Gozzini 2006

Journal of Global History
http://journals.cambridge.org/JGH
Additional services for Journal of Global History:
Email alerts: Click here

Subscriptions: Click here
Commercial reprints: Click here
Terms of use : Click here
The global system of international migrations, 1900 and

2000: a comparative approach
Giovanni Gozzini
Journal of Global History / Volume 1 / Issue 03 / November 2006, pp 321 - 341

DOI: 10.1017/S1740022806003020, Published online: 10 November 2006
Link to this article: http://journals.cambridge.org/abstract_S1740022806003020
How to cite this article:

Giovanni Gozzini (2006). The global system of international migrations, 1900 and 2000: a
comparative approach. Journal of Global History, 1, pp 321-341 doi:10.1017/S1740022806003020
Request Permissions : Click here
Downloaded from http://journals.cambridge.org/JGH, IP address: 169.230.243.252 on 02 Apr 2015

Journal of Global History (2006) 1, pp. 321–341 ª London School of Economics and Political Science 2006
doi:10.1017/S1740022806003020
The global system of international

migrations, 1900 and 2000:
a comparative approach
Giovanni Gozzini
Department of Communication Science, University of Siena, Italy
E-mail: [email protected]
Abstract
The aim of this paper is to compare international migrations in two different periods of
history, both of them marked by a rapid increase in the trend: the first period runs from
1870 to 1914, and the second from 1965 to 2000. Historical and current migration system
maps are compared, together with their different combinations of push and pull factors,
and of coerced and free migrations. The various repercussions that international migrations
have on demographic structure and on the economic systems of the sending and receiving
countries are counterbalanced by a number of significant analogies that occur at the micro-
level (individual and community) of the mechanisms governing the decision to migrate, and
of migrants’ identities and behavioural patterns. The final section shows that migrants in
both periods retain ties to both their old and new countries (as indicated by remittances
and return migration), despite the major obstacles that stand in the way of the free circula-
tion of international migrants today. This suggests that an analysis of migration systems,
emphasizing ongoing interactions of various sorts between sending and receiving areas, is
superior to either a model based on the presumption that assimilation will occur over time
or one which presumes an irreducible ‘multiculturalism’.
Historical migration systems

