World Development Vol. 31, No. 4, pp.

667–683, 2003
Ó 2003 Elsevier Science Ltd. All rights reserved
Printed in Great Britain
www.elsevier.com/locate/worlddev 0305-750X/03/$ - see front matter
doi:10.1016/S0305-750X(03)00002-0

The Two Faces of Globalization: Against

Globalization as We Know It q
BRANKO MILANOVIC *
World Bank, Washington, DC, USA
Summary. — The paper shows that the current view of globalization as an automatic and benign
force is flawed: it focuses on only one, positive, face of globalization while entirely neglecting the
malignant one. The two key historical episodes that are adduced by the supporters of the
‘‘globalization as it is’’ (the Halcyon days of the 1870–1913, and the record of the last two decades
of development) are shown to be misinterpreted. The ‘‘Halcyon days’’ were never Halcyon for those
who were ‘‘globalized’’ through colonization since colonial constraints prevented them from
industrializing. The record of the last two decades (1978–98) is shown to be almost uniformly worse
than that of the previous two (1960–78).
Ó 2003 Elsevier Science Ltd. All rights reserved.

Key words — world, globalization, growth, development theory, convergence

1. THE MAINSTREAM VIEW only thing that a country needs to do is to open

up its borders, reduce tariff rates, attract for-
The mainstream view of globalization, at eign capital, and in a few generations if not less,
least among the people who ‘‘matter’’ in the the poor will become rich, the illiterate will
countries that ‘‘matter’’––the vast majority of learn how to read and write, and inequality will
economists, many political scientists, and po- vanish as the poor countries catch up with the
litical commentators––is that globalization is a rich. This is the view conveyed implicitly and
benign force leading us ultimately to the era of subliminally by many serious papers and pub-
converging world incomes (as poor countries lications as, for example, in the Dollar and
such as China open up to the world and see Kraay (2000) often-repeated statement that
their incomes rise), converging institutions as ‘‘the poor and the rich gain one-for-one from
democracy becomes a universal norm, and cul- openness,’’ 1 or in Sala-i-MartinÕs (2002) deri-
tural richness as people of different background sive statements about inequality and globaliza-
interact more frequently. The most famous, or tion. While, of course, the authors are careful
notorious, reflection of that Pollyannaish view enough not to explicitly make such statements
of the world was the early announcement by (e.g., Dollar & Kraay do acknowledge that
Fukuyama (1989) of the ‘‘end of history.’’ Al- gains ‘‘one-for-one’’ are expressed in percentage
though the ethnic warfare since then has not terms, so that a poor person whose income is
disproved FukuyamaÕs (or rather HegelÕs) view, one-hundredth of that of a rich person will also
since none of the ethnic warriors had an alter- gain one-hundredth of the rich personÕs gain), 2
native civilizational blueprint––a point which is
implied in HegelÕs hypothesis––the more recent
q
debates about globalization as well as the role The paper represents authorÕs own views only; the
of Islam––a society with an alternative blue- views should not be attributed to the World Bank or its
print––do show that the end of history is not affiliated organization.
around the corner. * I am thankful to Dennis Arens, Nancy Bidsall, David
It is only a slight caricaturization of this Dollar, Bill Easterly, James K. Galbraith, Carol Gra-
na€ıve view to state that its proponents regard ham, Elizabeth King, Ravi Kanbur, Mansoob Murshed,
globalization as a deus ex machina for many of Martin Ravallion, Dani Rodrik, and Michael Ward for
the problems, such as poverty, illiteracy or in- very valuable comments on an earlier draft of the paper.
equality that beset the developing world. The Final revision accepted: 24 November 2002.
667
668 WORLD DEVELOPMENT

they do leave their statement sufficiently am- ched above are two other views. One, the Left
biguous, thus allowing more explicit and wrong view, regards globalization as a malignant force
Pollyannaish views of globalization to find that leads to child labor in the South and takes
currency in the mainstream popular magazines away middle-class jobs in the North. For the
and newspapers. There the heavy guns of the Left, to be anti-globalization is a very difficult
globalization debate are not embarrassed by task since the Left is, by definition, internation-
the finer points of relative vs. absolute gains, or alist. But what the Left resents is that todayÕs
with percentages or logarithms: they simply globalization is led by a triumphant, and often,
state that globalization is good for everyone. unbridled capitalism. Unbridled capitalism does
For example, The Economist (2000, p. 82) in a produce the effects of which the Left complains:
review of the Dollar and Kraay article writes: destruction of environment, obliteration of in-
‘‘Growth really does help the poor: in fact, it digenous cultures (e.g., how many Mayas still
raises their incomes by about as much as speak Mayan?), and exploitation of the weak.
it raises incomes of everybody else.’’ This is The conservative, and often xenophobic,
deemed insufficient to carry the (misleading) Right also agrees that globalization is a ma-
message. In the next paragraph, they continue: lignant force. That view is more prevalent in
‘‘On average, incomes of the poor rise one-for- Europe, with its history of xenophobia, than in
one with incomes overall.’’ the United States. 3 In Europe, globalization
Moreover, the past too is harnessed to sup- engenders not only fear of losing jobs to the
port this dominant view of globalization. The poor masses in the South, but of losing cultural
period 1870–1913, the heyday of imperialism homogeneity that many European countries
and colonialism, is made to appear as the period have acquired through a long process of oblit-
of universal growth, and catch-up of poor coun- eration of local cultures (where are the French
tries as, for example, in Lindert and Williamson Bretons today?) and three centuries of capitalist
(2001, p. 1) who somewhat incredibly write: development. Their homogeneity is threatened,
‘‘globalization probably mitigated the steep rise moreover, by the people of different color,
in income gaps between nations. The nations culture, and way of life. Silvio BerlusconiÕs re-
that gained the most from globalization are cent quip about Islam, FallaciÕs (2002) diatribes
those poor ones that changed their policies to against Muslim immigrants, and HeiderÕs, Le
exploit it. . .’’ PenÕs and FortuynÕs political support is all part
Thus, globalization is regarded as a benign and parcel of the fear engendered by a more
and automatic force that, once certain precon- globalized society.
ditions are set in place (‘‘sound’’ macropolicies, Can these two views, the dominant one, and
protection of property rights etc.), will inexo- the critical too, be correct? Yes, they can be-
rably lead countries and individuals to a state cause globalization being such a huge and
of economic bliss. We show here that this view multifaceted process presents different faces to
of globalization is based on one serious meth- different people. Depending on where we live,
odological error: a systematic ignorance of the whether we are rich or poor, where we stand
double-sided nature of globalization, that is, ideologically, we are bound to see the process
systematic ignorance of its malignant side. We differently. But this is nothing new. Globaliza-
show, first, how this methodological error leads tion as it played out from the mid-19th century
to the misreading of the 19th century economic to 1914 was also a contradictory force, with
history; second, we argue that the Pollyannaish both its benign and malignant features. Thus,
view of globalization severely distorts the les- we believe, today too, as in the past, global-
sons of the most recent period, 1980–2000, and ization has two faces: the benign one, based on
third, we show how a more accurate and real- voluntary exchanges and free circulation of
istic reading of globalization requires, in many people, capital, goods and ideas; and the other
respects, different policies from the ones sug- face, based on coercion and brute force.
gested by the na€ıve (or self-interested?) global-
ization cheerleaders.
3. BY RAILROADS AND GUNSHIPS

