What do UK payments statistics tell us about Covid-19 measures?

Analysis of Payment Statistics for the UK to January-June 2020

Introduction

There has been a lot of discussion of the economic impact of the lockdown measures on consumers and businesses. Suggestions of significant impact to GDP have been shown in other data sets but one source of data has not been investigated in much detail: the inter-bank data from the main clearing and settlement systems.

Inter-bank payments in the UK are the life blood of UK commerce. These include supplier payments, salaries, bill payments including credit cards, house purchases and company acquisitions, card settlements to merchants, person-to-person transfers and mortgages and insurance.

In the first of a series of articles looking at these statistics I consider what is apparent from the set of interbank payments volumes which are published card-based and e-money transfer statistics are less available. Further articles will focus on specific clearing systems, such as Bacs and Faster Payments.

Inter-bank payments

The large payments systems which connect all banks and process the vast majority of the payments value are the focus of this article. Even card payments on UK-issued cards are mostly settled through these systems and, of course, many people pay off their credit or charge card balances through their bank using immediate payments or direct debit.

There are reasons (see Appendix) why analysis of this information is complicated and should be undertaken carefully, but there is a good, definitive source: each of the clearing companies themselves.

The ultimate sources of the statistics are:

• Bacs – Bacs Payment Schemes Limited
• Imaging Clearing Systems – Cheque and Credit Clearing Limited (Pay.UK)
• UK Faster Payments – Faster Payments Schemes Limited
• CHAPS – Bank of England

In addition, there are some useful assumptions which can be made which overcome some of the deficiencies. These are given at the end of the article.

For comparison purposes, figures are calculated as averages per calendar day for the month as some schemes operate only Monday to Friday. The statistics themselves are published on a monthly total basis.

The payment schemes

Each of the schemes operated by Pay.UK is more relevant for some types of payment rather than others. Recurring payments such as utility bills, mortgages, insurance premiums and other subscriptions tend to get paid by Bacs Direct Debit™ because it allows payers to forget about having to make the payment, if they are likely to have sufficient in their account.

Table 1. UK inter-bank payment schemes and their main uses

Can we see the impact of UK restrictions in the payments data?

The UK payments industry is generally a picture of slow change according to its statistics, enlivened by some month-on-month variability caused by, for example, tax payments to HMRC in January and remote banking payments used to open ISAs in March and early April. People and especially businesses are slow to change their habits and processes and industry projects take time to plan and deliver and have migration projects which are measured in years, for example the migration to Bacstel-IP ran from 2003 to 2005.

Firstly, let’s look at all the common, inter-bank payment methods (excluding retail card payments and the CHAPS high-value, real time system). The increase in e-commerce and contactless payments during isolation is a topic for another article. We would expect to see some effect on the schemes above if there was an impact on salaries, bill payments, collections (of subscriptions and regular bills) and change in the use of cheques or bank giro credits.

The graph below shows the total payments value excluding card payments, and CHAPS which is covered later. The need for a payment can recur weekly, monthly, annually or be driven by events. For this reason, I’ve decided to take a daily average per calendar day to adjust for variable lengths of months. It is arguable whether using a monthly total or a per-day average calculated for the month is better in any case, a monthly total graph is given later on.

Across the last two years the overall volume of payments has been fairly constant despite the introduction of image clearing for cheques and bank giro credits. It’s clear that there is a slight increase in overall value processed by these schemes which process around £20 billion per day on average in contrast to the £66.6 billion in March 2020 processed by all of the card schemes equal to £2.2 billion per day average.

What is apparent is that from a peak in March of £21.6 billion per day, roughly equal to the peak in July 2019, the April and May figures are lower than any others since January 2018, reaching a low of £16.6 billion per day in May, 10% lower than the previous minimum, and about 18.4% below the average for the 12 months to February 2020. In June, however this figure rose back up to £18.4 billion per day, which is similar to April’s.

It’s fairly clear that there was an effect from April to June although the overall trend is now back towards the expected total. Whether this will continue in July’s figures is unclear.

It’s reasonable to assume these are the effects of the COVID-19 pandemic, measures to control it and their consequent effects.

While it is clear from the total value graph that something is happening in April to June 2020, the impact on volume of payments is unclear with Bacs Direct Credit and Direct Debit showing little change and Faster payment recording a small reduction in April and May.

So the number of payments is the same but overall the average value is decreasing. This could be for a number of reasons:

• Reduced salary payments for furloughed staff
• Lower value for recurring bills including mortgages and rents
• Delays or decreases in supplier payments due to smaller orders coupled with a delay based on payment terms
• Changes in the reasons why people and businesses pay
• Increases in benefit payments for non-working freelancers and self-employed

To understand what is actually going on, it’s necessary to look into the individual payment methods one by one, which we will do in subsequent articles.

CHAPS RTGS (Real Time Gross Settlement) payments

CHAPS payments are used by businesses for high-value, critical payments including those to settle securities trades and mergers and acquisitions, and by individuals to pay for significant transactions such as house purchases, but they are also used to pre-fund or settle a number of the payment schemes and for a number of other financial market or infrastructure transactions.

CHAPS payments are handled individually and immediately and settled between central bank accounts. The retail and commercial market uses these transactions where confirmation of a payment is required such as in a securities transaction, an asset purchase or managing liquidity between accounts. Consumers will use CHAPS transactions most when buying houses. Financial institutions will also use CHAPS transactions for purchasing securities, for providing credit to other institutions and as a requirement of financial infrastructures such as clearing systems.