At the end of the nineteenth century international migrations took place in a context of
lesser inequality between countries and lesser demographic pressure than today.1 Notwith-
standing the increase of those ‘push’ and ‘pull’ factors along with the development of
the networks of communication and transport, the global volume of the international
movements of population is today proportionally lower than it was one century ago.
With 35 million migrants, the United States contains the largest number of foreign people
(12.4% of the total population) but in 1911 the proportion was larger (14.7%) and current
1 In 1870 the world population was almost one-fifth of the current one and the greater GDP per capita
differential (between United Kingdom and regional average of Africa) was 7:1, while in 1998 (between
United States and Africa) it was 20:1. See Angus Maddison, The world economy: a millennial perspective,
Paris: OECD, 2001, tab. B-10 p. 241, tab. B-21 p. 264.
I thank the journal’s editors and two anonymous referees for decisive comments on the initial draft.
321
j
322 j G I O V A N N I G O Z Z I N I
estimates by the United Nations suggest that this proportion will be roughly stable between
now and 2050, and then begin to decline.2
Where the intake of immigrants is concerned, the United States stands out as the leading
country in absolute figures today just as it did a hundred years ago: this has led historical
studies on international migrations to develop a ‘USA-centric bias’, which has pushed routes
taken by other migrant flows somewhat into the background.3 But the flow of migrants
that crossed the Atlantic a century ago bound for the American continents – despite its
pre-eminence (some 60 million migrants in the hundred years stretching from 1815 to
1914) – certainly does not account for the totality of the great late-nineteenth-century
migratory movements. According to the recent estimates by Adam McKeown, two further
and broadly comparable migrant flows occurred in addition to this one over the slightly
different periodization 1846–1940: from India and southern China to Southeast Asia, the
Indian Ocean Rim and South Pacific (estimated at between 48 and 52 million people);
from north-eastern Asia and Russia to Manchuria, Siberia, central Asia and Japan (between
46 and 51 million people).4
In Russia the migration movement was stimulated by the emancipation of serfs (1861),
the break-up of the system based on community land ownership and the deregulation
(1894) of the circulation of the domestic population. Often subsistence crises, famine and
epidemics (such as the cholera epidemic in 1891–92) were responsible for triggering migrant
movements, which were facilitated by the development of the railway network and by fall-
ing transportation costs. For the most part it was a one-way migration movement with
no return. Thus the migration movement from Russia was impacted in a major way by gov-
ernment measures deregulating internal circulation and, at the same time, shows a predomi-
nance of negative ‘push’ factors (flight from critical situations caused by shrinking
resources) over the positive ‘pull’ factor represented by the prospect of a higher salary or
improved living and working conditions.5
A significant minority (estimated at 10%) of migration movements from India, on
the other hand, was closely linked to the trade in coolies – according to one of the many
possible etymologies, a word which signifies members of the Kuli tribe, described as
2 Dianne A. Schmidley, Profile of the foreign-born population in the United States: 2000, Washington DC:
US Census Bureau, 2001, p. 9; United Nations, World population prospects. The 2004 revision: highlights,
New York: United Nations, 2005, p. 19.
3 J. D. Gould, ‘European inter-continental emigration 1815–1914: patterns and causes’, Journal of European
Economic History, 8, 3, 1979, pp. 604, 624. For an analogous criticism see Frank Thistlethwaite,
‘Migrations from Europe overseas in the 19th and 20th centuries’, in XIe Congrès International des
Sciences Historiques. Rapports, vol. 5, Histoire contemporaine, Göteborg: Almquist et Wicksell, 1960,
pp. 32–60; Dirk Hoerder, Cultures in contact. World migrations in the second millennium, Durham
and London: Duke University Press, 2002, p. 8; A. McKeown, ‘Global migration 1846–1940’, Journal
of World History, 15, 2, 2004, pp. 155–89.
4 McKeown, ‘Global migration’, tab. 1 p. 156.
5 Donald W. Treadgold, The great Siberian migration: government and peasant in resettlement from
emancipation to the First World War, Princeton: Princeton University Press, 1957; Barbara A. Anderson,
Internal migration during modernization in late-nineteenth-century Russia, Princeton: Princeton University
Press, 1980; David Moon, ‘Peasant migration, the abolition of serfdom, and the internal passport system,
c. 1800–1914’, in David Eltis, ed., Coerced and free migration. Global perspectives, Stanford, CA:
Stanford University Press, 2002, pp. 324–57.
j
T H E G L O B A L S Y S T E M O F I N T E R N A T I O N A L M I G R A T I O N S j 323
inferior, in Gujarati, the language of the Centre-West part of India6 – and to a system
for the management of salaried labour that arose in the East out of the ashes of the slave
system outlawed in India in 1833. Its place was taken by new kinds of contractual labour,
the most important of which was indentured labour: a multi-year labour contract (often
lasting five years) signed on an individual basis at the point of embarkation, providing
for a fixed salary plus board, lodging and health care on plantations and mines in other
British colonies (Burma, Sri Lanka, Malaysia and also South Africa and Jamaica). A large
amount of Indian and Chinese migrants moved under some form of contract and debt
obligation, wage labour, and more or less brutal recruitment systems. Women were a min-
ority (comprising some 10–30% of the total) split evenly between migrants’ wives and
widows seeking a fate better than the wretched prospects awaiting them at home. The
vast majority (close to 80%) of Indian migrants returned home when their contract
expired.7
In China the emperor Guangxu lifted a ban on emigration in 1893, thus clearly demon-
strating the Qing dynasty’s involvement in migrant trafficking. Between 28 and 33 million
Chinese migrated into Manchuria and Siberia, some 19 million people left the provinces
on the south-east coast (Fujian, Guangdong), bound for destinations (Thailand, the
Philippines, Indonesia, Oceania, Hawaii, Central America) that were largely complementary
compared to those reached by migrants of Indian origin.8 Most did not leave under
formal contracts of indenture, but debts that had to be paid off with money earned in the
migrant’s new location were not uncommon. The Chinese emigration was a topic of bitter
controversy right from the outset – the US Congress banned participation in the coolie trade
in 1863 – and it was only with the coming of the new century that this unbridled migration
was made subject to bilateral agreements between China and the target countries, which led
to a relative improvement in travelling and labour conditions. Historians continue to
debate the affinities between the economic coercion proper to the various forms of inden-
tured labour and the extra-economic coercion previously adopted by the slave system.
Interpretations suggesting some form of ‘continuum’ between the two, focusing on the
close similarities between slaves and coolies, have been countered by a variety of analyses
pointing to the migrants’ conscious and voluntarily signing of long-term contracts hiring
out their labour, to the relative improvement in sea-travel conditions (mortality rates
recorded in the 1880s oscillated between 10 and 25 per thousand, as opposed to the 10%
peaks hit in the middle of the century) and to the difficulty inherent in drawing any
clear distinction between economic and extra-economic coercion in the world of salaried
labour in the nineteenth century (including, to some extent, also in Europe and in North
6 Jan Breman and E. Valentine Daniel, ‘The making of a coolie’, Journal of Peasant Studies, 19, 3/4, 1992,
pp. 268–72.
7 See Hoerder, Cultures, p. 366; Kay Saunders, ed., Indentured labor in the British Empire 1834–1920,
London: Croom Helm, 1984; Piet C. Emmer, ed., Colonialism and migration: indentured labor before
and after slavery, Boston: Kluwer, 1986; Thomas Sowell, Migration and cultures: a world view, New
York: Basic Books, 1996, p. 314; Lydia Potts, The world labour market: a history of migration, London:
Zed Books, 1990, tab. 3, p. 70.
8 Robert L. Irick, Ch’ing policy toward the coolie trade 1847–1878, China: Chinese Materials
Centre, 1982;Yen Ching-Hwung, Coolies and mandarins: China’s protection of overseas Chinese during
the late Ch’ing period 1851–1911, Singapore University Press, Singapore, 1985; McKeown, ‘Global
migration’.
j
America).9 The high percentage of returns to the mother country and the overwhelming
predominance of lone males seem to suggest that these migrations were at least partly
‘strategic’, in other words linked to migrants’ capacity for making decisions, which was
admittedly slight but which nonetheless did exist.
The Atlantic migration system had behind it a history stretching back over many centu-
ries of seasonal workforce migration in Europe in the modern era, marked by the relative
geographical proximity of points of departure and destinations and by the overwhelming
force of attraction exerted by the urban expansion of the major cities.10 This link between
urbanization and emigration – which seems less significant in the Asian context – also
included many of the migrants’ ports of destination (New York, Chicago and Rio de
Janeiro).11 In the course of the nineteenth century certain ‘push’ factors either of an eco-
nomic nature (the famine of the ‘hungry forties’) or of a political nature (pogroms targeting
Jewish communities in Eastern Europe) prompted the Irish and the Jews to cross the
Atlantic. Other Europeans had already migrated from the centre and the north of the
continent (the British and the Germans in particular), and later, especially in the last two
decades of the century, they were followed by new emigrants hailing from Southern and
Eastern Europe who took advantage of both the increase in speed, reducing the Atlantic
ship crossing time from forty days to ten, and the drop in price as the changeover was
made from sail to steam power. It was during this phase that the ‘pull’ factor represented
by the hope of better work and pay opportunities came to the fore. Very early western
social sciences compared the late-nineteenth-century European migration to the strongly
restrictive elements present in the prior Irish and contemporary Jewish migrations; they
interpreted it as relatively ‘freer’ and more characterized by strategic choices, which took
into consideration the different national economic trends in the fields of employment and
salary.12
Yet economics cannot fully explain the migrant’s choice of destination. That choice
appears to be the result not only of economic considerations but also of specific chain
migration trends coming into play between local communities in the migrants’ original
9 The opposite interpretations are represented by Hugh Tinker, A new system of slavery: the export of
Indian labour overseas 1830–1920, London: Oxford University Press, 1974, and David Northrup,
Indentured labor in the age of imperialism, New York: Cambridge University Press, 1995. For an overview
of the debate see David Eltis, ‘Introduction: migration and agency in global history’, in Eltis, ed., Coerced,
pp. 1–31. For an historical survey that underlines the recent and conflicting character of the category of
‘free’ wage labor, see Robert J. Steinfeld, The invention of free labor: the employment relation in English
and American law and culture 1350–1870, Chapel Hill, NC: University of North Carolina Press, 1991.
10 Jan Lucassen, Migrant labour in Europe 1600–1700: the drift to the North Sea, London: Croom Helm,
1987; Leslie Page Moch, Moving Europeans. Migration in Western Europe since 1650, Bloomington:
Indiana University Press, 1992. One of the first essays on the matter is Ernest G. Ravenstein, ‘The laws of
migration’, Journal of the Royal Statistical Society, 48, 1885, pp. 167–227 and 52, 1889, pp. 241–301.
11 Maddison, World, tab. B-14 p. 248; Gilbert Rozman, ‘East Asian urbanization in the nineteenth century.
Comparisons with Europe’, in Ad van der Woude, Akira Hayami and Jan de Vries, eds., Urbanization in
history. A process of dynamic interactions, Oxford: Clarendon, 1990, p. 67.
12 Henry Jerome, Migration and business cycles, New York: National Bureau of Economic Research, 1926;
Dorothy S. Thomas, Social and economic aspects of Swedish population movements, New York:
Macmillan, 1941; Brinley Thomas, Migration and economic growth: a study of Great Britain and the
Atlantic economy, Cambridge: Cambridge University Press, 1954; Simon Kuznets, ‘Long swings in
population growth and related economic variables’, Proceedings of the American Philosophical Society,
102, 1958, pp. 25–52.
j
homeland and their new settlements in the receiving countries. From 1908 to 1914 the New
York Immigration Bureau recorded a broad majority of immigrants travelling to meet up
with relatives (79%) and friends (15%), while in 1910 the Immigration Commission Report
placed letters from migrants who had already settled in their new homeland and visits by
temporarily or definitively repatriated migrants at the top of the list of underlying causes
behind the migrant influx. Between 30% and 40% of immigrants into the United States
at the end of the century utilized tickets prepaid by previous migrants.13
The idea of a ‘migration system’ highlights the existence of historical (colonial, political,
commercial) links and of relations based on mutual exchange between countries at the
sending and receiving ends of international migratory flows. It does not mean a ‘hard’
historical structure regulated by predictable ways of working, but rather a dynamic (and
spatially delimited) field of research within which it is possible to analyse similarities and
differences at the micro-level of migrants’ behaviour and at the macro-level of socioeco-
nomic consequences of the international migrations.14 This vision based on connection
and interdependence tends to render obsolete the static and rigidly dichotomous view of
migrants’ identity and belonging that has long been fundamental to the study of migration:
on the one hand, the assimilationist approach (e pluribus unum) whereby original
ethnic identity must be forgotten in favour of the plan for a new life in the destination
country; and on the other the multiculturalist approach whereby immigrants’ culture must
be defended and kept separate from all possible contamination by the dominant native
culture. Rather, considered within a given migration system, a migrant’s identity is trans-
national and capable of conjugating multiple senses of belonging, of triggering processes
of exchange between sending and receiving countries and of changing over time with each
passing generation.15 Different historical migration systems show different mixtures of
free and coerced movements. But an initial cluster of movements between a region of origin
and a receiving region creates links which continue over a period of time, and establishes a
self-sustaining flow and network of information, money and further two-way movements of
people.16
13 Timothy J. Hatton and Jeffrey G. Williamson, The age of mass migration. Causes and economic impact,
Oxford and New York: Oxford University Press, 1998, p. 14; Derek Baines, Emigration from Europe
1815–1930, London: Macmillan, 1991, p. 42.
14 Mary M. Kritz and Hania Zlotnik, ‘Global interactions: migration systems, processes and policies’, in
Mary M. Kritz, Lin L. Lin and Hania Zlotnik, eds., International migration systems. A global approach,
Oxford: Clarendon, 1992, pp. 1–16.
15 For the assimilationist paradigm, see Robert E. Park, Race and culture, New York: Free Press, 1928; Oscar
Handlin, Boston’s immigrants: a study of acculturation, Cambridge, MA: Harvard University Press, 1951.
For the multiculturalist paradigm see John Bodnar, The transplanted. A history of immigrants in urban
America, Bloomington: Indiana University Press, 1985; Nathan Glazer and Daniel P. Moynihan, Beyond
the melting pot: the Negroes, Puerto Ricans, Jews, Italians, and Irish of New York City, Cambridge, MA:
MIT Press, 1970; Charles Taylor, Multiculturalism and ‘the politics of recognition’, Princeton: Princeton
University Press, 1992. For intermediate interpretations variously linked to the transnational approach see
David A. Hollinger, Postethnic America, beyond multiculturalism, New York: Basic Books, 1995; Stephen
Castles and Alastair Davidson, Citizenship and migration: globalization and the politics of belonging,
London: Macmillan, 2000.
16 Hoerder, Cultures, pp. 16–17.
j
Return migration and remittances

