer to a region or group of countries, they refer to the aggregate output of the respective region or group.
According to the WDI data-set, and with the major exceptions of China and India (and Bangladesh), there is both a clear overall downward trend in the rate of economic growth of developing countries after 1980, and a relative increase in the instability of this growth. Panel A of Figure 1 illustrates several related issues. First, the average rate of growth of world output declined from 4.6% between 1960 and 1980 to 2.8% between 1980 and 2000. So, if in the first 20-year period world output grew 2.5-fold, in the second it only managed a 1.7-fold increase. Second, one of the groups of countries where the slowdown of growth was more marked was the G6: their rate of growth fell by more than half -- from 5.1% to 2.4%. Third, even in the USA -- and despite the prosperity of the Clinton years -- the average rate of growth in the second 20-year period was below that of the first one: 3.5% and 3.2%, respectively. Fourth, and despite China’s and India’s size and remarkable growth rates, the overall rate of growth of output in the 59 LDCs included in this sample declined from 5.5% to 4.5%. Fifth, this decline, of course, is much more pronounced if one excludes China and India from the LDC-sample -- in this case the decline is from 5.8% to 3.7%. Sixth, of the 5 groups of developing countries identified in the graph, Latin America had the most pronounced drop in its rate of growth: from 5.6% to 2.2%; so, if in the first 20-year period Latin America trebled its overall output, in the second it only managed half that expansion. Seventh, in only 12 of the 59 LDCs of the sample there was an increase in the rate of growth in the second period -- just one country in Latin America (Chile); three in Asia (China, India and Bangladesh); and 8 Sub-Saharan African countries (Benin; Burkina Faso; Chad; Ghana; Guinea Bissau; Mauritius; Senegal; and Sudan). Finally, although only one in every 5 countries was able to accelerate its rate of growth in the second period, among these countries there were 3 of the 5 largest LDCs; however, the other 9 countries which accelerated their rate of GDP growth in the second period are relatively small, representing less than 7% of the remaining LDCs’ population.
Furthermore, Panel B of Figure 1 illustrates that not only the overall growth-trend is clearly downward, but also -- and again with the major exceptions of China and India -- there was an increased growth-instability in the G7, in the first- and second-tier NICs, in Sub-Saharan Africa and (especially) in Latin America.
Panel C and D indicate that the same phenomena, in particular the asymmetries between China, India and Bangladesh and almost every other country in the world, are found (and even to a greater extent) in the growth rates of GDP per capita in these two periods. In fact, if using other sources one adds to the WDI data-set data for the missing 15 -- mostly troubled -- Sub-Saharan countries for the period 1980-2000, the LDCs’ median GDP per capita growth for the second 20-year period falls to practically to zero.
Panel C shows that, first, the average rate of growth of the overall world GDP per capita fell by half in these two periods (from 2.6% to 1.3%). As a result, while in the first 20-year period this indicator expanded by 67%, in the second it managed an increase of less than 30%. Second, in the G6 the slowdown was remarkable, with per-capita-GDP-growth falling from 4.3% to 2%, respectively. Third, again even in the US there was a decline in the average rate of growth of GDP per capita in both periods (2.3% and 2.1%). Fourth, and despite both the outstanding performance of China and India, and the rapid slowdown in LDCs’ population growth, the rate of growth of GDP per capita in the 59 LDCs included in this sample declined from 3.1% to 2.7%. Fifth, excluding China and India the corresponding statistic for the remaining 57 LDCs declines from 3.2% to 1.5%. Sixth, of all groups and regions, rapidly-liberalising, rapidly-globalising and rapidly-reforming Latin America had by far the largest decline in GDP per capita growth: from 3% between 1960 and 1980 to a mere 0.4% between 1980 and 2000. So, if for most of the first period Latin America was almost managing to keep up with the increase in GDP per capita of the G7, in the second one Latin America had a rate which was only one-fifth that of the G7 (which, as mentioned above, was not very impressive in the first place). Also, if with the first rate it would have taken less than 25 years for the region to double its income per capita, at the rate of the second period the region would have needed 175 years to do the same! Seventh, even some of the most dynamic East Asian economies posted a lower per capita GDP growth during the second period -- a phenomenon, though, largely explained by the 1997 East Asian financial crisis. Finally, at an average rate of growth of GDP per capita of 8.3%, 3.6% and 2.6% during the second period China, India and Bangladesh, with a combined population of no less than 2.4 billion, were among a small number of countries that improved significantly their performance in the second 20-year period.
