The BNPL business model has so far shown at least the potential to be profitable – depending on the market in which it is offered and the strategy chosen by the provider. This is particularly true if BNPL is part of a broader product range. However, it is unclear how sustainable the BNPL business model is, as well as how it will hold up in the changed macroeconomic environment we are now in, with higher interest rates and increased competition. The analysis shows that the main risks of the BNPL providers’ business models come from two sources: credit risks taken by providers and the large technology investments required. Together, they limit profitability.

Profitable business model in Sweden

The BNPL providers we interviewed state that the business models for BNPL and the checkout solution are closely linked and fundamentally profitable in the Swedish market.^[31] The profitability of the business model is difficult to verify through publicly available data for the Swedish providers as they also have other operations in addition to BNPL and do not report the results for BNPL separately. They report that several factors combine to make this the case in both the Swedish market and a number of other Nordic markets. First, the infrastructure around BNPL is different in Sweden compared to other countries. BNPL itself is highly standardised and consumers are also somewhat accustomed to paying with BNPL at physical retailers. It is also relatively easy for online retailers to access public information about consumers and to identify them via their personal identity numbers. In addition, there is an infrastructure of debt collection agencies that can help them recover any debts incurred.

Several providers we have interviewed say that it is more difficult to achieve profitability with the business model as soon as you expand beyond Sweden, Norway and Finland because the conditions are different in other countries.

These factors have made it possible to automate processes and simplify credit assessments, reducing both costs and credit risks. Moreover, the widespread use of BankID in e-commerce means that the rate of fraud is lower in the Swedish market than in other markets without similar identification solutions.^[32] See, for instance, Sveriges Riksbank, Payments Report 2022: Are payments in Sweden safe? (2022).

Second, both BNPL services themselves and the providers’ overall activities are generally broader in Sweden than outside Sweden – at least when compared to global BNPL providers. In Sweden, BNPL services usually include both BNPL and the checkout solution as a whole, while outside Sweden it is more common for operators to specialise in only one of these products. This makes it possible for the Swedish providers to handle a larger share of an online retailer’s payments and to sell more different products to the online retailer. In addition, most Swedish BNPL providers also have other activities, such as corporate lending, property loans and credit cards, which means that they are less dependent on BNPL as a single product.

Being able to sell multiple products to online retailers has become particularly important as Swish has grown rapidly as a popular payment option in e-commerce. Some online retailers we interviewed said that more than half of their e-commerce payments are now made via Swish, making it important for BNPL providers to offer Swish as well in order to stay relevant. In addition to adding new payment options, the checkout solution is also being developed on an ongoing basis, which involves additional costs, such as integrating and updating the delivery selector so that the consumer can instantly see the price, days to delivery and pick-up points and similar information about the delivery to make the checkout experience even smoother.

Credit risks are significant and have increased in recent years

A key aspect of the BNPL business model is credit risk. BNPL providers’ overall loan loss level was around 4 per cent of their total lending in 2022.^[33] These figures refer to the entire business of BNPL providers, which, in most cases, also includes business activities other than BNPL and checkout solutions. According to annual reports for 2022, between 15 and 85 per cent of the providers’ revenues come from the Payments or Commerce segments. ^[34] Loan loss rate is defined as net loan losses divided by total loans. This can be compared with the credit loss level for the major Swedish banks amounting to a tenth of a per cent in the same year.^[35] The major Swedish banks are Handelsbanken, SEB and Swedbank. BNPL providers therefore take more credit risk than the major Swedish banks and their level of risk has also increased over the last five years (see Diagram 80). Moreover, they have a significantly higher share of non-performing exposures compared to the major Swedish banks.

