Most EMIs Fail to Use their Safeguarding Account Properly, Here’s Why
In July 2020, the FCA noted on its website that their surveys revealed that some firms (including UK Electronic Money Institutions, or EMIs) were commingling customer and firm funds, keeping inaccurate records and accounts, and not having sufficiently effective risk management procedures, thereby failing to use their safeguarding account properly, as set out in the guidelines given to them by the FCA. Another survey in 2019 among 11 non-bank payment service providers (PSPs) found that:
“Some firms were unable to explain which payment services they provided in certain situations or identify when they were issuing e-money, nor was there clarity about when they were acting as agent or distributor for another payment service provider. This meant that they could not accurately identify relevant funds and did not know whether they were safeguarding the correct amount of relevant funds.”
Unfortunately, this holds true not solely in the UK but in other European countries as well and the same problems can be noted in such fintech hubs as Lithuania. In this article, we review to what extent companies with E-money License in Europe are sticking to safeguarding rules and using the safeguarding account properly. In order to ascertain this, we use statistical data from the Bank of Lithuania’s database covering 43 Lithuanian EMIs for the period of 2019-2021.
The results show that 48% of the companies with E-money License in Lithuania in the sample incur discrepancies greater than 50%, these discrepancies being imbalances between the funds they hold in their safeguarding account or accounts and the e-money liabilities they have in issuance, which may take the form of shortages (less safeguarded funds relative to e-money) but more commonly of excesses (more safeguarded funds relative to e-money).
This exorbitant figure only points to the need on the part of EMIs to watch their safeguarding practices more carefully. This is particularly true for recent entrants to the EMI market. According to the same data, 62% of companies with so-called Small EMI License run discrepancies greater than 50%.
We start in Section 1 by explaining the balance sheet mechanics of customer funds safeguarding. In Section 2 we perform some statistical analyses.
Safeguarding account and segregation explained using balance sheets
A critical and perhaps not-well-known difference between credit institutions (which includes high-street, money-centre commercial banks) and electronic money institutions (EMIs) is that, while the former can commingle (mix) the funds deposited by customers with their own funds and use both for their own purposes (e.g., to extend credit), the latter must ring-fence all funds received from customers and keep them separate from their own on “segregated accounts“. While EMIs do have access to customer funds, they are forbidden from using them for purposes other than purely transactional ones involving the issuance and redemption of e-money. Click the following links to learn more about the difference between banks and EMIs and how E-money License is different from a banking license.
EMIs stick to this requirement in the following way. Say I want to transfer €1,000 from my current account at my bank to my e-money account with an EMI (I did a similar operation a few days ago). What these two financial institutions will do behind the scenes to ‘settle’ my transfer is something as follows. Let’s suppose the EMI holds customer (i.e., my) funds with the same bank where I hold my current account. To settle the transfer, the bank debits (subtracts from) my account and credits (adds to) the EMIs’ account (what’s called an “in-house settlement”, because no funds leave the bank) (Fig. 1).
Similarly, if I were to transfer funds out of my e-money account, the reverse operation would take place; for every €1 transferred out of my e-money account, the EMI would see a €1 flow out of its customer funds safeguarding account (provided that I was not transferring the funds to another of the EMI’s customers) (Fig. 2).
The transactional mechanics explained in Figs. 1 and 2 lead us to expect that, after many transfers in and out of the EMI’s customers’ e-wallets, at every moment the EMI would have kept the same amount of funds in its customer funds safeguarding account (its asset) as it had of e-money in issuance (its liability). In other words, we would expect parity between these two items, and the balance sheet of the EMI to look something this:
Reality is a bit messier, of course, and EMI regulations are not so stringent. For example, in the above example, I have assumed that the EMI holds the safeguarded funds ‘with’ another credit institution, but in practice, it may also hold them with a central bank or – less frequently – invested in low-risk, liquid assets.
Fig. 4 shows data for 43 Lithuanian EMIs. Whereas in 2019 the three holding methods were much evenly used, by Q1 2021 EMIs in Lithuania had shifted most of their safeguarded funds to accounts at credit institutions.
Not only that, but EMIs may also hold safeguarded funds denominated in currencies other than those in which their corresponding e-money liabilities are denominated. For example, a customer may place funds in its EUR e-money account, while the EMI where that account is held places the “same” amount (using the appropriate exchange rate) of funds in a USD safeguarding account.
I have also overlooked the fact that EMIs are given not one but two choices when deciding how to safeguard customer funds (see article 25 of Law on Electronic Money and Electronic Money Institutions). One is the ‘segregated funds’ method outlined above. The other is an ‘insurance policy’ method. The evidence suggests that EMIs overwhelmingly use the first method (Bank of Lithuania, 2019; PSP Lab, 2020); thus the reason for glossing over the second method in this article.
For these and other operational reasons, discrepancies are bound to (re-)occur, which can be defined as:
Situations where the amount of funds assets that the EMI holds on their safeguarding account differs from the amount of e-money liabilities it has on issuance.
