Submission to Treasury: Regulating Buy Now, Pay Later in Australia
Kadre welcomes the opportunity to provide this feedback to Treasury on the future regulatory framework for buy now, pay later (BNPL) arrangements under the National Consumer Credit Protection Act 2009 (the Credit Act).
Kadre is a community of risk, retail credit, compliance, and data science specialists that provides advisory and consulting services to the financial services industry. The Kadre community comprises former senior executives of many large institutions, fintechs and service providers in the consumer lending sector. Many of these executives were at the forefront of the development and implementation of major initiatives such as Comprehensive Credit Reporting and Responsible Lending Obligations. With this knowledge and experience, Kadre is well-placed to offer practical and well-informed feedback on critical areas highlighted in the Treasury Consultation paper.
In particular, our expertise lies in addressing the following regulatory issues defined in the Options Paper:
- Unaffordable or inappropriate lending practices
- Non-participation in Australia’s credit reporting framework
In response to the questions posed in the Options Paper:
Are the guiding principles appropriate and fit for purpose to inform the development of a BNPL regulatory framework? What other factors should be considered?
Kadre considers the guiding principles are appropriate and fit for purpose. However, in our opinion, the guiding principles should not single out BNPL, but rather should establish a consistent and less complicated means of regulating responsible provision of credit for other existing, new and future credit products
Of the three options below, which option do you think is most appropriate? Would you change any aspects of that option?
Kadre supports Option 3 for the reasons described in this paper.
What do you think are the issues with the other two options?
We have described our view of the issues with the other two options proposed in the Options Paper. In our opinion, the existing requirements in the Credit Act are sufficient to apply to all lenders providing there is a clarity about the definition of scalability and that all lenders can access the full extent of a consumer’s indebtedness through the credit reporting regime.
Require BNPL providers to hold an Australian Credit License
Option 1 described by Treasury involves strengthening the BNPL Industry Code and adding an affordability test which imposes a bespoke affordability assessment for BNPL providers. As noted in the Options Paper, BNPL is not regulated under the National Consumer Credit Protection Act 2009 (the Credit Act), because it typically falls under the exemptions available to certain types of credit in Schedule 1 of the Credit Act.
Having dealt with many Fintechs and lending start-ups in Australia, Kadre is of the view that BNPL can operate under the existing Credit Act as a licensed entity providing there is clarity about the definition of scalability in relation to assessing a customer’s financial situation and that all lenders can obtain reliable information about the level of a consumer’s indebtedness.
In our view, accepting the BNPL Industry Code and allowing an exemption to the Credit Act will set a precedent for continuous regulatory arbitrage in cases relating to innovative credit products. The nature of self-regulated Industry Code means that BNPL, and potential future innovative credit providers, would not be subject to the same level of public scrutiny, monitoring and enforcement as other credit providers that operate under the Credit Act.
In our experience, the principles of the Australian Credit License process are sound and reflect society’s expectation of providers of credit in areas such as:
- Fit and proper person to engage in credit activities.
- Adequate compliance arrangements.
- Supervising and training representatives.
- Adequacy of resources.
- Dispute resolution and hardship processes.
- Sound risk management practices.
- Appropriate and defined compensation arrangements.
An issue that does need to be addressed is the time required to process an ACL application. Working with many start-ups, we have found this to be a challenge for fintechs seeking to enter the market with innovative ideas related to provision of credit. In our view, the government should consider rectifying this situation with an appropriate flexible resourcing mechanism for ASIC in this area.
Option 2 described by Treasury proposes to bring BNPL within the Credit Act with a tailored version of Responsible Lending Obligations. In our view, adding more “carve outs” like those for SACC, MACC and Reverse Mortgage loans will further complicate regulations for existing lenders and create additional uncertainty for credit providers that are considering new and innovative products. Option 2 would set a precedent for each new market entrant to seek ‘their own rules’ claiming their product is ‘unique’ and should be treated differently based on a technicality, such as in the case of BNPL, rather than on the core premise of the product.