Figure 8. The level of loan losses is high and has increased in recent years, while the share of non-performing exposures has remained more constant Left: Loan losses, net/Total loans, per cent. Right: Stage 3 loans/Total loans to the public, per cent

Credit risks come mainly from consumers

Credit risks for BNPL providers come from two main channels. The first and largest is consumer credit risk. Consumer loan losses can occur if a consumer buys a product using BNPL and the online retailer delivers it according to contract, but the consumer then fails to pay. Based on an analysis by Finansinspektionen, around one in ten invoices are not paid on time (see Diagram 9).^[36] See Finansinspektionen, Swedish Consumer Credit 2022 (2022). This means that BNPL providers can make money from arrangement fees, reminder fees and the like, but that they have a higher risk of loan losses.

Figure 9. Percentage of invoices that are converted or for which a payment reminder is issued Per cent

Consumer credit risk is a competitive balancing act for providers. The more information providers ask for at checkout, the easier it is to evaluate consumers’ creditworthiness. This is especially true for new consumers. At the same time, the more information providers ask for at checkout, the more likely consumers are to abandon their shopping baskets, as it takes longer to complete the purchase. Moreover, the tighter credit granting gets, the fewer sales there will be for the online retailer. A certain trade-off thus has to be made, considering that providers sell BNPL to online retailers on the basis that it will increase their sales. Here, it is also possible that established providers with proven risk models, licences and the like have a certain advantage over newer providers who have to build all this from scratch.

Based on an analysis by Finansinspektionen, BNPL users under the age of 25 and those with a lower income (less than SEK 8,000 per month before tax) are the most likely to fail to pay on time, and they are also more likely to convert their invoices into loans.^[37] See Finansinspektionen, Swedish Consumer Credit 2022 (2022).

Given the short maturities, usually less than one month, providers state that, if necessary, they could adjust their lending behaviour and very quickly make an impact on the credit risk in their loan portfolios. But even if this was technically possible in practice, it is unclear what effect it would have on profitability. More restrictive lending by BNPL providers to consumers would imply a less favourable service offering to online retailers, which could hamper their sales. It is therefore not inconceivable that online retailers would look for other providers who have not tightened their lending in the same way and therefore have a more attractive service offering. This could lead to a negative spiral for the BNPL provider, in which it tightens up to reduce credit risks but, at the same time, reduces its opportunities for long-term growth and profitability if it starts losing online retailers as customers.

BNPL providers are more protected against risks from online retailers

In addition to credit risks from consumers, BNPL providers also have some exposure to credit risks from the online retailers themselves, as they take over the credit risk of the online retailer towards its customer. This means that the BNPL provider may incur loan losses if the online retailer does not deliver the goods to the consumer even though the BNPL operator has already paid the online retailer for the goods purchased. This risk is more significant in sectors where many goods are returned, such as the clothing sector. However, BNPL providers themselves consider this to be a minor risk factor which is also relatively easy to protect themselves against.

BNPL providers usually protect themselves against credit risk from online retailers by screening new online retailers from a credit risk perspective before signing contracts with them. This process can take several weeks depending on the size of the online retailer, its history and the sector in which it operates. They then monitor credit risk on an ongoing basis. To some extent, BNPL providers can also protect themselves against credit risk from the online retailer by applying delayed payments, so that the online retailer is paid for the consumer’s goods after a delay of a few days or when the goods have been dispatched instead of directly at the time of purchase.

Because BNPL providers are able to see online retailers’ payment flows when they take charge of the entire checkout solution, they consider it to be relatively easy for them to recognise if an online retailer is getting into financial trouble. It is therefore unusual for an online retailer to go bankrupt without the BNPL providers having had time to take some form of action. However, it seems particularly important for BNPL providers to monitor this now that the number of retail bankruptcies is increasing.

The number of bankruptcies in the retail sector increased by 34 per cent in the first half of 2023, compared to the previous year.^[38] UC, Konkursstatistik [Bankruptcy statistics], June 2023.

Significant economies of scale motivate operators to focus on growth

Offering a BNPL and checkout solution requires significant fixed investments both initially and on an ongoing basis, as it is a relatively technically complex product. This means at least two things. First, there are significant barriers to entry for new players wishing to enter the market, both in terms of time and money, given that the development of solutions normally takes several years.

