By Josh Bamfo
Since Nigeria Transfer Pricing Regulations (the Regulations) came into force in September 2012, with 2, August 2012 commencement date, transfer pricing has become one of the hottest tax issues in Nigeria. Taxpayers have complied with the Regulations by submitting transfer pricing returns on an annual basis, prepared contemporaneous transfer pricing documentation, and the Federal Inland Revenue Services (FIRS) has embarked on its first round of transfer pricing audits. However, one of the challenging aspects of the implementation of the Regulations is the lack of local and regional (African) external comparable data to assist with the application of some of the Regulation’s recommended transfer pricing methods.
Transfer pricing basically refers to the pricing of related party transactions involving tangible goods, intangible goods, services and financial services. Revenue authorities are concerned about transfer pricing because, without transfer pricing regulations, multinational enterprises (MNEs) will have an incentive to misprice their related party transactions to take advantage of corporate tax rate arbitrage across countries, resulting in profit shifting from entities in high tax jurisdictions to related entities in low tax jurisdictions. This will result in loss of tax revenue for countries with relatively higher corporate tax rates. Hence, the need for local transfer pricing regulations to ensure that related party transactions are reasonably priced to ensure fair allocation of system profits between related entities, and therefore, between the respective tax jurisdictions, from a tax revenue perspective.
To this end, most countries across the globe have adopted the arm’s length principle in line with the Organisation for Economic Cooperation and Development’s (OECD’s) Transfer Pricing Guidelines (OECD Transfer Pricing Guidelines) to ensure that related party transactions are reasonably and fairly priced. The arm’s length principle basically refers to the requirement that related party transactions should be priced as if the transactions were conducted with independent parties. This is based on the basic economic principle that, all other factors remaining the same, a rational investor with a profit maximization objective will not under-price its transactions with an unrelated party; hence, market prices are appropriate benchmarks to ensure that taxpayers do not intentionally misprice their related party transactions to take advantage of corporate tax arbitrage.
A critical aspect of the application of the arm’s length principle is the challenge of finding prices or returns from comparable third party transactions or companies to benchmark prices or returns earned from the related party transactions being analysed. To ensure that transfer pricing analyses yield reliable results, the OECD Transfer Pricing Guidelines provides five comparability factors that analysts are supposed to consider when performing transfer pricing analyses. These comparability factors include: product and service characteristics, functional analysis, contractual terms, economic circumstances and business strategy. Thus, a benchmarking analysis should consider these comparability factors in ensuring that the analysis yields reliable benchmark results.
The data constraint issue
Although the OECD Transfer Pricing Guidelines require an analyst to consider each of the comparability factors when performing a transfer pricing analysis, some of the comparability factors will be more important than others depending on the method selected as the most appropriate to apply. The method selected for testing the covered transaction is identified based on facts and circumstances relating to that transaction. Where analysts select the use of external comparable companies’ data to apply methods such as the Resale Price Method (RPM), the Cost Plus Method (CPM) or the Transactional Net Margin Method (TNMM), one of the relevant comparability factors that needs to be considered is comparability of economic circumstances. This comparability factor considers relevant issues such as the need for the selected comparable companies to come from comparable economies with similar market risk and similar expected market returns as the company whose returns are being tested. As a result, when benchmarking appropriate returns of a tested party from countries such as US, UK, Germany, Italy, China, and Poland, to mention a few, they ideally prefer or require taxpayers to use comparable companies from the same country as the tested party. However, other jurisdictions such as African countries, including Nigeria, have significant constraints with respect to publicly available comparable data.
The comparable data constraint in Africa is mainly due to the absence of legislation that requires private companies to publish their financials, difficulty in assessing ownership data of companies, and the relatively limited number of publicly listed companies in the local stock exchanges, such as the Nigeria Stock Exchange. While financial information is useful for testing pricing arrangements and returns, ownership data is useful for ensuring that selected comparable companies used in transfer pricing analyses are truly independent of influence from associated/related entities.
