Migration and development

A paper prepared for the Policy Analysis and Research Programme

of the Global Commission on International Migration

by
Dhananjayan Sriskandarajah
Institute for Public Policy Research

[email protected]

September 2005

The analysis provided in this paper is that of the author, and does not
represent the views of the Global Commission on International Migration.
Introduction1

Of all the mutual impacts between countries – trade, aid, foreign investment, communication,
transport, etc. – migration perhaps has the potential to have the most significant and lasting
impacts. Migration can transform the individuals who move, the societies they move into and
even the societies they leave behind. For that same reason, migration also has the potential to
be the most politically controversial issue, especially in the societies where immigrants settle.
As a result, it is very easy when discussing migration to focus solely on the impacts of
immigration. Yet, we know that emigration can also have significant impacts, especially on
some countries in the developing world. Therefore, good migration policies need to go
beyond local impacts and take account of these mutual, global impacts.

These mutual impacts are often in sharpest relief in the context of economic development.
Here, what we might call the ‘development-migration-development’ nexus becomes of great
interest. Put simply, we know that migration is often motivated by relative disparities in the
economic development of sending and receiving countries (though generally not in the case of
those move to seek political asylum). Yet, there is also evidence to suggest that migration
itself can have important impacts on economic development, especially on relatively poorer
countries experiencing significant outflows of migrants. As the scale and complexity of
migratory flows have grown, the mutual development impacts of the flows of people, skills,
knowledge, and remittances have received considerable attention in recent years from
researchers and policy makers. 1

This paper synthesises some of the findings of this growing literature and seeks to draw out
some key interventions that policy makers at both the national and multilateral levels might
want to consider. While the paper is certainly concerned with empirical and theoretical issues
arising from this literature, its primary focus is on policy options that optimise the
instrumental benefits of migration for enhancing economic development and poverty
reduction in developing countries. Adopting what some have called a ‘nationalist’ position
(see Ellerman 2003), this paper assumes that the adverse impacts of migration on particular
countries, even if that migration has benefited the individual migrant and improved global
economic efficiency, are worth investigating and, where appropriate, doing something about.

Here, the parallels between migration and other mutual impacts, especially with international
trade, are most relevant. While it is clear that trade almost always benefits all those involved
in a transaction, policy makers at all levels still need to pay attention to particular outcomes
where other public policy aims might be compromised and to ways in which the impacts of
trade can benefit the most vulnerable stakeholders. This paper argues that the challenge of
managing the migration-development nexus seems to be much the same.

There is also another important parallel between international trade and international
migration. Some 50 years ago, the global trading system was just beginning the slow but
steady process of liberalisation. At that time, the emergence of the current rules and
governance systems for international trade would have been unimaginable to many involved.
In many ways, the current rules governing international migration resemble the global trading

1
The author is indebted to Bente Nielsen for providing invaluable research assistance for this paper, and to
several colleagues at ippr and anonymous reviewers for comments.

1
system of 50 years ago. Those thinking about what a new international framework for
managing migration would look like face remarkably similar challenges: how to design a
system that leads to freer and fairer flows of people, skills and remittances.

This paper seeks to point the way in one aspect of that system, namely how to utilise the
impacts of migration for economic development and poverty reduction. Three key areas that
affect the relationship between migration and development are examined in detail:
• the immediate and long-term impacts of the emigration of people (the loss of skills,
the potential for circulation and return migration);
• the impact of financial flows (remittances and investments); and
• the role of diaspora populations (other links).

The paper identifies both general measures that apply to all migrants in all countries (e. g. the
cost of transferring money) and also specific contexts in which the relationship between
migration and development is particularly relevant (e. g. low-income, high-emigration
countries; countries experiencing ‘brain strain’; post-conflict situations). Before turning to
these policy options, the following section outlines some of the key features of the empirical
landscape.

Some empirical observations

Migration is a complex phenomenon. Migrants move within their own country and
between countries; some people move for short periods, others permanently; some are forced
to move, others do so willingly; some people move with high levels of financial and human
capital; others are not so well endowed; and so on. Making conclusive generalised
observations about these migratory flows is not easy, let alone making observations about
their impact on economic development. This paper recognises the empirical complexity of
the subject being discussed but seeks, nevertheless, to identify possible policy interventions
that could promote fairer flows of people, money, skills and knowledge.

Migration is a widespread and growing phenomenon. While flows can be controlled or

managed to a degree, people will migrate, especially where there is war and persecution,
where economic opportunities are distributed unevenly across borders, or where there are
large inequalities in standards of living around the world. In the last few decades there has
been significant growth in the numbers of international migrants (Table 1), especially in the
developed countries.

2
Table 1. International migrants by region of destination, 1960-2000, millions

Region 1960 1970 1980 1990 2000

World 75. 9 81. 5 99. 8 154. 0 174. 9

Developed countries 32. 1 38. 3 47. 7 89. 7 110. 3

Developed countries (excl. USSR) 29. 1 35. 2 44. 5 59. 3 80. 8
Developing countries 43. 8 43. 2 52. 1 64. 3 64. 6

Africa 9. 0 9. 9 14. 1 16. 2 16. 3

Asia* 29. 3 28. 1 32. 3 41. 8 43. 8
Latin America & the Caribbean 6. 0 5. 8 6. 1 7. 0 5. 9
North America 12. 5 13. 0 18. 1 27. 6 40. 8
Oceania 2. 1 3. 0 3. 8 4. 8 5. 8
Europe** 14. 0 18. 7 22. 2 26. 3 32. 8
USSR (former) 2. 9 3. 1 3. 3 30. 3 29. 5
Source: UN/DESA 2004: viii. * Excluding Armenia, Azerbaijan, Georgia, Kazakhstan,
Kyrgyzstan, Tajikistan, Turkmenistan and Uzbekistan. ** Excluding Belarus, Estonia,
Latvia, Lithuania, the Republic of Moldova, the Russian Federation and Ukraine.
NB. These are net numbers; gross flows of migrants may be higher, meaning that the total
stock of those who have been international migrants at some stage is likely to be much higher.

However, international migratory flows are relatively small compared to other flows such
as trade or financial flows. Despite considerable deepening of global flows of money, goods,
services and information, the mobility of people remains limited. While the share of
merchandise exports and trade in services in world GDP roughly doubled between 1960 and
2000 (from around 10 to 20 per cent and 3 to 5 per cent respectively), the share of
international migrants in world population has not risen that dramatically (from around 2. 5 to
3 per cent). As such, we should be careful not to exaggerate the importance of migration.

Economic disparities between countries influence the dynamics of international

migration. People often move to seek economic opportunities in more prosperous countries.
The fastest growth in immigrant numbers has taken place in developed countries (Table 1
above) and about a quarter of all international migrants are thought to have moved from non-
OECD countries to OECD countries (Table 2). However, even amongst the more substantial
movements between non-OECD countries, many people move from poorer to richer
countries, for example from many parts of Sub-Saharan Africa to South Africa and from
many parts of Southeast Asia to Singapore and Malaysia.

3
Table 2. International migration by region, millions and percentage.

From OECD From rest of the

Total
countries world
22. 2m 34. 1m 56. 3m
To OECD countries
(16. 2 %) (24. 9%) (41. 1%)
2. 5m 77. 9m 80. 4m
To rest of the world
(1. 8%) (57. 0%) (58. 8%)
24. 7m 112. 0m 136. 7m
Total
(18. 0%) (81. 9%) (100. 0%)
Source: Harrison et al. (2004:4).

Migration can have significant benefits for global economic welfare. When migrant
workers move between differently endowed countries (e. g. from a country where there are
large labour surpluses in one sector to another where there are labour shortages in that sector),
that movement can enhance economic conditions in both sending and receiving countries.
Indeed, one estimate suggests that if developed countries were to increase the proportion of
migrant workers in the labour force equivalent to 3 per cent, world welfare would increase by
over 150 billion dollars per annum (Winters 2002).

Migration can also have significant economic benefits for developed countries. Given the
preferences of the resident workforce in developed countries, migrant workers are likely to fill
vacancies in the so-called dirty, dangerous and difficult jobs. In the medium term, industries
in developed countries that face critical vacancies can benefit from taping into excess labour
supply from developing countries (e. g. health or IT sectors in recent years). Over the long
term, as dependency ratios in developed countries rise, there will be a need to attract migrant
workers to keep an economy dynamic.

Movements from poorer to richer countries can have adverse economic impacts on
sending countries. Where these flows lead to a drain of highly skilled people from
developing countries, the ability of those countries to develop may be compromised. The
absence of these key workers hampers the ability (‘brain strain’) of these countries to come up
with homegrown solutions to their problems. Where those migrants move and contribute to
economic dynamism in destination countries, there is a risk that migration can widen the gap
between richer and poorer countries.

These impacts are often worst in the poorest countries. For some poor countries that have
high rates of permanent emigration, especially of highly skilled people, migration can be a
significant threat. Where these countries have poor economic and financial infrastructure, the
potential for emigrants to contribute to development through remittances, investment, and
return/circulation migration is also hampered.

Yet, migration can also have positive economic impacts on countries of origin. The
money that migrants send home (remittances) can contribute significantly to the recipients’
welfare as well the receiving country’s economic well-being. Where migrants return home,
either permanently or for short periods, with new skills that they put to good use, they and
their communities can benefit. Even when they don’t return in person, members of a diaspora
can contribute to the development of their erstwhile homes through trade, investment,
networking, and skills transfer. Some of these effects are summarised in Table 3.
4
Table 3. A summary of the economic effects of emigration

Positive effects Negative effects

• Provides opportunities to workers not • Loss of highly skilled workers and
available in the home country. reduced quality of essential services.
• May ease effect on the domestic market • Reduced growth and productivity
of the supply of excess labour and because of the lower stock of highly
reduce unemployment. skilled workers and its externalities.
• Inflow of remittances (that increase • Lower return from public investments in
incomes and may lead to improved public education.
human development outcomes for • Selective migration may cause
recipients) and foreign exchange. increasing disparities in incomes in the
• Technology, investments and venture home country.
capital from diasporas. • Loss of fiscal revenue from taxation of
• Can contribute to increased trade flows workers.
between sending and receiving countries. • Remittances may diminish over time.
• Stimulus to investment in domestic • Inflationary potential of remittances,
education and individual human capital especially on real estate, in some areas.
investments. • A ‘culture’ of migration, disincentives to
• Return of skilled workers may increase invest locally.
local human capital, transfer of skills and
links to foreign networks.
• Charitable activities of diasporas can
assist in relief and local community
development.
Source: adapted by author from UN/DESA (2004: 97) and other sources.

The net economic effect of migration depends on the context and is not well understood.
Despite the best efforts of economists and others working in this area, there are no accurate
models for estimating or predicting the net economic benefit of international migration on
either sending or receiving countries. For sending countries, the net impact will depend to a
great deal on the balance between temporary versus permanent migration, the balance
between high-skill versus low-skill migration, the particular sectors and labour markets
affected by emigration, the scale of remittances and so on.

