Many transactions involve two or more unrelated parties that participate in providing goods or services to the customer. In some situations, it may not be clear which party has the performance obligation to provide the goods or services to the end consumer. The FASB attempted to simplify and enhance an entity’s determination of the performance obligations in these complex transactions by reducing the number of agent indicators that are used to determine if an entity is an agent or principal, and by providing an overarching control notion in its Accounting Standards Update (ASU) 2014-09 (also referenced as Accounting Standards Codification/ASC 606). The FASB has since issued ASU 2016-08, which clarifies the guidance on principal versus agent considerations.
This issue is significant to entities in many industries, especially software and technology entities that create or sell virtual goods on their websites or through smartphone applications. Entities that recognize revenue on a gross basis include the full amount of revenue with an associated cost of sales, whereas entities that recognize revenue on a net basis recognize a smaller revenue amount that is offset by the cost of sales. Therefore, the application of this standard is unlikely to change net income, but it can substantially impact the top-line revenue or gross profit percentages, which are often critical indicators used by investors to determine the value of an entity.
The core objective of this guidance is to help an entity determine whether it is a principal or an agent in a transaction. The classification of an entity as a principal or agent affects the amount of revenue recognized for the transaction: the principal recognizes revenues and costs associated with providing the goods or services at the gross amount, while an agent recognizes revenue at the net amount (i.e., the fee or commission the entity receives).
The primary difference between a principal and an agent is the performance obligation being satisfied—the principal has a performance obligation to provide the specified good or service to the end consumer, whereas the agent arranges for the principal to provide the specified good or service (ASC 606-10-55-36). This same guidance details that a contract with a customer may include more than one specified good or service, and in this case, an entity may be an agent for some and a principal for others. Additionally, a fundamental characteristic of a principal in a transaction is control: a principal substantively controls the goods or services before they are transferred to the customer (ASC 606-10-55-37).
An entity is a principal if, when involved in providing goods or services to a customer, that entity obtains control of any of following (ASC 606-10-55-37A):
- A good or another asset from the other party which the entity then transfers to the customer.
- A right to a service to be performed by the other party, which gives the entity the ability to direct that party to provide the service to the customer on the entity’s behalf.
- A good or service from the other party that it then combines with other goods or services in provided the specified good or service to the customer.
Many aspects of determining an entity’s extent of control are judgment based. ASC 606 offers three indicators to help an entity determine if it controls the specified good or service before it is transferred to the customer (ASC 606-10-55-39):
- The entity is primarily responsible for fulfilling the promise to provide the specified good or service;
- The entity has inventory risk before or after (i.e.; customer has a right of return) the specified good or service has been transferred to a customer;
- The entity has discretion in setting the price for the specified good or service;
Under ASC 606, no indicator is weighted heavier than any other indicator, and the list provided is not exhaustive. Other indicators not listed above may provide more convincing evidence for some contracts. Management should use judgment in determining which indicators provide the strongest evidence in determining whether the entity has control of the promised goods or services.
It is also important to note that an entity classified as a principal may satisfy a performance obligation by itself or it may subcontract another entity to fulfill the obligation on its behalf. A contractual side agreement such as this would not change the entity’s classification. However, if another party assumes the performance obligation in such a way that the entity is no longer required to fulfill the performance obligation, then the entity is no longer acting as a principal and does not recognize revenue for that performance obligation on a gross basis.
Under the new standard, there are no longer specific rules for shipping and handling. There is also no longer the policy election previously provided under the old standard for gross or net presentation of taxes. Tax collections should be presented based upon the substance of the tax arrangement instead of based upon a policy election.
Bond Entity is a company with a website, Bond.com, which provides online deals and coupons for customers. Various companies will offer their products and services on Bond.com, and based on contractual agreements will remit 10-30% of the sale price to Bond Entity. Silva Company agrees to sell its ergonomic chairs on Bond.com for $200 (the chairs cost Silva Company $100 to produce), and remit 10% of each sale to Bond Entity. Customers buy the chairs on Bond.com using credit or debit cards. Bond then arranges the shipping of the item from Silva’s warehouse to the customer. At the end of the year, Bond sold 1,000 chairs.
Bond Entity is an agent in the transaction.
(1) The entity is primarily responsible for fulfilling the promise to provide the specified good or service. Bond Entity is not responsible for providing the ergonomic chairs. Instead, it facilitates the sale of the ergonomic chairs from Silva Entity to the end customers.