To help overcome the comparable data constraint in Africa, in the past, comparable data from Pan-European countries have been used to perform such benchmarking analysis irrespective of the significantly different economies and the related market risk differential. This problem has come to the forefront of transfer pricing implementation in Africa as more and more African economies enact their local transfer pricing rules, and seek for more reliable transfer pricing analyses. One of the arguments used to buttress this concern is that a rational investor investing in an emerging economy such as Nigeria will have a higher expected rate of return on such an investment to cover the relatively high market risk in Nigeria compared to a similar investment in a Pan European country. As a result, using Pan-European comparables to benchmark returns for the local Nigerian entity could reduce the reliability of the benchmarking analysis.
Clearly, there are significant differences between the Pan-European economies, where the market risks are relatively low and industries are typically more matured with stable but lower growth rates and the emerging economies with relatively high market risks and industries that are typically in their early stages of development with relatively high but volatile growth rates. The telecommunication and banking industries are good examples of such significant disparity in average returns. On the average, Nigerian banks recorded growth in net profit margin of 13% for FY 2014. First Bank of Nigeria Plc. recorded an accelerated growth rate of 16% from 2013. Another big Nigerian bank, Guaranty Trust Bank Plc (GTB) recorded profit growth of 10% to end 2014 FY. On the flipside, the interim results of banks in the UK as reported by KPMG London in its publication on “UK Banks: Performance Benchmarking Report” for 2014 shows that the UK’s five biggest banks – Standard Chartered Bank Group, RBS, Barclays, HSBC and Lloyds Bank cumulatively recorded margins of approximately 8% lower than the corresponding period in 2013.
Thus, benchmarking appropriate returns that a Nigerian bank should have earned in a given year with comparables from UK would yield unreliable results. A similar trend can be observed in the telecommunications industry.
To help manage this obvious comparable data constraint issue in Nigeria and other African countries, we consider both short term and long term remedies. The objective is to find practically reasonable approaches that achieve the following:
Consider the economic circumstances comparability factor;
Do not impose significant analytical burden on taxpayers and tax administrators due to its complexity; and
Are consistent with similar best practices in other regions.
Short term remedies
Considering that it will take a while for the appropriate local and African comparable data to be compiled and made publicly available, the short term remedies considered in this article focus on how best we can make use of currently available data to address the comparable economies factor.
Country risk adjustment
In practice, it is common to find cases where extensive search for local comparables yield insufficient or no comparable data. As a result, one common approach employed by some analysts is to perform country risk adjustments to the returns of the available Pan European companies that have been identified and considered similar to the local (Nigerian or African) entity being tested. Where such adjustments are sufficiently carried out, it can be argued that variability in returns of the selected Pan European companies and the local entity that is due to the difference in economic circumstances have been adequately addressed in the analysis, therefore making the results more reliable – comparing apples to apples. In applying the said adjustment, several approaches could be adopted depending on the facts and circumstances of the case under review. For example, in benchmarking foreign currency denominated long term loan interest rates charged to a local entity, some analysts have relied on government bond spreads differential between the Pan European country and the country of the local entity (the borrower). The spread is then used as a proxy for the country risk premium that should be added to the interest rates charged on comparable loans made to Pan-European companies in establishing reliable benchmark interest rates that should be charged the local entity.
This approach has a number of limitations. For one, interest rates on government bonds capture different risk factors such as the time value of money in the form of an inflation premium, the opportunity cost of capital in the form of the interest rate risk premium, and a country risk premium, assuming the borrowings are in the same currency. Thus, there is a likelihood that the upward adjustment of the interest rates or returns of the Pan European comparable loans or companies may be overstated. Secondly, further adjustments to improve the reliability of the results render this approach complex and burdensome to both taxpayers and tax administrators. Thirdly, generally, where an approach requires that many adjustments, it is an indication that it might not be the most appropriate approach. Fourthly, the country risk adjustment approach is inconsistent with best practices in other regions where they face similar local or regional data constraints. For example, for tangible goods transactions and services, Latin American countries tend to expand their search for regional comparable companies to include companies from US and Canada without performing any country risk adjustments. Fifthly, the approach is not flexible and fungible enough to allow MNEs have a consistent benchmark policy for the Africa region for similar functions performed by affiliates across the region, considering that each country will most likely yield a different country risk adjustment. Finally, although the FIRS has not provided explicit guidance on how to address the local comparable data constraint, the complexity of this approach would most likely be less preferable to an alternative that is less complex but yields equally or more reliable results. As a result, we believe that the country risk adjustment should only be applied where there does not exist less complex approaches that can yield equally reliable results.
