Misuse of E-Commerce for Trade in Counterfeits

3. The trade in fakes: A first glance

This chapter presents the results of a quantitative exercise to study the relations between e-commerce and trade in counterfeit goods.

The qualitative evidence outlined in the previous chapter suggests that e-commerce provides an increasingly attractive means to facilitate the trade in counterfeit goods for a large range of product categories. The purpose of this analysis is to discern if there is a positive correlation between measures of e-commerce and trade in counterfeits. To accomplish this, three datasets from Eurostat, the World Bank, and the United Nations Conference on Trade and Development (UNCTAD) will be used to calculate correlations between e-commerce and the number and value of counterfeit customs seizures.

These correlations between e-commerce activity and illicit trade in fakes, become stronger for indicators of illicit trade in counterfeits misusing small parcels. This suggests that illicit goods purchased through on-line transactions are often shipped with small parcels, sent either by mail or express and postal services.

It is important to note that this section simply analyses correlations between e-commerce and counterfeit illicit trade. With that said, the information and correlations from this section justify that deeper analysis is necessary to further understand e-commerce’s role in shaping global illicit trade.

Data

Following the approach taken in the previous studies, including the (OECD/EUIPO, 2016[1]), (OECD/EUIPO, 2019[3]) and (OECD/EUIPO, 2021[4]) reports, the analysis in this report is based on international trade statistics and customs seizures of infringing products. It also uses aggregated indicators on e-commerce.

Trade data

The trade statistics are based on the United Nations (UN) Comtrade database (based on the value of merchandise assigned by customs officials, i.e. the landed customs value). With 171 reporting economies and 247 partner economies, the database covers a majority of world trade and is considered the most comprehensive trade database available. Products are registered based on the six-digit Harmonised System (HS) (an international commodity classification system, developed and maintained by the World Customs Organization [WCO]), which means the level of detail in the data is high. Data used in this study are based on landed customs value. In most instances, this is the same as the transaction value appearing on accompanying invoices. Landed customs value includes the insurance and freight charges incurred when transporting goods from the economy of origin to the economy of importation.

Seizure data

Data on customs seizures originate from national customs administrations. This report relies on customs seizure data from the WCO, the European Commission’s Directorate-General for Taxation and Customs Union (DG TAXUD) and from the United States Department of Homeland Security (DHS). The latter submitted seizure data from US Customs and Border Protection (CBP), the American customs agency, and from the US Immigration and Customs Enforcement (ICE).

In each year analysed (2017, 2018 and 2019), the total number of customs seizures of counterfeit and pirated goods worldwide consistently exceeded 130 000. Overall, the unified database on customs seizures of IP-infringing goods includes almost 465 000 observations, as compared to the 428 000 recorded from 2011-13 (OECD/EUIPO, 2016[1]).

The database contains a wealth of information about illicit goods that can be used for quantitative and qualitative analysis. In most cases for each seizure the database reports: date of seizure, mode of transport of fake products, departure and destination economies, general statistical category of seized goods as well as their detailed description, name of legitimate brand owner, number of seized products and their approximate value. In addition, some customs data contains information related to e-commerce. In particular, some EU countries indicate in their data if the seized product was purchased on-line. The data contains a check-box, where customs officers indicate if it was the case.

A detailed analysis of the data reveals a set of limitations. Some of these limitations deal with discrepancies between the datasets, while others involve differing product classification levels or outliers in terms of seized goods or provenance economies. All limitations were thoroughly discussed in the (OECD/EUIPO, 2016[1]), (OECD/EUIPO, 2019[3]) and (OECD/EUIPO, 2021[4]) reports, and a methodological way forward was proposed for each limitation. This report also relies on the same methodology presented and discussed in the 2021 study, and it employs the same solutions to the seizure-data limitations.

E-commerce data

Measuring e-commerce is challenging since e-commerce definitions differ in various contexts, as discussed in the introductory section. Just like there is no single definition, there is no single measure that captures the scope of e-commerce within an economy. However, existing measures of aggregate components can still be used to uncover correlations related to e-commerce. These measures can be found in some datasets that approximate several dimensions of e-commerce, from consumer-related aspects to the enabling environment for e-commerce. These proxies will be used to calculate the correlation between e-commerce and illicit trade. These datasets come from:

• Eurostat,

• the World Bank, and

• the United Nations Conference on Trade and Development (UNCTAD)

Eurostat. For EU Member States, the 2019 Eurostat ICT Survey on Access and Usage can be used in analysis related to e-commerce, both domestic and cross-border. Two sections in the survey reveal the share of individuals who have purchased a good or service online during the prior three months, a finding that can be used as a proxy for e-commerce within an economy. This variable was combined with global customs seizures data from destination countries to draw a correlation. In addition, different specifications within the dataset enable a deeper comparison between the frequency and the value of global customs seizures as well as global customs seizures shipped by mail and express courier (EC).

