From Fakes to Forced Labour

3. At first glance

To strengthen the argument about a link between illicit trade in counterfeits and labour exploitation, it is essential to rely on quantitative evidence. Anecdotal reports and case studies provide useful context, but they cannot establish broader patterns or correlations across economies. Only by employing systematic data can we assess whether countries with weak labour standards, higher incidences of exploitation, or deficient enforcement are those more likely to be involved in the production or transit of counterfeit goods.

The general concept underpinning the analysis is straightforward. The underlying premise is that more flexible or weakly enforced worker protections, coupled with unsafe and exploitative working conditions, may create an enabling environment for illicit trade activities. In such environments, criminals can minimise costs by disregarding labour rights and occupational safety, while simultaneously exploiting institutional weaknesses that reduce the risk of detection or punishment. The quantitative analysis therefore aims to test whether measurable indicators of labour exploitation and weak labour governance are statistically associated with indicators of counterfeiting.

3.1. Measuring exploitative and informal labour practices

There are various levels of labour exploitation. Some indicators reflect systemic governance failures and the overall degree of respect for law in a country, while others point to specific instances of abuse, such as child labour or forced labour. One useful way to conceptualise this is through Skrivankova’s (2010[22]) definition of the continuum of exploitation, which ranges from decent work to the most severe forms such as forced labour and human trafficking. At the outset, it is difficult to select a single dimension that might best capture the relevant dynamics. For this reason, the present analysis considers a broad set of variables, acknowledging that labour exploitation manifests in different forms depending on institutional and economic contexts.

The first group of indicators measures different forms of labour exploitation. These indicators collectively capture the most severe forms of labour exploitation and serve as proxies for environments where criminal networks can thrive by exploiting vulnerable groups. There are three proxies in this group, all sourced from the ILOSTAT database of the ILO.

• Child labour. This variable measures the proportion of children engaged in economic activities that are classified as child labour, according to the International Labour Organization (ILO) definitions. It is expressed as a percentage of the total population of children aged 5‑17 years. The indicator captures work that is mentally, physically, socially, or morally harmful to children, or that interferes with their schooling. Data are collected from national household surveys and harmonised by the ILO to ensure comparability across countries and years.

• Children in Hazardous Labour. This variable measures the proportion of children engaged in work that is considered hazardous according to ILO standards. Hazardous labour includes tasks or conditions that are likely to harm the health, safety, or moral development of children — e.g. work involving dangerous machinery, exposure to harmful substances, extreme temperatures, long hours, or night shifts. It is usually expressed as a percentage of the total child population aged 5‑17 years. Data are derived from national household surveys and harmonised by the ILO to ensure cross-country comparability.

• Forced Labour. This variable measures the prevalence of individuals engaged in forced labour, as defined by the ILO. Forced labour refers to all work or service that is extracted from a person under threat of penalty and for which the person has not offered themselves voluntarily. This includes situations of coercion, deception, restriction of movement, withholding of wages or identity documents, and physical or sexual violence. The indicator is expressed as the number of victims and converted for analytical purposes into a percentage of the active population. Data are compiled by the ILO from national surveys, administrative records, and specialised studies, and are harmonised to ensure comparability across countries and time periods.

A second set of indicators captures occupational safety risks, which may also serve as indirect proxies for exploitative labour environments. High fatality rates may point to unsafe workplaces, weak inspections, and minimal employer accountability – conditions that overlap with environments conducive to counterfeiting operations. There is one proxy in this set, sourced from ILOSTAT.

• Fatal injuries. This variable measures the incidence of work-related fatal injuries, defined as any occupational injury resulting in death occurring in the course of work or while performing work-related duties. The indicator is typically expressed as the number of fatalities per 100 000 workers, covering employees and, in some cases, self-employed workers. Data are collected from national administrative records, labour inspections, and insurance schemes, and then harmonised by the ILO to improve international comparability.

The last set of indicators relates to the institutional framework governing labour relations and the degree of rule of law. These proxies assess whether strong legal frameworks and functioning social dialogue mechanisms mitigate the risks of labour exploitation and, by extension, involvement in illicit trade. There are four proxies in this set.

• Collective bargaining. This variable measures the proportion of employees whose working conditions are regulated by a collective agreement between employers (or employers’ organisations) and workers’ organisations, in line with ILO definitions. It reflects the extent of collective bargaining coverage within the workforce and serves as an indicator of the strength of social dialogue and labour relations in a country. The measure is expressed as a percentage of all employees and is derived from national labour force surveys or administrative records. Data are harmonised by the ILO to allow cross-country and time-series comparability.