Managing the impacts of migration will require global interventions. Individual states
often see migration in terms of local impacts and national interests. However, the global
nature of flows means that managing the mutual impacts of migration will need a robust
supranational framework. Moreover, managing the process for the mutual benefit of
developed and developing countries will require partnership between states.

5
Optimising movements of people

Evidence of large numbers of highly-skilled people emigrating from the developing world is
not hard to find:

• nearly one in ten tertiary-educated adults born in the developing world resided in North
America, Australia or Western Europe (Lowell et al. 2004:9);
• between a third and half of the developing world's science and technology personnel live
in the developed world;
• about 40 per cent of all African professionals have left the continent's shores in the post-
colonial period (Africa Recruit 2003);
• by the end of the 1990s, Indians working in the US on working visas accounted for 30 per
cent of the Indian software labour force (Commander et al. 2003);
• Jamaica loses about 20 per cent of its specialist nurses annually to mainly the US or the
UK (Wyss 2004); and
• the proportions of tertiary educated people amongst emigrants from some developing
countries is vastly higher than those in the resident population (Figure 1 shows, for
example, that 79. 8 per cent of emigrants from India have a tertiary education while only
2. 5 per cent of the overall Indian population has a tertiary education).

Figure 1. Propensity of the highly-educated to migrate

90

79.8
80
74.6
71.6
70
62.3 63.7

60
54.6 53.9
Percentage
50

40

30

20
14
10.4 9.2
10
2 2.5 2.3 2.8
1.1
0
Bangladesh Brazil China India Indonesia Mexico Sri Lanka Tunisia

Workers with tertiary education as a percentage of migrant population

Workers with tertiary education as a percentage of total population

Source: CPS, OECD, UNESCO cited in Kapur (2004).

These data confirm the fact that highly-skilled workers have a higher propensity to migrate
than other workers. Skilled workers are attracted to developed countries not only because of
higher wages, but also because of better working conditions with higher productivity. On the
demand side, developed countries are keen to attract skilled workers, which can fill gaps in
the labour force. The flight of skilled workers will further lower productivity and wages in
developing countries, hence increasing the problems that cause emigration in the first place.

6
This vicious circle is compounded by the fact that countries with a lack of skilled workers will
face difficulties attracting foreign direct investment.

However, the movement of highly-skilled workers itself is not something that we should
necessarily be concerned about. Indeed, there is emerging evidence to suggest that many
countries gain from the emigration of highly-skilled people (see, e. g. , Beine, Docquier and
Rapoport 2003; Commander, Kangasniemi and Winters 2003). Instead, we should be
concerned about the potential for these movements to have adverse consequences on the
development of a sending country, especially where it undermines progress towards key
public policy objectives such as the Millennium Development Goals (MDGs). We might
label instances where these adverse impacts arise as ‘brain strain’.

Box 1. Migration for development in the Philippines?

The Philippines has now surpassed Mexico as the largest exporter of migrant workers, with
about 8 million Filipinos (one tenth of the resident population) working abroad (Wehrfritz and
Vitug 2004). Such a high rate of emigration has helped the Philippines in two ways:
• it has helped eased the structural employment problems brought on by a high population
growth rate (around 2 per cent annually, with a population increase from 54. 3 million in
1985 to 81. 1 million in 2003 (ADB 2004)); and
• helped boost the Filipino economy through large-scale remittances (between US$14
billion and US$21 billion (Wehrfritz and Vitug 2004)).

The Filipino Government seeks to facilitate and promote temporary legal migration through
subsidised benefits such as pre-departure training, life and medical insurance, and emergency
loans (O’Neil 2004). It also encourages remittances to be sent through official channels: the
Overseas Workers Welfare Administration issues identity cards, which double as Visa cards
that can be used to access savings accounts and send home remittances for $3 or less per
transaction.

The question of the net impacts of emigration is an open one. There has certainly been a
seepage of highly skilled workers - the Philippine Software Association (PSA) estimates the
migration rate for IT professionals at about 30-50 per cent per year while the rate for
physicians going to foreign countries could be as high as 60 per cent (Alburo and Albella
2002). However, this must been seen against the fact that the unemployment rate for college
graduates in general has remained high (ranging from 12 to 16 per cent in the 1990s (id. )).
In other words, highly educated workers have not been absorbed in local labour markets and
therefore migration is not necessarily a direct loss to the economy, especially when
remittances are taken into account.

It is important to recognise that brain drain does not necessarily lead to brain strain. For
example, highly-skilled migrant workers who would not have otherwise been employed
effectively in the domestic labour market due to an oversupply of workers with those skills do
not necessarily represent a loss to the economy when they migrate. The sector they left may
be able to develop and thrive despite the leakage of these workers and the sending country’s
economy may actually benefit from remittances that migrant workers send home. Early
evidence from India’s IT sector suggests that the impact of mobility have not had clearly
adverse impacts (Commander et al. 2004). Hence, the Philippines continues to encourage its
skilled workers to migrate despite a very high proportion of the country’s workers already
being abroad (see box). Similarly, there have even been calls India for a special ‘emigration

7
ministry’ whose sole purpose would be to ‘export as many Indians as possible’ because it is
seen as in India’s interest to have a large, prosperous, influential and skilled diaspora. 2 In
other words, emigration of skilled workers must be seen in the context of the requirements of
the country in question. This importance of context means that the first step in addressing the
negative impacts of the emigration of highly-skilled workers is identification.

Box 2. Migration & the health MDGs

One of the clearest interactions between migration and the MDGs occurs in the area of health.
This is also where brain strain is likely to be most clearly manifested in some developing
countries. Apart from its implications for poverty reduction (goal 1), migration has important
ramifications for health-related MDGs such as halting and reversing the spread of HIV/AIDS,
reducing child mortality and maternal mortality. Mobility is often a key factor in the spread
of sexually transmitted diseases (such as HIV/AIDS) and other communicable diseases (such
as tuberculosis).

Also important though is the emigration of health professionals from developing countries.
The departure of key workers can have adverse impacts on the health situation and seriously
impede the delivery of health services. Both impacts can combine to undermine progress
towards the MDGs. In some Sub-Saharan African countries, this impact is particularly acute.
For example, Ghana, which is estimated to have lost 50 per cent of its nurses to the US, UK
and Canada within the last decade, increasing expenditure on the health sector by nearly 30
per cent in recent years has not prevented increases in infant and child mortality (Nyonator et
al. 2004). It is very likely that the depletion of trained staff through emigration is a
contributing factor.

What can be done to address these adverse impacts of migration? First, what evidence that
does exist on why health professionals migrate suggests that it is not just wage disparities that
need to be addressed. Other, non-financial factors such as working conditions, teamwork,
good supervision and training, and the opportunity to have study leave are also important
(Stilwell et al. 2004). Any successful policy interventions to promote skills retention in
developing countries must address these non-financial factors also.

Secondly, One way of helping to address these challenges that is advocated by the IOM and
other international organisations is mobilizing diaspora communities to return temporarily to
former home to work and train locals in health and other sectors (IOM 2004). Such
programmes could also be broadened to include non-diaspora members moving temporarily
to countries at risk.

Recommendation 1. Identify brain strain ‘hotspots’

There is a pressing need to identify particular sectors in particular developing countries that
risk being adversely affected by the emigration of skilled workers ('brain strain hotspots'). As
discussed above, context is vital: one-size-fits-all solutions such as blanket bans on highly-
skilled emigration may actually be counterproductive. Instead, we need to be able to identify
the scale, nature and impact of brain strain where and when in occurs. Only then can suitable

8
and effective interventions be designed and implemented. This, however, poses at least two
major challenges; one methodological and one operational.

The challenge of identifying brain strain hotspots will require a coordinated effort to pull
together a growing but disparate literature and develop a robust methodology. Understanding
the complete picture of the impacts of emigration will require combining quantitative (e. g.
modelling demand/growth in training in particular sectors, rate of growth in wages and
conditions) and qualitative (e. g. surveying migration intentions) efforts. It will also need
considerable context-specific evidence such as information on vacancy rates in key sectors,
historical and comparative changes in the distribution of certain key workers (e. g. teachers
per 1000 children), the size and nature of the sending economy, and the migration experience
of those who leave (relative incomes, remittances, return). Understanding the impact of brain
strain will only be possible through comparative analysis of progress towards achieving key
public policy targets in sectors of concern (e. g. reducing mortality or increasing literacy)
over time to see if particular outcomes are being compromised.

For such an exercise to fulfil its potential, it would also need fairly good coverage of
developing countries across the world. This would require a methodology that was suitably
robust but also a coordinated approach to assessing the extent of brain strain in different
sectors in different countries. This would probably involve setting up a network of in-country
experts who could monitor and report on worrying trends. Such a network could conduct a
‘global audit’ – something that would be an incredibly useful tool for policymakers.

Recommendation 2. Multilateral efforts to slow the flow from ‘hotspots’

Slowing the flow of key highly-skilled workers from vulnerable sectors in some developing
countries is a near impossible task: as more and more people leave, the demands on those who
remain become greater, forcing more to leave, and so on. Slowing the flow can often be
critical to stop haemorrhaging from some sectors in the immediate term and also an important
step in achieving sustainable growth in that sector. Yet, efforts to do this must avoid two
important pitfalls: compromising important public policy outcomes for the sake of preventing
emigration and infringing the rights of those who seek to emigrate.

Relatively successful developing countries often face a difficult dilemma: on the one hand,
education is an important aspect of achieving economic development, but the better an
education system is, the more easily locally-trained skilled people will be able to find more
lucrative work overseas. One temptation has been to change the education of workers to
make them unsuitable for working abroad but still adequate for meeting the needs of the
domestic economy. This has proved especially tempting in the health sector and some
African countries such as Ghana, Malawi, Tanzania and Zambia have toyed with this
approach. While this may seem like a sensible and cost effective idea (indeed, it could
actually save governments money), it may be unsustainable in the long-term and actually
compromise the quality of the services being delivered by those under-trained workers.

Another, less problematic approach, might be for developing countries to shift overall
resource allocation towards primary and secondary education rather than tertiary. Often, the
highest subsidies are spent on college and university students and spending per student in
tertiary education is much higher than in primary and secondary education. This is despite
the fact that the highly skilled have a greater tendency to emigrate and that the return to

9
investment in primary education and secondary is greater than that of tertiary education
(Psacharopoulos 1985:591; Martin 2003:28).

Yet, whatever they do, it is difficult to see how developing countries alone can mitigate
problem of brain strain given demand from developed countries for some types of workers
and huge wage disparities. Here, again underlining the principle of mutuality, developed
countries need to play a critical role. Since developed countries are currently benefiting from
migration of highly skilled workers, they also have a responsibility to ensure that those
benefits do not come at the detriment to the developing world. This can best be done through
restrictions on the active recruitment of key workers from brain strain hotspots (appropriately
identified as in Recommendation 1).

