See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/284298532

Migrants in Qatar: A Socio-Economic Profile

Article in Journal of Arabian Studies · December 2012

DOI: 10.1080/21534764.2012.735458

CITATIONS READS

17 2,268

1 author:

Ganesh Seshan
World Bank
27 PUBLICATIONS 199 CITATIONS

SEE PROFILE

All content following this page was uploaded by Ganesh Seshan on 18 January 2016.

The user has requested enhancement of the downloaded file.

 Journal of Arabian Studies 2.2 (December 2012), pp. 157–171

Migrants in Qatar:
A Socio-Economic Profile
GANESH SESHAN
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

Abstract: This first micro-level study of temporary migrants from developing countries in
Qatar uses data collected during the summer of 2007. The median migrant has the following
profile: male, thirty-one years of age, single and possessing a high school education. He
works ten hours a day, six days per week, and earns US$3,945 per annum of which over
half is remitted home to his parents. He incurred a job placement fee equivalent to 1.25
times his monthly salary. Using multivariate regression, we find that migrants’ capacity for
work is positively correlated with age, having a university degree, and satisfaction with
sleep and mental wellbeing. Estimates of a remittance function show that males send more
money home, and that migrants are motivated to save, therefore a tendency to transfer
higher amounts. We also find that older migrants remit more, suggesting that as the
likelihood of returning home increases with age, remittance level rises.
Keywords: temporary contract migrants, remittances, savings, placement fees, mental
wellbeing, work capacity, Qatar, Arabian Gulf

1 Introduction
The countries of the Gulf Cooperation Council (GCC) have one of the highest shares of foreign-born
labor force in the world and in terms of destination choice are ranked third globally for international
migrants after the European Union and North America.1 Annual remittances from the GCC countries
stood at US$54 billion in 2007, accounting for 14% of the global remittance market.2 Saudi Arabia led
the way with remittances worth US$26 billion, followed by Kuwait (US$9.8 billion), UAE (US$8.7
billion), Qatar (US$4.5 billion), Oman (US$3.7 billion) and Bahrain (US$1.5 billion).3
Despite the significance of Gulf migration in terms of destination choice and global remittance
flows, there has been a limited micro-level research on migrants in the region. Existing studies,
though still small in number, tend to look at broad patterns in terms of composition of the migrant
workforce.4 In one of the few micro studies of migrants in the GCC, Zachariah, Prakash, and

Ganesh Seshan is Assistant Professor of Economics at the Edmund A. Walsh School of Foreign Service in
Qatar, Georgetown University, PO Box 23689, Doha, Qatar, [email protected].
1
The GCC countries include Saudi Arabia, Kuwait, the United Arab Emirates (UAE), Qatar, Bahrain,
and Oman.
2
The World Bank’s Migration and Remittances Factbook 2011 reports total remittance flows in 2007 at
US$385 billion.
3
Totals are from the World Bank’s Migration and Remittances Factbook 2011 with the exception of
Qatar and UAE where figures were obtained from the Statistical Bulletins of their respective central banks.
4
See Fargues, “International Migration in the Arab region: Trends and Policies” (2006) and Kapis-
zewski, “Arab Versus Asian Migrant Workers in the GCC Countries” (2006), which are both working
papers submitted to the United Nations Expert Group Meeting on International Migration and Development
in the Arab Region (2006).

ISSN 2153-4764 print/ISSN 2153-4780 online

© 2012 Taylor & Francis
http://dx.doi.org/10.1080/21534764.2012.735458
http://www.tandfonline.com
 158 Ganesh Seshsan

