Global Financial Integrity


FAQs about GFI’s Value Gap Analysis Methodology

GFI frequently receives questions about certain aspects of our value gap methodology for estimating the scale of trade misinvoicing. Here are the most frequently asked questions and our answers:

Q: What is trade misinvoicing and why is it a problem?

A: Trade misinvoicing is a method of moving money illicitly across borders, which involves the deliberate falsification of the value, volume or quality of an international commercial transaction of goods or services by at least one party to the transaction. This typically occurs when exporters or importers submit false shipment information on customs invoices when shipping exports or receiving imports. There are four standard types of trade misinvoicing: under-invoicing and over-invoicing of both exports and imports. Two of these types of trade misinvoicing result in illicit outflows of money out of an economy (import over-invoicing and export under-invoicing) and two result in illicit inflows of money into an economy (import under-invoicing and export over-invoicing).

Trade misinvoicing is a problem because it results in a loss of desperately need public tax revenues in developing countries. The most common reasons for illicit outflows from developing countries are to evade corporate income taxes and shift wealth from weak currencies into hard currencies, while common reasons for illicit inflows into developing countries are for evading customs duties and value-added taxes (VAT), and laundering the proceeds from and/or financing illegal activities of transnational criminal organizations. While much attention is often given to the problem of illicit outflows from developing countries, the problem of illicit inflows is often just as corrosive as illicit outflows. In both cases, the result is that taxes are not being paid to governments, resulting in less revenue available for public health, public education and other essential government services.

Q: How large of a problem is trade misinvoicing?

A: Every year, trade misinvoicing creates a value gap of hundreds of billions of dollars in emerging market and developing countries which, due to massive losses of related duties and VAT, has a corrosive impact on their economies and their ability to reach the 2030 Sustainable Development Goals. GFI believes that developing country governments could substantially augment their domestic resource mobilization efforts by addressing the problem of trade misinvoicing.

Q: How does GFI identify and estimate trade misinvoicing?

A: In order to identify a country’s imports/exports that might have been misinvoiced, GFI conducts a “value gap” analysis. In such an analysis, GFI examines a country’s imports and exports with its partners in order to identify if one of four major types of trade misinvoicing is present: 1) import under-invoicing; 2) import over-invoicing; 3) export under-invoicing and 4) export over-invoicing. Most countries submit annual trade reports to the United Nations each year and GFI uses this publicly available data to cross-reference one country’s report with those submitted by its trading partners in order to undertake a comparative analysis. The United Nations Comtrade database, which collects an average of 15 million trade transactions per year, is our main source of data. With our customized program we can analyze millions of records to identify mismatches between what a country and its trading partners reports. As an example, if Egypt reported paying $5 million for alarm clocks imported from China in 2016, but China only reported exporting $3 million in alarm clocks to Egypt in 2016, this would represent a value gap of $2 million. With Egypt as our focus country, this would reflect a case of import over-invoicing by Egypt. After identifying all of the mismatches between what is reported by a country and its trade partners in a given year, we are able to identify the total “value gap” for the country. See GFI’s Methodology for a detailed description of estimating the scale of trade misinvoicing.

Q: Why did GFI switch from using the DoTS data set to using UN Comtrade?

A: In 2018, GFI switched from using the International Monetary Fund’s Direction of Trade Statistics (DoTS) database, because DoTS collects national-level data, while the United Nations Comtrade database, which provides much more granular, product-specific information on bilateral trade in goods. While there are strengths and shortcomings to both databases, Comtrade enables GFI to undertake much more detailed analyses of country trade data for use in our estimates.

Q: Can the value gaps identified in UN Comtrade data really be trusted?

A: Yes. Despite limitations in the data (see “Limitations of GFI’s Value Gap Analysis Methodology,” in GFI’s Methodology for a detailed description of the shortcomings of the official international trade data and how GFI addresses these concerns), today we have more data for countries available than in any previous period in history, and many national and international agencies are improving data collection each year. But because there are limitations in the international trade data, GFI underscores that we believe our final estimates are conservative, and that the scale of the problem of IFFs is likely to be much larger than can be shown by our analyses of trade misinvoicing. Our primary goal in this case is not exactitude, but rather to show the order of magnitude of the problem. Therefore, even if the limitations in the available data cause our estimates to be off by a certain degree, we are confident that they still illustrate the broad orders of magnitude regarding the deleterious problem of trade misinvoicing for countries around the world.

Q: When trying to identify an estimate of the global level of trade misinvoicing, wouldn’t one country’s import over-invoicing be reflected as another trading partner’s export under-invoicing? And if both are included in a global-level analysis, then isn’t there a risk of double-counting the actual level of IFFs?