Establishing codes of practice, such as the Commonwealth Code of Practice for International
Recruitment of Health Workers, which encourage countries and agencies to recruit
responsibly and make sure that migrant-sending countries benefit in the form of overseas
assistance and/or return migration. One of the most important principles of such a code
should be to avoid recruitment of workers from particular sectors in particular countries with
a shortage of skilled workers. Employers should be strongly encouraged if not required to
sign up to this code and there needs to be cooperation within and between countries. If only
some employers sign up to a code of practice, it will have a very limited effect. Similarly, if
only some countries address the problem of brain drain while others continue to recruit
indiscriminately, the desired effect will not be achieved. Hence, one of the strongest
recommendations must be for the establishment of multilateral agreements on this issue

In doing this, governments will also need to work in partnership with the private sector,
particularly international recruitment agencies, to ensure that codes of practice are adhered to.
This would have the advantage of ensuring that ethical guidelines are followed in recruitment
policies and also that migrant workers would be offered added safety against exploitation and
a lack of rights.

Recommendation 3. Bolster training and retention in ‘hotspot’ sectors

Slowing the flows may not always be an effective way of preventing key workers from
emigrating. On the other hand, preventing emigration opportunities may actually lead to
fewer people being attracted to the sectors in question and therefore not serve the intended
outcome of meeting the need for those key workers domestically.

Here, the obvious solution would be to boost investment in educating and training staff to fill
vacancies in critical sectors. Yet, as Carrington and Detragiache (1999) note, simply
improving educational opportunities to attract more people into a sector may be largely futile
unless there are incentives to remain in the country. This means a dual approach of increased
training and more incentives to stay must be undertaken.

Again stressing the principle of mutuality, this is perhaps the area where the need for
developed countries stakeholders to play a role is most straightforward. Here, developed
country governments (either bilaterally or through multilateral agencies) and business should
increase financial support for:

10
 • financing the education and training of additional key workers in those hotspots to
meet local demand and, if appropriate, with a view to meet international demand in the
future; and
• improving wages and conditions in hotspots so that more people are attracted into
those sectors, more trained personnel choose not to migrate, and more of those who do
migrate return either temporarily or permanently.

Where governments are involved, it is important that such finance does not ‘crowd out’ other
development assistance, something that is unconscionable given that brain strain is often the
result of developed country demand. Here, policy-makers may need to engage with wider
debates about optimising the impacts of ODA, perhaps calling for exceptions to allow aid to
be channelled into meeting recurrent government costs such as wages of health workers. If
there is sufficient appetite (and the necessary technical capacity) in developed countries, such
schemes could be overtly indexed to the taxes paid by migrant workers from particular
countries.

For the private sector, there is an obvious interest in ensuring a sustainable supply of key
workers, especially given an ageing population in many developed countries. Some form of
creative tripartite partnership between sending country, receiving country and private sector
stakeholders may be a useful way of sharing the burden of training key workers more fairly.

Recommendation 4. Promote circular migration

Migration is often viewed as a process in which migrants settle permanently in a new country
with their family. However, there is evidence of more circular patterns emerging in which
migrants return to their country of birth, once or many times over a period of time.
Encouraging such circularity can make sense for several reasons:

• Return migration means that the sending country can benefit from skills learnt by the
migrant and that human capital is not lost. Indeed, returning migrants with overseas
experience can be vital to the growth and success of the sector they left and returned to
(again, early evidence suggests that this may be the case for India’s IT sector
(Commander et al. 2004)).
• Migrants are more likely to remit if they have close links with their sending country.
This is particularly the case if temporary migration results in family separation.
• Circular migration is likely to be more acceptable to voters in developed countries
(HCIDC 2004:41). Hence effective temporary migration programmes may lead to
more workers in the developing world being able to make use of migration.

Both the migrant sending and hosting countries can benefit from circular migration by
implementing a range of policies:

• Making it easy for workers to obtain temporary work visas, as evidence shows that
this actually promotes return migration. O’Neil (2004) mentions the example of
Indonesian workers: of those who regularly cross the loosely-controlled border with
Thailand, few settle for long but those workers who make it across the closely-guarded
border with Malaysia tend to stay and even bring their families with them.
• Operating a centralised temporary visa-programme in which periods of work are a
period of a few years. As Lowell et al. (2004:19) argue ‘the terms of the working

11
 agreement need to be clear at the outset and should not foster unreasonable
expectations’. The US ‘J’ cultural exchange visa is an example of such a scheme.
Migrants can work in various areas such as research or health care with the purpose of
gaining experience but they are required to return to their home country for a
minimum period of two years before being able to reapply for another visa.
• Providing financial incentives for return (rather than disincentives to stay) such as
reimbursing part of income taxation upon return to the home country. This idea is
supported by the UK HCIDC, which found that ‘given that migrants who leave will
not be making a claim on their contributions, … there is some sense of fairness in this
suggestion’ (2004:45).
• Promoting assimilation back into the migrant-sending country by providing loans for
housing and cars and ensuring re-introduction into the local labour market at a level,
which corresponds to the skills and experience acquired by the migrant.
• Developed country governments might also be wise to support more programmes for
their nationals to spend time working in developing countries, especially in brain
strain hotspots. This would bolster the idea of mutuality and facilitate greater cultural
interaction.

An important aspect of circular migration that is not discussed very often is that of pensions.
Along with other financial incentives, pensions are an important determinant of where
migrants decide to work and obviously retire. The fear of losing pension entitlements from
exiting a country or not being able to earn pension rights from a temporary period of work can
be a significant obstacle circular migration. In recognition of this the European Union has
implemented a set of rules for migrants within the European Economic Area regarding the
entitlement to pensions in order to protect migrants and facilitate migration. However, apart
from this, very little cooperation exists in this area, and this adds to the difficulties of
migrating. Not only do migrants have to spend considerable effort in obtaining information
about pension schemes and entitlements, they are also in a vulnerable situation, since they are
rarely protected by an adequate legal framework or a body of representation. This is not an
easy area to tackle as pension schemes in most countries are complicated and rely on a
lifetime of work as a basis for how to calculate entitlement. However, in order to encourage
return migration and protect migrants, it is essential that governments cooperate on finding a
resolution to this problem. Temporary migrants should be able to both earn entitlements to
state pensions and to be able to transfers savings from occupational and personal savings
schemes to the country of retirement.

In addition to this, in order to encourage return migration in the long term, it is vital that many
migrant-sending countries focus on implementing good governance, anti-corruption measures
and sound policies that foster economic growth (HCIDC 2004:47). Countries wrought by
civil war or oppressive political regimes will not be attractive to highly-skilled workers,
despite strong family ties and other social links.

While promoting circular migration, it is also important to keep in mind that temporary
workers are often vulnerable, and that they are unaware of their rights. Bilateral agreements
of temporary migration between countries could be effective in protecting labour rights, as
embassies and consulates could already provide representation of the country’s citizens and
because national administration has more bargaining power in ensuring that workers are
treated properly. The agreement between Mexico and Canada on agricultural workers
presents a good example of how temporary migration can benefit both countries (see Box 3).

12
Box 3. Temporary Mexican agricultural workers in Canada

Since 1974, Mexican agricultural workers have been able to work on a temporary basis in
Canada through a formal circular migration programme (Igartua 2004). The number of
migrants is strictly determined by demand of Canadian Employers, and a central agency in
Mexico carries out administrative duties such as medical check-ups and matching workers
with employers. The programme provides both work and income for the migrants. Despite
average work duration of about 5 months per year, Mexican workers in Canada still earn more
than they would if they were working during the whole year in Mexico. Almost all workers
in the programme say that their family’s well-being has improved as a result of the program
and that they have been able to pay for schooling for their children, hoping that this will give
them better employment opportunities outside of agriculture.

Generally, highly skilled migrants are more likely to settle permanently, perhaps because it is
easier for them to assimilate in the new country, and because they are offered good working
conditions and are often permitted to bring their families with them. This means that policies
such as those mentioned above must be particularly well targeted at providing incentives for
skilled workers to return. However, it is the case for all workers that if circular migration is
to be successful, there must be both incentives and the potential for prosperity after return
(Lowell et al. 2004:21). Again, this leaves the low-income countries or countries with
stagnating economies in the most disadvantaged position, although they need the benefits of
return migration the most. Hence, developed countries which cooperate with developing
countries in difficult situations must be particularly focused at promoting return migration by
readily issuing visas but also enforcing return rather than provide permanent status.

Optimising financial flows

Remittances from migrant workers to developing countries are:

• large and rising: in 2003 they were estimated to total US$93 billion dollars, up from
US$88 billion dollars in 2002;
• an important source of income and foreign exchange earnings for many countries,
especially for those with foreign-exchange constraints;
• now approximately double the size of net official finance (World Bank 2004:169) and
rising relative to FDI;
• in Latin America and the Caribbean, the largest and fastest growing remittance-receiving
region in the world, remittances now exceed the combined flows of all FDI and net ODA
(IADB 2004);
• usually less volatile than private capital flows that tend to move pro-cyclically, and may
even rise during recessions (Ratha 2003: 157) helping to stimulate vulnerable economies;
• more likely to reach areas of economies and societies that are left relatively untouched by
ODA and private capital from overseas investors;
• particularly important for some regions and countries: Latin American countries and
South Asia take the lion share of global remittances between them (while Sub-Saharan
Africa only receives 4. 1 per cent); and Mexico, India and the Philippines are the largest
recipients of remittances in absolute terms;

13
• extremely important for some small countries with large diasporas: as a percentage of
national income, remittances account for as much as a third of GDP in some countries
(Figure 2);
• likely to be acutely important in situations of conflict or post-war reconstruction (HCIDC
2004:56); and
• are difficult to estimate and more needs to be done to improve the accuracy of remittance
data. 3

Figure 2. Remittances as share of GDP, selected countries

40
37.3

35

30
26.5

25
22.8
Percentage

20
17
16.2 16.1
15
15 13.8 13.8 13.6 13.5 12.8

9.7 9.3
10 8.9 8.9 8.5 7.9
7

0
lic
a
ua

de

lip u

s
s
va

l S on

a
en

ep o
o

do

ub
an

ca
ga

or
ia

vi

ra
ne

nk
t
R cc
th

ua
an

la

ad
em
ag

do

er
on

va
rd

ai

du
so

an ro
ba

La
pi
os
V

an
m

qu
lb
Jo

ar

ol
T

al

on
Y

ic o
Le

Le
A

Ja

ug
e
M

V
in M
ic

ri
E
hi
ap

S
N

P
E

om
D

Source: Ratha (2003).