Irudaya Rajan provide useful insights into the nature of Indian migrant workers in the UAE.5 This
paper focuses on the socio-economic profile of migrants from developing countries in the state of
Qatar, using data obtained in a survey supervised by the author among expatriate workers in
Doha, Qatar’s capital city, during May and June 2007. To our knowledge, this is the first
survey targeting a migrant population in the country.
What emerges from the data is the following profile of the median migrant: male, he is thirty-
one years of age, and single, possesses a high school education and has been in Qatar for 1.4
years. He works ten hours a day, six days per week, earning US$3,945 annually, of which
over half is remitted back home to his parents. To come to Qatar, he had to incur a placement
fee equivalent to 1.25 months of his salary. The median migrant has some accumulated
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

savings, and reports that he is satisfied with it. He also expresses satisfaction with his capacity
to work. In terms of self-reported wellbeing over the month prior to being interviewed, he was
satisfied with his amount of sleep, and was happy some of the time, but also felt a little nervous.
Section 2 provides some background on migration to Qatar while Section 3 presents the data.
An empirical analysis of the determinants of remittances is carried out in Section 4, and Section 5
concludes this contribution.

2 Background of international migration to Qatar

In 2007, approximately 87% of the population in the State of Qatar aged fifteen or older, were
foreign born, which at that time made Qatar the country with the highest share of immigrants
in the world.6 The majority of immigrants to Qatar originate from developing countries, with a
significant proportion from South and East Asia.7 Migration into Qatar and other oil-and-gas-pro-
ducing countries of the Arabian Gulf took off in the 1970s with rising oil prices that subsequently
fueled a substantial construction boom. This required a large pool of unskilled and semi-skilled
labors, which the national labor force was unable to provide.8 Since the 1970s, the share of skilled
and professional migrant workers has increased, but the foreign labor force is still dominated by a
sizeable share of workers in the construction sector. Most nationals in Qatar are employed in the

5
See Zachariah, Prakash, and Irudaya Rajan, “Indian Workers in UAE: Employment, Wages and
Working Conditions”, Economic and Political Weekly 39.22 (2004), pp. 2227–34.
6
Computed from Qatar’s Labor Force Survey 2007 conducted by the Qatar Statistical Authority. Qatar’s
total population in 2007 was estimated at 1.2 million [Qatar Statistical Authority, Qatar Annual Statistical
Abstract (2007), p. 3]. Qatar continues to rank at the top of immigration countries in 2010 (measured by the
population share of immigrants) as indicated in the World Bank’s Migration and Remittances Factbook
2011.
7
Kapiszewski, “Arab Versus Asian Migrant Workers in the GCC Countries”. Details of migrant stocks
or flows to Qatar or the Gulf region are difficult to find. Census data for Qatar do not provide a breakdown of
countries of origin for the migrant population. Local companies and recruitment agencies do not readily dis-
close such numbers. Hence, any attempt to provide such figures should be interpreted as an effort in good
faith. In recent years, some countries, such as India and the Philippines, have published statistics on emigrant
outflows to various destination countries, but in the absence of information on returning migrants, it is
impossible to compute sensible stock measures. Kapiszewski estimates that the largest share of migrants
in the country in 2002 came from India (20%) followed by Pakistan and Nepal. His figures for 2002 also
show that the Indian sub-continent, together with the Philippines made up 71% of the immigrant share in
Qatar. Unofficial estimates from the Indian Embassy in Qatar confirm the large presence of Indian migrants,
the majority of whom are from Kerala although greater flows have come from Uttar Pradesh in recent years.
An embassy official thought that there were approximately 200,000 Indian migrants in Qatar in 2007, which
is about 20% of the immigrant population.
8
Shah, “Recent Immigration Policies in the Oil-Rich Gulf: How Effective are They Likely to Be?” ILO
Asian Regional Programme on Governance of Labor Migration, Working Paper 3 (2008), p. 1.
 Migrants in Qatar 159

public sector, though in recent times, as in the rest of the GCC countries that are abundantly
endowed with oil and gas, there is a drive to increase their employment share in the private
sector.9
Migration to Qatar and to the neighboring GCC countries is temporary, with work contracts
stipulating the duration of stay, and at the end of the work contract, migrants are expected to return
to their home countries.10 These contracts are typically for one to two years and some may extend
to five years, although they are renewable at the discretion of the employer.11 Because a migrant’s
work permit or visa is sponsored by his employer, this effectively ties him to his sponsor. Only
workers earning an income above a minimum level are permitted to bring their families with them
–– in Qatar, the amount was QR7,000 a month in 2007, or the equivalent of US$23,000
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