We note and agree with the Australian Law Review Commission (ALRC) statement in its Interim report[1]:
“13. Many stakeholders have identified the navigability of the law as a key concern — it is too difficult to locate relevant parts of the law, and even experienced lawyers cannot always be confident that they are taking into account all relevant provisions and instruments on a particular issue without ‘missing something’.” The Credit Act and Regulatory Guidance (RG209) already contains provision for dealing with variations on Responsible Lending Obligations. For example, Regulatory Guidance (RG209.81) states
that “The process of determining the kinds of inquiries and steps that are reasonable has been described as ‘scalability’. You need to apply your own judgment in determining what is reasonable in the individual circumstances”.
In our view, BNPL products are credit and should be regulated consistently with other credit providers within the existing Credit Act. We feel it is incongruous that a $1,000 BNPL credit line requires no credit assessment while a $1,000 overdraft or credit card, which can have a materially lower impact on the cash flow of a consumer, requires substantial effort by the lender to assess the customer’s financial position. Clearly defining scalability in relation to reasonable inquiries about, and verification of, the customer’s financial position will provide a strong framework for ensuring lenders balance a low cost, efficient and streamlined process against avoiding reasonably foreseeable consumer financial harm.
Clarify what is meant by scalability
We support Option 3 proposed by Treasury that BNPL be required to hold an Australian Credit License and be regulated under the Credit Act. However, our support is dependent upon there being a clearer definition of scalability for the benefit of all lenders.
Even before it was introduced, the definition of “scalability” has been a point of contention among consumer lenders. When introduced for consultation in 2010, there was little definitive explanation of scalability leaving each lender to interpret thresholds for reasonable inquiries about, and verification of, the customer’s financial position. Competitive pressures and product economics led to a wide range of interpretations, resulting in an equally inconsistent level of consumer protection.
Many lenders implemented Income Validation Models and Expense Benchmarks for credit facilities that had combinations of low value and low risk so they could maintain efficient, low cost and streamlined processes. Over the subsequent few years, ASIC took action against a number of lenders for failing to comply with responsible lending obligations and additional requirements were added to regulations for SACC and MACC loans. The perceived compliance requirements and threat of regulator action meant that many lenders took a conservative approach to credit assessment resulting in more complex processes, greater customer friction and higher costs.
The scalability interpretation was further muddied by the seminal “Wagyu and Shiraz” case[1]. Westpac took a view that scalability could be applied using a risk measurement algorithm such that customers with good credit scores could have their financial situation assessed using a streamlined approach including the use of the HEM expense benchmark. While these findings from the case led to changes to the regulatory guidance, the issue of scalability remains open to interpretation.
These historical events illustrate the challengers with an “open ended” approach to scalability interpretation. Kadre has worked with many consumer lenders across the industry and has a deep knowledge of credit policies and credit assessment processes including the way in which they assess and verify a customer’s financial position. We are currently working with organizations such as the Sydney University Business School to determine appropriate empirical methods for defining scalability. Our base hypothesis is that customer vulnerability or likelihood of hardship is a function of their credit history, as reflected in a credit score, and their financial capacity. In simple terms, capacity is defined by the equation of net income, less expenses, less financial liabilities. Customers with high capacity are less likely to suffer financial hardship than those with low, or negative, capacity.
In our view, the degree of capacity can be used to determine the level of assessment needed into the customer’s financial position.
At one end of the spectrum, where the implications to the customer’s cash flow would be low, an efficient process can involve use of Income Validation Models and Expense Benchmarks together with liability information from Comprehensive Credit Reporting. At the other end, where the implications on customer cash flow are expected to be high, a more in-depth process requiring detailed analysis of income, expenses and liabilities would be more appropriate as described in the ASIC Regulatory Guide RG209.