Finally, Figures 2, 3 and 4 show in more detail the relative performance of some middle- and low-income LDCs vis-à-vis that of the G7.
● Source: regions and countries as in Figure 1. pc = per capita. G7 = G6 and the USA; N-1 = first-tier NICs (Republic of Korea; Singapore; and Taiwan); N-2 = second-tier NICs (Malaysia and Thailand); LA = Latin American and Caribbean countries); Indo+Phil = Indonesia and The Philippines; Sub S Africa = Sub-Saharan countries.
The superior economic performance of East Asia in both middle- and low-income countries becomes even more evident in Figure 2. In particular some Latin American countries, such as Argentina (even before its Wagnerian 2001-2002 crisis), seem to have moved in almost the exact opposite direction than the NICs-1. Furthermore, Figure 3 indicates that in Latin America Argentina was not alone in this.
However, Figure 4 illustrates that at least Brazil and Mexico during most of the first period, and Chile after the mid-1980s, did manage impressive episodes of ‘catching-up’ with the G7. However, as periods of slow growth and recession were equally as remarkable in their intensity, especially for the two largest economies in Latin America after the 1982 debt-crisis, none of these three countries reached the year 2000 with the same relative level they had in 1960.
● Mex/China = GDP per capita of Mexico as a multiple of that of China; Bra/India = that of Brazil vis-à-vis India; Arg/Korea = that of Argentina and Korea; and Chile/Mal = that of Chile and Malaysia.
All in all, no much long-term catching-up in this part of the world! In turn, of the 9 East Asian countries in the WDI data-set, only The Philippines -- Latin America’s honorary country in East Asia -- fails to catch up with the G7 during this period. The aim of the rest of the paper is to study some trade-related issues to this asymmetric regional growth, its regional dimensions, and especially the role played by their regional leader.
3.- Increased international economic integration in the world economy
One of the main characteristics of the post-Second World War international economy has been an increase in the share of export in most countries' GDP, particularly in OECD and East Asian countries. Figure 5 illustrates this phenomenon and the contrasting behaviour of Latin America.
● S Asia = South Asia; K+T = Korea and Taiwan; a = Singapore, Hong-Kong and Malaysia; and b = Oil Middle East.
● Source: World Bank Data Base and WDI (2003); current prices.
As Figure 5 shows, the trend towards an increased degree of international economic integration has been different across regions, with some East Asian countries having the fastest increase, and Latin America between 1950 and 1973 the largest decline in GDP export-shares; as a result, as Figure 6 shows, shares in international trade (measured now as market-shares in OECD imports2) have changed significantly during this period.
● Jap = Japan; N-1 = Korea, Singapore and Taiwan (from now on, Hong-Kong will be excluded from the NICs-1 due to its progressive transformation into a financial centre); N-2 = Malaysia and Thailand (in the analysis that follows the NICs-2 will only include these two countries); LA* = Latin America and the Caribbean (excluding Mexico and Brazil); Bra = Brazil; Mex = Mexico; S Asia = South Asia (includes India and Pakistan); Africa includes Sub-Saharan Africa, South Africa, and North Africa; and Oil ME = oil Middle East (includes Iran, Iraq and Saudi Arabia).
● Source: Trade CAN.3
Figure 6 shows that both industrialised and Third World countries had major changes in market shares in OECD imports during this period. Among the industrialised countries the changing fortunes of the USA and Japan stand out. Among the non-oil Third World countries the differences between Latin America and East Asia are probably the most remarkable ones -- with Latin America (excluding Brazil and Mexico) reducing its market-share in OECD imports by more than half in just 22 years -- from a market share of 6.22% in 1963 to one of just 2.75% in 1985; and again down to 2.14% in 2000. Figure 6 also illustrates the relative similarities in the decline in market shares of this sub-region and the USA.
