Over the past decade, the number of countries that require (or at least expect) transfer pricing documentation has increased dramatically. However, the nature of the regulatory requirements and the benefits/cost of providing documentation varies substantially by country. Given the cost of preparing documentation, many MNEs prioritize different countries, designating some as ones where documentation is needed either because of local requirements or to mitigate risk, while deciding not to provide documentation for legal entities in other countries, either because the transfer pricing risk is low or because budget constraints will simply not allow it. In many cases, this is done is done in a relatively ad hoc way, and is largely driven by the question of whether there is a legal or regulatory requirement for such documentation.

The preparation of a risk scorecard can help the process of identifying transfer pricing risks and prioritizing transfer pricing needs in a structured way. As an initial observation, there is a high level of uncertainty involved in establishing reliable priorities – are the French tax authorities more or less aggressive than the Spanish, is this difference consistent and knowable or is it subject to substantial uncertainty? And can 50 or more countries be reliably rank-ordered based on these judgments? How much weight should be given to the aggressiveness of local tax authorities as compared to the magnitude of the transactions at issue or other factor such as local operating margins?

The behavioral economics literature teaches us that, when confronted with such complexity, people tend to focus on one or two issues while either disregarding or giving very little weight to other potential factors. This approach, however, often leads to suboptimal results as it tends to overstate the importance of these one or two factors while understating the importance of the other factors. A more reliable approach is to select five or six factors that are likely to be important, ideally selecting factors that are not closely correlated with each other – operating profits as a percent of sales and operating profits as a percent of total costs are essentially providing the same information; sales and operating margin are often providing different types of information with sales providing a quantitative measure of likely importance and operating margin providing a measure of a likely trigger to tax authority interest.

The table below shows one set of risk factors that could be used in building a risk scorecard. The factors are divided into three groups based on the general source of risk. The first set of risk factors are based on country attributes — Is documentation required? What is the likelihood of an audit? Will documentation improve audit outcomes by lowering the size or likelihood of an adjustment, lower penalties, etc.? Does the tax department have confidence that it knows how to manage an audit if one should develop?

The second set of risk factors are based on specific legal entity attributes — How important is the legal entity in terms of sales? Are low local profits or losses likely to increase the risk of a transfer pricing audit? Is the local entity engaged in transactions with a principal or intangible holding company in a low tax jurisdiction, and does the entity in the low tax jurisdiction have substance or not?

A third category of risk factors is related to transaction type – the likelihood and magnitude of transfer pricing risk will vary with the type of transaction. For example, the allocation of administrative and IT services charges may generate a high likelihood of audit in many countries, but the magnitude of the adjustments may be limited. Conversely, the likelihood of an adjustment for tangible property purchased and resold by a routine distributor may be low, but the size of the adjustment may be much larger than in the case of services. Intangible transactions often have raise risks that are both relatively likely and relatively large in size.

Two additional comments are important. First, the information used to construct a risk scorecard is often hard to get and imprecise – for example, it is hard to know whether documentation either lowers the cost of an audit or improves audit outcomes in a specific country even though this is probably one of the more important reasons that companies are willing to invest in documentation. But even in this case, including it in the scorecard puts some structure around how to think about managing transfer pricing risk, and identifies the type of information that is needed to do so effectively. Moreover, MNEs that have to prepare a Country-by-Country report are almost certainly already collecting the information needed to capture the sources of risk that arise from legal entity attributes.

Second, there is no single ideal set of factors to include in a risk scorecard — for example, if all transactions are with a tax haven principal with no substance, this attribute will have an equal impact on all legal entities and therefore there is no reason to include it in the scorecard. Moreover, the specific factors that are included in the scorecard may vary with the purpose of the scorecard – is it being used to determine which legal entities/transactions to document, or is it being used to evaluate risk and to determine if a financial statement reserve is needed?

A risk scorecard is also most effective when used as a dynamic interactive tools rather than as a static one time ranking. Therefore, risk scorecards are another example of where visual dashboards can be used to process large amounts of information – click here or on the graphic below to see a video showing an example of such a dashboard based on the data for FakeCo, a large MNE with 297 different legal entities and three discrete supply chain.

 

The video above talks about building a risk scorecard; the one below about using one.

 

If you want to be notified of new posts to this blog, please subscribe.