A partnership with academia

Building knowledge for trade and development

Vi Digital Library - Text Preview

Globalization and economic convergence: An assessment

Discussion Paper by Robert Rowthorn &Richard Kozul-Wright, 2000

Download original document (English)

What: This paper offers a critical survey of a strong globalization thesis that predicts a direct link from more open trade and investment regimes to faster economic growth in developing countries and income convergence across the global economy. Its examination of recent experience suggests that while in a more open and integrated world economy both the quantity and the quality of investment are influenced by external factors. Who: For teachers and students of globalization issues. How: As a background reading paper on globalization and economic convergence.


Robert Rowthorn &
Richard Kozul-Wright

No. 131

March 1998




Robert Rowthorn &
Richard Kozul-Wright

No. 131

March 1998

Comments from colleagues on earlier drafts of this paper are gratefully acknowledged, and in
particular those from Yilmaz Akyüz, Jan Kregel, Will Milberg and Paul Rayment. Final
responsibility resides with the authors.


- ii -

The opinions expressed in this paper are those of the authors and do not necessarily reflect the
views of UNCTAD. The designations and terminology employed are also those of the authors.

UNCTAD Discussion Papers are read anonymously by at least one referee, whose comments are
taken into account before publication.

Comments on this paper are invited and should be addressed to the authors, c/o Editorial
Assistant*, Macroeconomic and Development Policies, GDS, United Nations Conference on Trade and
Development (UNCTAD), Palais des Nations, CH-1211 Geneva 10, Switzerland. Copies of the
UNCTAD Review, Discussion Papers and Reprint Series may also be obtained from this address. New
Discussion Papers are available on the web site at: http://www.unicc.org/unctad/en/pressref/prdis.htm

* Tel. 022-907.5733; Fax 907.0274; E.mail: nicole.winch@unctad.org

JEL classification: F100, F200, N000 and O400.

- iii -


Chapter Page






A. Trade, openness and convergence 21
B. Capital mobility and convergence 28



- 1 -

* Faculty of Economics, University of Cambridge, United Kingdom.

** United Nations Conference on Trade and Development (UNCTAD), Geneva.


Robert Rowthorn* and Richard Kozul Wright**

This paper offers a critical survey of a strong globalization thesis that predicts a direct link from
more open trade and investment regimes to faster economic growth in developing countries and
income convergence across the global economy. Its examination of recent experience suggests that
while in a more open and integrated world economy both the quantity and the quality of investment
are influenced by external factors the forces driving capital accumulation retain strong domestic roots
and remain open to the influence of various types of policy initiative.


Since the late 1970s there has been a fundamental change in economic policy, beginning in the

industrial economies, then in developing countries and finally - and most dramaticall - in the former

socialist economies. Emphasis has been placed on a minimal role for the state, greater reliance on

market forces, and increased openness and integration into the world economy. Technological advances

have on some accounts already eroded longstanding geographical, ideological and political obstacles

to cross-border transactions, and transnational firms have been identified as the new engines of growth

and development. These same forces are expected to generate faster economic growth, particularly for

poorer countries, leading to convergence of incomes worldwide. From this perspective, policies still

matter, but only to the extent that dirigiste economic regimes resist implementing rapid and

comprehensive trade and financial liberalization along with the deregulation of domestic activity needed

to take full advantage of the new growth opportunities.

Among the most ardent supporters of this strong globalization thesis have been practitioners of

the dismal science. Indeed, while a good deal of the globalization debate has been characterized by

loose and speculative discourse, conventional economists have been able to establish an authoritative

voice by introducing a tighter analytical framework and a mass of empirical evidence. Their analysis

of globalization has centred on the greater mobility of capital and has done much to revive the flagging

intellectual fortunes of neo-classical growth and trade theory.

- 2 -

This paper offers a critical survey of some of the analytical and empirical arguments linking

globalization and convergence, and takes issue with the claim that capitalist economic development has

entered a new phase in which the domestic determinants of growth have become subordinate to

international economic forces. In the absence of a spontaneous link from greater openness to faster

economic growth, the notion of a policy agenda centred simply on the elimination of the state from

economic affairs is rejected.


Over the past two decades growing cross-border linkages have exerted powerful influences on the

shape and direction of the world economy (table 1). From 1973 to 1994, the volume of world exports
grew at an average annual rate of around 4.5 per cent, compared with 3.1 per cent for world GDP, but
with a marked acceleration after 1985 to 6.7 per cent. As a consequence, world exports of goods and

services in relation to world output rose from 12.1 per cent to 16.7 per cent over this period.

Table 1

The growth of international economic activity, 1964-1994

Export World FDI International World
volume flows bank loans real GDP

1964-1973 9.2 -- 34.0 4.6

1973-1980 4.6 14.8 26.7 3.6

l980-1985 2.4 4.9 12.0 2.6

1985-1994 6.7 14.3 12.0 3.2

Source: IMF (various years); BIS (various years).

Notwithstanding, international trade has not been the main catalyst for accelerating global

economic integration. That role has been played by international capital. Cross-border financial flows

have risen spectacularly over the past two decades, and the scope and depth of financial integration has

far outpaced that in goods markets. The abandonment of fixed exchange rates in the early 1970s, along
with a gradual loosening of capital controls opened the flood gates to short-term capital flows; average

- 3 -

1 Paul Rayment has reminded us, however, that intra-firm trade was probably already high in the 1960s in the context
of North-North intra-industry trade, although statistical limits make it difficult to give a precise figure.

2 On the changing nature of international trade, see Krugman (1996); on trends in financial flows see Felix (1996), Kregel
(1994) and Akyüz (1995); FDI trends are fully documented in UNCTAD’s World Investment Reports.

daily trade in the global foreign exchange market rose from $15 billion in 1973 to $880 billion in 1992
and over $1,300 billion in 1995. From 1980 to 1993, cross-border sales and purchases of financial
assets rose from less than 10 per cent of GDP in the United States, Germany and Japan to 135, 170 and

80 per cent, respectively. International banking has also, over this period, grown considerably faster

than world trade or output.

Direct investment flows have also made a significant contribution to global economic integration

in the sphere of production, and at a pace considerably faster than trade in goods and services. During

the 1970s, annual flows of foreign direct investment (FDI) averaged $27.5 billion, rising to $50 billion
in the first half of the 1980s and to $166 billion in the second half. Following a dip in the early 1990s,
they reached $318 billion in 1995.

While there is evidence that some of these flows have accelerated since the mid-1970s, it

is unlikely that by themselves they constitute a structural break in the evolution of the world economy.

Indeed, as may be seen from table 1, international economic integration was just as rapid in the 1960s
and early 1970s. Consequently, observers have pointed to more qualitative changes in the nature of
international trade and capital flows than was previously associated with the process of economic

integration. These include: the rise in manufactured exports from low-wage to high-wage economies

and the growth of intra-firm trade accompanying a finer geographical separation of production

activities;1 a shift in the composition of private capital flows from bank-lending to equity and portfolio

investments, particularly in respect of capital flows to developing countries, along with a tremendous

pace of financial innovation designed to reduce investors’ exposure to credit, liquidity and exchange

risk; a steady shift towards FDI in services, which now accounts for well over half of the total stock of

FDI and an increase in the flow to developing countries (accounting for over one third of total inflows
in 1993-1996), much of it linked to export-oriented manufacturing.2

These more qualitative changes in the pattern of economic flows have been associated with

institutional changes at the microeconomic and macroeconomic levels that are said to be contributing

to a much wider and deeper process of economic integration. The triumph of the market over the state -

exemplified by the collapse of communism in Eastern Europe - has, from this perspective, not only

widened the geographical scope of the international economy but has, by expanding the entry and exit

options of capital, fundamentally altered the interplay of political and economic forces, thus greatly

diminishing the independence and influence of purely domestic actors. Secondly, capital has become

a much more complex factor of production. A more elaborate system of intra-firm flows of goods and

- 4 -

3 See Greider (1997), Rodrik (1997), Williamson (1997) and Pfaff (1997). In this context, various authors are
rediscovering the seminal work of Karl Polanyi (1944).

services, as well as inter-firm alliances of various kinds, has emerged, giving rise to a more complex

pattern of specialization linked to a new system of international production. Moreover, the importance

of human capital is seen to be of growing importance in this system. Finally, the triumph of market

forces has been accompanied by an emerging liberal international regulatory framework, which has

further restrained the influence of domestic policy actions (e.g. Mankiw, 1995; Lawrence, 1993; Cable,

Strictly speaking, these pressures should culminate in a truly global economy, where all firms and

financial institutions operate transnationally - i.e. beyond the confines of national boundaries. In such

a world goods, factors of production and financial assets would be almost perfect substitutes

everywhere, and it would no longer be possible to consider nation states as distinct economic identities

with autonomous decision-making power in the pursuit of national objectives. Those public goods that
are needed to maintain an open-market system, such as secure property rights and a stable monetary

system, would become a global responsibility. Overall economic performance would depend upon the

response of firms to global market incentives and the effectiveness of global regulations.

The political, social and moral dimensions of such a global economy are beginning to provoke

considerable debate and controversy.3 Economists have reacted to descriptions of conflictual global

market forces with a mix of disdain and indifference and, while such responses have not always been

unwarranted, they have done little to help generate a more constructive dialogue (Rodrik, 1997). In
fact, the world economy is still a very long way from this situation. A more apt description of the

current situation is global economic interdependence, where cross-border linkages between markets and

among production and financial activities are now so strong that economic developments in any one

country are influenced to a significant degree by policies and developments outside its boundaries.

However, the extent and nature of that influence continue to depend upon a country’s resource

endowments, institutional arrangements and domestic policy choices.

The prospects of this interdependent world economy over the medium to longer term will hinge

on its ability to deliver not only a faster pace of economic growth but also a level of prosperity which

is widely shared by different countries and classes. To date, conventional economic analysis has been

able to monopolize the discussion on global prospects with a forceful assertion of the strong and

spontaneous links between greater openness, faster growth, and economic convergence. The next

chapter will consider this case in greater detail.

- 5 -

4 An extensive historical literature has also pointed out that successful development experiences over the last 200 years
do not conform to the advice of free traders. See Panic (1988), Bairoch (1993), and Bairoch and Kozul-Wright (1997).


In its analysis of contemporary globalization trends, conventional economics has drawn upon -

and in important respects refined - two of its most distinguished tools of analysis; trade and growth

theory. The classical theory of international trade was elaborated by David Ricardo some 200 years ago

and embellished a century later by Eli Hecksher and Bertil Ohlin. This theory teaches that under free

trade countries will specialize in those activities in which they have a comparative advantage. Other

things being equal, a country which is rich in a particular resource will export products whose

production utilizes this resource intensively and import other products. The resulting efficiency gains

lead to a higher level of global welfare as well as to optimal resource use for each individual economy.

