All too often, whenever we discuss payments, we forget about the actual underlying liquidity. This is true of faster payments (FPS), here in the UK, SEPA in Europe and pretty much anywhere in the world. The most glaring error though is overlooking liquidity when we talk cross-border payments.
So why is it overlooked? I can only assume its because payments are typically all about payment messages, rather than actual flow of funds. When we talk cross-border payments we are constantly talking purely about the messaging. SWIFT GPI is seen as a cross-border payment component, but all it does is track a message. A message which isn’t actually the movement of funds….When we talk FPS we talk about the messages, the payment instruction, the acknowledgement and then notifying our end customer. When messages flow over the FPS infrastructure, money has not moved, simply because like most instant payment systems, they are clearing systems. These clearing systems then move money via a scheduled batch processes which moves the money and completes the settlement element of the payment. Essentially, FPS is a message that contains a “promise” to move money / make a payment. The promise is backed by locked in liquidity, so the recipient does know they will get the money.
In terms of payments though, just think what is happening from a liquid funds point of view. The sending bank has liquidity locked away to back up its promise to pay the crediting bank. The crediting bank is paying that money to the end beneficiary, who may want that in true liquid terms and withdraw cash. That means the crediting bank is leveraging its own liquidity to facilitate crediting the end beneficiary. So, my payment of £1 means £1 is locked away my bank, and the crediting bank has used its own £1 to complete the payment. In terms of liquid cash £2 could be being used for a single £1 payment. However, things get worse. For the sending bank, well the money that moves is not the money that is backing the promise, no, its liquid funds in the central reserve account that move as part of the settlement process. So, my £1 payment means my sending bank has locked £1 away, to back the promise of the payment, made the payment with an additional £1 taken from its reserve account, and the crediting bank has temporarily (potentially) stumped up its own £1 to pay the beneficiary – though it will receive £1 in its settlement account later. At a macro level then, moving £1 means I have for sure used £2 and may of needed £3…..
If you multiply this up based on the values and volumes that we see flow through FPS in the UK, we soon start to see that there are vast amounts of cash that is simply “locked”, it cannot be put to any real use. For the banks themselves, these are funds that could be leant, put to work, earning additional revenue for the bank. I wonder, has anyone looked at the true “costs” of processing an FPS transaction?
Liquidity costs money
When we started ClearBank, one of the big challenges was understanding what liquidity would be locked away. When you service other banks, this gets even more complicated as you need to take a bit of a “punt” on what money they wish to be moving. Treasury must put real liquid funds into a collateral account to back the promise of a payment in FPS, thankfully the same is not true of CHAPS (RTGS). If you get this wrong, guess what, your payments stop flowing and your customers get very unhappy (and rightfully so).
For smaller FinTech players, trying to access FPS directly is pointless. Though many are direct participants and no doubt many more will look to join. I say its pointless because a) you could just use ClearBank API and you get all the benefits of being directly connected and b) you have to make sure you have your own liquid funds to back the promise of the payments your customers will make.
If you’re a small FinTech then and your balance is say £5m, which is all your customers money, then how much of that are you willing to “lock” away? 1m, 2m, 3m? Well whatever you lock, your customers can no longer access. So, if you have customers that want to pay away pretty much more than half their balance, you simply cannot make those payments. So you have the £5m, customers want to pay out £2.6m well they cannot, because you can only lock away £2.5m max. The maths do not lie. So, you need to borrow money, or raise additional investment in your FinTech to support your customers potential payment flows. How much has that cost you? If you are borrowing it, then you are losing money at a macro level. If you are raising the funds then, what have you lost in terms of your stake in the company as a founder going forward? That 15% stake that you have just used to fund customer flows, what will that cost you if you get to say a £250m valuation????
For larger banks, the challenge is about lost revenue opportunities. Lock up too much liquidity and you are losing potential revenues, lock up too little and you cannot facilitate customer payments.
Cross border is tough.
When we look at cross-border payment flows, things get worse. For a bank, you must either have a credit line with another bank or you are “pre-funding” accounts based on forward forecasts of what payment flows may look like. (So, you are locking up liquidity – or paying for credit). Things get worse, because not only are you locking up liquidity with other banks, you’re actually also having to lock liquidity away in your own base currency to cover any associated risk of another bank holding your funds. This could be anything between 7%-30% of the balance you are holding with another bank.
When we talk cross-border payments, the discussion gets confused between the payment message, and the flow of funds. A cross-border payment message is pretty real-time (ish) if, and it’s a big IF, your bank has funds in the destination currency already, and that both banks and jurisdictions are open. In such a case the payment message gets there in real-time(ish) and the payment will probably get processed in a similar time frame. SWIFT GPI will even show you the message flowing over the network.
However, in reality though, your cross-border payment journey was triggered off by locking up liquidity and trying to get that right, based on a forward forecast. This is before we look at the fact that your bank does not have a relationship with the bank that holds the end beneficiary bank account, or your bank does not even have a relationship with a bank in that jurisdiction. By over-laying these additional factors, by adding in “intermediaries” and correspondents, we also must multiple out the liquidity challenge across all the banks in the correspondent network chain – just to process my single payment.
Locked liquidity and processing friction can be termed as financial friction. This is estimated to cost you, me, businesses, the banks themselves, the global economy if you will, some $15trn a year. This cost is pretty much down to the challenge of managing liquidity, the overlooked component of payments.