Accounting for credit card rewards programmes: the perceptions of managers and experts

Abstract

Purpose

International Financial Reporting Standard (IFRS) 15 required credit card rewards programmes (CCRPs) to reconsider their accounting practices. While Brink and Steenkamp (2023a, 2023b) developed a theoretical accounting model for CCRP transactions after the effective date of IFRS 15, this model should be validated and finalised as an accounting framework. Thus, the purpose of the present paper was to examine the validity of Brink and Steenkamp’s (2023b) model by interviewing CCRP managers and obtaining the opinions of experts in the field, and then develop a framework for accounting for CCRP transactions after the effective date of IFRS 15.

Design/methodology/approach

A qualitative exploratory approach within an interpretive paradigm was applied. Fifteen semi-structured interviews were conducted with South African CCRP managers, after which the Delphi technique (with 22 experts) was used. All data collected were analysed using thematic analysis, after which the CCRP accounting framework was finalised.

Findings

The study confirmed parts of the theoretical model, updated the model for what was evident in practice (e.g., not identifying interest as a relevant revenue stream, not differentiating between an open-loop and closed-loop structure and not including interest in the interchange fee) and improved the model by including alternative accounting treatments and additional guidance (e.g., to determine how the CCRP transaction should be viewed and to determine the value of award credits without an observable value).

Practical implications

The CCRP accounting framework provides practical guidelines for CCRP accounting and will assist managers of CCRPs in their decision-making processes and the application of judgement.

Originality/value

The study developed a CCRP accounting framework embedded in a decision tree and included all possible alternatives for accounting for CCRP transactions, which is a novel contribution to the field.

Citation

Brink, S., Steenkamp, G. and Odendaal, A. (2024), "Accounting for credit card rewards programmes: the perceptions of managers and experts", Qualitative Research in Financial Markets, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/QRFM-04-2024-0080

1. Introduction

The International Financial Reporting Standards (IFRS) provide regulations so that annual financial statements can be consistent, transparent and comparable around the world (IFRS, 2021). With the issue of IFRS 15 Revenue from Contracts with Customers (which became effective for accounting periods beginning on or after 1 January 2018), stakeholders raised various concerns stating that the standard provided minimal guidance on the accounting treatment of credit card rewards programme (CCRP) transactions (IFRS, 2012). In response to stakeholder concerns, the International Accounting Standards Board (IASB) advised management to consider the structure and functioning of these complex arrangements and then apply their discretion when accounting for CCRP transactions (EY, 2013; PwC, 2016). Therefore, after the effective date of IFRS 15, CCRP management must apply their judgement to determine how to account for these unique and complex agreements (EY, 2013).

As a starting point to address the research problem of accounting for CCRP transactions after the effective date of IFRS 15, Brink and Steenkamp (2023a, 2023b) considered existing literature (including IFRS) and developed a theoretical model for accounting for CCRP transactions. However, it was clear that clarity on the most appropriate accounting treatment of a CCRP transaction cannot be gained by only considering literature (Brink and Steenkamp, 2023b). It was evident that a comprehensive understanding of the experiences of CCRP management, as well as the opinions of experts in the field, was vital in reaching an answer regarding the most appropriate accounting treatment for CCRP transactions, because IFRS 15 leaves it to the discretion of management (Brink and Steenkamp, 2023a, 2023b). The aim of the present paper was to examine the validity of Brink and Steenkamp’s (2023b) model by interviewing CCRP managers and obtaining the opinions of experts in the field, and then develop a framework for accounting for CCRP transactions after the effective date of IFRS 15.

The study confirmed parts of the theoretical model, updated the theoretical model for what was evident in practice (e.g., not identifying interest as a relevant revenue stream, not differentiating between an open-loop and closed-loop structure and not including interest in the interchange fee) and improved the model by including alternative accounting treatments and additional guidance (e.g., to determine how the CCRP transaction should be viewed and to determine the value of award credits without an observable value). This study addressed the gap that was identified in the body of accounting knowledge by developing a CCRP accounting framework, embedded in a decision tree, which includes all possible alternatives for accounting for CCRP transactions - a novel contribution to the field. This study contributes to practice by providing stakeholders (CCRP management and auditors) with a best-practice CCRP accounting framework for the recognition and measurement of such transactions. Applying the CCRP accounting framework developed by this study could ensure faithful representation of the underlying CCRP transaction, and comparability between similar companies will be achieved or enhanced. A further contribution lies within the practice side of implementing new standards to complex transactions that require management judgement.

2. Literature review

Applying general, principle-based accounting standards to very specific and complex transactions comes with various challenges, as was evident with the issue of IFRS 15 and its application within CCRP transactions. This literature review will start by explaining the characteristics of CCRPs. Thereafter, the effect of the new accounting standard (IFRS 15) on the accounting treatment of CCRPs will be discussed, by specifically considering prior research on the matter.

2.1 Explaining characteristics of credit card rewards programmes

The characteristics of CCRPs that complicate the accounting treatment is the fact that these programmes are so complex with multiple parties involved (for example, the card issuer, the cardholder and the merchant) (Brink, 2017b; FASB and IASB, 2013). CCRPs also differ in their structure and functioning − CCRPs can be structured in the form of a simple CCRP (swipe only), a complex CCRP (linked to other products) and a co-brand CCRP (Brink et al., 2023). This results in CCRPs being unique. The basic functioning of a credit card arrangement (on which a typical CCRP is based) can be explained as follows: when a cardholder purchases goods or services from the merchant, the merchant receives an amount of cash from the card issuer that is slightly less than the original invoiced price for the goods and services acquired by the cardholder. This difference between the invoice price and the cash paid to the merchant is referred to as the merchant interchange fee (FASB and IASB, 2013). The card issuer may then also administer a CCRP as part of a credit card arrangement. In terms of the rewards programme, the cardholder receives award credits from the CCRP for each credit card purchase transaction from a merchant (FASB and IASB, 2013). Brink et al. (2023) documented the structure and functioning of simple, co-brand and complex CCRPs and found that CCRPs’ structure and functioning differ from each other and that their structure and functioning have a direct impact on the accounting treatment, justifying a variety of accounting treatments.

