Measuring decent work with statistical indicators.

The promotion of decent work has been the ILO's central objective and organizing framework since 1999 when this concept was first introduced and described in terms of "opportunities for women and men to obtain decent and productive work in conditions of freedom, equity, security and human dignity" (ILO, 1999, p. 3). Indeed, the ILO believes that decent work not only makes an important contribution to sustainable development, but that it is also an important objective in its own right. Since work is such a major part of life in terms of time, social integration and individual self-esteem, decent work is clearly a fundamental dimension of the quality of life. Productive work is also the main source of income for the vast majority of people.

Every person at work or looking for work, whatever his or her country, occupation or skill level has a notion of what "decency" at work stands for. But what is it exactly? How can it be measured so that the decency of different jobs, enterprises and countries can be compared and monitored?

This article takes up the challenge of objectively measuring decent work with statistical indicators. Its main objectives are: (i) to translate the general concept of decent work into easily understandable characteristics of work, (ii) to identify statistical indicators with which those characteristics can be measured right now with an acceptable degree of consistency, accuracy and cross-country comparability, and (iii) to suggest further statistical activities and indicators to improve the measurement of decent work in the future. The ultimate aim is to provide the basis for developing a core set of ILO decent work indicators in order to focus efforts on and monitor progress towards decent work around the world, while empirically documenting relationships between different aspects of decent work and between decent work, poverty and economic performance.

The first section of the article discusses some important policy issues which can be addressed with statistical indicators of decent work. The second looks at the concept of decent work and its implications for the construction of statistical indicators. The third section begins by translating the concept of decent work into easily understandable, broad characteristics of work. This section then goes on to introduce and describe specific statistical indicators for each of those characteristics of decent work, including not only suggested indicators for immediate measurement, but also possible developmental work for improved measurement in the future. A final section draws conclusions and provides recommendations about what needs to be done to compile the suggested indicators of decent work for a sizeable number of countries in all regions of the world.

It should be emphasized that this article does not provide a final set of ILO decent work indicators, even though substantial thought and internal ILO discussion have gone into identification of the suggested indicators. To arrive at a final core set of ILO decent work indicators, further measurement and testing will need to be taken into consideration. In any case, it is important to note that even after a core set of ILO decent work indicators has been established, it will not be final or complete in at least two very important ways. First, improved measurement over time will allow for a broader core set of indicators in the future. And second, countries, regions and technical programmes will require additional indicators to meet their specific needs.

Decent work indicators and policy

Data collection and statistics concerning labour have traditionally focused on employment and unemployment, with the latter grabbing most of the headlines. This is clearly inadequate. The volume of employment generated by an economy at any point in time does not tell us much about the quality-of-life or "person-enabling" characteristics of that employment. (1)

A few of the areas in which comprehensive decent work indicators and data could shed new light are discussed briefly below.

Poverty reduction

The United Nations Millennium Assembly adopted the ambitious target of reducing by half the number of persons living in extreme poverty by 2015 (from its 1990 level). Current knowledge about poverty strongly suggests that economic growth is a necessary precondition for sustained poverty reduction, but not sufficient in itself. It is the different possible combinations of economic, employment, social and poverty-alleviation policies that are of particular interest to countries committed to poverty alleviation. Indeed, aggregate data on the share of the population living in extreme poverty do not reveal whether individual cases of poverty result from insufficient employment, low pay, inability to work due to an uncompensated occupational injury, discrimination, lack of access to social protection, unavailability of a basic pension, or one of the many other labour market factors. Better measurement of decent work--especially of the scope of social protection and the incidence of low pay and excessive hours--will provide fresh insights into the many policy options for addressing poverty. Understanding decent work in context and its relationship to poverty, therefore, should be a critical aspect of the struggle to reduce extreme poverty.

Social dimensions of globalization and sustainable development

It is possible to point to a number of countries that have decisively reduced poverty. Establishing through quantitative analysis that decent work promotes high and sustained economic growth and social development is more difficult. The ILO view is that countries are in a better position to benefit from globalization with an appropriate balance between economic and social development, and that this in turn leads to more decent work. New data on decent work and empirical analysis are needed to substantiate these views. Similarly, empirical analysis is needed to assist policy-makers in finding where the appropriate balance lies for a given country--as between economic and social development--to help ensure sustained economic growth.

Better jobs, better lives

Better and more comprehensive measurement of decent work will allow for a more detailed assessment of the mechanisms by which economic growth translates into higher standards of human welfare, and how these in turn lay the ground for faster economic and social development. (2) Since decent work has many dimensions, the combinations and patterns prevailing among particular demographic and socio-economic groups are of special interest. Comprehensive data on decent work could usefully inform analysts, observers and policy-makers about many other relationships as well.