World Bank. The World Bank Global Findex is a survey carried out every three years on how adults save, borrow, make payments and manage risk in 140 countries. In this data set, the variable “used the internet to buy something online in the past year” can be used a proxy of online purchase. The information provided by this variable is quite like Eurostat data but differs in two key areas. Firstly, the World Bank Index database in this report refers to 2017 data. Secondly, the World Bank database has a broader scope, as it is a worldwide survey rather than a regional one. Like in the Eurostat analysis, this proxy of e-commerce purchase was combined with customs seizures data by destination country to calculate correlations.

UNCTAD. The UNCTAD E-commerce Index measures the extent to which economies are prepared to support online shopping. The index relies on four non-weighted factors:

1. Account ownership at a financial institution or with a mobile-money-service provider (% of population ages 15+) from the World Bank Findex database

2. Individuals using the Internet (% of population) from the International Telecommunication Union (ITU)

3. Postal Reliability Index from the Universal Postal Union (UPU)

4. Secure Internet servers (per 1 million people) from the World Bank

To draw correlations between e-commerce and illicit trade, the 2019 UNCTAD E-commerce index was combined with the global customs seizures by destination country.

The regression analysis will be divided into two subsections. One section will look at correlations between e-commerce and overall illicit trade flows. The second section will examine e-commerce’s correlation with illicit trade transported through small parcels. Since e-commerce has significantly contributed to the rise of small parcel shipping, it can be expected that the e-commerce’s correlation with small parcel illicit trade will be more significant than its correlation with overall counterfeit trade.

E-commerce and trade in fakes at a first glance

This section of the chapter looks to find observable patterns between e-commerce and illicit trade. To do this, the datasets will be used to calculate the degree of positive correlation between online purchase indicators and global seizures data.

Overall there is a positive and statistically significant correlation between the indicators of e-commerce activity in an economy, and imports of counterfeits to that economy. A positive and statistically robust relation is observed, no matter what proxy of e-commerce is used.

The results of the quantitative exercises are presented in figures below.

Figure 3.1. Proxy of online purchases (in log) and number of customs seizures (in log), 2019

Notes: Only European countries included, and each point corresponds to one European country for 2019; e-commerce activity is represented by “share of individuals who have purchased a good or service online during the prior three months” as reported by Eurostat. The estimated value of the intercept is 2.30 (1.82) and of the coefficient of the proxy of online purchase (in log) is 1.186 (0.49). The multiple R²  = 0.373

Source: Eurostat ICT Survey on Access and Usage and OECD customs seizures database

Shown by the figure, a relationship between the online purchase proxy and the number of customs seizures is visible. These data point at clear correlation between e-commerce and imports of counterfeit goods to an economy.

Figure 3.2. Proxy of online purchases (in log) and value of customs seizures (in log), 2019

Note: Only European countries included, and each point corresponds to one European country for 2019; e-commerce activity is represented by “share of individuals who have purchased a good or service online during the prior three months” as reported by Eurostat. The estimated value of the intercept is 11.03 (2.18) and of the coefficient of the proxy of online purchase (in log) is 1.11 (0.59). The multiple R²  = 0.3

Source: Eurostat ICT Survey on Access and Usage and OECD customs seizures database.

Like in the previous data, this figure, which compares e-commerce and the value of counterfeit trade, presents a correlation between online purchase and European illicit trade. A robust, positive correlation appears to be distinguishable, re-confirming the positive correlation between the e-commerce activity in an economy and imports of counterfeits to that economy.

Figure 3.3. Proxy of online purchase (in log) and number of global customs seizures (in log), 2017

Note: Each point corresponds to one economy in 2017. E-commerce activity approximated by the World Bank Global Findex. The estimated value of the intercept is 1.75 (0.83) and of the coefficient of the proxy of online purchase (in log) is 1.05 (0.25). The multiple R²  = 0.48

Source: World Bank Global Findex data and OECD customs seizures database.

