19 November 2024
ESMA74-2119945925-2059

ESMA
European Securities and Markets Authority

Final Report

Technical Advice on CSDR Penalty Mechanism

ESMA - 201-203 rue de Bercy - CS 80910 - 75589 Paris Cedex 12 - France - Tel. +33 (0) 1 58 36 43 21 - www.esma.europa.eu

Table of Contents

1 Executive Summary ......................................................................................................2

2 Introduction ...................................................................................................................9

3 Alternative parameters, when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available ......................................11

3.1 Background..........................................................................................................11

3.2 Feedback statement ............................................................................................13

3.2.1 Alternative parameters proposals for penalty rates applied in case of settlement fails due to lack of cash (Q1, Q2, Q3) .......................................................13

3.3 ESMA’s assessment ............................................................................................15

3.4 ESMA’s technical advice......................................................................................17

4 Treatment of historical reference data for the calculation of late matching fail penalties 19

4.1 Background..........................................................................................................19

4.2 Feedback statement ............................................................................................21

4.2.1 Accumulation of reference data issue (Q5) ..................................................21

4.2.2 Causes of late matching (Q6).......................................................................21

4.2.3 Establishment of reference data used for the calculation of cash penalties (Q7, Q8, Q9, Q10, Q11, Q12).............................................................................................23

4.3 ESMA’s assessment ............................................................................................25

4.4 ESMA’s technical advice......................................................................................26

5 Alternative methods for calculating cash penalties, including progressive penalty rates 29

5.1 Background..........................................................................................................29

5.2 Feedback statement ............................................................................................30

5.2.1 Functioning of the current penalty mechanism .............................................30

5.2.2 Proposed revision of the penalty mechanism ...............................................41

5.2.3 Additional considerations to simplify the penalty mechanism.......................49

5.3 Additional evidence..............................................................................................59

5.3.1 Settlement efficiency trends..........................................................................59

5.3.2 Securities lending and borrowing rates.........................................................65

ESMA
European Securities and Markets Authority

5.3.3 Duration of settlement fails in T2S................................................................67

5.4 ESMA’s technical advice......................................................................................68

6 Annexes ......................................................................................................................74

6.1 Annex I – Summary of Questions ........................................................................74

6.1.1 Alternative parameters, when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available ............74

6.1.2 Treatment of historical reference data for the calculation of late matching fail penalties......................................................................................................................74

6.1.3 Alternative methods for calculating cash penalties, including progressive penalty rates................................................................................................................77

6.2 Annex II – EC Mandate regarding Technical Advice on the parameters for the calculation of cash penalties for settlement fails caused by a lack of cash ....................87

6.3 Annex III – EC Mandate regarding Technical Advice on the treatment of historical reference data for the calculation of late matching fail penalties (LMFPs) .....................94

6.4 Annex IV – EC Mandate regarding Technical Advice on the potential calibration of the structure and severity of cash penalties to discourage settlement fails, incentivise their rapid resolution and improve settlement efficiency.......................................................100

6.5 Annex V: Cost-benefit analysis ..........................................................................106

6.5.1 Alternative parameters, when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available ..........106

6.5.2 Treatment of historical reference data for the calculation of late matching fail penalties....................................................................................................................108

6.5.3 Alternative methods for calculating cash penalties, including progressive penalty rates..............................................................................................................110

ESMA - 201-203 rue de Bercy - CS 80910 - 75589 Paris Cedex 12 - France - Tel. +33 (0) 1 58 36 43 21 - www.esma.europa.eu 2

1 Executive Summary

Reasons for publication

CSDR (Regulation (EU) No 909/2014) 1 was published in the Official Journal on 28 August 2014, and entered into force on 17 September 2014. CSDR was amended by CSDR Refit (Regulation (EU) No 2023/2845) 2 , which entered into force on 17 January 2024.

CSDR includes a set of measures to prevent and address failures in the settlement of securities transactions (settlement fails), commonly referred to as settlement discipline measures. They consist of reporting requirements, cash penalties for CSD participants in case of settlement fails, and mandatory buy-ins where a CSD participant fails to deliver the securities within a fixed extension period (which, according to CSDR Refit, have become a last resort measure where the rate of settlement fails in the EU is not improving and is presenting a threat to financial stability).

Although settlement fails cannot be totally eliminated, persistent settlement fails negatively affect the functioning and competitiveness of the capital markets and contradict the objectives of the Savings and Investments Union, which aims to improve the functioning of market infrastructures across the EU.

Cash penalties should not only deter participants from causing settlement fails, but also incentivise the failing party to rapidly resolve the settlement fail: the failing party is charged a daily penalty for each business day that a transaction fails to settle after the intended settlement date (ISD).

The European Commission (EC) is empowered to adopt delegated acts to specify the parameters for the calculation of a deterrent and proportionate level of cash penalties.

According to Commission Delegated Regulation (EU) 2021/70, cash penalties are being applied to failing settlement instructions in securities settlement systems operated by EU CSDs as of 1 February 2022. Based on the settlement fails reports it receives, ESMA notes a decrease of settlement fails since February 2022. This decrease can however be considered subdued, and certain asset classes, in particular ETFs, are still associated with high levels of settlement fails.

[1] Regulation (EU) No 909/2014 of the European Parliament and of the Council of 23 July 2014 on improving securities settlement in the European Union and on central securities depositories and amending Directives 98/26/EC and 2014/65/EU and Regulation (EU) No 236/2012

[2] Regulation (EU) No 2023/2845 of the European Parliament and of the Council of 13 December 2023 amending Regulation (EU) No 909/2014 as regards settlement discipline, cross-border provision of services, supervisory cooperation, provision of banking-type ancillary services and requirements for third-country central securities depositories and amending Regulation (EU) No 236/2012

In the context of the application of cash penalties in case of settlement fails caused by a lack of cash under CSDR, on 13 December 2022, ESMA received a request from the EC for technical advice on alternative parameters, when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available.

In addition, on 15 May 2023, ESMA received a second request for technical advice on specifying the treatment of historical reference data for the calculation of late matching fail penalties (LMFPs).

Last but not least, the full potential of measures to address settlement fails, in particular cash penalties, must be explored, also as a low level of settlement fails is essential in light of the ongoing discussions about a potential shortening of the settlement cycle in the EU.

In light of the above, the EC has sent ESMA a third request for technical advice to assess the effectiveness and proportionality of the current penalty mechanism and propose, if justified, changes to the structure or severity of the mechanism and consider alternative methods for calculating cash penalties, including by introducing progressive penalty rates.

ESMA published a Consultation Paper (CP) in December 2023 (the consultation period was open until the end of February 2024) to collect views, comments and opinions, as well as data and evidence from stakeholders and market participants on the effectiveness of the current penalty mechanism in discouraging settlement fails and incentivising their rapid resolution, and on ESMA’s preliminary proposals on the following topics:

a) alternative parameters, when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available;

b) the treatment of historical reference data for the calculation of late matching fail penalties;

c) alternative methods for calculating cash penalties, including progressive penalty rates.

Given the overlapping scope the three requests for technical advice, ESMA has combined them in one document.

Following the introductory section, this Final Report contains three main sections on the topics mentioned above. It covers a detailed summary of the feedback received to the CP as well as ESMA’s assessment, together with ESMA’s technical advice to the EC. The Final Report also includes the impact assessment regarding the proposed measures as well as the SMSG advice.

ESMA’s technical advice included in Section III recommends that, in the absence of an overnight interest credit rate due to the monetary policy of the central bank issuing the settlement currency, other comparable interest rates of the ECB and the relevant central bank could be used to calculate a proxy which a CSD can use to calculate the cash penalties due to lack of cash.

In Section IV, given the stakeholders’ input, in order to prevent the accumulation of reference data over time and to ensure the efficient operation of securities settlement systems, ESMA advises the EC to amend the relevant Level 2 provisions to allow CSDs to use the oldest available reference price for the calculation of the related cash penalties, where settlement instructions have been matched after the intended settlement date, and that intended settlement date is beyond 40 business days in the past from the matching date.

Section V on alternative methods for calculating cash penalties examines the feedback received on the functioning of the current penalty mechanism, on ESMA’s wide-ranging proposals for its revision, and on additional considerations to simplify the penalty mechanism. This feedback is complemented with additional evidence on settlement efficiency trends, on securities lending and borrowing rates and on the duration of settlement fails in T2S. ESMA also recognises that a significant increase of penalty rates may divert resources from expected investments and costs for the industry in the context of the move to T+1.

In light of the feedback received and the additional evidence gathered, ESMA proposes i) to maintain the design of the current penalty mechanism, i.e. not to introduce fundamental changes to the methods for calculating penalties, and ii) to introduce an overall moderate increase of the penalty rates for most of the asset classes.

Next Steps

The EC will consider ESMA’s technical advice when amending the Commission Delegated Regulation (EU) 2017/389. The powers of the Commission to adopt delegated acts are subject to Article 67 of CSDR that allows the European Parliament and the Council to object to a delegated act within a period of 3 months, extendible by 3 further months at the initiative of the European Parliament or of the Council. The delegated act will only enter into force if neither European Parliament nor the Council have objected on expiry of that period or if both institutions have informed the Commission of their intention not to raise objections.

Legislative References

CRR3	Regulation (EU) No 575/2013 of the European Parliament and of the Council of 26 June 2013 on prudential requirements for credit institutions and investment firms and amending Regulation (EU) No 648/2012
CSDR4	Regulation (EU) No 909/2014 of the European Parliament and of the Council of 23 July 2014 on improving securities settlement in the European Union and on central securities depositories and amending Directives 98/26/EC and 2014/65/EU and Regulation (EU) No 236/2012
CSDR Refit5	Regulation (EU) No 2023/2845 of the European Parliament and of the Council of 13 December 2023 amending Regulation (EU) No 909/2014 as regards settlement discipline, cross-border provision of services, supervisory cooperation, provision of banking-type ancillary services and requirements for third-country central securities depositories and amending Regulation (EU) No 236/2012
ESMAR6	Regulation (EU) No 1095/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Securities and Markets Authority), amending Decision No 716/2009/EC and repealing Commission Decision 2009/77/EC
MiFIR7	Regulation (EU) No 600/2014 of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Regulation (EU) No 648/2012

[3] ELI: http://data.europa.eu/eli/reg/2013/575/oj

[4] ELI: http://data.europa.eu/eli/reg/2014/909/oj

[5] ELI: http://data.europa.eu/eli/reg/2023/2845/oj

[6] ELI: http://data.europa.eu/eli/reg/2010/1095/oj

[7] ELI: http://data.europa.eu/eli/reg/2014/600/oj

MiFID II8	Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU (recast)
SFTR9	Regulation (EU) 2015/2365 of the European Parliament and of the Council of 25 November 2015 on transparency of securities financing transactions and of reuse and amending Regulation (EU) No 648/2012
Commission Delegated Regulation (EU) 2017/38910	Commission Delegated Regulation (EU) 2017/389 of 11 November 2016 supplementing Regulation (EU) No 909/2014 of the European Parliament and of the Council as regards the parameters for the calculation of cash penalties for settlement fails and the operations of CSDs in host Member States
RTS 211	Commission Delegated Regulation (EU) 2017/583 of 14 July 2016 supplementing Regulation (EU) No 600/2014 of the European Parliament and of the Council on markets in financial instruments with regard to regulatory technical standards on transparency requirements for trading venues and investment firms in respect of bonds, structured finance products, emission allowances and derivatives
RTS on Settlement Discipline12	Commission Delegated Regulation (EU) 2018/1229 of 25 May 2018 supplementing Regulation (EU) No 909/2014 of the European Parliament and of the Council with regard to regulatory technical standards on settlement discipline
Commission Delegated Regulation (EU) 2021/7013	Commission Delegated Regulation (EU) 2021/70 of 23 October 2020 amending Delegated Regulation (EU) 2018/1229 concerning the regulatory technical

[8] ELI: http://data.europa.eu/eli/dir/2014/65/oj

[9] ELI: http://data.europa.eu/eli/reg/2015/2365/oj

[10] ELI: http://data.europa.eu/eli/reg_del/2017/389/oj

[11] ELI: http://data.europa.eu/eli/reg_del/2017/583/oj

[12] ELI: http://data.europa.eu/eli/reg_del/2018/1229/oj

[13] ELI: http://data.europa.eu/eli/reg_del/2021/70/oj

standards on settlement discipline, as regards its entry into force

List of acronyms

BD	Business day
CP	Consultation Paper
CSD	Central Securities Depository
ESMA	European Securities and Markets Authority
EC	European Commission
EEA	European Economic Area
ETF	Exchange traded fund
EU	European Union
DvP	Delivery versus Payment
FoP	Free of Payment
ISD	Intended settlement date
ITS	Implementing Technical Standards
LMFPs	Late Matching Fail Penalties
MBI	Mandatory buy-in
MMI	Money Market Instrument
NCA	National Competent Authority
RTS	Regulatory Technical Standards
SSS	Securities Settlement System
SFT	Securities Financing Transactions
TA	Technical Advice
T2S	TARGET2-Securities
UCITS	Undertakings for Collective Investments in Transferable Securities

2 Introduction

1. CSDR was published in the Official Journal on 28 August 2014, and entered into force on 17 September 2014. CSDR was amended by CSDR Refit, which entered into force on 17 January 2024.

2. Articles 6, 7 and 7a) of CSDR, as amended by CSDR Refit, include a set of measures to prevent and address failures in the settlement of securities transactions (settlement fails), commonly referred to as settlement discipline measures. They consist of reporting requirements, cash penalties for CSD participants in case of settlement fails, and mandatory buy-ins where a CSD participant fails to deliver the securities within a fixed extension period (which, according to CSDR Refit, have become a last resort measure where the rate of settlement fails in the EU is not improving and is presenting a threat to financial stability).

3. In view of the preparation of an amendment of Commission Delegated Regulation (EU) 2017/38914, the European Commission (EC) has submitted three requests for technical advice to ESMA covering the following topics on cash penalties:

a) alternative parameters, when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available (request submitted on 13 December 2022);

b) treatment of historical reference data for the calculation of late matching fail penalties (request submitted on 15 May 2023);

c) alternative methods for calculating cash penalties, including progressive penalty rates (request submitted on 28 August 2023).

4. Initially, the EC asked ESMA to submit the technical advice on the three topics by 30 September 2024. However, given the recent developments related to the potential move to a shorter settlement cycle (T+1) in the EU, the EC and ESMA have agreed that it would be useful to align the delivery of the ESMA Technical Advice on the penalty mechanism with the ESMA Report on T+1.

[14] Commission Delegated Regulation (EU) 2017/389 of 11 November 2016 supplementing Regulation (EU) No 909/2014 of the European Parliament and of the Council as regards the parameters for the calculation of cash penalties for settlement fails and the operations of CSDs in host Member States.

ESMA European Securities and Markets Authority

5. The mandates mentioned above set out the principles which ESMA was invited to take into account when developing its advice, including proportionality and coherence within the regulatory framework of the Union. ESMA was invited to widely consult market participants in an open and transparent manner and to take into account the resulting opinions in its advice. ESMA was also invited to justify its advice by providing a quantitative and qualitative cost-benefit analysis of all the options considered and proposed.

6. ESMA has developed its draft technical advice having due regard to the principle of proportionality and being mindful about the possible costs the obligations they contain would create for market participants. The input from stakeholders has helped ESMA in finalising the technical advice and the relevant impact assessment.

