Cash savings profitability analysis

In July 2023 we published the Cash Savings Market Review 2023. The paper provided an overview of the current state of the cash savings market, focusing on whether savers were benefiting from rising interest rates and if competition was working effectively.

Since then, we have conducted in-depth analysis of the profits made on savings and their contribution to overall firm profitability. Profitability is key to firms being able to operate effectively and provide services to consumers.

Our analysis of firms’ profitability also forms an important part of our assessment of competitive conditions. When competition is effective, we would expect returns to be gradually driven down towards the minimum level required by investors. Where profits above this level are a persistent feature of a market, it could indicate low levels of competitive pressure.

We have compared profitability and other relevant measures between business model types and product types, as well as changes over time, particularly in light of changes to the UK base rate.

Our results should be considered relative to the period selected in the economic cycle. The period captured in our analysis (October 2021 to September 2023) could be characterised as a peak in financial results in this economic cycle. They are not necessarily indicative of long-term sustained results.

Although 2023 annual results for several firms show higher overall profitability compared with 2022, many informed us that monthly results as of late 2023 and their outlook for 2024 show a further decline in savings net interest margins (NIMs).

Financial performance in terms of net interest margins across 14 banks and 3 building societies improved from the point the base rate started to rise in December 2021 until the end of the period we have looked at (September 2023), with a rising contribution from savings.

Profitability can be measured for savings products by considering funds transfer pricing (FTP), an internal process through which many firms allocate income to savings product lines at a granular level. While FTP income does not signify external revenue, it represents the economic value of savings. A comparison of FTP income with interest expenses has allowed us to quantify cash savings profitability.

Savings net interest margins (NIMs) of our sample of firms increased rapidly between October 2021 (0.58%) and March 2023 (1.78%), before flattening and then declining slightly between July and September 2023 to 1.56%. This was echoed by increases in overall firm NIMs, which grew from 2.07% in 2021 to 2.76% in 2023.

This finding reiterates the analysis in the Cash Savings Market Review, where we noted higher levels of profitability in 2022 to 2023, ie, since the base rate began to rise, compared with pre-Covid levels.

Changes in savings NIM were not uniform across our sample of firms. We saw a reduction in the gap in NIM between the big 4, scale challengers and mid-tier firms, suggesting improvements in competition.

Firms’ savings divisions saw faster rises in FTP income than interest expenses for all products. FTP income adjusted to base rate increases faster than pricing as it refers to external benchmarks such as Sterling Overnight Index Average (SONIA) and changes daily. Increases in interest expense lagged compared to the base rate, as firms have more autonomy over rates offered to customers and decisions are subject to market conditions, competition, and internal requirements.

Profit before tax (PBT, which includes more costs than NIM, such as direct costs and allocated operating costs) margins for savings followed the trend in savings NIM. Despite the mirrored trends with savings NIM, savings PBT margins were low, peaking at 0.12% in March 2023. We found that non-interest costs (such as operational costs) did not rise in line with growth in savings balances or numbers.

There are signs that NIMs have peaked, indicated by the flattening and slight decline in savings NIM from July to September 2023. This trend was driven by interest expenses rising more quickly than FTP income from March to September 2023, as firms raised their savings rates.

This was potentially in response to commercial pressures such as increased price competition and regulatory pressure. Additionally, higher consumer engagement with the market as consumers looked to switch to higher-rate accounts, is likely to have contributed to a shift in the mix of products. There was increased uptake in term accounts, which generate lower NIMs, compared to easy access accounts, which generate the highest of all savings NIMs.

With interest expenses forming an increasing proportion of FTP income from April 2022 to September 2023, there are signs that the benefits firms earn from FTP have been increasingly passed through to consumers.

Type of data gathered

We have analysed data to identify the extent to which firms benefit financially from personal cash savings products. We have used data ranging from published financial statements to product-level data submitted to us by firms through a request for information (RFI).

The analysis also incorporates 2015 to 2020 data from the Strategic Review of Retail Banking Business Models and information from the CMA Retail Banking Market Investigation (2016).

Firms also submitted qualitative responses where they explained how they consider the value and profitability of cash savings, provided additional context to their financial results, and gave forward-looking outlooks.

