The information in this web page is based on personal communications with Françoise Carré and Joann Vanek (2010) of the WIEGO Statistics Programme.
The first measurement challenge is a basic one: what types of activities should be included as domestic work in the statistical definition used in collecting and tabulating data and how can this definition be implemented given the three major classifications for employment characteristics?
There is no common statistical definition across countries for what constitutes domestic work. In its report Decent Work for Domestic Workers, the International Labour Organization uses a statistical definition of domestic work that is based on the International Standard Classification of Occupations (ISCO) (ILO 2010). It is a comprehensive definition capturing most of the relevant categories of persons employed in domestic work.
Given the nature of the ISCO framework, there can be no discrete major level occupational category that encompasses all types of work done by domestic workers. The ILO statistical definition includes four occupational codes at the four-digit level of the classification: three under ISCO-88 Major Group 5: Service Workers and Shop and Market Sales Workersand one under ISCO-88 Major Group 9: Elementary Occupations, as follows:
- 5121 – Housekeepers and related workers
- 5131 – Child care workers
- 5133 – Home-based personal care workers
- 9139 – Domestic helpers and cleaners
There are challenges, however, in using this definition to collect, tabulate, and compile data on domestic workers. The definition requires tabulating four different occupational sub-sub-categories at the four-digit level. But data are often not collected/coded/tabulated at the four digit level of the occupational code, as financial and human resources are needed to do so. Another problem is that in preparing international databases, countries are usually not asked to supply data at the four-digit level of disaggregation. For example, the ILO collects and disseminates data on employment by occupation at only the one-digit level of ISCO in the LABORSTA database.
An additional problem is that not all workers who might be considered domestic workers are included in the occupations identified in the ILO report. In particular, for private security guards, chauffers and gardeners, ISCO includes codes but does not have separate codes for those employed by private households.
Given the problems in collecting/tabulating/reporting data at the four-digit level, an alternative approach to measuring domestic workers is based on the International Standard Industrial Classification of All Economic Activities (ISIC) Category P: Private Households with Employed Persons. This one-digit category includes the activities of private households employing all kinds of domestic personnel such as those we usually think of as doing domestic work – housekeepers and cleaners, cooks, waiters, valets, butlers, laundry workers – as well as others: gardeners, gatekeepers, stable hands, chauffeurs, caretakers, governesses, babysitters, tutors, and so on. Currently it is probably the best available approach for estimating the situation of domestic workers at the international level because it is available for more countries than the ISCO approach. It was used by the ILO in preparing data for the ILO report cited earlier. However, if the definition is accurately implemented, this ISIC-based measurement leaves out a growing number of domestic workers, especially in developed countries, who are employed through intermediary brokers or services to provide health care, maid, cleaning, or security services.
To improve data on domestic workers, statisticians and advocates need to address the following questions:
- Should the definition of domestic workers include workers who are hired and paid by an intermediary agency to provide domestic services? Are the characteristics and conditions of work sufficiently similar between domestic workers who are hired by individuals/households and those hired through agencies to include both within the same statistical category? Some argue that since these workers are not hired by private households, the nature of their employment arrangements is totally different than contract workers and, therefore, there should be two separate categories: one for domestic workers hired by agencies, the other for domestic workers hired by households. However in terms of policy goals both categories of workers would benefit from similar or equivalent access to social protection, decent wages and conditions of work.
- How significant is the number of domestic workers hired by agencies relative to domestic workers hired by households in developed countries? How many of these workers are not reflected in data as ISIC category P? This could perhaps be tested using micro-data files for a few countries.
In addition to these main measurement challenges, there are other basic statistical issues that need to be considered when compiling and analyzing data on domestic work. First, many domestic workers are migrants – often undeclared workers – while others, particularly younger workers, are relatives or kin “who do domestic work in exchange for ‘board and lodgings’” (Lund and Budlender 2009). These workers may not be captured in labour force surveys, particularly in developed countries.
Second, not all domestic workers are working informally. If, for example, an employer of a domestic worker makes a contribution to a public pension fund, the worker is considered to be formally employed. Delineating which domestic workers are formal or informal is not always easy depending on the questions asked of and the tabulations produced on domestic workers.
Third, not all countries collect data on where temporary or contract workers are employed. As a consequence, it would not be possible to determine which temporary or contract workers are working in homes belonging to others.
The ILO and the WIEGO network are working with the International Expert Group on Informal Sector Statistics (known as the Delhi Group),national statistical offices, and national data analysts and advocates to improve statistical measures and data on domestic work: the data on urban domestic work in Latin America presented on this website is one outcome of this joint effort.