In this post, the Ghana-based economics researchers present their experiences over the period 2017-20. Dr. Ama Asantewah Ahene-Codjoe and PhD candidate Angela Azumah Alu worked collaboratively with project coordinator and lead researcher, Dr Rahul Mehrotra at the Graduate Institute, Geneva to analyze and estimate illicit financial flows via trade misinvoicing in the export of Ghana’s top commodities (gold and cocoa beans). The researchers adopted an innovative, interdisciplinary approach comprising qualitative, quantitative and political economy analyses. The qualitative and political economy analyses helped to identify inherent risks for IFFs in selected commodities. These subsequently aided our application of the statistical price filter method based in turn on legal transfer pricing guidelines for conducting arm’s length research. This blogpost presents a summary of the Ghana team’s research findings, alongside an overview of the main challenges faced and solutions adopted.
Mispricing of international trade in natural resources contributes to significant tax base erosion from commodity-dependent developing countries. However, existing evidence and empirical methodologies to estimate the magnitude of this phenomenon remain limited. In our research, we apply an interdisciplinary approach motivated by legal rules for trade valuation and statistical price-filter methods informed by expert interviews to estimate the magnitude of abnormal pricing in gold and cocoa exports from Ghana i.e. exports valued outside an arm’s length price range that indicates fair market values. Using transaction-level microdata from Ghana Customs, our results indicate that undervalued exports for gold equalled 11% of its total exports between 2011 and 2017. Similarly, we estimate that 1.0% of total cocoa beans and 7.2% of cocoa paste exported within the same period were undervalued. Overall, these findings indicate significant tax base erosion through trade mispricing from Ghana (See Ahene-Codjoe, Alu and Mehrotra, 2020).
Delays in data acquisition was our first major challenge. The use of transaction level data from the Customs Division of the Ghana Revenue Authority (GRA) was one of the key aspects of the quantitative data needed for generating the abnormal pricing estimates. The data acquisition process began with a letter to the office of the Commissioner-General of GRA, followed by several visits to the said office to trace its progress to the unit with the data, that is, the Policy and Programmes Unit of the Customs Division. The delay slowed down the WP1 research. However, the team focused on other aspects of the research, including the political economy as well as building a network with the major players in the industry, whilst waiting for the data to be released.
The team also faced challenges in securing interviews with officials of stakeholder institutions for the qualitative data needed for our estimations and/or validating our findings. Sometimes meetings had to be rescheduled due to the absence of expert officials. Nevertheless, constant communication with these officials was upheld by the team. Again, there was a challenge with the number of institutions that had to be visited and the distances between them. Managing to schedule visits on the same day curtailed incessant travels, thereby saving time.
Another difficulty the team encountered was in the method of estimating abnormal pricing in the cocoa beans sector. Initially, the team utilised spot prices from the Ghanaian Cocoa Ex Dock NY Prices as the market reference price to generate the estimates. However, the feedback received after presenting the research to personnel of Ghana Cocoa Board (COCOBOD) was that the requisite market reference price is the prices listed on the London International Futures and Forwards Exchange (LIFFE). This is due to sales of Ghana’s cocoa beans using forward contracts. The team had to visit the Trading Room at the Cocoa Marketing Company (CMC) to better understand the trading system used. The estimates were redone based on the new information acquired and current estimates to a large extent reflect the situation pertaining to the export of cocoa beans from Ghana.
Furthermore, there was an issue raised about the purity of gold as used for the estimates of abnormal pricing in the export of gold from Ghana by the Ghana Chamber of Mines. To address this, an external database that provided a breakdown of the purity levels of gold from various mines in Ghana was used to supplement the data from Customs. This breakdown aided the team to better set the boundaries for the market price filters.
Insights and Conclusion
Lessons learnt from the project include the need for collaboration in research for development. This to ensure that the insights gathered by researchers as well as the inferences made accurately reflect the situations that pertain in the country.
Also, worthy of mention is the co-operation with the team in Switzerland that provided very important inputs for the Economics team in Ghana. Access granted to some databases such as the Thomson Reuters Datastream is much appreciated. Similarly, guidance in the intricate computer programming languages used for our analysis and regular skype meetings were very helpful. They particularly provided quick feedback and an opportunity to discuss research progress and brainstorm potential solutions to obstacles faced. These indeed led to a smoother work delivery within the proposed period of the work plan.
The team is very appreciative of the time and contributions from the overall leader of the research project, Prof. Gilles Carbonnier and Dr. Rahul Mehrotra (the project coordinator) – both of the Graduate Institute of Geneva; the leader of the Ghana team, Dr. Fred Dzanku of the Institute of Statistical Social and Economic Research (ISSER); Dr. Steve Manteaw of Integrated, Social Development Centre (ISODEC); Mr. Christopher Opoku Nyarko of the Ghana Chamber of Mines; Mr. Amponsah Tawiah, formerly of the Minerals Commission of Ghana; the late Dr. K. Opare- Hammond, former Managing Director of Precious Minerals Marketing Company (PMMC); Mr. Kofi Nti, former Commissioner-General of GRA; and Mr. Richard P. Yawutse, former Deputy Commissioner in charge of the Policy and Programmes Unit of the Customs Division of the GRA. Our sincerest thanks to Mr. Mate-Kodjo and Mr. Shaibu Musah both of the Customs Division of the GRA.
Finally, we are grateful to the officials of the various stakeholder institutions such as COCOBOD in general and Cocoa Marketing Company (CMC) in particular, the Ghana Extractive Industries Transparency Initiative (GHEITI), the Ghana Chamber of Mines, the Minerals Commission, the Bank of Ghana, Precious Minerals Marketing Company (PMMC), the Financial Intelligence Centre (FIC), Ghana Export Promotion Authority (GEPA), Ghana Investment Promotion Centre (GIPC) and the Policy and Programme Unit of the Customs Division of the GRA. Their contributions were very relevant for understanding the commodities, the risks for IFFs and the setting of the benchmark as well as the arm’s length price filters for the market reference price filters. Overall, the WP1 team in Ghana is very appreciative of the support received from all stakeholders.