2. THE TWO FACES OF GLOBALIZATION These two faces have been very clearly in
evidence during the previous period of global-
In contrast to the view of globalization as a ization a century ago. On the one hand, there
purely benign force which we have briefly sket- was a manifold increase in output and trade
 TWO FACES OF GLOBALIZATION 669

between Western European countries and their 10 million people, are only the worst excesses
overseas offshoots (the United States, Canada, (see Hochschild, 1998).
Australia, and New Zealand); there were mil- Economists who deal in models of individual
lions of Italian, Polish, or Irish migrants who rational behavior are not well equipped to treat
traversed the Atlantic in search of a better life conquests and plunder. Thus, they prefer to
(and found it), bringing moreover a wage and stick to the ‘‘nice’’ face of globalization, to
income convergence between Europe and the describe how the global working of the ‘‘invis-
United States by putting a downward pressure ible hand’’ brought late 19th century techno-
on wages in the United States, and allowing logical marvels. It is, for example, remarkable
European wages to go up (OÕRourke & Wil- that in an influential article on the 19th century
liamson, 1999). Telegraph cables and railroads globalization by two distinguished economists,
were built to bring the world closer and to ac- Jeffrey Williamson from Harvard and Peter
celerate the transfer of goods. In Cuba, the Lindert from University of California (2001)
main producer of sugar, railroads were built never once were the words ‘‘colonialism,’’
before any existed in Italy or Holland (Bairoch, ‘‘colony,’’ ‘‘slavery,’’ or ‘‘colonization’’ men-
1997, vol. 2, p. 574). Foreign capital flowed tioned. This omission is all the more interesting
from the capital-rich England and France to because 1870–1913 (or 1820 which they also
the lands capital-poor, yet rich in opportuni- choose as the beginning year) was not only the
ties, such as Argentina and Russia. In KeynesÕ epoch par excellence of colonialism, but of
(1998, 1918, pp. 11–12) famous phrase, wist- slavery too. Just pro memoria, in the British
fully regretting the passing away of a world that colonies, slavery was banned in 1833; in the
was destroyed by the Great War, a Londoner French colonies, after the 1848 revolution; in
the Dutch colonies, it continued until 1863; in
could secure. . . cheap and comfortable means of the (Southern) United States, it was abolished
transport to any country or climate without passport in 1865, while in Brazil, it went on until 1878. It
or other formality, could dispatch his servant (sic!) is thus, to say the least, very odd to ignore the
to the neighboring office of a bank for such supply
of the precious metals as might seem convenient,
existence of slavery when talking about glo-
and could then proceed abroad to foreign quarters, balization in the 19th century.
without knowledge of their religion, language, or cus- From this ‘‘clean-shaven’’ world of voluntary
toms. . . exchanges, the unpleasant facts of slavery and
conquest are simply banished. So, when we
While to the KeynesÕ Londoners, globaliza- reflect on what globalization then brought to
tion indeed presented that clean, friendly face, those who were enslaved, and to those who
was the same true for the others? Not really. could ‘‘send their servant to the neighboring
Globalization was brought to the many at the office of a bank’’ in London, are we surprised
‘‘point of a gun,’’ and many were ‘‘globalized’’ that people today might also have similarly
literally kicking and screaming, from Commo- divided views about globalization?
doreÕs Perry ultimatum which opened Japan,
to British and French gunboat diplomacy in
Tunisia, Egypt and Zanzibar, to the Opium 4. INCOME DIVERGENCE DURING
wars and gunboats that patrolled Chinese inter- THE 19TH CENTURY
nal waterways. Worst of all, for many millions
who were sold in slavery, or who toiled 16 h a The dominant, economistsÕ view, of the 19th
day on plantations from Malaya to Brazil that century globalization is indeed based on what
too was globalization. Globalization was not Williamson calls ‘‘the Atlantic economy,’’ that
merely accompanied by the worst excesses of is, the exchange of goods, migration, and cap-
colonialism; colonialism was not an accident. ital flows between Western Europe, and
On the contrary, globalization was colonialism Northern America (where Argentina and Uru-
because it is through being colonies that most guay too make a few cameo appearances). As
of the non-European countries were brought to already mentioned, economists are well-placed
the global world. The Dutch East Indies com- to deal with this benign face of globalization
pany that, according to conservative estimates because their key methodological construct is a
by Maddison (2001, p. 87) pillaged during self-interested individual, and when there is no
1868–1930 between 7.4% and 10.3% of Indo- external coercion (slavery or gunboats), econ-
nesiaÕs national income per year, 4 and the omists can best study how individuals, follow-
genocide in Congo that might have killed up to ing their own interests, bring out economic
670 WORLD DEVELOPMENT

changes that our textbooks tell us should hap- participating countries’’ (p. 18). But let us see if
pen. that was really so.
The problem with this approach is twofold. We have three sources of data on incomes
First, it applies only to a limited part of the (GDP per capita) for the period stretching from
world. Colonialism, pillage, and slavery were the early 19th century to 1913. They are pro-
no less part of globalization than the voluntary duced by Maddison (1995, 2001); Bairoch
movement of Irish peasants to the United (1997) and Prados de la Escosura (2000). The
States, or the voluntary transfer of British countries we want to include––and they are
funds to Argentina. So, if we want to discuss mostly the only ones for which the data are
the North Atlantic economy alone, the Wil- available––are those that were all part of the
liamson–Lindert approach is fine: they can af- broader Atlantic economy, the key participants
ford to ignore the rest of the world. But, if we in globalization. These are the rich WENAO
want to use the parable of the North Atlantic (Western Europe, North America, and Ocea-
economy to argue that this is what globaliza- nia) countries. Their number varies between 18
tion is, then it is wrong because it is only one, and 20 in BairochÕs series, 13 (but only in 1850)
and possibly a less important, facet of global- and 21 in Prados de la EscosuraÕs series, and 19
ization. in MaddisonÕs data. 5 Consequently, the coun-
Second, and more importantly, we have to try coverage is fairly standard and constant.
look more carefully at the claim that global- We look at whether there was convergence or
ization brings convergence of income among the not of mean incomes (GDPs per capita) by
participating countries with poor countries calculating Gini coefficient across GDPs per
growing faster and presumably catching up with capita of these countries, with each country
the rich. This is an important tenet in the my- being given the same weight. 6 If there is con-
thology of benign globalization because it is vergence, the Gini coefficient should go down.
supposed to show the benefits of globalization Yet as Figure 1 shows, the story is not at all
reaped by the poor countries. (Notice that the that simple.
proper unit of analysis here is country. We are According to Bairoch, during the peak pe-
not concerned with whether globalization makes riod of globalization 1870–1913, incomes be-
the world more equal or not, in which case we tween the rich countries continued to diverge:
would need to calculate inequality across world the Gini coefficient of their GDPs per capita
citizens, as for example done by Bourguignon increased by five Gini points, or by almost a
and Morrisson (2000). Here, we are simply third, rising from 15.8 in 1870 to 20.9 in 1913.
concerned with the so-called theory of conver- According to Maddison, inequality is about the
gence––namely that the poor countries, when same at the beginning and at the end of the
they open up, grow faster than the rich.) Indeed period. Both Bairoch and Maddison use GDPs
Lindert and Williamson (2001) make that con- per capita expressed in PPP (purchasing power
clusion by showing the wage convergence be- parity) terms anchored, respectively, in US
tween the densely-populated Western European 1960 prices and 1990 international prices
economies and the sparsely populated (and re- (Geary–Kramis dollars). Prados de la Escosura
source-rich) ‘‘New World.’’ As people migrate uses current PPP exchange rates––which means
from Western Europe to the United States or that his GDPs per capita are not comparable
Argentina, wages and income per capita con- across time––to derive the rankings of about
verge. Apodictically, Lindert and Williamson the same set of countries over 1850–1938. Only
write (2001, p. 13): ‘‘Real wages and living his data show an income convergence among
standards converged among the currently- the rich countries starting in 1860 and ending on
industrialized countries between 1850 and World the eve of the WW I. 7
War I.’’ They do accept that even as migration Thus, the evidence of income convergence
and trade contributed to wage equalization among the subset of rich globalization partici-
among the participants, capital flows which fa- pants which we were led to expect was the norm
vored the richer countries (that is, flowed to- during the previous globalization episode turns
ward the rich rather than toward the poorer out on a closer inspection to be far from
countries) were an ‘‘anti-convergence force’’ (p. watertight. We see that depending on the au-
17). Yet, on balance, their conclusion is that thor and on the PPP rates used, rich countries
‘‘prewar [World War I] globalization looks like show either a divergence, or stability, or con-
a force equalizing average incomes between vergence of their incomes during 1870–1913. 8
 TWO FACES OF GLOBALIZATION 671

25

20
Prados

Maddison

15

10

Bairoch

0
1800 1820 1830 1850 1860 1870 1880 1890 1900 1913 1929 1938

Figure 1. The Gini coefficient of GDPs per capita of rich countries, 1800–1938. Source: calculated from Bairoch (1997,
vol. 2, pp. 252–253), Maddison (1995, p. 194 & 2001, Appendix A), and Prados de la Escosura (2000, p. 24).