A: GFI’s approach to estimating trade misinvoicing avoids double-counting. Earlier approaches to calculating the levels of global trade misinvoicing using the “country-world comparison” could exhibit the problem of double-counting, i.e., one country’s import over-invoicing is reflected in another trading partner’s export under-invoicing, and if both are included in an analysis then there is a risk of double-counting. For this reason, GFI does not attempt to arrive at “country-world comparisons.” Instead, GFI focuses exclusively on analyzing trade between country pairs in a series of country-by-country analyses. Our annual IFF Update reports analyze trade transactions for 148 developing countries with a set of 36 advanced economies (as designated by the IMF). Our series of country-specific reports analyze a developing country’s trade with all of its trade partners. In both approaches, all four types of trade misinvoicing are calculated from the perspective of each developing country.

Q: Can GFI’s methodology detect misinvoicing of international trade in services?

A: The United Nations Comtrade database and most other types of available trade data covers only merchandise goods for most developing countries. Very little trade in services data in developing countries are reported to UN Comtrade. Therefore, even as trade in services as a percent of total world trade has been increasing, a range of misinvoicing in trade in services cannot be detected in our – or any – value gap analysis. Such trade misinvoicing in services could include falsified invoices for management fees, interest payments, licenses, consulting contracts, etc., which have become commonly used avenues for overcharges as a way to shift money out of emerging market and developing countries. Arguably, if trade in services data could be available and incorporated into our value gap analyses, our estimates of the value of trade misinvoicing would be much higher.

Q: Are all cases of trade misinvoicing that can be detected in the international trade data always due to people trying to move money illicitly, or are there other reasons why mismatches in the trade data between trading partners might show up?

A: Not all identified mismatches in the international trade data are due to purposeful trade fraud. There are various reasons that mismatches may occur in bilateral trade data between partners, such as time lags in reporting by individual countries which record transaction in different years, errors in calculating currency conversions, exchange rate fluctuations, inconsistent use of assigning Harmonized System (HS) product codes to goods, and simple human error at the statistical bureaus. GFI applies statistical treatments into our analyses to reduce the likelihood these occurrences will impact our estimate including by harmonizing trade margins, i.e., the difference between Freight on Board (FOB) prices (used by exporters) and Cost, Insurance, Freight (CIF) prices (used by importers); utilizing applied weighting to the data; adjusting for Hong Kong re-exports and the Swiss gold trade, etc. See GFI’s Methodology section for a detailed description of these discrepancies in the international trade data and the statistical treatments GFI applies in order to address them. Despite factors which might cause a false positive for trade misinvoicing, GFI believes the majority of the identified mismatches in the international trade data among two partners are due to purposeful trade fraud. As a bit of anecdotal evidence regarding the frequency of trade misinvoicing, GFI has been told by the Commissioner’s General of Customs in four countries that an estimated 80 per cent of the import invoices they receive are fraudulent.

Q: Can GFI’s methodology detect “same misinvoice faking” (when both the importer and exporter collude in advance of a trade to agree on falsifying invoices with the same fraudulent price of the goods being shipped)?

A: No, GFI’s methodology cannot detect “same invoice faking”. The anomalies between the actual price and a fraudulent price cannot be detected when both parties collude on trade fraud through misinvoicing. GFI has long acknowledged this problem with the international trade data, which is another reason why we caution that our estimates of trade misinvoicing are very likely to be conservative, and the actual scale of trade misinvoicing is likely to be much larger. In order to better address the problem of “same invoice faking” – and to crack down on trade misinvoicing generally – GFI developed its GFTrade tool for use by customs authorities to check the stated prices on the invoices submitted by importers and exporters to see if they are within the range of average prices for the same goods traded among 43 of the largest trading countries over the previous 12 months. Any invoices with prices that are orders of magnitude outside of this average range can be flagged for further investigation by authorities. Therefore, while “same invoice faking” may not be detected using the standard Partner-Country method with United Nations Comtrade data, it can be detected using our GFTrade tool. Contact Tom Cardamone at for more information on GFTrade.

Q: Couldn’t the limitations to GFI’s methodology create the possibility that GFI’s policy recommendations for developing countries may be unsupported?

A: No. GFI believes that even if its estimates of the scale of trade misinvoicing are off the mark due to limitations associated with its methodology, the orders of magnitude of the problem are still so great that the policy reforms and regulatory steps advocated by GFI are still warranted. The consistent and key message emanating from GFI’s reports and studies is that the magnitude of trade misinvoicing remains persistently large and is a chronic and ubiquitous problem. Therefore, the suggested policy reforms and regulatory steps are deeply warranted.

For further details, see GFI’s Methodology for estimating trade misinvoicing.