Moreover, according to traditional trade theory, even in the absence of capital and labour mobility,

convergence of factor prices should accompany greater openness; workers of comparable skill will be

paid the same in the developed and developing worlds and owners of capital will likewise obtain the

same rate of return on their investments.

Despite its vaulted textbook status, the Hecksher-Ohlin model has long been subject to theoretical
and empirical criticisms. Its underlying assumptions, in particular those of a common technology set

available to all producers, perfect competition and constant returns to scale in production activities, have

been longstanding targets (Kaldor, 1981). But ever since Wassily Leontief's seminal work on the factor
content of United States trade and with the rising share of intra-industry trade in total world trade,

empirical studies of trade flows have been against orthodoxy.4 These theoretical and empirical

challenges have extended to the idea of factor price convergence (Bhagwati, 1994).
Salvaging traditional trade theory has taken it in a direction consistent with globalization trends.

In particular, the increasing international mobility of capital has reduced the importance of differences

in the level of capital stock in determining a country’s comparative advantage, while the growth of trade

in goods with high knowledge and skill content has increased the importance of relative endowments

of skilled and unskilled labour. Clearly, at any given time, countries vary with respect to their

endowment of educated labour, and the logic of classical theory would teach that countries which are

well-endowed with human and physical capital will specialize in goods and services which utilize them

intensively, while those which are badly endowed will specialize in other types of production. Recent

trends in the world economy have been seen to accord well with this conclusion, particularly the rise

of North-South trade. Moreover, factor price convergence has resurfaced as a key feature of

globalization (Wood, 1994; Williamson, 1997).

- 6 -

5 A more complex description of the interlinkages between technology, trade and investment did emerge from efforts to
reduce the empirical residual discovered by growth empirics. But the idea that catching up by developing countries was an
uncertain and far from spontaneous process was, to a large extent, inconsistent with conventional analysis. See Gerschenkron
(1962), Abramovitz (1986), Maddison (1982) and Freeman (1993).
6 There is considerable room for choice of variables in convergence studies. Some of these studies look at real GDP per
head of population, while others focus on real wages, GDP per worker or working hour, or total factor productivity; some look
at the whole economy, others at a subsector of the economy, such as manufacturing or services. See Baumol et al. (1989) for
a useful discussion of convergence measures.

This revised Hecksher-Ohlin model has also opened more direct links to the neo-classical growth

theory, missing from earlier expositions. An original growth model developed by Robert Solow and

Trevor Swan associated with a well-behaved neo-classical production function generated a steady state

on which all countries would converge in the long run, assuming that the savings rate, population

growth and technologies were the same everywhere. In the transition, growth would be faster in poorer

countries than in the richer economies because capital scarcity in the former would generate a higher

rate of return to capital, a faster pace of capital accumulation, and consequently a faster per capita

growth rate. However, an extensive literature on growth accounting failed to uncover the predicted

convergence in per capita incomes, and subsequent research on economic growth focused more on

differences in growth performance across countries than on similarities.

A notable feature of this first generation of neo-classical growth models was their abstraction from

international economic factors and consequently from trade theory.5 However, a renewed emphasis on

capital mobility has helped cement a marriage between these two branches of conventional economic

analysis. This marriage has been reinforced by an extensive body of new comparative empirical data

(Summers and Heston, 1988).
The heart of this union has been a renewed emphasis on economic convergence. The chosen

methodology has been the testing of explanatory variables in cross-sectional regression analysis, where

rates of economic growth are the independent variable and several variables are introduced to control

for the steady state. This new convergence literature is built around the distinction between beta

convergence and sigma convergence (Sala-i-Martin, 1996). Beta convergence occurs when there is an
inverse relationship between the initial value of some variable (per capita income, productivity, etc.) and
its future growth.6 Thus, areas with a low initial value of this variable have a tendency to grow

relatively fast and catch up with more advanced areas. Sigma convergence is not concerned with

catching-up per se or with the advantages of backwardness, but with the dispersion of per capita

income, or some related variable, and with the evolution of dispersion through time. The dispersion of

a variable across areas is typically measured using the standard deviation of the log or the coefficient

of variation (standard deviation divided by the mean). Sigma convergence is said to occur when the
relevant measure declines through time. Note that an eventual stability in dispersion is consistent with

continued mobility in the ranking of individuals, since individuals may be changing places without

- 7 -

7 See Sachs and Warner (1995a) and World Bank (1996). This policy has recently been strongly promoted as a solution
to economic stagnation in Africa. See, for example, Collier and Gunnin (1997).

affecting the overall distribution of incomes. Beta convergence although a necessary, is not sufficient

reason for sigma convergence. Beta convergence may produce quite a strong regression toward the

mean, but this may be offset by a flow of random shocks which maintain dispersion at a constant level.

In the light of the limitations of earlier growth models, two types of beta convergence have been

distinguished: absolute and conditional. Under absolute convergence, catching-up is observed in reality,

so that backward areas actually do grow faster on average than more advanced ones simply because they

are poorer. However, as noted earlier, this result rested on a rather narrow set of determinants of the

steady state. Broadening that set and allowing economies to have different steady states suggests that

backward areas still have the potential to grow faster than advanced areas, but this potential will only

be realized if they satisfy certain conditions. Where these conditions are not satisfied, then the growth

rate in backward areas may be as slow, or even slower than in advanced areas. Moreover, because

economies converge on their own steady state there is no assumption about a final state where all

income levels are identical.

Testing these new trade and growth theories has generated two empirical findings which have been

dominating recent discussions of globalization. The first is the apparent uniformity of conditional

convergence rates. Although authors differ widely in the variables they include, the general result seems

to be that, controlling for other factors, convergence in per capita income occurs at around 2 per cent

a year. Table 2 reports the results obtained by Sala-i-Martin for a variety of countries and regions. A

second finding has suggested that among all the possible conditioning variables the most fundamental

is how open an economy is to the world economy. A recent and well-known paper in this tradition is

by Sachs and Warner (1995a), who use a five point system to classify countries into "open" and

After controlling for other variables and initial income levels, they find that open economies grew

faster than closed ones. They also find strong evidence of beta convergence (catching up) among open
economies, but no sign of it among closed ones. On the basis of these findings, they advocate that

countries should deregulate their economies, roll back government intervention and comprehensively

liberalize their external economic relations.7

- 8 -

Table 2

Estimates of the speeds of $-convergence for a variety of data sets

Long-run single regressiona Panel
estimates of





World (110 countries) - 1960-1990

OECD countries - 1960-1990

United States (48 states) - 1880-1990

Japan (47 prefectures) - 955-1990

Europe (total 90 regions) - 1950-1990

- Germany (11 regions)

- United Kingdom (11 regions)

- France (21 regions)

- Italy (20 regions)

- Spain (17 regions)
































Source: Sala-i-Martin (1996).
a The regressions use non-linear least squares to estimate equations of the form:

( 1 T ) ln (yit yi, t&T) ' a & [ ln (yi, t&T ] (1 T )
+ other variables. Where yi, t-T is the per capita income in country or region i at the beginning of the interval
divided by the overall CPI. T is the length of the interval; "other variables" are regional dummies and sectoral
variables holding constant temporary shocks, that may affect the performance of a region in a manner correlated
with the initial level of income.

Each column contains two numbers. The first one is the estimate of $. Underneath it, in parentheses, its standard
error. The constant, regional dummies and/or structural variables are not reported in the table. The coefficients
for Europe total include one dummy for each of the eight countries. Columns 1 and 2 report the value of $
estimated from a single cross-section using the longest available data; column 1 reports the coefficient when the
only variable held constant is the initial level of income; column 2 reports the value of $ estimated when additional
variables are held constant; column 3 reports the panel estimates when all the subperiods are assumed to have the
same coefficient $. This estimation allows for time effects. For most countries, the restriction of $ being constant
over the subperiods cannot be rejected (see Barro and Sala-i-Martin, 1995).

- 9 -

8 This is know as "Galton<s Fallacy" (see Friedman, 1992, and Quah, 1993). Galton found that the sons of tall men were
on average likely to be shorter than their fathers, from which he inferred that the proportion of tall men in the population would
decline and hence there would be a shortage of men suitable for army service. This inference ignores the fact that some
relatively short men will have tall sons, so that the pool of tall men will be constantly replenished by recruitment from below.
Thus, regression towards the mean, which tends to reduce dispersion, will be accompanied by random variation, which tends
to increase dispersion. In equilibrium, these two forces will exactly offset each other, so the distribution of heights will remain

9 For a good discussion of these issues, see Bliss (1997).
10 For the evidence of growing inequalities within countries, see UNCTAD (1997, Part II, chapter 3).

Despite the impressive display of empirical techniques, the idea of conditional convergence has

been challenged on various methodological grounds. The finding that poorer countries tend to grow

faster than rich economies may be perfectly correct, and it may also indicate some kind of causal

relationship whereby poor countries have a genuine growth advantage over rich ones. However, the

existence of such beta convergence does not automatically imply that international dispersion of per

capita incomes of productivity levels will fall in the course of time.8 If the initial dispersion is very

large, the existence of beta convergence may imply a future reduction in dispersion and hence a period

of sigma convergence. But this will not continue indefinitely, since individual countries are subject to
continuing stochastic shocks, whose effect is to increase dispersion. There will be an eventual

equilibrium in which beta convergence (regression towards the mean) is offset by continuing random
variation, and the distribution of per capita incomes will stabilize. Sigma convergence will then come

to a halt. This point is now well understood, and it is widely recognized that beta convergence is a

necessary but not sufficient condition for sigma convergence.9

A second criticism of the concept of conditional convergence dismisses it as vacuous, on the

grounds that ascertaining how far each country is from its own steady state ignores the more

fundamental question why some countries are rich and some are poor and has consequently failed to

grasp the long-run determinants of economic success (e.g. Quah, 1996a, 1996b; Hall and Jones, 1997).
Closely related to this is the role of technological progress, which has been identified as the Achilles heel

of the new convergence literature because - and much as in earlier growth models - technology is seen

as a public good freely available and useable everywhere, making it an exogenous influence on the

growth process (e.g. Romer, 1995; Bernard and Jones, 1996).
But perhaps of even greater significance to the immediate debate on globalization are serious

doubts about the empirical evidence on fast growth and convergence with the world economy becoming

more open. As table 3 suggests, there has been a persistent slowing down of the world economy as it

has become more closely integrated over the past two decades. Moreover, this slowdown in growth has

been accompanied by increased volatility of growth performance (Felix, 1996). Perhaps most
significantly, the pattern of economic growth since the late 1970s has coincided with ever larger income

gaps not only across countries but also within them.10

- 10 -

Table 3

The slowdown of growth in the world economya

Number of countries
1972-1981 1982-1991

Decadal growth higher than in 1960-1971 18 10

Decadal growth lower than in 1960-1971 39 47

Monotonically rising -- 4

Monotonically falling -- 17

Negative growth (3 in the 1960s) 5 17

Source: Felix (1996, table 2)
a Fifty-seven countries, including all OECD, Asian and Latin American economies whose 1983 GDP was above $10

billion, all Middle Eastern/North African economies with GDPs above $7.5 billion, and all sub-Saharan African
economies with GDPs above $5billion.