Having criticized the dominant approach for income was between 10% and 30% less to the
showing only a selective picture of globaliza- British per capita income (Bairoch, 1997, vol. 2,
tion, we need to extend it in two additional p. 845), while in 1800, Chinese per capita income
directions. First, we extend the picture over was equal or higher than the British, 9 it be-
time by considering the same WENAO coun- comes clear that, on a global scale, there must
tries during the period prior to 1870, on a have been divergence over 1800–1913, and that
well-founded assumption that globalization during the heyday of globalization in 1870–1913,
had started by the turn of the 19th century. divergence must have continued unabated.
There we notice, according to BairochÕs data, a Moreover, income declines among the non-
strong divergence during 1800–70––a diver- European participants in the globalization
gence which makes sense when one reflects that process were an integral part of the process it-
at the turn of the 19th century income differ- self: Indian deindustrialization is directly linked
ences between European countries were mini- to the British colonial commercial policy; large
mal. The Gini, according to Bairoch, more than transfers out of Indonesia and most of Africa
doubles from 6.2 in 1800 to 15.8 in 1870. Even were part and parcel of globalization. Most
MaddisonÕs data show a significant increase in important, a typical ‘‘colonial contract’’ or
inequality, with the Gini rising from 17.1 in (more properly called) a ‘‘colonial diktat’’ (see
1820 to 20.5 in 1870 (that is, by 20%). Thus, Bairoch, 1997, vol. 2, pp. 665–669) precluded
among the rich countries, once we extend our autochthonous industrial development of the
gaze past the peak period of globalization, conquered parts of the world. According to
there was a clear process of income divergence. Bairoch, the ‘‘colonial contract’’ was the main
Second, we need to extend the analysis in cause of nontransmission of industrial revolu-
space, by including other countries. Here we are, tion outside Europe since it implied that (a)
of course, on shakier grounds because none of colonies could import only products from the
the authors presents consistent series for both metropolis and tariff rates had to be low, nor-
the rich (WENAO) countries and some of the mally 0%, 10 (b) colonial exports could be made
most important future Third World countries. only to the metropolis from which they could
Yet, if we reflect that in 1760, Indian per capita re-exported (c) production of manufactured
672 WORLD DEVELOPMENT

goods that could compete with products of countries, there is no conclusive evidence that
the metropolis was banned, and (d) transport income differences did not widen. So, basically,
between colony and metropolis was conducted it is divergence all around that was brought by
only on metropolisÕ ships. Economic policy of the previous bout of globalization.
the colonies (to the extent that there was any Let us now move to the interpretation of the
independent economic policy) was therefore more recent economic record made by the un-
entirely subjugated to the interests of the me- conditional partisans of ‘‘real’’ globalization. 14
tropolis, the most important objective being to
prevent industrial competition from the col-
ony. 11
While we lack, as already mentioned, gener- 5. MISINTERPRETING THE RECENT
ally accepted estimates of GDP per capita for ECONOMIC RECORD
the future Third World countries, we do have
estimates of their levels of industrialization. Consider Tables 2 and 3. Let us then suppose
Since these are closely linked with GDP per that we show them to a Martian visitor en-
capita (and in the 19th century were even more dowed with elementary arithmetic knowledge
so), 12 we can observe not only the relative and tell him three things: first, that more
decline of the Third World but its absolute growth (higher income) is better than lower
impoverishment over the 19th century (see growth (and lower income); second, that WE-
Table 1 reproduced from Bairoch, 1997, vol. 1, NAO is the richest region and that we would
p. 404). ideally like to see differences between the rich
There we clearly see that other facet of glo- and poor regions decrease; and third, that there
balization: there can be little doubt that glo- are two periods of globalization. The first pe-
balization was responsible for the economic riod (1960–78) comprises ‘‘import substitution’’
decline of the countries that at the turn of the in Latin America and most of Asia, and Africa;
19th century were at about the same level of Communism in Eastern Europe/FSU, China,
development as Western Europe, that is, India Vietnam; and ‘‘welfare state’’ in the rich coun-
and China. For other conquered lands which tries. The second period is the era ‘‘structural
were less advanced than Western Europe, 19th adjustment’’ in Latin America and Africa,
century globalization brought colonialism, ‘‘transition to market economy’’ in Eastern
which prevented their industrialization and Europe/FSU, and ‘‘retrenchment of welfare
thus development. Now, this is not to argue state’’ in the rich world. Then we ask him to
that the underdevelopment of the Third World choose which period he thinks was better.
was the cause of the First WorldÕs development His decision should not be too difficult. He
as some hold (Frank, 1998). It suffices to take a would first observe that whether he looks at the
much more moderate and well-argued position world mean unweighted GDP per capita only
as BairochÕs and to see globalization and co- (so that each country counts the same) or at the
lonialism as a cause of Third World decline, but population-weighted world GDP per capita,
not as a cause of First World success––the growth rate was between two and three times
latter one having been essentially endogenous greater in the first period. Then, he will notice
to the West. 13 that whatever region he selects, and whatever
In conclusions, we find first, that during the concept of growth he uses, growth rate is al-
19th century, globalization was accompanied ways higher in the first period than in the sec-
by a growing divergence in income between the ond. That would provide him with some
countries of the world, and second, that even additional confidence that the first period was
among the leaders in this process, the rich better.

Table 1. Level of industrialization (manufacturing output per capita), 1800–1913 (UK 1900 ¼ 100)
1800 1830 1860 1880 1900 1913
Total developed countries 8 11 16 24 35 55
Total Third World 6 6 4 3 2 2
Memo:
United Kingdom 16 25 64 87 100 115
United States 9 14 21 38 69 126
Source: Bairoch (1997, vol. 1, p. 404).
 TWO FACES OF GLOBALIZATION 673

Table 2. Unweighted regional GDP per capita levels and growth rates, 1960–98a
GDP per capita (in 1995 international prices) Growth rate of GDP per capita
(%, p.a.)
Year 1960 Year 1978 Year 1998 1960–78 1978–98
Africa 1,514 2,147 2,432 2.0 0.6
Asia 1,971 5,944 7,050 6.3 0.9
Latin America 3,458 5,338 6,329 2.4 0.9
E. Europe/FSU 2,093 5,277 4,851 5.3 )0.4
WENAO 8,257 14,243 20,990 3.1 2.0
World 3,277 5,972 7,456 3.4 1.1
Source: Own calculations using the data from World Bank SIMA (Statistical Information Management and Ana-
lysis) database, countriesÕ statistical yearbooks, Maddison (2001) and Penn World Tables.
a
Each country is one observation.

