{Journals}sjpe/48_2/a198/makeup/a198.3d

Scottish Journal of Political Economy, Vol. 48, No. 2, May 2001 # Scottish Economic Society 2001, Published by Blackwell Publishers Ltd, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA

OVERSEAS WORK EXPERIENCE, SAVINGS AND ENTREPRENEURSHIP AMONGST RETURN MIGRANTS TO LDCS Barry McCormick and Jackline Wahba*

ABSTRACT This paper contributes to a small but rapidly growing literature concerned with the potentially substantial implications of international migration for economic development in LDCs. We study the linkages between overseas employment, savings and entrepreneurial activity on return. In an econometric model of the probability of entrepreneurial activity, we find evidence supporting the hypotheses that both overseas savings, and the duration of stay overseas increase the probability of becoming an entrepreneur amongst literate returnees to Egypt. Amongst illiterate returnees, overseas savings alone increase the probability of becoming an entrepreneur. The results for literates suggest that skill acquisition overseas may matter more substantially than overcoming a savings constraint in explaining how overseas opportunities influence entrepreneurship on return. For illiterates, who usually accept menial positions overseas that offer little opportunity for learning, the opposite obtains.

I INTRODUCTION The gradual integration of fragmented national labour markets into world markets provides unprecedented scope for workers to be employed Ðperhaps only temporarily Ðoutside their country of citizenship, in order to acquire new skills and ideas, and to accumulate savings at a faster rate than is possible in a low-income country. The `brain-drain' to high wage countries already provides the basis of widely discussed proposals to compensate LDCs for the loss of human capital using income taxation of migrant earnings Ðfor example, Bhagwati (1976). Our focus in this paper concerns another mechanism whereby the `brain-drain', and in particular its temporary nature, might contribute to economic development. We explore how overseas work experience in high wage countries enables return migrants to bring accumulated savings, new skills and ideas that might facilitate `modernisation' and economic growth in the origin country. In particular we examine two theories of how overseas employment might enable return migrants to provide an entrepreneurial contribution: first, *University of Southampton

164

{Journals}sjpe/48_2/a198/makeup/a198.3d

OVERSEAS WORK EXPERIENCE, SAVINGS AND RETURN MIGRATION 165

accumulated savings overcome domestic liquidity constraints, and secondly, that overseas work experience generates new skills and ideas. The possibility that return migrants from overseas bring additional experience and savings that might help to transform LDCs first received serious attention from sociologists concerned to understand rural social change. For example, Foster (1967) explores how the migration of Mexican Braceros to the US changed the living standards in their Mexican villages on their return, partly as a result of experiencing American culture, but primarily as a consequence of their accumulating US earnings. Gugler and Hannigan (1976) discuss the crucial role of return migrants in the social and economic transformation of West African villages. Lutifyya's (1966) study of return migrants to Jordan pointed out the implications of young people returning to their villages, after having worked in Western Europe and the US, for social norms towards marriage, extended family life, education, and the role of women, as well as economic undertakings. Nada (1990) and Reichert (1993) study the influence of temporary international migration on the socio-economic development of six villages situated in rural Egypt. Economists have shown only slight interest in the influence of international migration on the nature of economic change in the origin country, and this topic is not a focus of the development literature. It is unsurprising therefore that two recent major surveys of international migration, Razin and Sadka (1997) and Lalonde and Topel (1997) omit all references to the topic. Nevertheless, small theoretical and empirical literatures have emerged (unfortunately separate), the latter of which has begun to explore two channels whereby overseas migration may significantly alter the economic prospects of the origin countries. First, migrants may face capital market restrictions in the origin country so that overseas savings and remittances are subsequently able to fuel productive investments. Secondly, human capital or entrepreneurial ideas may be accumulated by working overseas, and which are used to start a business, or otherwise enhance earnings, when migrants return to their country. Given the low wages and market distortions prevailing in many LDCs, this accumulation might not have been possible without working overseas. Empirical analysis of these effects has been undertaken by numerous authorsÐ for example, Paine (1974), Lucas and Stark (1985), Adams (1991, 1998), Lucas (1985), Taylor and Adelman (1996), study the impact of remittances, and Escobar and Martinez (1990), Massey et al. (1987), Cornelius (1990), Durand and Massey (1992), Portes and Guarnizo, (1991), Lopez and Seligson (1991) study small business formation. More recently, Arif and Irfan (1997) and Ilhai (1999), have studied the impact of overseas savings on the occupational choice of return migrants to Pakistan. Insightful overviews of return migration and its development potential are given by Thomas-Hope (1999) for Jamaica, Diatta and Mbow (1999) for Senegal, and Rodriguez and Horton (1995) for the Philippines. There are only a few theoretical models of how international migration may influence welfare in the origin country which incorporate more than a `lost worker' effectÐ that is the consequences of the decline in the domestic labour stock. It had been thought that the effect of the `lost worker' on domestic # Scottish Economic Society 2001