Shown in this figure, our proxy of online purchase is positively correlated to the illicit trade indicator. With that said, the correlation between the e-commerce proxy and the trade indicator is stronger than similar indicators from Eurostat, which suggests that e-commerce might be a greater determinant of counterfeit trade when looking at a larger, and more heterogeneous sample of economies, than the European Union only. The finding that countries that do not have developed e-commerce market also do not report large volumes of seizures of fakes, re-iterates the claim of e-commerce being an important solution effectively abused by counterfeiters.

Figure 3.4. Proxy of online purchase (in log) and value of global customs seizures (in log), 2017

Source: World Bank Global Findex data and OECD customs seizures database.

Like the previous figure, this figure, which compares e-commerce and illicit trade proxies, presents a correlation that seem to be statistically stronger than the one found using Eurostat information, and limited to European countries only. Although the correlation found here is not as distinguishable as the one in the previous figure, the relationship uncovered by the World Bank is stronger than the relationship established using Eurostat data.

Figure 3.5. E-commerce Index and number of global customs seizures (in log), 2019

Note: Each point corresponds to one country in 2019. E-commerce activity approximated by UNCTAD E-commerce Index. The estimated value of the intercept is 46.17 (3.83) and of the coefficient of the number of customs seizures (in log) is 4.79 (0.75). The multiple R²  = 0.56

Source: UNCTAD E-commerce Index and OECD customs seizures database.

As shown in this figure, the E-commerce index is evidently correlated to global customs seizures data. Supporting the hypothesis created using World Bank data, there appears to be a stronger correlation between e-commerce and illicit trade globally when compared to the European Union. It also suggests that when looking at a broader set of economies, the “effect of e-commerce” (i.e. strong correlation of seizures and e-commerce activity) is clearly visible. It points at e-commerce being an important way of getting to a customer used by counterfeiters.

Figure 3.6. E-commerce Index and value of global customs seizures (in log), 2019

Note: Each point corresponds to one country in 2019. E-commerce activity approximated by UNCTAD E-commerce Index. The estimated value of the intercept is 9.75 (1.06) and of the coefficient of the E-commerce index is 0.048 (0.015). The multiple R²  = 0.3

Source: UNCTAD E-commerce Index and OECD customs seizures database.

This figure presents a clear correlation between e-commerce and seizure data, although the relationship seems to be weaker. Even though the results seem to be weaker as for other exercises using UNCTAD data, the correlation is still comparable to previously explored e-commerce and illicit trade correlation data.

These six quantitative exercises were conducted for all counterfeit customs seizures in all means of transport. The results from this regression analysis show a positive correlation between various e-commerce proxies and illicit trade indicators. This result suggests that e-commerce does play a role in the global illicit trade landscape, but further analysis is needed to better understand the magnitude of the problem.

In addition, these results point that the problem of abuse of e-commerce by counterfeiters is more pronounced in developed economies, such as the EU. In those economies, consumers often use e-commerce enjoying the flexibility it offers. Unfortunately, criminals abuse these preferences and to a great degree misuse e-commerce as a way to access the market with counterfeits.

Focus on small parcels

The rest of this chapter will analyse trends relating to e-commerce indicators and illicit trade seizures found in small parcels. Since small parcels tend to be the preferred mode of transportation for e-commerce orders, additional checks to link small parcel illicit trade and e-commerce would be useful to develop a heightened understanding of e-commerce and illicit trade. The same three datasets will be used to calculate correlations between these data points, although this section will specifically focus on customs seizures by mail and express couriers (EC).

Figure 3.7. Proxy of online purchase in European countries (in log) and number of customs seizures by mail and EC (in log)

Note: Each point corresponds to one European country in 2019 The estimated value of the coefficient of the proxy of online purchases in the last 12 months (in log) is 1.8 (0.91). The multiple R²  = 0.34

Source: Eurostat ICT Survey on Access and Usage and OECD customs seizures database.

The observations in this figure differ from the previously analysed Eurostat data by changing the scope of the survey results from 3 months to 12 months and specifying that the counterfeit goods were transported by mail and express and courier services. In the same manner as previous Eurostat information, when the e-commerce proxy is compared to illicit trade indicators, a positive correlation is found between the two factors. This clearly supports the claim that most fakes ordered online are shipped via small parcels.

