al employment evidence on internal and international ... · evidence on internal and international...

Ru

Ra

l em

plo

ym

ent

a joint product of Fao’s Statistics Division and Social policies and Rural Institutions Division

©FA

O /

Seba

stia

n Li

ste

– N

OO

R

KnoWleDGe mateRIalS

Evidence on internal and international migration patterns in selected African countries

Figure 1 | Number of African migrants in the world, by destination

Source: UNDESA database.

To the world

1990 1995 2000 2005 2010 2015

To Africa To Europe To North America

0

5 000 000

1 0000 000

1 5000 000

2 000 0000

25 000 000

30 000 000

35 000 000

IntroductionMigration is a complex phenomenon and a fundamental component of structural transformation in developing countries. Migration patterns vary substantially across regions and countries, and flows have changed

considerably over time. Many households consider migration as a strategy to improve their livelihood, minimize their risks and diversify their income sources.

Key messages• Migration is widespread in Africa: a large share of households – in the countries for which data are available –

have at least one member who migrated, either internally or externally.• Internal migrants mostly originate from rural areas, while international migrants mainly come from urban areas.• Migrants are predominantly male, aged between 15 and 34, and households with migrants show relatively

higher levels of education. • Employment is the main driver of migration. Education and family reasons are the second and third most

common reasons. Women migrate mainly for family-related reasons, contrary to what happens in the rest of the population.

• Households with migrants – especially with international migrants – are wealthier than households without migrants. Their higher expenditure power might be either a consequence of migration (due to migrants’ remittances) or the reason why migration was possible in the first place (only better-off families could afford the costs associated with the movement).

35 000 000

30 000 000

25 000 000

15 000 000

20 000 000

10 000 000

5 000 000


2

Table 1 | Internal and International migrants

Country HH with migrants, 100 = all HH

HH with internal migrants, 100 = all HH

HH with international migrants, 100 = all HH

HH with migrants, 100 = rural HH

HH with internal migrants, 100 = rural HH

HH with international migrants, 100 = rural HH

Burkina Faso 49% 19% 31% 49% 18% 32%

Ethiopia 7% 5% 2% 7% 5% 2%

Ghana 5% NA NA 7% NA NA

Kenya 65% 35% 37% 66% 37% 36%

Malawi 19% 12% 7% 16% 9% 7%

Nigeria 36% 29% 7% 29% 25% 4%

Senegal 55% 30% 32% 59% 36% 32%

South Africa 12% 10% 3% 22% 19% 4%

Uganda 48% 42% 11% 47% 44% 8%

Africa has a long history of internal and international migration, with migrants moving within and beyond the continent. In 2015, the UNDESA estimated that 14 percent of international migrants1 in the world originated from the continent. In the same year, African countries were the estimated origin of 12 percent of the international migrants in Europe. Figure 1 shows how the number of African international migrants by destination evolved between 1990 and 2015. The stock of African migrants in the world increased by 46 percent up to a total of 32 million people. Roughly half of them (16 millions) remained within the continent, corresponding to 80 percent of all African migrants in 2015. These data show the importance of the Africa-to-Africa migration corridor, which is the fifth largest in the world. Despite the growth of migratory flows in the continent and the increased political relevance, information on flows, and particularly on internal migration, is limited, albeit being the most dominant flow. Of particular importance is data that can shed light on socio-economic conditions associated with migration in rural and urban areas, connected to adequate breakdown by age, sex and origin of emigrants. This type of information is seldom available.

This note summarizes findings on outmigration patterns of selected African countries obtained from two groups of household surveys, both provided by the World Bank. In Nigeria, Senegal, Uganda, Burkina Faso, Kenya and South Africa the note relies on data from the “Migration and Remittances Household Surveys” (MRHS2), which focuses

specifically on migration. In Ethiopia, Malawi, Nigeria3 and Ghana, instead, evidence is drawn from the “Living Standard Measurement Study” (LSMS4).

Main findings5

Characteristics of the migratory flows

From where to whereInternal migration is more frequent thaninternational migration.

