Mirror analysis, a risk analysis support tool for Customs administrations

By Roger-Claver Victorien Gnogoue, Financial Services Director, Côte d’Ivoire Customs

Twenty-first century challenges place demands upon Customs administrations throughout the world, particularly in developing countries. Customs services must, therefore, be effective and efficient if they are to meet their revenue and trade facilitation targets, the achievement of which requires substantial investment in reforms, modernization, and continued improvement.

Côte d’Ivoire Customs is aware of these challenges, and has introduced a range of reforms to ensure its thorough modernization. It is now an exemplary, up-to-date and high-performing service. The modernization programme driven by its development partners – the World Bank (WB) and the International Monetary Fund (IMF) – is founded on six projects.

Its enforcement and anti-smuggling initiative forms part of the reform process under Project 4. To reduce supply chain risks in matters of security and non-compliance with prevailing laws and regulations, Côte d’Ivoire Customs has put a series of modern working methods and techniques in place to improve and develop its risk management, and inspection, investigation and enforcement capabilities. It has, in particular, established a control and selectivity system based on risk analysis, and has introduced a simplified report procedure which computerizes declaration control reports, thus creating a database of Customs offences.

Risk analysis

Customs action has had to be rationalized in response to increased volumes of trade in recent years. Pre-clearance audits, which verify the nature and value of goods before importers or their representatives take possession of them, have had to be appropriately combined with post clearance audits (PCAs), which are carried out after goods are released or cleared at the border, the aim being to control less but better.

Pre-clearance audit, carried out by frontline services, involves a selection of operations with a potentially high risk of fraud. Côte d’Ivoire Customs, with technical assistance from the IMF Regional Technical Assistance Centre in West Africa (AFRITAC), has introduced a risk analysis-based control and selectivity system to ensure this process. The system’s performance over the previous 12 months was assessed in late 2015: 20% of transactions covering around 70% of recognized fraud findings were audited. Historically, less than 10% of declarations audited led to a finding of fraud.

PCAs, carried out by second-line services, are designed to ensure that completed Customs operations have been properly conducted. This involves an in-depth analysis, both of declarations and of the attached import documents, and may cover the three financial years preceding the audit. While sectors of activity and operators to be audited may be selected on the basis of a risk analysis, new methods are increasingly recommended by the WCO, including mirror analysis.

By definition, mirror analysis is a decision support tool designed to study differences in a country’s foreign trade. To that end, it analyses both import and export statistics. This technique is used to develop the capacity of Customs administrations to identify possible irregularities, such as under-valuations, incorrect classifications (tariff slippage), false declarations of origin, etc. In other words, the magnitude of the differences may point Customs administrations to fraudulent flows most commonly used over a period of time.

Some results of mirror analyses and their interpretation

Mirror analysis has been used in Côte d’Ivoire to study 2014 data, the most recent complete data available in the United Nations Commodity Trade Statistics Database (UN COMTRADE). This exercise highlighted significant disparities between data taken from our SYDAM WORLD computerized Customs clearance system and data stored in the UN database. These differences involve several products, and are more substantial for Asian countries. Our attention was drawn to two particular cases. Because of their sensitive nature, information regarding Harmonized System (HS) headings or the source countries of the products involved will not be disclosed.

The first case concerns a product that will be identified as tariff subheading “X1,” which is mainly imported from an Asian country. The 2014 import value declared for this subheading in the SYDAM is 127.24 billion CFA francs, compared to a mirror value of 211.3 billion CFA francs, i.e. a negative difference of 84.10 billion CFA francs. In addition, the prices declared at the Customs barrier are well below international prices (209 CFA francs/kg). Two particular operators declare prices, which are significantly atypical compared to average imports of the same product.

The second concerns product “X2”. While one country may have declared exports of this product to Côte d’Ivoire, the 2014 national Customs statistics show a weak flow of such goods that year. This product attracted a tariff of 20%. By contrast, a product in another subheading taxed at a lower rate, i.e. 10%, which will be referred to as “Y2,” accounted for a considerably greater number of declarations over the same period. In this particular case, this could represent a tariff slippage, leading to the avoidance of a proportion of the Customs duties.

The challenge is all the more significant because the introduction in 2015 of the common external tariff of the Economic Community of West African States (ECOWAS) brought about a 35% increase in the Customs duties on product X2, which could lead to greater revenue losses if corrective measures are not put in place.

Using the results of a mirror analysis

The findings of any mirror analysis lead to assumptions of fraud which must then be verified by investigations in the field or, failing that, in-depth document reviews. They point Customs administrations towards possible fraud, and make it possible to avoid sometimes cumbersome and costly investigations that may prove inconclusive and fruitless.

Côte d’Ivoire Customs issued 2,420 fraud reports in 2015. The offences most commonly recorded by frontline services are false values, false goods and false weight declarations, which alone account for 83.68% of total offences. The principal offences identified by second-line services are false value declarations (29.83%), undeclared imports (17.23%), and false goods declarations (7.32%).

The results of controls are recorded in the simplified report, allowing operations to be audited and control officers to be assessed. To ensure that Customs controls are increasingly effective, the statistical services that conduct mirror analyses must also cultivate close relations with Customs control services, and targeted products and origins must be regularly monitored during Customs clearance operations or PCAs.

In addition, the results of mirror analyses allow Customs statistics to be corrected. Statistical asymmetries are a very serious issue since they affect the reliability of the trade balance, the balance of payments, and national accounts. Data reconciliation is necessary to improve the quality of the foreign trade statistics that Customs administrations are responsible for compiling in many countries.

Data quality and mirror analysis in perspective

The pursuit of data quality is essential for a mirror analysis. Customs administrations must ensure this quality by collecting data from various sources, particularly foreign Customs administrations, port authorities, exporting countries, and businesses.

Many data accuracy problems have been highlighted repeatedly during mirror analyses. These may be goods classification errors by the country of origin, time lags between export and import declarations, the “misallocation” of imports to a third country, or exchange rate volatility. These problems can be resolved largely through Customs cooperation.

The WCO has drafted a model agreement on bilateral administrative assistance in traditional areas such as valuation, classification, and origin of goods, thus facilitating Customs cooperation which is essential to any mirror analysis. In response to the differences identified by these analyses, a Customs administration must organize meetings with the administrations of the countries concerned in order to reconcile its data. This will allow goods to be monitored and traced from their place of production to the country of final destination.

It should be noted here that, under the Trade Support and Regional Integration Programme (PACIR), the European Union (EU) undertook to foster regional integration in the West African Economic and Monetary Union (WAEMU) by interconnecting Customs computer systems. This project includes a data exchange protocol enabling goods transit operations to be managed more effectively.

It is also essential to ensure close cooperation between Customs administrations and businesses, in particular to initiate a dialogue on data reliability. The implementation of an authorized economic operator (AEO) programme may be beneficial in this respect. In Côte d’Ivoire, a public-private partnership between Côte d’Ivoire Customs and the private sector was developed through the Observatoire pour la Célérité des Opérations de Dédouanement (OCOD)*, a body set up to streamline Customs clearance processes.

This cooperation, which has already produced useful results (particularly two studies of Customs release times, trade facilitation measures and appeals against Customs decisions), could represent a starting point for greater dialogue on data quality with certain trusted traders. In addition, the WCO’s decision to dedicate 2017 to data analysis under the slogan “Data Analysis for Effective Border Management” is another opportunity to promote dialogue on the quality and use of Customs data.

*Observatory for the Speed of Customs Clearance Operations


More information