2.2 Challenge of implementing International Financial Reporting Standard 15

Implementing a new accounting standard, like IFRS 15, has various implications for preparers (Davern et al., 2019; Napier and Stadler, 2020; Onie et al., 2022). The following industries experienced challenges when applying the principle-based guidance of IFRS 15, which required management’s judgement: telecommunication, retail, oil and gas, construction and financial industries (Brink and Steenkamp, 2023b; Ogunode and Salawu, 2021; PwC, 2018; Usurelu and Dutescu, 2021; Van Wyk and Coetsee, 2020). For CCRPs, the main debate and challenge with the introduction of IFRS 15 was whether CCRPs were included in the scope of IFRS 15 (EY, 2014). Other concerns included the interaction with financial instruments, different treatments for cash versus non-cash rewards, identifying the relevant revenue stream and the value of award credits without an observable value (Brink, 2017b; IFRS, 2012; Brink and Steenkamp, 2023a, 2023b) considered existing literature in an attempt to address the research problem of accounting for CCRP transactions after the effective date of IFRS 15. Brink and Steenkamp (2023a) found that a CCRP transaction can be viewed in isolation as marketing or as an integral part of the credit card transaction. If the transaction is viewed as a marketing tool, the CCRP transaction should be accounted for as an expense and a liability. If the transaction is viewed as an integral part of the credit card transaction, the CCRP transaction should be accounted for as a reduction of income (deferred income) (Brink and Steenkamp, 2023a). A theoretical model for CCRP accounting (illustrated in Figure 1) was developed by Brink and Steenkamp (2023b).

The theoretical model has, however, not been updated for what is evident in practice, based on the experiences of CCRP managers and other experts. Insights from practice would clear up various unanswered questions relating to the model, including: whether the interchange fee includes interest and, if so, how to allocate the transaction; how the value per award credit without an observable value is determined; if interest or a combination of interest and interchange fee can be identified as the relevant revenue stream, and if so how to allocate the transaction; whether other relevant revenue streams are identified based on the type of CCRP and how to account for those transactions; whether both the merchant and the cardholder can be identified as the card issuer’s customer and if so how to allocate the transaction; if the structure of the CCRP (i.e. open loop vs closed loop) has an effect on the accounting treatment of the transaction; and if a contract with the merchant is a condition to identify the merchant as the card issuer’s customer.

3. Methods and data collection

A qualitative exploratory approach was used to obtain rich, descriptive information on how CCRP managers and experts perceive, understand and interpret the underlying CCRP transaction (McKenzie and Danforth, 2014; Myers, 2009). The interpretive paradigm (one that seeks to understand) was selected to investigate the experiences of CCRP management and the opinions of experts (Lather, 2006). The interpretive paradigm is well suited to accounting research and lends itself to exploring existing accounting practices and proposing recommendations for new ways of doing things (De Villiers et al., 2019). The CCRP accounting framework would not only provide one answer on accounting for CCRP transactions but rather incorporate various alternatives (falling within interpretivism). There were two phases of data gathering, namely, through semi-structured interviews and the Delphi technique, both of which are discussed later in the paper.

Brink et al. (2023) identified South Africa as a good starting point to address the problem of accounting for CCRP transactions. The people directly involved in the CCRP transaction (those who develop and run CCRPs and those responsible for accounting for the transaction) would be able to provide rich and descriptive data relating to the phenomenon under study. Selecting CCRP managers, including the head of the CCRP, the accountant responsible for recording the CCRP transactions, and, if applicable, the IFRS advisory team, will facilitate the selection of an information-rich sample from which the most can be learned regarding the phenomenon under study (Creswell and Guetterman, 2019; Merriam and Tisdell, 2016). There are 12 CCRPs operating in South Africa, and for the semi-structured interviews, the managers of all 12 South African CCRPs were purposively [1] selected and invited to participate in the study. CCRP managers were defined as the people directly involved in the CCRP transaction. Semi-structured interviews were used to gain rich and in-depth information (Polkinghorne, 2005; Ryan et al., 2009) from CCRP managers regarding their experiences in decision-making and applying their judgement to account for CCRP transactions after IFRS 15. Durocher (2009) suggested the use of interviews in accounting research to facilitate the understanding of the nature and characteristics of the decision-making process by users in practice. Hossain et al. (2022) highlighted the importance of stakeholders’ perceptions on the accounting treatment of specific transactions and that semi-structured interviews can be applied to examine stakeholders’ perceptions. The managers of eight of the 12 South African CCRPs participated in the study – leading to 19 participants being interviewed by one of the researchers. In two instances, the head of the CCRP required the accountant(s) to sit in on a single interview, and separate interviews were not conducted. In such cases, a single participant number was allocated to represent all parties present in the interview. 14 Interviews (referred to as P1–P14) and a single follow-up interview were conducted, totalling 15 interviews. Refer to Appendix 1 for more details.

To allow the interviewees to relate their experiences of accounting for CCRP transactions after IFRS 15, while still encouraging flexibility in responding (Simon et al., 2014), a series of open-ended questions and probes were applied (Ryan et al., 2009). The researcher who conducted the interviews was able to explore any spontaneous issues raised by the interviewees and obtain clarifications for answers, where required. The interview guide was determined by the aim of the study (Ryan et al., 2009) and informed based on concepts identified in the literature (Barriball and While, 1994), specifically the key elements of the underlying CCRP transaction that impact the accounting treatment (Brink, 2017a, 2017c). The interview guide ensured that the data obtained from the interviews were sufficient to enable the researchers to validate the theoretical model and finalise a framework for accounting for CCRP transactions. Refer to Appendix 2. To document the data obtained in an accurate and factual manner, the interviews were recorded (totalling 12 h and 19 min of recordings) and transcribed, which resulted in 362 pages of data with a word count of 111,534. The researcher recorded her observations and personal thoughts by keeping field notes during the interviews (Creswell and Guetterman, 2019). The field notes were used to identify new probes for subsequent interviews. The first interview served as a pilot interview to ensure that the interview guide was properly developed and minor adjustments (for example, introducing the research in a less formal way and changing the order in which the questions were asked) were made for the interviews that followed. To develop the CCRP accounting framework, the Brink and Steenkamp (2023b) model was updated based on the data obtained from the semi-structured interviews.