Data sufficiency

Our request for information (RFI) covered October 2021 to September 2023. We recognise that this period was a time of macroeconomic change. Given this, we have used historical data from the CMA’s Retail Banking Market Investigation and the FCA’s Strategic Review of Retail Banking Business Models to look at changes in overall NIMs across a longer period.

It is important to note that the firms included in the FCA’s Strategic Review of Retail Banking Business Models report are not identical to the firms included in the cash savings RFI.

Our RFI was on a monthly basis, therefore any whole-year figures we refer to in this analysis represent annualised income statement figures (income and costs) for 2021 and 2023, and 12 months of RFI data for 2022. Average figures for the year in question have been used for balances and account numbers.

The depth of information received through the RFI varied in quality. This has forced us to impose limitations on which firms could be included for the various types of analysis that we conducted.

All analysis has been done on a weighted average basis. Savings NIMs and savings PBT margins are weighted by savings balances, while overall NIMs are weighted by average interest-earning assets.

The products in scope include personal savings accounts and personal ISAs. We have not included data on business accounts.

Funds transfer pricing

Banks’ income and expenses

As part of its business activities, a bank or building society receives interest from borrowers (such as loan or mortgage customers) and pays interest to depositors (for example, savers).

In financial reporting, there is only minimal revenue relating to cash savings, arising from minor fees and charges such as for early withdrawals. The predominant financial statement line relating to cash savings is the interest expense.

For many small firms this is the case, and profitability metrics like NIM can only be measured on an overall business level.

However, many large banks and building societies perform Funds Transfer Pricing (FTP) - an internal process that removes risks from business units (lending and funding) and centralises them into the treasury function. This process helps firms quantify the value that deposits bring to their business, and the cost that lending incurs.

How funds transfer pricing works

The treasury function aggregates funds centrally from the deposit-taking business unit, and then redistributes them to the lending business units. The funds transfer price is credited to the deposit-taking units, such as current accounts and savings, and charged to the ones which lend, such as mortgages. 

Figure 1: Illustrative example of funds transfer pricing

Overall business NIM rose moderately to beyond previous levels in 2023 

Using historic data, we observed relatively flat NIMs in some firms’ retail divisions between 2012 to 2018 for large firms, averaging around 2.5% (Analysis based on results from CMA Retail Banking Market Investigation 2016 and data gathered in the Strategic Review of Retail Banking Business Models).

NIMs slightly decreased between 2019 and 2021 from 2.18% to 2.07%, drivers of which include competitive pressures and the drop in the base rate. They rose from 2021 to 2023 to beyond pre-Covid levels.

Whilst there was variation in firms’ opinions, most acknowledge that the rise in the base rate has contributed to strong margins for savings and other deposits. Meanwhile, they noted that lending margins (for example, mortgage margins) have been squeezed, offsetting the improvement in performance of deposits.

Overall, total NIM in 2023 was on average higher for our sample market than in previous years, suggesting that firms had benefited from the high base rate environment, at least in the short term.

Looking at firms’ 2023 published results, multiple firms saw increased levels of net interest income. Some have noted that this has been influenced by the base rate rises and higher interest rate environment.

Cash savings NIM rose between 2021 and 2023 

For all cohorts, savings NIMs increased between 2021 to 2023, with the big 4 and scale challengers generating the highest savings NIMs. This is due in part to their high proportion of easy access accounts which generate the highest NIMs, as will be discussed later.

Over the period, the gap in NIM between big 4 compared with scale challengers reduced, indicating that as time has passed, these types of firms have benefited to a similar degree from recent market conditions. Digital challengers overall had lower savings NIMs. As new entrants to the market, they offered higher interest rates to customers which increased their interest expense and therefore cut into their savings net interest margin.

Savings FTP income, which refers to external market rates, rose more quickly than interest expense, which is set at firms’ discretion

In the chart below, we look at savings monthly net interest income (NII), split into its components of interest income (ie FTP income) and interest expense (ie interest paid out to customers).

Looking at the firms’ savings cost of funding against the base rate, we can see that the cost of funding for all cohorts, particularly the big 4, started to diverge from the base rate around August 2022. Firms’ savings cost of funding is interest expense divided by balances, a representation of the weighted average interest rate paid out to customers. This suggests a lag between base rate changes and rises in savings interest rates.