Thus the migration system category manages to counterbalance an exclusively economic
interpretation of migration movements and it helps to explain ‘the great unwritten story
of the history of migrations’17: the return migration. In the Indian migration system
the particularly high return migration rate (80%) reflected release from a long and burden-
some labour relationship. In the Atlantic circuit ‘departing aliens’ were not officially
recorded by the US authorities until 1908, a form of lip service to the assimilationist
myth of the melting pot. But about one-third of immigrants to North America, about
50% of immigrants to Latin America and just over 10% of immigrants to Australia returned
to their native country after a stay lasting on average no more than five years. The reasons
underlying these ‘remigrants’ decisions could be radically antithetical in nature, ranging
from the successful build-up of sufficient capital for investment back in the home country,
to failure to socially integrate in their new homeland, homesickness, political, religious, or
ideological opposition to the social system in the new homeland, or unfavourable economic
circumstances.18
Return migration represents a historic constant in migration systems. It is estimated that
about one-third of immigrants remigrated from the United States between 1900 and 1980,
and several case studies conducted between the 1980s and the present day show extremely
varied remigration trends: very low (3%) for Asians in the United States, but far higher
(from one-third to two-thirds) for Mexicans in the United States and Turks in Germany.
Remigrants extended the duration of the decision made on departure to make a break
and move, triggering social climbing processes based on home and land ownership. This
phenomenon was marked in Italy, Sweden, Hungary and Poland (where there was a marked
growth in peasant smallholdings in the first decade of the century) one hundred years ago
and apparent in Pakistan, Malta and Hong Kong today. The remigrants’ experience was
that of ‘conservative adventurers’ who – even though they had industrial training and
work experience in the receiving country – chose, once back in their homeland, to boost
their wealth and security, but in a relatively traditional manner.19 Many scholars today con-
sider return migration to have had a relatively low macroeconomic impact on economic
17 Russell King, ‘Generalizations from the history of return migration’, in Bimal Ghosh, ed., Return
migration: journey of hope or despair?, Geneva: United Nations – International Organization for
Migration, 2000, p. 7.
18 J. D. Gould, ‘European inter-continental migration. The road home: return migration from the USA’,
Journal of European Economic History, 9, 1, 1980, pp. 41–112; Mark Wyman, Round-trip to America.
The immigrants return to Europe 1880–1930, Ithaca and London: Cornell University Press, 1993; Eva
Morawska, ‘Labor migration of Poles in the Atlantic world economy’, in Dirk Hoerder and Leslie Page
Moch, eds., European migrants. Global and local perspectives, Boston: Northwestern University Press
1996, pp. 170–208; Sowell, Migration, p. 24; King, ‘Generalizations’, pp. 20–1.
19 Robert Warren and Jennyfer M. Peck, ‘Foreign-born emigration from the United States 1960–1970’,
Demography, 17, 1, 1980, pp. 1–84; Belinda I. Reyes, Dynamics of immigration: return migration to
western Mexico, S. Francisco: Public Policy Institute, 1997; George J. Borjas and Bernt M. Bratsberg, Who
leaves? The outmigration of the foreign-born, Working Paper 4913, Cambridge, MA: National Bureau of
Economic Research, 1994; Amelie Constant and Douglas S. Massey, ‘Return migration by German
guestworkers: neoclassical versus new economic theories’, International Migration, 40, 4, 2002, pp. 5–38.
The image of ‘conservative adventurers’ belongs to Diego Cinel, The national integration of Italian return
migration 1870–1929, Cambridge, MA: Cambridge University Press, 1991.
j
development in the sending countries.20 But several micro-analyses have come up with
different results, showing majorities of remigrants opting for self-employment and indepen-
dent work in Turkey, in the rural villages of Mexico, and in the Indian Punjab.21
One recurrent historical pattern concerns the role of women. It was their conduct that
displayed the most significant anthropological changes: with the return of the first migrants,
traditional costumes began to disappear from the daily life of many villages in the south of
Italy. Similarly, in Turkey in the 1990s remigrant women clashed with their communities
of origin by introducing anti-traditionalist views on such controversial personal issues as
contraception, the veil, premarital sex, abortion, and independence in making decisions
regarding work and family spending.22 In the history of international migrations women
have consistently played a decisive role in changing the nature of migration: when they
stay home the movement of men is very often temporary, but when they move, the migra-
tion of the entire family becomes permanent. For many developing countries this is a novelty
among novelties: the entry into the global circuit of migrations upsets even domestic
equilibria, profoundly modifying the roles and hierarchies of family and clan.
Emigrants’ remittances represent another enduring aspect of chain migration. It seems
that their volume has diminished in relative terms: Italian emigrants’ remittances in 1906
were worth over one-third of the value of overall exports,23 while today that figure is higher
only in Egypt (90% of exports), but lower in India (27%), Turkey (18%), Pakistan (13%)
and Mexico (6%). Yet, unlike foreign direct investments (two-thirds of which are concen-
trated in wealthy countries), three-fourths of remittances are located in developing countries
and they exceed Official Development Assistance by 44%.24 Just like a hundred years ago,
so today, too, this capital is mostly earmarked for ‘conspicuous consumption’ rather than
for investment. The main use of remittances appears to be the purchase of durable consumer
goods and, in particular, a home. In some instances (e.g. the Philippines) this has had
20 Demetrios G. Papademetriou and Philip L. Martin, eds., The unsettled relationship: labor migration and
economic development, New York: Greenwood, 1991; David O’Connor and Leila Farsakh, eds.,
Development strategy, employment, and migration: country experiences, Paris: OECD, 1996.
21 Bimal Ghosh, ‘Economic migration and sending countries’, in Julien van der Broeck, ed., The economics
of labour migration, Cheltenham: Edward Elgar, 1996, pp. 77–114; Christian Dustmann and Oliver
Kierchkamp, ‘The optimal migration duration and activity choice after remigration’, Journal of
Development Economics, 67, 2002, pp. 351–72; A. S. Oberai and H. K. Manmohan Singh, ‘Migration,
production and technology in agriculture: a case study in the Indian Punjab’, International Labour Review,
121, 3, 1982, pp. 327–43; Wayne A. Cornelius, Labour migration to the United States: development
outcomes and alternatives in Mexican sending communities, Working Paper n. 38, Washington DC:
Commission for the Study of International Migration and Cooperative Economic Development, 1990.
22 Amalia Signorelli, ‘Movimenti di popolazione e trasformazioni culturali’, in Storia dell’Italia repubblicana,
Turin: Einaudi 1995, pp. 650–1; Lincoln H. Day and Ahmet Icduygu, ‘The consequences of international
migration for the status of women: a Turkish study’, International Migration, 35, 3, 1997, pp. 337–61.
Analogous empirical evidence in Michele R. Gamburd, Kitchen spoon’s handle: transnationalism and Sri
Lanka’s migrant housemaids, Ithaca and New York: Cornell University Press, 2000.
23 Gino Massullo, ‘Economia delle rimesse’, in Piero Bevilacqua, Andreina De Clementi and Emilio Franzina,
eds., Storia dell’emigrazione italiana, vol. 1, Rome: Donzelli, 2001, pp. 161–83. Similar data are available
for the remittances from Argentina by Spanish and Portuguese migrants since 1905 through 1912: see
Magnus Hörner, ‘Immigration into Latin America, especially Argentina and Chile’, in Piet C. Emmer and
Magnus Hörner, eds., European expansion and migration. Essays on the intercontinental migration from
Asia, Africa, and Europe, New York and Oxford: Berg, 1992, p. 238.
24 Peter Gammeltoft, ‘Remittances and other financial flows to developing countries’, International
Migration, 40, 5, 2002, pp. 181–211. The non-official remittances are estimated about one-half of the
official ones.
j
a positive – albeit temporary – impact on employment in the construction industry, while in

others (e.g. Thailand) remittances have had more impact on levels of education and on the
purchase and farming of land. Often the particular uses made of remittances seem to be
negotiated with the elders of the community, who are concerned to avoid upheaval and a
break in the traditional status of hierarchies and social balances.25 That said, the remit-
tances really do reach the situations of need, and in their entirety, unlike that which fre-
quently occurs with other official channels of finance: in several Mexican cities
remittances account for one-fifth of the capital invested in micro-businesses. The Indian
state of Kerala, which is achieving unquestioned results in the struggle against poverty
and in the efficiency of its health system, is the state from which the largest number of
Indian migrants hail and to which the majority of remittances are sent.26
It seems possible to argue that return migration and remittances show more or less
important impacts if they are considered from a micro (and not only economic) or macro
point of view, both in the past and in the present day. However, above and beyond their
economic and cultural impact, remigration and remittances underscore the transnational
character of migrants and the major historical continuity of the migration systems in terms
of both individual and group behaviour.
Old and new geography