annually.12 Labor markets are tightly regulated in Qatar, as in other GCC countries, and migrants
are not permitted to change employers unless permitted by their current job sponsor.
Remittance services in Qatar are dominated by money exchange houses that tend to offer
cheaper rates than local banks. Exchange houses are also more accommodating towards the
migrant clientele by being open in the evenings and stationing tellers who can speak foreign
languages. Western Union ranks ahead in terms of its agent coverage and density in the countries
from which most migrants in Qatar originate, but charges a relatively higher remittance fee
(between US$7 and US$10) for their service, compared to other exchange houses (between
US$4 and US$5.5) that use their own proprietary money transfer system. In 2007, remittance out-
flows from Qatar totaled US$4.45 billion, which was approximately 5.5% of the country’s GDP.13

3 Data
The data set used for this empirical study was part of a pilot initiative to gather socio-economic
information on migrants in Qatar, since no such information was publicly available at the time. A
two-month survey was initiated in May 2007 with a final working sample of 128 migrants from
various developing countries.
The questionnaire used was designed and pre-tested with a group of migrant workers, after
which a team of fifteen university students from Georgetown and Carnegie Mellon University
in Qatar were trained to carry out face-to-face interviews. A convenience sample approach was
adopted, due to the absence of census information on the composition of migrants in Qatar by
country of origin, and because of the practical challenge of identifying where these communities
resided. The students, many of whom had grown up in Qatar and were themselves from expatriate
households, were instructed to use their social networks to access various migrant communities.
Detailed data were collected on the migrants’ mental health, the frequency and methods of
communication with families, and the remittances they sent home, in addition to savings and
loan decisions. A range of other socio-economic information, including demographics (age, edu-
cation, gender and marital status), country of origin, characteristics of family back home,

9
The 2007 Qatar Labor Force Survey found that 94.6% of Qatari nationals worked in the public sector.
10
Shah, “Recent Immigration Policies in the Oil-Rich Gulf”, p. 10.
11
There is no data on whether the duration of a contract varies depending on the migrant’s job or country
of origin. The length of the contract is not left entirely to the sponsor, as the country’s Ministry of Interior
may also place limits on the contract term.
12
Shah further states that family dependents are sponsored by the worker and in the case of expatriate
children, sons can stay in the country until they reach twenty-one and daughters until they get married [Shah,
“Recent Immigration Policies in the Oil-Rich Gulf”, p. 29]. The Qatari Riyal (QR) is pegged to the US dollar.
One dollar is equivalent to QR3.65.
13
Qatar Central Bank, Quarterly Qatar Statistical Bulletin (Mar. 2010), p. 68.
 160 Ganesh Seshsan

Table I: Selected summary statistics of migrant workers in Qatar.

(1) (2) (3) (4) (5)
Mean Std. dev Min Median Max No. obs
Male indicator 0.76 0 1 1 128
Age 32.13 8.21 18 31 60 128
Married indicator 0.46 0 0 1 128
without family in Qatar 0.73 0 1 1 59
Household size in Qatar 1.38 0.90 1 1 5 128
Household size in home country 4.11 2.58 0 4 13 122
Days worked per week 6.16 0.71 4 6 7 128
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

Hours worked per day 10.19 2.56 7 10 18 127

Migrant’s annual income (USD) 6,348 6,061 986 3,945 31,233 128
Qatar household’s annual income (USD) 8,931 11,684 986 4,027 58,192 128
Annual expenses (USD) 3,366 5,623 164 1,644 41,753 126
Annualized remittances (USD) 2,402 1,875 2.6 2,198 14,000 128
As share of household income 0.46 0.29 0.0 0.5 1.5 128
Paid placement fees indicator 0.64 0 1 1 125
Placement Fee (USD) 766 785 1.4 458 4,000 78
As a ratio of monthly income 2.51 2.70 0.01 1.25 12.1 78
As a ratio of annual income 0.21 0.23 0.00 0.10 1.01 78
Source: Author’s calculation using Qatar Migrant Survey (2007).

occupation, and earnings and expenditure, was also sought from the participants. Table I presents
summary statistics for a selection of the variables.