Not much is known about remittance behaviour, though we can observe that remittances to
developing countries are:
• often motivated by altruism (Lucas 2004:6);
• commonplace (as many as 7 out of 10 Latino migrants in the US report that they send
remittances at least once a year (IADB 2004) and 85 per cent of Zimbabweans surveyed
in South Africa and the UK said they remitted money home (Bloch 2005);
• usually sent in small amounts, around US$200-300 for US-Latin American remittances
(Suro 2003; IADB 2004);4
• mostly sent to close relatives;
• likely to reduce over time for permanent migrants, though a substantial proportion of
immigrants who have been away from home for long periods still send money back to
their relatives (IADB 2004); and
• are predominantly used to pay for common household expenditures such as food, clothing
and utilities (IABD 2004).

14
Table 4. Types of potential impacts of remittances and financial transactions

For individual remitter For ‘host’ country of remitter

• Strengthens ties with former home. • Relatively small leakage from developed
• May repay financial and other debts at countries (migrants on average remit a
home. small percentage of total income earned).
• Risks being an excessive drain on often • May serve as indirect compensation for
meagre disposable incomes. brain drain.
For recipients of remittances For country receiving remittances
• Increased incomes. • Important source of foreign exchange.
• Access to goods that migrants send or • May boost financial sector and
bring with them. investment.
• May fund increased human capital • May push exchange rates up decreasing
investment (e. g. education for family export competitiveness.
members). • May help to alleviate poverty, especially
• May provide capital for entrepreneurs. in particular regions or during
• Remittances may provide insurance emergencies.
against risk. • Opportunities for leveraging loans using
• Steady remittance income may act as expected remittances.
economic disincentive. • Migrants visiting home may bolster
tourism and related industries.
• Migrants’ demand for ‘nostalgia’
products may create new export markets
for domestic industry.
• Migrants may help fund relief,
development or community
organisations.

Any set of policies that seek to shape remittance flows must recognise some the complex
impacts of remittances on all those concerned (Table 4) but also some key principles:

• Remittances are almost always legitimate private transfers of post-tax income and
therefore should not be subject to undue governmental regulation. Regulators cannot and
should not seek to control the size, nature or use of remittances. At the very best all that
policy-makers can hope for is to expand the choices open to remitters to put their money
to good use.
• Formal, recorded transfers are to be preferred to informal flows to avoid clandestine
laundering networks, minimise risk of exploitation, improve our ability to account
properly for flows, and because deepening formal financial institutions can have other
positive economic impacts.
• It is better to use carrots rather than sticks to make sure that remittances are declared
because sanctions may reduce flows or increase the likelihood of informal transfers.
• The goal of remittance policies should not be to maximise remittances per se (there is no
clear economic argument to justify this) but rather to facilitate their transfer and optimise
their impact.

It is critical to acknowledge that there is no one-size-fits-all approach to increasing the scale

and improving the effectiveness of remittances. Areas for action range from improving

15
information available to low-wage temporary migrants to making inward investment into a
country more attractive to wealthier members of a diaspora.

Recommendation 5. Lower the cost of remittances

The cost of transferring remittances varies greatly depending on country of origin, country of
destination, means of transfer and amount sent (Orozco 2003). However, there is evidence
that migrants who remit relatively small amounts often end up paying disproportionately high
transfer costs, thereby eating into their income and/or the incomes of the remittance receiver.
For small amounts, fees sometime account for 20 per cent of the total sent (World Bank
2004:171).
Given the global scale of remittance flows, even a small decrease in transfer fees will result in
larger receipts for developing countries. Suro (2003) estimates that for remittances from the
US to Latin American countries alone, lowering those fees to about 5 per cent of total
remittances would free up an extra one billion dollars, which would benefit many poor
households. The high cost of transfer is usually ascribed to a lack of competition, lack of
information and high fixed costs for the banking sector (Suro 2003; HCIDC 2004). In
addition, a lack of access to the formal banking system by migrants hinders choice and a
move towards greater competition. 5

Recognising this, the Inter-American Development Bank has committed to working with a
network of participating stakeholders to help reach two goals by 2010: reducing by 50 per
cent the average cost of LAC remittance market transactions by promoting increased
competition, and increasing to 50 per cent the number of families receiving remittances
through the financial system (as opposed to less-formal means).

Box 4. Falling prices in the US market for transfers

Only a few years ago, the average transfer cost of remittances was around 15 to 20 per cent of
the amount sent (IADB 2004; Business Week 2003). A more recent study found that, in
February 2004, the average transfer cost from the US to Latin America and the Caribbean
stood at 7. 9 per cent for sending US$200 (IADB 2004). While transfer costs vary between
recipient countries, greater competition and better technology generally push transfer costs
down. If we assume that this has taken place across the world, then reduced transaction costs
could save remitters several billion dollars per year. Given that overall remittance volumes
are increasing, further reductions in transaction costs are likely to result in more billions being
saved.

The market for the transfer of remittances from US to Latin America has traditionally been
dominated by Western Union. Its reputation and number of branches has meant the company
has been able to charge commissions of 10% or more on the value of transactions, but it is
now facing stiff competition from other companies who see remittances as profitable
business. Major banks such as Bank of America and Citibank are cooperating with Mexican
banks in order to ease the process of and lower the costs of transfer. This also means that
relatives of migrants can now use the ATMs of local banks to access remittances sent via
major banks in the US.

Some US banks have also been allowed to use the Federal Reserve’s National Automated
Clearing House to transfer money to Mexico, and this means that transfer fees are almost as

16
low as for transfers within the US. In total, the increase in competition has meant that prices
have fallen. In 1999, ‘the average $300 transfer from the U. S. to Mexico cost $60 -- or
20%. Of that, about half was commission for the transfer and the remaining half was the cost
of exchanging dollars for pesos at the unfavourable rate charged by the companies. Today, it
costs only $10 to send the same $300. And prices are likely to fall even further. ’ As shown
by this example, increased competition and deregulation in the market for transfers can indeed
free up large amounts of funds to be used by the families of migrants.

Recommendation 6. Use remittances to strengthen financial systems

Remittances can be sent through many channels with varying levels of regulation and opacity:
banks, money-transfer organisations, hand delivery/couriering, or transfers that occur as part
of other commercial or charitable activities. Some of these latter channels have become
labelled as ‘informal’ (commonly known as hawala or hundi). These are flows are variously
estimated (roughly) to account for about a third of all remittances sent from the EU (Hussain
2005); half of all flows out of some countries (Puri and Ritzema 1999:10-11); and more than
half globally (World Bank 2004: 170). A common recommendation in the literature on
remittances is to encourage formal transfers over informal transfers, especially as there is
concern, especially since the events of 11 September 2001, that informal remittances are
being used to finance terrorism.

However, as Pieke, van Hear and Lindley (2005) usefully point out, the distinction between
‘formal’ and ‘informal’ remittances flows is blurred, changing and sometimes unhelpful.
They also point out that there seems to be no systemic difference in the developmental impact
of one sort of transfer over another, with ‘informal’ systems often offering considerable
financial and other advantages for migrants themselves. As such, we should not prioritise
only the so-called ‘formal’ avenues. Finally, they suggest (pp. 34-5) that governments
should adopt a ‘a firm but light touch in regulating money transfer’ so as not to squeeze out
small operators who may provide a specific niche in facilitating transfers to particularly
vulnerable, often going the ‘last mile’ that more regulated organisations may not be able to.
That said, the more we encourage recorded remittances of any kind, the better for data
collection and understanding the exact scale and nature of remittances.

What is perhaps more important than the route of transfer is the impact that transfer has on
the financial system of the recipient country. Here, basic economic theory tells us that a
robust, healthy financial infrastructure is a prerequisite for a healthy economy. In general, it
is clear that good financial infrastructure is essential for ensuring economic development. In
the financial sector in developed countries savings are used as investment funds. However,
in the case of remittances to developing countries, the relationship is not so clear-cut,
especially where money is sent through non-banking channels.

A greater use of the banking sector for transfers also opens up sources of finance for the
migrant-recipient, and it allows remittances kept as savings to be used as funds for
investment. Hence, there is a clear link between developing the banking sector in order to
capture informal flows of remittances, and channelling those flows towards investment (Puri
and Ritzema 1999:25). From the discussion above, it is clear that recommended policies to
encourage a switch away from informal transfer methods must focus on providing the means
and incentives for migrants to remit formally. Hence, these policies would include:

17
 • Making it easier for migrants to access the formal banking system (HCIDC 2004:59).
A greater number of transfers would lower average fixed costs faced by banks and
give migrants greater choice and hence increase competition.
• Encouraging greater transparency such that migrants are able to compare costs of
transferring money (Ibid:59). Increasing the amount of information available to
migrants is important, especially since there is evidence that their knowledge of the
banking sector and the actual cost of transferring is very limited. This role could be
filled by NGOs who are better positioned than government institutions to collect,
analyse and distribute information on specific remittance service providers (Carling
2004:5). Pre-departure financial training for migrants may also be useful (de Bruyn,
and Kuddus 2005).
• Facilitating cooperation between major international banks and smaller banks in
developing countries in order to connect networks of branches in migrant-hosting and
migrant-sending countries. One way of achieving this is to ensure that such
cooperation is legal (World Bank 2004:172).

At the same time, micro-entrepreneurs in developing countries cannot obtain credit at a

reasonable interest rate due to an underdeveloped banking network, and lack of bank accounts
and assets that can be offered as guarantees. This lack of access to credit can hamper
economic growth and create the dependency on remittances described above. Here, there
may be an important role for channelling remittances through micro-finance schemes that
supply the type of credit required by small-scale entrepreneurs (Puri and Ritzema 1999) (see
Box 5).

Box 5. Linking remittances and micro-finance

One example of linking remittances to finance is the Inter-American Development Bank’s
Multilateral Investment Fund that seeks to cooperate with Latin-American micro-finance
institutions to help them enter the market for remittances in order to create such links. 6
Another interesting and pioneering project is that of the Microfinance International
Corporation (MFIC), which was initially set up to cater for migrants from El Salvador in the
US but will be extended to other Latin American countries. Remittances are collected by the
firm to such a scale that they constitute a pool of low-cost funds, which can then be used in
lending programs by partner micro-finance institutions. MFIC also uses software which not
only allows instant access to remittances by the receivers, but which also provides the sender
with the opportunity to use remittances for school fees, utility bills or savings. Hence, MFIC
provides an excellent example of a profit-making firm, which nevertheless contributes to
economic growth and development in developing countries. 7 There are also clear
opportunities for encouraging established micro-finance institutions to extend their savings
and credit services to migrants and their families, or even for other organisations to offer these
sorts of services.

See also the International Remittance Network (IRNET) established by the World Council of
Credit Unions: http://www. woccu. org/prod_serv/irnet/.