4).- East Asia versus Latin America in terms of exports performances
The differences between the economic policies -- in particular trade and industrial policies -- followed by Latin America and East Asia since the 1960s are well documented4. The main benefit of these policies for East Asia was to allow these countries to increase massively both their shares of exports and of manufactures in GDP. By contrast, Latin America, which had been the region with the largest share of exports in GDP in the world throughout the first half of the century (and by 1963 still had a share of OECD imports larger than that of Japan in 2000), by the time of the first oil crisis had its exports reduced to a share of GDP almost half its 1950 level (see Figure 5). This decline followed both a weak demand for primary commodities in OECD markets since the Korean War, and a trade policy characterised by a strong anti-export bias. Both ECLAC and the political groups in power in most of Latin America during the 1950s and 1960s had an economic ideology characterised by a deep pessimism regarding the growth-enhancing potentials of unprocessed primary commodity exports, and a high degree of optimism regarding the growth potential of a massive drive towards a state-led process of import substituting industrialisation (ISI)5. As a result, and despite a strong growth performance in some countries, particularly in Brazil and Mexico, the ISI-led increase in the share of manufactures in GDP took place at the expense of the GDP export-share6. Figure 7 illustrates this phenomenon.
● [mf] share of manufactures in GDP; [x] of exports; and [p-c x] of exports of non-oil-primary-commodity. 1980 constant-prices. 3-year moving averages. Excludes Venezuela (and Mexico after oil-discoveries) -- oil is excluded to avoid the highly distorting effect of pricing oil throughout at 1980 prices.
● Source: Palma (2004a, using data from ECLAC Statistical Division.)
ECLAC had predicted that the initial negative impact of ISI on Latin America's primary commodity exports was soon going to be compensated for by a process of export diversification, allowing the region to change the structure of its exports away from primary commodities to manufacturing; this process was to take place first through rapid regional economic integration. However, after many years of implementation, this policy met only with partial success. Starting with the 1973 oil crisis, but mainly from 1982 onwards (as a result of both the debt crisis and the rapidly changing ideological environment in the region), Latin America embarked on one of the most radical policy shifts in Third World history. It moved away from its state-led ISI policy into a massive process of economic ‘liberalisation’ and de-regulation. However, Latin America's new orthodoxy has proved to be of a very different nature from that that emerged in the non-Anglo-Saxon industrialised countries and in the other regions of the Third World, particularly East Asia.
Latin America was badly affected by a series of negative external shocks at the end of the 1970s and the beginning of the 1980s, which found the region in a particularly vulnerable position due to its large current account deficits and its large stock of foreign debt. There are four main reasons for this. First, international interest rates began to rise rapidly in 1979 (following Paul Volcker’s tightening of monetary policy at the Federal Reserve, which led to a trebling of interest rates in the US and in the LIBOR-indexed Latin America foreign debt). Secondly, the terms of trade of the non-oil exporting countries of the region collapsed at the end of the 1970s (see Figure 13 below). Thirdly, with Mexico's default in 1982, voluntary lending to Latin America stopped abruptly in the second half of that year. Finally, recession and growing protectionism in most of the North during the 1980s complicated still further the economic environment within which the Latin American economies had to regain their internal and external macro-equilibrium following the 1982 debt crisis.7
As had happened in the 1930s, a massive and continuous external shock that found Latin America in an extremely vulnerable position not only brought about the need for a very painful internal and external macroeconomic adjustment, but also laid the foundations for a radical and widespread change in economic thinking. The resulting ideological transformation eventually led to a generalised change in the economic paradigm of the region. In this case, it was characterised by an extreme move towards trade and financial liberalisation, wholesale privatisation and market deregulation, along the lines begun in Chile in 1973. And a key element to understanding these reforms, particularly the ‘fundamentalist’ way in which they were implemented throughout the region, is that they were mostly carried out as a result of the substantial economic weaknesses of these economies. In other words, they were a desperate attempt to reverse capital flight, reduce run-away inflation and bring the economies out of recession. Its discourse ended up having as a compass a 'magnetic north' that was simply the reversal of as many aspects of the previous development strategy as possible.8 The mere idea that alternatives could exist met with a mixture of amusement and contempt.9 This fact helps to explain the rigidity with which the reforms were implemented in Latin America, as opposed to in East Asia.