3.1 Demographics
The average age of migrants in our sample was thirty-two years. Over three-quarters were male
and 81% were alone in Qatar without immediate family members. Forty-six percent of the sample
was married and among those married, 73% were in Qatar without their spouses and dependents.
Fifty-one percent of the migrants surveyed stayed in workers’ accommodation provided by the
employer, while 11.7% rented their own accommodation and another 18% shared a rental
place with non-relatives (Figures 1 and 2).
The majority of the migrants in the sample came from the Indian subcontinent (59.4%), fol-
lowed by the Philippines (25%), which coincides with our understanding of the migrant

Figure 1: Migrant’s age distribution.

 Migrants in Qatar 161
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

Figure 2: Migrants’ accommodation type.

Figure 3: Source of migrants in sample.

Figure 4: Migrants’ education attainment.

composition in Qatar.14 Nearly 77% of the migrants interviewed had completed schooling at the
high school level or greater. The average size of the household left behind in the country of origin
was four members (Figures 3 and 4).

14
The Indian subcontinent breakdown in terms of shares is as follows: India (25.8%), Sri Lanka
(17.2%), Nepal (10.2%), Bangladesh (4.6%), and Pakistan (1.6%).
 162 Ganesh Seshsan
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

Figure 5: (a) Migrants’ own income distribution and (b) migrant household’s income distribution.

3.2 Income and employment

The median yearly income of the migrant worker was US$3,945, while that of his or her house-
hold in Qatar was US$4,028.15 The median hours worked per day was ten hours over an average
of six days per week (Figure 5).

3.3 Cost of migration: placement fees

Approximately, 64% of the migrants had to pay a placement fee to secure their job in Qatar. This fee, on
average, amounted to approximately two and a half times their monthly salary or 21% of their annual
income. At the extreme end, the fee reported was the equivalent of a year’s worth of salary. Migrants
whose placement fees exceeded half their annual salary tended to originate from South Asia (Figure 6).

3.4 International telecommunications

With respect to international telecommunications, about which complete data were available for
109 of the individuals surveyed, migrants used several methods. The cell-phone was the most fre-
quent method at 56%, followed by the Internet phone (47.7%), text messaging (10.1%), Internet
cafe (7.3%), and fixed-line (5.6%).16 Excluding the use of text messaging and Internet cafe,
migrants spent approximately nineteen minutes a week on average calling their relatives and
friends abroad. The average international calling rate (per minute) was US$0.50. Call tariffs
are measured using posted (peak) cell-phone and landline prices by Qatar Telecom and the
reported price of Internet phone calls by the migrants.17 On average, migrants who used a
mobile phone for making international calls spoke for a duration of approximately
12.3 minutes per week. Internet phone users spent around 24.8 minutes a week on calls
abroad, while landline users spoke on average for 10.7 minutes per week.
Figure 7 illustrates an intriguing relationship between remittances and weekly international
calling times. The hypothesis is that there are important synergies between telecommunications
and remittances among migrants. Migrants with greater access to telecommunication may have
a higher demand for remittances, because more frequent communication with family and relatives

15
For local currency details, see footnote 12.
16
The Internet phone service was provided by unlicensed vendors and utilized VOIP technology. The
method of telecommunications was not necessarily mutually exclusive for each migrant.
17
In cases where the migrant used cell-phones and Internet phone to call abroad, I construct a weighed
average of tariffs, using the total time spend on cell phones and Internet phones by all migrants who used a
combination of both calling methods.
 Migrants in Qatar 163
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

Figure 6: (a) Placement fee (US$) and (b) ratio of placement fee to monthly income.

