Principal vs. agent – we need more!

As we covered here, the IASB recently called for stakeholder feedback to inform its review of the accounting standard for revenue from contracts with customers, IFRS 15.

This is part of the regular post-implementation review process “to assess whether the effects of applying the new requirements on users of financial statements, preparers, auditors and regulators are those the IASB intended when it developed the requirements.” Based on reviewing a few of the initial submissions, the perpetually intriguing area of identifying whether a party providing goods or services to a customer is acting as a principal or an agent could still use some fine-tuning. You may recall that a party makes this determination for each specified good or service promised to the customer, based on assessing whether it controls those goods or services before their transfer to the customer. The standard provides three main indicators that this is the case: that the entity is primarily responsible for fulfilling the promise to the customer; that the entity has inventory risk before the transfer to the customer; and that the entity has discretion in establishing the price for the specified good or service. Of course, these indicators may vary in relevance depending on the circumstances, and other indicators may provide more persuasive evidence in other contracts.

The Malaysian Institute of Certified Public Accountants provided the following list of challenges:

  • Unclear how control is transferred in relation to services in the context of principal vs agent consideration.
  • Unclear as to the duration an entity is considered obtaining legal title to a specified good only momentarily before the legal title is transferred to a customer. For example, an entity may be arranging transportation for goods to be delivered from its supplier’s premises to the customer’s premises.
  • Fragmented explanations or indicators … which make it difficult to understand and relate to the concept of control.
  • Tendency of entities to apply the indicators rather than the concept of control as it is easier to understand those indicators than the concept of control.
  • Applying indicators may point to different conclusions and as such, could result in different accounting treatments despite the same fact pattern.

The Swedish Financial Reporting Board agreed with that last point, citing a view among stakeholders that the control indicator relating to discretion in establishing price for the specified good or service lacks the same level of importance as the other two, and recommending that the IASB clarify if and how the indicators should be weighed. More generally, they said that “considering the development of business models over the past couple of years, we think the guidance needs to be supported by some new targeted illustrative examples to help preparers determine whether an entity is a principal or an agent. The need is in our view most important for assessment of electronical and digital services, and for intangible assets such as IP and licences (which) are often sold in bundles with tangible goods, supporting and integrating services as well as other separate services, which makes the control assessment complex.”

After also noting various issues, PricewaterhouseCoopers makes the following astute disclosure suggestion:

  • One further practical solution to mitigate the risk to comparability due to the inherent judgement in this area that we would propose to be considered is an enhancement to disclosure requirements. Sometimes two entities with similar fact patterns might still get to a different conclusion because there is inherent judgement present. In order to enhance comparability in such situations, if this was deemed to be a critical judgement and there is diversity in practice in the principal versus agent judgement for similar types of arrangements, the disclosure could require that;
    • If the entity has concluded that they are principal, then they should disclose what the revenue would have been if they had concluded they were an agent.
    • If the entity has concluded that they are an agent, then they should disclose what the revenue would have been if they had concluded that they were principal.

That might be the best there is within our current reporting model. It does rather illustrate though the limits of a restrictive either/or approach to accounting, in an era when we should be technologically and conceptually capable of moving beyond that. It seems that even with the best skill and integrity in the world, the principal vs. agent judgment may sometimes be inherently unclear, and one on which practitioners and auditors may differ; that’s even though the impact on key performance indicators such as revenue and gross profit of going one way rather than another might be about as material as any accounting judgment could be. Arguably, the main problem isn’t the difficulty of making a judgment, but that such a judgment has to be made at all, that a company can’t communicate its performance in a way that gives informed prominence to multiple ways of looking at its activities and allows users to understand the underlying complexities. That is, in the terms I used here, to deemphasize a circumscribed “story” in favour of an information “swarm”…

The opinions expressed are solely those of the author.

2 thoughts on “Principal vs. agent – we need more!

  1. Pingback: Post-implementation review of IFRS 15: we always knew it would be like this! | John Hughes IFRS Blog

  2. Pingback: Revisión posterior-a-la-implementación del IFRS 15

Leave a comment