Remittances should also be seen in the context of the broader financial context affecting many
developing countries. For a start, decisions on how to remit are affected by factors such as
the existence of parallel exchange rate markets and tariffs on the import of goods carried by
returning migrants. Secondly, there is considerable scope for leveraging finance using

18
expected remittances as collateral. While there are relatively few formal arrangements for
this in place, agencies like the IDB are looking into it. Thirdly, there is considerable scope for
financial institutions in recipient countries to securitise against expected remittances earnings.
This goes some way in overcoming the currency and emerging market risks that many
countries face by establishing an offshore collection account for foreign currency receipts.
The advantages of monetising future flows in this way is rarely discussed by the migration
research scholars, though banks in Latin America and Turkey have gone down this route and
agencies like the IMF, World Bank and regional development banks are advocating its use. 8

Recommendation 7. Maximise development impacts of remittances

Remittances usually constitute small payments sent by individual migrants to their relatives.
Since remittances are person-to-person flows, they are well targeted to meet the needs of the
recipient (World Bank 2004) and they have the ability to lift people out of poverty. Based on
a study of panel-data, Page and Adams (2003:20) conclude that ‘on average, a 10 per cent
increase in the share of international migrants in a country’s population will lead to a 1. 9 per
cent decline in the share of people living on less than $1. 00 per person per day’.

At the same time, remittances have been thought to increase inequality, since wealthier
workers are better able to pay the costs of migrating (Ratha 2003). However, well-educated
migrants are less likely to remit as they tend to settle permanently and bring their families
with them. 9 In addition, even if the poor do not receive remittances, they can still benefit
indirectly through increased demand for their labour, and this can explain the positive
correlation between remittances and alleviation of poverty described above. Hence, a number
of different factors are at play and the empirical evidence on this issue is very mixed (Lucas
forthcoming 2005).

Although remittances may have a positive effect on poverty within a country, the impact of
remittances on inequality between countries is problematic. As already noted, there are initial
costs of migrating, and the very poor countries tend to have a low number of migrants for this
reason. Indeed, since lower middle-income countries often have relatively high rates of
emigration, compared to the poorest countries (sometimes called the ‘migration hump’), these
countries stand to gain the most from remittances while very poor, low-emigration countries
stand to miss out on the potential economic benefits of migration. This can partly explain
why the share of global remittances received by Sub-Saharan Africa is so low. Although
low-income countries can arguably be said to need migrant remittances the most, they do not
receive nearly as much as middle-income countries. Therefore, remittances cannot be viewed
as an adequate substitute for ODA.

However, even considering the arguments above that remittances spent on consumption
generate multiplier effects and that it is difficult to distinguish between consumption and
investment, there is still a concern about the effect of remittances on long-term growth. In
particular, remittances, which are unearned income for the recipients, might have a negative
impact on the incentive to find work in the domestic economy. Levitt (1996: 7) studied the
impact of migration to the US on a town in the Dominican Republic. Remittances sent by
parents to their children still at school reduced the incentive to do well at school, since the
remittances was seen as a guarantee that there would be a continuous flow of income. In
addition, although the town relied on agriculture, young people were unwilling to become

19
agricultural workers and were planning to emigrate to the US to a life of easier work with
higher financial rewards.

Based on a study of Mexican villages with a large number of migrants in the village
population, Ojeda (2003: 13) argues that there is a causal relationship between the way that
remittances are spent and a continuous flow of migrants from Mexico to the US: low
productivity in agriculture in Mexico results in low local wages and pressure to migrate to the
US; when remittances are spent on consumption goods, there is greater chance of inflation
that will precipitate lower returns to investment in productive uses; this in turn reduces the
likelihood of new jobs being created, which then increases the pressure to migrate to the US.

Hence, in the long run, the impact of remittances on the effort level and quantity of workers
can potentially have a negative effect on economic growth. Chami et al. (2003) use panel-
data from 113 different countries to test the assumption that the disincentive to work from
remittances will indeed have a negative impact on economic growth. They find a negative
correlation between the size of remittances and economic growth, and this should be a cause
for concern. However, as already noted above, remittances can increase during times of
economic hardship, since they are based on altruism from the migrant, and this could also
explain the direct relationship between remittances and negative economic growth. Whether
or not the disincentive effect does indeed exist is difficult to examine, especially at a global
level since this is likely to be determined by a number of factors such as cultural influence,
family structures and gender of the sender and recipient.

The majority of remittances are spent on consumption goods such as food, clothing and health
care. Money is also spent on improving housing and buying durable consumer goods, but
generally only a small part is spent on investing in productive uses. One survey suggests that
only 6 per cent of remittances sent home by the African diaspora is reinvested (Banjoko
2004). Investment spending tends to have large ‘multiplier effects’ since money spent on
productive use will increase the number of jobs and output of the economy, and hence this has
been a matter of concern to many policy-makers. Although multiplier effects are also present
in consumption spending, it is generally thought to be smaller, and this is particularly the case
if remittances are spent on imported goods or if the increased demand for consumption leads
to inflation because it is not matched by an increase in supply. Thus, if migrant remittances
can somehow be geared towards productive investment, the economic gains are likely to be
greater (van Doorn 2002).

However, remittances are private funds and as argued by the HCIDC (2004:60):

‘we should be wary of using our ideas about what constitutes productive and
unproductive expenditure as a template for assessing decisions made by poor
households in desperate situations…the recipients of remittances are in a better
position than we are to make rational decisions based on the risks and
opportunities they face’.

In addition, goods which are usually viewed as consumption goods such as food and
education can also be considered investment, which will increase human capital. Hence, the
insistence on using classifications of goods used in economics can distort the picture and
make the economic impact of remittances appear less positive. Importantly, it is difficult,
especially at this early stage, to assess the long-term effect on the usage of remittances. As
an example, if remittances are spent on food, and this allows schoolchildren to focus more on

20
schooling and increases their ability to concentrate during class time, it can have a profound
effect on human capital in the future.

Given these parameters, two policy options emerge:

• Remote control. Asymmetric information between remitter and recipient often means
that remitters cannot control whether their money is being spent most productively
(Chami et al. 2003:4). The remitter may be led to believe that remittances are actually
saved rather than being spent on consumer products and handouts to needy relatives. 10
By harnessing new opportunities offered by technologies such as the Internet we may be
able to correct some of these asymmetries. More control by the remitter may be a worthy
goal in itself but may also be a better way of ensuring that remittances are used
effectively. There may also be options for offering remitters the opportunities to purchase
goods or services directly (e. g. health insurance (Carling 2004:6) or even groceries
(supermarkets in countries like Peru, Argentina and Senegal allow emigrants to shop
online for goods that are then delivered to families in their country of origin (AMAP
2004)).

• Matching funds. Although migrants will spend remittances in such a way that their
families benefit the most, this can coincide with aims of government policies to invest in
education, healthcare and improved housing. One way to do this would be through
governments in remittances-receiving countries (or even the governments of the countries
where migrants work) promising ‘matching-funds’ to induce migrants to spend some of
their remittances on productive or worthwhile areas. For example, a government could
promise to match any remittance spent on school fees in the country of origin with extra
investment in that particular school. This would provide an incentive to the migrant since
government funding would increase, hopefully improving the quality of education (not
just for their sponsor but all pupils) and perhaps even improve the esteem that the migrant
is held in by the community. It would also mean that the local area could benefit from
migration regardless of whether multiplier effects are present to distribute income. From
the discussion above, it is also clear that information plays a crucial role. Governments
and NGOs should be encouraged to provide information for migrants and their families on
how best to spend remittances such that long-term benefits are gained.

• Assign money to support the entrepreneurial activities of returning migrants, perhaps on a

semi-commercial basis. A venture capital fund, administered by and for returning
migrants may help provide much-needed seed capital and also facilitate a social and
commercial network. The IADB has recently funded such a project for workers returning
from Japan to Brazil (IADB 2004: 25).

Optimising diasporic potentials

The involvement of migrant communities, or ‘diasporas’ more generally, in the life of their
erstwhile homes goes well beyond financial flows. Diasporas can be the source of ideas,
behaviours, identities, and social capital that flows between countries (something Levitt
(1996:3) describes as ‘social remittances’). Similarly, migrants can transfer knowledge and
skills (sometimes called ‘technological remittances’) or even political identities and practices
(which Golding (2004: 805) calls ‘political remittances’).

21
Despite the potential significance of diaspora involvement, many mainstream development
institutions such as NGOs and governmental organisations have hitherto been reluctant to
engage with diaspora communities. The HCIDC (2004:68) encourages increased cooperation
between the diaspora and other players and argues that ‘diaspora organisations must not be
seen as marginal players in international development; rather, the Government, DFID, the
private sector and mainstream NGOs should work harder to involve them more fully’.

In recent years, there has been great optimism in the potential for what diaspora organisations
can do but, unfortunately, there seems to be very little systematic evidence of what role
diasporas do play. Below are three options for how diaspora involvement can be optimised.

Box 6. Examples of diaspora engagement

Africa Diaspora Investment Forum: UK-based forum aimed at promoting investment into
African economies, including through a new venture capital fund for SMEs in Africa.
http://www. africadiaspora. com

Homecoming Revolution: a not-for-profit NGO set up to attract diaspora members back to

South Africa and facilitate their resettlement. Provides a range of advisory and other services.
http://www. homecomingrevolution. co. za

Jamaican Diaspora Canadian Foundation: set up with the help of Jamaican government to
mobilise Jamaicans in the areas of law enforcement, development, education and health.
http://www. jamaicandiaspora. org/articles/0030. html

Kenya Development Network Consortium: aims to create development and business

opportunities and includes The Kenyan Edge Inc, an investor-owned organisation.
http://www. kdnc. org/

Tamil Eelam Economic Development Organisation: 20 year-old multinational network

assisting development projects in Northeast Sri Lanka, both during the civil war and since the
ceasefire. http://www. teedor. org/

Recommendation 8. Encourage collective remittances

Although the majority of remittances are sent by individuals, a small part is sent by collective
diaspora organisations, which can be linked by place of origin or religious or political beliefs.
A common type of diaspora organisation is a Hometown Association (HTA), which is
organised around links to a specific area such as a village. Despite the modest economic base
of these organisations, there is evidence that villages connected to HTAs tend to have better
roads, electricity and employment opportunities (Soerensen 2004:17). Hence, these
associations should be encouraged, and one way of doing so is to provide ‘matching funds’,
which, by adding to the funds already being spent by HTAs, provide an incentive to save and
invest. ‘One example is a project in the Mexican state of Zacatecas, where each dollar
contributed in remittances is matched by three dollars (one from the municipality, one form
the state and one from the federal government)’ (Van Doorn 2002). As an indication of the
success of this programme, it will be extended to cover the whole of Mexico (Goldring
(2004:802) quoting Amador (2002)). Matched-funding programmes could also be carried

22
out by NGOs or governmental organisations such as DFID or USAID. This would also
strengthen the links between these players, which was called for in the above section.