(2) The entity has inventory risk before or after the specified good or service has been transferred to a customer. Bond Entity does not hold the inventory and does not lose money on unsold units.
(3) The entity has discretion in setting prices. Bond Entity agrees to sell the chairs at the price Silva offers to customers through the website, and cannot change that price.
Bond Entity does not seem to have control of the goods or services before they are transferred to the customer. Additional factors that could be included in determining control are (1) the 10% commission structure of Bond Entity’s revenue in this transaction and (2) the very limited consumer credit risk that Bond Entity will experience in credit card transactions. Thus, Bond Entity will recognize net revenue of $20,000. Silva Company will recognize gross revenue of $200,000 and a commission expense for the amount remitted to Bond Entity.
Diversity in Thought
Several potential implementation issues have been brought forth by stakeholders. These have been discussed in TRG memos, as well as ASU 2016-08. These matters are largely continuations of issues that have arisen under ASC 605, and are as follows:
- Issue 1: Application of the agency indicators
- Issue 1a: Interaction of the agency indicators with the concept that a principal controls the good or service before its transfer to the customer
- Issue 1b: Application of the agency indicators to complex contracts in which the indicators provide contradictory evidence
- Issue 2: What amount of revenue should the entity recognize if it determines that it is the principal, but received a net amount of cash and does not know the gross amount?
- Issue 3: How should the transaction price allocation guidance be applied to transactions in which the entity is a principal for some of the deliverables and an agent for others?
The first issue arises because of the new, explicitly stated control guidance. In the new standard, the determination of principal versus agent is founded upon the idea of control. If an entity controls the specified good or service that are ultimately passed through to the customers, that entity is considered the principal to the transaction, whereas an entity that does not possess control of the delivered good or service is merely an agent in facilitating the transaction. There are two primary views that were discussed in detail in the TRG meetings.
View A. The control criterion should be the determinative criterion for classifying an entity as principal or agent. The standard states that an entity is a principal if it controls the promised goods or services before transferring the goods/services to the end customer. Furthermore, the Board’s basis for conclusions indicates that the nature of the promised goods or services (i.e., the performance obligation) may not be readily apparent.
This complexity often occurs when control of the goods or services are transferred almost instantaneously, or when it is unclear what rights the intermediary holds for the promised goods or services. “For that reason, the Boards included indicators . . . to help an entity determine whether the entity controls the goods or services before transferring them” (BC par. 382). There is a strong case that if an entity has control of the goods or services then the indicators are not needed to evaluate whether the entity is the principal or the agent.
View B. The indicators are tools used to determine if an entity has control. That is, it would be impossible to separate the control criterion from the indicators because the indicators are suggestive of control (or the lack thereof). Proponents of this view argue that the analysis of the indicators is very similar to the old standard. In the old standard, there was no notion of control; however, a general concept of control was embedded within the indicators (e.g., primary obligor and general inventory risk). Thus, the stakeholders with this view would argue that this difference between the old and new standard in terms of application of control is primarily the Board’s explicitly stating the control notion. Accordingly, the indicators should be applied in a similar manner as the old standard, with the indicators providing evidence of control and determining the classification of the entity as a principal or agent.
These issues were discussed by the TRG and the ASU 2016-08 update states that the indicators in ASC 606-10-55-39 are to support or assist with the assessment of control, not determine it outright. ASC 606-10-55-320 through 323 (Example 46) highlights an example where the determination of control is reached without a discussion of the indicators. It also states that the list of indicators is not exhaustive and that no single indicator has more or less weight within the standard—judgment is required in each instance and the facts of the situation will determine weight given to the indicators. View A is more in accordance with ASU 2016-08 than View B; ASC 605 focused more on risk and reward with its indicators (the pre-update indicators in 606) and ASC 606 introduces control as the main criterion.
An equally difficult situation arises for entities when no obvious indication of control exists and the indicators provide contradictory evidence. One example of this is a virtual good sold by an intermediary to an end customer. In some situations, the originator will receive a fixed amount, and the intermediary can choose the selling price of the good. However, the originator is ultimately responsible for providing the promised virtual good to the customer. In this circumstance and in similar situations, one entity seems responsible for providing the ultimate good or service to the end customer, but another entity seems to be responsible for pricing (i.e., controlling the benefits of the good or service).