Comparables from similar emerging economies
This approach considers searching for comparable companies from similar emerging economies, whilst taking into account, the importance of geography in considering similar markets. Specifically, for tangible goods and services transactions, the approach starts with a search for functionally comparable companies in similar industries from similar emerging economies in Africa, Middle East and Eastern Europe. Where the search yields limited to no accepted comparable companies, the search is expanded to include similar emerging economies in the Asia-Pacific region. The approach clearly takes into account similarity in market risk as well as the general global practice of expanding searches to countries within similar geographical region. An additional merit of this approach is that it does not involve additional burden to both taxpayers and tax administrators, since similarity in market risk is captured by selecting companies from similar emerging economies as opposed to performing complex country risk adjustments for each accepted company. Further, this approach is flexible and fungible enough to allow MNEs to determine an appropriate and consistent benchmark for their affiliates performing similar functions, bearing similar risks and utilizing similar assets across the African region. As a result, we recommend the comparables from similar emerging economies approach as the most appropriate short term remedy for the local comparable data constraint problem in Nigeria and the rest of Africa.
Some analysts, however, point to the fact that choosing comparable companies from broadly comparable emerging economies considers market risks and other market characteristics, as such, an approach that explicitly focuses only on market risk will yield more reliable results. As a result, their proposed approach uses publicly available country risk ranking data to determine comparable countries to use for the analysis. For example, some analysts use the Euromoney numerical country risk rating to determine the rating of the country of the tested party and that of other countries, and determine a range of comparable market risk countries by choosing plus and minus 10% of the rating of the tested party’s country. As a result, a country like South Africa could have a Pan European country such as Portugal or Ireland as a comparable economy from a market risk perspective. Although this approach explicitly uses a proxy for market risk to screen comparable economies, it is highly limited in its application for tested parties in other African countries such as Nigeria and Ghana. African countries such as Nigeria, Ghana, Kenya and Uganda with transfer pricing regulations but relatively low country risk ratings have most of their comparable economies, based on the plus and minus 10% of the Euromoney country risk rating, being countries with limited to no comparable data. Therefore analysts often end up with the same data constraint problems they were trying to address in the first place. As a result, we believe that the applicability of the approach based on country risk ratings is limited to a few African countries with relatively favourable country risk rating such as South Africa, but not applicable to most of the other African countries such as Nigeria, Ghana, Kenya and Uganda with relatively poor country risk rating.
Long term remedy
Developing a reliable local comparable data
As a first step, the FIRS and other African revenue authorities should take the lead in advocating for legislation that enables private data publishing companies to publish the financials of publicly traded and private companies. This will enable reputable private data publishing companies such as Bureau van Djik to compile and publish the relevant comparable data and assess ownership data of the companies. Secondly, the governments of the affected African countries are advised to invest in, develop and enrich their respective trading floors. The belief here is that as local stock markets develop and have more listings, the quality of the reported financial information of listed companies will improve. Finally, international agencies such as the United Nations and the African Tax Administration Forum (ATAF) have demonstrated their willingness to provide assistance to the development of transfer pricing in Africa. Their assistance in this area will help overcome the comparable data constraint in the Africa region. For example, ATAF, through the formation of the ATAF Working Group on Transfer Pricing has held several symposiums, organized to address the issue of inadequate comparables in the region. Realisation of the solutions we hope to receive from ATAF is likely to foster implementation of transfer pricing in Africa.
The data constraint problem facing the application of the transfer pricing regulations in Nigeria and other African countries such as Ghana, Kenya, and South Africa, to mention a few, is real and should be taken seriously. Although a long term and sustainable solution in the form of developing local African comparable data is needed, a short term remedy in the form of the use of comparable companies from similar emerging economies within broadly similar geographical region is a viable option. The latter appears to be emerging as the generally accepted best practice in the short term for most African revenue authorities, including the FIRS, as they push back against the use of Pan-European comparable companies to benchmark appropriate returns for local African companies.
Bamfo is Associate Director and Nelson Osahon Idemudia, Senior Tax Adviser, KPMG Advisory Services, Lagos Nigeria