Box 7. A diaspora university in Somaliland

‘The development of the University of Hargeisa in Somaliland was a project spearheaded by
the UK Somali community from that region. Against all odds and to much national and
international acclaim, the newly developed University of Hargeisa (UoH) in Somaliland
opened its doors in 2000 to the first batch of access course students in preparation for a full
start in September 2000. Initiated in mid-1997, this effort united Somalis in Somaliland itself
with Somalis in the diaspora as far-flung as Australia, Sweden, Kuwait, the United States, and
Britain. The project enjoyed support by the government of Somaliland, a territory still
without international recognition. A steering committee in London that combined Somali
expertise and leadership with British know-how and experience worked in close collaboration
with an interim council in Somaliland. Local businesses in Somaliland took full
responsibility for rehabilitating the government-donated dilapidated old-school building that
was in fact home to over 500 returned Somali refugees. Somalis in Sweden provided 750
chairs and tables; Kuwait-based Somalis sent computers. In the project's second year, the
Somaliland Forum, a cyberspace-based global network of Somalis formed taskforces to tackle
specific elements, raised money, maintained email groups, and hosted real-time e-
conferences…
UoH threw a brain drain into sharp reverse. One-third of the students on the access course
returned from the Gulf, the UK and Canada to attend. High school students who would either
have had to leave Somaliland to pursue further studies or drop out now have the option to
stay. The university's first vice chancellor, an eminent Somali scientist who worked in
Canada for a number of years, took up post to work pro bono to oversee UoH's crucial first
few years’ (Chikezie 2000).

In addition, communication between the diaspora organisations and the government in the
country of origin should also be promoted. As an example, the government could help to
identify specific small-scale projects, which could then be sponsored by the diaspora.
However, as Van Hear et al. (2004:22) write, diaspora members ‘are primarily interested in
the advancement of their own particular group or sectional interest’. As a consequence,
policy-makers must be not loose sight of the impact on equity and equality in the country
supported by the diaspora. As an example, if most migrants have come from a particular
region of the country, relying heavily on the diaspora for both organisation and contacts can
lead to aid assistance exacerbating inequality in e. g. infrastructure and services.

Recommendation 9. Facilitate diaspora trade and investment

Lucas (2004:3) argues that migrants can promote both trade and investment in their country of
origin. Firstly, they are better informed about trading and investment opportunities and
secondly, they are able to enforce contracts through a network of contacts at home. This can
be important in countries where a legal framework for conducting business is missing. Some
migrant-sending countries do indeed offer very good opportunities for migrants to become
investors. As an example, a case study of foreign-born professionals in Silicon Valley
showed that 76 per cent of Indian and 73 per cent of Chinese immigrants would consider
starting a business in their country of birth, and a majority of them cited the availability of
skilled labour as one of the most important reasons for doing so (Saxenian et al. 2002).

23
This finding highlights the fact that social ties are not enough to ensure that migrants choose
their country of birth as a target for investment. More generally, in order to attract
investment, a healthy business environment is important (Soerensen 2004:24). This includes
adequate physical and financial infrastructure and a sound legal framework. A skilled labour
force with low labour costs is also an important asset. This means that low-income countries
with many obstacles to investment such as poor infrastructure and a lack of an adequate legal
framework and skilled workers will struggle to make migrant workers in the developed world
invest in their home country. As highlighted before, low-income countries that are already
severely disadvantaged will be in a poor position to make use of migration and the consequent
linkages as a strategy for development.

Even for countries that can offer an attractive business environment, other obstacles remain.
Despite the information and contacts possessed by migrants, many of them lack the
experience of entrepreneurship and knowledge of business methods. Hence, training
programmes and business counselling could be a useful addition in trying to use the potential
of migrants. However, Puri and Ritzema (1999:24-5) cite the case of such a programme
launched in Sri Lanka in 1982 that showed that

…return migrants belonging to higher occupational categories are suitable for an

orientation program of this type leaving out unskilled return migrants. Second,
the possibilities for guiding candidates into business are limited unless
accompanied by measures to facilitate the access to capital. Third, the ability to
identify and develop a project, as well as managerial skills needed to run a
business, cannot be imparted only through a program of class instruction
(Ibid:25).

Hence, a better approach might be to encourage migrants to invest indirectly, through savings.
Presently, many migrants may keep their savings in bank accounts outside their home
country, especially if overseas banks can offer higher interest rates on savings or a lower risk
of inflation and exchange rate devaluations. If saved in the migrant-sending country, they
could provide foreign currency and funds for investment. Various policies are described,
which have been developed in order to try and encourage migrants to save in their home
country (Ibid:19-22):

• Mandatory remittances: Migrants are required to send a certain percentage of their

savings to their country of origin. This has only been successful in the case of
Korean workers, who are mainly employed overseas by Korean companies for a
limited period of time. As this is difficult to replicate, such a policy cannot be
recommended.
• Foreign currency accounts and foreign currency-denominated bonds: These options
pose no risk of exchange rate devaluations and can be offered in combination with an
attractive premium over world interest rates, which can attract savings by migrants.
However, as the authors note, ‘a major limitation of foreign currency accounts as a
tool for stimulating remittances is that they, by their very nature, are attractive only
to migrate belonging to professional and higher-skilled categories who earn relatively
higher incomes’ (Ibid:21). It is also important to consider whether the premium that
has to be paid to attract savings in this way is worth paying.

The case for providing micro-credit as a tool for economic development has already been
presented in relation to remittances. However, another approach, which involves this type of

24
finance, would be to encourage permanent migrants, who do not remit but who have both
savings and the intention to contribute to economic development in the home country, to
invest in micro-finance institutions. Such a scheme would make use of the intention to both
help the country while generating a return on investment.

Recommendation 10. Consolidate knowledge networks

Networks of highly-skilled workers have been suggested as being one way of addressing
brain strain. Meyer (1999) calls this the ‘diaspora option’ and examines several networks
between nationals working mainly in science or engineering. The Internet is the main tool of
communication for these networks and perhaps due to difficulties of organising a network
before the use of the Internet became available, many of these networks are very recent. This
also means that the empirical evidence on the effect of these networks on the economic
development of the country of origin is very scarce. Nevertheless, attempts at projects in
which expatriates and the national community cooperate have been made. These include
research projects, technology transfer and expert consulting. Examples of initiatives to
encourage cooperation between the diaspora, countries of origin and development institutions
include the UNDP’s Transfer of Knowledge Through Expatriate Nationals (TOKTEN)
programme, IOM’s Migration and Development in Africa (MIDA), and World Bank/IOM
initiatives on return of qualified Afghans. TOKTEN is currently running successfully in 35
developing countries (Van Hear et al. 2004:28). On a smaller scale, individual diaspora
communities are often actively involved in these sorts of networks. Within the Sri Lankan
Tamil community, there have been important efforts to facilitate short-term diaspora return in
the post-ceasefire context in that country (for example, the Tamil Eelam Consultancy House -
http://www. techcanada. org/).

Despite the potential gain from these networks, it is important to keep in mind that not all
countries are equally well poised to gain from these.

‘It seems that for technology transfer through migrants and their networks to
be a significant factor in economic development at home, three basic sets of
conditions must hold: the migrants must be employed in sectors, occupations
and countries that grant access to useful information; a knowledge network
must emerge in some form, permitting transfer of that information; the home
country must be in a position to take advantage of the new information’
(Lucas 2004:15).

Therefore, even if there is a potential economic benefit from knowledge transfers,

very poor countries are usually not in a position to make use of this.

Challenges ahead

It might be useful at this stage to outline some of the major challenges facing researchers and
policy-makers in this area. This is by no means an exhaustive list; it is simply an attempt to
flag up some key themes that need exploring.

25
Anticipating future flows

Demographic transition in developed countries throws the question of migration and

development into sharp relief, and highlights most clearly the importance of mutuality in this
area. Yet, one of the key questions is just what future flows of people will look like.

Given demographic booms in the low-wage sending regions compared to high-wage regions,
and the likelihood of continued real wage gaps between sending and receiving regions,
international migration is likely to remain a feature of the global economy (Hatton &
Willamson 2003). Under conditions of successful development, poverty eradication and
growing equality between rich and poor countries, we might expect that any poverty
constraint on potential emigrants from the poorest parts of the world to ease, thus leading to
greater migration. On the other hand, if levels of inter-country inequality do not diminish, the
demand for emigration is likely to remain in place and needed to be kept in check by tighter
controls guarding entry into the richer parts of the world. International migration is likely to
play an increasingly important role regardless of which scenario emerges.

It is also a feature that is likely to result in greater global per capita consumption growth than
further increases in the mobility of factors such as capital (IMF 2004: 164-7). As initiatives
such as Mode 4 of the General Agreement on Trade in Services (GATS) take shape, more
labour mobility may be inevitable. There are also likely to be larger flows between
developing countries, especially as faster-growing countries suck in workers from slower-
growing neighbours.

In a world in which economic and demographic characteristics point strongly towards greater
international migration, immigration policies remain the important determining variable
shaping where, when and how large actual flows are. Given that links between migration and
development are likely to remain important for some time, getting these policies right will be
incredibly important.

Promoting versus restricting mobility

A key choice facing policy-makers in designing such policies is whether to promote or restrict
mobility. The apparently obvious solution to the development challenges thrown by
migration – and one frequently espoused by ‘compassionate racists’ who want to ‘help’ the
developing world – is to place severe restrictions on the movement of highly-skilled people
from developing countries. But simply closing the door or clogging the drain could have
deeply troubling implications for the human rights of the people involved, might not be
effective, would limit the benefits that immigration can deliver for the receiving country
(especially where migrant workers are instrumental in delivering critical services in that
country) and may deny some countries access to the positive impacts to be gained from
emigration (e. g. remittances and investment flows from diasporas).

A more sustainable response to these challenges should instead start from the recognition that
migration can have positive impacts for those who move, the societies they move to and even
the societies they leave behind. Managing migration then becomes less about limiting
mobility and more about optimising mutual impacts (making sure that migration fuels growth
in all countries involved while addressing brain strain in the sectors of countries where it is
occurring or likely to occur).

26
A crucial guiding principle in optimising the migration-development nexus should be about
promoting more real choices for migrants (and potential migrants) rather than limiting choice.
This is a common theme that links almost all the positive approaches in this area. For
example, it is better to give potential migrants a real choice between staying in their country
of origin rather than feeling compelled to leave for economic (or political) reasons. Similarly,
it is better for migrants to have a real choice between staying on in their current destination or
returning home (rather than feel that they have no future in their country of origin). When it
comes to remittances, opening up more choices for transfer will lead to more efficiency and
more choices in where money can be invested productively will lead to better utilisation of
the potential for remittances. Checking whether policy interventions promote or restrict
choices for individuals concerned might be just the sort of simple but effective principle upon
which good policies in this area can be built.