• Informal employment. This variable measures the proportion of employed persons whose jobs are not covered by national labour legislation, income taxation, social protection, or entitlement to employment benefits. It includes both self-employed workers and employees in informal jobs, whether in the formal sector, informal sector, or households. The indicator is expressed as a percentage of total employment and is based on data from national labour force surveys or population censuses. The ILO harmonises these figures to ensure cross-country and time-series comparability.

• Trade union rate. This variable measures the proportion of employees who are members of a trade union, based on national definitions. It reflects the degree of unionisation within the workforce and serves as an indicator of workers’ capacity for collective representation. The measure is expressed as a percentage of all employees and is derived from national labour force surveys, administrative records from trade unions, or labour ministries. The ILO harmonises the data to enhance comparability across countries and over time.

3.1.1. Limitations of the data

Several caveats must be acknowledged. First, not all countries that appear in customs seizure records are producers of counterfeit goods. Some countries serve primarily as transit hubs, leveraging logistics infrastructure that is exploited by counterfeiters. Producers are predominantly concentrated in Asia, but transit economies are geographically dispersed, ranging from advanced trading hubs to developing economies. This heterogeneity makes it difficult to assign labour exploitation dynamics directly to counterfeiting outcomes.

Second, transit activities themselves can be labour-intensive. Counterfeiters often use warehouses in transit points to relabel, repackage, or assemble counterfeit products. Such activities – although occurring outside the production stage – can still involve abusive or unsafe labour practices. As a result, transit economies may appear prominently in counterfeiting indicators, while also facing risks of labour exploitation.

Third, other structural factors – such as weak IP enforcement, corruption, broader deficiencies in rule of law, or policies that encourage the use of forced labour – may be correlated with labour exploitation and with counterfeiting. This overlap may generate correlations among the explanatory variables, which could make it difficult to identify individual effects. The primary aim of this exercise is descriptive: documenting statistical relationships and patterns in the data. Therefore, coefficients should be interpreted with caution as they may capture correlations rather than causal effects.

3.2. Quantifying the scope and scale of counterfeit trade

The labour-related indicators are combined with OECD/EUIPO data on counterfeiting, which relies primarily on customs seizure data collected from:

• the World Customs Organization (WCO),

• the European Commission’s Directorate-General for Taxation and Customs Union (DG TAXUD), and

• the United States Department of Homeland Security (DHS), including CBP and ICE.

From these sources, three proxies of counterfeiting are used:

• The number of seizures by provenance economies. This index is based on the OECD customs seizure dataset. See Annex A of OECD/EUIPO (2025[23]) for the methodology used to construct this dataset.

• The GTRIC-e score, which is based on OECD measures of the relative likelihood an economy exports counterfeit goods. It is a weighted indicator combining (i) the absolute value of counterfeit and pirated exports and (ii) their share in the economy’s total trade. For example, although Benin’s absolute value of counterfeit exports is low, the high proportion within its total exports results in a high score. In general, economies with high GTRIC-e scores either export large absolute values of counterfeit goods or have a high share of such goods in their total exports.

• The estimated value of counterfeit goods by provenance economies results in a methodology developed by the OECD which is based on the GTRIC-e described above. The estimated values of fakes used in this study are derived from OECD/EUIPO (2025[23]).

Together, these measures provide a comprehensive picture of economies’ roles in counterfeiting, which can then be analysed alongside labour exploitation and governance indicators.

3.3. Patterns and links between labour abuse and counterfeiting

As a first step, simple correlations are presented between various proxies for labour market conditions and measures of illicit trade. These correlations provide an initial indication of whether a positive or negative association exists between the variables under consideration, without implying any direct causal relationship.

3.3.1. Illicit trade and labour exploitation

Forced labour

The analysis reveals a clear positive correlation between the estimated value of counterfeit goods exported by a country and the number of victims of forced labour (Figure 3.1) reported by this country. In other words, countries with higher estimated values of fakes are more frequently associated with higher levels of forced labour.

As shown in Figure 3.2, a similar relationship emerges when examining enforcement data. Specifically, the number of seizures identifying the economies of provenance for counterfeit goods also shows a positive correlation with the prevalence of forced labour.