Figure 3.8. Proxy of online purchase in European countries (in log) and the value of customs seizures by mail and EC (in log)

Note: Each point corresponds to one European country in 2019 The estimated value of the coefficient of the proxy of online purchases in the last 12 months (in log) is 4.23 (0.7). The multiple R²  = 0.74

Source: Eurostat ICT Survey on Access and Usage and OECD customs seizures database.

This figure uses the same proxy of e-commerce as in the last example but combines that data with the value of regional customs seizures by mail and EC, rather than the quantity. When making this comparison, an evident trend emerges. This figure provides a clear correlation between e-commerce and illicit trade. And since this relationship is strong, a significant positive correlation between online purchase and the value of customs seizures by small parcels is justifiable.

Figure 3.9. Proxy of online purchase (in log) and number of global customs seizures by mail & express courier

Note: Each point corresponds to one country in 2017 The estimated value of the coefficient of the proxy of online purchase is 1.54 (0.33). The multiple R²  = 0.58

Source: World Bank Global Findex data and OECD customs seizures database.

The correlations presented in this figure point at a clear links between World Bank index of online purchase and global seizures. Continuing the trends discovered from previous observations, the World Bank correlation is stronger than the comparable Eurostat correlation. Put it differently, looking at a larger set of countries, including those, where e-commerce is used less intensely, one sees a much clearer correlation between the degree of seizures and e-commerce intensity. For the EU-only the correlation is less visible, as e-commerce is popular in the EU, and hence most European countries are more intensely targeted by counterfeiters that abuse the on-line environment.

Figure 3.10. Proxy of online purchase (in log) and number of global customs seizures by mail & express courier

Note: Each point corresponds to one country in 2017 The estimated value of the intercept is 7.01 (1.49) and of the coefficient of the proxy of online purchase (in log) is 1.54 (0.43). The multiple R²  = 0.46

Source: World Bank Global Findex data and OECD customs seizures database.

When using the World Bank proxy of e-commerce and the value of global customs seizure by mail and EC, the correlation peculiarly weakens. Since this identical comparison presented the strongest correlation using Eurostat data, it could be assumed that this comparison would yield the strongest relationship for World Bank data as well. In fact, breaking from the previously observed pattern, the World Bank figure showed a weaker correlation than the comparable one based on the Eurostat data.

Figure 3.11. E-commerce Index and number of global customs seizures (in log)

Note: Each point corresponds to one country in 2019 The estimated value of the coefficient of the E-commerce index is 0.08 (0.012). The multiple R²  = 0.59

Source: UNCTAD E-commerce Index and OECD customs seizures database.

The correlation depicted in this figure between the UNCTAD E-commerce index and the global illicit trade indicator is quite strong. Clearly shown by the figure, the data follows the constructed regression line and present a clear positive correlation between e-commerce and global illicit trade in small parcels. In fact, this comparison of the E-commerce index on the number of global customs seizures by mail and express courier (EC) is the strongest correlation observed using UNCTAD information.

Figure 3.12. E-commerce Index and value of global customs seizures (in log)

Note: Each point corresponds to one country in 2019 The estimated value of the coefficient of the E-commerce index is 0.07 (0.014). The multiple R²  = 0.46

Source: UNCTAD E-commerce Index and OECD customs seizures database.

The final check, based on UNCTAD data and shown in this figure, presents a moderate correlation between the e-commerce index and the value of global seizures by mail and EC. The result from this dataset was similar to the result from the World Bank dataset. This finding makes sense since both analyses capture world data, compared to Eurostat which uses regional data, for e-commerce proxies and compares it to the same set of customs seizures.

Comparing the regressions between overall customs seizures and customs seizures by mail and EC, an interesting trend emerges. Each correlation involving small parcels is either similar or more significant than the corresponding correlation reached in the overall seizure section. This finding suggests that e-commerce has noticeable impacts in increasing trade in counterfeits since small parcels are the predominant mode of transportation for e-commerce deliveries.

Moreover, this pattern is particularly strong for developed economies, such as the EU. As e-commerce is popular in the EU, most European countries are intensely targeted by counterfeiters that abuse the on-line environment, and use small parcels as the conveyance method.

References

[41] Amazon (2021), Brand Protection Report, https://assets.aboutamazon.com/96/a0/90f229d54c8cba5072b2c4e021f7/amz-brand-report.pdf.