One of the first major differences, on the other hand, concerns the recent rapid increase in
the number of countries of migrant origin and destination. The most important sending
countries in 1990 numbered 55 whereas they numbered only 22 in 1914 (and 29 in
1970), while receiving countries numbered 67 (as opposed to 27 in 1914 and 39 in 1970).
From 1965 to 1995 the most important countries of immigrant origin (accounting for
75% of the total) rose from 20 to 26 for the United States (they numbered 11 in 1914),
from 14 to 24 for Canada, from 6 to 12 for Australia, and from 5 to 19 for Germany.27
Today experts point to four major migration systems: from Mexico towards the United
States; across the Mediterranean towards Europe; between Asia and the Persian Gulf; and
25 Richard H. Adams, The effects of international remittances on poverty, inequality and development in
rural Egypt, Research Report 86, Washington DC: International Food Policy Research Institute, 1991;
Mohammed Nishat and Nighat Bilgrami, ‘The impact of migrant workers’ remittances on the Pakistan
economy’, Pakistan Economic and Social Review, 29, 1, 1991, pp. 21–41; Björn Gustafsson and Negatu
Makonnen, ‘Poverty and remittances in Lesotho’, Journal of African Economies, 2, 1, 1993, pp. 49–73;
Philippe L. Delville, La riziére et la valise. Irrigation, migration et stratégies paysannes dans la vallée du
fleuve Sénégal, Paris: Syros, 1991, p. 35; Katy Gardner, Global migrants, local lives: travel and
transformation in rural Bangladesh, London: Clarendon, 1995; Peter Stalker, Workers without frontiers:
the impact of globalization on international migration, London: Rienner, 2000, p. 81; Ninna Nyberg-
Sørensen, Nicholas van Hear, and Poul Engberg Pedersen, ‘The migration-development nexus: evidence
and policy options’, International Migration, 40, 5, 2002, pp. 14–15.
26 Christopher Woodruff and Rene Zenteno, Remittances and microenterprises in Mexico, Graduate School
of International Relations and Pacific Studies Working Paper, San Diego: University of California, 2001;
Ashwani Saith, ‘Absorbing external shocks: the Gulf crisis, international migration linkages and the Indian
economy, 1990 (with special reference to the impact on Kerala)’, Development and Change, 23, 1, 1992,
pp. 101–46.
27 Hania Zlotnik, ‘Trends of international migration since 1965: what existing data reveal’, International
Migration, 37, 1, 1999, p. 29; Stalker, Workers, p. 7.
j
between the shores of the Pacific Ocean. The geographical distribution of the foreign popu-
lation stock in 1914 was concentrated in North America (to the tune of about 40%), in
Southeastern and Central Asia (35–40%) with a further pole in South America (15%) and
residual nuclei in Oceania (2%) and in Africa (2%). Today foreign population has its two
largest density poles in south-central (India, Pakistan) and western Asia (Saudi Arabia,
Israel) (27%) and North America (27%), but in addition we now have Europe (20%) and
Africa (11%), while South America (3%) has dropped to the status of a residual pole along
with Oceania (3%).
After 1945 the European continent cast off the role that it had played as the main source
of transatlantic emigration in the days of the great migration and it became instead the thea-
tre of major intracontinental migration movements (from Italy towards Germany and
Switzerland, and from Spain and Portugal towards France), which were superseded after
1970 by intercontinental migration movements from North Africa, Asia, and (to a lesser
degree) from Central and South America. Latin America, in turn, shifted from being a
land of immigration to a land of emigration. With the 1973 oil crisis a new pole of attrac-
tion for population movements emerged: the Persian Gulf region was experiencing strong
economic development linked to the influx of ‘petrodollars’ and it became a goal for migra-
tion movements (some 6 million people) from South Asia, the Far East and East Africa. The
newly industrialized countries (Hong Kong, Taiwan, South Korea and Singapore) took in
growing numbers of migrants from central and southern Asia who began to choose these
destinations alongside the traditional ongoing goals of North American and Oceania.
Underlying this increase in the number of migration routes lies greater ease of
transportation and communication on a worldwide scale: transport costs by sea and by
air dropped by 70%–80% between 1920 and 1990. But migration has not grown over
the last three decades (annual growth rate þ2.2%) at the same rate as trade in goods and
services (þ5%) or as capital flows (þ20%). Some scholars suggest that trade, investment,
and migration are substitutes for one another, but the current example of China (major
sending country of migrants, major attractor of foreign direct investments and marked by
an impressive growth of merchandise exports) seems to contradict that hypothesis. One cen-
tury ago the macroeconomic factors displayed similar trends: the great historical migration
was accompanied by a remarkable increase of international trade and capital flows.28
Yet international migration today takes place in a context of greater global inequality.
Europe’s poorest areas in 1914 had a lower gross national product per capita than the
United States or Canada (1:3.5) but it was significantly closer to that of Argentina and at
least equivalent to that of Brazil; meanwhile most Asian migrants departed from countries
(Russia, China, India) that were richer or at least not obviously poorer than the target
countries of destination.29 Historically, major migration movements have been dominated
by north–north or south–south routes, incapable of crossing the economic divide between
the northern and southern hemispheres. Today’s wealth differential (1:9 between Western
28 For the standard model of substitution between trade and factor mobility see Wolfgang F. Stolper and Paul
A. Samuelson, ‘Protection and real wages’, Review of Economic Studies, 9, 1941, pp. 58–73. For the data
on international trade and capital flows see Maddison, World, tab. F-5 p. 363; William Woodruff, Impact
of western man. A study of Europe’s role in the world economy 1750–1960, London: Macmillan 1966,
tab. IV/1 p. 150.
29 Maddison, World, tab. 3-1b, p. 126.
j
Offshoots and Asia minus Japan), on the other hand, translates into wider wage gaps
between sending and receiving countries: these gaps can rise as high as 1:5 between Eastern
Europe and the rest of the western world, while they can hit the 1:30 mark for farming and
the 1:4 mark for industry between Mexico and the United States, and 1:7 between Indonesia
and Malaysia. A skilled worker’s wage in India can be up to three times higher than that of
an unskilled worker, but a qualified engineer in Bombay earns on average 1/24th of the sal-
ary earned by his counterpart in New York: as recent surveys show, this suggests that the
inclination to emigrate is far higher (as much as fourteen times higher) among educated
Indians. The ‘pull’ factor represented by higher salaries on the work force appears to be
far greater today than it was in the past. An average Italian salary was about one-third of
an average US salary in 1913 and just under two-thirds of an average Argentine or German
salary, while in the intra-European migration movements after the Second World War the
salary differentials oscillated between 35% and 50%.30
Historical migration systems were built so as to avoid significant flows from poorest to
richest countries; this was done by directly refusing the admission (as happened to Chinese
immigrants to Western Offshoots) or by constraining the movements along the trade routes
of the colonial empires (as happened to the indentured labourers in the Indian and Pacific
Ocean). By contrast, the current migration systems are connecting poor (but not the poor-
est) sending countries and rich receiving countries, in spite of the restrictive immigration
policies developed by many governments.
Obstacles to labour migration