There are various forms of action research that can be applied to confirm the CCRP accounting framework (Given, 2008), including a single case study, multiple case studies (Naslund, 2002), follow-up interviews (Skinner, Edwards and Corbett, 2014) and the Delphi technique (Vernon, 2008). Using a single case study approach, the researchers would only observe one CCRP applying the framework. Although in-depth feedback and suggestions would be obtained from the selected CCRP, this will not represent all CCRPs in the South African market. Selecting multiple CCRPs for simulation purposes (a multiple case study approach) would provide a wider range of suggestions and feedback, but still a more comprehensive form of confirming the framework was sought. Follow-up interviews focus more on member checking, i.e. the interview participants confirming that their perceptions have been accurately captured and included in the framework. Instead of just confirming the data obtained from the participants, the researchers wanted to add an additional layer by asking subject experts (including participants other than the interview participants) to review the conceptual soundness of the entire accounting framework. Since this study aimed to make a practical contribution, the researchers wanted to gain confirmation across different contexts. The Delphi technique offered a confirmatory method that included not only all South African CCRPs but also subject matter experts in confirming the framework, and it was thus selected as a confirmation technique in this study.

The Delphi technique was therefore considered appropriate to validate the theoretical accounting model, as the aim was to gain consensus to address complexity and uncertainty in an area where knowledge was imperfect, incomplete or unknown (Amos and Pearse, 2008; Donohoe and Needham, 2009). The Delphi technique required that a panel of experts be selected (Clayton, 1997), and the success of the Delphi technique depended principally on this selection (Chan et al., 2001). The 12 South African CCRP’s managers and the auditors of the 12 South African CCRPs were purposively selected as they possessed different expertise and skills relevant to the validation of the CCRP accounting framework and could, therefore, each contribute in a different way. The heads of the CCRPs could be relied upon to understand the dynamics of the programme and the underlying transaction; the accountants and members of the IFRS advisory team were experts in recording the transactions in accordance with the IFRS; and the auditors were able to give an opinion on whether or not the transaction was faithfully recorded and met the requirements of the IFRS. In total, 21 experts participated in round one of the Delphi technique (referred to as E1–E21) and 12 experts (referred to as E22–E33) in round two. Eleven of the experts who participated in round one also participated in round two. Refer to Appendix 3 for more details.

The Delphi technique used sequential rounds, each one building on the findings of the previous rounds, resulting in consensus in the final round (De Villiers et al., 2005). For the purposes of round one, both the CCRP accounting framework and a questionnaire were sent to the panel of experts who agreed to participate. The first round of the survey included a questionnaire with questions relating to the flow, accuracy and completeness of the developed CCRP accounting framework. Feedback was obtained from 21 experts, and the data were analysed, and, where necessary, the CCRP accounting framework was updated.

The revised version of the CCRP accounting framework was presented to the same panel members in round two and was accompanied by a detailed description of adjustments based on perspectives received during round one. Also, when adjustments were not made, the reasons were provided. The second round questionnaire enquired whether the panel members were satisfied with the revised CCRP accounting framework and, if not, recommendations and any other comments relating to decision-making were required. In the second round of the Delphi technique, feedback was obtained from 12 experts. Even though a high attrition rate was evident, which is typical in further rounds of the Delphi technique (Donohoe and Needham, 2009), no contradictory views between panel members or major issues were identified during both rounds. Also, only minor adjustments to the CCRP accounting framework were made. After two rounds of the Delphi technique, optimal data saturation, expert opinion stability and collective agreement were achieved, and another round was not necessary (Vernon, 2008).

This study followed Braun and Clarke’s (2006) step-by-step guide for thematically processing and analysing the data obtained from the semi-structured interviews and the Delphi technique. The transcriptions of the interviews and the feedback received from experts were read and re-read facilitating familiarisation with the data. During the process, initial areas of interest were identified through writing down preliminary ideas and observations (Step 1). The data was coded by using ATLAS.ti and coding interesting features of the data in a systematic fashion across the entire data set, and collating data relevant to each code (Step 2). During Step 3, the researchers searched for themes and meaningful patterns in the data through a process of clustering codes around patterns of meanings that captured important principles in the data that directly related to the research aim. Examples of codes identified included “Identifying a new revenue stream” (from the interviews) and “Explaining to whom the framework applies” (from the expert feedback). Themes were identified with reference to the key elements of the underlying CCRP transaction, including for example “Customer in relation to the card issuer”. The codes and themes identified were reviewed and refined by checking whether they worked in relation to the coded extracts and the entire data set, generating a thematic map of the analysis (Steps 4 and 5). In accordance with Step 6 the findings were structured according to the themes that were determined. The assigning of the various quotes that were coded (representing vivid, compelling extract examples) to the themes identified were reconsidered to craft the story that the data were telling.

A combination of an inductive and deductive approach was followed with prominence given to the inductive approach to analyse the data. The researchers mainly coded from the data, allowing for codes and groups to be identified based on participants’ insights and experiences (inductive). The data were further categorised into themes that were determined deductively (Hsieh and Shannon, 2005). This analysis prioritised data-based meaning (a bottom-up approach) over theory-based meaning (a top-down approach) (Braun and Clarke, 2012). This approach provided a rich description of the data (inductive) and not just a detailed analysis of some aspects of the data (deductive) (Braun and Clarke, 2006). The identified themes were applied to finalise and validate the CCRP accounting framework. Institutional permission was obtained from the organisations that ran a CCRP and the audit firms, and informed consent was obtained from the CCRP managers and audit partners prior to the interviews. Ethical clearance was obtained prior to commencing with the research.

4. Findings and discussion

This section explains the two phases used to examine the validity of Brink and Steenkamp’s (2023b) model and to develop a framework for accounting for CCRP transactions. The first phase describes CCRP managers’ experiences in accounting for CCRP transactions after IFRS 15 (in subsections 4.1–4.4). The second phase obtains the opinions of experts in the field (in subsection 4.5). Based on these findings, the CCRP accounting framework is presented (see Figure 2 in subsection 4.6).