[16] Andrenelli, A. (2019), “Electronic transmissions and international trade - shedding new light on the moratorium debate”, OECD Trade Policy Papers, Vol. 233, https://doi.org/10.1787/57b50a4b-en.

[32] DHS (2020), Combating Trafficking in Counterfeit and Pirated Goods, US Department of Homeland Security, Washington, https://www.dhs.gov/sites/default/files/publications/20_0124_plcy_counterfeit-pirated-goods-report_01.pdf.

[40] eBay (2021), 2020 Global Transparency Report, http://www.ebaymainstreet.com/sites/default/files/2021-05/2020-eBay-Global-Transparency-Report.pdf.

[30] EC (2017), Overview of the functioning of the Memorandum of Understanding on the sale of counterfeit goods via the internet, SWD(2017) 430 final, https://ec.europa.eu/docsroom/documents/26602.

[29] EC (2016), Memorandum of Understanding: 21 June 2016, Ref. Ares(2016)3934515 26/07/2016, European Commission, Brussels, https://ec.europa.eu/docsroom/documents/43321/attachments/2/translations/en/renditions/native.

[28] EC (2013), Report from the Commission to the European Parliament and the Council on the Functioning of the Memorandum of Understanding on the Sale of Counterfeit Goods via the Internet, COM(2013), 209 final,, European Commission, Brussels,, http://Ref. Ares(2016).

[27] EC (2020b), Report on the functioning of the Memorandum of Understanding on the sale of Counterfeit Goods on the internet, SWD(2020) 166 final/2, European Commission, Brussels, https://ec.europa.eu/docsroom/documents/42701.

[20] EUIPO (2021), Monitoring and analysing social media in relation to IPR infringement Report, https://euipo.europa.eu/tunnel-web/secure/webdav/guest/document_library/observatory/documents/reports/2021_Monitoring_and_analysing_social_media_in_relation_to_IPR_Infringement_Report/2021_Monitoring_and_analysing_social_media_in_relation_to_IPR_Infringem.

[19] EUIPO (2021), New and existing trends in using social media for IP infringement activities and good practices to address them, https://euipo.europa.eu/tunnel-web/secure/webdav/guest/document_library/observatory/documents/reports/2021_Social_Media/2021_Social_Media_Discussion_Paper_FullR_en.pdf.

[39] GCIG (2016), One Internet, Centre for International Governance Innovation and Chatham House, http://www.cigionline.org/documents/1045/gcig_final_report_-_with_cover.pdf.

[21] Goldberg, S. (2019), “Regulating Privacy Online: An Economic Evaluation of the GDPR”, SSRN Working Paper, http://dx.doi.org/10.2139/ssrn.3421731.

[31] Group, I. (2020), IP Crime and Enforcement Report, IP Crime Group Secretariat, Concept House, Newport, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/913644/ip-crime-report-2019-20.pdf.

[22] INTA (2020), Addressing the Sale of Counterfeits on the Internet, https://www.inta.org/wp-content/uploads/public-files/advocacy/committee-reports/Addressing_the_Sale_of_Counterfeits_on_the_Internet_June_2021_edit.pdf.

[9] OCDE/EUIPO (2021), Misuse of Containerized Maritime Shipping in the Global Trade of Counterfeits, Éditions OCDE, Paris,, https://doi.org/10.1787/e39d8939-en.

[35] OECD (2021), COVID-19 vaccine and the Threat of Illicit Trade, Chair’s Summary Note, https://www.oecd.org/gov/illicit-trade/summary-note-covid-19-vaccine-and-the-threat-of-illicit-trade.pdf.

[26] OECD (2021), Illicit trade: Discussions with stakeholders, unpublished.

[10] OECD (2020), Illicit Trade in a Time of Crisis. Chair’s Summary Note, https://www.oecd.org/gov/illicit-trade/oecd-webinar-illicit-trade-time-crisis-23-april.pdf.

[11] OECD (2020), OECD Digital Economy Outlook 2020, OECD Publishing, Paris, https://doi.org/10.1787/bb167041-en.

[34] OECD (2020), Trade in Fake Medicines at the Time of the Covid-19 Pandemics. Chair’s Summary Note, https://www.oecd.org/gov/illicit-trade/oecd-fake-medicines-webinar-june-10-summary-note.pdf.