Counting decent jobs and decent establishments

How many decent jobs and workplaces are there in a country? What percentages of a country's workers and workplaces have decent jobs? These are important questions that can only be answered using individual-job-level or establishment-level data. Indeed, to do so, one first needs to determine the decency of each job and workplace, and this requires information on all aspects of decent work for each individual or establishment (or a sample of them). First though, it is necessary to define what constitutes a "decent job" and a "decent workplace". For example, is a job decent if it embodies some aspects of decent work but not others, such as high pay and the right to organize but limited social protection? What if a job has high pay, the right to organize and social protection but excessively long hours? Since it is possible to set criteria of "decency", the first two questions above can be answered with appropriate micro-level data. This has been done by the European Union. Using micro-level data on three dimensions of job quality (job security, access to training and career development, and hourly wages), researchers have defined what they describe as dead-end, low pay/low productivity, reasonable and good jobs and estimated that they respectively account for 8, 17, 37 and 38 per cent of all jobs in the European Union (European Commission, 2001). Whenever possible, it is a good idea to answer the above questions using micro-level data and transparent criteria.

Assessing country performance with a decent work index

Is decent work more prevalent in country A or country B? Has decent work improved in country A and/or country B? Such questions are commonly asked by the media, the public and local and national leaders. They are typically answered using average national values for a series of indicators that have been aggregated into an index. In this sense, an ILO decent work index--analogous to the UNDP's Human Development Index--could have considerable value as it would help to broaden the common view of labour issues beyond the current overemphasis on employment and unemployment. However, there are major technical problems associated with composite indexes. These include the subjective judgement required to weight different indicators; situations where national data are missing for a specific indicator; and the need to maintain simplicity and transparency while covering all aspects of decent work. Failure to address such problems adequately could damage the credibility of the index. Accordingly, the ILO is now investigating the possibility of developing a decent work index by weighing its large potential value against the major technical and practical difficulties that would be involved in constructing and applying it. (3)

Conceptualizing decent work

The six dimensions of decent work

The definition of decent work--as "opportunities for women and men to obtain decent and productive work in conditions of freedom, equity, security and human dignity"--explicitly includes six dimensions.

First, opportunities for work refers to the need for all persons who want work to be able to find work, since decent work is obviously not possible without work itself. The underlying concept of work is a broad one, encompassing all forms of economic activity, including self-employment, unpaid family work and wage employment in the informal and formal sectors.

Second, the idea of work in conditions of freedom underscores the fact that work should be freely chosen--i.e. not forced on individuals--and that certain forms of work are not acceptable in the twenty-first century. Specifically, this means that bonded labour, slave labour and the worst forms of child labour should be eliminated in accordance with applicable international Conventions. It also means that workers should be free to join workers' organizations and free from discrimination.

Third, productive work is essential for workers to have acceptable livelihoods for themselves and their families, as well as to ensure sustainable development and the competitiveness of enterprises and countries.

Fourth, the notion of equity in work represents workers' need to enjoy fair and equitable treatment and opportunity in work. It encompasses absence of discrimination at work and in access to work, and the possibility of balancing work with family life.

Fifth, security at work is a reminder of the need to safeguard health, pensions and livelihoods, and to provide adequate financial and other protection in the event of sickness and other contingencies. It also recognizes workers' need to limit the insecurity associated with the possibility of loss of work and livelihood.

Sixth, dignity at work requires that workers be treated with respect at work, and that they be able to voice their concerns and participate in decision-making about their own working conditions. An essential aspect of this is workers' freedom to represent their interests collectively.

The first two dimensions of decent work--opportunities for work and freedom of choice of employment--are concerned with the availability and basic acceptability of work. The other four dimensions--productive work, equity, security and dignity--are concerned with the extent to which available and freely accepted work is "decent". In many ways, this is similar to what the European Union calls "quality of employment" (see, for example, Paoli and Merlli6, 2000; European Commission, 2001, ch. 4). In addition to these six dimensions of decent work, the socio-economic context is important too, since this partly determines both what constitutes decency in a given society and the extent to which the achievement of decent work enhances national economic, social and labour market performance.

Implications of the decent work concept for statistical indicators

At the outset, it is important to highlight the main implications of the decent work concept for the identification of statistical indicators.

Firstly, decent work, and therefore decent work indicators, should be relevant to everyone--to men and women, to people in high- and low-income countries, and to work performed in the modern and traditional sectors alike.

Secondly, some aspects of decent work are absolute in that the same standard applies to everyone in every country (e.g. fundamental rights at work), while other aspects are relative in that each country and society evolves its own norms of decency. For example, the level of pay and working conditions considered to be decent differs across countries (even though the principle that as many people as possible should …