{Journals}sjpe/48_2/a198/makeup/a198.3d

166

BARRY MCCORMICK AND JACKLINE WAHBA

welfare hinges solely on the presence of distortions in the domestic labour marketÐ for example, Bhagwati and Hamada (1974). However, recent analysis due to Rivera-Batiz (1982), has shown that in models with non-traded goods in the origin economy, the welfare of those remaining in the origin country is reduced by international migration due to the loss of trading opportunities involving non-traded goods, with those who have migrated. Subsequent analysis incorporating remittances has extended these models with non-traded goods, and shown that if remittances are sufficiently great then those remaining behind can benefit from overseas migration ÐDjajic (1986). It is also possible to show that if remittances per migrant are higher than domestic wages Ðas for example has been found for Egypt and Korea Ðthen endogenously determined equilibrium overseas migration may not only raise domestic welfare, but also be less than that which is socially optimal ÐMcCormick and Wahba (2000). Thus laissez-faire migration may no longer be socially optimal with remittances. However, in these models the origin country is affected solely by a `lost worker' and in some models, by remittances. No model to date has sought to capture the channels whereby overseas employment influences entrepreneurial activity. The purpose of this paper is to explore empirically the entrepreneurial activity of return migrants to Egypt, contrasting their situation with that prior to migration, and exploring the characteristics and overseas experience of returnees that engage in entrepreneurial activity. In particular we explore the extent to which returnees to Egypt become entrepreneurs and the influence on this process of overseas savings, overseas work experience, and pre-migration formal education. The evidence indicates how overseas migration to higher wage economies, for even comparatively short spells, can enable future entrepreneurs to accumulate financial capital on a scale not otherwise possible, and to accumulate new skills that are useful on return. We discuss how this development mechanism differs for the least educated and for those that were entrepreneurs before migrating. The economic contribution of return migrants in this sample supports the view that overseas work experience in a high wage country can play a useful role in the development process. Unlike most studies of return migration we use data that is drawn from the whole origin country, and do not, for example, limit our attention to certain rural areas. The data used in this study are drawn from the October 1988 special round of the Labour Force Sample Survey (LFSS), which was carried out by the Central Agency of Public Mobilisation and Statistics (CAPMAS) in Egypt. The 1988 included a supplementary survey module on workers who are return migrants, where a migrant refers to someone who spent a minimum of six months overseas and has been overseas only for employment purposes. This return migration module describes the main characteristics of 1526 returnees in the labour market before and after migration, in addition to details on migrationÐ country of destination, migration duration, and wages and savings whilst overseas. Section II provides an overview of international migration, remittances, and economic activity amongst return migrants that occurs in Egypt. In Section III we analyse the behaviourÐ both overseas and in EgyptÐ of return migrants who become entrepreneurs, distinguishing those who were not entrepreneurs # Scottish Economic Society 2001

{Journals}sjpe/48_2/a198/makeup/a198.3d

OVERSEAS WORK EXPERIENCE, SAVINGS AND RETURN MIGRATION 167

before international migration.1 A simple econometric model of the probability that a return migrant is an entrepreneur is analysed in Section IV. Finally, the main findings are summarised in Section V. II

INTERNATIONAL MIGRATION FROM AND TO EGYPT

In 1988 the Egyptian labour force was approximately 18
6 millions while another 1
78 millions were estimated to be working overseas, primarily in Saudi Arabia, Jordan, Iraq, Kuwait and Libya. These working migrants and others overseas (estimated at 185 000 non-working migrants) generated a considerable TABLE 1 The characteristics of returnees All returnees

Old entrepreneurs

New entrepreneurs

Mean of age in 1988 Male (%) Married in 1988

Individual characteristics 38
84 94
32 87
54

43
27 98
28 94
47

39
83 98
47 86
12

Illiterate Read and write Primary Preparatory Secondary University and higher

Education in 1988 29
70 18
06 9
56 20
10 4
25 18
30

57
14 27
65 9
68 1
84 0
46 3
23

27
35 20
00 13
88 16
73 3
27 18
77

Location 49
81 49
48

27
14 26
23

55
36 56
11

Overseas experience 2
56

2
50

4
90

Urban origin Urban destination Mean of years spent overseas Average monthly overseas savings (LE) Technical and scientific Management Clerical Sales Services Agriculture Production Beneficial to current job (%) Sample size

594

441

1093

Occupation overseas 20
96 0
49 3
70 4
16 8
35 13
75 48
58

3
65 0 1
74 3
71 4
94 26
05 59
91

15
28 1
33 4
49 6
42 8
46 16
77 47
25

Skills acquired abroad 36
12

18
67

47
97

217

246

1528

1 The influence of overseas work experience return migrants earnings is discussed by Co, Gang and Yun (2000).

# Scottish Economic Society 2001

{Journals}sjpe/48_2/a198/makeup/a198.3d

168

BARRY MCCORMICK AND JACKLINE WAHBA

direct impact on the Egyptian economy as a result of their foreign earnings that permitted a higher level of non-inflationary domestic expenditure than would otherwise have been possible. Since 1983 remittances have not dropped below $3
1billion and have reached as much as $6
1billion. The sums have reflected the ebb and flow of overseas migration, but throughout the period remittances have provided substantial resource flows, not considerably different to the value of merchandise exports, and in the order of 5± 11% of GDP. In the period 1974 ± 88, 1
1±1
2 million migrants are estimated to have returned to Egypt, with another 170 000 returning before 1974, so that about 7% of the labour force had returned from overseas work in 1988. Although we now explore micro-evidence, the scale of both return migration and remitted capital is such that the aggregate consequences of the activity of return migrants may be substantial. TABLE 2 The characteristics of returnees' work in Egypt All returnees Before migration