Some stakeholders have argued that inventory risk does not apply to situations for which there is no physical inventory (e.g., virtual goods or services). Other stakeholders have posited that the inventory risk relates to the contract being cancellable or non-cancellable. Thus, if an intermediary makes a purchase for a nonrefundable amount, it has inventory risk for that minimum amount. As a brief illustrative example, if an intermediary in a virtual goods transaction paid a fixed, nonrefundable amount for access to a certain amount of virtual goods from an originator, some stakeholders believe that the intermediary has inventory risk for those virtual goods, even though that intermediary does not have physical inventory.
Entities must exercise significant judgment when indicators of control conflict. An entity should evaluate which indicators are most suggestive of control in the specific transaction (these may change for different types of transactions) and make a classification based upon which indicators strongly suggest control or the lack thereof.
Another challenging situation arises in contracts for which the entity determines that it is the principal, but it does not know the final transaction price. For example, a steakhouse or movie theater could sell gift cards to Costco at a certain price, and then Costco could choose to sell those gift cards at whatever price it chooses. If the entity (the steakhouse or movie theater) performs an analysis of control and determines that it is the principal in the transaction, should it recognize gross revenue at the amount it sold the gift card to Costco, or the amount at which Costco ultimately sells the gift card? This issue is further complicated when the entity does not know the price paid by the end customer.
View A. In situations where the entity providing goods or services may not know the end price paid by the customer, the intermediary is considered the end customer. In essence, there are two principals in the transaction: the entity is the principal to the intermediary, and the intermediary is the principal to the end customer. This view may be justified by explaining that the principal’s primary performance obligation is to give the intermediary the right to provide its customers access to a certain amount of its goods or services. The intermediary can then sell that access to its customers at whatever price it chooses. Under this view, the transaction price should reflect the amount of consideration to which it is entitled from the intermediary.
For example, if Carmike Cinemas sells a gift card to Costco, proponents of this view would argue that Carmike is the principal to the transaction and Costco is the end customer. Then, if Costco later sold that gift card to a Costco member, then Costco would be the principal in that transaction and the store member would be the end customer. Both Carmike and Costco would recognize gross revenue in their respective transactions.
View B. These transactions should be considered variable consideration with no constraint because the entity’s customer is the end customer. Proponents of this view believe that the amount to which the entity is entitled is the amount charged to the end customer, which is variable because the intermediary may charge different prices for the goods or services. Thus, an entity would have to estimate the amount of variable consideration to which it expects to be entitled (see Variable Consideration article for more detail).
View C. These transactions should be considered variable consideration, except that the constraint on variable consideration should apply. Proponents of this view believe that the amount to which the entity is entitled is the amount charged to the end customer, which is variable because the intermediary may charge different prices for the goods or services. However, these stakeholders believe that the constraint should apply because the amount of consideration is very susceptible to outside factors, the uncertainty about the amount of consideration is not expected to be resolved for a long time, and the contract has a large number of possible consideration amounts.
View D. There is no variable component to these transactions, and rather the consideration is fixed. Unlike View A, these stakeholders assert that the entity is ultimately providing goods or services to the end customer (and not the intermediary). However, unlike Views B and C, the fixed amount the entity is entitled to from the intermediary would be the transaction price. The supporters of this view argue that the additional amounts paid by the end customer above the fixed amount the entity receives from the intermediary do not qualify as variable consideration. Any variability is controlled by, and benefits, the intermediary.
In the July 2014 TRG meeting, most of the members seemed to coalesce around Views A and D, for which the entity would recognize the amount sold to the intermediary as the gross amount, and would disregard the transaction price of the intermediary to the end customer. This appears to be for three reasons. First, recognizing revenue at the fixed amount of consideration provided by the intermediary would be less costly for an entity to estimate than trying to determine the end price that would be charged by the intermediary to the end customers. Second, some members of the TRG seemed concerned with situations in which the intermediary provides goods or services at a discount (or for free) as part of a promotional tactic. Some TRG members argued that in such circumstances, it did not seem appropriate for the entity to recognize less revenue than it received from the intermediary simply because of an operational or marketing choice by the intermediary. Finally, most participants seemed to indicate that in practice, entities are not typically estimating the end price paid by the customer. Instead, they are recognizing the revenue received by the intermediary.
