The first employment-related indicator of the MDGs addresses the creation of productive employment by measuring the annual growth rate of GDP per person employed.6 The rationale for including labour productivity in the set of MDG indicators is its close and positive relationship with wages. According to neo-classical theory, wages equal marginal labour productivity. Empirical studies confirm a close – albeit far from perfect – relationship between wages and labour productivity (Luebker 2011).
Nevertheless, labour productivity growth presents several drawbacks that limit its validity as a measure of productive employment. Most importantly, in a significant number of developing economies, other production factors – particularly natural resources and land – generate a considerable share of value added, with factor income being earned by their respective owners. In more developed economies with high capital-to-labour ratios, gross domestic product (GDP) growth may be driven mainly by capital accumulation (not necessarily higher returns), and the earned factor incomes then typically accrue to capital owners. This implies that levels of labour productivity – not to mention growth rates – can only be meaningfully compared across economies that share similar factor endowments and sectoral compositions.
Furthermore, labour productivity only makes a relevant measure of decent pay if there are no large differences in labour productivity and corresponding wages between workers. Many poor economies, however, may employ highly productive labour with high wages in some sectors (e.g. mining), while a large share of their workforce is engaged in low-productivity and low-wage occupations, such as subsistence agriculture and work in the urban informal sector.
Wide wage differentials may also occur within sectors. In consequence, GDP growth rates may
6 A general problem in linking production and employment is the possible discrepancy between output measured by national accounts and people in work (but not employed) who generate output. While housework (unpaid, at home) is neither considered as employment nor measured in national accounts, this may be different for the self-employed and (part of) their produce. For example, smallholder production – though it may (partially) be for household consumption – is within the national accounts production boundary.
22
increase without any improvement of labour productivity and wages across large segments of the workforce.
2.3.2 Employment-to-population ratio
The employment-to-population ratio (EPR) is the share of a country’s working-age population that is actually employed. This indicator measures an economy’s ability to provide employment for those people who are able to work, i.e. the extent to which full employment is achieved. The ILO (2009) suggests a value range from 50 to 75 per cent. Yet the lack of a clear target value or even a guideline for the EPR renders the interpretation of this indicator difficult and ambiguous, particularly in the context of developing economies. For example, a high EPR may result from a low unemployment rate, which might, in turn, be the sign of a well-functioning and rapidly clearing labour market. However, low registered unemployment can also be a reflection of the lack of social support systems, let alone unemployment insurance, as is the case in many poor economies. People in such circumstances cannot afford to be unemployed and have to work for survival. Whether their work is counted as
“employment” will then depend on the market orientation of what they do for a living, i.e.
whether it is done in exchange for pay or profit. Similarly, in economies dominated by smallholder agriculture, a high rate of labour market participation and, consequently, a high EPR do not signal ample labour market opportunities, but rather that all available labour is required to operate the household farm. Finally, an EPR may also be low because a substantial share of the working-age population is out of the labour force attending school or university.
Taken together, these ambiguities make the EPR a dubious indicator that can hardly be expected to provide reliable information on progress towards full employment.
2.3.3 Working poverty rate
The ILO (2009) defines the working poverty rate (WPR) as the proportion of employed people living in a household with a per capita income below the nationally defined poverty line (or the international poverty lines of Int. $1.25 or 2 per day). The WPR provides a measure of productive employment by linking poverty and employment data. It represents the share of workers whose jobs do not generate sufficient income to lift them and their families out of poverty. Although this indicator explicitly considers only the income dimension of decent work, it implicitly provides a broader picture: it is reasonable to assume that jobs which provide people with insufficient means to meet their basic needs are unlikely to fulfil other
23
requirements of decent work (ILO 2009). Yet, working poverty may be caused by factors other than low pay (e.g. number of dependants), while other sources of income might mask low labour earnings (e.g. remittances and government transfers).
In principle, the WPR can be readily computed from micro data sets that provide information on poverty status at the household level and employment at the individual level. In practice, however, the WPR has often been calculated using aggregate statistics, as the product of the poverty rate and the labour force participation rate (macroeconomic approach). This simplification has been justified by a lack of micro data, but entails important drawbacks. The neglect of differences in demographic composition and labour force participation rates among poor and non-poor households results in biased estimates of the country-level incidence of working poverty. Using data for 15 sub-Saharan African countries, Kapsos (2011) finds an average difference of 8.6 percentage points between the macro- and the micro-based WPR estimates. Finally, this indicator suffers from the general drawbacks of a headcount index in that it does not account for intra-household inequality or for the depth of poverty. It is also plagued by the problems related to the definition of national and (more so) international poverty lines.7
2.3.4 Vulnerable employment rate
According to the ILO (2009), the vulnerable employment rate is the proportion of own-account and contributing family workers in total employment. Based on their employment status, these two categories of workers are termed vulnerable because they are believed to face high economic risks. Own-account workers and contributing family workers are assumed to be less likely to have formal work arrangements, and thus often to lack elements of decent work, such as adequate social security and effective social dialogue mechanisms (Sparreboom 2011).
Additionally, own-account workers in developing countries are typically perceived to earn a low and irregular income since they are mostly subsistence farmers and small-scale entrepreneurs operating in the informal economy. Indeed, in many cases, contributing family
7 The literature has identified challenges arising from different consumption patterns across countries/regions, the 2005 PPP factors based on 1988–2005 prices and changing basic needs, the strong bias towards African countries (only two out of the 15 poorest countries are in Asia), and urban bias because internationally comparable services and goods are mainly found in urban areas. For a detailed critique and the implications for Asian poverty calculations, see ADB (2014).
24
workers receive neither payment in cash nor remuneration in kind, but benefit indirectly from an increased household income.
However, the assumption that self-employed (family) workers are per se more vulnerable than employees is problematic, thus calling into question the specificity and relevance of the
“vulnerable employment” indicator. Own-account workers and the associated category of contributing family members form a very heterogeneous group in developing countries. It includes not only those engaged in agricultural and urban informal subsistence activities, but also those running productive and profitable micro enterprises. In fact, research has shown that self-employment is often – albeit not always, and conditional on context – chosen voluntarily (Maloney 2004). Furthermore, in many developing countries wage workers (not to mention casual or seasonal workers) are not covered by social protection and do not have legally enforceable contracts of employment, which might render their conditions just as precarious as those of some own-account and contributing family workers. Thus, vulnerable employment would better be defined in terms of employment conditions, rather than employment status.
2.3.5 Share of women in wage employment in the non-agricultural sector
This last employment-related indicator of the MDGs is part of the gender equality goal and can be found under Target 3.A, which aims to “eliminate gender disparity in primary and secondary education, preferably by 2005, and in all levels of education no later than 2015”.
The share of women in paid employment in the non-agricultural sector is expressed as the percentage of female workers in total non-agricultural wage employment.
The ratio of women in wage employment in the manufacturing and service sectors is meant to provide an indication of the extent to which female workers have access to the modern monetary economy and thus benefit from the structural transition of the economy. The underlying assumption is that women in such wage employment enjoy greater autonomy and self-reliance in personal development and decision-making (ILO 2009). While there is indeed some evidence that wage employment outside agriculture improves the situation of women (Kabeer 2005), the general discrimination against self-employment is subject to the same criticism as that expressed above in regard to “vulnerable employment”. Furthermore, the indicator ignores the value that women may attach to the flexibility afforded by work on a household farm or in a household enterprise. Indeed, wage employment may not only bring
25
about positive changes in women’s lives. Since women typically continue to be responsible for domestic work and childcare, gender inequalities in work burdens may intensify (ibid.).