In 1988

Old entrepreneus Before migration

New entrepreneurs

In 1988

Before migration

In 1988

Waged Employer Self-employed Unpaid family worker Unemployed Out of labour force

Employment status 62
26 60
94 Ð 10
10 19
16 58
89 8
24 10
06 41
11 6
21 3
27 Ð 6
82 6
58 Ð 6
39 Ð Ð

Ð 69
79 30
21 Ð Ð Ð

72
93 Ð Ð 13
67 6
20 7
20

15
88 50
61 32
66 0
85 Ð Ð

Government Public enterprises Private Other

Sector of Employment 25
91 29
32 Ð 44
49 8
59 54
52 28
27 59
39 45
48 1
33 2
70 Ð

Ð Ð 100 Ð

21
21 47
72 27
67 3
40

9
53 2
35 87
08 1
07

Technical and Scientific Management Clerical Sales Services Agriculture Production

22
55 0
55 5
07 4
01 4
56 31
62 31
63

Occupation 23
07 2
32 7
13 6
72 6
69 28
27 25
80

2
91 Ð Ð 7
90 1
60 65
67 21
92

2
50 Ð Ð 10
78 1
55 66
01 19
15

17
38 0
58 3
47 5
26 6
22 26
10 40
99

12
74 2
93 0
19 17
60 9
12 24
10 33
32

Agriculture Manufacturing and Mining Utilities Construction Trade Transport Finance Services

32
61 15
95 0
51 11
45 7
02 6
93 0
80 25
53

Industry 28
90 13
73 0
88 8
06 10
28 7
93 1
02 28
58

65
24 11
10 0
29 3
37 10
69 6
06 0
56 2
69

66
01 9
83 0
29 2
46 12
76 4
05 1
04 3
56

28
67 22
46 0
43 12
77 9
16 5
56 0
27 20
70

24
23 13
53 Ð 6
73 23
38 11
86 3
40 16
87

Sample Size # Scottish Economic Society 2001

1528

217

246

{Journals}sjpe/48_2/a198/makeup/a198.3d

OVERSEAS WORK EXPERIENCE, SAVINGS AND RETURN MIGRATION 169

In our sample most Egyptian international migrants went to either Saudi Arabia or Iraq (65%). The mean overseas spell length was around 2
6 years. In the period up to 1983 the migration was primarily from urban areas but this was sharply reversed in the post-1983 era. However, while only 9
7% of migrants went to Libya and 3
7% to Kuwait, these workers spent relatively long spells overseas Ð 4
3 and 5
7 years, respectively.

The characteristics of return migrants Table 1 describes the characteristics of all returnees, those who were entrepreneurs prior to migrating, and those who were entrepreneurs only after returning from overseas. We have defined entrepreneurs to be those either selfemployed or employers. International migrants returning to Egypt are drawn from a wide spectrum of formal educational backgrounds as shown in Table 1. In 1988 about 18% were university graduates, but only 23% had remained at school beyond 14 years, and 30% were illiterate. Almost all of the returnees were male (94%) and had a mean age of 39. Since on average 5
5 years had passed since a worker had returned from overseas, the mean age on return was 33. The distribution of return migrants by both regional and rural=urban locations of residence was almost identical to that prior to migration, with very few migrants TABLE 3 Transitional probabilities of returnees by employment status and sector of employment Employment status in 1988

Employment status before migration

Waged

Unpaid Selffamily Employer employed worker Total

Waged Employer Self-employed Unpaid family worker Unemployed Unemployed-new entrant to labour market

82
91 9
34 17
88 26
98 67
79 88
00

9
61 85
56 27
13 25
70 22
94 7
20

7
07 4
31 53
65 8
79 2
66 3
85

0
41 0
79 1
34 38
52 6
61 0
95

100 100 100 100 100 100

Out of labour force

72
69

10
03

10
52

6
76

100

Sector of Employment in 1988 Sector of employment before migration Government Public enterprise Private Other

Public Government enterprise 82
39 8
98 6
82 14
72

2
20 15
54 3
26 12
47

Private 13
50 72
97 88
05 48
17

Other Total 1
91 2
50 1
87 24
64

100 100 100 100

Note: 1 The table should be read horizontally with the data providing the proportions of workers in 1988 in each employment status (column) that before migration occupied the status given in the row. # Scottish Economic Society 2001

{Journals}sjpe/48_2/a198/makeup/a198.3d

170

BARRY MCCORMICK AND JACKLINE WAHBA

TABLE 4 Earnings and savings of waged and self-employed returnees whilst overseas (In 1988 LE) Percent of return migrants 1. 2. 3. 4. 5. 6.