4.1 How is underlying credit card rewards programme transaction viewed in practice?

The literature identified two possible management views on CCRP transactions, namely “in isolation as marketing” and as “an integral part of the credit card transaction” (Brink and Steenkamp, 2023a). How CCRP transactions are accounted for in practice was compared to the two possible management views. Participants 7, 10 and 11 concluded that the aim of their CCRP was to serve as a marketing tool and, in line with the literature (Brink and Steenkamp, 2023a), accounted for the transaction as an expense and liability. Participants 1, 2, 3, 4, 5 and 12 concluded that their CCRP was not purely a marketing tool but rather formed an integral part of a revenue transaction and, in line with the literature (Brink and Steenkamp, 2023a), accounted for the transaction as a reduction of revenue (deferred revenue).

Participants 9 and 13 (both managers of complex CCRPs) also viewed the transaction as an integral part of a revenue transaction, but instead of deferring revenue, a separate gross revenue and an expense and liability were recognised. Though this accounting treatment contrasted with existing literature, the CCRP transaction was still recognised by applying a revenue standard’s (IFRS 15) principles. In a complex CCRP which is linked to other products, there are multiple revenue streams coming from various cardholders relating to the several products, and there is no direct link between the multiple revenue streams and the award credits liability per cardholder (P8; P9; P14). The structure and functioning of these complex CCRPs create a disconnect between the revenue and the award credits liability, making the application of the deferred revenue model impractical (P8; P9). Participants 8, 9 and 13 applied their judgement and concluded that, even though there is a disconnect between the revenue stream and the award credits liability, the nature of the transaction indicates that it forms part of a revenue transaction and that IFRS 15 is applicable, even if IFRS 15’s deferral model is not applied.

Participant 11 provided some further guidance on distinguishing between viewing a CCRP as a marketing tool as opposed to an integral part of a revenue transaction. Participant 11 stated that for a rewards programme linked to other products (a complex CCRP) the CCRP “forms an integral part of the underlying business […] it’s what pulls their business together in terms of their different business lines”. Participant 11, therefore, believed that where a rewards programme is linked to other products of the financial institution, this would indicate that the CCRP transaction formed an integral part of a bigger revenue transaction. Specific guidance to determine whether the CCRP transaction is viewed as a marketing tool rather than as an integral part of the revenue transaction was missing from the literature (Brink and Steenkamp, 2023a, 2023b) and therefore included in the CCRP accounting framework. To keep the decision tree dedicated to the main decision criteria in accounting for award credits, it was decided that all additional guidance would be presented separately as an addendum (presented in Figure 3).

Considering the alternative to view the transaction as an integral part of a revenue transaction, Participants 2, 3, 5, 9, 12 and 13 indicated that, according to their experience, even if the transaction were viewed as part of a revenue transaction, there would always be some sort of marketing element included. To stay competitive in an environment where everybody else has rewards programmes, a card issuer cannot afford not to have a CCRP that automatically acts as a marketing tool (P3). Participants 5, 7 and 11 agreed, stating that a CCRP attracts new customers. The decision tree was updated to expand the definition of “viewed as an integral part of a revenue transaction” to include “a marketing element to stay competitive”. If the CCRP transaction is viewed as such, the transaction should be accounted for as a reduction of revenue (deferred revenue) or in terms of a revenue standard.

4.2 Credit card rewards programme transaction outside scope of International Financial Reporting Standard 15

If the CCRP transaction is viewed “in isolation as marketing”, the transaction is not regarded as a revenue transaction and would fall outside the scope of IFRS 15. If the CCRP transaction is viewed “in isolation as marketing” the nature of the benefits, which have an impact on measuring the award credits, should be considered first. If the nature of the benefits supplied is a non-cash reward, the award credits should be accounted for in terms of IAS 37 (expense and provision, measured at a value from the card issuer’s perspective adjusted with the expected redemption option ratios) (Brink and Steenkamp, 2023b; Brink et al., 2023). Participants 10 and 11 (managers of a simple CCRP), who viewed their CCRP transaction as a marketing tool that offers non-cash rewards, agreed with the IAS 37 accounting treatment in terms of recognition and measurement. Literature indicates that some CCRPs applying IAS 37 to account for the CCRP transaction recognise the cost of award credits as an offset to merchant interchange fee income instead of an expense (Brink, 2017b). This alternative was not evident from the interviews and, therefore, not included in the CCRP accounting framework (P6; P7; P10; P11) (Figure 2).

In contrast, if the nature of the benefits supplied is a cash reward, the award credits should be accounted for in terms of IFRS 9 (expense and financial liability, measured at the cash amount payable representing the fair value from the cardholder’s perspective, not adjusted for the expected redemption rate) (Brink and Steenkamp, 2023b; Brink et al., 2023). Participants 1, 3, 4, 5, 10 and 13 agreed, stating that if the nature of the benefits were a cash reward, that would automatically mean that IFRS 9 is applicable (whether the transaction is viewed as a marketing tool or as an integral part of a revenue transaction).

4.3 Credit card rewards programme transaction possibly within scope of International Financial Reporting Standard 15

In accounting for a CCRP transaction that is viewed as an integral part of a revenue transaction, management need to apply their judgement to identify the relevant revenue stream and the customer in relation to the card issuer. These key elements of the underlying CCRP transaction are discussed next.

4.3.1 Relevant revenue stream.

If management views the CCRP award credits primarily as an integral part of a revenue transaction, it is crucial to identify this relevant revenue transaction (revenue stream) to determine whether the transaction falls within the scope of IFRS 15 (Brink and Steenkamp, 2023b). According to Participant 5, the structure and functioning will determine which revenue stream is identified as the relevant revenue stream. Brink (2017b) determined that the relevant revenue stream can be identified with reference to the rationale behind the CCRP that gives rise to the CCRP transaction. Participants 3, 4, 11 and 13 agreed, as illustrated in the comment by Participant 3:

If we look at the principles that IFRS 15 is trying to encourage us to follow, we need to look at what revenue are we getting, what product are we providing, and then from that, you know, that’s what drives a loyalty programme […] So, I personally do think that should be a driver in the decision-making.