We believe that Views A and D most faithfully represent the economics of the transaction, and are also the views that are most practical to implement. The Board in ASU 2016-08 decided to not address this issue because it is not pervasive and to maintain convergence with IFRS 15, which IASB has not made amendments to regarding this issue. In addition, the Board reminds the reader that ASC 606-10-32 contains instruction on uncertainty within consideration.
The third area of concern currently discussed by stakeholders is the application of discounts to bundles of goods or services for which an entity is the principal for some of the goods/services and an agent for other goods/services in the bundle. The two views are to (a) allocate the discount to all performance obligations, regardless of whether the entity is the principal or agent (both gross and net amounts), and (b) allocate the discount only to the transactions for which the entity is the principal (only gross amounts).
View A. An entity should allocate the discount to all performance obligations, regardless of whether the element is a gross or net element. If the entity has a single customer in the transaction, some stakeholders think that applying the discount to all goods/services in the bundle is more consistent with allocation guidance in the revenue standard.
View B. An entity should recognize its fee or commission for any net elements and the discount should be applied only to gross elements. If the contractual arrangement of the bundle is such that the entity providing the goods/services has different end customers for the different promised goods or services, it is likely inappropriate to allocate the discount to all the goods and services. The contract combination guidance does not allow an entity to combine arrangements with two or more unrelated third parties. In such a circumstance, View B seems more appropriate. Thus, an entity should evaluate whether there is more than one customer privy to the contract.
The TRG did not spend much time discussing this issue. In fact, one TRG member expressed the opinion that this issue would likely be resolved by coming to a resolution for Issues 1 and 2. Our view is consistent with this outlook: it does not seem to be an area of great controversy. The application simply seems to differ based upon the determination of end customers to the entity.
Comparison to 605
The guidance under ASC 606 is fairly similar to ASC 605. Many of the complex issues that arose under ASC 605 are likely to arise in ASC 606 as well. However, there are some differences between the old and new standard.
The first key difference between ASC 605 and 606 is the concept of control. Unlike ASC 605, ASC 606 explicitly emphasizes the concept of control as a key determinative factor in whether an entity is a principal or an agent. Under ASC 605, entities are provided with a list of eight indicators (two of which are considered strong indicators) that are determinative in evaluating whether an entity is a principal or an agent. Similarly, ASC 606 has three indicators of principal relationship that are used to assist an entity in determining whether it has control or not.
Another difference between the two standards is that the three indicators in ASC 606 are not weighted and that ASC 606 allows entities to consider other indicators not explicitly listed in the guidance. However, as mentioned above in the Diversity in Practice section, the application of these indicators may be very similar under both standards for some complex transactions because some stakeholders think that weighting may be necessary in certain circumstances.
A final key difference is that ASC 605 provides specific rules for shipping and handling, as well as for sales and excise taxes. The only guidance provided under ASC 606 for shipping and handling fees or taxes is that the transaction price should exclude “amounts collected on behalf of third parties.”
Determining the principal/agent to a transaction under the new revenue standard is quite similar to the current practice of ASC 605. The concept of control is more prevalent under the new standard, and the indicators have been consolidated and given equal weight. There is continued debate regarding some key application issues in the new standard, such as the interaction between the agency indicators and the notion of control, as well as determining the gross amount recognized when an entity does not know the price charged to the end customer. We expect that these complex issues will continue to be discussed by the stakeholders, standard setters, and industries to determine how to best reflect the transfer of control and risk in transactions where goods or services pass through multiple entities to get to the final customer.
- Accounting Standards Codification (ASC) 606, Revenue from Contracts with Customers. ASC 606-10-55-36 to 40.
- Accounting Standards Codification (ASC) 605, Revenue Recognition ASC 605-10-45.
- FASB ASU 2016-08, Revenue from Contracts with Customers, “Principal versus Agent Considerations” (March 2016)
- EY Technical Line, A closer look at the new revenue recognition standard (June 16, 2014), Section 4.4.
- FASB Joint Transition Resources Group, Staff Paper “Gross versus Net Revenue” (July 18, 2014).
- KPMG Issues In-depth, Revenue from Contracts with Customers (September 2014), Section 10.3.
- PwC, Revenue from Contracts with Customers (Global Edition) (July 31, 2014), Chapter 10.
- PwC, Revenue from Contracts with Customers (Retail and consumer industry supplement) (September 8, 2014), pp. 17-19.