3 Alternative parameters, when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available

3.1 Background

EC request for ESMA technical advice (13 December 2022) ESMA is invited to provide technical advice to assist the Commission in amending Delegated Regulation (EU) 2017/389. In particular, this advice should specify which alternative rate or methodology should be applied in the calculation of cash penalties for settlement fails caused by a lack of cash where no overnight credit rate charged by the central bank issuing the settlement currency exists. In order to ensure a deterrent effect of cash penalties and incentivise timely settlement by failing participants, the penalty rate should reflect the borrowing costs for that currency. ESMA should ensure that the applicable interest rate is set such that the level of cash penalties provides incentives to failing participants to promptly settle failed transactions, without endangering the integrity of the EU capital market. Simultaneously when defining the alternatives their impact on the level of penalties and on the market should be considered. In particular, the proposed rate should not lead to further fragmentation of the single market for capital. Moreover, considering the automation of calculation of cash penalties the proposed alternative rate should be easy to source and compute. The Delegated Regulation notes that the most appropriate benchmark of borrowing costs in the calculation of a penalty rate is the official interest rate of the central bank issuing the settlement currency. Other potential substitute interest rates exist on the national and EU capital markets. Although some of them are set without the involvement of a central bank, they reflect the borrowing costs on the commercial inter-bank market and are used in several securities settlement systems. The technical advice should reflect upon the relevance of these proxy rates for the calculation of cash penalties in case of settlement fails caused by a lack of cash in light of the requirements of the Delegated Regulation (in particular Recital 12 and point 8 of the Annex).

7. Commission Delegated Regulation (EU) 2017/389 specifies the parameters and methodology for the calculation of the level of cash penalties that CSDs will impose on and collect from the failing participants in their securities settlement systems. Specifically, Article 2 states that “The level of cash penalties referred to in the third subparagraph of Article 7(2) of Regulation (EU) No 909/2014 for settlement fails of

ESMA - 201-203 rue de Bercy - CS 80910 - 75589 Paris Cedex 12 - France - Tel. +33 (0) 1 58 36 43 21 - www.esma.europa.eu 12

transactions in a given financial instrument shall be calculated by applying the relevant penalty rate set out in the Annex to this Regulation to the reference price of the transaction determined in accordance with Article 3 of this Regulation”. Accordingly, the Annex to the Delegated Regulation specifies penalty rates applicable to settlement fails. In the case of settlement fails due to a lack of cash (point 8 of the Annex) the applicable rate should be the official interest rate for overnight credit charged by the central bank issuing the settlement with a floor of 0 (“zero”). For instance, in the case of Euro-settled transactions this would be the rate on the marginal lending facility, which is the interest rate banks pay when they borrow money overnight from the European Central Bank (ECB).

8. Neither CSDR nor the Commission Delegated Regulation (EU) 2017/389 provide a common definition of the overnight credit rate to be applied by CSDs or an alternative proxy interest rate for calculating the cost of borrowing in case a central bank overnight lending facility does not exist for the settlement currency. This appears to be the case of Bulgaria and Denmark.

9. The current version of Commission Delegated Regulation (EU) 2017/389 was based on ESMA Technical Advice15. It should be mentioned that, when ESMA consulted on the draft Technical Advice, the fact that some central banks do not have an overnight lending facility was not raised.

10. Alternative calculation methodologies or rates have been proposed, as evidenced by the ECSDA CSDR Penalties Framework16. Please see table 17 published by ECSDA with the cash penalty rates.

11. At the same time, ESMA would like to point out that, according to Article 7(5) of CSDR, the Commission is empowered to supplement such Regulation by adopting delegated acts specifying parameters for the calculation of a deterrent and proportionate level of the cash penalties based on all of the following:

a. asset type;

b. liquidity of the financial instrument;

c. type of transaction;

[15] https://www.esma.europa.eu/sites/default/files/library/2015/11/2015-esma-1219_-_final_report_csdr_ta_incl_cba_for_ec.pdf

[16] https://ecsda.eu/wp-content/uploads/2021/10/2021_10_05_ECSDA_CSDR_Penalties_Framework.pdf

[17] 2021_10_05_ECSDA_Currencies_Discount_rates_26_08_20.xlsx (live.com)

d. duration of the settlement fail.

12. According to the above-mentioned provision, when specifying these parameters, the Commission will need to take the level of settlement fails per class of financial instruments and the effect that low or negative interest rates could have on the incentives of counterparties and on settlement fails into account. The parameters used for the calculation of cash penalties will need to ensure a high degree of settlement discipline and the smooth and orderly functioning of the financial markets concerned.

13. The Commission will also have to review the parameters for the calculation of the level of the cash penalties on a regular basis and at least every four years in order to reassess the appropriateness and effectiveness of the cash penalties in achieving a level of settlement fails in the Union deemed to be acceptable having regard to the impact on the financial stability of the Union.

3.2 Feedback statement

3.2.1 Alternative parameters proposals for penalty rates applied in case of settlement fails due to lack of cash (Q1, Q2, Q3)

Q1: Do you agree with ESMA’s proposal? Which Option is preferable in your view? Please also state the reasons for your answer.

Summary of responses

14. Several respondents stated that it is premature to review the penalty mechanism, as it only entered into force recently.

15. As a result, they generally support option 3 since it is most closely aligned with the current process in Denmark. Option 3 will not require a functional change to the T2S penalty mechanism, as opposed to option 1 and 4.

16. Similarly, many respondents would, in principle, support option 4 (without adopting the progressive penalty rates), because it is the simplest model. Progressive penalty rates are not supported by the respondents.

17. Most respondents point out that option 4 with progressive penalty rates requires a structural system change with an impact on CSDs, custodians and other market participants and on market liquidity and pricing. This would lead to additional costs for investors in the secondary market and issuers in the primary market. The rationale for such a change is unclear since there are not many late fails (after 4 business days less than 5 % of fails).

18. One respondent favours option 4, if higher rates would apply from the day after the relevant ISD. This should only be introduced if T2S is able to ensure a FIFO criterion. The oldest transaction should be settled when cash becomes available, but this would require a change in T2S. Incorrect application of FIFO should be included in the appeal procedure.

19. Respondents argued that the future move to T+1 in the EU will mean more settlement fails. Therefore, an introduction of progressive rates with option 4 would be even more detrimental to market participants.

20. One respondent was of the view that options 1, 2 and 3 fail to address issues related to “lack of cash” fails in non-euro currencies. Therefore, option 4 without progressive rates was preferred.

21. As for the specific case of DKK, in absence of a central bank intraday credit rate, there was broad support for option 3, which is in line with the current model. This option does not require any change of the T2S penalty mechanism.

Q2: Do you have other suggestions? If yes, please specify and provide arguments.

Summary of responses

22. Respondents reaffirmed that they suggest retaining the current methodology for DKK and BGN and to avoid introducing progressive rates in general.

23. One respondent suggested an exemption or relief for market makers because they are not responsible for the fails.

Late matching fail penalties resulting from delayed matching of settlement instructions after a CSD’s “cut off time” have the potential to punish the wrong party within a chain of failed settlements.

Q3: Do you agree with the approach followed for the Option you support to incorporate proportionality in the Technical Advice? If not, please provide an indication of further proportionality considerations, detailed justifications and alternative wording as needed.

Summary of responses

24. Respondents generally supported the current approach.

3.3 ESMA’s assessment

25. Option 4 seems to be the least preferable option for most of the respondents. As for the specific case of DKK with an absence of a central bank intraday credit rate, there is broad support for option 3 which is in line with the current model. This option does not require any change of the T2S penalty mechanism.

26. No common calculation method or agreement on the variables used to calculate an alternative rate has been developed, when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available. Currently different settlement currencies use different domestic rates or a combination of a domestic benchmark rate and spread of key ECB interest rates. Please see the examples below.

27. It should be mentioned that Bulgaria is expected to join the Eurozone in 2024/2025 (exact date to be confirmed), therefore it appears that soon the only EEA jurisdiction without an official interest rate for overnight credit for the national currency will be Denmark.

A) Bulgaria

28. BG FSC has confirmed the following approach which the Bulgarian Central Depository AD (CDAD) and the Bulgarian National Bank (BNB) implement due to the lack of an official interest rate for overnight credit in Bulgarian Leva.

29. The penalty rate (PR) is determined as the base interest rate (BIR) published monthly by the BNB, plus the spread calculated as difference between ECB marginal lending facility rate (MLFR) and the ECB Main refinancing operations rate (MROR). Thus, the applicable formula is: PR = BIR + (MLFR – MROR).

30. The PR in January 2023 was 1.67%. This figure is calculated by taking the BIR = 1.42%, as of 1 January 2023 (it was zero until 1 October 2022) and the spread between MLFR = 2.75 % and the MROR = 2.50% as of 21 December 2022. Thus, PR = 1.42 + (2.75 – 2.50) = 1.42 + 0.25 = 1.67% per year.

31. The method described above was proposed as an interim solution by the BNB and afterwards discussed with EC representatives, BNB, the Bulgarian Ministry of Finance and FSC. As a follow-up of the above-mentioned discussion, the method was adopted by both BG CSDs (in Ordinance No. 31 of BNB) and CDAD (in Article 30 of CDAD Rules and Regulations).

32. Furthermore, the determination of the domestic rate BIR is completely market-based. BNB calculates and publishes the LEONIA Plus index – LEV Overnight Index Average, which is determined as the average-weighted interest rate on all uncollateralised overnight deposits in Bulgarian Leva provided on the interbank market in Bulgaria. LEONIA Plus is calculated and published on a daily basis. The BIR is the average of all daily values of LEONIA Plus for the previous month. BIR is determined and published on the first day of each month and applies for the same month.

B) Denmark

33. DK FSA has confirmed the approach used by VP Securities, the Danish CSD, in the absence of an official interest rate for overnight credit in Denmark. Danmarks Nationalbank conducts a fixed exchange rate policy against the Euro. This means that the value of the Danish Krone is kept stable against the Euro. Danmarks Nationalbank does this by way of monetary policy. In a fixed exchange rate regime, monetary-policy interest rates are reserved for managing the exchange rate. Therefore, interest rates are kept relative to those of the ECB.

34. For the Euro currency, the ECB’s “marginal lending facility rate” is used to calculate penalties. Since Danmarks Nationalbank does not offer an overnight credit facility there is no interest rate for overnight credit. Instead, there is an interest rate for weekly credit which is equal to the ECB’s “main refinancing operations” rate (MROs). According to DK FSA, the interest rate of Danmarks Nationalbank would be too low compared to the ECB’s marginal lending facility rate, since it is weekly instead of overnight. A proxy used so far is based on Danmarks Nationalbank’s weekly lending rate plus the spread calculated as the difference between the ECB marginal lending facility rate (MLFR) and the ECB deposit facility rate. This proxy rate is then divided by 360, with a floor of zero. Such approach is linked to the fixed exchange rate policy.

35. An example of the proxy used by the Danish CSD to calculate the penalties is: Weekly lending rate + (ECB marginal lending facility interest rate – ECB deposit facility rate) = 3.10% per year + (4.25% per year – 3.50% per year) = 3.85% per year.

3.4 ESMA’s technical advice

36. ESMA believes that the penalty rate applied for a settlement fail due to lack of cash should ensure that it is cheaper to borrow cash to settle the transaction than to pay the penalties and obtain interest on the unpaid cash.

37. ESMA recommends the approach mentioned below when calculating cash penalties for settlement fails due to lack of cash, for the currency/currencies without an official interest rate for overnight credit.

38. Generally, the model is linked to the fixed exchange rate policy. This is the case of Denmark (and Bulgaria because it is preparing to join the Eurozone).

39. The interest rates are used to keep the exchange rates stable against the Euro and therefore there is no overnight credit rate.

40. Because the rates are always kept stable to the rates of the ECB, it is possible to calculate what an overnight credit rate would be if there was one. The calculation is performed by taking the current proxy of the weekly credit rate (the proxy of the ECB main refinancing operation rate) and adding the spread between the ECB overnight margin facility lending rate and the ECB weekly deposit facility lending rate.

41. As a result, the proxy is designed to ensure that the rate used to calculate the fines always fulfills the purpose that it should be more advantageous to borrow and pay on time instead of paying the penalty.

42. This approach takes the proportionality principle into account, as it does not go beyond what is necessary to achieve the objective of the cash penalties regime under CSDR. It is simple and avoids excessive financial, administrative or procedural burdens for the CSDs and the central banks concerned.

ESMA’s proposed approach

43. In the absence of an overnight interest credit rate due to the monetary policy of the central bank issuing the settlement currency, other comparable interest rates of the ECB and the relevant central bank could be used to calculate a proxy which a CSD can use to calculate the cash penalties due to lack of cash.

44. That alternative method must always ensure that the cash penalty rate costs more than it would cost to borrow the cash and pay for the securities on time. It is a precondition for this alternative method that the interest rates of the ECB and the relevant central bank are comparable. The proxy is calculated by the central bank issuing the settlement currency and must be recalculated whenever either the interest rates of the ECB and/or the interest rates of the central bank issuing the settlement currency are subject to changes.

Example:

STEP 1: identification of comparable interest rates between the ECB and the central bank.

These interest rates could be the ECB main refinancing operations rate (MRO) and the corresponding weekly credit interest rate of the central bank.

STEP 2: the addition to the corresponding weekly credit interest rate of the central bank of the spread between the ECB deposit facility rate and the ECB marginal lending facility rate.

This spread is added because a weekly credit interest rate will typically be lower than an overnight credit interest rate. Adding the spread will compensate for this difference, so that the interest rate used to calculate the cash penalties is as close as possible to the official interest rate for overnight credit of the ECB.

4 Treatment of historical reference data for the calculation of late matching fail penalties

4.1 Background

EC request for ESMA technical advice (15 May 2023) ESMA is invited to provide technical advice to assist the Commission in amending Delegated Regulation (EU) 2017/389. This advice should specify how to deal with reference data accumulation caused by the need to calculate LMFPs, in particular by suggesting appropriate methods to calculate settlement fails penalties and handle reference data underlying transactions that are matched after the ISD.

45. Article 7(2) of CSDR requires CSDs to apply cash penalties to participants that cause settlement fails, which shall be calculated on a daily basis for each business day that a transaction fails to be settled after its intended settlement date, but no longer than the actual settlement day.

CSDR Article 7

2. For each securities settlement system it operates, a CSD shall establish procedures that facilitate settlement of transactions in financial instruments referred to in Article 5(1) that are not settled on the intended settlement date. These procedures shall provide for a penalty mechanism which will serve as an effective deterrent for participants that cause settlement fails.

Before establishing the procedures referred to in the first subparagraph, a CSD shall consult the relevant trading venues and CCPs in respect of which it provides settlement services.

The penalty mechanism referred to in the first subparagraph shall include cash penalties for participants that cause settlement fails (‘failing participants’). Cash penalties shall be calculated on a daily basis for each business day that a transaction fails to be settled after its intended settlement date until the end of a buy-in process referred to in paragraph 3, but no longer than the actual settlement day. The cash penalties shall not be configured as a revenue source for the CSD.

46. The parameters for the calculation of cash penalties are defined in the Commission Delegated Regulation (EU) 2017/389:

Article 2 - Calculation of cash penalties

The level of cash penalties referred to in the third subparagraph of Article 7(2) of Regulation (EU) No 909/2014 for settlement fails of transactions in a given financial instrument shall be calculated by applying the relevant penalty rate set out in the Annex to this Regulation to the reference price of the transaction determined in accordance with Article 3 of this Regulation.

Article 3 - Reference price of the transaction

2. The reference price referred to in paragraph 1 shall be used to calculate the level of cash penalties for all settlement fails, irrespective of whether the settlement fail is due to a lack of securities or cash.

47. The CSDR settlement discipline regime implies that LMFPs must be calculated for settlement fails with an ISD for any point in time in the past as of 1 February 2022. This poses a challenge for any IT system with finite resources: to calculate settlement fails for any given day in the past means that the related historical reference data must be kept available and made use of by the system where the amount of reference data is gradually increasing every business day (with no possibility for historical reference data deletion/removal).

48. This accumulation of historical reference data may have an impact on all EU CSDs as well as, most notably, on TARGET2-Securities (T2S) where the accumulation of past data over time could degrade the performance of the system.

49. In light of the above, the EC has asked ESMA to suggest a possible amendment to Commission Delegated Regulation (EU) 2017/389. Such amendment should clarify the calculation method for LMFPs that prevents the accumulation of reference data over time and ensures the efficient operation of securities settlement systems.