For example, a customer with $100,000 income may only have a small surplus capacity if they are highly geared with multiple loans, while a customer with $50,000 income can have a much higher surplus capacity if they have no other loans. Similarly, even small BNPL debts can be burdensome to those on low incomes and the current “one size fits all” approach by offering the same limit to all new customers can lead to poor customer outcomes.
Accurate and reliable information about a consumer’s income is crucial in determining the level of assessment that should be undertaken. One of the challenges continually faced by lenders is the difficulty in verifying a customer’s income. Most methods are complex and unreliable with solutions including pay slips, bank statement transactions, superannuation data and other triangulation techniques. In our view, solving the income issue should be considered in the context of clarifying scalability and we suggest Treasury explore whether there is some way of accessing income data through a reliable government source such as Single Touch Payroll (STP).
In our engagements with lenders, we are constantly asked how the lender should assess a consumer’s liability in relation to a BNPL facility. Most BNPL accounts have a predetermined limit for aggregate debt a customer can access. Our view is that this limit could be used to determine a hypothetical instalment amount in a similar manner to the way that ASIC approached the assessment of Credit Card repayments in its CP303. In CP303, ASIC proposed that responsible lending assessments be based on the consumer’s ability to repay the credit limit within three years. A similar approach could be used by assuming a BNPL limit of up to say $3,000 be repaid over a period of twelve months. This would mean that a customer with a $1,200 BNPL limit would need to demonstrate that they have sufficient capacity to repay $100 per month. Scalability principles should apply to the full breadth of credit providers, not just BNPL. In our view, the existing Responsible Lending Obligations in the Credit Act are sufficient to apply to all lenders providing that the definition of scalability is made clear. It should provide consistency and certainty as to what is, and is not, acceptable, without the risk of impeding innovation.
Mandate contribution of Partial Data for Credit Reporting
Our suggested methodology for scalability and assessing financial capacity is dependent on having a full picture of a customer’s liabilities. This would require a more robust Comprehensive Credit Reporting (CCR) regime.
Many of the Kadre community were instrumental in the development and implementation of Comprehensive Credit Reporting. Most are disappointed that the many and varied compromises made during the legislative change process have significantly undermined the original benefits expected for lenders and consumers. Crucial to the success of the Comprehensive Credit Reporting regime was addressing the issue of information asymmetry. In original submissions on Credit Reporting to the Australian Law Review Commission (ALRC) around 2010, many of us made the point that a key causal factor in consumer financial harm is the lack of data available to the lender to assess a customer’s financial situation. Under the current
Comprehensive Credit Reporting regime, information asymmetry continues to exist because of issues including:
- Reporting credit information is optional for all credit providers except for the major banks (defined as large ADIs).
- Obtaining a complete picture of the customer requires access to all three Credit Reporting Bodies (CRBs).
- Information about repayment amount needs to be derived rather than given.
- True indebtedness is not available because only the credit limit is reported and not the actual balance.
While we will submit these issues as part of the planned review of the credit reporting framework scheduled for October 2024, in our opinion, some of these issues are relevant to the current debate about BNPL arrangements under the Credit Act and could be addressed immediately.
While some BNPL providers participate in CCR, there are others that do not. In addition, there are still many credit providers that do not provide their data to the CCR repository held by CRBs. Details of which lenders are providing data to the CCR repository is not transparent so it is not unusual for a lender to believe they have identified all of a consumer’s credit facilities when in fact, there can still be undisclosed and/or indeterminable debts. Some lenders have reverted to using bank statements in an attempt to identify loans that are not listed at the CRB, however this can be highly complex and of limited reliability even with the Consumer Data Right and Open Banking.
The CCR regime has been structured so that lenders can choose to provide and consume data at one of three levels as defined within the Principles of Reciprocity and Data Exchange (PRDE):
- Partial Reporting – credit liabilities including open accounts, limits and accounts closed (described as CCLI).
- Full Reporting – CCLI plus 24 months of Repayment History Information (RHI).
- Negative Only Reporting – credit enquiries, defaults, bankruptcies and court judgements.