Table 3. Population-weighted regional GDP per capita levels and growth rates, 1960–98a
GDP per capita (in 1995 international prices) Growth rate of GDP per capita
(%, p.a.)
Year 1960 Year 1978 Year 1998 1960–1978 1978–1998
Africa 1,539 2,007 2,033 1.5 0.1
Asia 963 1,945 3,967 4.0 3.6
Latin America 3,297 5,460 6,353 2.8 0.8
E. Europe/FSU 2,206 5,361 4,290 5.1 )1.1
WENAO 9,792 16,438 22,594 2.9 1.6
World 3,058 4,940 6,498 2.7 1.4
Source: Own calculations using the data from World Bank SIMA (Statistical Information Management and Ana-
lysis) database, countriesÕ statistical yearbooks, Maddison (2001) and Penn World Tables.
a
Each country is one observation, but each observation is weighted by countryÕs population.

He might then remember our instruction that capita growth rates of all countries in the world
we would also like regional incomes to con- (save the rich WENAO) during 1960–78 and
verge. Yet there too, he will notice that ac- 1978–98. Out of 124 countries, 95 grew faster in
cording to unweighted GDP per capita, in the the first period. Notice not only that most of
first period, two out of four poorer regions the dots are to the right of the 45-degree line,
grew faster than WENAO, while in the second, but also that there is a large number of the dots
all of them grew slower than WENAO. If he in the Southeastern quadrant. These are coun-
wanted to confirm that finding by looking at tries whose growth rates have switched from
what happened to an average citizen of each being positive––and often highly so––in the first
region, he would notice again that in the first period, to being negative in the second. 15
period, average per capita incomes in Eastern Then, our Martian visitor would come back
Europe/FSU and in Asia grew faster, and in to us, and na€ıvely announce that he has defi-
Latin America about the same, as in WENAO. nitely concluded that the first period was better
But in the second period, average incomes in since most countries grew faster then, and most
Africa, Latin America and Eastern Europe/ of the poorer regions tended to catch up with
FSU were about stagnant or mildly declining the rich world. He would think that the test was
(with per capita growth rates ranging from )1 rather easy and that he had done pretty well. 16
to +0.8 p.a.), while WENAO grew by 1.6% p.a., Unfortunately, our Martian is not a good
and Asia, mostly thanks to China, by 3.6% p.a. economist. Our mainstream economist would
Thus, he would conclude that, by the regional have to convince him that the second period––
convergence criterion, the first period was bet- the period of structural adjustment and glo-
ter. balization––was actually better. It would be a
In addition, we might provide our Martian hard sell, but it could be done. First, our
visitor with some further statistics. Consider economist would concede the fact that there
Figure 2 which shows the average GDP per was a divergence in countriesÕ performance
674 WORLD DEVELOPMENT

10.00%

8.00%

6.00%

4.00%
Growth rate 1978-1998

2.00%

0.00%
-4.00% -2.00% 0.00% 2.00% 4.00% 6.00% 8.00% 10.00%

-2.00%

-4.00%

-6.00%
Growth rate 1960-1978

Figure 2. Average real GDP per capita growth rates in 1960–78 and 1978–98 (124 non-WENAO countries). Source:
World Bank SIMA (Statistical Information Management and Analysis) database, countriesÕ statistical yearbooks,
Maddison (2001) and Penn World Tables.

since the end of the 1970s and that poor decade as openness ostensibly progressed. We
countries have tended to grow slower (or even shall show in detail, largely following Rodrik
to decline) than the rich countries. As shown in (2000), what is wrong with this selection crite-
Figure 3, the Gini coefficient of the GDPs per rion. But before we move to that, consider the
capita of all countries in the world, after being prelude. Since the catch-up is defined in terms
roughly stable during 1960–78, has inexorably of mean population––weighted income of the
risen since 1978, from a Gini of about 46 to a ‘‘globalizers,’’ and since China is among these,
Gini of 54 today––a huge increase of almost and since China has had such a remarkable
20%. growth record over the last two decades, the
The economist will claim however, that the authors should not have even bothered to in-
divergence in incomes is due to some ‘‘bad’’ clude other countries. All that is needed to
countries which, unwilling to globalize, have obtain the desired conclusions is that ChinaÕs
chosen the wrong policies. So, he would like to growth accelerate (as shown in World Bank,
expunge the world of these ‘‘bad’’ countries and 2002, Figure 1.12). 17
to show that there was indeed a convergence in Because China is a favorite example of the
incomes among countries that adopted ‘‘good’’ ‘‘openness is good for you’’ school, it is worth
policies and globalized. considering in somewhat greater detail. Now,
This approach, the ‘‘weeding’’ of the ‘‘bad’’ one may find it rather strange that the key
from the ‘‘good’’ countries was adopted by proof of beneficence of global capitalism is
Dollar and Kraay (1999) and by the recent provided by one of the few remaining Com-
World Bank report on globalization (World munist countries. Of course, the partisans of
Bank, 2002). These studies select countries that ‘‘real’’ globalization argue that China is a
are globalizers using the ratio of exports and Communist country in name only, and that
imports over GDP (that is, trade openness) and what matters is its integration with world
then show how highly open countriesÕ GDP per economy and de facto introduction of markets.
capita has either tended to catch up with rich Yet, the fact that a Communist countryÕs re-
countriesÕ GDP per capita, or how their growth cord is wheeled out to defend capitalism is not
rates have gradually accelerated from decade to merely a boutade. Almost one-third of ChinaÕs
 TWO FACES OF GLOBALIZATION 675

0.560

0.540

0.520

0.500

0.480
Gini coefficient

World
0.460

0.440

0.420

0.400

0.380

0.360
50

52

54

56

58

60

62

64

66

68

70

72

74

76

78

80

82

84

86

88

90

92

94

96

98
19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19
Year

Figure 3. Gini coefficient: Unweighted inter-national inequality, 1950–98. Source: Milanovic (2002). 144 countries
included. All current countries (e.g., Russia, Bangladesh, Serbia, etc.) projected backward in order to avoid spurious
Gini increase due to a greater number of observations/countries. Each country/year is one observation.

industrial output is still produced by state- hardly be taken as an example of success of the
owned enterprises, and almost 20% of total current mainstream economic policy prescrip-
GDP, a fraction higher than in any country in tions.
the world save for North Korea, Cuba, and a The very process of ‘‘selecting’’ the good
few former Soviet republics––a level of state globalizers based on trade ratios is flawed, as
involvement unlikely to be endorsed by main- argued by Rodrik (2000). He points to several
stream economists. 18 Second, one of the pre- technical and data-selection problems in the
conditions for ChinaÕs growth was arguably the Dollar and Kraay analysis, of which two seem
set of policies that is also anathema to todayÕs most important. First globalizers are selected
mainstream view: nationalization, widespread based on a combination of an outcome indi-
and free education at all levels, impediments to cator (trade over GDP) over which policy-
the free circulation of labor which kept lots of makers have no control and another which they
people from migrating into cities, land reform do control (level of tariff rates). 19 There is an
and abolishment of large landholdings––all additional problem with this selection criterion.
hardly a favorite policy prescription for a de- As Birdsall and Hamoudi (2002) show, most of
veloping country. Finally, little noticed is a ‘‘nonglobalizers’’ were unwilling nonglobalizers
paradox pointed out by Weitzman and Xu in the sense that their trade/GDP ratios had
(1997) that, by far, the most dynamic sector of declined because their exports were heavily
the Chinese economy is that of Township and dependent on natural resources and primary
Village Enterprises (TVEs) whose property commodities whose terms of trade declined in
rights are the very example of nontransparency: the 1980s. Consequently, countriesÕ export rev-
a TVE is legally owned by a ‘‘community,’’ enues dropped, and they in turn had to curtail
village or a township, is run by managers, or imports, reducing trade/GDP ratios on both
capitalists, and seeks private capital but pays accounts. In addition, they ran into balance of
no dividends. In effect, TVE is all that an effi- payments problems that required contract-
cient enterprise should not be. Yet it is this ionary policies, and it is therefore not surpris-
sector that shows the most significant progress. ing that there was a positive correlation
Thus, China, on these grounds alone, can between openness and growth. Birdsall and
676 WORLD DEVELOPMENT