Migration is a highly diversified phenomenon that shows substantial variability across countries. The percentage of households with migrants in the analysed countries varies from 5 percent in Ghana, to 65 percent in Kenya. The share of rural households with migrants is similar in most countries, with the exception of South Africa, where theshare of rural households with migrants is twice as large as the national figure6. Internal migration is more widespread than international migration in most of the countries considered, with the exception of Senegal, Kenya, and Burkina Faso. For households in rural areas, internal migration flows are more common than international migration, in every considered country, except in Burkina Faso (Table 1). International migration is usually expensive, and this can be one reason for its lower frequency. Furthermore, migration abroad is often the result of a dynamic process, whereby people decide first to move within their own country to raise the necessary funds to pay for the long-distance migration.

Source: FAO’s Statistics Division.


3

Figure 2 | Share of rural and urban households with internal and international migrants, 100 = all households with internal and international migrants

Kenya Malawi SenegalNigeria UgandaSouth Africa

From rural From urban

%

0

20

40

60

80

100

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Internal migrants mostly originate fromrural areas. International migrants, instead, come mostly from urban areas.

More than 50 percent of the households reporting at least one internal migrant are located in rural areas, with the exception of Nigeria, where the share is still 48 percent. The highest share of households with internal migrants living in rural areas is observed in Uganda, which is relatively less urbanized country compared to others in this pool, such as Nigeria and South Africa, where we observe otherwise.

International migration from rural areas is less common, as people may face more constraints than in urban areas. Travel is longer, and the required economic resources to migrate abroad are likely to be significantly higher than those required for migrating internally. Malawi shows the highest share of rural households among the households with international migrants (74%). Indeed, this is a country with high prevalence of rural areas. On the other hand, the smallest share of rural households with international migrants is observed in Nigeria.

Figure 3 below illustrates internal migratory flows from rural and urban areas. As expected, urban areas are the preferred destinations, for both migrants originating from rural and urban areas. This is the case in all countries considered. The share of people moving from rural to urban areas in the total migrants varies from 40 percent in Nigeria to 55 percent in South Africa. This difference can be the outcome of distress factors that are often more pronounced in rural areas, such as poverty and food insecurity, limited availability of decent job opportunities and climate change.

The share of people who migrate from urban to rural areas is usually limited in all the countries considered, while people movements from rural to other rural areas are more common. In particular, in Burkina Faso and Uganda, 36 percent and 33 percent of internal migrants, respectively, move from a rural to another rural area. Exceptions, in this respect, are South Africa and Kenya, where the percentage of migrants moving from an urban to a rural area is higher than in other countries.

Sources: Data from MRHS(*) and LSMS-ISA(**) surveys processed by FAO, Statistics Division, and from the RuLIS(***) project.


4

Figure 3 | Movements of internal migrants by origin and destination. 100 = total migrants from both urban and rural areas

Figure 4 | Proportion of age groups among migrants, national and rural

Total Rural Total Rural Total Rural Total Rural Total RuralBurkina Faso Ghana SenegalNigeria Uganda

%

0-14 15-34 35+

0

20

40

60

80

100

35

61

4

36

60

4

10

69

21

8

70

22

23

65

12

28

59

13

40

56

4

36

59

5

26

69

5

26

69

5

Burkina Faso Kenya SenegalNigeria UgandaSouth Africa

To rural To urban

%

0

20

40

60

80

100

From urbanFrom ruralFrom urbanFrom ruralFrom urbanFrom ruralFrom UrbanFrom RuralFrom UrbanFrom RuralFrom urbanFrom rural

Who migrates?

Migrants are most likely to be male and between 15 and 34 years old. 60 to 70 percent of all migrants are aged between 15 and 34 years. Young people are the most dynamic part of the population and are more prone

to move in search of new employment opportunities and better livelihood options. The share of children among migrants is similar in rural and urban areas.

In most countries considered, women migrate less frequently than men do, both internally and internationally (Figure 5). Female migrants are the majority only in two cases: among international migrants from Ethiopia, and

among internal migrants in Malawi. The percentage of women that migrate from rural areas is slightly lower than that of women who migrate from urban areas, especially in Senegal.