4.2 Feedback statement

4.2.1 Accumulation of reference data issue (Q5)

Q5: As a CSD, do you face the issue of accumulation of reference data related to Late Matching Fail Penalties (LMFPs), that may degrade the functioning of the securities settlement system you operate? If yes, please provide details, including data where available, in particular regarding the number and value of late matching instructions, as well as for how many business days they go in the past from the moment they are entered into the securities settlement system, and the percentage they represent compared to the overall number and value of settlement fails on a monthly basis (please use as a reference the period June 2022 – June 2023).

Summary

50. The CSDs responded that accumulation of reference data related to late matching fail penalties does not degrade the functioning of their systems. However, those CSDs that use the T2S penalty mechanism acknowledge that the Eurosystem have identified such a possible performance issue.

51. Some of the CSDs from the T2S environment fully supported the demand of the T2S penalty mechanism.

Conclusions/Actions to be taken

52. There is a general acknowledgement of a possible performance issue for the users of the T2S penalty mechanism, i.e. the CSDs that use the T2S penalty mechanism as service provider.

4.2.2 Causes of late matching (Q6)

Q6. What are the causes of late matching? How can you explain that there are so many late matching instructions? What measures could be envisaged in order to reduce the number of late matching instructions?

Summary

53. One respondent presented the general view, to prevent LMFPs, that instructions should be entered and matched at the CSD on trade date to enable the identification and resolution of any discrepancies prior to ISD.

54. Several respondents affirmed that their data does not indicate so many late matching instructions contrary to ESMA’s statement.

55. Many respondents suggested the following measures to reduce the number of unmatched settlement instructions:

further use of partial release/settlement. This could help to avoid re-instructing a previous settlement instruction and avoid late matching;
encouraging clients to send instructions as soon as possible and improving the information sharing between counterparties;
excluding market claims and transformations from the scope of penalties application as well as settlement fails reporting;
place of settlement should be a mandatory field in the matching process;
a process that includes affirmation and instruction of trades on trade date similar to the US market;
introducing a requirement for confirmation and allocation in a machine-readable format;
the use of UTI and SSI solutions could have substantial effect to reduce failure to match on time;
increasing support in the market for sending and receiving allegements and so-called “close” or “near”- matching reporting could be beneficial;
electronic processing of allocation-confirmation;
further use of tools for pre-matching, standard settlement instructions, UTI tracking/monitoring to reduce late matching issues;
auto-partialing across all CSDs and custodians;
use of an SSI repository.

Conclusion/Actions to be taken

56. The responses did not present a clear picture of whether late matching is a general problem and the magnitude of the problem.

57. However, stakeholders requested more market standards to ensure more data is in place on trade day.

4.2.3 Establishment of reference data used for the calculation of cash penalties (Q7, Q8, Q9, Q10, Q11, Q12)

Q7: Do you agree with ESMA’s proposal to establish a threshold beyond which more recent reference data shall be used for the calculation of the related cash penalties to prevent the degradation of the performance of the systems used by CSDs? Please also state the reasons for your answer.

Summary

58. Respondents generally agreed with the proposal to allow CSDs to use a threshold. They agreed it should not necessarily be mandatory but left at the discretion of each CSD.

59. It was also acknowledged that a threshold is necessary for CSDs that use the T2S penalty mechanism.

60. Even though respondents recognised the challenges of T2S and support a pragmatic solution through the recourse of a threshold, several of them also called for a more sustainable long-term solution.

Conclusions/Actions to be taken

61.Most of the respondents are in favour of setting a threshold.

Q8: Do you agree with the threshold of 92 business days or 40 business days in order to prevent the degradation of the performance of the systems used by CSDs? Please specify which threshold would be more relevant in your view:

a) 92 business days;

b) 40 business days;

c) other (please specify).

Please also state the reasons for your answer and provide data where available, in particular regarding the number and value of late matching instructions that go beyond 92 business days, 40 business days in the past or another threshold you think would be more relevant, and the percentage they represent compared to the overall number and value of settlement fails on a monthly basis (please use as a reference the period June 2022 – December 2023).

Summary

62. There was broad support to introduce a 40-day threshold, as long as it is left to the CSDs to decide if they need to introduce it.

Conclusions/Actions to be taken

63. A 40-day threshold could be introduced.

Q9: Do you agree that the issuer CSD for each financial instrument shall be responsible for confirming the relevant reference data to be used for the related penalties calculation? Please also state the reasons for your answer.

Summary

64. Several respondents agreed that the issuer CSDs should be responsible for confirming the relevant data to be used for the calculation of cash penalties.

65. However, most CSDs disagreed with the proposal, as they considered changing the current reference data sourcing as unnecessary.

66. Respondents suggested ESMA to centrally provide and publish all reference data needed to calculate penalties in a single database accessible to all stakeholders.

Conclusions/Actions to be taken

67. There are mixed views concerning the introduction of a requirement addressed to issuer CSDs.

Q10: In your view, where settlement instructions have been matched after the intended settlement date, and that intended settlement date is beyond the agreed number of business days in the past, the use of more recent reference data (last available data) for the calculation of the related cash penalties should be optional or compulsory? Please also state the reasons for your answer.

Q11: Do you have other suggestions? If yes, please specify, provide drafting suggestions and provide arguments including data where available.

Summary

68. Respondents suggested to create and maintain a central database containing the necessary information to calculate cash penalty amounts.

69. Respondents stressed the importance for a standardised approach across CSDs. They believe that the oldest available historical reference data should be used.

70. Some respondents indicated that CSDs are better placed to decide if reference data older than 40 days should be used.

71. Several respondents suggested that the CSDs should apply a time limit for the receipt of settlement instructions. A similar approach is already in place in the Markets Standards for Corporate Actions Processing and the T2S Corporate Actions Processing with respect to the detection period for market claims.

72. Some respondents underlined that the latest available reference price should be used to calculate any late matching penalties.

Conclusions/Actions to be taken

73. There is no clear way forward as the respondents’ views were not unanimous.

Q12: Do you agree with the approach followed to incorporate proportionality in the Technical Advice? If not, please provide an indication of further proportionality considerations, detailed justifications and alternative wording as needed.

Summary

74. There was a general support to incorporate the principle of proportionality.

Conclusions/Actions to be taken

75. The principle of proportionality should be taken into account.

4.3 ESMA’s assessment

76. ESMA supports the objective of ensuring a proportionate approach by not requiring CSDs to accumulate unlimited reference data in respect of LFMPs in the systems they use and to prevent the degradation of the performance of the systems used by CSDs.

77. As such, ESMA believes that the efficient and smooth operation of securities settlement systems should take precedence over the use of daily reference data for the calculation of LMFPs beyond a certain date in the past. ESMA also acknowledges the importance for CSDs to have predictability regarding the amount of data they need to manage, so that they can plan and develop the capacity of the systems they use accordingly.

78. At the same time, ESMA considers it important to ensure that the number and value of settlement fails for which the calculation of penalties may be impacted are very low, and that this should be a criterion for setting the threshold beyond which recent reference data (last available data) may be used for the calculation of the related cash penalties. Based on an overview of the number and share of LMFPs in T2S across July-October 2023: an average of less than 300 LMFPs monthly (<0.03% of total penalties) apply for business days older than 40 days, and an average of less than 100 LMFPs monthly (<0.008% of total penalties) apply for business days older than 92 days. To be precise, the July-October 2023 monthly average in T2S for 92+ days was 77 penalties. As a comparison, the monthly average in T2S for 40+ days was 279 penalties.

79. ESMA also believes that it is important that CSDs which are part of an interoperable link should establish arrangements for determining the relevant reference data to be used for the related penalties calculation. Currently, this would be the case of CSDs participating in T2S and for the interoperable link (“the Bridge”) between Euroclear Bank and Clearstream Banking Luxembourg.

80. Last but not least, ESMA would like to highlight that CSDs would still need to archive the reference data related to the calculation of penalties to ensure compliance with the relevant record keeping requirements under Article 29(1) of CSDR, according to which a CSD shall maintain, for a period of at least 10 years, all its records on the services and activities.

4.4 ESMA’s technical advice

81. ESMA’s advice is to amend Commission Delegated Regulation (EU) 2017/389 as per the proposal included below (please see in particular the added paragraph 3 in Article 3).

82. Given the stakeholders’ input, ESMA believes that 40 business days is an adequate threshold beyond which more recent reference data could be used for the calculation of cash penalties, bearing in mind the proportionality principle, in order not to go beyond what is necessary to achieve the objective of preventing the accumulation of reference data over time and ensuring the efficient operation of securities settlement systems, while also enabling the application of effective and deterrent penalties. ESMA also considers that CSDs that can keep and have access to older historical reference data should be able to use the respective data for the calculation of cash penalties for LMFPs going beyond 40 business days in the past.

Example:
LMFP to be computed with ISD 43 days in the past.
For penalties with ISD within 40 business days in the past → T2S will use the correct historical reference price.
For penalties with ISD beyond 40 days in the past → T2S will use the oldest available historical reference price, which is the price of the 40th day in the past. It will apply it for penalties corresponding to business day 41, 42, 43 in the past.
In case a CSD, using a different penalty mechanism, has older data available e.g. 92 days, it should use it and not be bound by the 40 business days threshold. Thus, in the example above, it will use the correct historical price.
While, if the CSD had to compute a LMFP with ISD of more than 92 business days in the past, it would use the oldest available historical reference price, i.e. the price of the 92nd business day in the past, also for penalties corresponding to business day 93, 94, 95, etc.

Proposed amendments to Commission Delegated Regulation (EU) 2017/389:

Article 2 - Calculation of cash penalties

Article 3 - Reference price of the transaction

3. Where settlement instructions have been matched after the intended settlement date, and that intended settlement date is within 40 business days in the past from the matching date, the actual daily reference price shall be used for the calculation of the related cash penalties.

By way of exception, where settlement instructions have been matched after the intended settlement date, and that intended settlement date is beyond 40 business days in the past from the matching date, the oldest available reference price may be used for the calculation of the related cash penalties.

In the cases mentioned in the second subparagraph, CSDs which are part of an interoperable link shall establish arrangements for determining the relevant reference price to be used for the related penalties calculation.

In the cases mentioned in the second subparagraph, CSDs shall communicate to their competent authorities the procedure determining the oldest available reference price they may use for the calculation of the related cash penalties.

5 Alternative methods for calculating cash penalties, including progressive penalty rates

5.1 Background

EC request for ESMA technical advice (28 August 2023) The Commission asks ESMA to suggest a possible amendment to Commission Delegated Regulation (EU) 2017/389. The Agency should assess the effectiveness and proportionality of the current penalty mechanism and propose, if justified, changes to the structure or severity of the mechanism and consider alternative methods for calculating cash penalties, including by introducing progressive penalty rates. In drafting its technical advice, ESMA should consider how the changing interest rate environment, including negative interest rates, affect a participant’s incentive to fail and how this could be mitigated. Furthermore, ESMA should reflect on the need for further flexibility with regards to penalties for settlement fails imposed on illiquid financial instruments. The proposed amendments to the structure and severity of the mechanism should effectively discourage settlement fails, incentivise their rapid resolution and improve settlement efficiency.

83. To respond to the EC request for ESMA technical advice (see above and Annex IV), ESMA has collected views, comments and opinions, as well as data and evidence from stakeholders and market participants on the effectiveness of the current penalty mechanism in discouraging settlement fails and incentivising their rapid resolution.

84. ESMA provided a preliminary assessment covering the effectiveness and proportionality of the current penalty mechanism, on the introduction of progressive penalty rates, the impact of changing interest rates, further flexibility in relation to illiquid financial instruments, automation of calculation of penalties, and ad-hoc measures for participants with high settlement fail rates.

85. Based on the preliminary assessment, ESMA proposed two main options for the revision of the penalty mechanism: as Option 1, the introduction of progressive penalty rates based on the current types of fails and a new type for fails due to the lack of Exchange Traded Funds (ETFs); and as Option 2, the introduction of progressive penalty rates with streamlined asset types and convexity.

86. Both options entailed a voluntarily significant increase of penalty rates to elicit stakeholders’ feedback. The CP also included numerous questions on additional considerations to simplify the penalty mechanism and on the costs and benefits of potential changes.

87. The EC will also be mandated to review the parameters for the calculation of the level of the cash penalties on a regular basis and at least every four years. Such regular reviews provide the opportunity for ESMA to further explore some of the suggestions outlined in the CP, even if they are not pursued in the short term.

5.2 Feedback statement

5.2.1 Functioning of the current penalty mechanism

Impact of current penalty mechanism (Q15, Q16, Q 20, Q21)

Q15: Impact of the penalty mechanism on reducing settlement fails

Summary

88. Most respondents agreed that the cash penalties mechanism in 2022 has positively impacted settlement efficiency in the EU, prompting the industry to improve post-trade processes. However, they believe it is too early to draw conclusions about its effect on the number of fails, due to the short observation period, that does not allow for a clear impact assessment on the settlement fail rates, and issues with data quality and consistency.

89. In particular, some respondents have highlighted that the methodology used to measure settlement fails has changed since the introduction of the cash mechanism, leading to challenges in comparing the results over time. Respondents also noted that the presentation of settlement efficiency rates in the latest ESMA Report on Trends, Risks and Vulnerabilities18 (TRV) at the time of publication of the CP has differed from the one used to produce the previous reports, and that in applying the current CSDR methodology, CSDs have introduced differences in the fail reporting parameters, leading to divergences even between CSDs in the same group.

[18] ESMA50-1389274163-2681 TRV 2, 2023 Risk Monitor (europa.eu) (Note: this was the latest ESMA TRV Report at the time of the public consultation on the draft Technical Advice on CSDR penalty mechanism.)

90. Very few respondents provided data on their own settlement efficiency rate and most respondents only made reference to T2S data19, public annual disclosure of CSDs and ESMA TRV reports, showing a slight decrease of the number of settlement fails (in value and in volume), in particular for equities (except ETFs). For instance, many respondents have highlighted in the T2S data that between January 2022 (the month before CSDR penalties went live) and January 2024, the settlement fails rates, in terms of value, fell from 6.6% to 3.8%.

91. According to the observations of one respondent, since the entry into application of the cash penalty mechanism, the total number of penalties has decreased from approximately 130,000 penalties per month to around 80,000 penalties. Moreover, the proportion of penalties related to Late Matching Fail Penalty (LMFP) transactions has decreased from approximately 13% to about 5%.

92. Based on a review of the available data from 10 CSDs’ annual public disclosure of settlement fails, one respondent observed that settlement efficiency has generally increased across these 10 CSDs, with an overall improvement of 3.96 % of the rate of settlement fails based on the value of settlement instructions.

93. Most respondents considered that the improvement of settlement efficiency cannot be solely attributed to the application of cash penalties. Respondents suggested that the slight improvement of the settlement efficiency in the EU over the last years could be explained by the following factors: mainly the evolution of the monetary policy (e.g. changes in interest rates, reduction of central bank purchase programs), but also other factors such as the use of partial settlement and other CSDs functions (e.g. “hold and release” function), and the efforts from market participants to enhance their post-trade processes with the application of CSDR and the settlement discipline regime beyond the cash penalties (e.g. improvements in the use of Standing Settlement Instructions (SSIs) and SWIFT message). Respondents also highlighted that it is difficult to determine the impact of each factor separately.

94. While some respondents considered that the cost of failing is sufficiently high in itself to discourage market players from failing, a few respondents considered that the level of cash penalties rates is too low to have an impact on settlement efficiency.

[19] TARGET2-Securities Annual Report 2022 (europa.eu)

Some respondents also considered that an important part of the settlement fails are due to structural causes, more than to behavioural causes and thus, that cash penalties may not have an impact on certain categories of settlement fails (for instance, settlement of ETFs).

95. One respondent from the asset management industry indicated that, in their view, there has not been any improvement in terms of settlement efficiency in their industry, mainly because settlement fails are due to the sell-side not delivering the securities. This respondent also considered that cash penalties represent an additional burden on asset managers, leading to increased operational costs, as they have to monitor every fail in order to claim it from their custodian / fiduciary function.