Hamoudi (2002, p. 5) write: ‘‘Dollar and Kraay growth responds to a myriad of factors other
have not isolated the benefits of Ôparticipating than tariff rates. But the figures illustrate the
in the global trading systemÕ, but rather the perils of a monocausal approach to the evi-
ÔcurseÕ of primary commodity dependence.’’ Se- dence.
cond in both India and China, which, as men- The authors of World Bank (2002) are aware
tioned, are used as the prime examples of ‘‘good’’ that their preferred causality, low tariff rate-
globalizers, the main trade reforms took place s ) high export and import growth ) high
after the onset of faster growth. The Chinese GDP growth cannot be proven. 21 They are
case is, as Rodrik (2000) writes, well known: aware also that both China and India grew
high growth began in the early 1980s, while behind very high protective walls. How then do
trade liberalization followed more than a de- they deal with these issues? With respect to the
cade later. Throughout the 1980s and until first point, they do so in a rather peculiar
1995, the average weighted tariff rate in China manner, as throughout the report there are
was about 40% (Figure 4)––a rate twice as high scattered statements denying that causality can
as the average for developing countries, and be inferred from or proven by their numbers.
more than four times the average of industri- But these statements are often ignored, and
alized countries. 20 It was only in 1996 that the there are a number of precisely such––causal––
average tariff decreased to 26%, and has since statements. 22 The second point is elegantly
decreased further to a level of about 16%. circumvented by showing the change in average
Rodrik (2000) shows that the same pattern tariff rates among the ‘‘globalizers’’ and ‘‘non-
holds for India: while growth accelerated in the globalizers’’ (World Bank, 2002, p. 36). But
early 1980s, trade reform did not start until since we saw that China and India had partic-
1991–93. There, too, growth and expansion of ularly high tariff rates, it is not surprising that
trade took place under the protection of an they reduced them more than say, Barbados or
even higher tariff wall than in China: in the Belize which started the 1980s with tariff rates
1980s, the weighted tariffs averaged 80–90%, of 15%.
and gradually came down, to the still very high Our conclusion regarding the most recent
level of about 40% (Figure 4). Dollar and period of globalization is twofold.
Kraay (1999) have clearly fallen prey to one of ––The last two decades, which witnessed ex-
(what Bairoch & Kozul-Wright, 1996 call) the pansion of globalization, are, in terms of
enduring myths of economic theory, namely overall growth and income convergence be-
that ‘‘liberalization [is] an important driving tween poor and rich countries, vastly less suc-
force behind rising trade.’’ On the contrary, cessful than the preceding two decades.
trade often increased the most during the mild ––The attempt to explain divergence of in-
protectionist phases, since the latter saw accel- comes by ‘‘eliminating’’ the countries with
eration in growth, and it is growth that gener- ‘‘bad’’ policies and focusing solely on those
ally leads to trade––not vice versa (Bairoch, with ‘‘good’’ policies is flawed because the
1997, p. 310; Yotopolos, 1996). successful countries, and China in particular,
How hazardous the Globalization, growth did not follow the orthodox economic advice.
and poverty reportÕs (World Bank, 2002) con- One can be pretty confident that if China had
clusions are can be observed from the two fig- exactly the same policies, but a miserable re-
ures below which chart ChinaÕs and IndiaÕs per cord of economic growth, those who hail it
capita growth rates and their average weighted now would flaunt it as an example of how
tariff rates over 1980–99. Notice that in the harmful to growth are state ownership, unde-
India graph, it is very difficult to see any cor- fined property rights in TVEs, and high tariff
relation between the two: the growth rate os- barriers.
cillates around 4% p.a., no matter what
happens to the tariff rate. The China graph is
not much different, except that there, if any- 6. THE TWO NARRATIVES AND THE
thing, we notice a correlation between the NEED FOR ‘‘READJUSTMENT OF
slowdown in growth rate in the last five years ADJUSTMENT’’
and a reduction in tariff rates––a relationship
that is exactly the opposite of the one the We can illustrate the difference between the
World Bank report claims to have found. We dominant focus on the benign aspect of glo-
cannot put much store by this finding: it obvi- balization alone from a more even-handed
ously covers a very short period, and the rate of presentation of globalizationÕs two sides: the
 TWO FACES OF GLOBALIZATION 677

60.0 0.160

0.140
50.0

0.120

40.0
0.100

30.0 0.080

0.060
20.0

0.040

10.0
0.020

0.0 0.000
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

120.0 0.100

0.080
100.0

0.060
80.0

0.040

60.0

0.020

40.0
0.000

20.0
-0.020

0.0 - 0.040
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

Figure 4. ChinaÕs (above) and IndiaÕs (below) real annual GDP per capita growth rate, and average weighted tariffs.
Source: Growth rates from World Bank SIMA database; traiff rates from Francis Ng (World Bank). Traiff rates are
bars (levels shown on the left axis), growth rates (in fractiles) are lines (levels shown on the right axis).

benign and the malignant. Consider the fol- The dominant narrative goes approximately
lowing two historical narratives of the same set as follows. Toward the end of the 18th century,
of events. there was Industrial Revolution that spread
678 WORLD DEVELOPMENT

from Europe slowly and unevenly, to the rest of suddenly become more tamed and civilized
the world. At the end of the Napoleonic wars, (‘‘social market economy’’) after the end of WW
the world entered a period of almost uninter- II? To answer these questions, consider the fol-
rupted peace lasting 100 years. During that lowing narrative of the same events.
period global capitalism appeared: it spread to After the technological and social revolu-
the rest of the globe, connecting Europe with tions occurred in Europe, its Northwestern part
the Americas, Asia, Africa. The leading coun- became the most advanced region of the world.
tries of the period grew the fastest, their in- It set out, at first timidly and often out of ad-
comes converged as trade blossomed, people venturism, then more seriously to conquer the
freely migrated to better places, and capital rest of the world. As Europe conquered other
flowed wherever it wanted. Then, suddenly, the countries, the winners established rules that
calamity of WW I struck, the world was in- were economically advantageous to themselves,
flamed, Communism and Fascism emerged, developed further the already-existing slave
nationalism and protectionism became ram- trade, and by flooding markets of their colonies
pant, trade declined, countriesÕ incomes di- (devoid of independent commercial and eco-
verged, until another, worse, calamity of the nomic policy) with their own manufactures,
WW II struck. For a period after the war, contributed to coloniesÕ deindustrialization. All
global capitalism could not get a free rein be- the while, gross coercion, wars, and even
cause large parts of the globe fell under the genocides went on in the colonies––perhaps
Communist sway. It is only in the 1980s, as not much noticed in Europe. So, the days
China liberalized and the Soviet empire broke of universal peace were quite far from being
up and abandoned Communism, that global- truly so.
ization, with its attendant growth for most, if European powers bent on conquest were, at
not all, could resurge. ‘‘Happy days are here the same time, in a struggle with each other.
again,’’ but we must not forget that the ogres of Their imperialism begot the Great War––the
nationalism and protectionism lurk behind the very war whose impossibility, because of intri-
corner. So give freedom to capital, let profit be cate economic ties between leading countries,
your guide, and growth is guaranteed to all. was proclaimed in the famous Angell book
This is, with some poetic embellishment, the published just years prior to the carnage. After
most common view of events of the last two a truce of 20 years, the Second WW erupted––a
centuries, perhaps (one might surmise) because straight continuation of WW I. Fascism was
the people who subscribe to that narrative have defeated but Communism came out stronger
tended to experience only the benign side of and spread to cover one-third of worldÕs pop-
globalization. The objective of that narrative is ulation. Under Communist threat from the
not to stimulate discussion, but to stifle it, outside, and under pressure of growing social-
similarly to the dominant narratives used in the democratic and Communist movements at
Communist countries where too the main pur- home, the capitalist regimes, already enfeebled
pose of the accepted view of history was to by the Great Depression, conceded to dramatic
generate the acquiescence to the dogma. The and far-reaching social reforms. The nature of
point is well made by Said (2002): wild capitalism of the 19th century changed
dramatically with the introduction of unem-
In this day, and almost universally, phrases such as ployment benefits and pensions, paid vacations,
‘‘the free market,’’ ‘‘privatization,’’ Ôless governmentÕ 40-h working week, guaranteed and free edu-
and others like them have become the orthodoxy of cation, health care for all, trade unions, and
globalization, its counterfeit universals. They are sta- protection of workersÕ rights. In the Third
ples of the dominant discourse, designed to create
consent and tacit approval. . . The main goal of this
World countries that became liberated, dreams
dominant discourse is to fashion the merciless logic of industrialization and catching-up could be
of corporate profit-making and political power into realistically entertained as countries grew
a normal state of affairs. quickly and import-substitution became the
dominant approach to development. But then
It is relatively easy to explode this rosy story under the shock of rising petroleum prices, high
of the world, told by the first narrative. One interest rates, and large debts, Third World
needs only to ask three simple questions: (a) where growth sputtered. In the West, the ideological
are conquest, colonialism, and slavery in this pendulum swung against the welfare state. The
narrative? (b) how does the narrative explain the social-democratic movement weakened, the
outbreak of WW I? and (c) why did capitalism collapse of Communism eliminated the external
 TWO FACES OF GLOBALIZATION 679