[33] OECD (2018), Governance Frameworks to Counter Illicit Trade, OECD Publishing, Paris,, https://doi.org/10.1787/9789264291652-en.

[37] OECD (2011), OECD Council Recommendation on Principles for Internet Policy Making, OECD, Paris, https://www.oecd.org/sti/ieconomy/49258588.pdf.

[13] OECD (2008), The Economic Impact of Counterfeiting and Piracy, Éditions OCDE, Paris, https://doi.org/10.1787/9789264045521-en.

[38] OECD (2016a), Ministerial Declaration on the Digital Economy, OECD, http://www.oecd.org/digital/Digital-Economy-Ministerial-Declaration-2016.pdf.

[17] OECD (2016b), Recommendation of the Council on Consumer Protection in E-commerce, OECD, Paris, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0422.

[4] OECD/EUIPO (2021), Global Trade in Fakes: a Worrying Threat, OECD Publishing, https://www.oecd.org/publications/global-trade-in-fakes-74c81154-en.htm.

[8] OECD/EUIPO (2020), Trade in Counterfeit Pharmaceutical Products, Illicit Trade, OECD Publishing, Paris, https://dx.doi.org/10.1787/a7c7e054-en.

[3] OECD/EUIPO (2019), Trends in Trade in Counterfeit and Pirated Goods, OECD Publishing, Paris,, https://doi.org/10.1787/g2g9f533-en.

[7] OECD/EUIPO (2018), Misuse of Small Parcels for Trade in Counterfeit Goods: Facts and Trends, OECD Publishing, Paris, https://doi.org/10.1787/9789264307858-en.

[5] OECD/EUIPO (2018), Trade in Counterfeit Goods and Free Trade Zones: Evidence from Recent Trends, OECD Publishing, Paris/EUIPO, Alicante, https://doi.org/10.1787/9789264289550-en.

[6] OECD/EUIPO (2018), Why Do Countries Export Fakes?: The Role of Governance Frameworks, Enforcement and Socio-economic Factors, OECD Publishing, Paris/EUIPO, Alicante, https://doi.org/10.1787/9789264302464-en.

[2] OECD/EUIPO (2017), Mapping the Real Routes of Trade in Fake Goods, Illicit Trade, OECD Publishing, Paris, https://doi.org/10.1787/9789264278349-en.

[1] OECD/EUIPO (2016), Trade in Counterfeit and Pirated Goods: Mapping the Economic Impact, Illicit Trade, OECD Publishing, Paris, https://doi.org/10.1787/9789264252653-en.

[18] Phaneuf, A. (2021), Social Commerce 2021: Social media and Ecommerce Convergence Trends Brings Growth Opportunity for Brands, Insider Inc.,, http://www.businessinsider.com/social-commerce-brand-trends-marketing-strategies.

[42] S. Goldberg, G. (2019), “Regulating Privacy Online: An Economic Evaluation of the GDPR”, SSRN Working Paper, http://dx.doi.org/10.2139/ssrn.3421731.

[15] Sorescu, S. (2021), “Trade in the time of parcels”, OECD Trade Policy Papers, Vol. 249, https://doi.org/10.1787/0faac348-en.

[23] Tian, H. (2018), Bullet-proof payment processors, https://ieeexplore.ieee.org/document/8376208.

[14] UNCTAD (2021), Estimates of global e-commerce 2019 and preliminary assessment of COVID-19 impact on online retail 2020, United Nations Conference on Trade and Development, Geneva, https://unctad.org/system/files/official-document/tn_unctad_ict4d18_en.pdf.

[12] UNICRI (2020), “Cyber-crime during the COVID-19 Pandemic”, http://www.unicri.it/news/cyber-crime-during-covid-19-pandemic.

[25] UPU (2020), E-commerce Guide, Universal Postal Union, Berne, https://www.upu.int/UPU/media/upu/publications/Final-November-update-UPU-E-Commerce-Guide_2020_EN.pdf.

[24] USPS (2020/21), FY 2020 Annual Report to Congress, United States Postal Service, https://about.usps.com/what/financials/annual-reports/fy2020.pdf.

[36] WCO (2020), Illicit Trade Report 2019, WCO, Brussels,, http://www.wcoomd.org/-/media/wco/public/global/pdf/topics/enforcement-and-compliance/activities-and-programmes/illicit-trade-report/itr_2019_en.pdf?db=web.