Despite the growth in wage gaps, the percentage of foreign population in the world (2.3%
in 1965, 2.4% in 2000) appears to be substantially stable.31 Several factors work together to
keep its size down. The first is the poverty debarring areas of Africa and Asia from taking
part in international migration systems. International migrants do not hail from poor and
isolated circumstances with no links to the world’s market places, but rather from regions
and countries experiencing rapid change and development on account of their integration
into the global trade, information, and production networks.32 The list of forty-nine least
30 Stalker, Workers, pp. 21–2; William Easterly, The elusive quest for growth. Economists’ adventures and
misadventures in the Tropics, Cambridge, MA: MIT Press, 2001, p. 81; Timothy J. Hatton and Jeffrey
G. Williamson, ‘Late-comers to mass emigration: the Latin experience’, in Timothy J. Hatton and Jeffrey
G. Williamson, eds., Migration and the international labor market 1850–1939, New York: Routledge,
1994, p. 56; Eva Morawska, ‘Labor migration of Poles in the Atlantic world economy 1880–1914’, in Dirk
Hoerder and Leslie Page Moch, eds., European migrants. Global and local perspectives, Boston:
Northwestern University Press, 1996, p. 182; Federico Romero, Emigrazione ed integrazione europea
1945–1973, Roma: Edizioni Lavoro 1991, p. 127.
31 UN figures for 2000 (175 millions of migrants which accounts for 2.9% of world population) include
29 million persons born outside their country for all the countries that used to be part of USSR: they are
the outcome of a change in classification instead of a real movement of population. See United Nations,
International Migration Report 2002, New York: United Nations, 2002, p. 6.
32 Douglas S. Massey et al., Worlds in motion: understanding international migration at the end of
millennium, Oxford: Clarendon 1998, p. 9. Riccardo Faini and Alessandra Venturini, ‘Trade, aid, and
migration: some basic policy issues’, European Economic Review, 37, 1993, pp. 435–42, estimate at 4,000
US dollars the mean annual income per capita above which the dynamics of international migration can
start.
j
developed countries in the world includes four of the sixteen most important sending coun-
tries, but only one of them (Bangladesh) belongs to the labour migration category, while the
others (Burundi, the Congo, Sudan) are part of the circuit of refugees and victims of conflict
and catastrophe. The connection between civil war and food crises in sub-Saharan Africa
has become the leading cause for migration.33 This is a particular type of coerced migration,
enforced by circumstance and short-range in terms of distance because the migrants are
obliged to go where the refugee camps are organized and they always hope to return to their
homeland once the political or economic crisis is over. The number of refugees rose from 2
to 10 million between 1965 and 2003 – after peaking at 18 million during the Balkan wars
in 1992 – and it accounts for 9% of the global migrant stock today. This kind of population
began to concentrate to the tune of over 50% in Asia (especially in Afghanistan, Iraq, and
Vietnam) in the nineties and to the tune of over 25% in Africa (Burundi, Sudan, Sierra
Leone, Somalia, Angola, and Eritrea): an overwhelming majority of these refugees simply
moved from one developing country to another. UN figures show that by the end of 2003
the overall number of refugees has dropped to just below 10 million on account of a signif-
icant return movement involving some 3.5 million people (2.6 million of whom returned to
Afghanistan following the fall of the Taliban regime). In terms of gender and class the
makeup of this refugee population does not differ much from the global average, high-
lighting the ‘normal’ and family-based character of these victims of conflict: about half of
the refugees are women and just over 40% of them are under the age of 18.34
The anti-immigration policies being implemented in receiving countries today are a
second factor helping to contain migration. The French Revolution proclaimed freedom
to migrate, triggering a cycle of national legislation which began with a pioneering British
measure in 1827 and spread legal recognition of that right to most European countries.
But between 1880 and 1914 all the most important receiving countries – except for the
United Kingdom and Brazil – changed their legislation in a restrictive sense where migrant
reception was concerned, despite the fact that there was far more room then than there is
today. (Population density in the Americas was roughly one-sixth of today’s figure.) Canada
debarred the Chinese in 1901; Australia raised the landing tax for blacks, Chinese, and
Southern Europeans to £200 sterling, as against the £5 tax for northern Europeans. The
United States banned prison inmates, the mentally deranged, the destitute and the Chinese
from entering the country in 1882, carriers of contagious diseases in 1891, and anarchists
and subversives in 1903; they introduced a literacy test in 1917, and a stringent quota sys-
tem was enforced to regulate migrant influxes from a variety of countries in 1924. Each of
these legislative changes evinces a strong link with periods of economic crisis and of growth
in wage inequality.35 The capitalist golden age after 1945 led to relative deregulation in the
33 Reginald Appleyard, ‘Emigration dynamics in developing countries’, International Migration, 33, 3/4,
1995, pp. 293–311; Myron Weiner, The global migration crisis. Challenge to states and to human
rights, New York: Harper Collins, 1995, p. 16.
34 UN High Commissioner for Refugees, 2003 global refugee trends, New York: United Nations, 2004.
35 Claudia Goldin, ‘The political economy of immigration restriction in the United States, 1890 to 1921’, in
Claudia Goldin and Gary D. Libecap, eds., The regulated economy. A historical approach to political
economy, Chicago: Chicago University Press 1994, pp. 223–57; Andrew Timmer and Jeffrey G.
Williamson, ‘Immigration policy prior to the Thirties: labor markets, policy interactions and globalization
backlash’, Population and Development Review, 24, 1998, pp. 739–71.
j
cross-border circulation of people (exemplified by the intra-European migratory dynamic), a

trend that was reversed by the economic crisis which began in 1973. The number of wealthy
countries with laws designed to cut the immigrant influx volume rose from six to thirteen
between 1976 and 1986 (today they number twenty-one) but the number of poorer coun-
tries with anti-immigration barriers at their borders also grew (from four to twenty). On
the other hand, several oil-producing countries (Saudi Arabia, Kuwait, Venezuela, and
Nigeria) opened up their borders to immigrants, only to then seal them again in the early
1980s when the price of crude oil dropped. Nigeria expelled over two million illegal immi-
grants, hailing above all from Ghana and attracted by the oil boom in the country, in 1983.
Unlike a century ago, the number of sending countries tending to restrict their migrant
outflow also rose from nineteen to thirty-seven (they number forty-one today): they include
many of the newly industrialized countries, but not China which, on the contrary, has
recognized its citizens’ personal right to emigrate since the 1980s and has today become
the leading sender country in the world in absolute terms, with a mean annual outflow of
380,000 migrants.
Another curb on the decision to migrate today is represented by the employment
offshoring strategy being adopted in the world of industry in both the northern and
southern hemispheres – between 1950 and 1990 the developing countries increased their
share in world people employed in the secondary sector from 34% to 62%.36 Another is
the existence of national welfare systems that simply were not there a hundred years ago.
This is the case with the economies in transition in the former Communist countries. Despite
the fact that this was the only area in the world showing a backsliding trend in per capita
gross national product, an abnormal spread of poverty, and a stagnant average life expec-
tancy figure in the 1990s, emigration flows towards wealthier countries were relatively
small: nearly 2.5 million (three-quarters of whom came from Russia, Kazakhstan and the
Ukraine). Moreover, these emigrants’ destinations were mostly dependent on processes of
ethnic or family reunification: Germany (about half the overall figure, mostly of German
ethnic origin), Israel (about one-fourth) and the United States (one-tenth). Since 1994 immi-
gration flows in Russia and in other countries in the region (Croatia, Slovenia, Czech
Republic, and Slovakia) have begun to prevail once again over emigration. The major causes
of that remigration seem to be not only the amount of insecurity in many Soviet successor
states and the tightening of immigration and asylum policies by the main receiving
countries, but also the basic family benefits (public health system, free education, social
insurance) provided by post-communist welfare states which make the migrants’ choice
more difficult.37
Migration and development: continuity and rupture

The main difference between today and the past, however, concerns international migra-
tions’ economic convergence effects. The history of European migration is associated with
36 See International Labour Organization, Labour Statistics Database, http://www.laborsta.ilo.org

37 United Nations, Population Division, International migration from countries with economies in transition:
1980–1999, New York: United Nations, 2002.
j
the onset of industrialization by substitution of capital for labour, privatization of land and
creation of markets both displacing traditionally stable communities and spurring on urba-
nization. On average, for each European nation, there was an interval of twenty-eight years
between the development of railway track length over the threshold of 1,000 kilometres and
the peak of international departures.38 Demographic growth and the boost to the work
force triggered by immigration in many receiving countries such as the United States –
unlike the colonial economies that attracted the Asian migration systems – imparted a fresh
thrust to investment in housing, in the railways and in training, which proved capable of
sparking a unique ‘population-sensitive’ process for the formation of capital.39
The Atlantic migration system epitomizes a theoretical model formulated by Heckscher
and Ohlin (two economists who lived in Sweden, an important source of migrants) in the
1930s: mass emigration alleviates demographic pressure on the labour market in sending
countries and triggers a rise in real salaries, while the opposite applies in the receiving coun-
tries where the arrival of migrants causes congestion in the labour market and a drop in sal-
aries. Thus international migrations tend to level out productive factors (land, labour,
capital), rather like a hydraulic system of communicating vessels. To a far greater extent
than the international movement of goods or capital, and to a far greater extent than the
spread of technological innovation, large-scale migration had a crucial impact (which can
be set at around the 60% mark) on leveling out real wage gaps between Italy and Argentina
(45%) and between Italy and the United States (102%) from 1870 to 1910. According
to O’Rourke and Williamson’s counterfactual hypothesis, those differences would have con-
siderably increased (by þ75% and þ32% respectively) without international migration.40
Also the European migration cycle in the 1950s and 1960s proved capable of triggering
rapid economic growth in countries with major work force availability such as Germany,
Italy, Switzerland and The Netherlands.41 In economic theory, migration flows provide a
rebalancing thrust between backward and advanced sectors of the economy. An inverted-
U relationship kicks in between emigration and development: migrants and salaries grow
at the same rate to the point where they trigger a situation in which the decision to migrate
is no longer seen as being as advantageous as it was originally, and expectations of local
employment once again take priority over the human price of departure, of cutting one’s
roots and of breaking ties of family and friendship.42
38 Douglas S. Massey, ‘Economic development and international migration in comparative perspective’,

