Employment choice and mobility in multi‐sector labour markets: Theoretical model and evidence from Ghana

• Published date: 01 December 2013
• Author: Olumide TAIWO
• DOI: http://doi.org/10.1111/j.1564-913X.2013.00189.x
• Date: 01 December 2013

International Labour Review, Vol. 152 (2013), No. 3–4

Copyright © The author 2013

Journal compilation © International Labour Organization 2013

Employment choice and mobility

in multi-sector labour markets:

Theoretical model and evidence

from Ghana

Olumide TAIWO*

Abstract. Using detailed household survey data on male workers in Ghana,

the author tests a theoretical model incorporating capital market failure and la-

bour market segmentation into sectors of (largely formal) wage employment and

(largely informal) self-employment and family enterprise employment, in which

credit-constrained individuals draw self-employment capital from family assets. The

data show very low rates of mobility across the three sectors, the highest mobility

being observed among family enterprise workers, and the lowest, among the self-

employed. The ndings show no robust evidence that wage earnings ease liquidity

constraints, while suggesting that both liquidity and skill transferability constraints

are important for mobility.

E

conomic mobility can mean different things in different contexts. Because

employment income is the most important component of personal and

household income, especially in developing countries, analysis of economic

mobility typically focuses on the labour market. In the textbook labour mar-

ket context, mobility simply means higher income. In some dual labour market

contexts, mobility may involve moving from the rural sector to the urban sec-

tor or, more generally, from the non-wage sector to the wage sector. In most

developing societies, however, the non-wage sector is not homogeneous but

internally partitioned on the basis of production process (whether a worker

is being added to an existing production process or has to create a new pro-

cess) and income determination, i.e. whether earnings can be considered as

marginal product (e.g. non-agricultural self-employment) or average product,

as in family agriculture. Mobility can thus also refer to transition within the

non-wage sector. The multisector nature of labour markets in less developed

countries makes mobility an important indicator of economic development;

*

Centre for the Study of the Economies of Africa (CSEA), Abuja; email: otaiwo@gmail.com.

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

publication does not constitute an endorsement by the ILO.

International Labour Review470

hindrances to labour mobility across sectors would clearly be a sign of their

uneven development.

The literature on employment choice and mobility is dominated by two-

sector models in which individuals choose between wage employment and self-

employment, and workers typically transit from the former to the latter (Evans

and Jovanovic, 1989; Hurst and Lusardi, 2004; Buera, 2009). Such models ad-

equately describe the experiences of high-income countries, but they are un-

suitable for the study of developing economies where the opposite happens.

Because existing theories assume that in equilibrium labour markets always

clear and individuals can nd wage employment whenever they want, they

do not provide sufcient analytical insights into non-clearing labour markets.

Indeed, the majority of self-employment in low-income countries consists of

informal back-up activities pursued by individuals who are unable to secure

(formal) wage employment.

These differences have important policy implications. For instance, the

two-sector model considers credit constraint to be the principal limitation to

self-employment and labour productivity growth. In the three-sector model

I analyse in this article, however, work experience also limits mobility because

of the risk of failure in self-employment and the non-marketability of experi-

ence across sectors. A wage worker would hesitate to enter self-employment

not only because of the risk of temporary wage loss in the event that self-

employment fails (Kihlstrom and Laffont, 1979), but also because of the risk

of permanent wage loss associated with the uncertainty of returning to wage

employment. Conversely, workers that began with self-employment might have

difculty obtaining wage employment if their experiences are not market-

able in the wage employment sector on account of substantial inter-sectoral

technology gaps.

In contrast to the received doctrine, I model a labour market that is

partitioned into wage employment, self-employment and family enterprise

sectors. The empirical analysis based on this theoretical model uses data

from Ghanaian household surveys to examine employment choice and inter-

sectoral mobility among men (who are typically more mobile than women).

My ndings emphasize the importance of education premia in wage employ-

ment and the inuence of family size and wealth on employment choice. In

terms of mobility, the results show that workers in self-employment are the

least mobile. In particular, those engaged in apprenticeship-led enterprises

are trapped in low-wage activities, while family enterprise workers are the

most mobile. There is no evidence that wage employment earnings ease the

liquidity constraints that delay transition into self-employment. The results

thus suggest that it is not sufcient to pay attention to liquidity constraints

alone; policy-makers must also pay attention to the marketability of skills

across sectors.

The remainder of the article is structured into six sections. The rst re-

views the literature and existing models, and discusses their shortcomings. The

second presents the model used here and considers its implications. The third