On a global level, the evidence on dispersion is remarkably consistent and clear. The international

dispersion of per capita incomes has been increasing rapidly for more than a century. The same is

presumably true for productivity, although comprehensive evidence on this does not seem to be

available. In an interesting study for the World Bank, Pritchett (1995) estimates what per capita incomes
in a wide range of countries were in 1870 and how they have altered. He finds that the ratio of GDP

per capita of the richest to the poorest country rose from 8.7 in 1870 to 38.1 in 1960 and 51.6 in 1985.
The standard deviation of log per capita income, which was between 0.513 and 0.636 in 1870, rose

from 0.867 in 1960 to 1.025 in 1985. Equally telling, of the 98 developing countries for which data
are available, 40 countries had income levels of over 20 per cent of the average per capita income of the

G-7 countries in 1960 and 14 countries over 40 per cent. By 1990, the figures had fallen to 29 and 11,
respectively. More generally, failure to catch up in this period is confirmed by the fact that the number

of developing countries with per capita incomes over 80 per cent, 60 per cent and 40 per cent of the
average per capita income of the G-7 countries all decreased between 1960 and 1990. There is little

evidence to suggest that this trend has been reversed in the past decade or so. Indeed, if anything, the

evidence points to a strengthening of the polarization forces (IMF, 1997).
Nor is more detailed examination of the historical evidence linking globalization and convergence

very convincing. This is particularly true for the period before 1913, which has recently been revisited

in search of lessons for the contemporary era (IMF, 1997). One recent study examining this period
found mixed results: convergence in a core of dynamic economies driven by catch-up growth in a small

- 11 -

11 See UNCTAD (1997). On the other hand, some poor countries have experienced fast growth, including such populous
countries such as China and Indonesia. This should be borne in mind when evaluating the finding that global dispersion has
been increasing rapidly. Such a finding may be biased by the experience of small countries and, if weighted by population
increase, the conclusion might be different.

12 Among the first authors to examine beta convergence was Baumol (1986), using data on 16 developed countries over
the period 1870-1979. His method was to regress the annual growth rate of labour productivity against the logarithm of the
initial value of this variable. He found a highly significant negative relationship, indicating that on average countries with
initially low productivity grew faster. These findings were criticized by De Long (1988) on grounds of sample bias and
measurement error. Such criticisms were accepted in principle by Baumol, but later studies by Baumol et al. (1989) and others,
who avoid the pitfalls identified by De Long but have still found strong evidence of beta convergence in the relatively
advanced economies. This has given rise to the idea that a certain threshold had to be crossed to become a member of the
convergence club.

13 See Levine and Renelt (1992); Dollar and Wolff (1993); O’Leary (1995).
14 It is interesting to note that, despite growing dispersion within individual sectors, the dispersion in economy-wide
productivity did not increase in the European Union during the 1980s. This was because structural shifts from low to high
productivity sectors offset lagging performance in individual sectors (see Rowthorn, 1992).

group of Western European late-industrializing economies was accompanied by divergence for the

world economy generally as well as for Continental Europe (UNCTAD, 1997).
Growing dispersion over the past four decades has been most noticeable in the developing world,

accompanied by a significant increase in the absolute income gap between the richest and poorest

developing countries and a near doubling of the ratio of maximum to minimum per capita.11 Moreover,

there were considerably fewer developing countries with income levels ranging from 40 per cent to 80

per cent of the income of the richest developing country in 1995 than in 1965, and a much larger

number of countries had fallen by 195 into the category of the poorest 20 per cent. These forces of
polarization appear to have intensified in the early 1980s (IMF, 1997, table 17).

Even among the "convergence club" of OECD countries the picture is rather more mixed than is

often assumed. Since 1870 dispersion in the OECD as a whole has been falling steadily, although

relative to the United States the position of many countries has fluctuated a great deal in terms of both

labour productivity and per capita income (table 4).12 Prior to 1950, the United States surged ahead of
Europe, but much of this gap was closed during the long post-war boom.

The most striking example of this has been that of Japan, whose per capita income is now similar
to that of Switzerland and only a few per cent behind the United States. However, evidence since the
mid-1970s appears more mixed.13 One striking feature of the recent period has been the failure of most
of the poorer countries of the European Union (EU) to sustain the impressive performance they
exhibited during the 1960s. With the exception of Ireland, which has grown rapidly over the past
decade, per capita income in these countries is much the same now relative to the EU average as it was
20 years ago (figure 1). Similarly, while Barro and Sala-i-Martin (1995) show that within most OECD
countries there has been a massive and prolonged reduction in regional disparities, they also show that
this process of convergence has come to a halt in recent times and in some cases has even gone into
reverse. Finally, across a broad range of industries, disparities in labour productivity within the
European Union diminished during the 1970s, but have widened again since then (table 5).14

- 12 -
















Table 4

GDP per capita for 16 currently developed countries
(Percentage of US level)

1870 1910 1950 1970 1990

Australia 140 102 69 76 74
Austria 64 56 33 59 71
Belgium 90 69 64 73
Canada 59 70 71 80 93
Denmark 69 63 61 75 77
Finland 42 34 40 61 77
France 70 53 49 72 78
Germany (West) 55 50 41 72 78
Italy 54 42 33 62 72
Japan 27a 24 19 64 88
Netherlands .. 65 55 73 72
Norway 53 41 53 65 84
Sweden 62 55 66 84 81
Switzerland .. 66 76 95 86
United Kingdom 120 86 66 70 74
United States 100 100 100 100 100

Source: Maddison (1991)
a 1890

Figure 1

GDP per capita of EU periphery (EU-15 = 100)

- 13 -

15 Gerschenkron’s Economic Backwardness in Historical Perspective was, after all, a classic application of conditional
convergence theory. There is a noticeable failure among contemporary convergence theorists to acknowledge this earlier

Table 5

GDP per worker hour in the EU-11: 1970-1990
(Coefficient of variation - percentage)


1970 1980 1990 1970-80 1980-90

Whole economy 33.13 27.22 27.54 -5.91 0.32
Agriculture 38.59 37.20 44.80 -1.39 7.60
Services 27.79 21.56 24.72 -6.23 3.16

Manufacturing 24.84 22.24 27.68 -2.60 5.44
of which:

food 31.31a 22.66 21.10 -8.65 -1.56
textiles 23.85a 25.87 33.11 2.02 7.24
paper 23.12a 17.64 24.47 -5.48 6.83
chemicals 31.58a 32.08 35.02 0.50 2.94

mineral products 26.60a 22.59 22.76 -4.01 0.17
basic metals 34.67a 36.43 27.89 1.76 -8.54
machinery & equipment 16.49a 19.76 27.09 3.27 7.33

Source: O'Leary (1995)
a 1972


These theoretical and empirical challenges to the conventional analysis of economic convergence

are considerable. However, that analaysis has helped to reassert the idea that being backward has

advantages, and that if certain conditions are satisfied then backward countries will develop rapidly.

This is an important insight, and the challenge which it presents is to identify precisely what conditions

must be satisfied in order to exploit catch-up opportunities and achieve rapid growth. This is not


different from the traditional activity of development economists, and the theory of conditional

convergence is really just the traditional approach in a new guise.15

- 14 -

16 Most economists have suggested that exogenous technological changes can explain the coincidence of globalization and
rising inequality. However, this argument has surprisingly little empirical support. Moreover, it is difficult to reconcile with
the growth slowdown.

However, the use of complex econometric techniques has, at most, given rise to a vast empirical

literature and a large number of potentially significant variables, without shedding much light on

endogenous growth dynamics.16 Indeed, according to one sympathetic reviewer of this literature:

"Policy makers who want to promote growth would not go far wrong ignoring most of the vast

literature reporting growth regressions. Basic theory, shrewd observation and common sense are surely

more reliable guides for policy" (Mankiw, 1995, pp. 307-308).
In response, several authors have examined the dynamics of global growth and income distribution

using transition matrices. Countries are classified into N groups according to some measure, such as

per capita income, qualitative stage of development or growth rate. A matrix P is then constructed

which specifies, for each j and k, the probability pjk that an arbitrary country in group j will move into
group k within a certain length of time, normally a year. These probabilities may be estimated on the

basis of historical evidence or they may be chosen for theoretical reasons.

The first interesting feature of simulation work in this area is the sensitivity of outcomes to the

underlying assumptions. This is brought out clearly by Pritchett (1995), who considers a sample
consisting of 14 "rich" countries" and 103 "developing" countries. In the former group per capita

income is assumed to grow at a constant g = 1.8 per cent per year. For developing countries, the growth

rate is a stochastic variable which belongs to one of four categories: stagnation g = 0.5 per cent; plateau

g = 1.8 per cent; boom g = 1.8*(rich country income/y), which implies that a boom is faster the poorer
a country is when it starts, but that growth eventually settles down to the "rich" country rate; and

an"implosion" period in which g = - 1.8*(y - 250)/y, which creates the possibility of large recessions
(especially for poorer countries), but with no country going below the minimum. All developing
countries begin the simulation in stagnation. Starting with the existing distribution of world income,

Pritchett shows how different, plausible, assumptions about the probability of transition out of

stagnation can generate radically different outcomes, ranging from extreme polarization of incomes to

fairly rapid convergence. His results are summarized in table 6.

A second interesting feature is the importance attributed to the time frame of the analysis. Quah
(1993) divides countries into a number of groups, depending on their per capita income relative to the
global average. From historical evidence for the period 1962-1984, he then calculates the probability
that a country in one group will move into another group in the following year. The resulting transition

matrix is shown in table 7. If the probabilities shown in this table are held constant and the evolution

of income distribution through time is examined, it turns out that incomes will eventually stabilize in a

"twin-peaked" distribution with both many poor and many rich countries.