Source: Data from MRH surveys processed by FAO, Statistics Division.



5

Figure 5 | Percentage of women who migrate within and outside the country

Malawi SenegalNigeriaKenyaEthiopia UgandaSouth Africa

%

From rural areas National level

Burkina Faso

0

10

20

30

40

50

60

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Inter

natio

nal m

igran

ts

Inter

nal m

igran

ts

Drivers of internal and international migration

Reasons for migration are mainly work-related, especially from rural areas. Family and education are the next most common reasons. Women migrate mainly for family-related reasons.

In all analysed countries7, except in Ethiopia, the most common reason to migrate is the search of better employment opportunities (Figure 6). This is the case for both internal and international migrants. In rural areas the

percentage of internal and international migrants leaving their households in search of a better job is higher than in urban areas. This reflects the limited access to social protection and the discrepancy of decent employment opportunities between urban and rural areas in many developing countries. Professional reasons are relatively more prominent for international migrants, with the exception of Kenya, where the percentage of people migrating for work related reasons is higher among internal migrants, especially in rural areas.

Figure 6 | Reasons for internal and international migration

Total Rural Total Rural Total Rural Total Rural Total Rural

Ethiopia Nigeria UgandaSenegal Kenya

%

Work Education Family Other

0

10

20

30

40

50

60

70

80

90

100

Inter

natio

nal

Inter

nal

Inter

natio

nal

Inter

nal

Inter

natio

nal

Inter

nal

Inter

natio

nal

Inter

nal

Inter

natio

nal

Inter

nal

Inter

natio

nal

Inter

nal

Inter

natio

nal

Inter

nal

Inter

natio

nal

Inter

nal

Inter

natio

nal

Inter

nal

Inter

natio

nal

Inter

nal

47

26

0.8

78

2

1

27

21

113

1

2

27

30

4

17

24

2

23

28

312

24

3

18

12

110

616

422

7

1

30

17

4

13

15

6 4

30

18

49

7

15

9

68

5

14

16

65

211

39

48

212

13

73

39

38

50

67

8374

93

49

6651

85

2819

42

5647

61

Sources: Data from MRHS and LSMS-ISA surveys processed by FAO, Statistics Division, and from the RuLIS project.



6

Family reasons for migration include family reunion, marriage and divorce. Together with education, family reasons are the second most important driver of migration. Few people migrate for other reasons, such as health, security, and as a consequence of natural disasters8. In Kenya and Nigeria, a high percentage of migrants is moving to study, both within the country and internationally. Uganda and especially Ethiopia, instead, show a higher share of migrants for family-related reasons compared to other countries.

Sex disaggregated data also reveal interesting findings (Table 2). In most of the selected countries, women migrate primarily for family reasons, especially within their own country. After marriage, women are more likely to

relocate to where the husband is living than the other way around. In rural areas, the percentage of women migrating for work-related reasons is higher compared to urban areas. Data for younger age classes show that family-related reasons for internal migration are more important in Ethiopia, Nigeria and Uganda compared to other countries. As it happens for the entire the main driver of migration is work (Table 2). Kenya shows a slightly different pattern compared to the other countries, with a higher share of international young migrants moving for education-related reasons, from both rural and urban areas.

Table 2 | Reasons for internal and international migration

Ethiopia Nigeria Senegal Uganda Kenya

Total Rural Total Rural Total Rural Total Rural Total Rural

Inte

rnal

Inte

rnat

iona

l

Inte

rnal

Inte

rnat

iona

l

Inte

rnal

Inte

rnat

iona

l

Inte

rnal

Inte

rnat

iona

l

Inte

rnal

Inte

rnat

iona

l

Inte

rnal

Inte

rnat

iona

l

Inte

rnal

Inte

rnat

iona

l

Inte

rnal

Inte

rnat

iona

l

Inte

rnal

Inte

rnat

iona

l

Inte

rnal

Inte

rnat

iona

l

Work

Total 28 19 51 85 42 56 47 61 67 83 74 93 49 66 49 68 65 47 72 50

Female population 20 44 29 58 17 39 20 41 17 32 19 45 28 62 29 48 66 40 57 42

Young population (15-34) 46 72 58 92 40 52 43 54 61 74 71 87 49 66 48 66 65 40 74 46

Education

Total 26 2 21 1 30 24 28 24 12 6 7 2 17 15 18 9 16 39 13 38



Family

Total 47 78 27 13 27 17 23 12 18 10 16 4 30 13 30 15 14 11 12 9



Other

Total 0 1 1 1 2 4 2 3 3 1 2 1 4 6 4 7 6 3 3 3



Source: FAO Statistics Division.