Average monthly wage abroad Less than 500 8
22 500 ±999 33
52 1000±1999 33
66 2000±2999 10
99 3000 ± 4999 7
44 5000 or more 6
17

1. 2. 3. 4. 5. 6. 7.

Average monthly savings while overseas None 32
29 Less than 100 3
81 100 ±499 31
55 500 ±999 16
25 1000±1499 9
65 1500±2999 3
25 3000 or more 3
20

1. 2. 3. 4. 5. 6.

Average monthly wage after return in 1988 Less than 50 3
37 50 ±99 22
28 100 ±149 24
97 150 ±249 27
26 250 ±499 17
00 500 or more 5
13

Note: 1 The table should be read horizontally with the data providing the proportions of workers in 1988 in each employment status (column) that before migration occupied the status given in the row.

not returning to their former region. Any concern therefore that international migration develops a taste for either `big-city' life or the facilities of urban areas, is not supported in these data. More noticeable changes occurred in our migrants' industry of occupation ÐTable 2. About 4% fewer worked in agriculture, and about 6% more in trade, transport and services. Nevertheless the migrants returned to broadly similar industrial patterns of employment. An examination of the public=private sector of work before and after migration suggests more striking changes. Whereas 44% of migrants had worked in public enterprise before migration only 9% did so on return, Table 2. In contrast, about one third of return migrants enter the private sector having previously not been employed there. This massive reduction in the public sector employment is an important aspect of understanding the determinants of international migration from Egypt. The proportion of employers in our sample rises from 10 to 19% between the pre-migration and post-return periods ÐTable 2. This increase arises largely because 10% of those who were waged workers prior to migration (waged workers comprise 62% of our sample prior to migration) become employers Ða significant transformation. This striking aspect of the behaviour of return # Scottish Economic Society 2001

{Journals}sjpe/48_2/a198/makeup/a198.3d

OVERSEAS WORK EXPERIENCE, SAVINGS AND RETURN MIGRATION 171

migrants, which we explore in more detail in the next section, is closely related to downsizing of public sector, and the need for young well educated public employees to secure private sector employment. Otherwise, the patterns of occupation and industry after return are very similar to those before migration, with a modest shift from agriculture and production. Table 3 documents how formally employed workers normally returned to their former labour market status, but that the transitions to self-employment and employer arise from various origins. Table 4 describes the monthly wage while abroad and after return, together with an estimate of monthly overseas saving, all in LE 1988. While overseas these migrants earned on average LE 1400 and saved on average LE 600Ða savings rate of approximately 40%. While this is a large proportion of the wages earned overseas, approximately one in three workers were unable to make any savings at all. The wedge between earnings overseas and that after returning to Egypt is considerable. Whereas only 8% of our sample earned less than LE 500 overseas, only 5% of our sample earned more than LE 500 on returning to Egypt. Uses of savings by migrants and their families have received a lot of attention. In our sample, as in many other studiesÐ see for example, Adams (1991), Nada (1990) and Fergany (1988) Ða large proportion (46%) of returnees invested in housing. However, of interest to this paper, is that 10% of returnees invested in economic projects. III

WHO ARE THE NEW ENTREPRENEURS?

We begin by contrasting characteristics of those returning from overseas who become entrepreneurs without having had previous entrepreneurial experience, with those who had entrepreneurial experience prior to overseas work, and with those who did not become entrepreneurs. We then construct an econometric model of the probability that an overseas worker becomes an entrepreneur allowing for the individual's history of entrepreneurship before working overseas. Table 1 gives an overview of the contrasting characteristics of those return migrants who undertook entrepreneurial work prior to migration, those who began entrepreneurial work only after returning to Egypt, and other returnees. The new entrepreneurs are on average four years younger and less likely to be married than the old entrepreneurs, but remain older than non-entrepreneuring returnees. The new entrepreneurs are slightly more likely to be urban rather than rural in both their origin and destination region whereas the old entrepreneurs are overwhelmingly rural in both background and region of return. About 22% of the new entrepreneurs are well educated having completed secondary or university education, whereas the old entrepreneurs only have 4% educated to this level and are primarily working in occupations where illiteracy is no bar. Even so, a considerable proportion (27%) of the new entrepreneurs were illiterate. The significant overall feature of those becoming entrepreneurs, following overseas migration, is that they are broadly drawn from all educational categories. The old entrepreneurs had on average spent longer spells overseas of 4
9 years relative to a mean figure for the old entrepreneurs 2
5 years. The new # Scottish Economic Society 2001