Table 6

Simulations of alternative paths of divergence or convergence, depending
on assumptions about transition out of stagnation of developing countries

Column I II III IV



Exogenous (at 1.5
per cent per year)

Exogenous increase
in 1995 due to
better policies

IIncreasing with
higher absolute


Decreasing with the
relative distance
from the leader

1870 1995 2120 1870 1995 2120 1870 1995 2120 1870 1995 2120

to maximum

6.6 40.7 340 7.7 44.8 13.4 7.7 46.8 2.8 7.3 45.2 276

Std. dev. in

0.49 1.1 1.6 0.54 1.14 0.472 0.53 0.99 0.15 0.53 1.15 1.46

Source: Pritchett (1995)

- 16 -

Table 7

Transition probabilities

Year t+1

Group 5 Group 4 Group 3 Group 2 Group 1

Group 5 0.97 0.03
Group 4 0.05 0.92 0.04

Group 3 0.04 0.92 0.04
Group 2 0.04 0.94 0.02

Group 1 0.01 0.99

Source: Quah (1993). In Quah's paper, grouping is based on real GDP per capita (relative to the world average). Group 5
consists of countries with less than or equal to B14 of world average income; groups 4, 3, 2 and 1 have upper 4 equal
to B12 , 1, 2 and times average income. Note that row totals may not sum to unity because of rounding errors.

Quah defines his country groupings with respect to average world income, so that convergence
towards the top of the range is impossible, since not all countries can simultaneously enjoy per capita
incomes several times the world average. With Quah’s criterion of classification, the direction of
convergence can only be towards the middle of the range. This property makes it inconvenient for

exploring variations on his approach, so we shall assume that the country groupings are defined

according to some criterion which allows for convergence towards extreme points. One possibility is

to group countries according to their stage of development, for example underdeveloped, lower

intermediate, intermediate, higher intermediate, developed. With such a classification, it is theoretically

possible for countries to cluster anywhere in the range, and convergence can occur towards the top or

bottom. With this re-interpretation, Quah’s estimated transition matrix tells us that in the long run there
will be many underdeveloped countries and many highly developed ones, with relatively few countries

in between. It also implies, with probability one, that any underdeveloped country will eventually move

up through all of the stages to become fully developed, and conversely that any developed country will

eventually move down through all the stages to become underdeveloped. Thus, Bangladesh will

eventually reach the most advanced stage of development, while the United States will fall back to the

least developed stage. It is this two-way movement which produces Quah’s stable long-run distribution.
The experience of the past 150 years suggests that countries do move downwards, but only to a

limited extent. Even ruinous wars, economic cataclysms or prolonged stagnation rarely reduce the per

capita income of countries by more than 50 per cent, and there is no recorded case of a country moving

from being highly or even moderately developed into the lowest category. There seems to be some kind

- 17 -

of ratchet effect at work, and the question is how do we allow for it within the framework of transition

matrices? As an illustration, let us assume that the economic status of a country can deteriorate by only

a limited amount, so that a country in group N can fall back into group N+1, but having done so cannot

fall back still further into group N+2. Since there are only five groups altogether, this is a fairly weak

assumption. Yet its impact is dramatic. Assuming that all other probabilities are as in Quah, we find
that all countries will eventually end up in the top two groups.

Starting with all countries initially in the bottom group, figures 2 and 3 compare how the

distribution of countries evolves, using Quah’s original transition probabilities and when these are
modified by means of the partial ratchet effect described above. With a ratchet present, the initial

polarization of incomes is even more rapid than in Quah’s case, but eventually the gap begins to narrow
as more countries leave the bottom and migrate into the upper groups.

However, this eventual convergence is very slow, and from the initial starting point it takes 330

years to reach the stage where 95 per cent of countries are in the top two groups. Convergence can, of

course, be speeded up by policy measures which increase the probability of upward mobility. Figure

4 shows what happens when the probability of upward movement by each country into the group above

is increased by 1 percentage point per year. Within 150 years from the initial starting point, 80 per cent
of all countries are in the top two developmental groups, and within a further 80 years the proportion

rises to 95 per cent. Thus, quite plausible modifications to Quah’s transition probabilities generate rapid
global development. Figure 5 shows what happens to income dispersion under the various scenarios.

Such accelerated development is consistent with the experience of the East Asian economies and

with the earlier generation of late industrializers in North America, Western Europe and Japan. In all

these cases the threshold to convergence was crossed during a very fast pace of growth over just a few
decades. Moreover, this analysis is equally consistent with the possibility of economies getting stuck,

especially when it is recognized that capitalist dynamics can involve technological and institutional

upheavals which mark a structural break.

The transition matrix provides a useful technique for summarizing the diversity of historical

experience and, as such, might be a better guide to future possibilities. However, since the matrix is

concerned with the experience of a diversity of countries affected by a wide variety of economic and

political factors, the probabilities in this matrix may be a poor guide to the prospects facing any

particular country. For example, the matrix may show that over a certain period of time, the average

poor country had a probability x of moving into a higher developmental stage. This probability may

be derived from the experience of a small number of successful countries belonging to a certain region,

say East Asia, and following some particular policy, say strategic trade intervention. For a country

belonging to some other region and following a quite different trade policy, the probability of

developing may be virtually nil without some major shift in policy or some radical change in the world

- 18 -

Group 1
(Highly developed)

Group 2

Group 3

Group 4

Group 5







0 50 100 150 200 250 300 350 400 450 500













Group 1
(Highly developed)

Group 5







0 50 100 150 200 250 300 350 400 450 500

Years elapsed











Group 2

Group 3

Group 4

Figure 2

Evolution of income distribution: Quah probabilities

Figure 3
Evolution of income distribution:
Quah probabilities plus ratchet

- 19 -

Group 1
(Highly developed)

Group 2

Group 3
Group 4

Group 5







0 50 100 150 200 250 300 350 400 450 500

Years elapsed

















0 50 100 150 200 250 300 350 400 450 500

Years elapsed













with ratchet

Quah probabilities

with ratchet & more
upward mobility

Figure 4

Evolution of income distribution:
ratchet plus increased upward mobility

Figure 5

Global income dispersion
Quah probabilities plus ratchet

- 20 -

17 However, this coefficient declines in magnitude and loses significance when certain other variables are included in the
regressions. In the former paper, this occurs when a variable for openness is included, and in the latter when a variable for life
expectancy is present. It is a mystery why investment should lose its significance when a variable for life expectancy is present,
and this result must be a statistical fluke. Barro and Sala-i-Martin (1995) also found that economic growth is negatively
associated with secondary and higher education for women. This implausible result puts a question mark over all of their
findings and those of similar studies.

environment. Thus, transition matrices may be useful as a historical summary and for conducting

thought experiments, but they cannot be a substitute for a careful assessment of the policies required to

achieve faster growth and distributional objectives.


The discussion so far suggests that, contrary to the strong assertions of many conventional

economists, the links from increased global economic integration to economic convergence are neither

linear nor spontaneous, and that any assessment of future prospects will be sensitive to initial conditions,

policy choices and the time frame of the analysis. It is also clear that choosing the appropriate starting

point for analysing an interdependent global economy is very important. The concern with static

efficiency gains from getting market prices right is limiting in this respect. Moreover, much of the

empirical literature does suggest an alternative and more appropriate place to begin - and on which there

is a consensus - namely capital accumulation, including investment in human capital.

The central role of investment is confirmed in a paper by Levine and Renelt (1992), which
examines whether the conclusions of existing studies are robust or fragile to small changes in the

conditioning information sets. They find that almost all results are fragile, and that the only robust

correlations are between growth and the share of investment in GDP, and between the investment share

and ratio of international trade to GDP. More recently, Sachs and Warner (1995b) have found
investment to be statistically significant. Sachs and Warner (1995a) and Barro and Sala-i-Martin (1995)
also find that investment has a positive and statistically significant coefficient in some their equations.17

This starting point, moreover, is in line with the earlier work on economic catch-up as well as on recent

work on the East Asian miracle, which informs much of the convergence and globalization literatures.

For the vast majority of developing countries, sustained economic growth requires shifts in the structure
of production, productivity growth and technological upgrading, all of which have very strong and

complementary links to a fast pace of capital formation. Putting capital accumulation back into the

globalization debate requires looking at international economic flows in a rather different light.

- 21 -

A. Trade, openness and convergence

The contribution of trade to accelerated growth in developing countries comes through the familiar

efficiency gains associated with greater uniformity in prices for internationally traded goods, as well as

more dynamic gains linked to heightened international competitiveness and the advantages of

specialization. However, analysis of the links between trade and economic growth has long been

hampered by ambiguities surrounding the categorization of national trade policies. Two concepts,

openness and outward orientation, have often been used interchangeably for this purpose. The tendency

to equate these two concepts in describing and assessing trade policy has its roots in the conventional

trade theory that policies combining protection with export promotion produce the same result as those

relying on full-scale import liberalization, because they are equivalent in terms of incentive structure -

a proposition that is valid only under very special conditions. Openness describes a situation

characterized by the absence of restrictions on flows of goods across national borders, notably on

imports, while outward orientation usually depicts a strategy of emphasizing world markets as an outlet

for domestic producers, and is often synonymous with export promotion. Countries which can be

described as outward- or export-oriented in the above sense can have important, but selective,

restrictions on imports, even though such barriers might be lower than in countries emphasizing

domestic rather than world markets.

The tendency to equate these two concepts has also given rise to a confusing proliferation of

empirical indicators, which are used interchangeably in connection with trade policy. However, findings

of a recent survey examining six common measures of trade policy indicate that countries can be very

differently ranked according to the measure used, and there are few significant cross-country

correlations between these measures (Pritchett, 1996). These findings raise obvious questions about the
reliability of the various indicators in capturing some common aspects of trade policy, and about the

interpretation of the empirical evidence on the relation between trade and economic performance.

