Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
Repo is one of those areas — collectively referred to as ‘financial plumbing’ — which can be safely ignored until the metaphorical ballcock breaks. Until then, it’s mostly bond geeks sucking their teeth about potential disasters that don’t happen in a language seemingly designed to exclude actual humans. However, something interesting is in the pipes.
Readers will recall that Andrew Bailey, in his LSE speech on the future of the Bank of England’s balance sheet, seemed to reckon that repo was pretty much the answer to life, the universe and everything.
To recap, the Bank spent years growing its balance sheet by purchasing assets (mostly gilts), and funding this through the issuance of liabilities (central bank reserves, aka electronic money) in a process we know as quantitative easing. For the past couple of years, the Bank has been both letting gilts mature and actively selling them, in a process we know as quantitative tightening — effectively retiring central bank reserves. And so its balance sheet has been shrinking.
But every Bank of England liability is an asset for commercial banks. And the commercial banks need central bank reserves for regulatory and prudential reasons. Moreover, the level of central bank reserves needed by the commercial banking system is understood to have increased a LOT since 2007, although the exact amount of reserves required for BAU is not actually known. So unwinding quantitative easing without something, somewhere blowing up might be tricksy. And as the Federal Reserve found at the end of 2018, even shrinking the balance sheet down to a still-very-large-number can cause problems.
Enter Short-Term Repo.
Right now, there are still vast quantities of central bank reserves. But quantitative tightening means they will become increasingly less abundant. So, the Bank of England’s Short Term Repo facility was set up in August 2022 as a place where commercial banks and building societies could rock up and borrow as many reserves as they reckoned they might need. In central bank parlance, this would help transition from a supply-driven operational framework to a demand-led system.
Banks need to pay interest on reserves that they borrow on the Short Term Repo facility. And they have to pledge more ‘Level A collateral’ (basically gilts, bills, bunds, OATs and USTs) than they borrow. But, to return to the plumbing analogy, the facility is a kind of neat overflow pipe that at least flags shortages of central bank reserves across the commercial banking system. And it should both prevent the Bank from discovering that it has lost control of its overnight interest rate, and also break the link between reserve shortage and financial instability. Yay.
Let’s check in on the overflow pipe:
MainFT’s Mary McDougall covered this back in May when usage had spiked to £15bn. Since then, it has doubled: almost £30bn of reserves are being borrowed from the Bank on a seven-day basis. (£30bn sounds a lot. But it’s easy to lose track of zeros when looking at central bank balance sheets. To put it in context, £30bn is around the quantity of reserves that have been withdrawn over the past year as the counterpart to active bond sales in the Bank’s quantitative tightening programme.* So not nothing.)
The Bank doesn’t publish data relating to specific counterparties, so we don’t know who it is doing the borrowing. And borrowing short term reserves from the Bank under this facility is specifically not supposed to carry any stigma. As such it would probably be silly to infer that someone is in trouble. But it’s kind of weird that it’s being tapped this much when the Bank’s balance sheet is still so large.
TD Securities’ Pooja Kumra, a gilt market strategist (/plumber), took a look last week at whether this leap in short term repo usage is cause for alarm. In a note for clients, she concludes after a review of the Sonia vs Base basis, GC-OIS term structure, RONIA volatility, and cable basis (me neither) that:
Reserves are certainly plentiful. But money market signals still argue for more caution on how much QT [quantitative tightening] the system can take, especially as central banks are in less of a hurry to cut rates in this cycle.
So is quantitative tightening done for?
Victoria Saporta, Executive Director for Markets at the Bank, gave a speech yesterday to the Association for Financial Markets in Europe in which she went out of her way to actively welcome the spike in Short Term Repo facility usage.
Moreover, she picks up the baton from Bailey in his LSE speech to argue that this is pretty much exactly what the bank expects to see and wants to see. She encourages banks to make greater use of the Indexed Long-Term Repo facility to source non-emergency reserves, and hinted that this facility will soon be made more attractive for banks to use later this year.** Or, in her words:
Some firms currently don’t have access and should be thinking actively about getting that plumbing sorted … let’s get ready to repo!
We can see that greater use of long-term repo would suit the Bank well. It fits the transition to a repo-based operational framework. But we still think it’s a bit weird that Short Term Repo use is picking up quite so quickly while the balance sheet is still around £750bn in size. £750bn is way, way larger than the Bank’s estimate of c£345-490bn for the so-called Preferred Minimum Range of Reserves — the bite-point below which the Bank of England’s balance sheet needs to be to prevent bad things from happening. We don’t completely understand why it’s being tapped, but reckon there are two main possibilities.
First up, Moyeen Islam — Barclays’ gilt guru — reckons there’s a chance that the ramp up in Short Term Repo usage is a sign that the PMRR is actually much higher than the Bank’s estimates. The overflow pipe is gushing for a reason. This would be a big deal for financial plumbers everywhere.
But it’s possible also that central bank reserves are increasingly just in the wrong place. While aggregate reserves might be abundant, reserve shortages could be biting in some institutions — and the pricing of the STR makes it more attractive to borrow from the Bank than from other commercial banks with reserve abundance. Why tap the market at a RONIA rate of 5.30 per cent when the Bank will lend to you at 5.25 per cent?
We’re frankly not geeky enough to read the runes of sterling money markets and work out whether something is about to break from too much quantitative tightening, but will be checking in from time to time on the STR’s weekly auction results to see whether escalating usage continues to suggest — against our expectations — that the Bank’s steady-state balance sheet is far larger than (almost) everyone imagines.
*£30.748bn APF sales since 24 Jul 2023 vs current draw of £29.076bn on STR facility.
**“We are therefore also in the process of reviewing the calibration of the ILTR to ensure it is effective and attractive enough so that it is used in business-as-usual (BAU) consistent with our policy objectives.”
https://www.ft.com/content/fa21693c-e077-427b-b712-effeb0e1d931