{Journals}sjpe/48_2/a198/makeup/a198.3d

172

BARRY MCCORMICK AND JACKLINE WAHBA

entrepreneurs were more likely to have been employed in a technical or scientific job overseas than the old entrepreneurs and less likely to have been employed in agriculture and production. Skills acquired overseas were judged to be beneficial by approximately one third of all entrepreneurs, and particularly beneficial to new entrepreneurs, amongst whom 48% judged the skills required abroad to be beneficial. (We return to this below.) Overseas average monthly savings amongst entrepreneurs amount to about six times the average post return wage in our sample. However, the importance of total savings is much greater for the new entrepreneurs who saved on average more than LE 50 000 in comparison to only LE 11 000 saved by the old entrepreneurs. We shall explore in our model below how far savings played a role in explaining the birth of new entrepreneurial activity amongst the return migrants. The picture of the pattern of work characteristics both before and after migration for the entrepreneurs in our sample is provided in Table 2. Amongst the new entrepreneurs on return to Egypt, 73% had been waged prior to migration, and 14% had been in unpaid family employment. In 1988 amongst the new entrepreneurs half employed others, one third were self-employed and 16% were waged workers with their entrepreneurial project activity an additional source of economic activity. Before migration 21% of the new entrepreneurs had been government employees, 48% in the public sector, and 28% in the private sector. In 1988 there remained a few entrepreneurs in the government sector who, having retained their waged employment, undertook economic projects as an additional activity. However the bulk of the new entrepreneurs were now unsurprisingly, in the private sector (87%). New entrepreneurs were more likely than the old to have a technical sales or services background whereas the old entrepreneurs are much more likely to have an agricultural background. IV AN ECONOMETRIC MODEL OF ENTREPRENEURIAL ACTIVITY AMONGST RETURN MIGRANTS In our study we have defined an entrepreneur to be either an employer, a selfemployed individual, or someone with a business project in addition to their usual economic activity. We have shown how entrepreneurship is related to working overseas with 29% of our sample being either employers or selfemployed in 1988 in comparison with only 18% before migrating overseas. Most of the return migrants who were entrepreneurs before migrating had little education. However, those beginning entrepreneurial activity for the first time, are a far more representative sample of the educational background of all overseas migrantsÐ Table 1. We have noted also the large proportion of the new entrepreneurs with university education who have had long spells overseas and have made substantial overseas savings. Our intention now is to construct a simple econometric model of the probability that a return migrant is an entrepreneur. We have two primary hypotheses. First that individuals who have made higher total savings whilst overseas are more likely to become entrepreneurs on return since for them the # Scottish Economic Society 2001

{Journals}sjpe/48_2/a198/makeup/a198.3d

OVERSEAS WORK EXPERIENCE, SAVINGS AND RETURN MIGRATION 173

opportunity costs of capital is less than for those who either must borrow in local capital markets or are liquidity constrained.2 Secondly, we conjecture that the length of time spent overseas will influence whether the returnee becomes an entrepreneur because the longer an individual spends overseas, total savings held constant, the greater the relevant work experience and opportunity for skill acquisition. We assume that the pay-off from becoming an entrepreneur is an unobserved variable y *, and that: y * ˆ 0 x ‡ ln S ‡ ln t ‡  where x is a vector of the individual and market level control variables, S is savings accumulated overseas, t is months of overseas work experience, and  is normally distributed error term with mean zero and variance one.3 Since we do not observe y *, only whether or not a returnee becomes an entrepreneur: y ˆ 1 if

y * > 0;

y ˆ 0 if

y * Æ 0:

We wish to investigate whether the rate of return to overseas savings and work experience varies with pre-migration general human capital. Given that 30% of our sample of returnees are illiterate, and illiteracy, I, may crucially limit the ability to benefit from working overseas, we allow:

ˆ 0 ‡ 1 I;  ˆ 0 ‡ 1 I: Our major hypotheses are that both overseas savings and work experience matter in explaining y *in the general population, thus ;  > 0. However, we recognise that the domestic returns to overseas work may be less or zero for illiterates, who often must take menial positions. Thus, we expect 1 < 0. The influence of overseas savings may also be less for illiterates if these workers are more risk averseÐhaving lower Egyptian incomes. However, if liquidity constraints are important in the LDC and these liquidity constraints are tighter for badly educated workers, then overseas savings will matter more for the less educated. We have explored our hypothesis in two different models. First, we have studied the probability that a returnee becomes an entrepreneur, allowing for pre-migration entrepreneurial history. Second, we have studied the probability that returnees become entrepreneurs upon return conditional upon not being entrepreneurs before migrating overseas. The results of estimating probit models of these two specifications, with a full set of education controls, but setting

1 ˆ 1 ˆ 0, are given in Columns 1 and 2 of Table 5. In Columns 3 and 4 we allow general human capital prior to migration to influence the impact of overseas savings and length of overseas work experience, 1 ; 1 6ˆ 0: In each 2

This is explored formally by Evans and Jovanovic (1989). Returnees have mean savings of LE 18, 248, but 30% report zero savings. For this latter group we set savings at LE 1. 3

# Scottish Economic Society 2001

{Journals}sjpe/48_2/a198/makeup/a198.3d

TABLE 5 Marginal effects of the probabilities of returnees being entrepreneurs in 1988 Entrepreneur1Entrepreneur2Entrepreneur1Entrepreneur2 (1) (2) (3) (4) Log total savings ( 0 )