ESMA’s assessment / Conclusions / Actions to be taken

96. ESMA understands that the short observation period and the inconsistency in the data make it difficult to accurately assess the impact that the introduction of cash penalties may have had on the level of settlement fails in the EU. ESMA also recognises that a strict correlation between the improvement in terms of settlement efficiency over the last few years and the introduction of cash penalties may not have been established so far.

97. However, ESMA notes that most respondents welcome the introduction of the cash penalty mechanism as an incentive for the industry to enhance the efficiency of their settlement efficiency practices, alongside external factors. ESMA therefore believes that a moderate increase of cash penalties rates (using as reference the average securities lending and borrowing rates) could ultimately lead to an improvement in terms of settlement efficiency.

98. A general decrease of settlement fail rates can be observed in the indicators presented in ESMA’s successive TRV Reports, either with lesser granularity until 202320 (corporate bonds, equities, government bonds) or with more granularity from 202421 (bonds, sovereign, UCITS, equity, MMI, ETF).

[20] ESMA50-165-2406_TRV 1-23 Statistical annex.pdf (europa.eu) and ESMA50-1389274163-2681 TRV 2, 2023 Risk Monitor (europa.eu)

[21] ESMA50-524821-3107 Report on Trends, Risks and Vulnerabilities, n.1, 2024 (europa.eu) and ESMA50-524821-3444 Trends, Risks and Vulnerabilities (TRV) Report, No. 2, 2024 (europa.eu)

99. Regarding the comments made by respondents on the different methodology for reporting and calculating settlement efficiency rates as reflected in ESMA’s TRV Reports, ESMA would like to point out that this is due to the introduction of the more granular methodology under Article 7(1) of CSDR (applicable as of 1 February 2022 following the entry into force of the RTS on Settlement Discipline). Given the need to run data quality checks on the new CSDR reports on CSDs’ settlement efficiency, ESMA has started using the new datasets beginning with the TRV Report 2, 202322 . Prior to that, the indicators on settlement efficiency published in the ESMA TRV Reports were based on voluntary reports submitted by CSDs to their NCAs, which in turn shared them with ESMA.

Q16: Deterrence and proportionality of the penalty mechanism

Summary

100. Respondents expressed mixed views on whether the current penalty mechanism works as a deterrent and is proportionate and effectively discourages settlement fails and incentivises their rapid resolution.

101. The majority of respondents considered the current cash penalties mechanism to be a deterrent and effectively discouraging settlement fails and pushed back on the introduction of substantial changes or additional complexities. Many of these respondents believe that the remaining fails are mainly due to structural causes and that such fails could not be addressed via the application of increased cash penalties.

102. Nonetheless, some of these respondents showed openness to a proportionate increase/recalibration of the cash penalties rates. In addition, around a third of the respondents stated that the current penalties rates are too low, and a re-calibration could be considered, subject to certain precautions e.g. for illiquid financial instruments.

103. Many respondents also highlighted that the costs associated with borrowing the securities are generally higher than the costs associated with paying a penalty due to the failure to deliver securities. As examples, respondents mentioned an average

[22] ESMA50-1389274163-2681 TRV 2, 2023 Risk Monitor (europa.eu)

penalty amount of EUR 80 within the ESES CSDs (Euroclear France, Netherlands and Belgium) in December 2023; 87% of penalties in Denmark ranging between 0-100 DKK for the fourth quarter of 2023; and 80-90% of penalties being below €5 according to a group of banking institutions.

104. Regarding the proportionality of the mechanism, many respondents, notably from the asset management industry, highlighted that the costs of implementation and maintenance of the process necessary to manage cash penalties are high and not necessarily proportionate compared to the low amounts of penalties received.

105. For instance, one trade association representing banking institutions estimated that EUR 5 million were invested in technological and additional staff resources. Respondents from the banking industry also listed examples where the penalty mechanism could lead to delays in the resolution of settlement fails, such as in cases of misaligned CSD batch times, when certain CSD functionalities are missing (e.g. a CSD not offering auto-partial, with lengthier manual instructions leading to latematching fail penalty), or when a mismatch on the place of settlement is not identified and not resolved before the intended settlement date (which could provide incentives for one participant to claim the late matching fail penalty due to rebooking).

106. Similarly, one association representing asset managers highlighted that the costs of implementation and maintenance of the cash penalties management regime are significant. In more detail, the yearly running costs of the CSDR penalty mechanism is estimated at EUR 300k, with EUR 150k in HR costs (regulatory watch, claims, queries, reconciliations), EUR 100k in IT costs (IT feeds, disclosures, documentation updates) and EUR 50k in custodial fees (filtering, pass-on, provision of settlement messages). The same association noted that 80% of cash penalties implying investment funds are to be received from brokers because of the late delivery of the securities, and therefore concluded that the current penalty mechanism is not proportionate and contrary to the interest of end-investors.

107. Most respondents argued against the introduction of any substantial changes to, or further complexity in, the current cash penalties framework.

108. A few respondents also questioned the proportionality of the current scope of the cash penalties, advocating for a reduced scope not covering “internal” transactions and other specific transactions such as transfers between the same participant accounts; portfolio transfers between the same and/or different accounts/participants; market claims/ transformations; investment funds redemptions/subscriptions orders settlement; primary market transactions; and auto-collateralisation operations.

Conclusions / Actions to be taken

109. ESMA notes that respondents concur in signalling that the deterrent effect of cash penalties is reduced if the level of cash penalties is lower than the cost of borrowing the securities in order to avoid settlement fails, and that some respondents consider that the current level of cash penalties is too low.

110. In light of the emphasis respondents put on the significant implementation and maintenance costs, ESMA considers that at this stage, structural changes should be avoided to preserve the proportionality of the cash penalties mechanism.

111. The scope of application of the cash penalties, and the use of additional tools to improve settlement efficiency will be addressed in separate workstreams.

Q20 and Q21: Proportionality of the penalty rates, and available data by asset type

Summary

112. Respondents expressed mixed views on the proportionality of the current penalty rates, with some respondents stating that the penalty mechanism has not been in place for a sufficient time to fully assess the proportionality of the current rates.

113. A third of the respondents consider the current cash penalties rates are proportionate, with some suggestions for a better calibration, including the creation of specific cash penalties rate for ETFs, harmonised methods to calculate penalties for ICSDs, and further explanation on the rationale behind the differences in rates per category.

114. A few respondents expressly indicated that the current cash penalties rate is not high enough and should be increased. Those respondents considered that cash penalties have not yet reached a level of total dissuasiveness and that the current rates, particularly for liquid assets, are too low.

115. Moreover, respondents stated that it would be not proportionate to increase penalty rates i) for illiquid instruments (particularly for bonds for which no repo market is available), ii) in case where fails can be linked to structural causes or root causes upon which participants cannot act and iii) if it may result in a permanent source of income for buyers, deemed more likely to receive cash penalties.

116. While most non-CSD respondents welcomed more transparency on the number and value of cash penalties by asset type, CSD respondents highlighted that this would require a common approach to data collection, underpinned by consistent methodologies to calculate the penalties, and complemented with data on the average duration of settlement fails.

117. As a result, most CSD respondents did not provide specific data or stated they were not able to do so. However, one respondent provided the breakdown of penalties by asset type in a given month and reported an average duration of fails between 1 and 3 days across CSDs.

Conclusions / Actions to be taken

118. ESMA has taken into account the comments received when identifying the asset types for which increased penalty rate may incentivise an improvement of settlement efficiency. ESMA will also look to collect more granular data on penalties by asset type and on the duration of fails in the medium term.

Q45: CSD participants passing on penalties to their clients

Summary

119. Most respondents claimed that penalties are passed on to clients, with only a few of respondents not agreeing with the claim. Some respondents posited that smaller penalties or penalties targeting retail investors are not passed on. One respondent argued that rules on passing-on penalties should be made more prescriptive.

120. Some respondents also presented investment funds as receiving a differential treatment, whereby the fund receives the penalties, but the manager of the fund pays them. These respondents also mentioned specific cases, for example, funds where intermediaries provide contractual settlement with the end investors not affected by late matching or late settlement, or situations where asset managers and their clients are open to foregoing a net credit of CSDR cash penalties if it means foregoing the operational cost of investigating and processing of the penalties and the potential regulatory impact of undue debits hitting a client account.

Conclusions / Actions to be taken

121. ESMA notes the feedback received and the specific cases mentioned, and believes that they may be further explored in the context of potential amendments to the RTS on settlement discipline, which covers the collection and distribution of cash penalties, given that currently there is no requirement for CSD participants to pass on the penalties to their clients.

Root causes and international comparison (Q17, Q44)

Q17: Root causes for settlement fails

Summary

122. Many respondents agreed that settlement failures due to a lack of cash are rare, representing less than 1% of total failures, and typically resolved within hours. One respondent believed that the lack of cash could be explained by the requirement to remove excess funding from T2S/CSD just before the deadline.

123. Regarding settlement fails due to a lack of securities or settlement instructions on hold, the variety of reasons mentioned can be grouped in two categories: on the one hand, the causes of fails that can be attributed to market participants’ behaviour/set-up, and, on the other hand, structural causes of fails that are more attributable to the set-up of market infrastructures and to the limited standardisation.

124. The causes of fails attributable to market participants can be grouped in three sub-categories. The first sub-category covers the reasons explaining why settlement instructions are sent late to the CSD. Many factors can explain the late sending of instructions:

o Limited automation and Straight Through Processing (STP), leading to communication delays and low data quality due to the use of non-standard channel and manual processing;

o Latency in confirming, allocating or booking the transaction/the OTC trade;

o Data quality issues, such as missing reference data required for booking a trade; missing data required to generate and provide onward SSIs; delays related to instrument reference data (e.g. new ISIN not being available); changes in SSIs not being communicated or updated in good time; or instructions rejected by the CSD or custodian due to incorrect format;

o Time zone issues, which may occur when a client is instructing the custodian in its own operating hours rather than the operating hours of the CSD, resulting in missing the cut-off time, or when a client is only able to provide allocations of transaction on T+1, leading to processing delays;

o Workflow management challenges, such as manual booking error or technology issues on the part of the market participant), or delays in opening the account of the client.

125. The second sub-category covers the reasons explaining why settlement instructions do not match at the CSD level. In particular, instructions cannot match at the CSD level where there are divergences in the economic data (e.g. ISIN, nominal, value date, cash amount, incorrect calendar holiday leading to value date mismatches, UNT/FAMT confusion in the quantity fields, etc.), and in the non-economic data, such as place of settlement (although this information is not a mandatory matching criteria at trade level).

126. The third sub-category covers the reasons why a party may not have sufficient securities to deliver. Failure to deliver securities mainly occurs where:

o Inventory/operational internal issues lead to a failure to realign securities between different locations or accounts in sufficient time. For example, the trading party has sufficient securities but the securities are not immediately available for settlement as they are located across the different trading venues / brokers where they have been purchased; o There is a real shortage of securities. For instance, where there is a lack of trading-level liquidity in the securities and the market maker is unable to source the securities;

o The trading party has arranged to have sufficient securities in time but it has not received the purchased securities on time.

127. By contrast, respondents have highlighted examples of causes of fails that are outside the direct control of custodians and their clients but are rather attributable to the existing market structure and the barriers to post-trade integration.

128. Many respondents discussed the “cascade effect” or “chain of fails”, i.e. that the success of the settlement of one transaction was generally contingent on the settlement of numerous transactions. and noted that brokers are dependent on their purchases settling to deliver their buy-side client. One respondent also noted that numerous intermediaries in the settlement chain increase the delay in sending instructions.

129. In the views of respondents, more specific structural causes with respect to CSDs include misaligned batch and cut-offs times between CSDs; misalignment between DvP and FoP batch times; CSDs not offering partial settlement, partial release or hold and release functionalities (in derogation of Article 8 of the RTS on Settlement Discipline); differing use and acceptance of ISO transaction types in settlement instructions messages leading to the rejection of instructions at the CSD’s SWIFT gateway; and certain ISINs are not eligible to settle in every EU CSD.

130. Respondents also mentioned the different cut-off times across CCPs as a structural cause of settlement fails. Furthermore, two respondents observed that the application of restrictive measures against serious human rights violations and abuses may result in settlement fails where the accounts and transactions are blocked for settlement.

Conclusions / Actions to be taken

131. ESMA takes due note of the root causes of settlement fails, and of the solutions that respondents put forward, ranging from enhancement and automation of confirmation, allocation and inventory management processes, to streamlined market practices and increased use of functionalities such as partial settlement, partial release and ‘hold and release’.

132. These suggestions will feed into ESMA’s upcoming Consultation Paper on measures to prevent settlement fails in order to increase settlement efficiency, which may consider amendments to the RTS on Settlement Discipline and to the Guidelines on standardised procedures and messaging protocols under Article 6(2) of CSDR.

133. Beyond regulatory measures that could be taken, ESMA strongly encourages all market participants to continue their efforts to increase settlement efficiency in the EU, also in light of a shortening of the settlement cycle.

Q44: Settlement efficiency in other jurisdictions

Summary

134. A few respondents stated that the levels of settlement efficiency are higher in the US and in the UK, while other respondents argued that these levels are either similar or lower. Most respondents however highlighted that the UK and the US are not comparable with the EU, and that they should not be used as benchmarks. These respondents noted that producing comparable statistics on settlement efficiency, across jurisdictions is challenging due to different methodologies in measuring and dealing with settlement fails.

135. In more detail, respondents described as main reasons for limited comparability the high consolidation of Financial Market Infrastructures in the US, as opposed to a more dispersed landscape in the EU; the impact of different liquidity levels in security borrowing and lending markets; additional complexity for cross-border transactions in the EU linked to multiple currencies; and a higher proportion of transactions centrally cleared in US markets, with a ‘continuous net settlement’ model which nets new and outstanding instructions.

136. Respondents also noted that in the US, the CSD can operate with ‘delivery without matching’, and that certain types of transaction can be returned to the delivering party after settlement date under the “don’t know” procedure, which leads to higher levels of reporting of settlement efficiency. By contrast, EEA CSDs operate with bilateral matching and irrevocable settlement finality.

137. A few respondents mentioned as a potential inspiration the introduction of the “dynamic fails charge”, as part of the TMPG Fails Charge framework introduced in the US since May 2009, which sets the charge of settlement fail on a daily basis, taking into account a reference rate linked to the monetary policy context23 . note that the EU may also display a potential correlation between the monetary policy context and settlement efficiency. In this light, respondents suggested to introduce stronger links between the level of cash penalties and the ECB’s key interest rates.

138. Similarly, respondents highlighted that many markets in Asia require the pre-placement of securities and cash prior to the execution of a transaction. This could contribute to lower levels of settlement fails, although potentially at the expense of increased costs and barriers to participation in these markets.

Conclusions / Actions to be taken

139. ESMA acknowledges the many factors leading to limited comparability between the EU and other jurisdictions. ESMA will however look to draw inspiration from initiatives in other jurisdictions, such as the introduction of the dynamic fails charge in the US.

5.2.2 Proposed revision of the penalty mechanism

Progressive penalty rates, with and without convexity (Q22, Q23, Q24 and Q28)

Q22: Progressive penalty rates increasing with settlement fail length

Summary

140. The vast majority of respondents did not support the introducing of progressive penalty rates increasing based on the length of the settlement fail.

141. Respondents noted that the impact of progressive penalty rates cannot be assessed with data available at present, as the current methodology for reporting from CSDs to NCAs does not provide for granular enough data on the duration of

[23] Cf. Section III in Frequently Asked Questions: TMPG Fails Charges https://www.newyorkfed.org/medialibrary/microsites/tmpg/files/TMPG-Fails-Charge-FAQ-10-19-2020.pdf

settlement fails (e.g. a three-day settlement fail is reported in the same way as three separate one-day settlement fails for the same quantity/notional value).

142. Moreover, the proposal for progressive penalty rates is generally viewed as unlikely to result in increased settlement ratios as it is not expected to lead to fundamental changes in market operations, while higher fees could increase spreads and reduce liquidity, potentially leading to higher costs of investors with a negative impact on the competitiveness of EU markets.