Despite these familiar concerns, a number of studies have reported that economic openness

facilitates catching-up by backward countries and leads to a reduction in the dispersion of per capita

incomes. The work of Sachs and Warner (1995a) has become an established reference point in these
discussions. However, their analysis is subject to several criticisms. They are probably correct to argue
that economic openness, in their sense of the term, is normally conducive to catching-up. However, this

is not very helpful because their definition is too broad to be really useful. In the case of trade barriers,

an economy is classified as open if its average tariff rate is less than 40 per cent and non-tariff barriers

cover less that 40 per cent of trade. According to this criterion, most Asian NIEs are classified as

consistently open over the past 25 years, despite their well-documented interventions in the sphere of

trade, finance and direct investment. The Republic of Korea and Taiwan Province of China, for

example, pursued vigorous strategic trade policies designed to stimulate local industry and promote

- 22 -

18 The theoretical foundations of dynamic gains from free trade are indeed controversial. See, for example, Rodrik (1997,
p. 30), and the articles on international competitiveness in Oxford Review of Economic Policy (Vol. 12, autumn 1996).
19 There is also plenty of evidence to suggest that a switch from primary to manufactured exports if not properly managed
can very easily get stuck. See Hirschman (1995, ch. 2) and UNCTAD (1992).
20 See UNCTAD (1996, ch. II) for an extended discussion of these interactions; also Akyüz et al. (1998).

exports through the use of controls on inward FDI and selective import restrictions. Such policies were

highly selective and biased towards machinery and intermediate goods imports. Tight restrictions on

some imports were accompanied by low or zero tariffs on others, so the average tariff rate and overall

quota coverage were quite low. By using a measure of openness which conceals these facts, Sachs and

Warner are ignoring the central issue of policy debate at the present time. To say that countries will

suffer if they have an average tariff rate or quota coverage of more than 40 per cent tells us nothing

about the impact of strategic trade policies involving more selective forms of intervention. It is

unjustified to use the findings of the Sachs-Warner study to support claims that comprehensive
deregulation is good for economic growth and that countries should refrain from East Asian-style

intervention. Indeed, there is ample evidence that price distortions in these economies were pervasive,

even more so than in many other developing countries with a much less impressive economic

performance. Rather, trade policies were designed to attain dynamic gains from linking trade to capital

accumulation and technological change.18

Other authors have refined the openness and convergence argument in a number of ways. One

approach has been to link the importance of trade to country size. Small developing countries have the

most to gain from globalization by allowing them to reap economies of scale by specialization, and then

to switch into increasingly capital-intensive production sectors before diminishing returns appear in the

original labour-intensive lines of specialization. This argument is usually illustrated with reference to

the successful East Asian experience (Findlay, 1996).
While this is in many respects a more reasonable argument than that of Sachs and Warner, the idea

that the East Asian success was the spontaneous outcome of market forces is not a tenable one and their

avoiding getting stuck required determined policy interventions as economies began to upgrade from

simple labour manufactures to those with a higher capital and technology content.19 At all levels of

development sustained export expansion depends on the creation of additional production capacity in

industry as well as productivity growth, which is dependent upon new investment. Thus a sustainable

growth process requires mutually reinforcing dynamic interactions among savings, exports and

investment.20 Such a process needs to be characterized by continuously rising exports, domestic savings

and investment, both in absolute terms and, for most of the industrialization process, as a proportion of

GDP. Initially, investment is likely to exceed domestic savings by a large margin, with the difference

being financed by net inflows of capital, but over time the external gap should narrow as exports and

savings grow faster than investment.

- 23 -

21 Like most developing countries, primary exports were the initial source of foreign exchange earnings in the Republic
of Korea and Taiwan Province of China in the early 1950s. Also, as in many other developing countries at that time, efforts
were made from an early date to move into manufactured goods, initially in wood and paper products more significantly, and
as was earlier the case in Hong Kong in textiles and clothing. In both the Republic of Korea and Taiwan Province of China,
labour-intensive exports as a share of their domestic production rose continuously and rapidly from the mid-1960s, reaching
60 per cent in the early 1980s, although in the case of Taiwan Province of China resource-intensive manufactured exports
remained relatively important as a source of foreign exchange for a longer period of time. See Hentschel (1992).

In East Asia, the export-investment nexus was built around traditional low-skill industries.

However, the emergence of successful export-oriented industries does not follow spontaneously from

rapid capital accumulation.21 Although these countries enjoyed a comparative advantage in labour-
intensive manufacturing industries, this did not mean that investments in these industries were

automatically forthcoming. In most cases, they still needed government support in financing,

international marketing, and sometimes even export subsidies in order to become viable in the world

market. Nevertheless, it is true that the development of these industries was largely based on

encouraging productive investments, and that at this stage the East Asian countries were not as

extensively engaged in a selective industrial policy as they were at a later stage. However, even if

investments are made in industries with static comparative advantages arising from existing resource

endowments, including labour, constraints on the room for productivity growth means that rising

domestic wages and loss of competitiveness can quickly bring a halt to an export drive. In anticipation

of such potential difficulties, East Asian economies, from the early stages of their development,

gradually and purposefully nurtured new generations of industries with a greater potential for

innovation, exporting and productivity growth in the long run.

Thus, in these countries measures to promote investment were linked to the establishment of

domestic capital and intermediate goods industries and technological upgrading (UNCTAD, 1994).
These included reinstituting import restrictions, rolling back tax exemptions on the import of certain

intermediate and capital goods and granting higher investment tax credits to businesses purchasing

domestically produced machinery. In addition, a policy was pursued of building up a technological

capacity at the national, industry and even firm level. In these economies, tax and other incentives for

enterprise training were complemented by a more detailed national training programme which placed

greater emphasis on technical subjects at higher levels of education and on greater industry involvement
in vocational training schemes. Measures to facilitate local R&D, including direct financial subsidies,

have also been extensively used.

Given that technological and organizational leadership was located abroad, the transfer and

adaptation of foreign technology was, from an early stage, recognized by policy makers in East Asia

as a critical link in the process of industrial upgrading. Acquiring new technologies required generating

foreign exchange, and it was for this purpose, as well as to provide an outlet for industrial production

without promoting domestic consumption, and to expose the recipients of rents to a measured degree

of competition on world markets, that East Asian governments encouraged exports. As a result of these

- 24 -

22 This is integral to the flying geese model of development (see Korhonen, 1994).
23 However, the dismantling of wartime controls on trade did begin earlier and was a gradual process aided by the
European Payments Union (1950-1958).

various policy initiatives, the first-tier NIEs have been particularly successful in upgrading their structure

of manufacturing output towards scale- and skill-intensive activities. By the second half of the 1980s,
the share of these activities in total manufacturing output had passed that of resource and labour-

intensive activities. The rising share of these scale and skill-intensive goods in total manufacturing

exports (although initially lagging their share of total output) also began to accelerate rapidly from the
mid-1970s. These goods not only comprise the majority of manufactured exports from the first-tier
NIEs but have also gained substantial shares in world markets (UNCTAD, 1996).

The importation of sophisticated products may enlarge the domestic demand for these products to

the point where it becomes eventually profitable for domestic firms to start producing them;22 imports

may also play a vital role in forcing domestic producers to upgrade. On the export side, trade can

stimulate economic growth by forcing domestic firms to develop the technical skills and organizational

capacities required to conquer and retain foreign markets. Such dynamic aspects of trade are mentioned

in passing by most trade theorists, but are rarely given the serious attention they deserve in a literature

which is dominated by the traditional concerns of static trade theory.

A second approach to examining the links between openness and convergence looks more closely

at the timing of liberalization episodes. Ben-David (1993) has examined the convergence experience
of three regional groupings from this perspective: the original EEC countries, the EFTA countries, and

North America. Ben-David concedes that in the case of EFTA income convergence preceded trade

integration, but for the other groupings he argues that trade integration was a major causal factor
promoting income convergence. In the case of Canada and the United States, there is some evidence

for this claim: between 1964 and 1973 there was an explosion of trade between Canada and the United

States, while at the same time growth in Canada accelerated and the per capita income gap between the

two countries narrowed significantly. In the case of the EEC, however, Ben-David’s claim is

contradicted by his own data. These data show that the dispersion of per capita incomes among the

original EEC members fell dramatically over the period 1951-1963, and more gradually thereafter. Yet

the process of reducing tariffs and eliminating quotas only began in 1958 and took some years to
complete, so that tariffs were still quite high in 1963.23 Moreover, most of the spectacular expansion

in intra-EEC trade took place after 1963, when post-war income convergence was almost complete.
Imports from within the EEC rose from 4 per cent of the GDP of EEC countries in 1951 to only 6 per

cent in 1963. It is difficult to believe that such a modest increase in trade, mostly concentrated towards
the end of the period, could have been a major factor behind the impressive convergence of incomes
which occurred between 1951 and 1963. A more convincing argument is that rapid growth and income
convergence throughout the EEC region led to a mood of optimism and created the political conditions

- 25 -

required to dismantle trade barriers, which in turn led to a rapid expansion of trade. This rapid growth

in trade may have consolidated and accelerated economic growth in the EEC. It may have also

stimulated further income convergence. However, it was not the main driving force behind post-war

income convergence, most of which occurred when trade between the EEC countries was still fairly

modest. In all these countries, a very rapid pace of capital accumulation, particularly in manufacturing,

beginning in the early 1950s provided the well-spring of the convergence process - a process which was
reinforced by closer integration linked to the growth of intra-industry trade.

Ben-David (1995) extends his earlier argument to cover a much wider range of countries. For each
of 25 middle- and high-income countries, he estimates how rapidly its per capita income has converged

with the average per capita income of its trading partners. He then compares this estimate with the rate

of convergence with an equal-sized group of randomly chosen countries. In a number of cases, it turns

out that convergence is significantly faster with trading partners than with a randomly chosen group of

countries. This result seems to be fairly robust. However, there is one qualification: the countries for

which this finding holds were all economically quite advanced in 1960, when the period covered by the
study began, and they were mainly European. The only non-Europeans were Australia, Canada and

Japan. In the case of Canada, Ben-David’s analysis suggests that his convergence results refer mainly

to its catch-up with the United States, for which independent evidence suggests that trade did play a

significant role. Only five non-OECD countries are covered by his study - Argentina, Chile, Mexico,

South Africa and Uruguay. It is striking that for none of these countries is there evidence of significant

convergence towards the income level of its main trading partners. This suggests that the relationship

between convergence and trade is valid only for relatively advanced countries. Moreover, as Ben-David

concedes, his evidence does not establish the direction of causality, and is consistent with the view that

income convergence stimulates trade rather than vice versa.

A final approach is to look directly at factor price movements to assess whether these have moved

in a manner consistent with the predictions of orthodox trade theory. The body of accumulated evidence

across developed and developing countries does not show any long-run wage convergence trend across

the world economy (Leamer and Levinsohn, 1995). Considering samples of countries at similar levels
of development, a number of studies focusing on the European Union have found some support for

cross-country wage convergence (Gremmen, 1985; Van Nourik, 1987). However, the evidence is more
mixed for the OECD countries. From 1961 to 1984, rising international trade flows among OECD
countries was associated with wage convergence. After 1984, no clear convergence pattern is observed,

and the average real wages of production workers in a number of industrialized countries, including the

United States and United Kingdom, diverged from the highest wage countries such as Germany. A

comparison of wage trends in eight advanced and five middle-income developing countries across

similar industries shows a statistically significant convergence in this sample of industry-relative wages

- 26 -

24 A recent study of manufacturing real wage growth in 32 developing countries for 1973-1990 confirms that the impact
of trade on wages cannot be delinked from investment and productivity performance, and that greater openness to trade in this
group of countries in the 1980s did not coincide with stronger wage growth (Paus and Robinson, 1997). This is also the
conclusion reached for the OECD countries by Dollar and Wolff (1993).

towards the worldwide mean owing to greater international trade. However, convergence is modest and

limited to high-income economies (Davies, 1992).
Over the past two decades most developing countries have seen wage gaps with the North rise,

including in some cases from already very low wages either because real wages have declined in

absolute terms (as in Latin America and much of Africa in the 1980s) or because they have lagged the
wage growth in the advanced countries (as has been the case in much of South Asia). The exception
is the first-tier of East Asian NIEs, where manufacturing wages have converged on those of the North

at a quite rapid pace. These widening wage gaps in manufacturing both between developed and

developing countries and also among developing ones have coincided with a period of greater openness

in the developing world.