Characteristics of Overseas Stay 0
015 0
014 (3
95) (4
31) 0
056 0
064 (3
42) (4
74) Ð Ð

Log months spent overseas (0 ) Illiterate *log total savings ( 1 ) Illiterate *log months spent abroad (1 ) Dummy = 1, if returned for less than 2 years

Previously entrepreneur Male Age

Less than primary Primary Preparatory Secondary University Higher diplomas Less than secondary Secondary or higher

Ð

Ð

0
011

0
029

0
017 (3
52) 0
082 (4
17) 0
005 (0
62) 0
086 (2
54) 0
012

(0
32)

(0
97)

(0
33)

Individual characteristics 0
540 Ð (14
90) 0
164 0
162 (2
15) (2
53) 0
004 0
001 (2
42) (0
98)

0
550 (15
38) 0
180 (2
39) 0
004 (2
56)

Education (ref.: none) 0
015 0
032 (0
39) (0
95) 0
039 0
047 (0
81) (1
21) 0
093 0
051 (2
13) (1
47) 0
187 0
124 (2
27) (1
95) 0
130 0
095 (2
65) (2
37) 0
091 0
071 (1
03) (1
06) Ð Ð Ð

Ð

Constant Sample size Log likelihood Tests of:3 H0 ˆ 0 ‡ 1 ˆ 0 H0 ˆ 0 ‡ 1 ˆ 0

Region of origin 0
027 0
009 (0
93) (0
39) 0
744 0
639 (7
22) (7
45) 1467 1213 653
42 519
86 Ð Ð

Ð Ð

(1
04) Ð 0
179 (2
79) 0
002 (1
36)

Ð

Ð

Ð

Ð

Ð

Ð

Ð

Ð

Ð

Ð

Ð

Ð

0
294 (3
42) 0
400 (4
27)

0
138 (1
84) 0
233 (2
89)

Employment characteristics before migration (ref.: Private sector) Government sector 0
160 0
126 0
187 (3
67) (3
74) (4
34) Public enterprise sector 0
032 0
023 0
026 (1
08) (0
91) (0
91) Urban

0
014 (3
68) 0
076 (4
75) 0
003 (0
48) 0
043 (1
50) 0
031

0
028 (0
99) 0
596 (5
23) 1467 651
48 3
85 0
02

0
146 (4
39) 0
018 (0
75)

0
012 (0
50) 0
573 (5
88) 1213 522
08 4
40 1
83

Notes: 1 Entrepreneur is defined as employer, self employed or operating a business venture using overseas savings. 2 This sample excludes those who were entrepreneurs prior to migration. 3 The  2 (1) statistics are reported in the table.  2 (1) at 5% significance level is 3.84. Notes: Returnees before 1970 are excluded. Absolute values of t-statistics are in parentheses. Marginal effects show the increment in the probability and are calculated at the reference set of individual charactertistics and sample means.

{Journals}sjpe/48_2/a198/makeup/a198.3d

OVERSEAS WORK EXPERIENCE, SAVINGS AND RETURN MIGRATION 175

model we have included a range of demographic characteristics capturing the age of the returnee in 1988, the educational level achieved by the returnee, gender, the urban or rural origin of the individual, and whether the individual was originally working in the government or public enterprise sector. In addition for the two models where we include the entrepreneurs prior to overseas migration, we allow for this pre-migration activity. In Columns 1 and 2 of Table 5 we provide probit estimates of and , the mean influence of overseas work experience and savings for all returnees. The results for both of the estimated models show that both total overseas savings and time spent overseas have positive and highly significant effects on being an entrepreneur. Using the estimates in column 2, the probability of becoming an entrepreneur, amongst non-entrepreneurs before migration, is enhanced at sample mean from 0
190 to 0
206 if the individual has savings of LE 30 000 rather than LE 10 000 (mean savings is about £18 000). The effects of length of the overseas work experience appear more substantial: an increase from 6 to 30 months of overseas work increase the probability of a non-entrepreneur on return from a mean of 0
19 to 0
29. Before commenting further on these marginal effects we will consider the interaction effects described in columns 3 and 4. We find that in both models 3 and 4, more savings has an influence on illiterates' entrepreneurial probability that is not significantly different for literates. Furthermore, we can reject (at the 5% level) the null hypothesis that

0 ‡ 1 ˆ 0, so that in each model we conclude that incremental total savings increases the probability of becoming an entrepreneur for both literates and illiterates, and by similar marginal rates. The influence of overseas work experience is not found to be significantly different from zero for illiterates in any of our models. We cannot reject the null hypothesis that 0 ‡ 1 ˆ 0. The converse implication of this is that the rates of return to overseas work experience are greater for the literates than were forecast for the returnees generally on the basis of columns 1 and 2 of Table 5. Thus, if overseas work is increased from 6 to 30 months we estimate an increase in the mean rate of literates becoming entrepreneurs, amongst those not entrepreneurs prior to migration, from 0
19 to 0
32. A similar, but plausible, increase in savings from LE 10 000 to LE 30 000 increases the probability that a literate becomes an entrepreneur by only 0
01 to 0
20. It may be argued that both savings, and time spent overseas, are chosen prior to migration as part of a plan to become entrepreneurs on returnÐfor example, Dustmann (1995). To address this possibility we test for the exogeneity of savings and overseas duration using the Smith-Blundell (1986) exogeneity test. In both cases we are able to accept the null hypothesis that the variables are exogeneous. (The observed  2 (1) = 0
95 in equation (1) for savings,  2 (1) = 2
14 for duration overseas, and  2 (2) = 2
19 for both savings and duration overseas.) The usefulness of overseas work experience We have found that the primary economic influence of overseas work on the probability of entrepreneurial activity amongst literate returnees arises from the # Scottish Economic Society 2001