Population and Development Review, 14, 3, 1988, pp. 383–413.
39 Alan Green and M. C. Urquhart, ‘Factor and commodity flows in the international economy of 1870–
1914: a multi-country view’, Journal of Economic History, 36, 1, 1976, pp. 217–52.
40 Kevin H. O’Rourke and Jeffrey G. Williamson, Globalization and history. The evolution of a nineteenth-
century Atlantic economy, Cambridge, MA: MIT Press 1999, chapters 8–9; Peter H. Lindert and Jeffrey G.
Williamson, ‘Does globalization make the world more unequal?’, in Michael D. Bordo, Alan M. Taylor,
and Jeffrey G. Williamson, eds., Globalization in historical perspective, Chicago: Chicago University Press,
2003, pp. 241–2.
41 Charles P. Kindleberger, Europe’s postwar growth: the role of the labor supply, Cambridge MA: Harvard
University Press, 1967.
42 Arthur W. Lewis, ‘Economic development with unlimited supplies of labour’, Manchester School of
Economic and Social Studies, 22, 1954, pp. 139–91; John R. Harris and Michael P. Todaro, ‘Migration,
unemployment and development: a two-sector analysis’, American Economic Review, 60, 1, 1970,
pp. 126–42.
j
Today, on the other hand, emigration in sending countries has an almost imperceptible
impact on the growing demographic trend towards expansion of the active population in
Latin America, North Africa and Asia. All of today’s most important sending countries
(China, Egypt, Indonesia, the Philippines and Turkey) show stable or rising unemployment
trends.43 Whereas a century ago migrants would set out from countries with surplus labour
and a shortage of capital bound for underpopulated countries in the process of industriali-
zation, today they tend to head for densely populated and fully industrialized countries
while setting out from countries far more heavily overpopulated than a century ago and
on a basis of greater relative inequality with their target countries.
However differences are counterbalanced by analogies. According to the data provided
by the 1910 US Census the illiteracy rate among immigrants arriving in the United States
was more than three times the rate of the native white population (13% as against 4%),
but less than three to five times the rates in their sending countries: migrants did not
come from the poorest social classes.44 Today migrants to the wealthy countries in the
OECD boast a far higher level of education (about seven years longer in school) than the
average in their countries of origin. A growing number of migrants today are highly quali-
fied people with tertiary education. Asian countries (India, China and the Philippines)
account for the highest absolute figure in this category, but a list compiled on a percentage
basis comparing tertiary educated migrants with their tertiary educated compatriots staying
behind shows that the countries worst hit by the ‘brain drain’ today are the countries of
central Africa.45
From 1870 to 1913 demographic growth in receiving countries was more than double that
in sending countries, thanks to the input provided by the Atlantic and Asian migration sys-
tems. Net immigration rates are far lower today (4.5 per 1,000 inhabitants in the
United States between 1995 and 2000 compared to 10.2 in the first decade of the twentieth
century), but the arrival of immigrants makes up for a constant drop in the proportion of min-
ors and an equally constant rise in the proportion of senior citizens, thus making a decisive
contribution to rebalancing the active and inactive sectors of the population. Yet this does
not mean that one can point to a clear connection between an increase in immigrant influx
and rising unemployment. The average annual net immigration rate in the United States
rose from 4.0 per 1,000 in the period 1990–95 to 4.5 per 1,000 in 1995–2000, while the
unemployment rate dropped from 6.4 to 4.8 per 1,000; the immigration rate in Germany
dropped from 6.6 to 2.3 per 1,000, while unemployment rose from 8.1 to 9.2 per 1,000.
43 In Mexico the rate of unemployment is falling (from 4.7% to 1.7% since 1995 through 2003) jointly with
the increasing figure of employees in the maquiladoras (the duty-free manufacturing plants located at the
northern border) from 470,000 to more than 1 million since 1995 through 2000: see Philip L. Martin,
‘Trade and migration: the Mexico-United States case’, in Slobodan Djajic, ed., International migration:
trends, policies and economic impact, London: Routledge, 2001, pp. 89–109.
44 George J. Borjas, Friends or strangers? The impact of immigrants on the US economy, New York: Basic
Books 1990, p. 137. In 1910 Southern Europe illiteracy rates ranged from 40% (Italy) to 70% (Greece).
45 Lutz Hendricks, ‘How important is human capital for development? Evidence from immigrant earnings’,
American Economic Review, 92, 2002, pp. 198–219; William J. Carrington and Enrica Detragiache,
How big is the brain drain?, Working Paper 98/102, Washington DC: International Monetary Fund, 1998;
Frederic Docquier and Abdeslam Marfouk, Measuring the international mobility of skilled workers 1990–
2000, Working Paper 3381, Washington DC: World Bank, 2004.
j
Dual labour markets

Between 1970 and 1990 average per capita income of US immigrants’ countries of origin
dropped by one-third, because of the gradual inclusion of poorer nations.46 And as the pro-
portion of migrants to the United States that come from developing countries has increased,
their mean level of education has tended to fall. Consequently immigrant workers find
themselves increasingly in direct competition with less-qualified members of the native
work force. By contrast, these immigrants are increasingly complementary to, rather than
replacements for, those native workers with higher levels of specialization – a segment of
the labour force that is growing in any case as the United States, like all developed countries,
concentrates increasingly on the production of services.47 Thus, the arrival of immigrants
today has a very minor impact on the overall salary dynamics in receiving countries, unlike
a hundred years ago. And despite what the common perception often suggests, neither is
there any clear link today between higher immigrant presence, the crisis of the welfare state
and rising taxes.48
A metaphor was coined to define immigrant workers a hundred years ago – they were
called ‘birds of passage’ – and it subsequently began to be used by scholars to point not
only to their precarious position but above all to the existence of a dual labour market split
between a regular primary group of jobs and salaries earmarked mainly for native workers,
and an irregular secondary group mainly set aside for immigrants and consisting of jobs
often described as ‘3-D’ (dirty, dangerous, difficult).49 In the Asian migration systems of a
century ago this separation of occupational circuits and thus of wage levels took the shape
of the indentured labour performed by migrants on plantations and down the mines. In the
Atlantic migration system, on the other hand, it was often the predominantly rural back-
ground of the migrants (except for the urban migrants from England and Scandinavia)
that led to disqualification and marginality in the target country’s social context.50
Yet unlike in Asia and in a large part of Latin America, almost 50% of immigrants in the
United States were involved in industry (which was going through an era of major expan-
sion), their level of qualification was not particularly low (40% were skilled workers as
46 George J. Borjas, Heaven’s door: immigration policy and the American economy, Princeton: Princeton
University Press, 1999, p. 47; Zlotnik, ‘Trends of international migration’, p. 27.
47 Adrian Wood, North–south trade, employment, and inequaliy, Oxford: Clarendon 1994.
48 George J. Borjas, ‘The economics of migration’, Journal of Economic Literature, 23, 1994, pp. 1667–717;
Michael J. Greenwood and Gary L. Hunt, ‘Economic effects of immigration on native and foreign-born
workers: complementarity, substitutability, and other channel of influence’, Southern Economic Journal,
61, 1995, n. 4, pp. 1076–97; Assaf Razin, Effraim Sadka, and Phillip Swagel, Tax burden and migration: a
political economy theory and evidence, Working Paper 6734, Cambridge, MA: National Bureau of
Economic Research, 1998, tab. 1.
49 William B. Bailey, ‘The birds of passage’, American Journal of Sociology, 18, 1912, pp. 391–7; Michael J.
Piore, Birds of passage: migrant labor in industrial societies, New York: Cambridge University Press, 1979.
50 Bodnar, Transplanted, pp. 8–17; Walter D. Kamphoefner, ‘At the crossroads of economic development:
background factors affecting emigration from nineteenth-century Germany’, in Ira A. Glazier and Luigi
De Rosa, eds., Migration, pp. 174–201; Derek E. Baines, Migration in a mature economy. Emigration and
internal migration in England and Wales 1861–1900, Cambridge: Cambridge University Press, 1986. For
the two-stage physiognomy (first rural then rural-cum-urban through domestic migration) of the
Scandinavian migrants see I. Semmingsen, ‘Emigration from Scandinavia’, Scandinavian Economic History
Review, 20, 1, 1972, pp. 45–60.
j
against 46% of native workers) and their average salary was only one-tenth lower.51 In the
United States today, on the other hand, over half the number of immigrants are spread out
over the more fragmented and hard-to-unionize jobs involved in the less-qualified areas of
the service industry (personal care and assistance, seasonal crop-picking work in the farming
industry, cleaning businesses and shops in cities). Despite boasting a constantly rising level
of education with a proportion of graduates (28%) not lagging too far behind native levels
(30%), the negative differential in average earnings appears to be growing (from 9% in
1980 to 23% in 1998). This deterioration in the situation is due above all to immigrants
hailing from Central America, and more especially from Mexico, whose average family
income, by comparison, stands at 68% and 66% respectively of the average earning figure
for native families, while families of Asian origin can boast a higher average income
(124%), and the figures for European (101%), Latin American (98%) and African (88%)
families come close to the figure for native families.52 To explain US immigrants’ earnings
differentials, the theory has been aired that today’s migration systems select migrant quality
(and potential success) also on the basis of distance. The greater the distance, along with the
amount of the (economic and sentimental) investment required for the trip, the greater
the expectation of, and commitment to, success for a decision to migrate which, while not
irreversible, at the very least has a major impact on the migrant’s life: in many ways this
is the case with Asian immigrants to other continents. Shorter-range trips, on the other
hand, which ironically have been given a fresh boost by today’s globalization process
(Mexico–US, Turkey–Germany, North Africa–Europe) appear to be less capable of making
quality or quantity selections, with a lesser impact overall both on the country of arrival and
on the country of departure.53
Yet it is unlikely that current international migrations can exert a convergence power
similar to the one displayed in the Atlantic economy according to the O’Rourke-Williamson
model, and it is equally difficult to identify any specific causal relationship between emigra-
tion and development in today’s developing countries. Whilst a century ago emigration con-
tributed to raising real wages, thus supporting the industrial take-off in many European
sending countries, today industrialization processes in the developing countries appear to
be linked to other factors: the opening up of international trade, a boost to the export
of products with high labour content and low technology content, and the attraction of
foreign investment. In other words, trade and emigration today do not seem to be alterna-
tive to one another but, rather, complementary.54 Thus for the time being, at least, it
appears that today’s international migrations are not capable of triggering the kind of
51 Christopher Hanes, ‘Immigrants’ relative rate of wage growth in the late 19th century’, Explorations in
Economic History, 33, 1996, pp. 75–64; Barry J. Eichengreen and Henry A. Gemery, ‘The earnings of
skilled and unskilled immigrants at the end of the 19th century’, Journal of Economic History, 46, 1986,
pp. 441–54.
52 Borjas, Heaven’s, pp. 21–80; Schmidley, Profile.
53 Timothy J. Hatton and Jeffrey G. Williamson, International migration in the long-run: positive selection,
negative selection and policy, Working Paper 10529, Cambridge, MA: National Bureau of Economic
Research, 2004.
54 Riccardo Faini, Jaime de Melo, and Klaus Zimmermann, Migration. The controversies and the evidence,
Cambridge, MA: Cambridge University Press, 1999, chapter 1; Wood, North, p. 186; Stalker, Workers,
p. 38.
j
processes leading to manufacturing convergence that occurred between the Old and New
Worlds in the late nineteenth century and between Southern and Northern Europe during
the 1960s.
Migrant categories and choices