Figure 7: Plot of annualized remittances (US$) against international calling

time (minutes/week) and call tariffs (US$, cost per minute).
Source: Author’s calculation using Qatar Migrant Survey (2007).

strengthen ties and raise altruistic motives for remitting funds. At the same time, encouraging
migrants to remit more may also stimulate demand for telecom services, if migrants want to
monitor the expenditure patterns of the recipients of remittance money.
Returning to Figure 7, and consistent with the stated hypothesis, we observe that annual remit-
tances increased with additional calling time up to around twenty-five minutes per week. An additional
US$1,120 is remitted abroad per year when weekly calls last between fifteen to twenty-five minutes,
compared to less than five minutes. Conceivably migrants do not require more than twenty-five
minutes a week to communicate with their relatives abroad and to be acquainted with their welfare
or remittance usage to decide subsequently upon additional transfers home. This relationship needs
to be more carefully studied to determine the direction of causality, but it points to the potential of
more affordable telecommunications, not just for transnational families to stay regularly in touch
but also to decide jointly on how remittances should be used in a way that enhances welfare.18

18
Elsewhere, I examine the relationship between international calls and remittances using a multivariate
regression approach, and find a strong association between the two variables [Seshan, “Does More Time
Spent Calling Home Correlate with Higher Remittances? Evidence from Migrants in Qatar”, B.E. Journal
of Economic Analysis and Policy 12.1 (2012).].
 164 Ganesh Seshsan
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

Figure 8: Remittance against income.

Figure 9: Remittance share against income.

3.5 Remittances
It could be taken for granted that most employed migrants in Qatar send some money home on a
regular basis. Remittances on an annualized basis tend to rise, albeit at a decreasing rate with
household income, and the share of remittances to income is relatively higher for lower
income migrant households. Annual remittance averaged US$2,402, which represented approxi-
mately 46.7% of a migrant’s income or 43.8% of the migrant’s household income in Qatar
(Figures 8 and 9).
We also observe a pattern of remittances rising with age, though at a decreasing rate up to the
mid-thirties, when it levels off. This may be attributed to the higher earnings of older migrants, but
also suggests that older migrants who face increasing prospects of returning home permanently
may increase the amount of remittances sent back. We investigate this further in a subsequent
empirical analysis, controlling for income and a variety of other factors that impact on remittance
levels (Figures 10 and 11).
Nearly 59% of remittance recipients were parents, followed by spouses at 28% and to a lesser
extend siblings and children at 12.5% and 5%, respectively. Slightly over a third of the migrants
surveyed were unaware of what the remittances were used for, which may reflect the tendency of
some migrants to let family members back home decide on the uses of remittances. This seems to
be true when parents are the recipients, and represented 57% of all cases where migrants were
unaware of how their transfers were utilized. Nonetheless, 12.5% of the sample indicated that
 Migrants in Qatar 165
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

Figure 10: Income against migrant’s age.

Figure 11: Remittance against migrant’s age.

a portion of their remittances was spent on education and 23.4% stated that some of their remit-
tances were kept aside as savings. A smaller fraction of migrants, that is, 3%, stated that a share of
their remittances was used to finance investments, which included house purchase, house con-
struction, or supporting a business (Figures 12 and 13).
The vast majority of migrants sampled sent money home through money exchange houses.
Sixty percent of the respondents who remitted money used a money exchange house to send
money home, and 35% used a bank transfer. Only a small fraction (3.5%) sent money home
with a friend or relative who was traveling back.19

3.6 Savings
Migrants in the study were also asked about their own savings behavior. Nearly 72% indicated
that they had some form of savings, either in Qatar or in their home country. This may suggest
that a larger percentage of migrants were sending money home to save than was expressed

19
A representative survey conducted by the Social and Economic Survey Research Institute (SESRI) in
2010 of approximately 682 migrant workers residing in labor camps in Qatar found similar proportions using
money exchanges (57%) and bank transfers (36%) [SESRI, “Migrant Labor Workers in Qatar: Demographic
Profile, Employment and Working Conditions, Remittances, Quality of Life and Future Outlook”, Executive
Summary Report, Mar. 2011, p. 7].
 166 Ganesh Seshsan
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

Figure 12: Remittance recipients.

Figure 13: Reported use of remittances.

when they reported the uses of their remittances. When asked about how satisfied they were with
their level of savings, 45% responded that they were less than satisfied, 32% were satisfied while
only 5.5% were very satisfied with their accumulated savings.
Of migrants who had savings, 52% indicated that they were saving up to buy or build a house,
38% stated that they were saving for their children’s education, 33% were setting money aside to
buy land in their home country, and 21.7% were saving to start or expand a business back home
(Figure 14).