{Journals}sjpe/48_2/a198/makeup/a198.3d

176

BARRY MCCORMICK AND JACKLINE WAHBA

period of time spent in overseas work rather than the consequences of the level of accumulated savings. Amongst illiterates the effect is through savings. There are three possibilities why overseas work experience is important for literates. First, work experience conveys skills that are useful on return to the origin country. Secondly, living overseas conveys ideas, experience with a more advanced commercial environment. Third, the correlation may reflect an unobserved characteristic: for example, those who work longer overseas may be more talented Ð securing offers of extended workÐand thus be more able to profit from entrepreneurial activity on return. To check that there exists an economic line of causation we have explored data which asks the following question: Have you benefited from your overseas work experience upon your return? If there is a link between duration overseas and domestic productivity then the proportion answering the question positively should increase with duration spent overseas. This we find is confirmed in the data: of those overseas for up to one year, 24% answered `yes'; of those overseas 1 ±3 years, 41% answered `yes'; of those overseas for longer, 59% answered `yes'. Thus there would appear support for the idea that longer work-spells overseas raise the likelihood that overseas skills will be useful on return. Turning to the other variables, we find, that previous entrepreneurs are about 55% more likely to become entrepreneurs after returning. The age of the returnee has only a small influence on the probability of becoming an entrepreneur: ten additional years results in less than half a percent higher probability of becoming an entrepreneur and marginally less likelihood of investing in a project; a young worker is as likely as an older one to become an entrepreneur on return.4 In Egypt, it is generally the case that the more educated are less likely to be entrepreneurs amongst return migrants than the less educated. This is also true amongst those who were not entrepreneurs prior to overseas migration. The role of public employment is an interesting one. Workers in the official government sector are perhaps unsurprisingly less likely to become entrepreneurs, but those in the government owned public enterprise sector, which was being downsized in the period, are not surprisingly less likely to become entrepreneurs as those in our reference group Ðprivate sector employees. Our earlier descriptions pointed to the higher incidence of entrepreneurship in the rural sectors on the Egyptian economy, and so it is interesting to explore how far amongst return migrants the probability of entrepreneurship is affected by the region from which the individual originated. The history of the individual as an entrepreneur prior to migration is allowed for either by an individual characteristic or by estimating the model with only those without previous entrepreneur experience. In both cases we find that having an urban region of origin makes no significant difference to the probability of becoming an entrepreneur in this sample of return migrants. 4 This is consistent with Evans and Leighton (1989) but not with career-shopping models in which younger persons begin by trying risky occupations.

# Scottish Economic Society 2001

{Journals}sjpe/48_2/a198/makeup/a198.3d

OVERSEAS WORK EXPERIENCE, SAVINGS AND RETURN MIGRATION 177

V

CONCLUSION

We have found that variations in overseas employment opportunities have significant effects on the probability of those returning migrants becoming entrepreneurs in the origin LDC. For literates these effects arise from both the influence of total savings overseas, which may help avoid liquidity constraints in the LDC, and from the length of overseas employment, which may reflect skill and ideas acquisition. The estimates suggest that the length of overseas employment may be economically a more important influence on variations in the likelihood of entrepreneurial work. Amongst illiterates we find no influence from longer periods overseas on the likelihood of becoming an entrepreneur, but a return on overseas savings that equalled that found for literates. This supports the conjecture that the constraints on entrepreneurship in LDCs might not only arise from widely discussed capital market imperfections, but also failures in the markets for acquiring human capital. ACKNOWLEDGEMENT The authors would like to thank two anonymous referees and participants at the RES=SES Annual Conference in St. Andrews, July 2000 for helpful comments and suggestions. REFERENCES ADAMS, Jr., R. (1991). The economic uses and impact of international remittances in rural Egypt. Economic Development and Cultural Change, 39, pp. 695±722. ADAMS, Jr., R. (1998). Remittances, Investment and rural asset accumulation in Pakistan. Economic Development and Cultural Change, pp. 155± 73. ARIF, G. M. and IRFAN, M. (1997). Return migration and occupational change: the case of Pakistani migrants returned from the Middle East. Pakistani Development Review, 36, pp. 1± 37. BHAGWATI, J. (1976), Taxing the Brain Drain, North Holland. BHAGWATI, J. and HAMADA, K. (1974). The brain drain, international integration of markets for professionals and unemployment: a theoretical analysis. Journal of Development Economics, 1, pp. 19± 42. CO, C. Y., GANG, I. N. and YUN, M. S. (2000). Returns to returning. Journal of Population Economics, 13, pp. 57± 80. CORNELIUS, W. (1990). Labour migration to the United States: development outcomes and alternative in Mexican sending communities, Commission for the Study of International Migration and Co-operative Economic Development. Washington DC. DIATTA, M. A. and MBOW, N. (1999). Releasing the development potential of return migration: the case of Senegal. International Migration, 37, pp. 243±64. DJAJIC, S. (1986). International migration, remittances and welfare in a dependent economy', Journal of Development Economics, 21, pp. 229± 34. DURAND, J. and MASSEY, D. S. (1991). Mexican migration to the United States: A critical review. Latin American Research Review, 27, pp. 3± 42. DUSTMANN, C. (1995). Savings behaviour of return migrants; a life-cycle analysis. Zeitschrift fur Wirtschafts- und Soczialwissenschaften, 115, pp. S.511 ±33. ESCOBAR, A. and MARTINEZ, M. (1988), Small-scale industry and international migration in Guadalajara, Mexico. Working paper No. 53. Washington DC: Commission for the Study of International Migration and Co-operative Economic Development. # Scottish Economic Society 2001