According to the literature, there are five possible revenue streams to consider in a CCRP transaction (Brink, 2017b; Brink et al., 2023):

• consideration for taking on the award credits liability in a co-brand CCRP;

• a pool of revenue in a complex CCRP (linked to other products);

• a membership fee in a simple CCRP;

• the interest on the credit card loan; and

• the interchange fee income.

The first three revenue streams listed above were identified from practice (Brink et al., 2023), and were thus included in the CCRP accounting framework (Figure 2). In co-brand CCRPs, where the CCRP (retailer) takes over the award credits liability from the card issuer and receives consideration for it, that consideration received is identified as the relevant revenue stream (Brink et al., 2023). If the transaction in the co-brand CCRP is viewed as an integral part of a revenue transaction and the nature of the benefits is a non-cash reward, IFRS 15 will be applicable (Brink et al., 2023). In terms of IFRS 15, the retailer shall recognise the contract liability (for the award credits granted) at the value of the consideration received (transaction price). This represents a value from the cardholder’s perspective, adjusted for the expected redemption rate (Brink et al., 2023).

In a complex CCRP (linked to other products), access fees are received by the rewards programme from the various product houses, resulting in a pool of revenue, and this pool of revenue is identified as the relevant revenue stream (Brink et al., 2023). If the transaction in the complex CCRP is viewed as an integral part of a revenue transaction and the nature of the benefits is a non-cash reward, IFRS 15 will be applicable (Brink et al., 2023). Gross revenue will be recognised for the access fees received from the various product houses, as well as a separate expense and an IFRS 15 liability for the award credits granted. The award credits should be measured at a value from the cardholder’s perspective, taking into account the expected redemption rate (Brink et al., 2023).

Some simple CCRPs (swipe only) identify the membership fee as the relevant revenue stream in the decision-making process. These simple CCRPs identified the relevant revenue stream by considering what funds the CCRP. According to Participant 1, the membership fee funded the CCRP and was therefore identified as the relevant revenue stream. If the transaction in the simple CCRP is viewed as an integral part of a revenue transaction, and the nature of the benefits is a non-cash reward, IFRS 15 would be applicable (Brink et al., 2023). In terms of the deferral model of IFRS 15, the consideration received must be allocated between revenue (membership fee) and deferred revenue (award credits). The allocation is based on relative stand-alone selling prices, and the stand-alone selling price of the award credits is a value from the cardholder’s perspective, adjusted for expected redemption rate (Brink and Steenkamp, 2023b; Brink et al., 2023).

The last two potential revenue streams in the numbered list above (interest on the credit card loan and interchange fee income) were considered for inclusion in the CCRP accounting framework, based on whether they were evidenced by the data from the interviews. In practice, no participants identified interest as the relevant revenue stream, contradicting the literature (Brink, 2017b). Participants provided reasons why the rationale behind the CCRP was not to increase interest and, consequently, why interest was not identified as the relevant revenue stream in a CCRP transaction. For example, participant 4 explained: “Most banks have […] [a] 55-day interest-free period, so I’m inclined to say that interest will not be the relevant revenue stream, or a definite revenue stream for each CCRP transaction”. The CCRP accounting framework (Figure 2), therefore, does not provide for the alternative of identifying interest as the relevant revenue stream.

In a credit card arrangement, the card issuer receives an interchange fee with each cardholder purchase transaction. Specifically, a credit card arrangement functions differently from a debit card arrangement. Brink (2017c) mentioned that the interchange fee charged when using a debit card is significantly lower than when using a credit card. This might indicate that the interchange fee (that can be identified as the relevant revenue stream) not only relates to the service of electronic payment facilitation but also includes a fee representing compensation for providing credit to the cardholder (interest). If this is the case, then the interest part included in the interchange fee should be accounted for in terms of IFRS 9 (Brink and Steenkamp, 2023b). Three aspects missing from the literature are: establishing whether a part of the interchange fee in a credit card transaction represents interest; how to determine this part; and how to account for it (Brink, 2017c).

The customer in relation to the card issuer for the interchange fee will play a role in determining whether interest is included in the interchange fee. If the merchant is being provided with a service of greater access to potential customers, the merchant is identified as the customer for the interchange fee. According to Participant 11, even though the card issuer carries credit risk in a credit card transaction (compared to a debit card), the merchant paying the interchange is not compensating the card issuer for taking on the credit risk. In such a scenario the interchange fee, therefore, does not include interest. If, however, the cardholder is identified as the customer in relation to the card issuer for the interchange fee, that interchange fee might represent compensation for the service of electronically transferring the cash to the merchant (this will be the same for a debit and credit card) and the service of providing credit (the reason for the difference in the interchange fee). The service of providing credit, however, represents interest-free credit, as interest will only accumulate after a certain time period (P3; P4; P14). None of the participants (whose card issuers offer a debit and credit card) were aware that the interchange fee included interest. It was therefore decided not to include this possibility in the CCRP accounting framework (presented in Figure 2).

4.3.2 Customer in relation to card issuer.

If management conclude that the rationale behind the CCRP is to increase interchange fee income, then interchange fee income is identified as the relevant revenue stream. If this is the case, it then becomes important to identify the customer in relation to the card issuer for the interchange fee. This would determine whether the CCRP transaction falls within the scope of IFRS 15. The scope of IFRS 15 states that an entity must apply the standard to a contract only if the counterparty to the contract is a customer. The customer in relation to the card issuer will therefore have a direct impact on the accounting treatment of the CCRP transaction (Brink, 2017a; Brink and Steenkamp, 2023b).

The customer in relation to the card issuer can be determined with reference to what service is being rendered in exchange for the interchange fee and to whom (Brink and Steenkamp, 2023b), but applying this definition in practice may be troublesome (Brink, 2017a). Brink (2017a) found that the cardholder (being provided with the service of electronic payment facilitation), the merchant (being provided with the service of greater access to customers) or both the cardholder and the merchant can be identified as the customer in relation to the card issuer. As Participant 4 considered it theoretically possible to identify both the cardholder and the merchant as the customer, this alternative was initially included in the framework, but was later deleted based on the comments received from a Delphi participant (E17). The remainder of the section, therefore, focuses on identifying only the cardholder, the merchant or another party as the sole customer for the interchange fee.

