The role of internal migration in accessing a first job: A case study of Uganda

• DOI: http://doi.org/10.1111/ilr.12123
• Published date: 01 December 2018
• Author: Delphine BOUTIN
• Date: 01 December 2018

International Labour Review, Vol. 157 (2018), No. 4

Copyright © The author 2018

Journal compilation © International Labour Organization 2018

* Laboratoire d’analyse et de recherche en économie et nance internationales (LAREFI),

University of Bordeaux, email: delphine.boutin@u-bordeaux.fr. The author wishes to thank Sara

Elder and the participants in the 2016 ILO Work4Youth Global Research Symposium and the

22nd Annual Meetings of the Society of Labor Economists (SOLE 2017) for their valuable com-

ments and suggestions. The author also acknowledges the support of the French National Research

Agency (ANR), through the programme “Investissements d’avenir ANR-10-LABX-14-01”, and the

Foundation for Studies and Research on International Development (FERDI).

Responsibility for opinions expressed in signed articles rests solely with their authors, and

publication does not constitute an endorsement by the ILO.

[Correction added on 6 November 2020, after initial online publication. A duplicate of

this article was published under the DOI 10.1111/ilr.12077, this duplicate has now been deleted

and its DOI redirected to this version of the article.]

The role of internal migration

in accessing a rst job:

A case study of Uganda

Delphine BOUTIN*

Abstract. Does experiencing internal migration hasten access to the labour mar-

ket? This article studies the gap in length of transition to a rst job between intern-

al migrant and non-migrant youth in Uganda. According to the specic context

of this developing country, three transition starting points are considered: date of

birth, minimum legal working age and school exit. Extended Cox proportional

hazard models suggest that migrants experience shorter transitions. However, when

excluding child labourers or measuring school-to-work transitions, signicant gaps

disappear. Decomposition of transition length gaps reveals the importance of ob-

servable and unobservable factors related in particular to area of origin, gender

and access to education.

Internal migration is generally conceptualized as a form of human capital

investment: individuals (and/or their households) weigh the current costs of

migration against its expected returns in terms of wages or better employment

transitions with a view to maximizing their welfare (Todaro, 1969; Harris and

Todaro, 1970; Fields, 1975). At destination, however, the labour market per-

formance of migrants may be affected by discrimination or by differentials in

pre-migration factors such as age, work experience, education and psychology.

Empirical evidence of the job search process of internal migrants is scarce

(especially in developing countries) mainly because of data limitations. This

article aims to assess whether the internal migration experience in Uganda is

International Labour Review632

an impediment or a driver for rst entry into the labour market, by studying

the length of transition periods. In relation to youth, speed of transition to a

rst employment is used as a proxy for the difculties that they may encounter

upon entering the labour market.1 Transition paths and lengths are thus useful

in identifying the effect of potential discrimination on young migrants, going

beyond static information regarding employment rate or occupational status.

An unsuccessful transition results in a longer entry period and may be associ-

ated with a degradation of human capital. This could be seen by employers as

a signal of low productivity, making the individual increasingly less “employ-

able” over time. Studying how internal migration impacts transition to a rst

job consists of the following two steps: rst, the time taken to nd a rst job

is calculated according to three duration methodologies outlined in section 3;

second, these estimated transition lengths are decomposed into explained and

unexplained components. In this second step, the main difculty lies in adapt-

ing a Blinder–Oaxaca style decomposition to non-linear models.

We focus on Uganda for a rst analysis of this question, taking advan-

tage of the comprehensive school-to-work transition surveys conducted by the

Work4Youth (W4Y) partnership between the International Labour Organiza-

tion (ILO) and the Mastercard Foundation.2 This country has experienced sub-

stantial internal migration due to the 20-year ethnic war in northern Uganda

(IOM, 2015). Voluntary international (628,00 0 people in 2013 (UBOS, 2012))

and internal migration (1.3 million according to the 2002 census) are also

common, mainly driven by high population growth (3.2 per cent a year) and

underemployment.3 Demographic pressure particularly affects people below

the age of 30, who account for 78 per cent of the total population (ibid.). As

commonly observed, the rate of unemployment among young people is higher

than among older groups (ILO, 2015). While youth unemployment is low in

comparison to other regions, it is negated by conditions of working poverty

and widespread vulnerable employment. Furthermore, the timing and length

of the transition to a rst job vary considerably across regions, socio-economic

contexts and individual characteristics.

In fact, it is uncertain whether it will take migrants relatively longer to

nd a rst job. It may take longer if skills acquired in their region of origin are

not perfectly transferable to the destination market, or are of a lower quality

1 This article does not study the “quality” of the job obtained, nor the possibility of choos-

ing to remain unemployed in the hope of getting a “good” job. This “luxury unemployment” as-

sumption (Udall and Sinclair, 1982) seems inappropriate in the Ugandan context as the proportion

of youth that have been employed in the formal sector is low for both migrants and non-migrants

(less than 5 per cent of those aged 15–29 have been in formal employment at some point in their

employment history).

2 See: www.ilo.org/w4y [accessed 4 September 2018].

3

Marriage also explains population movements, especially in rural zones and for girls (IOM,

2015; Kudo, 2015). Migration is also linked to the transition to adulthood (Beegle and Poulin, 2013)

or could be an attempt to diversify sources of income (Rosenzweig and Stark, 1989).