EUROSTAT: Smuggling of Migrants


Migrant smuggling is a profitable business for criminal networks that exploit the desperation and vulnerability of migrants seeking to escape armed conflict, persecution and deprivation. Facilitators offer a broad range of services such as transportation, accommodation and fraudulent documents at excessively high prices. According to Europol, more than 90 % of irregular migrants use these ‘facilitation services’ and in 2015 alone the estimated annual turnover related to migrant smuggling was reached EUR 3–6 billion, with possible scenarios projecting a double or even a triple increase1.

Migrant smuggling therefore is a source of significant financial flows between residents and non-residents. Failing to capture or record these flows correctly could potentially cause significant distortions in the external sector statistics. Furthermore, facilitation of illegal migration is an illegal economic activity which falls within the production boundary of national accounts and therefore should in theory be included in the statistical estimates of economic output.

Scope, Methodology, Compilation Practices, and Data Sources

In the European Union, Council Directive 2002/90/EC of 28 November 20022 provides a common definition for assisting illegal immigration. It includes the following infringements:

  • Assisting intentionally a non-EU national to enter or transit through an EU country, in breach of the law.
  • Assisting intentionally, and for financial gain, a non-EU national to reside in an EU country, in breach of the law.
  • Instigating, taking part in or attempting to commit the above acts.

It should be noted that migrant smuggling is different from human trafficking whereas the former is an activity into which the parties involved enter by mutual agreement (i.e. with the consent of the persons being smuggled). By contrast, human trafficking is an activity against an individual with no mutual agreement. In other words, migrant smuggling is a transaction where irregular migrants are not forced to move and it is a resident-non-resident transaction. If the migrant is forced to move it is classified as human trafficking, not as an illegal economic activity3.

There are two main types of agreement for migrant smuggling:

  • In the 'pay-as-you-go' agreement, no final destination is predetermined, and the speed and direction depend on migrants’ ability to pay at each step. In this case, the role of the smuggling coordinator could be reduced from the base model.
  • The ‘full package’ agreement is a less common model, where migrants pay the fee in their country of origin to a smuggler that arranges several services to have the migrant transported to the destination country.

Given that ‘pay-as-you-go’ is the dominant mode of migration, data compilers in the destination and transit countries would be interested in the transactions between the resident smugglers and the migrants, who by definition are non-resident. Therefore, from an EU perspective, models on migrant smuggling could be reduced to estimating the effects of border-crossing and transiting-through. These models could disregard migrants’ consumption of smuggling services before the point of entry. Frontex data suggest that EU border crossing is in many cases facilitated by non-resident smugglers4 , therefore only a part of the related transactions is relevant for the EU Member States balance of payments. However transiting through the EU is more likely to be fully operated by resident service providers and in this case transactions between the migrants and the facilitators should be recorded as transportation and travel services.

Data sources on migrant smugglers’ fees could be police reports, interview-based media publications and information from social media. Prices may depend on factors such as the border type (land or sea), the types of services provided by the smugglers, and the risks they bear. Prices are further affected by seasonality and by supply and demand shifts.

On the number of smuggled migrants, EU data compilers could use data on detections by police authorities and Frontex. However, these data should be adjusted with detection rates to reflect the true numbers. In more global terms data on irregular migration is available at the International Organization for Migration. There are also statistical methods suggested in the context of the US-Mexico illegal border crossing5 which could produce output needed for adjusting the nominal detection rates. These methods include:

  • capture-recapture methods6
  • stratified sampling of border crossings 7
  • surveys and respondent-driven sampling 8
  • synthetic and proxy measures 9

While at the moment there are no national practices that could be mentioned, for the balance of payments and national accounts purposes a demand-side models could be more feasible since, although seemingly distorted, there is a relatively better data on the numbers of irregular border crosses and the prices involved. As with other indirect estimates, it is recommended to stratify the data along the various available categories and possibly to construct price and quantity matrices which after multiplication should produce the total turnover from irregular migration. Some supply-side data, for example numbers of facilitators, could also be available and used to verify various assumptions and adjust the final estimation models. These data could also be essential for geographical allocation of the transactions as well as a possible input for estimating the associated intermediate costs.

Current Challenges and Conclusions

Trafficking of migrants is not among the so-called 'core' illegal economic activities for the compilers of macroeconomics statistics in the EU, i.e. the minimum set that is obligatorily estimated for the purposes of national accounts and balance of payments including drugs, prostitution and smuggling of tobacco and alcohol. There are currently no official estimates and country case studies that could be cited. However, the financial flows related to migrant smuggling have an increasing significance especially for Member States that are in the main migration routes.

While data on irregular migrant headcounts and on service prices constitutes the main issue for the statistical compilation, there are also other challenges in capturing correctly the related economic effects in the balance of payments statistics.

Firstly, smuggling of migrants is often closely related to corruption and bribery which is an illegal activity of a different category. Furthermore, in certain cases, for example accommodation services for irregular migrants, there is not a clear delineation between illegal and underground economy. The overlap of the different concepts and categories therefore brings a potential risk of double counting.

Secondly, due to the specific nature of illegal migration the issue of migrants’ consent to be smuggled is a not straight-forward one and it is 'possible that smuggled migrants might retract their consent during a smuggling operation' . There are thus borderline cases for which the concept of an economic transaction may or may not hold and corresponding adjustments would be necessary.

Last but not least, there is the issue of the irregular migrants' residence. In many cases the process of migration could take several months, if not years. Irregular migrants might spend a considerable amount of time in transit hubs until they eventually reach their destination with a clear intention to change their centre of economic interest. As a result tracking and timing the irregular migrants' change of residence could be a complex task.

1European Commission, Europol,

2Council Directive 2002/90/EC of 28 November 2002 defining the facilitation of unauthorised entry, transit and residence,

3European Commission (2018), ‘Handbook on the compilation of statistics on illegal economic activities in national accounts and balance of payments’, Luxembourg.

4European Border and Coast Guard Agency (2017), ‘FRAN Quarterly Q1 2017’, p. 25.

5Morral, A., Willis, H. and Brownell, P. (2011), ‘Measuring Illegal Border Crossing Between Ports of Entry: An Assessment of Four Promising Methods’, RAND Corporation Homeland Security and Defense Center.

6In a nutshell the model is built on estimating the probability on repeated attempts to irregular entry of border and using it to adjust the nominal detection rates. A major assumption in this model is to account for discouraged migrants, i.e. those who do not attempt further entries once turned back by border control.

7In this method detection rates could be adjusted depending on known parameters. For example, if border control resources are spread across various regions (strata) of the border depending on the risk of crossing in these regions, the proportions of resource allocation could be used to adjust total detection rate.

8Ibid. 'respondent-driven sampling begins with a nonrandom sample of individuals from the population of interest, interviewing them about their characteristics of interest (...) and then asking them to distribute invitations to participate in the survey to their friends'.

9For example, indicators derived by expert-based judgments, econometric models, mathematical simulations, etc.

10United Nations Office on Drugs and Crime (2011), Smuggling of Migrants: A Global Review, p.6