Figure 3.1. Correlation between the value of fakes (in log) and the number of forced labour victims (in log)

Note: Data are shown on a logarithmic scale to better compare values that differ widely in magnitude. Each point corresponds to one country for 2021. Correlation coefficient = 0.42.

Source: ILOSTAT and OECD global customs seizures.

Figure 3.2. Correlation between the number of seizures (in log) and the number of forced labour victims (in log)

Note: Data are shown on a logarithmic scale to better compare values that differ widely in magnitude. Each point corresponds to one country for 2021. Correlation coefficient = 0.38.

Source: ILOSTAT and OECD global customs seizures.

Child labour

The analysis reveals a positive correlation, albeit moderate, between the likelihood of a country being identified as a source of illicit trade and the prevalence of child labour. In other words, countries with a higher probability of being associated with illicit trade activities also tend to report higher percentages of child labour. Although the relationship is not strong, this relationship suggests that underlying socio-economic and governance factors may simultaneously drive both the incidence of illicit trade and the exploitation of child labour.

Figure 3.3. Correlation between child labour and GTRIC-e

Note: Data are shown on a logarithmic scale to better compare values that differ widely in magnitude. Each point corresponds to one country for 2021. Correlation coefficient = 0.11.

Source: ILOSTAT and OECD global customs seizure.

Children in hazardous labour

The analysis reveals a positive correlation between the number of seizures in which a country is identified as a source of illicit trade and the percentage of children engaged in hazardous labour. In other words, countries more frequently cited as origins of illicit trade tend to report higher proportions of children involved in work that poses significant risks to their health, safety, or moral development.

Figure 3.4. Correlation between children in hazardous labour and number of seizures

Note: Data are shown on a logarithmic scale to better compare values that differ widely in magnitude. Each point corresponds to one country for 2021. Correlation coefficient=0.12.

Source: ILOSTAT and OECD global customs seizure.

3.3.2. Illicit trade and labour safety

Fatal occupational injuries

Data indicate a positive correlation between the number of seizures linked to a country as a source of illicit trade and its rate of fatal occupational injuries per 100 000 workers, suggesting that countries more frequently tied to illicit trade also face higher workplace fatality risks.

Figure 3.5 Correlation between number of seizures (in log) and fatal occupational injuries per 100 000 workers

Note: Data are shown on a logarithmic scale to better compare values that differ widely in magnitude. Each point corresponds to one country for 2021. Correlation coefficient=0.34.

Source: ILOSTAT and OECD global customs seizure.

The comparison between Figure 3.5 and Figure 3.6 suggests that the correlation between injuries and the monetary value of seizures is considerably weaker than that between injuries and the overall number of seizures. This pattern may imply that injuries are particularly associated with the production of lower-value counterfeit goods, perhaps reflecting more hazardous or less controlled manufacturing environments characteristic of low-end counterfeit production.

Figure 3.6. Correlation between value of fakes (in log) and fatal occupational injuries per 100 000 workers

Note: Data are shown on a logarithmic scale to better compare values that differ widely in magnitude. Each point corresponds to one country for 2021. Correlation coefficient=0.1.

Source: ILOSTAT and OECD global customs seizures.

3.3.3. Labour rights and protection

Informal employment

The analysis shows a positive correlation between the likelihood of a country being identified as a source of illicit trade and the percentage of informal employment within its labour market. In other words, countries with a higher probability of being associated with illicit trade tend to report greater shares of workers employed in informal, unregulated, or unprotected jobs.

Figure 3.7. Correlation between GTRIC-e and informal employment

Note: Each point corresponds to one country for 2021. Correlation coefficient=0.32.

Source: OECD global customs seizure.

Trade union membership rate

The analysis reveals a negative correlation between the likelihood of a country being identified as a source of illicit trade and its trade union membership rate (Figure 3.8). In other words, countries with a higher probability of being associated with illicit trade activities tend to have lower levels of trade union representation within their workforce. Figure 3.9 also indicates a negative correlation between the number of seizures identifying countries as provenance of counterfeits and countries’ presence of trade union.

Figure 3.8. Correlation between GTRIC-e and trade union rate

Note: Each point corresponds to one country for 2021. Correlation coefficient=-0.2.

Source: ILOSTAT and OECD global customs seizure.

Figure 3.9. Correlation between number of seizures (in log) and trade union rate

Note: Data are shown on a logarithmic scale to better compare values that differ widely in magnitude. Each point corresponds to one country for 2021. Correlation coefficient=-0.19.