{Journals}sjpe/48_2/a198/makeup/a198.3d

178

BARRY MCCORMICK AND JACKLINE WAHBA

EVANS, D. and JOVANOVIC, B. (1989). An estimated model of entrepreneurial choice under liquidity constraints. Journal of Political Economy, 97, pp. 808±27. EVANS, D. and LEIGHTON, L. (1989). Some empirical aspects of entrepreneurship. American Economic Review, 79, pp. 519± 35. FERGANY, N. (1988). In Pursuit of Livelihood. Beirut: The Centre for Arab Unity Studies. FOSTER, G. (1967). Tzintuntzan: Mexican Peasants in a Changing World. Little, Brown, Boston. ILAHI, N. (1999). Return migration and occupational change. Review of Development Economics, 3, pp. 170 ± 86. LALONDE, R. and TOPEL, R. (1997). Economic impact of international migration and the economic performance of migrants. In M, Rosenzweig and O. Stark (eds.), Handbook of Population and Family Economics, 2. Elsevier, The Netherlands. LOPEZ, J. and SELIGSON, M. (1991). Small business development in El Salvador: The Impact of Remittances. In S. Diaz-Briquets and S. Weintraub. (eds.), Migration, Remittances and Small Business Development: Mexico and Caribbean Basin Countries. Westveiw Press, USA. LUCAS, R. E. B. (1991). Emigration to South Africa's mines. American Economic Review, 77, pp. 313 ± 30. LUCAS, R. E. B. and STARK, O. (1985). Motivations to remit: evidence from Botswana. Journal of Political Economy, 93, pp. 901± 18. LUTIFYYA, A. M. (1966). Baytin a Jordanian Village: A Study of Social Institutions and Social Change in a Folk Community. MASSEY, D. S., et al. (1987). Return to Aztlan: The Social Process of International Migration from Western Mexico, University of California Press, USA. McCORMICK, B. and WAHBA, J. (2000). Overseas employment and remittances to a dual Economy. The Economic Journal, 110, April, pp. 509± 34. NADA, A. (1991). Impact of temporary international migration on rural Egypt, Cairo Papers in Social Sciences, 14, 3. Cairo: The American University in Cairo Press. PAINE, S. (1974). Exporting Workers: The Turkish Case. Occasional paper 41. Cambridge University Press, Great Britain. PORTES, A. and GUARNIZO, L. (1991). Tropical capitalists: US bound immigration and small enterprise development in the Dominican Republic. In S. Diaz-Briquets and S. Weintraub. (eds.), Migration, Remittances and Small Business Development: Mexico and Caribbean Basin Countries. Westveiw Press, USA. RAZIN, A. and SADKA, E. (1997). International migration and international trade. In M. Rosenzweig and O. Stark (eds.), Handbook of Population and Family Economics, 2. Elsevier, The Netherlands. REICHERT, C. (1993). Labour migration and rural development in Egypt: a study of return migration in six villages. Sociologia Ruralis, 33, pp. 42± 60. RIVERA-BATIZ, F. (1982). Non-traded goods and economic welfare in source country. Journal of Development Economics, 11, pp. 81 ±90. RODRIGUEZ, E. and HORTON, S. (1995). International return migration and remittances in the Philippines. University of Toronto, Department of Economics Working paper 95-01. SMITH, R. and BLUNDELL, R. (1986). An exogeneity test for a simultaneous equation Tobit model with an application to labour supply. Econometrica, 54, pp. 679±86. TAYLOR, J. E. and ADELMAN, I. (1996a). Village Economies: The Design, Estimation and Use of Village-Wide Economic Models, Cambridge University Press, Cambridge. THOMAS-HOPE, E. (1999). Return migration to Jamaica and its development potential. International Migration, 37, pp. 183± 205.

Date of receipt of final manuscript: 20 February 2001.

# Scottish Economic Society 2001