Traditional migration systems in the past developed to a large extent along ‘cluster’ lines:
one-third of Finland’s emigrants came from a single province, an overwhelming majority
of Chinese emigrants came from only two provinces, and fully half the number of
Austro-Hungarian emigrants hailed from Galicia. The decision to up and go seemed to
spread through local networks like a kind of contagious fever, and in its turn it triggered
chain migration between emigrants’ local communities of origin and the local communities
in which immigrants had gathered and settled. A two-directional flow of people was kept up
over the years between the localities of origin and destination consisting of both repatriation
and new departures, but also of a constant flow of money and letters packed with news.
Migrants’ identity was both local and national; the local cohesion of the sending
Gemeinschaft strengthened the cohesion of the receiving Gemeinschaft (‘Little Italy’,
‘Chinatown’), showing surprising staying power on a worldwide scale. The majority of
migrants to the Americas migrated to industrial occupations and towns rather than frontier
homesteading, preferring the protection of their fellow countrymen to the free individual
opportunities.55
This, too, is a major historical constant: ‘the migrant stock already in the United States is
the most important predictor of the level of emigration from the country of origin’.56 Even
today there are several instances of ‘transplanted communities’ such as the Chinese village
of Houyu, fully four-fifths of which has transferred to New York. Almost all Chinese emi-
grants hail from three areas on the east coast (Guangdong, Fujian, and Zhejiang), and over
half the immigrants of Asian origin in the United States are concentrated in four cities alone
(Los Angeles, San Francisco, New York, and Honolulu). Indian emigrants hail mostly from
the state of Kerala, while Pakistani emigrants tend to come from the Mirpur district in the
region of Kashmir, where fully 90% of the local population has migrated.57
That does not mean, however, that the individual or family character of migrants has
failed to change. A hundred years ago, working-age males largely predominated over
55 John S. MacDonald and Leatrice D. MacDonald, ‘Migration, ethnic neighborhood and social networks’,
Millbank Memorial Fund Quarterly, 42, 1964, pp. 82–97; J. D. Gould, ‘European international
migration: the role of ‘‘diffusion’’ and ‘‘feedback’’’, Journal of European Economic History, 9, 1980, n. 2,
pp. 267–315; Massimo Livi Bacci, L’immigrazione e l’assimilazione negli Stati Uniti secondo le statistiche
demografiche americane, Milano: Giuffrè, 1961, p. 8; Robert P. Swieringa, ‘Local patterns of Dutch
migration to the United States in the mid-nineteenth century’, in Rudolph J. Vecoli and Suzanne M. Sinke,
eds., A century of European migrations 1830–1930, Urbana: University of Illinois Press, 1991, p. 150;
Hoerder, Cultures, p. 362.
56 UN, Department of Economic and Social Affairs, Population Division, World population monitoring.
International migration and development, New York: United Nations, 1998, p. 148.
57 Philip Q. Yang, Post-1965 immigration to the United States: structural determinants, Westport, CO:
Praeger, 1995, p. 119; Reed Veda, ‘The changing face of post-1965 immigration’, in David Jacobson,
ed., The immigration reader. America in a multidisciplinary perspective, Oxford: Blackwell, 1998, pp. 73–
4; Charles Tilly, Transplanted networks, in Virginia Yans-McLaughlin, ed., Immigration reconsidered:
history, sociology, and politics, Oxford University Press, New York 1990, pp. 25–42.
j
females among European migrants bound for the Americas, in proportions which stretched
from two-thirds to three-fourths depending on the ethnic group in question, and almost half
the migrant flow consisted of isolated individuals; male predominance was, moreover, even
more marked in Asian migration systems. Today, on the other hand, women account for
almost half the overall total in the world’s foreign population stock. The number of women
is on the increase among migrants of Asian origin, currently accounting for well over half
the overall total: some 60% of Sri Lankan citizens resident overseas and two-thirds of
Filipino migrants are women. In many of these cases the decision to emigrate is pegged to
a job in domestic service in wealthy countries and it appears to be the symptom of a process
of emancipation that has developed in the migrants’ home societies, in a context based on
the rational economic management and basic legality of the migration movement. What fre-
quently emerges is an active decision-making role played by women in the micro-businesses
abounding in Chinese communities abroad. In other instances (Nigeria, Eastern Europe)
female migrant flows are often controlled by organized crime, and to a great extent they
end up fueling the prostitution racket.58
This tragic circumstance affects only a minority of migrant women today. Two-thirds
of the overall immigrant inflow into France, Canada, and the United States and one-
fourth of the immigrant flow bound for Australia are linked to family reunification
processes.59 This is a new development even compared to European continental emigration
in the 1950s and 1960s, which was characterized by the predominance of lone males and
which was of a markedly temporary nature encouraged by geographical proximity, by the
seasonal nature of the jobs on offer (especially in the construction industry) and by the
deregulation of the circulation of people. In exactly the same way as occurred a hundred
years ago, this female presence is linked over time to a drop in the fertility rates in immi-
grant communities, which gradually became increasingly similar to those of the population
of the host country, together with a greater propensity to marry outside their ethnic
group of origin (e.g. between the second and third generations of Italian immigrants into
the United States the population grew on average from half to over two-thirds): thus the pre-
sence of women has shown itself to be a major factor for integration both in the past and
today.60
One hundred years ago, the process of taking the nationality of the receiving country
gained currency only within the Atlantic migration system; it is still less common in most
other circuits than it was in that one. Almost one-third of immigrants had naturalized in
58 Sopemi (Système d’observation permanente des migrations), Trends in international migration, Annual
Report 2000, Paris: OECD, 2000, p. 25; Jan Ryan, ‘Chinese women as transnational migrants: gender
and class in global migration narratives’, International Migration, 40, 2, 2002, pp. 93–116; Gregory
Kelson and Detra DeLaet, eds., Gender and immigration, London: Macmillan, 1999; Janet H. Momsen,
Gender, migration and domestic service, London: Routledge, 1999.
59 United Nations, International, p. 25.
60 Donna R. Gabaccia, From the other side: women, gender, and immigrant life in the United States 1820–
1990, Bloomington: Indiana University Press, 1994; Alice Scourby, ‘Three generations of Greek
Americans: a study in ethnicity’, International Migration Review, 14, 1980, pp. 43–52; Livi Bacci,
L’immigrazione, p. 87; Avery M. Guest, ‘Fertility variation among the U.S. foreign stock population in
1900’, International Migration Review, 16, 1982, pp. 577–94; Stanley Lieberson and Mary C. Waters,
From many strands: ethnic and racial groups in contemporary America, New York: Russell Sage
Foundation, 1988, p. 199.
j
the United States by 1914, but over half could not speak English.61 The education differen-
tial between natives and immigrants gradually fell in subsequent generations: 84% of first-
generation Mexican women immigrants in the United States spoke Spanish, while 84% of
the third generation speak English. Education and linguistic assimilation were connected
to improvement in economic status: the positive average income differential of second-
generation immigrants has been estimated at 6.5% over the figure for first-generation immi-
grants.62 The naturalization process in the United States today involves approximately half
the overall influx of legal immigrants and there is a significant link between it and the rate
of female immigrant participation in the work force, average income growth, a diminution
in the poverty rate and home ownership.63 The present panorama also comprises signals of
a contradictory kind. In the course of the 1990s the annual number of acquired citizenships
rose in Germany, Canada, Great Britain, Sweden, Denmark, Finland, France and Spain,
while it fell in Australia, Japan, Holland and Norway. Unlike the Gulf countries (which
deny temporary guest workers any opportunity to become citizens) many democratic
nation-states are moving from an approach chiefly based on jus sanguinis and the restriction
of naturalization toward a more inclusive one, cautiously opening the access to citizenship
to long-term residents and granting some basic provisions for family reunification. But
none of them has ratified the 1990 International Convention on the Protection of the Rights
of all Migrant Workers.64
One of the most important obstacles to migrant integration today is illegal immigration.
Random surveys and censuses set the average annual flow of undocumented migrants at
1–2 million people, which accounts for about half the overall movement.65 Data sources
(population censuses, border statistics, household surveys) are seriously affected by that
phenomenon and they can yield a reliable estimate of international migration only by
gathering and comparing all information available in a comprehensive manner. The link
between organized crime and migrant trafficking was a typical feature of Asian post-
slavery migration systems in the past. The shipping company agents who often resorted to
deceit to persuade European migrants to cross the Atlantic a hundred years ago were
acting within the bounds of legality, working closely with legitimate private business
or even with public institutions (such as the governments of Brazil and of Argentina),
and their action was rapidly recognized and endowed with a legal framework in many
61 Immigration Commission, US Congress, Report of the Immigration Commission, US Government Printing