Kadre has worked with many small lenders who have found that implementation of Full Reporting of CCR (including RHI) is difficult and expensive. As a result, a number of ADIs and other credit providers are yet to supply some, or all of their CCR data to a CRB. The issue is exacerbated by the inconsistent provision of data by the BNPL sector.
In our experience, the difficulties for small lenders in providing Full Reporting of CCR are valid but we are of the view that provision of data at a Partial Reporting level should be possible for most lenders. We suggest that Treasury consider the concept of mandating that all ACL holders provide Partial Reporting of CCR. This would go a long way to helping lenders to understand a customer’s level of indebtedness and therefore better understand a customer’s financial position.
A secondary issue with CCR involves the ability of lenders to access the data for use in making an assessment of the customer’s financial position. While the data of the major banks is required to be reported to CRBs with which they have an agreement (typically all three CRBs), other lenders can choose to provide their data to any number of the three CRBs. Most smaller lenders only provide CCR data to a single CRB. As a consequence, a lender needs to enquire on all three CRBs to obtain the most complete view of the customer’s liabilities. Due to the costs associated with obtaining credit reports, compounded in the case of obtaining multiple credit reports, smaller lenders generally only access a single CRB for CCR data meaning that they may not obtain the most complete information about a customer’s credit facilities if this data is spread across multiple CRBs. Larger lenders such as the major banks are able to negotiate differential (lower) pricing with the CRBs based on volume discounts which means that small lenders
are at a cost disadvantage if they were to access multiple CRBs to support their Responsible Lending Obligations.
It is essential for responsible lending purposes and consumer protection that all lenders be able to access at least Partial Reporting of CCR. Therefore, as part of mandating provision of Partial Reporting, we suggest one of two options for pricing of credit reports, either:
- Partial Reporting of CCR by credit providers be reported to all CRBs who can then price competitively, or
- Price parity be mandated across all CRBs for access to at least Partial Reporting of CCR so all lenders obtain at least that level of data at the same price
Due to the way in which BNPL providers deal with customers who do not meet their repayment obligations, we are of the opinion that Partial Reporting of CCR data should be supplemented with an indicator, similar to the existing account default flag, so that a lender can understand whether a terminated facility was closed by the customer or closed by the provider as a result of non-payment (these could be referred to as “closed by customer” versus “closed by provider”). This is particularly important for BNPL facilities so lenders can identify potentially vulnerable customers whose facilities were closed by the BNPL provider. We suggest that mandating Partial Reporting of CCR data across all credit providers including BNPL be considered as an interim step to improve the CCR regime pending review of the credit reporting framework scheduled for October 2024.
Conclusion
Our Kadre community believes that the Government has a great opportunity to address the increasing potential of consumer financial harm and reduce the cost of credit while continuing to promote innovation that will emerge from technology advancements. In our opinion, the existing requirements in the Credit Act are sufficient to apply to all lenders providing there is a clarity about the definition of scalability and that all lenders can access the full extent of a consumer’s indebtedness through the credit reporting regime.
The Kadre Community
The Kadre community includes former CROs of major banks and other financial institutions including fintechs, former senior executives of Credit Reporting Bodies, founding members and former directors of ARCA and subject matter experts who have been involved in ASIC investigations, helping lenders deal with enforceable undertakings and remediation programs. The community also includes leading data scientists who have a deep understanding of how to use data to achieve successful business outcomes.
Kadre has provided this submission as an independent group seeking only to protect consumers from financial harm by creating a fair credit regime that is in the best interests of the Australian economy.
Contributors to this paper from the Kadre community include Steve Johnson, David Grafton, Tim Brinkler, Paul Jewell, Paul Abbey, Michael Hartman, Thushare Dissanayake, Zana Glavanic, Ewa Baranowska and Sebastian Lopez.
The contact point for Kadre is Steve Johnson (steve@kadre.com.au M: 0419368326)