3.7 Physical and mental health

While the benefits of migration are often highlighted, there may be costs that migrant families
incur that can offset the benefits of migration and remittances. Relationships between spouses
may be strained when one partner goes abroad to work, children may suffer psychological
trauma due to extended separation from a migrant parent or parents, and migrants themselves
may suffer in terms of their mental health, due to separation from spouse and children as well
as conditions in their work and place of residence. Their poor mental condition may in turn
affect their ability to be productive at work. A number of self-report mental health questions
 Migrants in Qatar 167
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

Figure 14: (a) Migrants’ satisfaction with savings and (b) migrant’s purpose for savings.

Figure 15: (a and b) Migrant’s satisfaction with sleep and work capacity.

were put to the participants. Respondents in the survey were asked about their self-reported phys-
ical and mental health over the past month. Over two-thirds reported that they were in good or
very good health, with 6.25% stating that their health was ‘bad’. In terms of sleep, 61.2% of
the migrants stated that they were satisfied or very satisfied with their sleep patterns.
We also asked the migrants about their satisfaction with their capacity for work over the
past month, to which 69% indicated that they were satisfied or very satisfied. Satisfaction
with sleep was correlated with satisfaction with work capacity, where 83% of individuals
who were at least satisfied with their work capacity reported being satisfied or very satisfied
with sleep (Figure 15).
In terms of happiness, 43% of the migrants stated they were happy at least most of
the time, while 48.4% reported feeling calm and peaceful at least most of the time. Nearly
50% of the migrants reported feeling nervous and downhearted at least some of the time
(Figure 16a–e).20
To obtain a preliminary idea of what influenced a migrant’s satisfaction with his or her work
capacity, we estimated a simple OLS regression of the level of satisfaction with work capacity on
age, a male indicator, satisfaction with sleep, state of happiness and nervousness, income levels,
and country of origin indicators as independent variables. The results in Table II should be taken
as an initial finding that merits more careful study. The estimates indicate a positive and statisti-
cally significant relationship between satisfaction with sleep and work capacity. Higher periods of

20
I did not document reasons for the subjects’ mental state. Future work should explore the correlates of
mental wellbeing amongst migrant workers.
 168 Ganesh Seshsan
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

Figure 16: (a–e) Migrant’s self-reported mental wellbeing.

Table II: Migrant’s satisfaction with work capacity OLS regression.

Age 0.210*** (0.05)
Age squared −0.003*** (0.00)
Male −0.403*** (0.13)
University degree 0.319* (0.16)
Sleep satisfaction 0.362** (0.13)
Calm and peaceful periods −0.223** (0.08)
Nervousnesss periods 0.170** (0.06)
Annual income 0.000 (0.00)
R-square 0.52
Observation 127
Note: *denotes statistical significance at the 10% level, **at the 5% level and ***at the 1% level. Regression includes
country of origin dummies. Standard errors are in parenthesis and correct for clustering by country of origin.

feeling calm and peaceful were also related to greater satisfaction with work capacity, while fewer
periods of nervousness were positively associated with more satisfaction with work capacity.
Older migrants tended to express more satisfaction with their capacity for work, while male
migrants were less satisfied than females with their work capacity. Holding a college degree
was also found to be positively related to satisfaction with work capacity, whereas income
levels appeared to be uncorrelated with work capacity satisfaction.
Again the tentative nature of these findings is emphasized, as there are issues related to omitted
variable bias and simultaneity bias. For example, an observed characteristic of a migrant (say a
death of a close relative) could influence both his sleep and work capacity. In addition, low satis-
faction with one’s work capacity could, in turn, affect sleep habits and mental wellbeing.

