Drawing on a quantitative analysis of recent liquidity shocks, a Bank of England (BoE) research paper concludes that wider use of central counterparty clearing will reduce the impact of repo trading activities on dealer capital ratios and extend their capacity to provide intermediation in financial markets.
While the high level findings come as no revelation — the core benefits and costs of CCP clearing have been widely discussed in SFT and elsewhere — its data-based analysis of how CCP clearing would have impacted dealer balance sheets during recent liquidity shocks is worth detailed reading.
Looking back at the ‘dash for cash’ that accompanied the onset of the Covid crisis, the research paper finds that for the gilt repo market, wider use of CCP clearing would have eased the impact of gilts repo exposures on UK dealer balance sheets by close to 40 per cent and would have improved their aggregate leverage ratio by 3bps.
The paper, released today, is published as Bank of England Staff Working Paper No 1026, The Potential Impact of Broader Central Clearing on Dealer Balance Sheet Capacity: a case study of UK gilt and Gilt Repo Markets.
It is authored by six BoE research analysts specialising in securities finance, money markets and monetary policy, notably Yuliya Baranova, Eleanor Holbrook, David MacDonald, William Rawstone, Nicholas Vause and Georgia Waddington.
The researchers reflect on recent periods of volatility in secured and unsecured financing markets, beginning with the repo markets spike of Q4 2019 in US money markets.
Much of the authors’ quantitative analysis is based on data from the March 2020 Covid crisis, where an initial flight to safety was followed by a sharp ‘dash for cash’, evidenced by sharp selling pressures in typically liquid markets such as US treasuries and UK gilts.
In both of these cases, access to liquidity was impacted by the reluctance or inability of dealers, principal trading firms and other sell-side liquidity providers to ‘intermediate’ the market owing to balance sheet constraints associated with their repo lending.
The BoE team also reflects on the impact of the 2022 ‘gilts’ crisis, when liability-driven investment (LDI) funds came under pressure in the face of sharply rising gilts yields to sell assets to meet margin calls on their derivatives exposures. They note that, again, some dealer firms reined in their appetite to provide intermediation as liquidity conditions tightened, with the Bank of England stepping in to extend liquidity through its asset purchase programme.
Their quantitative analysis evaluates the degree to which wider use of central clearing could ease balance sheet constraints for the dealer community and facilitate wider dealer intermediation.
They note that wider use of central clearing may also unlock other potential benefits including lower settlement fail rates and wider incentives for “all-to-all” trading, thereby improving liquidity access by enabling a wider range of market participants to engage with each other without intermediation.
For the gilt repo market, the paper finds that had comprehensive central clearing been employed throughout the ‘dash for cash’ (DFC) shock in March 2020, this could have reduced the balance sheet impact of gilt repo trading for UK dealer firms by 3.8bps, or close to 40 per cent, in aggregate (p 25). This compares with a 3.3bps (or 35 per cent) reduction in the leverage ratio impact of gilts repo trading if comprehensive central clearing had been in place over the DFC shock.
However, their research notes that this impact varies significantly across dealer firms, with central clearing reducing the leverage ratio impact for some dealers by as much as 10bps but having much less effect for other dealers.
This variation across dealer firms is explained largely by the structure of their repo books and the differing business models they employ. Generally, dealers whose repo books are more closely matched, in terms of repo and reverse repo positions having similar sizes and maturities, were most likely to benefit from large netting benefits.
The BoE paper finds that with further steps to promote standardisation of maturity dates for repos and reverse repos, a further £116 billion, or 14 per cent of outstanding gilt repo, would have been eligible for netting benefits. This assumption recognises that, as a condition for trade netting, offsetting transactions should have the same maturity date.
Maturity standardisation would have increased the stock of nettable repo by 4 per cent of outstanding repo, or approximately £40 billion, the paper finds (p 26).
It notes that “moderate standardisation” of maturity dates could have lowered the impact of gilt repo trading on dealers’ aggregate leverage ratio by a further 2.0 bps during the ‘dash for cash’ crisis and 1.7 bps pre-DFC, raising the total liquidity ratio improvement through central clearing by 5.8 basis points during the ‘dash for cash crisis’, and 5.0 bps for pre-DFC.
“Although the 5.8 bps reduction in leverage ratio impact seems relatively small compared with the total buffers, the extra capacity could be material in the context of the balance sheet capacity of the repo desk,” the paper concludes.
Whether, in practice, the introduction of further central clearing and maturity standardisation does result in greater ability and willingness of dealers to provide intermediation, particularly at times of liquidity tightening, will be dependent on the dealer’s business model and how they choose to utilise this additional balance sheet capacity realised by CCP clearing.
Inevitably, if the dealer chooses to reallocate this additional headroom to other business areas, for example, this is unlikely to result in little additional balance sheet capacity offered to the repo desk and little improvement in the dealer’s willingness to offer repo market intermediation during times of liquidity tightening.
Looking at the operational benefits of central clearing in gilts markets, the research team concludes that the extension of CCP clearing for cash gilts could reduce gross settlement obligations by 48 per cent on an average trading day and 53 per cent on a peak day (p 28).
This, it believes, is likely to reduce settlement fails rates and lower the capital charges for dealers associated with trade failures, along with settlement penalties under the settlement discipline regime of the Central Securities Depositories Association.
The paper finds that the settlement fail rates for gilts and gilt repo markets rose from around 2 per cent in Jan-Feb 2020 to peak rates of 6-7 per cent with the ‘dash for cash’ shock in March 2020. However, settlement fails rose much less in the cleared space over this shock period, with fail rates peaking at approximately 2 per cent for CCP-cleared transactions (p 29).