143. In the views of some participants, unintended consequences of the proposed changes include difficulties for participants to calculate, forecast and reconcile penalties, potentially undermining the principle of immunisation, as well as an incentive not to send settlement instructions until securities are available to participants, as the late matching fines would be lower than settlement fail penalties.

144. Most respondents also argued progressive penalty rates would have a limited impact on overall settlement efficiency levels, as the volume of settlement fails is highly concentrated in the first two days after the intended settlement date (ISD). These respondents referred to the T2S Annual Report 202224, highlighting that the volume of settlement fails decreased significantly overtime, from almost 25% of settlement fails recycled and carried over on the first business day following ISD, to less than 5% on the fourth business day, and 0.8% after 16 business days.

145. An additional argument against the proposal is the assumption that settlement fails lasting longer are more likely to be linked to factors outside of the participants’ control, for which a higher penalty rate as an incentive to change participants’ behaviour would not resolve the root cause of the settlement fail.

146. ETFs were often used as an example of financial instrument requiring a longer time to settle due to their intrinsic characteristics: e.g. when the ETFs are issued in a different time zone and/or with different trading times than its underlying instruments.

147. Very few respondents supported the proposal. One participant stated that the introduction of the new mechanism should allow to deter market participants to shortsell securities that will be borrowed, while another respondent deems that the

[24] https://www.ecb.europa.eu/press/intro/publications/pdf/ecb.t2sar2022.en.pdf

progressive penalty rates would be justified as long as the final rates are higher than the securities borrowing rate.

Conclusions / Actions to be taken

148. ESMA acknowledges the consensus from the vast majority of respondents against the introduction of progressive penalty rates, noting that there is limited evidence of the potential benefits of this approach, compared to the expected increase of operational complexity and costs for all market participants.

149. ESMA does not propose the introduction of progressive penalty rates at this stage and will look to collect additional data on the length of settlement fails to support further reflections on the effectiveness of linear penalty rates.

Q23 and Q24: Convexity in the progression of penalty rates

Summary

150. All respondents opposed the introduction of convexity in the progression of penalty rates, i.e. rates increasing in a first phase and decreasing in a second phase, both for liquid and for illiquid financial instruments. Respondents concurred in noting that convexity would increase complexity, with limited evidence that it would improve efficiency, under the assumption that settlement fails lasting many days suggest structural issues. Respondents recalled the necessity of a simple and predictable approach to penalty rates and argued that the cost/benefit ratio of introducing convexity is too low to justify such a change.

151. Some respondents also argued that introducing convexity in penalty rates could increase the number of settlement fails, by incentivising delayed corrective actions as lower penalty rates for days further than the ISD could lead to lower levels of penalties on average overtime.

Conclusions / Actions to be taken

152. ESMA acknowledges that increasing the complexity of the penalty mechanism could reduce its effectiveness in terms of communication, comprehension, and adherence by market participants. The introduction of convexity in penalty rates will not be pursued at this stage.

Proposed levels of penalty rates (Q25, Q26, Q34, Q38 Q46)

Q25, Q26, Q38: Levels of penalty rates

Summary

153. A large majority of respondents expressed their strong opposition to the proposed increase of penalty rates. These respondents argued that such a significant increase could lead to unnecessary costs and harm the competitiveness of EU markets.

154. In more detail, respondents noted that a significant increase in penalties could lead to unintended consequences, such as distortive effects on trading behaviours for the relevant securities (assuming that the impact of penalties is priced in) and incentives to settlement outside of the market infrastructures subject to the penalty mechanism.

155. Respondents also stated that significantly high penalties could disproportionately impact participants even if they are not ultimately found responsible for settlement fails, as they would be strongly impacted by an increase in the value of bilateral claims (when linked to the level of penalty rates). They also argued that significantly high penalties could encourage purchasing parties to refuse partial settlement or to initiate buy-ins (e.g. in cross-border non-cleared bond markets), assuming that they could obtain high returns from penalties if full matching does not occur.

156. Most respondents supported maintaining the current system with a moderate increase of the penalty rates before considering structural adjustments to the penalty mechanism, such as the introduction of progressive penalty rates. They suggested that this recalibration should take interest rates and broader market conditions into account and should ensure that the costs of failing to deliver a security are higher than the costs of delivering the security, including through securities lending/borrowing. Respondents also anticipated that a shorter settlement cycle may increase settlement fails rates.

157. One respondent suggested to introduce a dynamic approach to penalty rates to take the interest rate environment into account. They proposed that penalty rates are reviewed on a periodic basis against a dynamic reference rate, such as the euro short-term rate (€STR), associating each type of settlement fail with a multiplier and a floor which could be anchored in the existing calibration. depending on the specific financial instruments.

158. Another respondent proposed that market-making activities are exempted from the penalty mechanism, and that flat rates, which could be based on external benchmarks, are introduced for all other activities.

Conclusions / Actions to be taken

159. ESMA acknowledges the strong pushback on the proposed increase of penalty rates, which was voluntarily significant to elicit stakeholders’ feedback. ESMA also recognises that a significant increase of penalty rates may divert resources from expected investments and costs of moving to T+1.

160. As an alternative, ESMA favours a moderate increase of penalty rates, using the securities lending and borrowing rates observed in 2022 and in 2023 as reference to ensure that the costs of penalties would remain on average above the costs of borrowing securities to resolve the fail.

161. At this stage, ESMA has not taken forward the proposals of a dynamic approach to penalty rates and/or the introduction of flat rates, as they may be considered a structural change to the penalty mechanism. These options are however reflected in the costs/benefits analysis in Annex V and could be explored during the next review of the penalty mechanism.

Q46: Minimum penalties across all fails types

Summary

162. Most respondents objected to the introduction of minimum penalties, arguing that it may bear a distortive impact on smaller trades, that it may breach the immunisation principle in case of block order settlement, partial settlement and netting, and that the overall implementation costs would not be proportionate compared to the expected benefits.

163. A few respondents suggested to explore a model whereby the CSDs withhold the credit and debit to CSD participants for cases where the costs of processing the penalty would be higher than the value of the penalty itself.

Conclusions / Actions to be taken

164. ESMA will not pursue the introduction of minimum penalties at this stage.

Categorisation of types of fails (Q27, Q29, Q31, Q35, Q36)

Q27, Q29 and Q31: Existing vs. new types of fails, and treatment of ETFs

Summary

165. A large majority of respondents favoured maintaining the existing categorisation of type of fails (Option 1), positing that a distinction by asset type and by liquidity status reflects the different post-trade processes.

166. Many respondents provided detailed feedback on specific characteristics of ETFs and how to improve settlement efficiency for this asset class. While some respondents supported the introduction of a new type of fail due to the lack of ETFs as an appropriate tool to the specific characteristic of this asset class, other respondents argued that ETFs are adequately penalised in the current ‘catch-all’ type of fail (covering ETFs, transferable securities referred to in point (c) of Article 4(1)(44) of MiFID II; units in collective investment undertakings, emission allowances and all other financial instruments).

167. One respondent suggested that fails due to lack of ETFs could be targeted under the current methodology, by assigning a specific penalty rate based on CFI codes typically used for ETFs.

168. By contrast, most of the respondents opposed the proposal to assign single penalty rates for all liquid financial instruments and for all illiquid financial instruments (Option 2), highlighting that this complete overhaul of the current methodology does not reflect the differences across asset classes and would not allow to rely on the CFI codes of each security. Only one respondent was in favour of the categorisation of types of fails as proposed under Option 2.

169. Beyond the few suggestions for a specific treatment of ETFs, only a few respondents supported additional granularity on the categorisation of fails, suggesting an exemption for market-making activities or a differentiation according to the type of counterparty to the trade.

Conclusions / Actions to be taken

170. ESMA notes the overall support for the existing categorisation of types of fails, and the mixed views on whether to introduce a specific type of fail due to lack of ETFs.

Q35 and Q36: Treatment of illiquid financial instruments

Summary

171. The majority of respondents were against the proposal for a new type of fail due to lack of illiquid bonds, noting that this would add complexity with no discernible benefits, as most of the bonds are classified as illiquid.

172. The few respondents in favour of a differentiation between bonds based on their liquidity noted that this would require as a prerequisite better calibrated penalty rates, more granular data collection on bonds, and a central ESMA database of all the data needed to calculate penalties.

173. Respondents also suggested that further flexibility in relation to penalties for illiquid financial instruments could include:

o Excluding highly illiquid bonds and/or illiquid financial instruments commonly arranged in contingent packages and as complex trades;

o Allowing for greater flexibility for market making activities; o Sponsoring the creation of a securities lending market on illiquid instruments, possibly including clearing by CCPs and encourage more lending of securities, such as extending auto-lending/borrowing facilities across all EU CSDs.

Conclusions / Actions to be taken

174. In light of this feedback, ESMA will not introduce a new type of fail specifically targeting illiquid bonds.

175. ESMA also notes that the suggestions for further flexibility for illiquid financial instruments are outside of the scope of the Technical Advice.

Potential consequences and implementation period (Q47)

Q34 and Q37: Risks of increased use on FoP settlement instructions, and for ‘net-long’ cash payments

Summary

176. Almost all respondents deem that the risk that participants could shift from DvP to FoP due to higher penalty rates negligible, with only one respondent mentioning it as possible. In their views, DvP and FoP address different needs with FoP mainly used for realignment of securities between accounts, pledging of collateral, and cross-border settlements where DvP in the relevant currency is not made available by the CSD.

177. Most of the respondents deemed low to non-existent the risk that underlying parties that end up with “net long” cash payments may not have incentives to manage their fails or bilaterally cancel failing instructions as they may “earn” cash from penalties. These respondents pointed that the risks of such practices on participants’ reputation and broker relations, and for their best execution obligations and for broker relations are too high compared to the low expected revenues, and that participants systemically misusing CSD functionalities receive an undue cash penalty would be subject to bilateral claims.

Conclusions / Actions to be taken

ESMA - 201-203 rue de Bercy - CS 80910 - 75589 Paris Cedex 12 - France - Tel. +33 (0) 1 58 36 43 21 - www.esma.europa.eu

178. ESMA notes that potential unintended consequences of a moderate increase of penalties rates are very low.

Q47: Time needed for CSDs and market participants

Summary

179. Only some respondents provided estimations of the time needed for CSDs and market participants to implement changes to the cash penalty mechanism, ranging from 6-9 months to 3-4 years, with an overall preference for longer timespans.

Conclusions / Actions to be taken

180. ESMA is of the view that an increase in penalty rates without structural changes to the penalty mechanism can be implemented without a significant implementation period.

5.2.3 Additional considerations to simplify the penalty mechanism

Changes to the parameters defining the rate levels (Q30, Q31, Q41, Q42)

Q30: Penalty rates based on the value of the settlement fail

Summary

181. Respondents were almost unanimous in objecting to defining rate levels based on the value of the failed instructions, invoking its general complexity and associated costs, with no demonstrable improvement on settlement efficiency. In particular, respondents identified as potential unintended consequences: mismatch between penalty paid and received and breaches of the principle of immunisation (according to which only the participants truly in default are penalised), in turn penalising retail investors and impairing the competitiveness of EU markets.

182. Only one respondent posited that the proposal to have settlement fails based on instructions with a lower value charged with a higher penalty rate than those with a higher value would be interesting as it would tackle the introduction of a threshold under which non-material penalties would not be passed on, non-reconciled or claimed.

Conclusions / Actions to be taken

183. In light of the feedback received, ESMA will not explore this approach at this stage.

Q41: Penalty rates based on transaction types

Summary

184. Most of the respondents stated they are against differentiated rates by transaction type, with similar arguments as above on the complexity and costs of such a change. As a potential unintended consequence, they highlighted that participants may choose specific transaction types solely based on their penalty implications.

185. Moreover, a few respondents called for certain types of transactions to be exempted from the cash penalty mechanism altogether, such as market claims and transformations, instructions generated by CSDs, transactions linked to mandatory buy-ins, allocations and instructions linked to corporate actions (e.g. Payment Free of Delivery (PFODs) and market claims (notably market claims on cash distributions). One respondent also proposed a one-day “grace period” for settlements related to new issuances, in light of inconsistent settlement deadlines across time zones.

Conclusions / Actions to be taken

186. In light of the feedback received, ESMA will not explore the introduction of penalty rates based on transaction types. The scope of transactions subject to the settlement discipline regime will be addressed in a separate Technical Advice, based on the feedback received via a dedicated Consultation Paper25 .

[25] 25https://www.esma.europa.eu/sites/default/files/2024-07/ESMA74-2119945925-1976_CSDR_Consultation_Paper_on_Technical_Advice_on_Scope_of_Settlement_Discipline.pdf

Q42: Penalty rates dependent on securities borrowing fees

Summary

187. The vast majority of respondents were against the suggestions to make penalty rates dependent on securities borrowing fees. However, they noted that the cost of paying penalties should be overall higher than the cost of borrowing securities. Respondents highlighted the operational complexity of this approach, due to the absence of a centralised source of securities borrowing rates and given that not all market participants have access to securities lending.

188. In particular, they noted that the variations of securities borrowing rates are due to numerous factors not necessarily linked to settlement efficiency, which could reduce the predictability of the penalty mechanism, and could bring about unintended consequences in particular for less liquid securities.

Conclusions / Actions to be taken

189. In light of the feedback received, ESMA will not explore this approach. ESMA has however used securities lending/borrowing rates under the SFTR as a benchmark for the proposed increase of cash penalty rates.

Changes to the calculation of the value of the failed instructions (Q32, Q33)

Q32: Calculation based on the instruction value on the first day of the settlement fail

Summary

190. Regarding the use of the market value of the first day of the settlement fail as a basis for the calculation of penalties for the entire duration of the fail, the main criticisms from respondents (beyond the complexity and cost of this change) were related to the breach of the immunisation principle, and to the decreased transparency for investors due to the dependence on the price for a single day.

Conclusions / Actions to be taken

191. In light of this feedback, ESMA will not explore this option further, although noting that some respondents stated that it may have been beneficial if implemented from inception.

Q33: Calculation of the value of free of payment (FoP) instructions

Summary

192. The overwhelming majority of respondents supported the current methodology of the valuation of FoP instructions, i.e. the use of the same reference price for DvP and FoP instructions.

Conclusions / Actions to be taken

193. ESMA recommends to the EC to amend Commission Delegated Regulation (EU) 2017/389 to include the use of the nominal value as an alternative to the methodology referred to in point c) of Article 7 of Commission Delegated Regulation (EU) 2017/389. This should further enable a consistent approach across CSDs regarding the valuation of instructions covering financial instruments that are not admitted to trading or traded on a trading venue. This should apply irrespective of the type of settlement instruction (e.g. DvP or FoP) and would continue to allow the use of the same daily reference price for the same ISIN for both DvP and FoP instructions.

Approach based on settlement efficiency targets (Q19, 39, 40)

Q19: Appropriate level(s) of settlement efficiency

Summary

194. There was a consensus amongst the respondents to say that 100% of settlement efficiency may not be a realistic target and that a higher settlement efficiency rate is better than a lower one.

195. Most respondents indicated that it is very challenging or even impossible to define quantitative targets of settlement efficiency, and suggested to focus instead on a sustainable reduction of settlement fails. Those respondents explained that the settlement efficiency rate is depending on many factors (e.g. market structure, structural causes of fails, asset-specific liquidity) and that an explicit target could notably impact liquidity in EU markets.

196. By contrast, only a few respondents stated they are in favour of regulators defining quantitative thresholds for settlement efficiency, with two respondents suggesting 95% and 99% as appropriate targets across all asset types.

197. The vast majority of respondents considered that settlement efficiency rates targets should be specific for each category of assets, given the settlement efficiency rate for asset type is dependent on the volume of trading, liquidity, cost and availability of borrowing securities, while objecting to CSD/SSS-specific targets.

198. Respondents highlighted that the process to determine appropriate targets should consider the duration of settlement fails, analyse the settlement fails rates over a meaningful period of time, and consider the systemic risk implications (e.g. when would high levels of fails will start creating a systemic risk for the EU, how could a systemic external shock arise).