threat, and made global capitalism again, as in being set without effective participation by the
the 1870s, entirely free to pursue unhindered its less-developed countries (LDCs), and the high
objectives of profit maximization––without costs of raising and pursuing a trade dispute.
much regard for social consequences. We can compare the last point to the problem
To question the profit objective is not to that Jewish survivors in Eastern Europe faced
denigrate its importance, much less to argue in trying to get the money impounded by the
that it should not be an important, perhaps the Swiss banks: how is a grand-mother surviving
most important, criterion. But it should not be on $100 a month, and not speaking English or
the sole criterion. It needs to be tempered by French, going to sue a Swiss bank?
other considerations, akin to the way that na- Because we are now dealing with global
tional capitalisms after WW II were ‘‘civilized’’ capitalism, the role of ‘‘moderator’’ can no
by the role of the state and strong social-dem- longer belong to the nation-state, but to inter-
ocratic parties. The erection of ‘‘financial via- national (global) actors. It is where the inter-
bility’’ as the only acceptable norm will not lead national financial institutions (IFI), such as the
to imperialist wars as it did in 1914, but will World Bank, enter. Continued misinterpreta-
exacerbate the negative effects of global capi- tion of the disastrous results brought to most of
talism which we already see, and which have Africa, Latin America, and Eastern Europe by
grown in importance during the last decade or about two decades of unabashedly free market
so––precisely the period during which the ear- policies will not prompt a review of these pol-
lier constraints on the free play of capital were icies, and will, on the contrary, allow their
weakened or abandoned. 23 Let me mention a continuation with probably equally bad re-
few of these effects: very high and/or increasing sults. 27 It is therefore incumbent on us to ex-
spatial and interpersonal inequality, blatant amine the actual results, and not the ideology
theft of public resources masquerading under of what these policies should have brought had
the name of privatization and cheered on by they worked as originally intended. We must
most economists and international organiza- thus address some uncomfortable issues. Let
tions, growth of slums, deteriorating labor me mention but three.
conditions, return of the long-forgotten dis- (i) How to explain that after sustained in-
eases such as tuberculosis, declines in education volvement and many structural adjustment
enrollment rates, dramatically increased mor- loans, and as many IMFÕs Stand-bys, African
tality in most of the former Soviet republics and GDP per capita has not budged from its level
Africa, deforestation, growth of worldwide of 20 years ago? Moreover, in 24 African
networks of mafias and drug cartels, even countries, GDP per capita is less than in
modern-day slavery through development of 1975, and in 12 countries even below its
piracy and abduction of women and children 1960Õs level. 28
for prostitution. 24 Capitalism left to itself will (ii) How to explain the recurrence of Latin
always produce these effects. If people want to crises, in countries such as Argentina, that
sell themselves, why should not they? If parents months prior to the outbreak of the crisis
do not want to send children to school, why not are being praised as model reformers. 29
allow them the choice? If university education (iii) How to explain that the best ‘‘pupils’’
is no longer free, perhaps a child from a poor among the transition countries (Moldova,
family can borrow to pay for it? If people do not Georgia, Kyrghyz Republic, Armenia) after
have money to pay for a cure or a drug, what setting out in 1991 with no debt at all, and
else can be done to ensure cost recovery? 25 following all the prescriptions of the IFIs,
While overt colonialism is a thing of the past, find themselves 10 years later with their
the rules are far from being even-handed as GDPs halved and in need of debt-forgive-
between the poor and the rich countries. They ness?
are slanted in favor of those who wield power. Something is clearly wrong. 30 Maintaining
Khor (2001) gives some examples from the that globalization as we know it is the way to
multilateral trading system: the well-known go and that, if the Washington consensus pol-
example of intellectual property rights, 26 dif- icies have not borne fruit so far, they will surely
ferential treatment of subsidies (subsidies for do so in the future, is to replace empiricism
R&D are exempt from counteraction while the with ideology. Unfortunately, it has been done
subsidies used by developing countries, for in- before, but the consequences were not very
dustrial upgrading, are not), standards that are good.
680 WORLD DEVELOPMENT

NOTES

1. In effect, the very first sentence of the abstract reads: 8. By the way, even the alleged divergence (Lindert &
‘‘Income of the poor rises one-for-one with overall Williamson, 2001, pp. 18–20) during the interwar ‘‘glo-
growth.’’ balization backlash’’ is not evident: according to Bai-
roch, incomes converged during that period, according to
2. In another of their papers, Dollar and Kraay Prados de la Esconsura, they diverged.
(2002) do make a point that for (a) countries cursed
by ‘‘poor geography’’ (e.g., Mali or Chad) or (b) those 9. Braudel (1984, p. 534), using BairochÕs calculations,
with inefficient or exploitative institutions, trade liber- gives Chinese GDP per capita as $282 (at US 1960
alization alone cannot be expected to bear much prices). According to Bairoch (1997, vol. 2, pp. 252–
fruit. 253), British GDP per capita in 1800 was $240.
Maddison (2001, p. 90) estimates British GDP per
capita in 1820 at $2,121 (in 1990 international dollars),
3. As Anderson (2002) rightly points out, to the
ChinaÕs GDP per capita at $600, and IndiaÕs at $533. If
European antonym: internationalism vs. nationalism,
we then set United Kingdom ¼ 1 in both Bairoch
the United States, somewhat uniquely, presented a
and Maddison, China is 1.17 in Bairoch and 0.29 in
different one: internationalism vs. isolationism. Hence
Maddison; India is 0.7–0.9 in Bairoch, and 0.25 in
specifically European xenophobia rooted in ethnicity
Maddison. Although the differences between the two
and ‘‘blood and soil’’ was never much of an ideology in
authors are often large for other countries as well (e.g.,
the United States.
Maddison gives AustraliaÕs GDP per capita in 1850 at
$3,070; recalculated in the same prices, BairochÕs
4. Even in terms of the Dutch income at the time, the estimate for the same year is $1,680), differences in the
amounts were staggering: the transfers amounted to estimates of the Chinese and Indian GDP per capita are
between 5.5% and 8.9% of Dutch GDP over the period even larger.
of 60 years. This dwarfs the Marshall plan whose net
transfers were about 4% of recipientsÕ countries GDP 10. Their maximum was often set at 5%, but at times
over the period of about five years (Bairoch, 1997, vol. 3, when such a maximum was imposed for fiscal reasons
p. 120). Of course, it makes puny todayÕs official aid (as in India in 1894), British industrial interests de-
contributed by the rich countries which is about 0.3% of manded that a similar local tax be imposed on Indian
their GDPs. products so ‘‘as not to discriminate British exports’’
(Bairoch, 1997, vol. 2, p. 860). After the first Opium war,
5. The countries are: Austria–Hungary, Belgium, Den- Britain imposed to China a maximum tariff range
mark, Finland, France, Germany, Greece, Ireland, Italy, between 5% and 9%. In a historical curiosum, note that
Netherlands, Norway, Portugal, Spain, Sweden, Swit- similar tariff preferences were imposed by Venice, and
zerland, United Kingdom in Western Europe; United later by Genoa, on the declining Eastern Roman Empire
States and Canada in North America, and Australia and from 12th century onward (see Runciman, 1932).
New Zealand in Oceania. Note that Finland, Ireland
and Norway were not independent countries for most of 11. Parts of the ‘‘colonial contract’’ (e.g., ban on
the period; Greece, up until 1830, while Germany and production of competing manufactured products) ap-
Italy are presumably included in their post-Unification plied to the European offshoots as well. This was, in
shapes. effect, one of the main motivations behind the drive for
American independence. North American producers
6. This is a variant of the so-called r convergence, but were not allowed to process pig iron, and had to sell it
a preferred one, we hold, because Gini is a better and to Great Britain only where of course it would be
more common measure of inequality than standard processed and reexported (Bairoch, 1997, vol. 2, p. 667,
deviation. & vol. 1, p. 462).