Figure 3. The average (mean) daily amount processed by CHAPS since 2018.

The news that the Bank of England made available £200 billion of funding to buy government bonds in March (https://www.theguardian.com/world/2020/jun/22/britain-nearly-went-bust-in-march-says-bank-of-england) may partially explain the increase in value from an average of £5,119 billion per month for the previous 12 months to £6,400 billion in March 2020. It appears that in March 2020, CHAPS processed £23 billion per day over the previous highest daily average for a month.

The use of CHAPS by financial institutions is complicated. In comparison with the average March 2019-Februiary 2020, March’s figure was 12% higher, April’s 3% lower, May’s 13% lower and June’s 3% higher. Some activities will definitely relate to transfer related to the Bank of England’s support and some may relate to reduced transaction volumes of other schemes or reduced dealing in securities. It certainly appears to be a complicated picture.

In contrast and again related to the average March 2019 to February 2020, the volume of CHAPS transactions made by retail and commercial customers is a clear picture of downturn, similar to other business and consumer payment mechanisms. In comparison with March-April 2019, the same months in 2020 saw a 17% decrease in total volume.

Comparing March 2020-June 2020 with the same period in the previous year, wholesale financial transactions are up by 18% for total value but only 15% in total volume. In contrast, the same comparison for retail and commercial use of CHAPS is up 16% in terms of value but down 16% in total volume. This is difficult to immediately explain as there are overlapping causes:

• increased support from Bank of England for the financial system will increase values but not necessarily volumes
• continued commercial activities will require liquidity management, especially with volatile currency markets
• acquisitions and purchase of bankrupt businesses may increase both volume and value of transactions
• securities trading is difficult to predict and may have increased in some months but not others

Part of the drop in average retail transactions may be due to lower volumes of house sales completing. According to HMRC around 100,000 property sales of over £40,000 completed per month at the end of 2019. Any drop would have a small contribution to the 600,000-700,000 per month decrease in transaction volumes seen in April and May.

What are the implications for GDP and economic recovery

ONS statistics have suggested that there was a 20.3% decrease in productivity in April 2020. Do the payment figures bear this out and what do June’s figures suggest for any recovery?

Firstly, it is difficult to assess GDP from payments activity. Because of the long supply chain getting goods from raw materials to consumers, many payments are involved along the way. These payments are frequently on terms of 30 to 60 days, sometimes longer. So tying activity and production to realisation of value can take significant time. In addition, while Bacs Direct Credit is primarily used by businesses, consumer payments for services go more often via Faster Payments or Direct Debit. Tax also makes the position unclear.

The answer to the question “how can we predict of calculate GDP from these statistics?” is almost certainly inaccurately and unwisely. So while the value of Direct Credit contracted by about 20% in April, it is not wise to equate it with an impact on GDP and it’s probably just a coincidence,

Andy Haldane, a Bank of England economist told the Treasury Select Committee that the UK is on course for a “V-shaped” economic recovery. Do these figures bear that out?

While the graphs do show a “v” over the last couple of months it is too early to early and the data are not granular enough to clearly show this, although they don’t rule it out. It’s also not clear what a V-shaped recovery might look like in terms of the payments mechanisms. So the conclusion will have to be “wait and see”. July’s figures will be a good test of Haldane’s belief.

Conclusion

We can see that there is some impact on UK payments, both for businesses and consumers and it’s clear that the impact started in March for CHAPS but April for retail payments; this when ordinary bank payments experienced reduced volumes and values.

Since the system is complicated by different uses, purposes, values and counterparties, further investigation by payment type is indicated.

While there are positive indications of some degree of a return to normal payment operations, it is by no means certain. Consumers and businesses should each play their part to reduce the risk of a second wave and consequent re-imposition of restrictions.

Appendix

It’s not surprising that these statistics have not been investigated in any detail as statistical information comes with a number of caveats:

• the data includes only inter-bank transactions, say from HSBC to Barclays, but not most transactions within a bank;
• one of the systems, CHAPS, is used to pre-fund or assist in settling the value of payments in other payment schemes (Bacs, Cheque Imaging, some acquirer settlements for card payments, …). It’s not double-counting in every case because of the way Bank of England settlement accounts work, the full value is not transferred and not for each payment, but there is an issue of increasing CHAPS volumes and values based on other schemes payments;
• a small proportion will be payments returns, for example payments made and then returned by the receiving bank, in some cases due to error or fraud;
• Banking groups may, or may not, use clearing for some or all payments between accounts at their banks.
• Months have different numbers of days and not all schemes are open all days. Analysing per calendar day rather than per-month seems to give more stability to averages but it may introduce effects or conceal useful month-on-month detail.

Assumptions

• “on us” payments not in the figures will be similar to those covered in the statistics
  Intra-bank payments (within the same bank) which do not go through the schemes above are unlikely to be significantly different from those between banks , taken as a whole
• payment users don’t change the method each time they pay
  Each payment scheme has a slightly different set of purposes it supports, for example Bacs Direct Debit payers and payees are used for regular payments
• card payments value (but not volume) are largely included in the totals
  Settlement payments from acquirers to merchants would appear in the figures and equated to around £66.6 billion in March 2020 equal to around one eighth of FPS and Bacs Direct Credit total value.
• credit and charge card bill payments don’t affect the figures significantly
  These payments may also be in the figures but relate to £14.5 billion of card payments in March; some of these will be made as intra-bank “on us” payments or account transfers and most which are not will be made using Faster Payments.

Random TEXT Block