199. One respondent also noted that a specific definition of settlement efficiency, as well as harmonised and comparable settlement efficiency statistics produced by CSDs, are pre-requisites to the introduction of settlement efficiency targets.

Conclusions / Actions to be taken

200. In light of the feedback received, ESMA will not explore the definition of settlement efficiency targets at this stage, noting that such targets would be more suitable at the level of asset types rather than at CSD/SSS level.

201. ESMA will continue to steer regulatory and supervisory action to tackle structural causes and enhance settlement efficiency in the EU and calls on all CSDs and market players to contribute to this objective at their respective level.

Q39 and Q40: Penalties for CSDs/SSSs and asset types with higher settlement fail rates

Summary

202. The vast majority of the respondents disagreed with differential penalties based on settlement efficiency, with only one respondent that favoured higher penalties for CSD/SSSs with higher settlement fails rates.

203. Respondents highlighted similar reasons as above to object to this approach: overall complexity and associated costs, breach of the immunisation principle, multiple drivers of settlement inefficiency not all linked to the performance of the CSD/SSS, and eventually detrimental impact on EU’s competitiveness.

Conclusions / Actions to be taken

204. In light of the feedback received, ESMA will not explore this approach.

Ad hoc measures for CSDs participants with high settlement fail rates (Q48, Q49, Q50, Q51)

Q48 and Q50: Detection, suspensions and working arrangements for participants with high settlement fail rates

Summary

205. Of the few CSDs that responded, all stated that only a small number of participants had been identified as failing consistently and systematically between 2022 and 2024, ranging from none to 11 clients per year, and without a significant impact of the SSSs according to one group of CSDs. No suspensions of any of these participants were reported.

206. Similarly, only few CSD respondents detailed the working arrangements they have put in place for participants with the highest rates of settlement fails. Only one group of CSDs have implemented monthly working arrangements to collect feedback on fail reasons and measures applied, with groups ranging from 6 to 47 clients depending on the CSD.

207. Another group of CSDs indicated that they have not implemented any working arrangements with the top 10 failing participants, but they have implemented working group or meetings to improve settlement efficiency. Another CSD stated it carries out monthly written consultations with the participants who are systemically important.

Conclusions / Actions to be taken

208. ESMA is of the view that CSDs should improve their monitoring of participants failing consistently and systematically and, where appropriate, consider suspensions in accordance with Article 7(7) of CSDR and Article 39 of the RTS on Settlement Discipline.

209. ESMA also reminds CSDs that the implementation of working arrangements for participants with the highest rates of settlement fails is a mandatory requirement under Article 13(2) of the RTS on Settlement Discipline and invites NCAs to ensure that these requirements are effectively implemented by the CSDs they supervise.

Q49: Special penalties for participants with high settlement fail rates

Summary

210. Almost all respondents opposed the introduction of special penalties applied to participants with high settlement fail rates, arguing that such special penalties would go against the principle of immunisation, could sanction participants that do not control the causes of fails, and would add complexity to the penalty mechanism.

211. Only three respondents, notably from the asset management sector, expressed interest in the proposal as an additional incentive to improve settlement efficiency. One respondent also suggested considering two additional criteria to identify the relevant participants: (i) the recurrence of late /failed trades from a given counterparty on the previous 12 months and (ii) the number of late/failed trades from a given counterparty on the previous 12 months.

Conclusions / Actions to be taken

212. ESMA acknowledges the feedback and will not purse the introduction of special penalties for participants with high settlement fail rates at this stage.

Q51: Settlement efficiency discussed at the CSDs’ User Committees

Summary

213. The majority of respondents welcomed discussions on settlement efficiency at the CSDs’ Users Committees and indicated this has already been the case for a long time. Best practices mentioned by respondents included the provision of data on settlement efficiency, settlement fail penalties and late matching penalties to the Users Committees and/or to other relevant industry working groups.

214. Respondents suggested that CSD’s participants should be invited to explain the context of changes of the fail rates, given the settlement efficiency rate of any individual market participant can affected by the settlement efficiency rate of other market participants. Respondents also highlighted that that any changes to CSD functionalities and CSD daily timetables aiming at improving settlement rates should be discussed at the CSD’ User Committees.

215. A few respondents called for setting up a broader committee on settlement efficiency going beyond CSDs and their participants to capture all relevant market stakeholders.

Conclusions / Actions to be taken

216. ESMA encourages the continuation of discussions on settlement efficiency at CSDs’ Users Committees, invites CSDs and their participants to put in place the best practices mentioned above, and will consider whether regulatory and/or supervisory actions are needed in this light.

Additional suggestions (Q18, Q43)

Q18: Key tools to improve settlement efficiency

Summary

217. Respondents identified several tools and actions that CSDs and market participants could implement to improve settlement efficiency.

218. Many respondents supported an increased use and improvement of CSDs functionalities such as (i) partial settlement, (ii) hold and release, (iii) partial release, (iv) auto-partialling, with several respondents also recommending the use of “shaping” of instructions.

219. These respondents were in favour of mandating all EU CSDs to offer partial settlement, hold and release, as well as partial release functionality (notably for omnibus accounts), if needed with an option to opt-out and/or exemptions for specific cases. To this end, they suggested the removal of the derogation provided for in Article 12 of the RTS on Settlement Discipline.

220. Moreover, many respondents have identified several structural issues at CSDs level that they consider should be tackled to improve settlement efficiency. They recommended that all EU CSDs provide fully automated, continuous real-time matching of settlement instructions, improve realignment processes between CSDs, align CSD cycles and market cut-offs, improve interoperability and standardisation to achieve a single market and seamless cross-border securities flow, and share their holidays calendar ahead of the next calendar day to allow trading parties and custodians to include them in their systems.

221. Some respondents emphasised that industry practices could improve thanks to standard and electronic means of communication to allocate and confirm trades, increased automation in firms’ internal processes, and higher granularity of data when setting up client accounts.

222. Respondents also highlighted the need for improved management of Standard Settlement Instructions (SSIs), through their automated use and population to avoid manual templates and call-backs, and with the help of centralised SSI databases as hubs to share and update information across all participants.

223. These respondents also suggested that mismatches between the economic elements of the trade and SSIs are identified before settlement instructions are sent to the CSDs, thanks to pre-settlement matching tools such as vendor matching platforms. Such platforms should align their matching criteria with those of the CSDs, including for the “place of settlement” (PSET).

224. A few respondents also recommended the use of transactions monitoring tools such as Unique Transaction Identifiers (UTIs) and / or Unique Product Identifiers (UPIs), to track of settlement instructions and their errors in the upstream parts of the settlement chain.

225. Some respondents put forward recommendations on intermediaries’/custodians’ practices, such as sending settlement instructions in real-time/ intra-day (to make instructions available at the CSD on trade date, and therefore to make matching discrepancies visible to trading parties via their custodians as early as possible); and making mandatory the reporting by custodians of field 94a (place of safekeeping) on MT535 (statement of holding messages).

226. Some respondents also considered that the use of securities lending by CSDs’ participants could improve settlement efficiency, for instance through custodians’ and CSDs’ auto-borrowing programs, enabling timely acquisition of securities.

227. In many responses, stakeholders stressed the challenges and complexities related to settlement on a cross border basis. One respondent even recommended that all trades should be settled via Target 2 Securities Platform (T2S) and that cross-border settlement outside T2S should only be authorised for cross border settlement in non-EU CSDs.

Conclusions / Actions to be taken

228. The suggestions provided will feed into ESMA’s further reflections on measures to enhance settlement efficiency, in line with the mandate in Article 6(5) of CSDR as amended by CSDR Refit. ESMA intends to launch a dedicated consultation on this topic in early 2025.

Temporary suspension of the penalty mechanism

Summary

229. Some respondents argued that a shortening of the settlement cycle could potentially lead to an increase of settlement fails and, consequently, to an increase of the cash penalties incurred by market participants. They noted that cash penalties are a specificity of the EU toolkit on settlement discipline, which may create an additional cost for failing counterparties when implementing a shortening of the settlement cycle in the EU compared to other jurisdictions like the US or the UK, and highlighted that funds, and in particular ETFs, may face specific problems (see relevant section in the Report on ESMA assessment of the shortening of the settlement cycle in the EU).

230. These respondents advocated for a temporary suspension of the penalty mechanism, which, in their view, would allow market participants to focus their resources on adapting to the shortening of the settlement cycle in the EU, without sustaining a temporary increase of the levels of penalties during the transition to T+1.

Conclusions/ Actions to be taken

231. ESMA believes that more substantial evidence, including quantitative data, is needed to support the assumption that the shortening of the settlement cycle in the EU would lead to an increase of the overall level of cash penalties.

232. ESMA advises the EC to consider ways to smooth out the impact of the transition, specifically related to the levels of settlement fails during the early implementation phase, including the possibility for a time-limited suspension of the application of cash penalties, or an alternative mechanism to alleviate the potential increase of the overall level of cash penalties in the context of the shift to T+1. However, this would need to be supported by substantial evidence from market participants, in the context of the governance put in place to help coordinate the shift to T+1 in the EU.

233. Recommendations on the timeframe and the design for a potential temporary suspension, or any alternative mechanism, could be provided in due course, given the agreed timeline for the move to T+1 in the EU.

5.3 Additional evidence

234. Considering the feedback received in the consultation paper, ESMA has conducted in-depth data analysis to provide additional evidence to support its technical advice. This additional evidence should be read in conjunction with regular publications on settlement efficiency, in particular ESMA’s TRV Reports.

5.3.1 Settlement efficiency trends

235. Since the introduction of the CSDR penalty mechanism, ESMA has closely monitored the trends in settlement efficiency and in the functioning of the cash penalties, with a comprehensive set of indicators based on data collected from CSDs, as submitted by NCAs to ESMA.

236. Some of these indicators supported ESMA’s assessment for this technical advice, notably on the volume and value of settlement fails per type of financial instruments. These indicators are presented without outliers.

0
2
4
6
8
10
12
14
16
18
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA EEA 1M-MA EEA 1Y-MA EEA Note: Settlement fails as a % of total value of settlement instructions at EEA level.

FIGURE 1: RATIO OF ALL SETTLEMENT FAILS (VALUE)

0
2
4
6
8
10
12
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA EEA 1M-MA EEA 1Y-MA EEA Note: Settlement fails as a % of total volume of settlement instructions at EEA level.

FIGURE 2: RATIO OF ALL SETTLEMENT FAILS (VOLUME)

237. Since the introduction of the penalty mechanism in February 2022, a general decrease of the ratio of all settlement fails, both in value and volume, can be observed.

FIGURE 3: RATIO OF SETTLEMENT FAILS FOR BONDS (VALUE)

0
10
20
30
40
50
60
70
80
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA CllctvInvstmtUdrtkgs 1M-MA CllctvInvstmtUdrtkgs 1Y-MA CllctvInvstmtUdrtkgs Note: Settlement fails as a % of total value of settlement instructions at EEA level for units in collective investment undertakings, other than ETFs.
Sources: ESMA.

0
5
10
15
20
25
30
35
40
45
50
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA Bd 1M-MA Bd 1Y-MA Bd Note: Settlement fails as a % of total value of settlement instructions at EEA level for bonds or other forms of securitised debt, including depositary receipts in respect of such securities.
Sources: ESMA.

FIGURE 4: RATIO OF SETTLEMENT FAILS FOR BONDS (VOLUME)

0
2
4
6
8
10
12
14
16
18
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA Bd 1M-MA Bd 1Y-MA Bd Note: Settlement fails as a % of total volume of settlement instructions at EEA level for bonds or other forms of securitised debt, including depositary receipts in respect of such securities.
Sources: ESMA.

FIGURE 5: RATIO OF SETTLEMENT FAILS FOR UNITS IN CIUS (VALUE)

0
2
4
6
8
10
12
14
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA CllctvInvstmtUdrtkgs 1M-MA CllctvInvstmtUdrtkgs 1Y-MA CllctvInvstmtUdrtkgs Note: Settlement fails as a % of total volume of settlement instructions at EEA level for units in collective investment undertakings, other than ETFs.
Sources: ESMA.

FIGURE 6: RATIO OF SETTLEMENT FAILS FOR UNITS IN CIUS (VOLUME)

0
5
10
15
20
25
30
35
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA Eqty 1M-MA Eqty 1Y-MA Eqty Note: Settlement fails as a % of total value of settlement instructions at EEA level for shares in companies and other securities equivalent to shares in companies, partnerships or other entities, and depositary receipts in respect of shares. Sources: ESMA.

FIGURE 7: RATIO OF SETTLEMENT FAILS FOR SHARES (VALUE)

0
2
4
6
8
10
12
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA Eqty 1M-MA Eqty 1Y-MA Eqty Note: Settlement fails as a % of total volume of settlement instructions at EEA level for shares in companies and other securities equivalent to shares in companies, partnerships or other entities, and depositary receipts in respect of shares. Sources: ESMA.

FIGURE 8: RATIO OF SETTLEMENT FAILS FOR SHARES (VOLUME)

0
2
4
6
8
10
12
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA MnyMktInstrm 1M-MA MnyMktInstrm 1Y-MA MnyMktInstrm Note: Settlement fails as a % of total value of settlement instructions at EEA level for money-market instruments, other than sovereign debt.
Sources: ESMA.

FIGURE 9: RATIO OF SETTLEMENT FAILS FOR MMIS (VALUE)

0
2
4
6
8
10
12
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA MnyMktInstrm 1M-MA MnyMktInstrm 1Y-MA MnyMktInstrm Note: Settlement fails as a % of total volume of settlement instructions at EEA level for money-market instruments, other than sovereign debt.
Sources: ESMA.

FIGURE 10: RATIO OF SETTLEMENT FAILS FOR MMIS (VOLUME)

0
1
2
3
4
5
6
7
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA SvrgnDebt 1M-MA SvrgnDebt 1Y-MA SvrgnDebt Note: Settlement fails as a % of total value of settlement instructions at EEA level for sovereign debt.
Sources: ESMA.

FIGURE 11:RATIO OF SETTLEMENT FAILS FOR SOVEREIGN BONDS (VALUE)

0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA SvrgnDebt 1M-MA SvrgnDebt 1Y-MA SvrgnDebt Note: Settlement fails as a % of total volume of settlement instructions at EEA level for sovereign debt.
Sources: ESMA.

FIGURE 12:RATIO OF SETTLEMENT FAILS FOR SOVEREIGN BONDS (VOLUME)

0
5
10
15
20
25
30
35
40
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA XchgTraddFnds 1M-MA XchgTraddFnds 1Y-MA XchgTraddFnds Note: Settlement fails as a % of total value of settlement instructions at EEA level for exchange-traded funds (ETFs).
Sources: ESMA.

FIGURE 13: RATIO OF SETTLEMENT FAILS FOR ETFS (VALUE)

0
5
10
15
20
25
30
35
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 1W-MA XchgTraddFnds 1M-MA XchgTraddFnds 1Y-MA XchgTraddFnds Note: Settlement fails as a % of total volume of settlement instructions at EEA level for exchange-traded funds (ETFs).
Sources: ESMA.

FIGURE 14: RATIO OF SETTLEMENT FAILS FOR ETFS (VOLUME)

238. The granularity of asset classes in these indicators differs from the categories of type of fails foreseen by the penalty mechanism: both MMIs and ETFs are not targeted by a specific category of fail type; shares can be covered in three fail types depending on their liquidity status and venue of execution; and bonds that are not sovereign bonds can be covered in two fail types depending on their venue of execution.

239. The indicators are nonetheless useful to assess general trends. The overall decrease in settlement fails is particularly noticeable for bonds, shares and MMIs, but remains more modest for units in collective investment undertakings (CIUs), sovereign bonds and ETFs.

240. The aggregated ratios of settlement fails across all EEA CSDs are varied depending on the asset classes, ranging in June 2024 from low levels for sovereign bonds (~2% of the total value of settlement instructions, and ~2.5% of the total volume (number) of settlement instructions) to very high levels for ETFs (~15% of the total value of settlement instructions, and ~20% of the total volume (number) of settlement instructions).