What apparatus?

For policy-makers, one of the critical decisions in this area is what sort of framework and
institutions can best address the challenges of optimising the impact of migration on
development and poverty reduction. The most obvious answer is that we need global
approaches to tackling them.

There is much scope for creative approaches in this area, especially given that migration is
one of the few areas of international public policy in which all (migrants themselves, sending
countries and receiving countries) stand to gain if managed appropriately. While some of the
policies recommended in this report and elsewhere could be pursued bilaterally, international
coordination will be essential in overseeing such things as the identification of brain strain
hotspots, creating and implementing restrictions on recruitment from those hotspots, and
(ideally) financing the required activities.

Some form of international research, regulatory and financial apparatus for coordinating these
efforts would seem appropriate. At the very least this could be some form of policy network
that included, at the very least, relevant UN agencies and programmes, the IOM, the Bretton
Woods institutions and major regional organisations. More ambitiously, such an apparatus
could be a formal partnership not just between developed and developing country
governments but also including the private sector, trades unions, migrant organisations and
civil society. Such partnerships are exactly the sort of creative and effective initiatives that
will be required to achieve the MDGs and other global policy challenges.

Who pays?

Assuming that extra funds are needed, at the very least, to address brain strain or, more
ambitiously, to fund a formal apparatus for research and policy on migration and
development, the question of who pays arises. It would seem that the notion of direct
compensation to developing countries for the loss of skilled workers only has the support of a
few and, if initiated, would likely only ever occur at a bilateral level and in particular sectors
(the World Health Organization has suggested that a compensation scheme might be
appropriate for migrant health workers). That said, other, more creative ways of funding
policy interventions in this area might be receive more universal support and, in any case, be
more effective. Here, at least three key questions emerge:

27
• Who pays - migrants themselves, their employers in host countries, the host government,
or some combination of these three?
• Who benefits - the migrant’s country of origin or at-risk developing countries in general;
sending country governments or non-governmental organisations; particular sectors of the
sending country’s economy (e. g. education)?
• What channels are used - direct transfers to sending countries or indirect administration of
funds collected by host country or intermediary institutions?

An extra tax on migrant workers that is then repatriated to the country of origin seems at first
to be an attractive proposition, most famously articulated by Jagdish Bhagwati (1976).
However, in general, requiring migrants themselves to pay extra (over and above any private
remittances they send) to their country of origin as compensation for lost skills should be
avoided for several reasons:

• extra tax burdens on migrant workers may create disincentives for migrants to migrate,
work legally once they migrate, and/or declare their income;
• mandatory remittance schemes implemented by the Philippines, Thailand and Pakistan
among others, have fared badly due to difficulties of implementation (Puri and Ritzema
1999:19); and
• taxing some workers higher than other could be in conflict with constitutional law (Martin
2003:28).

A more viable alternative might be for migrants to be taxed at the same rate as nationals of the
country, but for part of that revenue to be repatriated to either sending-country governments
or to fund particular development-related initiatives. In many countries (e. g. Netherlands,
Denmark, Sweden) temporary workers pay a lower tax rate than residents, partly in order to
attract foreign workers, but also partly because it is recognised that they are less of a drain on
public expenditure in areas such as child care and pensions.

Conclusion

The relationship between migration and development is at once incredibly significant but also
marginal in many ways. It is significant for all the reasons outlined above. Yet, it would be
naïve not to recognise that the relationship is also somewhat marginal. Development is
marginal to migration: when we consider all the various aspects of managing migration, the
development impacts on countries of origin is rarely going to be as prominent a research or
policy issue as issues such as immigrant integration, migrant rights and the economic impacts
of immigration. Similarly, migration is marginal to development: of all the factors that shape
the development potential of an economy, migration is almost always going to be relatively
insignificant.

That said, it is clear that migration has the potential to shape significantly the economic
development of some sectors in some countries. It is clear from the growing evidence in this
area that whether the impacts of migration are positive or negative will depend very much on
the context and how the situation is managed. Much will depend on the rate and nature of
emigration. The complexity of migratory flows as noted at the very beginning mean that no
one set of policies will be universally applicable. Table 5 shows a tentative sketch of what
sort of policies might be appropriate for particular contexts.

28
Table 5. Managing threats and opportunities for developing sending countries

Permanent emigration Temporary emigration

High- Low-skill/wage High- Low-skill/wage
skill/wage skill/wage
Need to
Ensure Ensure key
maximise
depleted sectors are not
Brain strain N/A skills-
skills are depleted
acquisition
replaced through limits
while abroad
High
potential for High total
Low and Highest
remitting but remittance
Remittances falling potential potential for
may fall off potential but at
for remittances remittances
after some lower amounts
time
Can lead to
High Ensure return
brain gain of Potential for
potential, migrants use
Diaspora/ emigrants gain skills and
especially skills and
Circulation skills while information
through experience
abroad and transfer
circulation gained abroad
circulate

While it is important to recognise that all four combinations of these streams (e. g. temporary
high-skill; permanent high-skill etc. ) are likely to remain important into the future (albeit to
different degrees for different countries), policy-makers face a choice of shaping the flows
themselves and/or shaping the impacts of these flows. Here, the key question that emerges is
whether migration should be both focus and locus for policy-makers. That is, while the
impacts of migration on development may be worth understanding, it is not a given that
appropriate policy interventions are best aimed at shaping migratory flows themselves.
Though it may seem counterintuitive, it may actually be more efficient and straightforward to
use levers such as trade, aid or foreign investment to correct for any adverse impacts of
migration rather than, say, tamper with migration flows themselves (given that greater
mobility is almost always a good thing). This is an area that deserves exploring further.

Some countries have recognised the potential held by migration while others risk losing out
or, worse still, being adversely affected. One of the most striking yet under-researched
aspects of the relationship between migration and development is the potential for migration
to affect inequality between individuals and at a global level. While politically-stable,
middle-income countries with a growing economy and financial infrastructure stand to benefit
immensely from international migration, the very poorest countries stand to lose the most and
gain the least from migration (brain strain, low return migration potential, poor environment
for the productive use of what remittances flow in etc. ). This differential impact of migration
may compromise the ability to pursue poverty reduction and achieve the Millennium
Development Goals (MDGs). Importantly, where migration leads to the depletion of key
health and education workers in developing countries, their ability to make progress on
critical MDGs may be compromised.

29
On the other hand, cooperation between states to manage migration may be an exemplar of
the sort of partnership that will be required to achieve the MDGs and other global policy
challenges. Individual states often see migration in terms of local impacts and national
interests. However, the global nature of flows means that managing the mutual impacts of
migration will need a robust supranational framework. Moreover, managing the process for
the mutual benefit of developed and developing countries will require partnership between
states.

Box 8. A Migration-Development Index (MDI)?

Akin to the UNDP’s hugely successful Human Development Index (HDI), perhaps there is
scope for a similar index to measure the development impacts of migration. While such a
measure could never capture the complexity of what is involved (in the same way that the
HDI does not capture all that human development is about), it could be a useful tool for
researchers and policy-makers.

A Migration-Development Index (MDI) could be a composite index that sought to capture the
scale and net economic impacts of migration. Analysis could include the net flows of
migrants, human capital, remittances and other financial transactions for each country
included and assess the impact of each on income, poverty, inequality and other development
outcomes. While such a measure would require improvements in the quality and
comparability of data across the world, the primary challenge would initially be
methodological.

It is hoped that this paper has highlighted three fruitful exchanges that need to be pursued
further. First, in research there needs to be more intellectual exchange between the growing
field of ‘migration studies’, to which the issue of development in countries of origin is
somewhat marginal, and development economics, to which the impact of migration is, at best,
marginal. There is much to be done in order to present a comprehensive empirical picture of
the nexus between migration and development. Global organisations such as the United
Nations may have a unique and important role to play in generating good quality, comparable
data and analysis that can inform policy-making.

Secondly, in policy-making, it is clear that there needs to be more effective dialogue between
arms of governments that do not traditionally work together very often. In the developed
world, this means departments responsible for home or immigrant affairs and those
responsible for international development.

Finally, in terms of international cooperation, there is also a need for more productive
partnerships between developed and developing countries to address areas of mutual interest
in the management of migration. This is likely to require a considerable overhaul of the
normative and institutional infrastructure of multilateral migration management. This may be
some way off, but action on shared development objectives may be one relatively easy step
forward. Given the global and interdependent scope of the challenges, it would seem wise to
have global and interdependent approaches to tackling them.

30
REFERENCES

Adams W (ed) (1968) The Brain Drain, New York: Macmillan

Adams R and Page J (2003) International Migration, Remittances, and Poverty in Developing
Countries, Washington, D. C. : World Bank
http://econ. worldbank. org/files/31999_wps3179. pdf

Africa Recruit (2003) Submission to the HCIDC, London: HCIDC

http://www. parliament. the-stationery-office. co. uk/pa/cm200304/cmselect/cmintdev/uc79-
v/uc79m03. htm

Alburo F A and Abella D I (2002) Skilled Labour Migration from Developing Countries:
Study on the Philippines, International Migration Papers 51, Geneva: ILO

Amador L (2002) ‘Mexico amplia el programa Tres por Uno. El gobierno destina dinero
para acelerar las obras sociales auspiciadas por paisanos de EU’ La Opinion, 13 February.

Accelerated Microenterprise Advancement Project (AMAP) (2004) Migrant Remittances

Newsletter 1(1) June

Asian Development Bank (ADB) (2004) Key Indicators: Poverty in Asia: Measurement,
Estimates, and Prospects, Manila: ADB

Banjoko T (2004) African Diaspora: Survey on Investment, London: Africa Recruit

http://www. africadiaspora. com/downloads/ADIF%20Survey-
%20Mobilising%20the%20Diaspora. ppt

Beine M, Docquier F and Rapoport H (2003) Brain Drain and LDCs’ Growth: Winners and
Losers, IZA Discussion Paper No. 819, Bonn: Institute for the Study of Labour (IZA)
ftp://repec. iza. org/RePEc/Discussionpaper/dp819. pdf

Bhagwati J (1976) ‘The brain drain’ in International Social Science Journal, Vol. 28, No. 4,
pp, 691-729, UNESCO.

Bloch A (2005) Understanding the links between the Zimbabwean diaspora and home,
Geneva: IOM
http://www. iom. int//DOCUMENTS/PUBLICATION/EN/MRS17_Zimbabwe_lr. pdf

de Bruyn T and Kuddus U (2005) Opportunities to improve the wider impact of remittances
in Bangladesh, Geneva: IOM
http://www. iom. int//DOCUMENTS/PUBLICATION/EN/MRS%2018%20Bangladesh_ebk.
pdf

Business Week (2003) Can Western Union Keep on Delivering? December 22

http://www. businessweek. com/magazine/content/03_51/b3863153_mz035. htm

Carling J (2004) ‘Policy Options for Increasing the Benefits of Migration’, Paper presented at
Ninth International Metropolis Conference, Geneva

31
Carrington W J and Detragiache E (1999) How Extensive is the Brain Drain? In Finance and
Development, June, 36(2)

Chami R, Fullenkamp C and Jahjah S (2003) Are Immigrant Remittance Flows a Source of
Capital For Development? , Washington, D. C. : IMF

Chikezie C (2001) Supporting Africa’s regional integration, African Knowledge Networks

Forum, 17-18 October, Addis Ababa.