7

Figure 7 | Employment status before and after migration from rural areas

Kenya Nigeria South Africa Uganda

Employed Self-Employed Unemployed Student Housewife Other

%

0

1

2

3

4

5

6

7

8

After migrationBefore migrationAfter migrationBefore migrationAfter migrationBefore migrationAfter migrationBefore migration

The comparison of migrants’ employment status before and after migration shows that individuals who moved away from rural areas have found better employment opportunities. Figure 7 shows this in Nigeria, Kenya, South Africa and Uganda: both wage- and self-employment are higher after migration. The difference is particularly pronounced in Kenya and South Africa. In the same vein, unemployment rates in all the countries decreased significantly after migration, along with the share of students. Consistent with the higher percentage

of women migrating for family-related reasons, the proportion of housewives increases after migration, although the variation is moderate in Kenya and South Africa. Similar patterns emerge when looking at the employment status before and after migration from urban areas. The higher importance of work as a reason to migrate, compared to study, is also confirmed by Figure 7: the percentage of people in education is higher before than after migration in all the countries considered.

Socio-economic conditions of householdswith migrants

Households with internal and international migrants are wealthier and more educated than those without migrants, in both rural and urban areas.

Poverty is one of the key drivers of migration. However, the poorest often lack the resources required for migrating. Similar considerations have led to an intense debate on the relationship between migration and development. While job opportunities and economic development can dampen the reasons for migration, they can also stimulate migration, by providing the resources needed to move away from migrants’ original area.

In the countries analysed, households with at least one migrant are, on average, wealthier than households without migrants, as they show a higher consumption expenditures. This emerges both for rural and urban households. Households with international migrants are the wealthiest (Figure 89). This should not be considered an evidence of causality, as higher expenditures might actually be the consequence of migration: remittances from migrated relatives often play a large role in relaxing financial constraints. At the same time, migration itself is expensive, indicating that only relatively better off households can afford it.



8

Figure 8 | Average annual total expenditures by type of household (international dollars I$)

Burkina Faso Nigeria UgandaSenegal MalawiEthiopia

HH Without migrants HH With internal migrants HH With international migrants

0

1 000

2 000

3 000

4 000

5 000

6 000

7 000

8 000

9 000

Rural areasNational levelRural areasNational levelRural areasNational levelRural areasNational levelRural areasNational levelRural areasNational level

The same reasoning can be applied for education levels within a household. The head of households with one or more migrants have completed on average a higher level of education than heads of households without migrant-members. This association is even more pronounced in households with international migrants. Correlated with

income, this indicates that more educated households, who are often the wealthiest households, are more likely to send a member away to seek education or more rewarding economic opportunities elsewhere, as they have the financial means to do so.