0%
20%
40%
60%
80%
100%
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 SvrgnDebt Bd Eqty XchgTraddFnds MnyMktInstrm CllctvInvstmtUdrtkgs OthrTrfblScties Othr EmssnAllwnc Note: Monthly value of failed settlement instructions at EEA level by type of financial instruments, in % of total.

FIGURE 15: FAILED SETTLEMENT INSTRUCTIONS BY FINANCIAL INSTRUMENT TYPE (VALUE)

0%
20%
40%
60%
80%
100%
Feb-22 Jun-22 Oct-22 Feb-23 Jun-23 Oct-23 Feb-24 Jun-24 Eqty Bd XchgTraddFnds SvrgnDebt OthrTrfblScties CllctvInvstmtUdrtkgs MnyMktInstrm Othr EmssnAllwnc Note: Monthly volume of failed settlement instructions at EEA level by type of financial instruments, in % of total.

FIGURE 16: FAILED SETTLEMENT INSTRUCTIONS BY FINANCIAL INSTRUMENT TYPE (VOLUME)

241. The analysis of the failed settlement instructions by type of financial instruments shows that sovereign bonds and bonds other than sovereign bonds constitute the largest portion of the monthly value of settlement fails, with both instrument types alternatively constituting up to 40% of such value.

242. By contrast, the monthly volume of failed settlement instructions is more evenly distributed across types of financial instruments, with equity and ETFs constituting the top 3 financial instrument types together with bonds other than sovereign bonds.

243. From these indicators, ESMA concludes that the penalty rates can continue to bear a positive impact on settlement efficiency overall. The analysis also confirms that tailoring the penalty rates with categories primarily based on asset classes remains relevant, in light of the observed divergences across asset classes.

5.3.2 Securities lending and borrowing rates

244. Respondents pointed to the discrepancy between the cost of incurring the penalty for failing to deliver a security and the costs of borrowing the same security to resolve the settlement fail.

245. To substantiate this assumption, ESMA has drawn up a comparison between the current cash penalty rates and the securities lending and borrowing rates observed in 2022 and in 2023, based on the reporting of Securities Financing Transactions (SFTs) to Trade Repositories (TRs) under the SFTR.

Types of fail	Current penalty rate	Equivalent yearly funding rate26	Annual average securities lending/borrowing volume-weighted rates
1. Settlement fail due to a lack of liquid shares	1 basis point	3.65%	1.51% in 2022; 2.49% in 2023
2. Settlement fail due to a lack of illiquid shares	0.5 basis point	1.83%	2.62% in 2022; 5.76% in 2023
3. Settlement fail due to a lack of financial instruments other than debt instruments traded on a SME growth market	0.25 basis point	0.91%	Insufficient data observed to provide a meaningful comparison
4. Settlement fail due to a lack of sovereign bonds	0.1 basis point	0.37%	0.28% in 2022, 3.28% in 2023
5. Settlement fail due to a lack of bonds other than sovereign bonds and not	0.2 basis point	0.73%	0.13% in 2022; 2.74% in 2023

[26] Daily rate (percent) x 365 days

traded on a SME growth market
6. Settlement fail due to a lack of bonds traded on a SME growth market	0.15 basis point	0.56%	No data observed in 2022 and 2023
7. Settlement fail due to a lack of other financial instruments	0.5 basis point	1.83%	1,76% in 2022; 6.07% in 2023
Settlement fail due to a lack of ETFs	0.5 basis points	1.83%	2.83% in 2022; 5.91% in 2023

TABLE 1: PENALTY RATES AND SECURITIES LENDING/BORROWING RATES PER FAIL TYPE

246. ESMA has compiled the volume-weighted average securities lending/borrowing rates for 2022 and for 2023 based on the lending fee reported in field 2.67 from trade activity reports under SFTR, removing the list of ISINs that have their principal trade venue in a third country as notified under the Short Selling Regulation, and excluding outliers (approx. 2.5% of total records).

247. The observed rates should only be understood as reference points rather than an accurate representation of the actual costs that market participants may occur when borrowing a security to resolve a settlement fail. These rates are influenced by many factors, not least the interest rate environment (as evidenced by the different values observed in 2022 and 2023), and the actual cost of borrowing a security is also dependent on the combination of other rates (e.g. a rebate rate is used for securities lending/borrowing against cash).

248. With these caveats in mind, ESMA nonetheless notes that the penalty rates have been lower than securities lending and borrowing rates for the following financial instruments: illiquid shares, sovereign bonds, bonds traded on a SME growth market and other financial instruments, in particular ETFs.

249. In light of this analysis, ESMA deems it necessary to consider increasing the penalty rates for the corresponding types of settlement fails.

5.3.3 Duration of settlement fails in T2S

250. Respondents challenged the effectiveness of the progressive penalty rates, i.e. an increase of penalty rates according to the length of the fails, by highlighting that most of the settlement fails are resolved within 2 business days (BD) of their intended settlement date (ISD).

251. To gather evidence on the duration of settlement fails, ESMA has asked the T2S Operator to provide an analysis covering a calendar month (March 2024). More details regarding the applied methodology are included below.

252. The duration of settlement fails has been calculated on a daily basis for each business day that a transaction fails to be settled after its intended settlement date, as follows: Duration of settlement fails=BD-ISD. At the end of the day, any transactions that are partially settled, unsettled, or on hold will be classified as failed.

253. High-level principles considered in the methodology for the calculation include:

o All internally generated transactions (e.g. T2S realignment transactions, auto-collateral transactions, etc.) are subtracted from the calculation;

o All reasons for non-settlement shall be considered as settlement fails;

o Failed transactions are grouped into multiple clusters according to their settlement delay (e.g. the cluster BD=ISD contains all transactions where the settlement fail occurred on ISD; the cluster BD=ISD+1 contains all transactions that are still pending 1 day after their ISD, etc.);

o For each business day on which a transaction fails to settle after its ISD, it is included in the reporting (e.g. in case the transaction fails to settle on “ISD”, it is counted in the cluster BD=ISD. In case the transaction fails again on “ISD+1”, it is also counted in the next cluster BD=ISD+1.This means that a transaction can be counted multiple times within the month);

o Transactions failed for more than 10 days are grouped in the cluster BD=ISD+n, n>10.

254. The duration of settlement fails has been calculated at T2S level (aggregated data) in value and volume, with data referring to March 2024, and all pending and recycling settlement transactions in T2S, including those submitted before 1 March 2024.

Cluster	Volume (number)	Value (remaining amount)	% volume	% value
BD=ISD	556.670	396.823.110.408,20 €	61%	72%
BD=ISD+1	106.545	44.775.951.553,96 €	12%	8%
BD=ISD+2	37.864	11.849.431.383,23 €	4%	2%
BD=ISD+3	36.898	15.208.091.555,87 €	4%	3%
BD=ISD+4	26.944	12.106.673.343,38 €	3%	2%
BD=ISD+5	19.889	7.295.337.061,37 €	2%	1%
BD=ISD+6	17.370	6.147.658.180,38 €	2%	1%
BD=ISD+7	15.711	5.635.636.198,11 €	2%	1%
BD=ISD+8	9.486	3.988.517.821,90 €	1%	1%
BD=ISD+9	5.474	1.733.068.687,95 €	1%	0%
BD=ISD+10	5.198	1.790.521.987,08 €	1%	0%
BD=ISD+n, n>10	77.490	44.306.518.945,49 €	8%	8%

TABLE 2: DURATION OF SETTLEMENT FAILS IN T2S (MARCH 2024)

255. Based on this ad-hoc analysis, ESMA notes that the vast majority of settlement fails in T2S are resolved within 2 BD of the ISD, both in terms of volume and value, and that similar trends may be safely assumed for all settlement instructions subject to the penalty mechanism.

256. ESMA therefore concludes that the introducing of progressive penalty rates may only have limited impact on the overall levels of settlement efficiency.

5.4 ESMA’s technical advice

257. In response to the EC’s request for advice, ESMA proposes to maintain the design of the current penalty mechanism, i.e. not to introduce fundamental changes to the methods for calculating penalties.

258. In more detail, ESMA discards at this stage the introduction of progressive penalty rates (with or without convexity), as well as changes to the parameters defining the rate levels (e.g. introducing an explicit link between the penalty rates and interest rates for settlement fails due to lack of financial instruments). These policy options are however included in the Cost Benefit Analysis in Annex V and may be further explored in a future review of the penalty mechanism. Similarly, ESMA does not propose a new calculation method linked to settlement efficiency targets or with a specific focus on CSDs participants with high settlement fail rates. At the same time, ESMA encourages CSDs, their participants, and competent authorities to continue working towards further improving settlement efficiency, in particular in light of the shortening of the settlement cycle in the EU.

259. Regarding the calculation of the value of failed instructions, ESMA suggests adding a reference to the possibility to use the nominal value for settlement instructions covering financial instruments that are not admitted to trading or traded on a trading venue. The proposed amendment in this respect to Article 3(1) of Commission Delegated Regulation (EU) 2017/389 is included below.

Article 3

Reference price of the transaction

1. The reference price referred to in Article 2 shall be equal to the aggregated market value of the financial instruments determined in accordance with Article 7 for each business day that the transaction fails to be settled, or to the nominal value for financial instruments referred to in point c) of Article 7 where the market value cannot be established.

Article 7

Determination of market values

The market value of financial instruments referred to in Articles 3, 5 and 6 of this Regulation shall be determined as follows:
(a)for financial instruments referred to in Article 3(1) of Regulation (EU) No 600/2014 admitted to trading on a trading venue within the Union, the market value of the relevant financial instrument shall be the closing price of the most relevant market in terms of liquidity referred to in Article 4(1)(a) of Regulation (EU) No 600/2014; (b)for financial instruments admitted to trading on a trading venue within the Union other than those referred to in point (a), the market value shall be the closing price derived from the trading venue within the Union with the highest turnover; (c)for financial instruments other than those referred to in points (a) and (b), the market value shall be determined on the basis of a predetermined methodology approved by the competent authority of the relevant CSD that refers to criteria related to reliable market data, such as market prices available across trading venues or investment firms.

260. To further incentivise settlement efficiency, ESMA proposes an overall moderate increase of the penalty rates for most of the asset classes in the case of fails due to lack of financial instruments, using as a benchmark the average securities lending and borrowing rates (with reference to SFTR data available to ESMA). This increase would however not apply to fails due to lack of liquid shares (which were already subject to a penalty rate higher than the relevant average securities lending and borrowing rate), and due to lack of financial instruments traded on an SME Growth Markets, in order to maintain the incentive for market financing for SMEs.

261. Regarding the specific case of ETFs, ESMA notes mixed views for the introduction of a new type of fail due to the lack of an ETF type, as defined in article 4(1)(46) of MiFID II, and identified with specific CFI codes in the MiFIR identifier table27 available on ESMA’s website. ESMA also considers that the introduction of this new type of fail can be considered a more fundamental change to the penalty mechanism.

[27] https://www.esma.europa.eu/sites/default/files/library/2016-1523annex9.11_cfi-rts2_field_mapping_rev.2.xlsx

262. In addition, while ETF fail rates remain considerably high, a more thorough analysis may be needed to assess to what extent the low settlement efficiency levels may be due to structural reasons and to the intrinsic characteristics of ETFs, also, to the extent possible, by comparison with other markets (US, UK).

263. ESMA therefore proposes to increase the penalty rate by 50% to 0,75 basis point for the current fail type 7 (for all other financial instruments not covered by the other types of fails), without creating a new type of fail. This moderate increase aims to incentivise a significant reduction of the levels of settlement fails for ETFs, while recognising that frictional inefficiencies may originate from the intrinsic characteristics of ETFs.

264. In summary, ESMA proposes to introduce the following changes to penalty rates (see the summary table further below):

o maintain the current penalty rates for settlement fails due to lack of liquid shares (type 1), and due to lack of instruments traded on an SME growth market (types 3 and 6);

o increase penalty rates by 50 % for settlement fails due to lack of illiquid shares (type 2), bonds other than sovereign bonds (type 5) and all other financial instruments including ETFs (type 7); and by 100% for settlement fails due to lack of sovereign bonds (type 4);

o increase the floor from 0 to 1 for the penalty rate for settlement fails due to lack of cash (type 8).

Type of fail	Current penalty rate	Proposed change
1. Settlement fail due to a lack of shares that have a liquid market within the meaning of point (b) of Article 2(1)(17) of	1.0 basis point	Maintain at 1.0 basis point
Regulation (EU) No 600/2014, excluding shares referred to in point 3
2. Settlement fail due to a lack of shares that do not have a liquid market within the meaning of point (b) of Article 2(1)(17) of Regulation (EU) No 600/2014, excluding shares referred to in point 3	0.5 basis point	Increase by 50% to 0.75 basis points
3. Settlement fail due to a lack of financial instruments traded on SME growth markets, excluding debt instruments referred to in point 6	0.25 basis point	Maintain at 0.25 basis point
4. Settlement fail due to a lack of debt instruments issued or guaranteed by: (a) a sovereign issuer as defined in Article 4(1)(60) of Directive 2014/65/EU; (b) a third country sovereign issuer; (c) a local government authority; (d) a central bank; (e) any multilateral development bank referred to in the second subparagraph of Article 117(1) and in Article 117(2) of Regulation (EU) No 575/2013 of the European Parliament and of the Council; (f) the European Financial Stability Facility or the European Stability Mechanism.	0.10 basis point	Increase by 100% to 0.20 basis point
5. Settlement fail due to a lack of debt instruments other than those referred to in points 4 and 6	0.20 basis point	Increase by 50% to 0.30 basis point
6. Settlement fail due to a lack of debt instruments traded on SME growth markets	0,15 basis point	Maintain at 0.15 basis point

Settlement fail due to a lack of all
other financial instruments not covered in points 1 to 6 0,5 basis point Increase by 50% to 0.75 basis point
Settlement fail due to a lack of cash Official interest rate for
overnight credit charged by the central bank issuing the settlement currency with a floor of 0 Official interest rate for overnight credit charged by the central bank issuing the settlement currency with a floor increased to 1.
28

TABLE 3: PROPOSED CHANGES IN PENALTY RATES PER FAIL TYPE

265. ESMA recommends this moderate increase in some of the penalty rates as an additional incentive for market participants to take the necessary steps (such as through further automation of processes, increased standardisation and harmonisation of market practices) to improve settlement efficiency and to address the root causes of settlement fails, in line with the CSDR Refit objective. Such efforts are also considered an integral part of preparing for a shortening of the settlement cycle in the EU. This moderate increase is without prejudice to a more significant increase of penalty rates and/or more structural changes to the CSDR penalty mechanism in the medium term, depending on the evolution of settlement efficiency trends.

[28] These penalty rates are applied irrespective of the currency, with alternative parameters applied when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available (see section 3).

6 Annexes

6.1 Annex I – Summary of Questions

6.1.1 Alternative parameters, when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available

Q1: Do you agree with ESMA’s proposal? Which Option is preferable in your view? Please also state the reasons for your answer.

Q2: Do you have other suggestions? If yes, please specify and provide arguments.

Q4: What costs and benefits do you envisage related to the implementation of each Option? Please use the table below. Where relevant, additional tables, graphs and information may be included in order to support some of the arguments or calculations presented in the table below.

Option
	Qualitative description	Quantitative description/ Data
Benefits
Compliance costs: - One-off - On-going
Costs to other stakeholders
Indirect costs

6.1.2 Treatment of historical reference data for the calculation of late matching fail penalties

Q6. What are the causes of late matching? How can you explain that there are so many late matching instructions lasting during a very long period? What measures could be envisaged in order to reduce the number of late matching instructions?

a)92 business days;

b)40 business days;

c)other (please specify).

Q9: Do you agree that the issuer CSD for each financial instrument shall be responsible for confirming the relevant reference data to be used for the related penalties calculation? Please also state the reasons for your answer.

Q11: Do you have other suggestions? If yes, please specify, provide drafting suggestions and provide arguments including data where available.

Q12: Do you agree with the approach followed to incorporate proportionality in the Technical Advice? If not, please provide an indication of further proportionality considerations, detailed justifications and alternative wording as needed.