Table 8 shows changes in the wage gap between various developing countries and the United

States for a number of low, medium and high skill industries between 1980 and the early 1990s. In all
these countries (except Kenya) the share of exports and imports in GDP has risen quite strongly over
this period. However, in most countries, and across all skill levels, there has been a strong decline in

wages in developing countries relative to those in the United States. The exception to this trend has been

in the Asian NIEs, particularly those in the first-tier such as the Republic of Korea, where wages across

all skill levels have converged on the United States.

It is difficult to explain these wage trends simply in terms of differences in factor endowments,

including differences in educational attainment. Rather, for most developing countries, strong growth

in wages depends upon industrial expansion and upgrading bringing higher levels of employment and

rapid productivity growth both to the economy as a whole as well as to specific industrial sectors.24 As

table 8 shows, those countries in East Asia which exhibited strong wage convergence also exhibited

strong productivity convergence. But this has been attained in the context of a managed investment-

export nexus, rather than as a result of spontaneous global market forces.

- 27 -

Table 8

Relative wage and productivity trends in selected countries
(Relative to US level)

Early Early
1980 1990s 1980 1990s

Textiles Clothing

A. Wages (annual wage per employee)
Germany 1.24 1.32 1.31 1.45
Mexico 0.43 0.21 0.49 0.24
Chile 0.37 0.23 0.45 0.26
Colombia 0.23 0.11 0.19 0.1
Hong Kong 0.38 0.57 0.52 0.64
Korea, Rep. of 0.21 0.51 0.23 0.61
Malaysia 0.1 0.17 0.07 0.19
Indonesia 0.04 0.04 0.06 0.05
Turkey 0.21 0.33 0.24 0.24
India 0.08 0.05 0.08 0.05
Kenya 0.13 0.03 0.14 0.03
Morocco 0.24 0.13 0.14 0.14

B. Productivity (value added per employee)
Germany 0.94 1.19 1.26 1.33
Mexico 0.43 0.25 0.84 0.22
Chile 0.57 0.32 0.72 0.34
Colombia 0.48 0.27 0.29 0.13
Hong Kong 0.35 0.43 0.42 0.41
Korea, Rep. of 0.29 0.63 0.28 0.65
Malaysia 0.16 0.2 0.12 0.14
Indonesia 0.08 0.07 0.05 0.07
Turkey 0.4 0.4 0.34 0.41
India 0.07 0.03 0.07 0.077
Kenya 0.14 0.02 0.13 0.03
Morocco 0.2 0.11 0.1 0.09


Transport equipment Printing & publishing


A. Wages (annual wage per employee)
Germany 0.97 1.08 1.3 1.32
Mexico 0.37 0.17 0.38 0.21
Chile 0.25 0.15 0.5 0.35
Colombia 0.13 0.07 0.16 0.09
Hong Kong 0.29 0.45 0.3 0.44
Korea, Rep. of 0.18 0.45 0.28 0.52
Malaysia 0.1 0.13 0.12 0.19
Indonesia 0.07 0.03 0.07 0.04
Turkey 0.26 0.27 0.28 0.33
India 0.06 0.04 0.07 0.05
Kenya 0.07 0.03 0.24 0.04
Morocco 0.3 0.14 0.32 ..

- 28 -

Table 8 (continued)

Early Early
1980 1990s 1980 1990s

Transport equipment Printing & publishing


B. Productivity (value added per employee)
Germany 0.86 0.83 0.95 0.82
Mexico 0.5 0.35 0.4 0.22
Chile 0.55 0.2 0.84 0.44
Colombia 0.25 0.17 0.26 0.2
Hong Kong 0.23 0.29 0.28 0.33
Korea, Rep. of 0.24 0.54 0.29 0.57
Malaysia 0.13 0.19 0.16 0.2
Indonesia 0.17 0.16 0.07 0.08
Turkey 0.27 0.33 0.27 0.46
India 0.05 0.03 0.05 0.03
Kenya 0.05 0.01 0.15 0.03
Morocco 0.23 0.13 0.19 ..

Source: UNIDO (1996)
Note: Figures for 1980 are from 1975.

B. Capital mobility and convergence

While much of the analysis of greater capital mobility has been handled within international trade

theory, its impact on growth and convergence is clearly much broader and more direct. Indeed, capital

mobility is now seen as playing the leading role in many accounts of the contemporary globalization

process. A rapid pace of financial liberalization has been advocated on the grounds that financial

repression, typified by nominal interest rates below the rate of inflation, not only leads to inefficiency

in the allocation of resources and discourages savings, but also redistributes wealth at the expense of

savers. Free capital movements should allow savings to be pooled and allocated globally, improving

the international allocation of resources and equalizing rates of return on capital (adjusting, of course,
for differences in risk), as capital moves from low-return locations in the North to high-returnlocations
in capital-scarce developing countries. Simultaneously with these gains, capital mobility, particularly

in respect of direct investments, should further accelerate growth through higher rates of accumulation

and the transfer of technology and organizational skills.

- 29 -

25 See Felix (1996), Kregel (1996), Akyüz (1995), and UNCTAD (1991) for more detailed analysis of these links.

However, like trade, the impact of capital mobility is more complex than suggested by conventional

analysis. Global financial deepening has occurred, while the link between financial flows and foreign

investment has considerably weakened. These financial flows are rarely associated with the flows of

real resources - i.e. capital equipment embodying best-practice production techniques and other resource

inputs seeking the highest available rates of return in the production of goods and services. Rather, they

are primarily related to the purchase and sale in secondary markets of liabilities created for the financing

of already existing real assets. A large proportion of these flows consists of liquid capital attracted by

short-term arbitrage margins and prospects of speculative capital gain, rather than by long-term yields

on productive investment. They are extremely volatile and subject to bandwagon effects, capable of
generating gyrations in security prices, exchange rates and trade balances. They make little contribution

to the international allocation of savings or diffusion of technology and hence to a reduction in

international disparities in per capita income.25 Indeed, the combination of financial and trade

liberalization can very easily upset the domestic accumulation dynamic by shifting incentives towards

the non-tradable goods sector and placing a premium on more liquid but less productive assets

(UNCTAD 1997, ch. 4).
In the light of the uncertain impact of short-term capital flows, many globalization enthusiasts have

singled out FDI as the driving force behind economic convergence (IMF, 1997; World Bank, 1996).
Unlike other capital flows, FDI can be a less volatile source of financing for the accumulation process

and can have more direct links to economic growth than other cross-border flows, because it facilitates

the transfer of technology and generates spillovers into other sectors. However, there are a number of

aspects of the current globalization experience that suggests that the impact of FDI on growth and

convergence requires careful assessment before designing policies that aim at maximizing the benefits

from hosting transnational corporations.

The economic impact of recorded increases in FDI flows is likely to differ in significant ways as

between greenfield FDI, which involves a firm constructing a new production facility abroad financed

by capital raised in the home country, the acquisition of a controlling interest in an already existing firm,

which seems to correspond more to shorter-term capital flows and FDI in the form of retained earnings

(Kregel, 1996). Moreover, many of the changes in global financial markets that have facilitated capital
mobility and the increase in FDI flows have also made it more difficult to evaluate their stability.

Evidence suggests that even when FDI is governed by long-term considerations, such as real rates of

return and securing market shares, aggregate FDI flows can respond rapidly to changes in short-term

economic conditions. This is particularly true for non-repatriated earnings on existing stocks of FDI,

which have constituted in recent years a more important source of asset acquisition abroad by United

- 30 -

26 The recent financial crisis in South-East Asia did not discriminate between countries with a greater or smaller reliance
on FDI flows.

27 Blomstrom et al. (1994) do find such a causal link. On the other hand, a study by Dutt (1997) finds no evidence of

28 See Borensztein et al. (1995), where the threshold is linked to the level of human capital.

States and United Kingdom firms than capital outflows from those countries (Claessens et al., 1993,
p. 22).26

Any examination of the impact of FDI on economic growth and convergence must also bear in

mind a number of other considerations. Although evidence suggests a positive association between FDI

and growth, it is difficult to determine causation.27 A realistic interpretation of the evidence suggests that

there is a threshold level of income which needs to be crossed before FDI can make a significant

contribution to overall growth performance.28 Such a conclusion is also broadly consistent with the

finding that technology and other spillovers from TNCs become significant only when there is already

in place an appropriate level of local capabilities (Lall, 1995; Aitken et al., 1996). Thus, to a very large
extent FDI is attracted to economies with a proven growth record, be it to seek markets or cost

advantages, in which case it is part of a virtuous growth circle (Kozul-Wright and Rayment, 1997).
Much of the developing world remains outside the expanding universe of international production

because even when cost advantages are present, they alone do not offset low productivity levels or the

absence of productive assets needed to complement firm-specific plant and equipment. Furthermore,

risks associated with investment tend to be inversely related to the stage of development reached. While

incurring these risks may be justified by the expectation of monopolistic profits associated with the
exploitation of a particular natural resource in a poor developing country, there will be no such

exceptional or monopolistic elements in manufacturing or services. Any attempt in these countries to

improve cost advantages by easing their entry conditions for FDI, or by reducing wages further, will

almost certainly fail to offset other disadvantages which cause the risk premiums to be high (Kregel,


This paper has surveyed some of the issues which have arisen during the recent debate on

globalization and in particular those associated with the revival of neo-classical growth and trade theory.

The central claim of a strong globalization thesis has been that faster growth under open trade regimes

offers favourable prospects for developing countries leading to income convergence across the global

economy. The key to this analysis has been the beneficial role of increased capital mobility.

- 31 -

The issues are important and the debate has been interesting, if only because so much quantitative

evidence is now available, given the great improvement in international statistics over the past decade.

However, many of the conceptual issues have been long familiar to development economists and

surprisingly little that is really new has emerged. More importantly, the empirical evidence in support

of this strong globalization thesis is very weak and the domestic determinants of economic growth

remain significant.

The broad policy conclusion from the above analysis is thus a simple one. In a more open and

integrated world economy, both the quantity and the quality of investment are increasingly influenced

by external factors. However, the forces driving capital accumulation retain strong domestic roots and

remain open to the influence of various types of policy initiative. The successful economies of East

Asia, far from illustrating the virtues of rapid liberalization and unfettered global market forces, confirm

the complexities of policy-making in an interdependent world.

