Central bank payments systems and the CPMI-IOSCO PFMIs

This is the first in a series of articles considering central bank payment systems self-assessments against the PFMIs.

Introduction and the Bank of England Example

In response to the financial crises of 2007-2009, at the behest of G20 governments, the Financial Stability Board and its constituent bodies developed broad global standards to shore up a system that had proven all too fragile – though it must be said that the public, regulated marketplaces did function throughout (except in isolated cases where for a few days their governments closed them for fear of collapsing prices). The same cannot be said of the freezing up of the far larger OTC and banks’ market operations in that period, which was the source of the economic and social damage inflicted.

Together, in the securities and financial derivatives markets areas, central bankers and supervisory agencies elaborated a critical body of standards in this concerted effort, the CPMI-IOSCO Principles for Financial Markets Infrastructures (the ‘PFMIs’). CPMI-IOSCO designated four types of institutions for detailed oversight: central securities depositories, central counterparty clearing houses, trade repositories, and payment systems. The Principles were built as a tool for organising the information both officials and the marketplace would need to have at hand on their workings.

With this firm’s historic focus on post-trade risk management in asset custody and market infrastructure operations, Thomas Murray has been involved in PFMI work since its published commentary sent to officials during their drafting in 2011. The PFMIs were promulgated in 2012.

The PFMIs are self-assessment reports written by the four types of infrastructures; they are not third-party audits or formal reviews, though relatively recently teams of central bank and capital markets authorities representatives have begun to assess the work done in other markets, and have begun to organise a strict schedule for this essential, if still gentle, form of enforcement.

Building off its years of infrastructure risk assessments, Thomas Murray has worked closely with infrastructures preparing their PFMI self-assessments, offering an external view to assist in the thinking and evaluation of materials. A key difficulty with the PFMIs is how one can know with some precision whether the principle is being partly or broadly or fully observed – how does one justify what is effectively a qualitative judgment? In response, this firm developed a model to input the responses, and give tentative weightings to the key considerations underpinning each principle, and then weightings for the principles themselves to come up with an aggregate “observes” or “broadly observes,” or whatever the result gave. The weightings are adjusted by institution, the point being not the final percentage adopted but the reflection required to arrive at that conclusion.[1]


Central bank payments systems are crucial to the daily functioning of every economy, though they are perhaps the least well know type of infrastructure. (We exclude here trade repositories, which have yet to reach a mature state with an accumulated world-wide track record that could be reviewed.)

In July 2016, the Bank of England published its self-assessment for its Real-Time Gross Settlement Service. It did not establish an overall level of observance of the 24 PFMIs. It declared its observance of Principles:

1 – Legal basis

4 – Credit risk

5 – Collateral

6 – Margin

8 – Settlement finality

9 – Money settlements

13 – Participant default rules and procedures

15 – General business

16 – Custody and investment risks

17 – Operational risk

18 – Access and participation requirements

21 – Efficiency and effectiveness

22 - Communication procedures and standards

23 – Disclosure of rules, key procedures, and market data

The Bank of England declared that it broadly observed Principles:

2 – Governance

3 – Framework for the comprehensive management of risks

The Bank declared that the other Principles did not apply to the case of payment systems, and that it had none that were partly or not observed.

Each Principle has a subset of key considerations, and it is worth taking a moment to dive into the text to see how the Bank proceeded in these evaluations. We will have a look at Principle 16 on custody and investment risk, which is judged to be fully observed.

The first point to make is that the Bank does not cite or reference any documents in support of what it is attesting. Unlike other FMIs, it affirms observance without supporting materials or links to them. Nonetheless, it addresses three of the four key considerations, the last one not being applicable.

It is worth citing the text to underscore the high level of generality at which the text remains throughout. For key consideration 16.1 on holding its own and participants’ assets: “CREST – the UK CSD – along with Euroclear Bank and Clearstream, the two ICSDs, are subject [sic] prudential supervision. The Bank also reviews the ISAE 3402 audits for these institutions.” This firm simply does not find this informative, the more so as in its work it must learn more about how the ICSDs and CREST themselves interact with various other central bank payment systems. We know they do, but this does not even leave a clue as to how. On Principle 16, the Bank of England says it is fully observant with the three key considerations, drops the fourth as irrelevant, and declares overall that it is observant. It does not explain that; it simply wrote that.

In conclusion, this firm would want to know more, and more information could be disclosed without jeopardising the security of the RTGS operations. Sterling payments are a key component of the global financial system, and central bank payment systems continue to evolve – we get echoes to that effect, but actually rather little information on that subject escapes a narrow group of directly interested parties. There is a balance to be struck between divulging too much in terms of security, and informing the market as to a broad trend or questions to be raised. We are not certain that balance has been struck yet.


In this series on the PFMIs for central bank payment systems, we shall look at two other cases where the self-assessments have been published. The question to be explored is whether there are similar problems central banks are experiencing that require remedies to be fully observant of any of these principles, or whether the question of observance varies across these infrastructures.

To conclude, we shall look at which central banks have not yet made their self-assessments public, as has been encouraged by the G20 from the outset.


The author, Thomas Krantz, is Senior Advisor, Capital markets, in the firm of Thomas Murray; and served as Secretary General of the World Federation of Exchanges (2000-2012). The views expressed are his own, and not necessarily those of the firm.

[1] Not all the principles apply to every kind of infrastructure. In the case of payment systems, physical deliver which are essential for commodities derivatives clearing most certainly do not apply. The same would be true for liquidity management, since central bankers know how to solve that problem.

Tags: CPMI-IOSCOPrinciples for Financial Markets InfrastructuresPFMIBank of EnglandderivativesCentral Securities DepositoriesCSDCentral CounterpartyCCPTrade Repositoriespost-trade risk managementinfrastructure risk assessmentscentral bank payments systemsPFMI self assessment