7. One explanation of the fact that inequality measured 12. The correlation between level of industrialization
by using current exchange rates (Prados) declines, while and GDP per capita in both 1900 and 1913 is about 0.7
inequality measured using PPP-constant exchange rates (calculated from Bairoch, 1997).
increases (or stays the same) is that price structures
between the countries have become more similar (see 13. Furthermore, if we extend the origin of globaliza-
Dowrick & Akmal, 2001, p. 16). tion back in time, say dating it from the European
 TWO FACES OF GLOBALIZATION 681

conquest of the Americas, then the same conclusion is 19. As Rodrik (2000, p. 1) writes: ‘‘Saying that
only reinforced. The Spanish conquest produced a Ôparticipation in world trade is good for a countryÕ is
dramatic decline in population and average incomes in as meaningful as saying that Ôupgrading of technological
the South and Central America (note that prior to the capabilities is good for growthÕ (and equally helpful to
conquest, Central AmericaÕs urbanization rates were policy makers).’’
probably greater than EuropeÕs, Bairoch, 1997, vol. 2, p.
546), while growth of slave trade did the same for 20. Based on World Bank calculations by Francis Ng
Africa. (downloadable from <www.worldbank.org/research/
As Bairoch (1989, p. 238) writes: ‘‘. . .I hasten to insist trade>).
on the fact that if colonization did not play an important
role in explaining Ôwhy we [the West] became rich,Õ it
21. The choice of this particular causality is all the
played a crucial role in explaining Ôwhy they [the Third
more intriguing since there is no reason whatsoever why
World] remained poorÕ and even why, at a certain stage
high exports (themselves a components of GDP) or
of history, Ôthey became poorer.Õ’’
imports should be bad for growth. I do not know if
anyone has ever made such a claim. At issue is precisely
14. For those who have not had the chance to follow the low tariffs ) high growth causality, which would be
Communist jargon, the ‘‘real’’ is a pun on the ‘‘real very hard to prove.
socialism,’’ the appelation invented by the Soviets in the
1970s, and similarly meant to convey the feeling that 22. I do not know how to interpret otherwise state-
their Communism, like todayÕs globalization, was the ments such as: ‘‘As they reformed and integrated with
only right one––because ‘‘real.’’ the world market, the Ômore globalizedÕ countries started
to grow rapidly, accelerating steadily from 2.9% in the
1970s to 5% throughout the 1990’’ (World Bank, 2002,
15. All the current countries are projected backward
p. 36), or the statement approvingly taken from Lindert
using their past republican/provincial growth rates. This
and Williamson (2001), ‘‘We infer that this is because
therefore represents probably the most detailed country
freer trade stimulates growth in Third world economies
growth database (see Milanovic, 2002). The main build-
today, regardless of its effects before 1940’’ (World
ing blocks for the database were World Bank SIMA,
Bank, 2002, p. 37). Or as Dollar and Kraay (2001) write:
countriesÕ statistical yearbooks, Penn World Tables, and
‘‘We provide evidence that, contrary to popular beliefs,
Maddison (2001). All GDPs per capita are expressed in
increased trade has strongly encouraged growth and
1995 international dollars.
poverty reduction and has contributed to narrowing the
gaps between rich and poor worldwide.’’
16. The first, to my knowledge, to have noticed and
discussed, with a great wealth of detail and economet- 23. Gunter Grass (2002) puts it as follows: ‘‘In the
rics, the discrepancy between the ‘‘improved’’ policies in fifties, sixties, and even in the seventies, a relatively
LDCs during the last two decades, and more than successful attempt to civilize capitalism was made across
disappointing results (worse than in the previous two Europe. If one assumes that socialism and capitalism are
decades) is Easterly (2001a,b). both indegenous, wayward children of the Enlightment,
they can be regarded as having imposed certain checks
on each other. Even capitalism was obliged to accept
17. This is incidently the wrong way to formulate the certain responsibilities. In Germany this was called the
convergence question. Convergence is always defined in social market economy. . . The consensus broke down in
terms of countries. If we were interested in whether the the early eighties. And since the collapse of the Com-
world were becoming a more equal place, the proper munist hierarchies, capitalism––recast as neoliberal-
way would be to study distribution of income among all ism––has felt it could run riot, as if out of control.
citizens of the world. The criterion used in World Bank There is no longer a counterweight to it. Today even the
(2002) is neither, and is moreover the only one capable few remaining responsible capitalists are raising a
of producing the desired results. warning finger. . . and see neoliberalism repeating the
mistakes of communism––issuing articles of faith that
deny that there is any alternative to the free market and
18. These numbers refer to 1998 and include only
claiming infallibility.’’
the value added of industrial and construction sector
State-owned enterprises (SOEs). They do not include
mixed-ownership sector or TVEs. Calculated from the 24. Kanbur (2001) writes of the spread of ‘‘obnoxious
Statistical Yearbook of China (1999), pp. 55, 432, 473. goods.’’
682 WORLD DEVELOPMENT