Figure 9 | Highest level of education of the household head

National Rural National Rural National Rural National Rural National Rural

Malawi Nigeria South AfricaSenegal Uganda

%

Primary of lower Secondary Tertiary

0

10

20

30

40

50

60

70

80

90

100

Inter

natio

nal

Inter

nal

Non m

igran

t

Inter

natio

nal

Inter

nal

Non m

igran

t

Inter

natio

nal

Inter

nal

Non m

igran

t

Inter

natio

nal

Inter

nal

Non m

igran

t

Inter

natio

nal

Inter

nal

Non m

igran

t

Inter

natio

nal

Inter

nal

Non m

igran

t

Inter

natio

nal

Inter

nal

Non m

igran

t

Inter

natio

nal

Inter

nal

Non m

igran

t

Inter

natio

nal

Inter

nal

Non m

igran

t

Inter

natio

nal

Inter

nal

Non m

igran

t

3

21

75

10

35

56

4

32

65

1

16

83

5

29

66

2

23

75

17

21

62

21

28

51

27

35

38

7

13

80

11

23

66

29

21

49

14

33

53

8

45

47

6

47

47

26

74

2

16

82

110

89

8

14

78

113

86

24

44

32

29

89

113

86

19

62

19

12

14

74

13

15

72

34

15

51

7

13

81

8

14

77

19

17

64




9

Figure 10 | Percentages of households receiving remittances

%

Rural National

0

10

20

30

40

50

Burkina Faso Senegal EthiopiaUganda MalawiGhana

46 17 5 35 442

Figure 10 presents information on households receiving remittances, at national level and in rural areas. Also households without migrants but receiving remittances from non-household members have been included in the computation. This percentage varies from the 46 percent

in Senegal to 4 percent in Malawi. With the exception of Senegal, the share of rural households that receive remittances is, on average, similar to that of urban households. Households that receive remittances are relatively wealthier (Figure 11).

Figure 11 | Average level of remittances received by households and annual total expenditures (values expressed in international dollars I$)

Burkina Faso Ethiopia GhanaUgandaSenegal Malawi

I$

Rural National

0

1 000

2 000

3 000

4 000

5 000

6 000

7 000

8 000

9 000

10 000

Tot.

Exp

(HH w

ith re

m)

Tot.

Exp

(HH w

ithou

t rem

)

Avera

ge re

mitt

ance

s

Tot.

Exp

(HH w

ith re

m)

Tot.

Exp

(HH w

ithou

t rem

)

Avera

ge re

mitt

ance

s

Tot.

Exp

(HH w

ith re

m)

Tot.

Exp

(HH w

ithou

t rem

)

Avera

ge re

mitt

ance

s

Tot.

Exp

(HH w

ith re

m)

Tot.

Exp

(HH w

ithou

t rem

)

Avera

ge re

mitt

ance

s

Tot.

Exp

(HH w

ith re

m)

Tot.

Exp

(HH w

ithou

t rem

)

Avera

ge re

mitt

ance

s

Tot.

Exp

(HH w

ith re

m)

Tot.

Exp

(HH w

ithou

t rem

)

Avera

ge re

mitt

ance

s

Sources: Data from MRHS and LSMS surveys processed by FAO, Statistics Division, and from the RuLIS project.

Sources: Data from MRHS and LSMS surveys processed by FAO, Statistics Division, and from the RuLIS project. the RuLIS project.


10

Figure 12 | Percentage of households owning an house, by type of holding

SenegalNigeria UgandaKenya

%

From rural areas National level

0

20

40

60

80

100

HH with

at le

ast o

ne m

igran

t

HH with

out m

igran

ts

HH with

at le

ast o

ne m

igran

t

HH with

out m

igran

ts

HH with

at le

ast o

ne m

igran

t

HH with

out m

igran

ts

HH with

at le

ast o

ne m

igran

t

HH with

out m

igran

ts

91 68 74 94 95 75 8376

Figure 12 shows that the percentage of families owning a house is similar or slightly higher in the group of households with at least one migrant. But for Senegal and Nigeria the differences are negligible. Overall, the ownership of a house tend to be more common in rural areas compared to the national level. A similar message emerges looking at the percentages of households with and without migrants owning land, derived from the MRHS (Figure 13). Most households in the surveyed

countries, between 51 percent in Senegal and 96 percent in Burkina Faso own land. As expected, in all countries considered, land ownership is more frequent in rural areas. The differences, except in Senegal and Kenya, are small though as also many urban dwellers own land. Differences in land-owning between households with and without migrants are negligible in most countries, indicating that remittances and migration are not a defining factor for owning land or not.

Figure 13 | Percentage of households owning land, by type of holding

Nigeria SenegalKenya UgandaBurkina Faso

%

Rural National

0

20

40

60

80

100

HH with

out m

igran

ts

HH with

at le

ast o

ne m

igran

t

HH with

out m

igran

ts

HH with

at le

ast o

ne m

igran

t

HH with

out m

igran

ts

HH with

at le

ast o

ne m

igran

t

HH with

out m

igran

ts

HH with

at le

ast o

ne m

igran

t

HH with

out m

igran

ts

HH with

at le

ast o

ne m

igran

t

97 8876 88 89 89 82 89 9096

96

70

61

7976

5158

83 83

93




11

What is FAO’s role?