Previous studies have not gathered and analysed the experiences of CCRP management and perceptions on identifying the customer in relation to the card issuer for the interchange service. Participant 3 stated:

I can see why people had debates about this […] it’s really complex. You’ve got four parties [cardholder, card issuer, acquirer bank and merchant], […] and if they weren’t each doing something for each other, they wouldn’t be in the loop …

Participant 3’s observation confirms the literature surrounding the uncertainty of whether CCRP transactions are included in the scope of IFRS 15 (Brink, 2017a; EY, 2014; PwC, 2012). From the interviews, it was clear that some participants identified the cardholder as the customer, while others did not. The decision-making processes used to reach each of these conclusions are discussed separately.

4.3.2.1 Cardholder as customer for interchange fee.

Some participants identified the cardholder as the customer, as the cardholder was provided with a service of electronically transferring cash. As Participant 3 stated: “[I]n swiping a card, … we are facilitating the transfer of their money in a safe and secure manner […] I think that’s definitely an interaction between the customer and the issuing bank”. If the cardholder is identified as the customer, the nature of the benefits, having an impact on measuring the award credits, needs to be considered next. Literature indicated that if the nature of the benefits was a cash reward, IFRS 9 would be applicable, recognising an expense and financial liability, measured at the cash amount payable representing the fair value from the cardholder’s perspective, not adjusted for the expected redemption rate (Brink and Steenkamp, 2023b). Participants 12 and 14 (managers of a complex CCRP offering direct cash-back rewards), who viewed their CCRP transaction as an integral part of a revenue transaction and offered a cash reward, agreed with this IFRS 9 accounting treatment in terms of recognition and measurement. If the nature of the benefits supplied were a non-cash reward, IFRS 15 would be applicable (Brink and Steenkamp, 2023b). In terms of the deferral model of IFRS 15, the interchange fee received must be allocated between revenue (interchange fee income) and deferred revenue (award credits) based on relative stand-alone selling prices. The allocation is based on relative stand-alone selling prices, and the stand-alone selling price of the award credits is a value from the cardholder’s perspective, adjusted for expected redemption rate (Brink and Steenkamp, 2023b). Participants 3 and 4 (managers of a simple CCRP), who viewed their CCRP transaction as an integral part of a revenue transaction and offered a non-cash reward, agreed with this IFRS 15 accounting treatment in terms of recognition and measurement.

4.3.2.2 Other parties as customer for interchange fee.

Some participants identified the merchant as the customer, as the merchant received access to customers or verification services. Participant 5 provided evidence of the increased access to customers, and emphasised that it was the merchant who is contractually bound to provide the interchange fee (cash flow received by the card issuer), by stating:

This goes back to: who do we actually have the contract with? And for me the best way to answer that is: who is liable to pay that interchange fee to STU [2]? […] which is the merchant […] So, once you get the person that pays you the money, it’s then easier to go and break down to say: what am I actually receiving this money for? And for me, because we are receiving the payment from the merchant, the service is linked with the ability to provide […] the platform to be able to make the card payments.

Participant 10 also indicated that the cardholder cannot be the customer as they are not aware of the interchange fee being earned, stating “They don’t know what they’re swiping for”. Participant 9 stated that the merchant is the customer for the interchange fee and that the service being rendered was more than merely the service of greater access to potential customers.

The service is verification, because the way the systems work with each other, merchant contacts VISA, VISA contacts us, and we confirm information like: yes, it is my customer; yes, I am the bank; yes, there is available funds; yes, you can transact. So, I almost thought like that is the service we are rendering to the merchant, is validity. And technically, I mean, I do agree, the merchant then gets greater access, because particularly in the environment we’re now in where people maybe don’t want cash, they get access to tap into that network but they need trust, and VISA itself is not the only trust in that network.

If the service being rendered includes access to customers and verification to the merchant, then literature indicates that whether the CCRP is structured in an open-loop or closed-loop structure could become relevant in determining whether the merchant could be identified as the customer in relation to the card issuer (Brink, 2017a). In a closed-loop structure, the card issuer has a contractual relationship with the merchant, but in an open-loop structure, the card issuer does not (Brink and Steenkamp, 2023b). If a contact is required to identify the merchant as customer, then in an open-loop structure the merchant cannot be identified as such (FASB and IASB, 2013). However, Participants 1, 5, 9, 10, 11 and 14 concluded that there is no requirement for a physical contract between the card issuer and the merchant to identify the merchant as the card issuer’s customer for the interchange fee. According to Participant 5, no matter whether the credit card operates in an open-loop structure or closed-loop structure, “the accounting treatment in the background is exactly the same”. Participant 9 provided the following explanation:

The problem is […] who is actually paying [for] the service because VISA is not the entity that’s paying us. The merchant is actually the one that has to pay us for the service. And, therefore, it wasn’t a case of […] we need to have a contract with the merchant. It was the fact that our contract with VISA and VISA’s contract with the merchant, is creating the contractual relationship between us and the merchant.

Participants 5 and 13 agreed. Even though there is no written contract, the contract is implied by the financial institution having a contract with the acquirer bank, and the acquirer bank having a contract with the merchant. It is not a requirement for a contract to be written; oral or implied undertakings during the entity’s customary business practices also qualify as a contract (EY, 2014). Therefore, the CCRP accounting framework makes no differentiation between identifying the merchant as customer in an open-loop and closed-loop structure (Figure 2).

Participant 7 identified the acquirer bank (VISA) as the customer for the interchange fee. For accounting purposes, there is no effect whether the merchant or the acquirer bank is identified, or not, as the customer in relation to the card issuer for the interchange fee. Both routes will result in the CCRP transaction falling outside the scope of IFRS 15 simply because the cardholder is not identified as the card issuer’s customer for the interchange fee. Based on the nature of the benefits, IAS 37 (for a non-cash reward) and IFRS 9 (for a cash reward) will be applicable. Identifying the merchant (for access to customers and verification) or the acquirer bank (for verification) as customer for the interchange fee was included in the CCRP accounting framework developed (Figure 2).