- 32 -


ABRAMOVITZ, M. (1986), "Catching up, forging ahead, and falling behind", Journal of Economic History,
Vol. 66.

AITKEN, B. et al., (1996),"Wages and foreign ownership: A comparative study of Mexico, Venezuela and the
United States", Journal of International Economics, Vol. 40.

AKYÜZ, Y. (1995), "Taming international finance", in J. Michie and G. Grieve Smith (eds.), Managing the Global
Economy (Oxford: Oxford University Press).

AKYÜZ, Y. et al. (1998), "New perspectives on East Asian development", Journal of Development Studies,

BAIROCH, P. (1993), Economics and World History: Myths and Paradoxes (London: Harvester).

BAIROCH, P. and R. KOZUL-WRIGHT (1997), "Globalization myths: Some historical reflections on integration,
industrialization and growth in the world economy", in R. Kozul-Wright and R. Rowthorn (eds.),
Transnational Corporations and the Global Economy (London: Macmillan).

BARRO, R.J. and X. SALA-I-MARTIN (1995), Economic Growth (New York: McGraw-Hill).

BAUMOL, W.J. (1986), "Productivity growth, convergence and welfare: What the long-run data show", American
Economic Review, Vol. 76, pp. 1073-1085.

BAUMOL, W.J., S.A.B. BLACKMAN and E.N. WOLFF (1989), Productivity and American Leadership: The
Long View (Cambridge, MA: MIT Press).

BEN-DAVID, D. (1993), "Equalizing exchange: Trade liberalization and income convergence", Quarterly Journal
of Economics, Vol. 108, pp. 653-79.

(1995), "Trade and convergence among countries", CEPR Discussion Paper, No. 1126, February.

BERNARD, A. and C. JONES (1996), "Technology and convergence", The Economic Journal, Vol. 106,
pp. 1037-1044.

BHAGWATI, J. (1994), "Free trade: Old and new challenges", The Economic Journal, Vol. 104, March,
pp. 231-246.

BIS (various years), Annual Report (Basle, Switzerland: Bank for International Settlements).

BLISS, C. (1997), "Galton’s fallacy and economic convergence", mimeo (Oxford: Nuffield College).

BLOMSTROM, M. et al. (1994), "What explains developing country growth?", NBER Working Paper, No. 4132
(Cambridge, MA: National Bureau of Economic Research).

BORENSZTEIN, E. et al. (1995), "How does foreign direct investment affect economic growth?", NBER Working
Paper, No. 5057 (Cambridge, MA: National Bureau of Economic Research).

CABLE, V. (1995), "What future for the state?", Daedalus, Spring.

CLAESSENS, S., M. Dooley and A. WARNER (1993), "Portfolio capital flows: Hot or cool?", World Bank
Discussion Paper, No. 228 (Washington, DC: The World Bank).

- 33 -

COLLIER, P. and W. GUNNIN (1997), "Explaining African economic performance", mimeo (Oxford: UK Centre
for the Study of African Economies, University of Oxford).

DAVIES, S. (1992), "Cross-country patterns of change in relative wages", NBER Annual Macroeconomic Review.

DE LONG, J.B. (1988), "Productivity growth, convergence and welfare: Comment", American Economic Review,
Vol. 78, pp. 1138-54.

DOLLAR, D. and E. WOLFF (1993), Competitiveness, Convergence and International Specialization
(Cambridge, MA: MIT Press).

DUTT, A. (1997) "Direct foreign investment, transnational corporations and growth: Some empirical evidence and
a North-South model", in R. Kozul-Wright and R. Rowthorn (eds.), Transnational Corporations and the
Global Economy (London: Macmillan).

FELIX, D. (1996), "Financial globalization versus free trade: The case for the Tobin tax", UNCTAD Review, Sales
No. E.97.II.D.2 (New York and Geneva: United Nations).

FINDLAY, R. (1996), "Modelling global interdependence: Centers, peripheries and frontiers", American
Economic Review: Papers and Proceedings, Vol. 86, No. 2 (May), pp. 47-51.

FREEMAN, C. (1993), "Interdependence of technological change with growth of trade and GNP", in M. Nissanke
and A. Hewitt (eds.), Economic Crisis in Developing Countries (London: Pinter).

FRIEDMAN, M. (1992), "Do old fallacies ever die?", Journal of Economic Literature, Vol. XXX, December,
pp. 2129-32.

GERSCHENKRON, A. (1962), Economic Backwardness in Historical Perspective (Cambridge MA: Harvard
University Press).

GREIDER, W. (1997), One World Ready or Not: The Manic Logic of Global Capitalism (New York: Simon and

GREMMEN, H. (1985), "Testing the factor price equalization theorem in the EC: An alternative approach",
Journal of Common Market Studies, Vol. 24.

HALL, R. and C. JONES (1997), "Levels of economic activity across countries", American Economic Review,

HENTSCHEL, J. (1992), "Productivity growth and structural change in industry and trade in five South-East
Asian countries", Technological Dynamism Project (Geneva: United Nations, UNCTAD Technology

HIRSCHMAN, A. (1995), A Propensity to Self-Subversion (Cambridge, MA: Harvard University Press).

IMF (various years), International Financial Statistics Yearbook (Washington, DC: International Monetary Fund).

(various years), Balance of Payment Statistics Yearbook (Washington, DC: IMF).

(1997), World Economic Outlook (Washington, DC: IMF).

KALDOR, N. (1981), "The role of increasing returns, technical progress and cumulative causation in the theory
of international trade and economic growth", Economie Appliquée, No. 4.

KORHONEN, P. (1994), "The Theory of the Flying Geese Pattern of Development and its Interpretation", Journal
of Peace Research, 31, 1. 93-108.

- 34 -

KOZUL-WRIGHT, R. and P. RAYMENT (1997), "The institutional hiatus in transition economies and its policy
consequences", Cambridge Journal of Economics, Vol. 21.

KREGEL, J. (1994), "Capital flows: Globalization of production and financing development", UNCTAD Review,
Sales No. E.94.II.D.19 (New York and Geneva: United Nations).

(1996), "Some risks and implications of financial globalization for national policy autonomy", UNCTAD
Review, Sales No. E.97.II.D.2 (New York and Geneva: United Nations).

KRUGMAN, P. (1995) "Growing world Trade: Causes and consequences", Brookings Papers on Economic
Activity, 1.

LALL, S. (1995), "Industrial strategy and policies on foreign direct investment in East Asia", Transnational
Corporations, Vol. 4, No. 3 (December).

LAWRENCE, R. (1993), "Futures for the world trading system and their implications for developing countries",
in M. Agosin and D. Tussie (eds.), Trade and Growth: New Dilemmas in Trade Policy (London: Macmillan).

LEAMER, E. and J. LEVINSOHN (1995), "International trade theory: The evidence", in G. Grossman and
K. Rogoff (eds.), Handbook in International Economics, Vol. III (Amsterdam, Elsevier).

LEVINE, D. and D. RENELT (1992), "A Sensitivity Analysis of Cross-Country Growth Regressions", American
Economic Review, Vol. 82, No. 4 (September), pp. 942-963.

MADDISON, A. (1982), Phases of Capitalist Development (New York: Oxford University Press).

MANKIW, G. (1995), "The growth of nations", Brookings Papers on Economic Activity, Vol. 1.

O’LEARY, E. (1995), Productivity Convergence: A Study of European Union Countries at the Aggregate,
Sectoral and Industry Levels, 1960 to 1990, Ph.D. dissertation, (Dublin: National University of Ireland).

PANIC, M. (1988), The National Management of the International Economy (London: Macmillan).

PAUS, E. and M. ROBINSON (1997), "The implications of increasing openness for real wages in developing
countries, 1973-90", World Development, Vol. 25, No. 4.

PFFAF, W. (1997), "Don’t forget globalization creates losers, too", International Herald Tribune, 13 November.

POLANYI, M. (1944), The Great Transformation (Boston: Beacon Press).

PRITCHETT, L. (1995), "Divergence, big time", Policy Research Working Paper, No. 1522 (Washington, DC:
World Bank).

PRITCHETT, L. (1996), "Measuring outward orientation in LDCs: Can it be done?", Journal of Development
Economics, Vol. 49, pp. 307-335.

QUAH, D. (1993), "Galton’s fallacy and tests of the convergence hypothesis", Scandinavian Journal of
Economics, Vol. 95, No. 4 (December), pp. 427-443.

(1996a), "Twin peaks: Growth and convergence in models of distribution dynamics", Economic Journal,
Vol. 106, July, pp. 1045-55.

(1996b), "Empirics for economic growth and convergence", European Economic Review, Vol. 40, No 6
(June), pp. 1353-1376.

RODRIK, D. (1997), Has Globalization Gone too Far? (Washington, DC: Institute for International Economics).

- 35 -

ROMER, P. (1995), "Comment on G. Mankiw: The growth of nations", Brookings Papers on Economic Activity,
Vol. 1.

ROWTHORN, R. (1992), "Productivity and American leadership", Review of Income and Wealth, Vol. 38,

SACHS, J. and A. WARNER (1995a), "Economic Reform and the Process of Global Integration", Brookings
Papers on Economic Activity, Vol. 1, pp. 1-118.

(1995b), "Natural Resource Abundance and Economic Growth, NBER Working Paper, No. 5398, December.

SALA-i-MARTIN, X. (1996), "The classical approach to convergence analysis", Economic Journal, Vol. 106,
July, pp. 1019-1043.

SUMMERS, R. and A. HESTON (1988), "A new set of international comparisons of real product and prices:
Estimates for 130 countries, 1950-1985", Review of Income and Wealth, March.

UNCTAD (various years), World Investment Report (New and Geneva: United Nations).

UNCTAD (1991), Trade and Development Report (New York and Geneva: United Nations).

(1992), Trade and Development Report (New York and Geneva: United Nations).

(1994), Trade and Development Report (New York and Geneva: United Nations).

(1996), Trade and Development Report (New York and Geneva: United Nations).

(1997), Trade and Development Report (New York and Geneva: United Nations).

UNIDO (1988), Handbook of Industrial Statistics (Paris: United Nations Industrial Development Organization).

(1996), International Yearbook of Statistics (Paris: UNIDO).

VAN NOURIK, A. (1987) "Testing the factor price equalization theorem in the EC: A comment", Journal of
Common Market Studies, Vol. 26.

WILLIAMSON, J.G. (1997), "Globalization and inequality: Past and present", The World Bank Research
Observer, Vol. 12, No. 2 (August), pp. 117-35.