4 An empirical model of remittance behavior

An emerging literature in economics attempts to address the factors that influence the level of
remittances sent home by migrants. This includes an investigation concerning the motives for
remittances. Prior studies have examined various motives for remittances that consist of
 Migrants in Qatar 169

transfers to support families and relatives in the country of origin, and savings or investments
for future consumption back home. Remittances to family and kin can be informed by altruistic
behavior but could also be the price to be paid for the option to return home at a future date, or as
an ‘insurance’ to be welcomed back by the home community permanently.21 Remittance flows
to purchase assets or to be kept as savings in the home country can be motivated by an inter-
temporal allocation of consumption or investments into a durable good across national
borders.22
We specify a model of remittance determinants for Qatar migrants with the level of annual-
ized remittances as the dependent variables, and several independent variables, motivated by
the literature, that include a collection of migrant characteristics such as age, its squared
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

term, gender, income, and country of origin. The inclusion of age and its square term is to
examine whether older migrants tend to send more home since the likelihood of returning
home permanently rises with age. Most of the existing literature refers to the ‘remittance
decay hypothesis’ which states that remittances decrease with the length of time migrants
spend in the host country since ties to their country of origin diminish. While this may be
the case for permanent migrants or those who migrate with the intention of eventually settling
in the host country, it may not be applicable to temporary migrants who arrive with work con-
tracts defined over a specific period of time, such as those in the GCC countries.23 The inclusion
of annual migrant’s household income in Qatar is made on the basis that remittances are
expected to be positively related to income levels. Country of origin indicators capture any vari-
ations that might be attributed to country-specific attributes of the migrant that could influence
remittance levels, such as exchange rates, cost of living, and the extent and coverage of remit-
tance-receiving institutions.
The results in Table III show that remittance rises with income level –– each additional US
$100 of annual income raises annualized remittances by US$6.5. Male migrants remit an extra
US$482 dollars annually compared to female migrants, possibly a reflection of their roles as
breadwinners for their families back home. We also see remittances rising with age but at a
decreasing rate which suggests, contrary to the ‘remittance-decay’ hypothesis, that older migrants
who face greater prospects of returning to their home country may increase remittances as they
draw closer to their actual return date. Relative to other remittance recipients, lower remittances
are sent when the recipients are children or grand children. Migrants who reported having some
savings tended to remit an additional US$593, compared to those who had no savings. This
suggests that migrants who have financial goals or are able to pursue them are likely to remit
more, all things being equal, compared to those who do not.

5 Conclusion
This paper presents one of the first empirical studies of migrants in Qatar from developing
countries utilizing an original data set collected during the summer of 2007. Foreigners

21
For an early model of altruism with respect to remittances, see Lucas and Stark, “Motivations to Remit:
Evidence from Botswana”, Journal of Political Economy 93.5 (1985), pp. 901–18. For a survey on the private
transfer literature, see Laferrère and Wolff, “Microeconomics Models of Family Transfer”, in Handbook of the
Economics on Giving, Reciprocity and Altruism, vol. 2, ed. Kolm and Ythier, chapter 13, pp. 889–961.
22
Dustmann and Mestres, “Remittance and Temporary Migration”, Journal of Development Economics
92.10 (2010), pp. 62–70.
23
An earlier study that tested this hypothesis is Brown, “Do Pacific Island Migrants’ Remittances
Decline over Time? Evidence from Tongans and Western Samoans in Australia”, The Contemporary
Pacific 10.1 (1998), pp. 107–51.
 170 Ganesh Seshsan

Table III: Annualized remittance OLS regression.

Age 313.28*** (87.64)
Age squared −4.55*** (1.26)
Male 482.28* (278.61)
Annual migrant’s household income 0.06*** (0.02)
Remittance recipient indicator: children −1121.58 (746.71)
Remittance recipient indicator: grandchildren −1705.65*** (370.39)
Indicator for having savings 592.68* (296.73)
Indicator: remitting for education 5.15 (178.32)
Indicator: remitting for asset accumulation 86.37 (324.39)
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

R-square 0.60
Observation 128
Note: * denotes statistical significance at the 10% level, ** at the 5% level and *** at the 1% level. Regression includes
country of origin dummies. Standard errors are in parenthesis and correct for clustering by country of origin.