Commander S, Kangasniemi M and Winters L A (2003) The Brain Drain: Curse or Boon?
IZA Discussion Paper No. 809, Bonn: Institute for the Study of Labour (IZA)
ftp://repec. iza. org/RePEc/Discussionpaper/dp809. pdf

Commander S, Chanda R, Kangasniemi M and Winters L A (2004) Must Skilled Migration

Be a Brain Drain? Evidence from the Indian Software Industry, IZA Discussion Paper No.
1422, Bonn: Institute for the Study of Labour (IZA)
ftp://repec. iza. org/RePEc/Discussionpaper/dp1422. pdf

Ellerman D (2003) Policy Research on Migration and Development, World Bank Policy
Research Working Paper 3117, August, Washington D. C. : World Bank
http://www. economics. ucr. edu/seminars/fall04/10-13-04. pdf

Faini R (2002) ‘Migration, Remittances and Growth’, paper presented at the Conference on
Poverty, International Migration and Asylum, Helsinki, September
http://www. wider. unu. edu/conference/conference-2002-3/conference%20papers/faini. pdf

Harrison A, Britton T and Swanson A (2004) Working Abroad: the benefits flowing from
nationals working in other economies, Paris: OECD
http://unstats. un. org/unsd/tradeserv/TSG%2009-04-Paris/Working%20abroad. pdf

House of Commons International Development Committee (HCIDC) (2004) Migration and

Development: How to make Migration Work for Poverty Reduction, Sixth Report of Session
2003-4, Vol. 1, United Kingdom: House of Commons.

Hussain M (2005) Measuring migrant remittances: From the perspective of the European
Commission, Paris: UN DESA Statistics Division
http://unstats. un. org/unsd/tradeserv/TSG%2002-05-Paris/tsg0502-5. pdf

Goldring L (2004) ‘Family and Collective Remittances to Mexico: A Multi-dimensional

Typology’ in Development and Change 35(4): 799-840

Hatton T J and Williamson J G (2003) What Fundamentals Drive World Migration?,

Discussion Paper 2003/23, Helsinki: WIDER
http://www. wider. unu. edu/publications/dps/dps2003/dp2003-23. pdf

Igartua G V (2004) The Temporary Mexican Migrant Labor Program in Canadian

Agriculture, working paper No 90, San Diego: University of California

32
Inter-American Development Bank (IADB) (2004) Sending Money Home: Remittances to
Latin America and the Caribbean, Washington, D. C. : IADB
http://www. iadb. org/mif/v2/files/StudyPE2004eng. pdf

International Monetary Fund (IMF) (2004) World Economic Outlook:

The Global Demographic Transition, Washington, D. C. : IMF
http://www. imf. org/external/pubs/ft/weo/2004/02/index. htm

International Organization for Migration (IOM) (2003) World Migration 2003: Managing
Migration: Challenges and Responses for People on the Move, World Migration Report
Series, Vol. 2, Geneva: IOM.

International Organization for Migration (IOM) (2004) The Millennium Development Goals
(MDGs) And Migration, New York: UN/POP/PD/2004/18
http://www. un. org/esa/population/ publications/POPaspects/IOM_POPMDG18. pdf

Kapur D (2004) Remittances: The New Development Mantra, G-24 Discussion Paper Series
29, Geneva: UNCTAD

Levitt P (1996) Social Remittances: A Conceptual Tool for Understanding Migration and
Development, Working Paper Series (96. 04),Cambridge, MA: Harvard.

Lowell B L (2002) Policy Responses to the International Mobility of Skilled Labour,

International Migration Papers 45, Geneva: ILO

Lowell B L, Findlay A and Stewart E (2004) Brain Strain: Optimising highly skilled
migration from developing countries, Asylum and Migration Working Paper 3, London: ippr.

Lucas, R E B (forthcoming 2005) International Migration Regimes and Economic

Development, Study commissioned by the Expert Group on Development Issues (EGDI),
Swedish Ministry for Foreign Affairs and Edward Elgar

Martin P (2003) Highly Skilled Labor Migration: Sharing the Benefits, Geneva: ILO
http://www. ilo. org/public/english/bureau/inst/download/migration2. pdf

Meyer J B and Brown M (1999) Scientific Diasporas: A New Approach to the Brain Drain,
MOST Discussion Paper 41, Geneva: UNESCO
http://www. unesco. org/most/meyer. htm

Nyonator F, Dovlo D and Sagoe, K (2004) The Health of the Nation and the Brain Drain in
the Health Sector. Paper presented at the UNDP Conference on Migration and Development
in Ghana, 14-16 September .
http://www. undp-gha. org/Document/MD%20CONF%20-
%20PAPER%20BY%20NYONATOR%20DOVLO%20SAGOE. pdf

Ojeda R H (2003) Transnational Migration, Remittances and Development in North America:

Globalization Lessons from the OaxaCalifornia Transnational Village/Community Modeling
Project, Washington, DC. : IADB.
http://www. iadb. org/mif/v2/files/Hinojosa2003MEeng. pdf

33
O’Neil K (2003) Using Remittances and Circular Migration to Drive Development,
Washington, D. C. : Migration Policy Institute.
http://www. migrationinformation. org/Feature/display. cfm?ID=133

O’Neil K (2004) Labor Export as Government Policy: The Case of the Philippines,
Washington, D. C. : Migration Policy Institute.

Orozco M (20030 Worker Remittances in an International Scope, Washington, D. C. : Inter-

American Dialogue Working Paper
http://www. iadialog. org/publications/country_studies/remittances/worldwde%20remit. pdf

Pieke F, Van Hear N and Lindley A (2005) Synthesis Study: Informal Remittance Systems in
Africa, Caribbean and Pacific countries, Oxford: COMPAS
http://www. compas. ox. ac. uk/publications/papers/Synthesis%20050115. pdf

Psacharopoulos G (1985) ‘Returns to education: A further international update and

implication’, Journal of Human Resources, 20(4): 583-604

Puri S and Ritzema T (1999) Migrant Worker Remittances, Micro-Finance and the Informal
Economy: Prospects and Issues, Working Paper 21, Geneva: ILO

Ratha D (2003) ‘Workers’ Remittances: An Important and Stable Source of External

Finance’, Global Development Finance report 2003, Washington, D. C. : World Bank
http://www. worldbank. org/prospects/gdf2003/gdf_ch07_web. pdf

Saxenian A with Motoyama Y and Xiaohong Quan (2002) Local and Global Networks of
Immigrant Professionals in Silicon Valley, San Francisco: Public Policy Institute of
California.
http://www. ppic. org/content/pubs/R_502ASR. pdf

Soerensen N N (2004) The Development Dimension of Migrant Remittances, Copenhagen:

Danish Institute of Development Studies.
http://www. iom. int/documents/publication/en/remittances_development_dimensions. pdf

Stillwell B, Diallo K, Zurn P, Vujicic M, Adams O and del Poz M (2004) ‘Migration of health
care-workers from developing countries: strategic approaches to its management’, Bulletin of
the World Health Organization, 82: 595-600

Straubhaar T (2000) International mobility of the highly skilled: Brain Gain, Brain Drain or
Brain Exchange, HWWA Discussion Paper 88, Hamburg.

Suro R (2003) Latino Remittances Swell Despite US Economic Slump,

Migrationinformation, Washington, D. C. : Migration Policy Institute.
http://www. migrationinformation. org/USfocus/display. cfm?ID=89

United Nations (2004) World Economic and Social Survey 2004, New York: UN DESA,
http://www. un. org/esa/policy/wess/

Van Doorn J (2002) ‘Migration, remittances and development’ in Labour Education, 4 (129),
pp 48-55

34
Van Hear N, Pieke F and Vertovec S (2004) The Contribution of UK-based Diasporas to
Development and Poverty Reduction, Oxford: COMPAS

Wehrfritz G and Vitug M (2004) ‘Philippines: Workers of the World’, Newsweek, 4 October.

Winters L A (2002) The Economic Implications of Liberalising Mode 4 Trade. Paper

prepared for the Joint WTO-World Bank Symposium on ‘The Movement of Natural Persons
(Mode 4) under the GATS’, WTO, Geneva, 11-12 April.
http://www. wto. org/english/tratop_e/serv_e/symp_apr_02_winters_e. doc

World Bank (2004) Global Development Finance, Washington, D. C. : World Bank

http://www. worldbank. org/prospects/gdf2004/

Wyss B (2004) ‘Global Poaching: Jamaica’s Brain Drain’ Econ-Atrocity Bulletin, 30 January.

35
Endnotes

1
See, e.g., http://www.cdr.dk/ResTHEMES/conflict/migdevfinal.htm;
http://www.migrationinformation.org/issue_jun03.cfm;
http://www.iom.int/en/what/migration_and_development.shtml; http://www2.gtz.de/migration-and-
development/english/; www.publications.parliament.uk/ pa/cm200304/cmselect/cmintdev/79/79.pdf
2
Editorial from Times of India,18 December 2002
3
Current initiatives to collect better, more comparable data on remittances include an Inter-Agency Remittance
Task Force, co-chaired by World Bank and DFID UK, an ‘Action Plan’ by G8 countries, and Technical
Subgroup on the Movement of Natural Persons – Mode 4 by the UN Statistical Commission.
4
US-Mexico remittances hit record high in 2004 totalling US$15 billion approximately, at US$327 per
transaction on average.
http://www.el-universal.com.mx/pls/impreso/noticia.html?id_nota=8201&tabla=miami
5
Suro (2003) describes how Latin American migrants use informal channels because they do not have the
necessary information about the banking sector or because they are unable to open bank accounts due to lack of
minimum income or identification papers.
6
http://www.iadb.org/NEWS/Display/PRPrint.cfm?PR_Num=87_04andLanguage=English
7
http://www.iadb.org/NEWS/DISPLAY/WSView.cfm?WS_Num=ws10204andLanguage=English
8
See, e.g., http://www.adb.org/Documents/Events/2004/Private-Sector-Dev/securitization.pdf
9
Faini (2003:7) shows that remittances decline as the proportion of migrants with a tertiary education increases.
10
According to a study cited in Wehrfritz and Vitug (2004), the typical domestic worker in Saudi Arabia will
remit most of her salary. Her husband then uses the money to buy consumer products and to support the
extended family, and often the overseas worker only discovers that nothing has been saved on her return.

36