The next figure shows a typical example of how a discrepancy may arise. The balance sheet in Fig. 5 shows the before and after of the EMI charging a €50 fee to one of its customers, “Customer X”. The EMI debits Customer X’s e-money account by €50 and keeps the proceeds in its safeguarding account, thereby creating a discrepancy of €50.
Regulators, while cognizant of the fact that eliminating these discrepancies completely may be operationally burdensome and not worth the stretch, do nevertheless ask EMIs to carry out reconciliations as often as is practicably possible. For example, the FCA stipulates that where “discrepancies arise as a result of reconciliations, [EMIs] should identify the reason for those discrepancies and correct them as soon as possible by paying in any shortfall or withdrawing any excess, unless the discrepancy arises only due to timing differences between internal and external accounting systems”, and in “no circumstances would it be acceptable for corrections to be made after the end of the business day.” (FCA, 2021, §10.65).
“The reconciliation should result in the amount of funds or assets safeguarded being: (1) sufficient to cover the amount that the institution would need to safeguard before the next reconciliation; and (2) not excessive (to minimise risks arising from commingling)” (FCA, 2021, §10.61).
In the case of European law, no rule similar to that of the FCA seems to be in place. Nevertheless, Article 7(1) of Directive 2009/100/EC does establish the requirement to safeguard customer funds. In Lithuania, this is transposed in Article 25 of the Republic of Lithuania Law on Electronic Money and Electronic Money Institutions.
As explained in our Short Guide to Safeguarding, the reason for this insistence on reconciliation lies in the fact that if not carried properly, tracking the funds becomes difficult. Furthermore, as a consequence of commingling funds of customers with funds of institution and subsequent insolvency of the institution the funds may form part of the institution’s insolvency estate. A self-defeating fact, as it eliminates the very benefits that safeguarding is aimed to achieve.
To continue with our previous example, in order to correct the discrepancy, the EMI would have to transfer €50 from its safeguarding account to its own funds account by the end of the business day when the discrepancy arose, ideally restoring the parity between e-money and safeguarded funds in doing so, as shown in Fig. 6.
But how well do EMIs stick to safeguarding requirements in practice? In the next section, we look at the empirical data to find out.
What the data say
A straightforward way to test whether EMIs run imbalances between the safeguarded customer funds they hold and the e-money they have in issuance is to simply look at their balance sheets and compare these two items.
The next two figures show data for 43 Lithuanian EMIs. Fig. 7 shows the scatterplot of segregated funds (horizontal axis) and total e-money liabilities outstanding (vertical axis). It can be seen that the values for the two variables are similar and highly correlated, but rarely identical.
Fig. 8 gives a closer look at the imbalances between the two variables by computing their ratio for March 2021. As can be appreciated, discrepancies are the norm, and quite substantial in magnitude, although skewed to the excess side (the median ratio of segregated funds to e-money liabilities for the period Dec-2019 to March-2021 is 1.26).
Given that both types of discrepancies (i.e., excesses and shortages) are undesirable from the point of view of regulators, we can calculate what percentage of these EMIs run imbalances greater than 50%, meaning they hold an amount of safeguarded funds 50% smaller or bigger than the e-money they have in issuance.
The numbers are stunning: on average, 47.8% of these EMIs run imbalances greater than 50% (Fig. 9).
If we break down the sample of EMIs into two groups, small EMIs (i.e., those with e-money in issuance of less than €5mn) and the rest, we find that small EMIs are the worst offenders: on average, 61.9% of them run discrepancies greater than 50%, while the same figure for bigger EMIs is somewhat lower, 27.1%, as shown in Fig. 10.
These numbers are big and alarming. Consider the following hypothetical analogy: 50% of credit institutions in the euro area have a Liquidity Coverage Ratio (LCR) lower than 100%, and 27% of G-SIBs (Global Systemically Important Banks) have a Leverage Ratio (LEV) lower than 3%, both lower thresholds set by the Basel III regulations.
Although – unlike the UK’s FCA – the Republic of Lithuania Law on Electronic Money and Electronic Money Institutions does not establish a maximum number of working days during which EMIs must correct their discrepancies, the data do suggest that these EMIs, especially small ones (those with e-money liabilities smaller than €5mn), are doing a poor job in their safeguarding practices.
As mentioned in the introduction, UK-based EMIs seem to be no exception to this, and poor safeguarding practices and procedures seem widespread in the EMI sector more generally.
How can PSP Lab help you?
If you are an existing EMI or a newcomer to the business, you may want to contact PSP Lab, since we are a niche consultancy company that helps companies in getting e-money licenses in the EU and the UK. We can also help you better understand the operational and legal aspects of safeguarding, safeguarding accounts and the e-money issuance business more generally. We’re here to help you not only navigate the complex EMI licensing landscape but also to help you understand how you can develop your e-money business.
Here is Short Guide to Safeguarding and the article explaining top 5 dangerous mistakes that an EMI can make in its safeguarding practices.
Contact us for a free consultation and we will identify and analyse your possibilities of obtaining the license in Europe as well as help you to understand the regulatory requirements related to the electronic money licensing process in the most suitable jurisdiction for you.