FAO has a key role to play in producing evidence on the relation between migration and rural livelihoods to support policy making at national and regional level. Migration is part of countries’ development. As rural economies undergo structural transformation, the movement of people towards urban areas or across international borders is inevitable. While large population movements from rural areas may generate large distress and deprive original areas from valuable manpower, they often correspond to improvements in resource allocation and in productivity increases.

In this framework, FAO works to support governments in designing and implementing evidence-based policies to accompany rural transformation processes. Better evidence on the determinants of migration and its consequences in rural areas is key to inform policy decisions. Today there are significant knowledge gaps on rural outmigration trends, which need to be tackled. This is particularly the case for migration driven by distress, when people do not perceive there is any other viable livelihood option except to migrate.

Reliable data, disaggregated by sex, age, origin and destination is necessary to understand socio-economic conditions associated with migration. At the moment, these data are scarce. Existing household surveys do not consistently include questions related to the drivers and impacts of migration in rural areas. Labour force surveys can help, but they may not adequately cover rural areas and agricultural-based livelihoods. Furthermore, these surveys are generally not suitable to assess circular and seasonal migration, which are typical of agriculture. Even when available, survey data are collected using different methodologies and questionnaires in different countries, which hinders international comparisons.

The FAO is developing an extensive work agenda to support policy making and more reliable information on the following topics: (i) rural migration dynamics, (ii) migration and the labour absorption capacity of the rural economy, (iii) migration, rural and agricultural livelihoods, (iv) migration and climate change, (v) migration and protracted crisis and (vi) migration and social protection. In this endeavour, FAO is collaborating and partnerships with other international organizations – such as the IOM and the World Bank, the Global Migration Group (GMG) and many research institutions.

At national level, FAO supports member countries in building their capacities to collect and analyse data. The organization could also assist countries in conducting labour market needs assessments to collect, process and disseminate data on rural labour markets. This will enable countries to design policies and mechanisms to better manage labour mobility, replace and/or reintegrate workers and facilitate the investments of remittances in on-farm and off-farm productive activities, thus harnessing the potential of migration for local development.

©FA

O /

Th

om

as H

ug

OR

[email protected]

www.fao.org/rural-employment


[email protected]

[email protected]

fao.org/rural-employment

fao.org/economic/ess

© F

AO, 2

017

1 The UNDESA estimates the international stock of migrants in a given country taking the number of foreign-born or foreign citizens in a country at mid-year. Estimates are computed every five years and time series for 1990, 1995, 2000, 2005, 2010 and 2015 are accessible at http://www.un.org/en/development/desa/population/migration/data/index.shtml

2 The considered MRHS are: 1) the Nigerian Migration Household Survey of 2009; 2) the Senegalese Migration and Remittances Household Survey of 2009; 3) the Ugandan Migration Household Survey of 2010; 4) the Burkinabe Enquête Ménage sur la Migration et les Transferts de Fonds of 2010; 5) the Kenyan Migration and Household Survey of 2009, and 6) the South African Migration and Household Survey of 2009.

3 In the case of Nigeria, we had access to both type of surveys: due to the different survey structure, these data complemented each other rather than being redundant.

4 The LSMS surveys selected are: the Ethiopian Rural Socioeconomic Survey of 2013-14; the Malawian Integrated Household Panel Survey of 2013; the Nigeria General Household Survey of 2012-2013, and the Ghanaian Living Standard Survey of 2013.