4.4 Value of award credits without observable value

Some CCRP award credits that are redeemable for non-cash rewards cannot be linked to a specific currency value (for example, when the award credits can be redeemed from a list of options, with different currency values depending on the programme partner and option). Determining the value of the award credits is the starting point to measure the award credits (as a liability) for accounting purposes (Brink and Steenkamp, 2023b), when the nature of the benefits are goods and services (and thus IFRS 15 or IAS 37 is applied). The value of award credits without an observable value could theoretically be estimated with reference to the weighted average cost price or selling price per redemption option (the estimated cost price or selling price per redemption option multiplied by the estimated redemption rate) (Brink and Steenkamp, 2023b). How CCRP management determine the value in practice has not been addressed in previous research (Brink, 2017b, 2017c; Brink and Steenkamp, 2023b; Chun et al., 2020).

There are three CCRPs operating in South Africa whose award credits do not have an observable value. Three participants (P2, P10, P11) from two such CCRPs were interviewed. The value per award credit is determined internally by these CCRPs, and a weighted average value per award credit is calculated – in line with the suggestions of Brink and Steenkamp (2023b). Participants 10 and 11 (from a simple CCRP) explained how this weighted average value per award credit is calculated:

[We use] conversion exercises – for every 1 000 points, 500 will be used for this, 400 will be used for this and the remaining 100 will be used for this. So, that’s how we then go, and we predict and say look, this would be the conversion ratio more or less, and we will then price it accordingly […] it’s a weighted average.

A weighted average redemption cost per award credit expected to be redeemed (using the cost of supplying the rewards and the cost paid to programme partners for supplying the goods or services) is therefore calculated (value from the card issuer’s perspective).

Participant 2 used the same estimation technique as Participants 10 and 11 to calculate the value per award credit internally. These values per award credit were determined separately for award credits expected to be redeemed at the co-brand CCRP and at various programme partners. For example, if an award credit is expected to be redeemed at the CCRP, the award credit is worth R1, while if an award credit is expected to be redeemed at one of the programme partners the award credit is worth R0.80 (being the value from retailer’s perspective). This value per award credit is calculated on the basis of the weighted average redemption cost per award credit redeemed and the costs of running the programme. Guidance in this regard was included in the addendum to the decision tree (Figure 3).

From the above findings it is clear that CCRP management were required to apply their judgement in accounting for complex CCRP transactions and in some cases experienced the adoption of IFRS 15 as challenging, aligning with literature (Brink and Steenkamp, 2023b; Ogunode and Salawu, 2021; PwC, 2018; Usurelu and Dutescu, 2021; Van Wyk and Coetsee, 2020).

4.5 Opinions of experts in the field

The CCRP accounting framework was well received by the Delphi experts. No contradictory views among panel members or major issues were identified. The framework was perceived and described by experts as excellent (E1), accurate and complete (E2; E7; E5; E8; E11; E12; E15; E16), relevant (E13) very comprehensive (E7; E9; E14), thorough (E5; E11), detailed (E15), insightful (E8), clear (E13), easy to follow (E7; E19) and very well laid out (E2). Sixteen panel members stated that the framework includes all the key variables of the underlying transaction to arrive at the most appropriate accounting treatment and 14 panel members indicated that the framework represents their typical decision heuristics in accounting for a CCRP transaction. Thirteen panel members agreed with the suggested accounting treatment included in the framework.

Even though the CCRP accounting framework provides a clear and concise structure for accounting for various types of CCRPs, assisting management within the decision-making process, management is still required to apply their judgement when answering some of the questions presented in the decision tree (E2; E3; E12; E17; E20) and resort to answers that may not always be optimal. The questions that specifically require the judgement of management are clearly indicated in the framework, namely (1) how the CCRP transaction is viewed and therefore distinguishing between a marketing tool versus an integral part of a revenue transaction; (2) how to apply IFRS 15 in a complex CCRP transaction. To answer these two questions, CCRP management should consult with IFRS advisory team members, auditors or other accountants.

Based on panel members’ suggestions, an introduction to the CCRP accounting framework was added (for example clarifying who the framework was developed for, linking the framework and the addendum using references, and including relevant IFRS paragraphs). To clarify that the membership fee (if identified as the relevant revenue stream) is recognised over time (E3) an explanatory note was added (Figure 3). During round two of the Delphi technique, consensus and optimal data saturation were reached as all 12 panel members were satisfied with the revised version of the CCRP accounting framework (Vernon, 2008).

4.6 Credit card rewards programme accounting framework

For application purposes, the CCRP accounting framework was embedded in a decision tree and included all possible alternatives for accounting for CCRP transactions (presented in Figure 2). To keep the decision tree dedicated to the main decision criteria in accounting for award credits, additional guidance was presented in an addendum to the decision tree (illustrated in Figure 3).

Introduction to the CCRP accounting framework:

The CCRP accounting framework can be applied by card issuers operating a CCRP (for example, a complex CCRP and a simple CCRP) and by retailers in a co-brand CCRP. The framework includes initial recognition and measurement of award credits in a CCRP transaction and excludes any other CCRP-related transactions (for example de-recognition of award credits). The CCRP accounting framework provides a structure for accounting for various types of CCRP transactions but still requires management to apply their professional unbiased judgement based on their knowledge of the structure and functioning of the CCRP (the questions that specifically require the judgement of management are clearly indicated in the framework). The CCRP accounting framework incorporates what was evident in literature and information from interviews with the managers of eight South African CCRPs. However, it is possible that there are scenarios which were not covered in the framework as CCRPs are unique and vary in their structure and functioning. The CCRP accounting framework consists of a decision tree as well as an addendum. The decision tree contains references (indicated by numbers 1–5 in superscript) to the addendum, which provides additional guidance on applying specific components of the decision tree.