Second, it is advantageous for BNPL providers to have as large a customer base as possible to offset their high fixed costs. They therefore have strong incentives to pursue growth strategies which may sacrifice profitability at least in the short term, for example, through lower fees from online retailers in order to bring more online retailers into their portfolio. In the medium term, there is a risk from a financial stability perspective that BNPL providers may prioritise growth over profitability, reducing their financial resilience. These growth strategies also contribute to increased competitive pressures and, in some cases, to an ongoing need for various forms of additional capitalisation.

In terms of competition, BNPL providers appear largely to use pricing to attract new online retailers. However, they themselves consider that other aspects of their service offerings are just as important. For example, this may include the level of customer service they offer to the consumer, the visibility of the BNPL provider to the consumer, the percentage of invoices that go to debt collection or the ancillary services they offer. It is also not entirely uncommon for different BNPL providers to need to cooperate to a degree, for example if a customer wants one operator’s checkout solution but another’s invoice purchase solution in the same checkout system.

Our interviewees find it difficult to assess competition from outside Sweden, such as Apple Pay Later, Mastercard Installments and PayPal.

While competition has increased in recent years, there are also various lock-in effects that, to some extent, hamper competition by reducing online retailers’ incentives to switch providers once they have entered into a contract. First, it is very costly for online retailers to change checkout providers when it comes to full integration, as opposed to the simpler fallback solutions that can be used in the short term that we discussed earlier. Depending on how the online retailer chooses to integrate with the BNPL provider, the process can take anywhere from a few days to several months. In the typical case, however, it takes at least two to three months and thus entails significant costs for the e-commerce retailer – especially since it is common to have relatively low margins in e-commerce.^[39] The effective working time can be around 2–3 weeks, but the projects often take longer when other priorities are higher or because it is difficult to find time with key employees. Online retailers are therefore normally reluctant to change supplier unless they have strong incentives. Second, contracts between providers and online retailers are often locked for at least one year, but often two to three years

A tougher macroeconomic situation will test the business model

It is likely that BNPL as a business model will be tested in the near future due to the deteriorating macroeconomic environment that we now find ourselves in, and that looks likely to hit both commerce in general and e-commerce hard. For e-commerce, this comes after a period of strong growth as mentioned earlier. Retail trade is expected to decline by 4.7 per cent measured at constant prices in 2023, according to the June forecast by the National Institute of Economic Research, and e-commerce turnover was already down in 2022 compared with the previous year, which was boosted by the transition during the pandemic into e-commerce.^[40] Based on households’ consumption expenditure in retail trade according to forecast published in June 2023. ^[41] Se Dagens Handel, Dyster prognos för 2023 – tillväxt drivs av prisökningar[Gloomy outlook for 2023 - growth driven by price increases], March 2023. In, Diagram 10 you can see that some segments are hit harder than others, and therefore BNPL providers with higher exposure to these segments should reasonably experience a greater deterioration in payment volumes if they do not take measures.

Figure 10. Growth of the e-commerce segments Growth rate in different e-commerce segments on an annual basis, per cent

BNPL providers generally expect reduced e-commerce transaction volumes, which will challenge their profitability. To some extent, it will therefore be necessary for BNPL providers to increase market share in order to maintain or improve profitability. The resulting increase in competition may put downward pressure on fees charged to online retailers and increase marketing costs In addition, BNPL providers’ financing costs are rising, meaning that they need to increase their prices to consumers in their offers in order to improve profitability.

At the same time, the tougher situation for online retailers means that BNPL providers expect more online retailers to try to renegotiate contracts to reduce their costs for the checkout solution. In addition, there are other payment alternatives today that may be more attractive to online retailers because they are cheaper than BNPL at the same time as they are popular among consumers. It is therefore not unthinkable that margins will fall in the future, depending on what the BNPL providers and online retailers agree on.