working in the country are temporary migrants who arrived with work contracts with a stipu-
lated employment period. We find that migrants’ satisfaction with their capacity for work is
positively associated with age, having a university degree, and satisfaction with sleep and
mental wellbeing. The results also show that remittances sent home increases with income
levels and men send more money home as compared to their female counterparts. Older
migrants also remit more, all things being equal, which suggests that as the likelihood of return-
ing home increases, remittance levels rise. Migrants with savings motives also tend to transfer
more funds back home.
There is much scope for further micro-level research on migrants working in the countries of
the Arabian Gulf. We hope that some of the data presented in this paper on migrants in Qatar will
encourage further investigation of Gulf migrants. No doubt, most migrants are in the region to
earn a living that is at least equal to or better than what they would receive in their home
country. They may have financial goals in mind, such as educating their children or accumulating
savings to build a home or start a business. To what extent are they able to work towards their
goals? What is the role of their families back home in striving towards these goals? Are the remit-
tances sent home used productively or spent primarily on durable goods and home enhancements?
Little is known of the extent to which remittance levels may even be attenuated by the limited
ability that migrants have to monitor or control how remittances are used by recipients. These
are some topics that would be interesting to pursue from the perspective of Gulf migration.

Bibliography
Brown, Richard P.C., “Do Pacific Island Migrants’ Remittances Decline over Time? Evidence from Tongans
and Western Samoans in Australia”, The Contemporary Pacific 10.1 (1998), pp. 107–51.
Dustmann, Christian, and Josep Mestres, “Remittance and Temporary Migration”, Journal of Development
Economics 92.10 (2010), pp. 62–70.
Fargues, Philippe, “International Migration in the Arab region: Trends and Policies”, working paper sub-
mitted to the United Nations Expert Group Meeting on International Migration and Development in
the Arab Region (2006), available online at www.un.org.
Kapiszewski, Andrzej, “Arab Versus Asian Migrant Workers in the GCC Countries”, working paper sub-
mitted to the United Nations Expert Group Meeting on International Migration and Development in
the Arab Region (2006), available online at www.un.org.
 Migrants in Qatar 171

Laferrère, Anne, and François-Charles Wolff, “Microeconomics Models of Family Transfer”, in Handbook
of the Economics on Giving, Reciprocity and Altruism 2, edited by Serge-Christophe Kolm, and Jean
Mercier Ythier (Amsterdam, The Netherlands: North-Holland, 2006), chap. 13, pp. 889–961.
Lucas, Robert E., and Oded Stark, “Motivations to Remit: Evidence from Botswana”, Journal of Political
Economy 93.5 (1985), pp. 901–18.
Qatar Statistical Authority, Qatar Annual Statistical Abstract (2007), available online at www.qsa.gov.qa/
eng/GeneralStatistics.htm#Annual_Abstract_2011.
Seshan, Ganesh, “Does More Time Spent Calling Home Correlate with Higher Remittances? Evidence from
Migrants in Qatar”, B.E. Journal of Economic Analysis and Policy 12.1 (2012), available online at www.
degruyter.com/view/j/bejeap.2012.12.issue-1/1935-1682.3262/1935-1682.3262.xml.
Shah, Nasra, “Recent Immigration Policies in the Oil-Rich Gulf: How Effective are They Likely to Be?” ILO
Asian Regional Programme on Governance of Labor Migration, Working Paper 3 (2008).
Downloaded by [Georgetown University], [Ganesh Seshan] at 12:14 13 December 2012

Social and Economic Survey Research Institute (SESRI), “Migrant Labor Workers in Qatar: Demographic
Profile, Employment and Working Conditions, Remittances, Quality of Life and Future Outlook”,
Executive Summary Report, March 2011.
World Bank, Migration and Remittances Factbook 2011 (2011), available at www.siteresources.worldbank.
org/INTLAC/.../Factbook2011-Ebook.pdf.
Zachariah, K.C.; B.A. Prakash; and S. Irudaya Rajan, “Indian Workers in UAE: Employment, Wages and
Working Conditions”, Economic and Political Weekly 39.22 (2004), pp. 2227–34.

View publication stats