As I have argued elsewhere (Palma 2004 a and b), it is not that East Asia did not implement its economic reforms partly out of necessity (and also because of mounting external political pressure from the United States government to do so); but its economic weaknesses were very different in nature and intensity from those in Latin America. As is well known, since the 1960s the East Asian countries integrated their economies into the increasingly complex world division of labour in a very different way than Latin America. Instead of accepting their traditional, ‘exogenously’ given comparative advantages (i.e., given by their existing and not by a potential -- more growth-enhancing -- resource endowment), they gained a different type of endogenously-created comparative advantage, mainly by being able to create an institutional environment that allowed them to follow a ‘flying geese’ pattern of production and upgrading.10 Following Japan’s example, this was achieved through an increasing export penetration of OECD markets for manufactured goods, within a process of the regionalisation of production clearly led by Japan (see especially Figure 19 below). The extraordinary success of the East Asian economies was based on several factors. On the external front, the openness of OECD markets, especially the USA, to their manufactured exports (with the exception of post-NAFTA Mexico, this openness was clearly not extended to Latin America, and especially not to Brazil); and the fast rate of expansion of international trade in these goods. On the internal one, to be able to build a structure of property rights, a political settlement and an institutional capability that would allow them to produce manufactures that could compete globally; to be able to continuously upgrade their exports through the above-mentioned 'flying geese’ path (which helped them, and particularly the first-tier NICs, to develop a remarkably positive interaction between increases in productivity and wage growth); their being able to generate the high levels of investment and saving required for this upgrading; and their achieving an effective co-ordination of this investment through different forms of industrial and trade policies, which simultaneously succeeded at insulating domestic markets and outwardly orienting tradable production.
However, problems for many East Asian economies emerged in the late 1980s and early 1990s. One of the most important was the result of their own success: some of their most important exports, particularly electronics, began to experience excess supply and rapidly falling prices. In part this was also the result of the increased standardisation (or 'commoditisation’) of inputs to the electronics and micro-electronics industry, in which many of these economies had concentrated their exports. As a response to this, their corporate sector massively expanded investment in new productive capacity in an attempt to turn falling prices to their advantage via increased market shares. In fact, the combined result was to exacerbate the global excess supply and to put further downward pressure on prices.11 An obvious casualty of this increased struggle for market shares was profitability; and declining profitability led to a changing composition of the finance for investment, away from internally generated profits and towards (domestic and foreign) debt. This was clearly reflected in rising debt/equity ratios, which particularly in Korea reached heights that even for this part of the world should have produced feelings of vertigo. This necessity to have access to an ever-growing amount of finance was clearly one of the key domestic pressures behind the drive towards financial liberalisation.
Another problem was that in the same period, China became a formidable competitor in many of the markets that were crucial to the second-tier East Asian NICs, a process that also affected their profitability and led to an increased need for external finance. At the same time, given the changing international division of labour, some of these economies, particularly Malaysia and Thailand, were reaching a point where further upgrading of exports to higher value added products was becoming increasingly difficult. In particular, it was becoming more and more complicated to break away from a 'sub-contracting’-type of industrialisation, and to advance further along the path towards the form of industrial development that characterised the first-tier NICs. So, in a more pragmatic way than in Latin America, they increasingly looked towards progressive forms of trade and financial liberalisation and economic de-regulation as a way of strengthening and accelerating their existing ambitious growth strategy.
In Latin America, however, the economic environment in which the reforms were being implemented was very different. It was one characterised by an attitude of ‘throwing-in the towel’ vis-à-vis their previous growth strategy of state-led ISI. As a result, a widespread ‘born-again’ type of neo-liberalism emerged, which sought totally to reverse almost every aspect of the growth strategy previously followed by the region.12 For example, it was not only taken for granted that trade and financial liberalisation would switch fairly automatically the engine of growth towards domestically financed private investment in tradable production, but it was also assumed that budgetary balance and unregulated market signals would be sufficient conditions for macroeconomic equilibrium and microeconomic efficiency. At the macro level it was believed that both a decline in the progressive nature of taxation and fiscal balances would inevitably release private saving for more productive uses in the private sector; and, at the micro level, market deregulation and trade liberalisation would not only stimulate tradable production but would also significantly increase private investment, while higher interest rates would boost domestic saving and reverse capital flight.
