Chapter 6: How to enforce minimum wages

6.2 Measuring non-compliance

In order to design, put in place and pursue sound strategies and measures for compliance and enforcement it is important to analyse the extent and patterns of non-compliance in a manner as detailed as the available data permit. This is not only relevant for enhancing the effectiveness of interventions but also to make strategic choices in allocating available public resources.

Studies can look at the extent of non-compliance according to urban and rural areas, the formal and informal economies, different economic sectors and occupations, size of the enterprise, and how non-compliance affects women and men. Compliance gap assessments can also look at the situation of groups more vulnerable to workplace discrimination, such as indigenous and tribal peoples, ethnic minorities, migrants or people with disabilities.

Compliance can be estimated in three ways:
  • The first method, which is extensively used in empirical studies, is to calculate from statistical data the percentage of workers earning less than the legal minimum wage in different jobs and sectors. This method provides rough estimates and contains measurement error.
  • A second measure is based on violations of minimum wage laws detected during workplace inspections. However, an analysis based on labour inspections can be only partial, as it refers exclusively to detected cases.
  • A third measure could be the number of complaints made by workers to enforcement bodies and courts. However, it is difficult to obtain the necessary administrative data, and not everyone who is underpaid lodges a complaint.

An illustration of the statistical methodology1

The Figures below use the first method and provide an estimate of the proportion of legally covered workers earning less than the legal minimum wage, taking into account imperfect data on their hours of work. Information on wages comes from household or labour force surveys, and wages are compared to the applicable legal minimum wages. Such a method allows comparison across countries and over time. It must be kept in mind, however, that they are only statistical estimates.

Using such a method, compliance rates vary widely from country to country. Figure 1 shows that the share of covered workers earning less than the legal minimum wage varies from less than 10 per cent in Viet Nam, Chile, Uruguay or Mexico, up to about 50 per cent in Turkey or Indonesia.

Figure 1. Non-compliance rate (percentage of employees paid less than minimum wage), latest year for which data are available
Source: Rani et al., 2013; Marinakis (2014).

Different population groups

Within countries, there are significant differences in non-compliance rates across different groups of the population.

The case of India illustrates a situation that is common across countries: non-compliance is more widespread in rural than in urban areas, and in the informal than in the formal economy. Women are also frequently more likely to be underpaid than men, as are disadvantaged ethnic or social groups such as so-called scheduled castes and scheduled tribes in India (Figure 2).

Similarly, if we look at non-compliance across different industries, the case of South Africa shows non-compliance to be particularly high in construction and agriculture (Figure 3).

Looking at the non-compliance rate by size of the enterprise, then the case of Costa Rica shows that non-compliance tends to be higher in smaller than in large enterprises (Figure 4).

Figure 2. Non-compliance rate in India, by different categories of workers, 2009–10
Source: Rani et al., 2013

Figure 3. Non-compliance rate in South Africa, by industry groups, 2011
Source: Rani et al., 2013.

Figure 4. Non-compliance rate in Costa Rica, by size of enterprise
Source: ILO based on labour force surveys.


1 Rani et al., (2013). “Minimum wage coverage and compliance in developing countries”. International Labour Review, Vol.152, No.3-4, pp.381-410.