We considered reasons for the lag in our Cash Savings Market Review. We noted that coming out of a period of exceptionally low base rates, when banks were determined not to impose negative rates on their savings customers, it might be expected that savings rates would take time to adjust back to their historic differential norms against the base rate (Cash Savings Market Review 2023, page 26).

The big 4 banks have had the lowest cost of funding, while scale challengers were in the middle of the range. This may be due to their ability to leverage a large current account customer base, and as some firms mentioned, a potential oversupply of funding following Covid and ring-fencing.

Challengers and building societies which have no (or lower) current account numbers need to compete with the large players by offering better rates on savings accounts. This echoes why specialist lenders, building societies and digital challengers had the highest cost of funding throughout this period.

Short-term accounts like easy access generated the highest NIM, whilst 2+ year term accounts generated the lowest 

As per Figure 7, both easy and limited access products achieved significantly higher NIMs than term accounts across the period, especially in 2022 and 2023. The table also shows that 2+ year term accounts achieved positive NIMs in 2023, prior to which they had been consistently negative.

Easy and limited access accounts generate the highest NIMs because of a combination of high FTP income and low interest rates. Firms offer low rates on these accounts because these balances in theory present a liquidity risk, in that they can be withdrawn at any moment.

The high FTP income on these products stems from the long ‘behavioural terms’ applied to these products, despite having no contractual term. Banks use behavioural modelling to predict the tenor or ‘stickiness’ of these balances. Many apply a behavioural tenor of several years to a large portion of the balances. The FTP income reflects that of longer-term funding, such as funding which provides better liquidity.

For term accounts, as accounts’ contractual terms increase, the interest rates offered to consumers tend to be higher. Therefore, the interest expense becomes closer to or higher than the associated fund transfer pricing income, resulting in squeezed or negative savings NIMs.

The mix of tenors changed, with a growing proportion of term accounts compared with easy access

Table 1: Average tenor mix across sample

_{Includes all firms within scope of cash savings RFI, big 4 banks, 4 scale challengers, 2 mid-tiers, 2 digital challengers, 3 building societies and 2 specialist lenders.}

As shown in table 1, easy and limited access accounts continued to make up much of the savings mix (based on average balances) across firms.

However, there was a larger uptake in term accounts between 2022 and 2023, with consumers actively seeking higher-paying savings accounts to improve returns during the higher base rate period. This led to a decrease in the proportion of balances held within easy access accounts.

This shift in mix has contributed to the slowing of FTP income growth and the flattening in NIM. Banks now have an increased share of less profitable fixed-term products (eg 4.4% to 10% for 1-year fixed term) and a decrease in the more profitable easy access products (84.4% to 75.9%).

Big 4 shared the lowest proportion of benefit with consumers, while digital challengers shared the highest  

In Figure 8 above, we analysed the percentage of the benefit recognised by firms (FTP income) that was passed on to savings consumers via interest payments (interest expense). This was calculated by dividing the interest expense paid out by FTP income. A larger percentage paid out in interest expense is therefore a reasonable proxy for the percentage of savings benefit being passed on to savings customers.

Digital challengers offered the largest benefit-shared percentage, likely a result of high rates offered to gain market share as new entrants. Scale challengers’ benefit-shared percentages improved from 41% to 63%.

At the start of the period (October 2021), the big 4 banks exhibited the lowest percentage of savings benefit passed to consumers at 11%. Their large market shares with high levels of consumer stickiness reflected the low level of competition they faced from challengers. Their large current account book and ability to access wholesale funding meant they had high levels of funding from outside of the savings book, with therefore a lower incentive to grow savings balances.

Base rate increases began from December 2021 onwards but the steeper increase in benefit shared did not take effect immediately. The percentage in benefit shared with customers saw an acceleration from around May 2022 onwards. This acceleration was likely due to new market entrants which increased competitive pressure in the market. The big 4’s benefit shared percentage became closer to that of other firms over the period, which may have been driven by this increased competitive pressure.

Overall, banks have increased their percentage of benefit passed on through the period and consumers saw more of the benefits that come with an increase in base rates.