Source: ILOSTAT and OECD global customs seizure.

Collective bargaining coverage

The analysis reveals a negative correlation between the likelihood of a country being identified as a source economy for illicit trade – as well as the number of seizures attributed to that country – and its collective bargaining rate. In other words, countries more frequently linked to illicit trade, whether in terms of probability or actual recorded seizures, tend to have lower levels of collective bargaining coverage among their workforces.

Figure 3.10. Correlation between GTRIC-e and collective bargaining coverage rate

Note: Each point corresponds to one country for 2021. Correlation coefficient=-0.27.

Source: ILOSTAT and OECD global customs seizures.

Figure 3.11. Correlation between the number of seizures and collective bargaining coverage rate

Note: Data are shown on a logarithmic scale to better compare values that differ widely in magnitude. Each point corresponds to one country for 2021. Correlation coefficient=-0.17.

Source: ILOSTAT and OECD global customs seizures.

These inverse relationships may reflect underlying factors such as weaker labour institutions, limited worker protections, and reduced collective bargaining capacity, all of which can create environments where illicit economic activities are more likely to flourish. Lower trade union density may also be indicative of broader governance and enforcement gaps that enable both labour exploitation and the growth of illicit trade networks.

3.4. What correlations suggest for enforcement and countering labour abuse

The empirical analysis identifies several statistically significant correlations between indicators of illicit trade and labour market variables. These correlations do not establish causality, but provide important first insights into the structural conditions in which illicit trade may occur.

Positive correlations emerge in several dimensions. Countries with higher probabilities of being identified as sources of counterfeit trade display higher levels of child labour, including hazardous forms of child labour and a greater prevalence of informal employment. Similarly, seizure data – capturing the number of cases in which a country is identified as a source of counterfeit goods – correlates positively with indicators such as average weekly working hours, incidence of fatal occupational injuries, prevalence of forced labour, and weaker labour rights protections. In addition, a positive correlation is observed between the estimated value of counterfeit goods and the prevalence of forced labour.

Negative correlations were also identified. Higher probabilities of being a source of counterfeit trade are associated with lower levels of trade union membership, while both the probability of being a source country and the number of seizures linked to a country show negative correlations with collective bargaining coverage. However, these results should be interpreted with caution. Unionisation rates and bargaining coverage are influenced by a wide range of country-specific factors – including industrial structure, labour law design, and cultural traditions of social dialogue – which may be omitted variables related both to the labour conditions and the dynamics of illicit trade.

The underlying drivers of these observed associations merit closer attention. High levels of child labour and forced labour are likely to reflect broader deficiencies in regulatory enforcement and weak institutional capacity, which can simultaneously lower the cost of production for illicit operators and reduce the probability of detection. Detecting forced labour is challenging even within legitimate supply chains subject to audits and inspections; in illicit activities, where such mechanisms are absent and concealment is deliberate, it becomes exponentially more difficult. Similarly, high levels of informality tend to undermine compliance with national labour standards and tax regimes, creating structural incentives for illicit trade to proliferate in parallel with informal economic activity. Excessive working hours and high occupational injury rates may capture broader patterns of poor occupational health and safety enforcement, reflecting weak governance frameworks that also extend to trade oversight.

The negative correlations with trade union membership and collective bargaining coverage suggest that weak labour representation may reduce the bargaining power of workers, lowering wage floors, and weakening protections against exploitation. Such conditions could facilitate the establishment and operation of illicit trade networks by reducing the likelihood of labour-related resistance or exposure. In this sense, weak institutionalised forms of social dialogue may contribute to a permissive environment for illicit activities.

It is important to emphasise that these are correlations, not causal mechanisms. Observed relationships may be influenced by third factors, such as broader governance quality, economic structure, or patterns of industrial development. For example, countries with a large informal economy may simultaneously face challenges of weak labour protection and higher exposure to illicit trade, without one necessarily driving the other directly. Similarly, correlations with hazardous child labour and forced labour may partly reflect underlying levels of economic vulnerability and limited alternative income opportunities.

In conclusion, while the patterns observed provide strong indicative evidence of links between illicit trade and adverse labour market conditions, they remain preliminary. An economic modelling approach – e.g. applying multivariate regression analysis or structural equation modelling – is needed to test the robustness of these associations and to control for confounding variables. Such an approach, presented in the next section, assesses whether the correlations identified here hold once broader macroeconomic, institutional, and governance factors are accounted for.