5 Caution must be exerted in assessing the data presented in this note. The starting point data used to compute the indicators presented in the Tables and Charts are derived from different surveys, which are

not harmonized and present different recall periods for migration and difference in the questions concerning migration. In most instances, household are asked to report on whether they have at least one ancient member who is currently living somewhere else. The recall period of LSMS-ISA surveys is variable, and coincides with the timespan from one wave to the next. Limited comparability is the reason why certain tables and charts do not report data for all the countries considered. More specifically, data for Kenya, South Africa, Ethiopia, Malawi and Ghana had to be computed without considering sampling weights. In the first two cases, the vector of the sampling weights was not included in any data file. For what concerns the three LSMS, their samples were not designed to estimate migration; as a consequence, they did not provide representative estimates at the national level. Hence the estimates only refer to the sampled households, and not to the entire population. Missing values for Ghana in Table 1 are due to the absence of a variable that specifies the destination of migrants. In the LSMS-ISA surveys for Ethiopia, Malawi and Nigeria migrants are identified by considering people who left the household from one wave to the next. This approach does not allow capturing households in which all members migrated between two waves.

6 It must be noticed that the surveys used in this note rely on national definitions of rural

areas, which are not homogeneous across countries.

7 The information on reasons to migrate was not available in the Malawian survey. We also decided to exclude Ghana and South Africa from the following graphs due to the high number of missing values.

8 The very low percentages of people migrating for security reasons may be due to the fact that household surveys are generally conducted only in safe areas of the countries, where it is possible to send survey enumerators. Furthermore, migration for security reasons is more likely to involve movements of entire.

9 We decided to exclude Malawi, Kenya and South Africa from this graph due to the poor quality of expenditure data.

(*) The Migration and Remittances Household Surveys (MRHS) are a series of surveys conducted for the Africa Migration Project jointly undertaken by the African Development Bank and the World Bank.

(**) Living Standard Measurement Surveys – Integrated Surveys on Agriculture (LSM-ISA), World Bank.

(***) The Rural Livelihoods Information System (RuLIS) is a joint FAO, World Bank and IFAD initiative aimed at gathering and standardizing rural income and livelihoods data from household surveys.

Endnotes

References Carletto, C., Larisson, J. and Özden, C. 2014. Informing Migration Policies: A Data Primer. Washington, DC, World Bank.

FAO. 2016. Migration, Agriculture and Rural Development. Addressing the root causes of migration and harnessing its potential for development. Rome (available at http://www.fao.org/3/a-i6064e.pdf).

World Bank. International Migrant Stock. URL: http://microdata.worldbank.org/index.php/catalog/mrs

World Bank. Living Standard Surveys (from Ethiopia, Malawi and Nigeria). URL: http://econ.worldbank.org/WBSITE/EXTERNAL/EXTDEC/EXTRESEARCH/EXTLSMS/0,,contentMDK:23512006~pagePK:64168445~piPK:64168309~theSitePK:3358997,00.html

World Bank. Migration surveys (from Nigeria, Senegal and Uganda). URL: http://microdata.worldbank.org/index.php/catalog/mrs

World Bank. 2006. World Development Report 2007 – Development and the Next Generation. Washington, DC, World Bank.

UN Department of Economic and Social Affairs, Population Division. 2016. International Migration Report 2015: Highlights (ST/ESA/SER.A/375). New York, U.S.A. (available at http://www.un.org/en/development/desa/population/migration/publications/migrationreport/docs/MigrationReport2015_Highlights.pdf).

I7468EN/1/06.17

mailto:gender%40fao.org?subject=

http://www.fao.org/gender

http://www.fao.org/economic/ess/en/

http://www.un.org/en/development/desa/population/migration/data/index.shtml



http://microdata.worldbank.org/index.php/catalog/mrs

http://www.fao.org/3/a-i6064e.pdf

http://www.fao.org/3/a-i6064e.pdf



http://econ.worldbank.org/WBSITE/EXTERNAL/EXTDEC/EXTRESEARCH/EXTLSMS/0,,contentMDK:23512006~pagePK:64168445~piPK:64168309~theSitePK:3358997,00.html




http://microdata.worldbank.org/index.php/catalog/mrs\

http://microdata.worldbank.org/index.php/catalog/mrs\

http://www.un.org/en/development/desa/population/migration/publications/migrationreport/docs/MigrationReport2015_Highlights.pdf



al employment evidence on internal and international ... · evidence on internal and international...

Documents