5. Conclusion

The aim and contribution of this study was to develop a framework for CCRP accounting after the effective date of IFRS 15. Brink and Steenkamp’s (2023b) theoretical model was enhanced by conducting and analysing 15 semi-structured interviews with CCRP managers. The resultant framework was validated using two rounds of the Delphi technique and provides a clear graphical representation of the key decision-making processes that are required from CCRP management when accounting for CCRP transactions. Questions that specifically require the judgement of management were clearly indicated in the framework. To answer these two questions, CCRP management should consult with IFRS advisory team members, auditors or other accountants. This paper emphasised the extent to which additional guidance is required to adequately account for CCRP transactions after the effective date of IFRS 15.

This study revealed that clarity on the most appropriate accounting treatment for CCRP transactions cannot be gained by only considering literature (Brink and Steenkamp, 2023b). Instead, the findings emphasise that insights from CCRP management experiences and expert opinions are crucial for determining the optimal accounting approach for these transactions. Parts of the model developed by Brink and Steenkamp (2023b) were found to be valid, but this study updated the model for what was evident in practice. It was found that interest cannot be identified as a relevant revenue stream in a CCRP transaction and that the interchange fee does not include a component of interest. For purposes of a CCRP transaction no differentiation needs to be made between an open-loop and closed-loop structure. For the first time clear guidance on how the CCRP transaction should be viewed and how to determine the value of award credits without an observable value were provided.

CCRPs could apply the CCRP accounting model to determine the appropriate accounting treatment (recognition and measurement) of CCRP transactions, to eliminate inconsistencies in practice and ensure faithfully representation of CCRP transactions. Another contribution lies within the practice side of implementing new standards in niche areas of business where the outcomes or consequences might be different from that which was envisioned in the standard. Anomalous results always lead to information that is important for standard-setters and practitioners. The paper raises some important questions for standard setters and their approach to standard-setting to consider as part of a post implementation review of IFRS 15, namely: Do standard setters accept the diversity in accounting practices evident in CCRP transactions, given that diversity in accounting practices is inconsistent with comparability? If not, should additional guidance be provided to narrow the interpretation of accounting standards and constrain potential diversity in accounting practices.

The scope of this study included only the initial recognition and measurement of award credits at the date that the award credits are granted. Future studies could consider the derecognition of award credits when they are redeemed. A further area for future research is developing illustrative examples and case studies for various types of CCRPs.

Figures

Figure 1.

CCRP Theoretical model

Figure 2.

CCRP accounting framework: decision tree on accounting for CCRP transactions

Figure 3.

Addendum to the decision tree on accounting for CCRP transactions

Table A1.

Summary of interviews conducted with CCRP managers

		Number of people interviewed who were:
Interview	Participant number	Accountants	Heads of programme	IFRS advisory team members
Interview 1	P1*	3	1
Interview 2	P2		1
Interview 3	P3		1
Interview 4	P4	1
Interview 5	P6		1
Interview 6	P8		1
Interview 7	P5			1
Interview 8	P7*	2
Interview 9	P9			1
Interview 10 (follow-up)	P9
Interview 11	P12		1
Interview 12	P10*	1	1
Interview 13	P11			1
Interview 14	P13			1
Interview 15	P14	1
Totals	14	8	7	4

Note:

* Even though more than one person was interviewed, a single participant number represents all parties that were interviewed

Source: Created by authors

Table A2.

Summary of participants of the Delphi survey

	Number of participants
Expert number	Accountant	Head of the programme	IFRS advisory team	Audit partner
E1, E22		1
E2			1
E3				1
E4, E24	1
E5, E29		1
E6				1
E7, E30	1
E8, E33	1
P9, E32	1
E10		1
E11, E23		1
E12				1
P13		1
E14			1
E15, E25				1a
E16			1
E17		1
E18		1
E19, E28	1
E20, E26				1
E21, E27				1
E31	1
Total	6	7	3	6

Notes:

^aThe audit partner was not available to participate, but the associate director involved in a CCRP audit was; In total 21 experts participated in round one of the Delphi technique (referred to as E1–E21) and 12 experts (referred to as E22–E33) in round two. Eleven of the experts that participated in round one also participated in round two

Source: Created by authors

Notes

1. Purposive sampling is “associated with small, in-depth studies, with research designs that are based on the gathering of qualitative data and focused on the exploration and interpretation of experiences and perceptions” (Matthews and Ross, 2010) and was therefore considered as an appropriate technique for selecting the CCRPs for participation.

2. For purposes of confidentiality the financial institution’s name was removed in the quotation.

3. Various probes were used during the interviews to obtain clarification for answers and to elicit valuable and complete information.

Appendix 1

Table A1 includes the participant number for each interview and details about the CCRP managers who were interviewed. This was the only demographic information deemed important and obtained from the participants.

Appendix 2. Interview guide

Questions [3]

Accounting treatment introductory question

1. Can you tell me about your experience and engagement with the accounting treatment of CCRP transactions?

2. Can you tell me a little more about the decision-making process you followed to account for CCRP transactions post IFRS 15?

3. Can we for a moment discuss whether you experienced any changes in accounting for CCRP transactions after the effective date of IFRS 15 and the impact of these changes?

4. How do you view the CCRP transaction – is it regarded as a marketing tool or is it seen as part of an integrated revenue transaction?

Relevant revenue stream.

5. What do you believe is the rationale behind the CCRP?

6. Do you believe that the rationale of a CCRP should be considered when determining the relevant revenue stream and why?

Customer in relation to the card issuer.

7. In your opinion what service is being rendered in exchange for the interchange fee and to whom?

8. Can you tell me a little more about the decision-making process you used to determine the customer in the CCRP transaction?

Nature of the benefits supplied.

9. Can we for a moment discuss the nature of the benefits supplied (cash vs non-cash rewards) - Do you view the award credits as a cash or a non-cash reward?

Measurement of the award credits.

10. Can we discuss the measurement of the award credits and how this value is determined?

Value of award credits.

11. What are your perspectives on the process of determining the value of award credits without an observable value?

Concluding question.

12. Can you think of anything else that would be important or that we have missed in deciding how to account for CCRP transactions?

Appendix 3

Table A2 provides details about the experts that participated in the Delphi technique.