Office, Washington DC 1911, vol. 1 p. 474.
62 Borjas, Heaven’s, p. 140; Frank D. Bean et al., ‘Educational and sociodemographic incorporation
among Hispanic immigrants to the US’, in Barry Edmonston and Jeffrey S. Passel, eds., Immigration and
ethnicity: the integration of America’s newest arrivals, Washington DC: Urban Institute, 1994, pp. 73–100;
Chris Minns, ‘Income, cohort effects, and occupational mobility: a new look at immigration to the United
States at the turn of the 20th century’, Explorations in Economic History, 37, 2000, pp. 326–50.
63 Schmidley, Profile.
64 Sopemi, Trends in international migration. Annual Report 2003, Paris: OECD, 2003, tab. B.I.6, pp. 339–
46; United Nations, International migration policies, New York: United Nations, 1998; Patrick Weil,
‘Access to citizenship: a comparison of twenty-five nationality laws’, in T. Alexander Aleinikoff and
Douglas Klusmeyer, eds., Citizenship today. Global perspectives and practices, Washington DC: Carnegie
Endowment for International Peace, 2001, pp. 17–35.
65 International Centre for Migration Policy Development, 2002 yearbook on illegal migration and
trafficking in Central and Eastern Europe, Vienna: ICMPD, 2003.
j
countries.66 Starting in 1888, the US Congress set up a number of working committees to

probe allegations that European sending countries were ‘exporting’ convicts: but on each
occasion they found confirmation of the fact that crime was less widespread among immi-
grants than among US citizens.67
That does not mean, of course, that immigrant Gemeinschaft in receiving countries were
not prepared to even step outside the law in order to protect their own compatriots (the
Italo-American example of ‘Cosa Nostra’ is emblematic in this sense). But surveys con-
ducted in Europe in the 1950s and 1960s also tended to confirm that crime among immi-
grants was lower than among the native population. That picture changed radically,
however, with the arrival of the economic crisis in the 1970s and of more restrictive immi-
gration laws. Just as happened a century earlier, xenophobia among the population in
wealthy countries went hand-in-hand with a deteriorating economic picture and with fear
of impoverishment, rather than as a reaction to any peaks in the immigration trend.68 But
the connection between anti-immigration policies and illegal immigration led in several
European countries to a rise in the number of foreigners being found in the ranks of orga-
nized crime, and thus to a link between immigration and crime: negligible or nonexistent
for some immigrant communities (Chinese, Vietnamese, Filipino), but in increasing order
of seriousness for others (Latin American, Eastern European, North African). These same
surveys reiterate the fact that the link between immigration and crime concerns only a nar-
row minority which is often opposed and isolated by its own immigrant community, but
whose presence appears to be closely linked not only to illegal immigration channels but
also to the absence of active policies or conditions for the reunification of migrant
families.69
Concluding remarks
Compared to a hundred years ago the world has become a smaller and more interconnected
place. This complicates and multiplies the geography of international migration. Asia’s post-
slavery circuits of indentured labour no longer exist and many developing countries play a
role in global labour migration today. Many areas of sub-Saharan Africa and central
Asia are still debarred from it, today as in the past, but the coerced short-range migration
of civilians fleeing from civil war and environmental disasters in those areas is growing.
66 Baines, Emigration, pp. 50–2.

67 Edith Abbott, The problem of crime and the foreign born, in National Commission on Law Observance
and Enforcement, Report on crime and foreign born, Washington DC: US Government Printing Office,
1931, pp. 23–77.
68 Kevin H. O’Rourke and Richard Sinnott, ‘What determines attitudes towards protection? Some cross-
country evidence’, in Susan Collins and Dani Rodrik, eds., Brookings Trade Forum 2000, Washington
DC: Brookings Institution, 2001, pp. 157–206; Anna Maria Mayda and Dani Rodrik, Why are some
people (and countries) more protectionist than others?, Working Paper 8641, Cambridge, MA: National
Bureau of Economic Research, 2001.
69 Peter A. Albrecht and Christian Pfeiffer, Die Kriminalisierung junger Ausländer, München: Juventa, 1979;
Michael Tomry, ed., Ethnicity, crime and immigration. Comparative and cross-national perspectives,
Chicago: University of Chicago Press, 1997; Ineke Haene Marshall, ed., Minorities, migrants, and crime.
Diversity and similarity across Europe and the United States, London: Sage, 1997.
j
In comparison to the past, the force of attraction exerted by much higher salaries appears
today as a decisive ‘pull’ factor but these greater wage differentials are now accompanied
by policies to contain and lower immigration flows which have been enforced by a vast
majority of receiving countries. Undocumented migrants account for about half of the glo-
bal volume of international migrations, and the criminal organizations increasingly exploit
the trafficking of human beings. Those novelties are accompanied by historical similarities
which the migration systems display in terms of individual and group behaviour: today as
in the past, return migration and remittances show that the typical migrant’s character is
a transnational character capable of maintaining powerful links between his/her sending
and receiving country over time and distance. This similarity is accompanied by naturaliza-
tion processes that are more widespread than in the past; and this despite the negative
synergy between anti-immigration policies and migrant trafficking. Today as in the past,
women are very much the leading players in these integration processes.
Yet international migration no longer has the macroeconomic impact that it once had. A
hundred years ago its contribution was crucial to fostering economic convergence between
Europe and the Americas. Today its impact is important to counterbalance the relative age-
ing rate in receiving countries but it is to all intents and purposes irrelevant in terms of tak-
ing the pressure off the labour market and of boosting real wages and economic
development in the sending countries. Much lesser equalizing effects are resulting from
the international migrations. The signs of entrepreneurial dynamism sparked by remigrants
and remittances in their communities of origin at the micro-economic level appear to be
restricted and specific. Despite the constant growth in foreign population flows and stock
in the world today (for all that it is slower than the growth rate in international trade),
we would be hard put today to argue that international migrations play such a clear and
specific role in international economic development as they did a century ago.
Giovanni Gozzini is Associate Professor of Contemporary

History at the University of Siena, Faculty of Humanities,
Department of Sciences of Communication.

Gozzini 2006

Uploaded by

Copyright:

Available Formats

Gozzini 2006

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Gozzini 2006

Uploaded by

Copyright:

Available Formats

Journal of Global History

Additional services for Journal of Global History:

Email alerts: Click here

The global system of international migrations, 1900 and

Journal of Global History / Volume 1 / Issue 03 / November 2006, pp 321 - 341

Link to this article: http://journals.cambridge.org/abstract_S1740022806003020

How to cite this article:

Request Permissions : Click here

Downloaded from http://journals.cambridge.org/JGH, IP address: 169.230.243.252 on 02 Apr 2015

The global system of international

Historical migration systems

Return migration and remittances

a positive – albeit temporary – impact on employment in the construction industry, while in

Old and new geography

Obstacles to labour migration

cross-border circulation of people (exemplified by the intra-European migratory dynamic), a

Migration and development: continuity and rupture

36 See International Labour Organization, Labour Statistics Database, http://www.laborsta.ilo.org

38 Douglas S. Massey, ‘Economic development and international migration in comparative perspective’,

Dual labour markets

Migrant categories and choices

61 Immigration Commission, US Congress, Report of the Immigration Commission, US Government Printing

countries.66 Starting in 1888, the US Congress set up a number of working committees to

66 Baines, Emigration, pp. 50–2.

Giovanni Gozzini is Associate Professor of Contemporary

You might also like