So far, this process of reform has turned out to be far more complex than predicted, and its results (at best) more mixed. The main achievement of the new policies was a temporary relaxation of the external financial constraint. This led to remarkable reductions in inflation and the end of the long post-1982-debt-crisis period of recession (although in many countries, especially in Brazil, growth turned out to be short-lived). There was also an important inflow of foreign direct investment, often directed (as in Brazil) towards the privatisation of utilities. Exports of many primary commodities increased significantly and there were sharp reductions in public deficits (again, except in Brazil).13
However, these reforms also had many negative effects. In particular, financial liberalisation not only greatly increased the likelihood of shocks, but (mainly because it was implemented with the wrong ‘sequencing’ -- i.e., simultaneously with drastic stabilisation and trade liberalisation programmes) also altered the fundamentals in a way that threw the export-led growth strategy off course. In particular, those countries that shifted their domestic imbalances into the external sector by using the nominal exchange rate as a price 'anchor’ were the ones where the positive results of the process of policy reform proved to be less sustainable. Their exchange rate policy produced a substantial appreciation of their currencies, which distorted the whole of the export-led growth strategy -- switching the engine of growth away from the expected domestically financed private investment in tradable production, and towards private consumption, asset-price bubbles, and externally financed private investment in non-tradable production and services. This switch, coupled with over-optimistic expectations of future performance (set in motion mainly by easy access to credit and by the massive ‘spin’ put on the economic reforms both at home and in the Washington institutions), had many serious negative effects, particularly on the balance of payments, national savings, non-residential investment and employment -- and, particularly in Brazil, also on the public sector accounts. As a result, the Latin American economies in general have ended up being more vulnerable to domestic and external shocks than at any time since the 1920s.14
4.b).- The changing degree of export-competitiveness of Latin America and East Asia in international markets
The main limitation to the use of traditional trade statistics (as those available in the United Nations Com-trade on-line database and the CAN software) to study issues of export-competitiveness and trade-induced growth is that they refer to gross value of exports and not to value added or to net-exports. This problem is particularly important in the study of the so-called ‘maquila’ industry (or purely assembly-type export-operations) of Mexico and some parts of Central America; in fact, the use of these data would totally distort the comparative-analysis attempted below between Latin America and East Asia. As a result, there is little option but to exclude Mexico in most of the following comparative analysis of supply-upgrading and international demand-adaptation between these two regions (except for the brief analysis that follows in section 4.b.1), and to analyse properly Mexico’s peculiar export-led growth experience in a separate place; this study is done elsewhere (Palma 2003).
4.b.1).- A brief summary of Mexico’s ‘maquila’ exports.
One (somehow unexpected) outcome of widespread trade liberalisation among LDCs has been an increase in the country-concentration of exports, particularly manufactured exports -- in fact, in 2002 just nine countries account for about 90% of all manufactured exports from LDCs (one of the nine is Mexico with a figure of well over US$ 150 billion in 2002).15 However, Mexico is the most extreme case of an ‘asymmetric’ or ‘unbalanced’ relationship between export expansion and GDP growth among this group of ‘successful’ LDC-exporters. Economic reform and NAFTA certainly moved Mexico' s ‘engine of growth’ towards the export sector; however, for any engine to be at all effective, the power it generates must be properly 'harnessed' -- as Figure 8 illustrates, this certainly has not happened so far in Mexico.
● Exports do not include oil; 1980 prices. For the convenience of the graph, before transforming them into logs the original data were first made into a 3-year moving average and then into index-numbers with 1981 as a base year (equal to 20.1; the Ln of this number is 3).
● Source: Palma (2003).
In the case of the Mexican economy, its inability to harness the power generated by its remarkable export expansion seems to be its most important economic failure since President De la Madrid implemented his radical trade liberalisation policies in the early 1980s. In fact, as Figure 8 shows, not only did the long-lasting nearly 1-to-1 relationship between exports and GDP (built during ISI) practically disappear after 1981, but the sharp acceleration in the rate of growth of exports has been associated with a sharp decline in the rate of growth of the economy leading to a collapse of the implicit ‘export-multiplier’ -- from 0.9 (during the period 1970-81) to 0.1 (between 1981 and 2001). This is not exactly the promised land of the advocates of the reform!16
Proximity to the United States and NAFTA opened up opportunities for multi-product Transnational Corporations to move some highly labour-intensive, assembly-end part of their complex ‘value chain’ to Mexico, particularly to the frontier states. These NAFTA-opportunities provoked a dynamic ‘Lewis-type’ response, based on both Mexico’s almost unlimited supply of cheap (and relatively unskilled) labour, and on a large amount of ‘complementary investment’ (in the form of infrastructure, telecommunications, energy, other utilities, services, etc.) built mostly by the public sector during the ISI period. After twenty years, as the Lewis model predicts, even this particularly dynamic growth (16% per year, reaching US$ 80 billion in 2000, a figure significantly larger than all Brazil’s exports of goods and services that year) can still be entirely accounted for by additional labour; in fact, there has been neither productivity growth nor wage growth associated with it.
Figure 9 clearly illustrates the main weakness of the Mexican-style ‘maquila’ industry -- the relative low levels of value added and domestic inputs in its operations.