The rate of benefit passed on was lower for easy access accounts compared with term accounts

We analysed the benefit shared percentages by tenor.

Easy access accounts’ proportion of benefit shared increased steadily throughout the period, albeit the proportion of benefit shared was consistently the lowest when compared to term accounts.

Term accounts saw a decline in the percentage of benefit shared until around July 2022, before they started to rise.

We observed convergence from February 2022 as the gap benefit shared between the account tenors narrowed. Firstly, easy access accounts’ proportion increased steadily. A possible reason for this is a steeper increase in easy access rates offered, with consumer engagement and commercial pressures in the easy access market intensifying. Greater competition from challengers also played a part.

As well as this, 3-year term accounts’ proportion steeply declined before moving in line with other term accounts. Firms noted that sufficient liquidity may have led to lower firm demand for 3-year funding. 

Savings NIM rose more quickly than overall NIM, indicating savings performance improved more quickly than other business lines

We compared an average savings NIM with an average overall NIM for a sample of firms to identify differences between the performance of savings divisions with that of their overall performance. Given the savings NIM is lower than overall NIM over this period, a low differential between them would indicate that savings had a significant impact on overall NIM, and/or that savings are performing very similarly to other business lines.

As shown in Figure 10, we found that both the average savings NIM and average overall NIM increased from October 2021 to September 2023. Similar to NII, we first saw periods of rising NIMs before a flattening off and slight decline across 2023.

What is clearer here is that the savings NIM increased more quickly than overall NIM across the first half of the period. As noted above, this suggests savings performance grew at a faster rate than other business lines, which is in line with firms’ assertions, particularly around compressed mortgage margins.

This strong performance in savings indicates that savings have been increasingly contributing to improvements in overall profitability.

The overall NIM above includes all other business lines within the retail banking (ring-fenced, if relevant) group, including business banking, lending, and non-consumer interest such as on deposits held at other banks.

Savings net interest income as a share of total net interest income increased, showing growing positive contribution from savings

Figure 11 shows Savings NII as a proportion of total NII, which we have used as a proxy for the contribution that savings made to overall NII. Savings made a positive and growing contribution to overall net interest income (NII), from 11% in October 2021 to 29% in September 2023. We observed signs of this contribution flattening from March 2023, before slightly decreasing.

Given that savings’ net interest income as a proportion of total NII has grown, this also suggests savings grew in materiality to the business. This is reiterated by the analysis on savings as a proportion of total liabilities.
 

We found that for all cohorts, savings were a significant proportion of all liabilities. Building societies had the highest proportion of savings relative to their total liabilities at 74%, this is because they fund their lending via savings and do not offer current accounts.

In contrast, the big 4 and scale challengers had the lowest proportion of savings relative to their total liabilities at 29% and 42% respectively. This is likely to be due to their higher current account balances and ability to access wholesale funding. Therefore, savings would make up a smaller proportion of total liabilities when compared to building societies.

How firms allocated operating costs

Firms described their indirect costs as standard central overheads including marketing, HR, accounting, and IT. For the savings division, many firms used a method to allocate indirect costs to the division. However, they could not allocate them across product type.

Firms tended to interpret allocated indirect costs in one of three ways:

• Costs that were identified as being directly traceable to the savings division but not to directly producing the products, for example, branch costs or staff costs.
• Estimating and then allocating a proportion of wider firm central overheads to the savings division.
• A balance of the two above. For example, a firm may allocate indirect costs that they could identify as being directly related to the savings division and then allocate a further percentage of the overall central overheads.

Regarding their other direct non-interest income or costs, firms had a consistent understanding that this only related to costs that were directly attributable to savings.

Operating costs remained broadly consistent over the period

Our analysis shows that costs stayed consistent across the months with a marginal compound monthly increase of 0.03%. The big 4 had the lowest average monthly growth (0.34%) and scale challengers experienced a slight decline (-0.37%).

Building societies and mid-tiers had slightly higher figures with monthly growth of 1.2% and 1.29% respectively. These figures suggest that larger firms were able to maintain flatter overhead levels compared with smaller firms.

Profit before tax (PBT) followed the same pattern as Net interest income (NII)