WOOD, A. (1994), North-South Trade, Employment and Inequality: Changing Fortunes in a Skill-Driven World
(Oxford: Clarendon Press).

WORLD BANK (1996), Global Economic Prospects and the Developing Countries (Washington, DC: The World

- 36 -

* Out of stock.

UNCTAD Discussion Papers

No. 42, October 1991 Georg KELL & Technology and competitiveness in the textile industry

*No. 43, February 1992 Yilmaz AKYÜZ On financial deepening and efficiency
*No. 44, July 1992 Cem SOMEL Finance for growth: Lessons from Japan

No. 45, July 1992 Sebastian SCHICH Indebtedness, sovereign risk and the spread: The
example of Hungary and the Euromarkets

No. 46, July 1992 Raju J. SINGH An imperfect information approach to the structure
of the financial system

*No. 47, August 1992 V. BHASKAR Privatization and the developing countries: The
issues and the evidence

*No. 48, August 1992 Ha-Joon CHANG & Public enterprises in developing countries and
Ajit SINGH economic efficiency

*No. 49, October 1992 Ajit SINGH The stock-market and economic development:
Should developing countries encourage stock-

No. 50, November 1992 Ulrich HOFFMANN & Demand growth for industrial raw materials and its
Dusan ZIVKOVIC determinants: An analysis for the period 1965-1988

No. 51, December 1992 Sebastian SCHICH Indebtedness, creditworthiness and trade finance and
payment arrangements

No. 52, December 1992 Dwight H. PERKINS China's 'gradual' approach to market reforms

No. 53, December 1992 Patricio MELLER Latin American adjustment and economic reforms:
Issues and recent experience

No. 54, January 1993 Trevor GARDNER The present economic situation in Zambia and the
role of privatisation in improving its economy

No. 55, February 1993 Alexandre R. BARROS Prospects for world sugar trade

No. 56, March 1993 Yilmaz AKYÜZ Financial liberalization: The key issues
No. 57, April 1993 Alice H. AMSDEN Structural macroeconomic underpinnings of effec-

tive industrial policy: Fast growth in the 1980s in
five Asian countries

No. 58, April 1993 Celso ALMEIDA Development and transfer of environmentally sound
technologies in manufacturing: A survey

No. 59, May 1993 Ali-Reza NIKPAY Privatization in Eastern Europe: A survey of the
main issues

No. 60, July 1993 Jean-Marc FONTAINE Reforming public enterprises and the public sector
in sub-Saharan Africa

No. 61, July 1993 Korkut BORATAV Public sector, public intervention and economic

No. 62, July 1993 Roberto FRENKEL Growth and structural reform in Latin America:
Where we stand

No 63, July 1993 Machiko NISSANKE & Mobilization and allocation of domestic savings: A
Priya BASU study on Bhutan

- 37 -

* Out of stock.

No. 64, July 1993 Machiko NISSANKE & Mobilization and allocation of domestic savings: A
Priya BASU case study on Nepal

No. 65, August 1993 Ercan UYGUR Liberalization and economic performance in Turkey

No. 66, August 1993 Yilmaz AKYÜZ Maastricht and fiscal retrenchment in Europe
No. 67, September 1993 Cem SOMEL The State in economic activity: Problems of

economic policy-making

No. 68, September 1993 Andrew CORNFORD The role of the Basle Committee on Banking Super-
vision in the regulation of international banking

No. 69, September 1993 Sebastian SCHICH The level and volatility of external financial posi-
tions and the costs of export credit insurance

No. 70, October 1993 Veena JHA, Ecolabelling and international trade
Simonetta ZARRILLI

No. 71, October 1993 Adolfo CANITROT The exchange rate as an instrument of trade policy

No. 72, October 1993 Xiaoning J. ZHAN North American economic integration and its impli-
cations for the exports of China and Hong Kong

No. 73, November 1993 J.H. REICHMAN Implications of the Draft TRIPS Agreement for
developing countries as competitors in an integrated
world market

No. 74, November 1993 Priya BASU & Fiscal adjustment in the Gambia: A case study

No. 75, November 1993 William W.F. CHOA The relevance of market structure to technological
progress: A case study of the chemical industry

No. 76, December 1993 Ajit SINGH The Plan, the market and evolutionary economic
reform in China

No. 77, January 1994 Shigehisa KASAHARA A rescue plan for the post-bubble Japanese econ-
omy: The establishment of the Cooperative Credit
Purchasing Company

No. 78, January 1994 Jean K. THISEN The European Single Market and its possible effects
on African external trade

No. 79, February 1994 Kálmán KALOTAY & Emerging stock markets and the scope for regional
Ana María ALVAREZ cooperation

No. 80, February 1994 Edouard DOMMEN Développement durable: Mots-déclic
*No. 81, March 1994 Juan A. DE CASTRO The internalization of external environmental costs

and sustainable development


No. 83, May 1994 Yilmaz AKYÜZ & Regimes for international capital movements and some
Andrew CORNFORD proposals for reform

No. 84, May 1994 David FELIX Industrial development in East Asia: What are the
lessons for Latin America?

No. 85, July 1994 S.M. SHAFAEDDIN The impact of trade liberalization on export and
GDP growth in least developed countries

No. 86, July 1994 Raju J. SINGH Bank credit, small firms and the design of a finan-
cial system for Eastern Europe

- 38 -

No. 87, July 1994 Thomas ZIESEMER Economic development and endogenous terms-of-
trade determination: Review and reinterpretation of
the Presbisch-Singer Thesis

No. 88, August 1994 Sebastian SCHICH The payment arrangements in the trade of CEECs
and LDCs between 1986 and 1994

No. 89, September 1994 Veena JHA & Are environmentally sound technologies the
Ana Paola TEIXEIRA Emperor's new clothes?

No. 90, October 1994 Manuel R. AGOSIN Saving and investment in Latin America

No. 91, October 1994 Yilmaz AKYÜZ & The investment-profits nexus in East Asian
Charles GORE industrialization

No. 92, November 1994 Charles GORE Development strategy in East Asian newly indus-
trializing economies: The experience of post-war
Japan, 1953-1973

No. 93, December 1994 J. F. OUTREVILLE Life insurance in developing countries: A cross-
country analysis

No. 94, January 1995 XIE Ping Financial services in China

No. 95, January 1995 William W.F. CHOA The derivation of trade matrices by commodity
groups in current and constant prices

No. 96, February 1995 Alexandre R. BARROS The role of wage stickiness in economic growth

No. 97, February 1995 Ajit SINGH How did East Asia grow so fast? Slow progress
towards an analytical consensus

No. 98, April 1995 Z. KOZUL-WRIGHT The role of the firm in the innovation process

No. 99, May 1995 Juan A. DE CASTRO Trade and labour standards: Using the wrong
instruments for the right cause

No. 100, August 1995 Roberto FRENKEL Macroeconomic sustainability and development
prospects: Latin American performance in the 1990s

No. 101, August 1995 R. KOZUL-WRIGHT Walking on two legs: Strengthening democracy and
& Paul RAYMENT productive entrepreneurship in the transition econ-


No. 102, August 1995 J.C. DE SOUZA BRAGA Financing the public sector in Latin America
& Sulamis DAIN

No. 103, September 1995 Toni HANIOTIS & Should governments subsidize exports through export
Sebastian SCHICH credit insurance agencies?

No. 104, September 1995 Robert ROWTHORN A simulation model of North-South trade

No. 105, October 1995 Giovanni N. DE VITO Market distortions and competition: the particular
case of Malaysia

No. 106, October 1995 John EATWELL Disguised unemployment: The G7 experience

No. 107, November 1995 Luisa E. SABATER Multilateral debt of least developed countries

No. 108, November 1995 David FELIX Financial globalization versus free trade: The case
for the Tobin tax

No. 109, December 1995 Urvashi ZUTSHI Aspects of the final outcome of the negotiations on
financial services of the Uruguay Round

No. 110, January 1996 H.A.C. PRASAD Bilateral terms of trade of selected countries from
the South with the North and the South

No. 111, January 1996 Charles GORE Methodological nationalism and the misunderstand-
ing of East Asian industrialization

- 39 -

No. 112, March 1996 Djidiack FAYE Aide publique au développement et dette extérieure:
Quelles mesures opportunes pour le financement du
secteur privé en Afrique?

No. 113, March 1996 Paul BAIROCH & Globalization myths: Some historical reflections on
Richard KOZUL-WRIGHT integration, industrialization and growth in the world


No. 114, April 1996 Rameshwar TANDON Japanese financial deregulation since 1984

No. 115, April 1996 E.V.K. FITZGERALD Intervention versus regulation: The role of the IMF
in crisis prevention and management

No. 116, June 1996 Jussi LANKOSKI Controlling agricultural nonpoint source pollution:
The case of mineral balances

No. 117, August 1996 José RIPOLL Domestic insurance markets in developing coun-
tries: Is there any life after GATS?

No. 118, September 1996 Sunanda SEN Growth centres in South East Asia in the era of

No. 119, September 1996 Leena ALANEN The impact of environmental cost internalization on
sectoral competitiveness: A new conceptual frame-

No. 120, October 1996 Sinan AL-SHABIBI Structural adjustment for the transition to disarma-
ment: An assessment of the role of the market

No. 121, October 1996 J.F. OUTREVILLE Reinsurance in developing countries: Market struc-
ture and comparative advantage

No. 122, December 1996 Jörg MAYER Implications of new trade and endogenous growth
theories for diversification policies of commodity-
dependent countries

No. 123, December 1996 L. RUTTEN & Collateralized commodity financing with special
L. SANTANA-BOADO reference to the use of warehouse receipts

No. 124, March 1997 Jörg MAYER Is having a rich natural-resource endowment detri-
mental to export diversification?

No. 125, April 1997 Brigitte BOCOUM The new mining legislation of Côte d'Ivoire: Some
comparative features

No. 126, April 1997 Jussi LANKOSKI Environmental effects of agricultural trade liberaliz-
ation and domestic agricultural policy reforms

No. 127, May 1997 Raju Jan SINGH Banks, growth and geography
No. 128, September 1997 Enrique COSIO-PASCAL Debt sustainability and social and human develop-


No. 129, September 1997 Andrew J. CORNFORD Selected features of financial sectors in Asia and
their implications for services trade

No. 130, February 1998 Matti VAINIO The effect of unclear property rights on environ-
mental degradation and increase in poverty


Copies of UNCTAD Discussion Papers and Reprint Series may be obtained from the Editorial Assistant,
Macroeconomic and Development Policies, GDS, UNCTAD, Palais des Nations, CH-1211 Geneva 10, Switzerland
(Tel. 41-22-907.5733; Fax 41-22-907.0274; E.mail: nicole.winch@unctad.org).