25. The day after I distributed the first version of feta cheese. But I remember how Armenian cognac,
this paper, a newspaper article in the Washington Post known to all under such a name, had to change its
tried to answer the question why, more than a decade appelation because ‘‘cognac’’ is a registered trademark.
after the end of Communism in the Soviet Union,
and two decades after the rejection of Maoist legacy 27. For a review of these policies see Easterly
in China, a Maoist movement in Nepal (a multi-party (2001a,b).
democracy) is making progress and can claim sup-
port among most of NepalÕs peasantry. The expla-
nation (Odenheimer, 2002) is worth quoting in 28. Meanwhile, from a much higher level, US GDP per
extenso capita has increased by a third since 1975, and has
‘‘The World Bank and the International Monetary doubled since 1960.
Fund often made economic conditions worse for poor
Nepalese. Heeding advice from the Bank and the IMF, 29. Including in the World Bank report on globaliza-
the Nepali government cut state subsidies, including tion, issued a month before the Argentine crisis, where
those that helped farmers buy fertilizer and seeds. The Argentina proudly belongs to the group of ‘‘well-
countryÕs education and health systems were privatized known’’ reformers (World Bank, 2002, p. 35). It has
to the point that most Nepalese, even if they work 14-h been demoted from that august group though in Dollar
days, cannot afford to send their children to school or and Kraay (2002) published in February 2002. By then
take them to the doctor when they are sick. Meanwhile, the crisis was all too obvious. Note that that in 1999 and
the World Bank supported huge hydroelectric and other 2000, The Heritage Index of Economic Freedom, an
massive infrastructure projects that brought windfalls to ultra right-wing think-tank, scored ArgentinaÕs eco-
international companies and corrupt Nepali officials, nomic policies about the same as Chile’s, that poster-
while utility costs for the average Nepali continued to child of the neoconservatives. Even in 2001, Argentina
rise. In the face of this poverty and corruption, the was scored only marginally worse (2.25 vs. ChileÕs 2), yet
Maoists have been playing the role of Robin Hood. much better than 3.25 given to Brazil.
Tenant farmers told me that they had been freed from
the grip of their landlords after a few well-placed Maoist
30. The typical excuse that the policies were right but
threats. Maoists have swooped down on agriculture
were badly implemented is wrong and is a very lame
banks and recaptured the land deeds that had been put
excuse indeed. It reminds me of the constant litany
up for collateral by poor farmers who had taken devel-
under Communism, that the Communist ideas were very
opment loans that they couldnÕt repay. The Maoists set
good, but were either poorly implemented, or people
up peopleÕs courts where disputes were tried without fees
were too wicked for such beautiful ideas. (I saw through
or bribes. Women used the peopleÕs courts to success-
that when I was less than 20. I am surprised that many
fully prosecute cases of wife beating and rape. Agents
smart people do not see through similar excuses today;
who enticed village girls to India and then sold them as
but then it is true that, at 20, I did not have a stake in not
prostitutes in Bombay––which happens to about 5,000
seeing the truth.) A policy that does not take into
young Nepalese women a year––were caught and pun-
account the actual situation and people as they are is
ished. Previously they often escaped by giving a cut of
inadequate. Furthermore, it is not true, even on IFIÕs
their profits to officials.’’
reckoning, that the governments always failed to fully
26. That poor countries have no money and expertise to implement the programs. Even when they did implement
enforce even the rules that may favor them is well known. them, the results––as in the transition countries––were
I have recently noticed that there is such a thing as French often relentlessly bad.

REFERENCES

Anderson, P. (2002). Internationalism: a breviary. New Bairoch, P., & Kozul-Wright, R. (1996). Globalization
Left Review, March–April, downloaded from Avail- myths: some historical reflections on integration,
able: <http://www.newleftreview.net/index.shtml>. industrialization and growth in the world economy.
Bairoch, P. (1989). The paradoxes of economic history: United Nations Conference on trade and develop-
economic laws and history. European Economic ment, Discussion paper no. 113.
Review, 33, 225–249. Birdsall, N., & Hamoudi, A. (2002). Commodity
Bairoch, P. (1997). Victoires and deboires (3 vols.). Paris: dependence, trade and growth: When ÔOpennessÕ is
Gallimard. not Enough. Center for Global Development Work-
 TWO FACES OF GLOBALIZATION 683

ing Paper No. 7; Available: <http://www.cgdev.org/ Khor, M. (2001). The multilateral trading system: a
rp/publications.html>. development perspective. Mimeo, World Network,
Bourguignon, F., & Morrisson, C. (2000). The size December.
distribution of income among world citizens, 1820– Lindert, P., & Williamson, J. (2001). Does globalization
1990. American Economic Review. make the world more unequal. National Bureau of
Braudel, F. (1984). The perspective of the world: Civili- Economic Research, Working paper 8228. Cam-
zation and capitalism, 15th–18th century. New York: bridge MA: NBER.
Harpers & Row. Maddison, A. (1995). Monitoring the world economy
Dollar, D., & Kraay, A. (1999). Trade, growth and 1820–1992. Paris: OECD, Development Centre Stud-
poverty. Policy research working paper no. 2199, ies.
World Bank, Washington DC. Maddison, A. (2001). World economy: a millennial
Dollar, D., & Kraay, A. (2000). Growth in good for the perspective. Paris: OECD Development Centre Stud-
poor. Policy research working paper no. 2587. ies.
Journal of Economic Growth, 7(3), 195–225. Milanovic, B. (2002). Worlds apart: the twentieth centuryÕs
Dollar, D., & Kraay, A. (2001). Trade, growth and promise that failed. Manuscript, downloadable from
poverty. Finance and Development, 38, (3). Down- <http://www.worldbank.org/research/inequality/>.
loaded from http://www.imf.org/external/pubs/ft/ Odenheimer, M. (2002). Where Stalin has admirers, and
fandd/2001/09/dollar.htm. Maoists fight on. Washington Post, April 7.
Dollar, D., & Kraay, A. (2002). Spreading the wealth. Rodrik, D. (2000). Comments on ÔTrade, growth and
Foreign Affairs (January/February). povertyÕ by D. Dollar and A. Kraay. Mimeo,
Dowrick, S., & Akmal, M. (2001). Contradictory trends October. Downloadable from <http://ksghome.har-
in global income inequality: a tale of two biases. vard.edu/~.drodrik.academic.ksg/>.
Version March 29, 2001. Downloaded from <http:// OÕRourke, K., & Williamson, J. (1999). Globalisation
ecocomm.anu.edu.au/economics/staff/dowrick/world- and history: The evolution of a nineteenth-century
inequ.pdf>. Atlantic economy. Cambridge, MA: MIT Press.
Easterly, B. (2001a). The lost decades: developing coun- Prados de la Escosura, L. (2000). International compar-
triesÕ stagnation in spite of policy reforms, 1980–1998. ison of real product 1820–1990. Explorations in
Mimeo. Downloaded from <http://www.worldbank. Economic History, 37, 1–41.
org/research/growth/pdfiles/lost%20decades_joeg.pdf>. Runciman, S. (1932). Byzantine civilisation. London:
Easterly, B. (2001b). The elusive quest for economic Edmond Arnold.
growth: EconomistsÕ adventures and misadventures in Said, E. W. (2002). The public role of writers and
the tropics. Cambridge and London: MIT Press. intellectuals. The 15th Jan Patocka Memorial Lec-
Fallaci, O. (2002). La rabbia e lÕorgoglio. Corriere della ture, October 24, delivered at IWM Institute, Vienna.
Sera, September 29. Sala-i-Martin, X. (2002). The disturbing ÔriseÕ in global
Frank, A. G. (1998). Re-orient. Berkeley: University of income inequality. Mimeo, March.
California Press. The Economist (2000). Growth is good. May 27, 82.
Fukuyama, F. (1989). The end of history. National Weitzman, M., & Xu, C. (1997). Chinese Township–
Interest, Summer. village enterprises as vaguely defined cooperatives.
Grass, G. (2002). The ÔprogressiveÕ restoration: a In: J. Roemer (Ed.), Property relations, incentives
discussion with Pierre Bourdieu. New Left Review, and welfare. Proceedings of a Conference held
March–April. Downloaded from <http://www.new- in Barcelona, Spain, by the International Eco-
leftreview.net/index.shtml>. nomic Association, June 1994, London: MacMillan
Hochschild, A. (1998). King LeopoldÕs ghost. Boston: Press.
Houghton Mifflin Co. World Bank (2002). Globalization, growth and poverty:
Kanbur, R. (2001). On obnoxious markets. Downloaded building an inclusive world economy. Policy Research
from <http://www.people.cornell.edu/pages/sk145/ Report. Washington, DC: World Bank.
papers.htm>. Yotopolos, P. (1996). Exchange rate parity for trade and
Keynes, J. M. (1998). Economic consequences of the development. Cambridge: Cambridge University
peace. New York: Penguin Books. Press.