Q13: What costs and benefits do you envisage related to the implementation of the approach proposed by ESMA? Please use the table below. Where relevant, additional tables, graphs and information may be included in order to support some of the arguments or calculations presented in the table below.

Approach proposed by ESMA
	Qualitative description	Quantitative description/ Data
Benefits
Compliance costs: - One-off - On-going
Costs to other stakeholders
Indirect costs

Q14: If applicable (if you have suggested a different approach than the one proposed by ESMA), please specify the costs and benefits you envisage related to the implementation of the respective approach. Please use the table below. Where relevant, additional tables, graphs and information may be included in order to support some of the arguments or calculations presented in the table below.

Approach proposed by respondent (if applicable)
	Qualitative description	Quantitative description/ Data
Benefits
Compliance costs: - One-off - On-going
Costs to other stakeholders
Indirect costs

6.1.3 Alternative methods for calculating cash penalties, including progressive penalty rates

Impact of current penalty mechanism

Q15: Based on your experience, what has been the impact of CSDR cash penalties on reducing settlement fails (by type of asset as foreseen in the Annex to Commission Delegated Regulation (EU) 2017/389 since the application of the regime in February 2022? Please provide data and arguments to justify your answer.

Q16: In your view, is the current CSDR penalty mechanism deterrent and proportionate? Does it effectively discourage settlement fails and incentivise their rapid resolution? Please provide data and arguments to justify your answer.

Q17: What are the main reasons for settlement fails, going beyond the high level categories: “fail to deliver securities”, “fail to deliver cash” or “settlement instructions on hold”? Please provide examples and data, as well as arguments to justify your answer.

Q18: What tools should be used in order to improve settlement efficiency? Please provide examples and data, as well as arguments to justify your answer.

Q19: What are your views on the appropriate level(s) of settlement efficiency at CSD/SSS level, as well as by asset type? Please provide data and arguments to justify your answer.

Q20: Do you think the penalty rates by asset type as foreseen in the Annex to Commission Delegated Regulation (EU) 2017/389 are proportionate? Please provide data and arguments to justify your answer.

Q21: Regarding the proportionality of the penalty rates by asset type as foreseen in the Annex to Commission Delegated Regulation (EU) 2017/389, ESMA does not have data on the breakdown of cash penalties (by number and value) applied by CSDs by asset type. Therefore, ESMA would like to use this CP to ask for data from all EEA CSDs on this breakdown, including on the duration of settlement fails by asset type.

Progressive penalty rates

Q22: In your view, would progressive penalty rates that increase with the length of the settlement fail be justified? Please provide examples and data, as well as arguments to justify your answer.

Q23: What are your views regarding the introduction of convexity in penalty rates as per the ESMA proposed Option 2 (settlement fails caused by a lack of liquid financial instruments)? Please justify your answer by providing quantitative examples and data if possible.

Q24: Would it be appropriate to apply the convexity criterion to settlement fails due to a lack of illiquid financial instruments as well? Please justify your answer by providing quantitative examples and data if possible.

Q25: What are your views regarding the level of progressive penalty rates:

a) as proposed under Option 1?

b) as proposed under Option 2?

Q26: If you disagree with ESMA’s proposal regarding the penalty rates, please specify which rates you believe would be more appropriate (i.e. deterrent and proportionate, with the potential to effectively discourage settlement fails, incentivise their rapid resolution and improve settlement efficiency). Please provide examples and data, as well as arguments to justify your answer. If relevant, please provide an indication of further proportionality considerations, detailed justifications and alternative proposals as needed.

Q27: What are your views regarding the categorisation of types of fails:

a) as proposed under Option 1?

b) as proposed under Option 2?

Do you believe that less/further granularity is needed in terms of the types of fails (asset classes) subject to cash penalties? Please justify your answer by providing quantitative examples and data if possible.

Q28: What costs and benefits do you envisage related to the implementation of progressive penalty rates by asset type (according to ESMA’s proposed Options 1 and 2)? Please use the table below. Where relevant, additional tables, graphs and information may be included in order to support some of the arguments or calculations presented in the table below.

Progressive penalty rates (by asset type) - ESMA’s proposal Option 1	Please see ESMA’s proposed Option 1 in Section 5.3 of this CP.

	Qualitative description	Quantitative description/ Data
Benefits
Compliance costs: - One-off - On-going
Costs to other stakeholders
Indirect costs
Progressive penalty rates (by asset type) - ESMA’s proposal Option 2	Please see ESMA’s proposed Option 2 in Section 5.3 of this CP.
	Qualitative description	Quantitative description/ Data
Benefits
Compliance costs: - One-off - On-going
Costs to other stakeholders
Indirect costs

Q29: Alternatively, do you think that progressive cash penalties rates should take into account a different breakdown than the one included in ESMA’s proposal above for any or all of the following categories:

(a) asset type;

(b) liquidity of the financial instrument;

(d) duration of the settlement fail.

If you have answered yes to the question above, what costs and benefits do you envisage related to the implementation of progressive penalty rates according to your proposal? Please use the table below. Where relevant, additional tables, graphs and information may be included in order to support some of the arguments or calculations presented in the table below.

Progressive penalty rates – respondent's proposal (if applicable)
	Qualitative description	Quantitative description/ Data
Benefits
Compliance costs: - One-off - On-going
Costs to other stakeholders
Indirect costs

Q30: Another potential approach to progressive penalty rates could be based not only on the length of the settlement fail but also on the value of the settlement fail. Settlement fails based on instructions with a lower value could be charged a higher penalty rate than those with a higher value, thus potentially creating an incentive for participants in settling smaller value instructions at their intended settlement date (ISD). Alternatively, settlement fails based on instructions with a higher value could be charged a higher penalty rate than those with a lower value. In your view, would such an approach be justified? Please provide arguments and examples in support of your answer, including data where available. What costs and benefits do you envisage related to the implementation of this approach? Please use the table below. Where relevant, additional tables, graphs and information may be included in order to support some of the arguments or calculations presented in the table below.

Progressive penalty rates – based on the length and value of the settlement fail	Settlement fails based on lower value settlement instructions could be charged a higher penalty rate than those based on higher value settlement instructions	Settlement fails based on higher value settlement instructions could be charged a higher penalty rate than those based on lower value settlement instructions
	Qualitative description	Quantitative description/ Data	Qualitative description	Quantitative description/ Data
Benefits
Compliance costs:
- One-off - On-going
Costs to other stakeholders
Indirect costs

Q31: Besides the criteria already listed, i.e. type of asset, liquidity of the financial instruments, duration and value of the settlement fail, what additional criteria should be considered when setting proportionate and effective cash penalty rates? Please provide examples and justify your answer.

Additional considerations to simplify the cash penalty mechanism, while ensuring it is deterrent and proportionate

Q32: Would you be in favour of the use of the market value of the financial instruments on the first day of the settlement fail as a basis for the calculation of penalties for the entire duration of the fail? ESMA would like to ask for the stakeholders’ views on the costs and benefits of such a measure. Please use the table below. Where relevant, additional tables, graphs and information may be included in order to support some of the arguments or calculations presented in the table below.

Use the market value of the financial instruments on the first day of the settlement fail as a basis for the calculation of penalties for the entire duration of the fail
	Qualitative description	Quantitative description/ Data
Benefits
Compliance costs: - One-off - On-going
Costs to other stakeholders
Indirect costs

Q33: How should free of payment (FoP) instructions be valued for the purpose of the application of cash penalties? Please justify your answer and provide examples and data where available.

Q34: Do you think there is a risk that higher penalty rates may lead to participants using less DvP and more FoP settlement instructions? Please justify your answer and provide examples and data where available.

Q35: ESMA is considering the feasibility of identifying another asset class subject to lower penalty rates: “bonds for which there is not a liquid market in accordance with the methodology specified in Article 13(1), point (b) of Commission Delegated Regulation (EU) 2017/583 (RTS 2)”. The information on the assessment of bonds’ liquidity is published by ESMA on a quarterly basis and further updated on FITRS. However, ESMA is also aware that this may add to the operational burden for CSDs that would need to check the liquidity of bonds before applying cash penalties. As such, ESMA would like to ask for the stakeholders’ views on the costs and benefits of such a measure. Please use the table below. Where relevant, additional tables, graphs and information may be included in order to support some of the arguments or calculations presented in the table below.

Applying lower penalty rates for illiquid bonds
	Qualitative description	Quantitative description/ Data
Benefits
Compliance costs: - One-off - On-going
Costs to other stakeholders
Indirect costs

Q36: Do you have other suggestions for further flexibility with regards to penalties for settlement fails imposed on illiquid financial instruments? Please justify your answer and provide examples and data where available.

Q37: How likely is it that underlying parties that end up with “net long” cash payments may not have incentives to manage their fails or bilaterally cancel failing instructions as they may “earn” cash from penalties? How could this risk be addressed? Please justify your answer and provide examples and data where available.

Q38: How could the parameters for the calculation of cash penalties take into account the effect that low or negative interest rates could have on the incentives of counterparties and on settlement fails? Please provide examples and data, as well as arguments to justify your answer.

Q39: To ensure a proportionate approach, do you think the penalty mechanism should be applied only at the level of those CSDs with higher settlement fail rates? Please provide examples and data, as well as arguments to justify your answer. If your answer is yes, please specify where the threshold should be set and if it should take into account the settlement efficiency at:

a) CSD/SSS level (please specify the settlement efficiency target);

b) at asset type level (please specify the settlement efficiency target); or

c) other (please specify, including the settlement efficiency target).

Q40: Please specify what costs and benefits you envisage regarding the application of the penalty mechanism only at the level of the CSDs with higher settlement fail rates. Please use the table below. Where relevant, additional tables, graphs and information may be included in order to support some of the arguments or calculations presented in the table below.

Application of the penalty mechanism only at the level of CSDs with lower settlement fail rates
	Qualitative description	Quantitative description/ Data
Benefits
Compliance costs: - One-off - On-going
Costs to other stakeholders
Indirect costs

Q41: Do you think penalty rates should vary according to the transaction type? If yes, please specify the transaction types and include proposals regarding the related penalty rates. Please justify your answer and provide examples and data where available. Please specify what costs and benefits you envisage related to the implementation of your proposal. Please use the table below. Where relevant, additional tables, graphs and information may be included in order to support some of the arguments or calculations presented in the table below.

Applying penalty rates by transaction types
	Qualitative description	Quantitative description/ Data
Benefits
Compliance costs: - One-off - On-going
Costs to other stakeholders
Indirect costs

Q42: Do you think that penalty rates should depend on stock borrowing fees? If yes, do you believe that the data provided by data vendors is of sufficient good quality that it can be relied upon? Please provide the average borrowing fees for the 8 categories of financial instruments depicted in Option 1 (i.e. liquid shares, illiquid shares, SME shares, ETFs, sovereign bonds, SME bonds, other corporate bonds, other financial instruments).

Q43: Do you have other suggestions to simplify the cash penalty mechanism, while ensuring it is deterrent and proportionate, and effectively discourages settlement fails, incentivises their rapid resolution and improves settlement efficiency? Please justify your answer and provide examples and data where available. Please specify what costs and benefits you envisage related to the implementation of your proposal. Please use the table below. Where relevant, additional tables, graphs and information may be included in order to support some of the arguments or calculations presented in the table below.

Respondent’s proposal (if applicable)
	Qualitative description	Quantitative description/ Data
Benefits
Compliance costs: - One-off - On-going

ESMA European Securities and Markets Authority

Costs to other stakeholders
Indirect costs

Q44: Based on your experience, are settlement fails lower in other markets (i.e USA, UK)? If so, which are in your opinion the main reasons for that? Please also specify the scope and methodology used for measuring settlement efficiency in the respective third-country jurisdictions.

Q45: Do CSD participants pass on the penalties to their clients? Please provide information about the current market practices as well as data, examples and reasons, if any, which may impede the passing on of penalties to clients.

Q46: Do you consider that introducing a minimum penalty across all types of fails would improve settlement efficiency? Is yes, what would be the amount of this minimum penalty and how should it apply? Please provide examples and data, as well as arguments to justify your answer.

Q47: What would be the time needed for CSDs and market participants to implement changes to the penalty mechanism (depending on the extent of the changes)? Please provide arguments to justify your answer.

Ad hoc measures for CSDs participants with high settlement fail rates

Q48: Since the application of the RTS on Settlement Discipline, how many participants have been detected as failing consistently and systematically within the meaning of Article 7(9) of CSDR? How many of them, if any, have been suspended pursuant to same Article?

Q49: In your view, would special penalties (either additional penalties or more severe penalty rates) applied to participants with high settlement fail rates be justified? Should such participants be identified using the same thresholds as in Article 39 of the RTS on Settlement Discipline, but within a shorter timeframe (e.g. 2 months instead of 12 months)? If not, what criteria/methodology should be used for defining participants with high settlement fail rates? Please provide examples and data, as well as arguments to justify your answer.

Q50: How have CSDs implemented working arrangements with participants in accordance with article 13(2) of the RTS on Settlement Discipline? How many participants have been targeted?

Q51: Should the topic of settlement efficiency be discussed at the CSDs’ User Committees to better identify any market circumstances and particular context of participant(s) explaining an increase or decrease of the fail rates? Please justify your answer.

6.2 Annex II – EC Mandate regarding Technical Advice on the
parameters for the calculation of cash penalties for settlement fails caused by a lack of cash

REQUEST TO THE EUROPEAN SECURITIES AND MARKETS AUTHORITY (ESMA) FOR TECHNICAL ADVICE ON A POSSIBLE AMENDMENT TO THE DELEGATED ACT SPECIFYING THE PARAMETERS FOR THE CALCULATION OF CASH PENALTIES FOR SETTLEMENT FAILS CAUSED BY A LACK OF CASH (Ref: Ares(2022)8651438 – 13/12/2022)

Commission Delegated Regulation (EU) 2017/389

With this mandate the European Commission seeks ESMA's technical advice on a possible amendment to the delegated act29 specifying the parameters for the calculation of cash penalties under the Central Securities Depositories Regulation (CSDR) 30 . This amendment to the delegated act should be adopted in accordance with Article 290 of the Treaty on the Functioning of the European Union (TFEU).

The Commission reserves the right to revise and/or supplement this mandate. The technical advice received on the basis of this mandate should not prejudge the Commission's final decision.

The mandate follows the CSDR, the Communication from the Commission to the European Parliament and the Council – Implementation of Article 290 of the Treaty on the Functioning of the European Union (the "290 Communication")31, and the Framework Agreement on Relations between the European Parliament and the European Commission (the "Framework Agreement")32 .

The European Parliament and the Council shall be duly informed about this mandate.

[29] Commission Delegated Regulation (EU) 2017/389 of 11 November 2016 supplementing Regulation (EU) No 909/2014 of the European Parliament and of the Council as regards the parameters for the calculation of cash penalties for settlement fails and the operations of CSDs in host Member States.

[30] Regulation (EU) No 909/2014 of the European Parliament and of the Council of 23 July 2014 on improving securities settlement in the European Union and on central securities depositories.

[31] Communication of 9.12.2009. COM (2009) 673 final.

[32] OJ L 304, 20.11.2010, p. 47.

In accordance with the Declaration 39 on Article 290 TFEU, annexed to the Final Act of the Intergovernmental Conference which adopted the Treaty of Lisbon, signed on 13 December 2007, and in accordance with the established practice within the European Securities Committee33, the Commission will continue, as appropriate, to consult experts appointed by the Member States in the preparation of possible delegated acts in the financial services area. In accordance with point 15 of the Framework Agreement, the Commission will provide full information and documentation on its meetings with experts appointed by the Member States within the framework of its work on the preparation and implementation of Union legislation, including soft law and delegated acts. Upon request by the Parliament, the Commission may also invite Parliament's experts to attend those meetings. The powers of the Commission to adopt delegated acts are subject to Article 67 of CSDR. As soon as the Commission adopts a possible delegated act, the Commission will notify it simultaneously to the European Parliament and the Council.

1. Context

1.1 Scope

The Central Securities Depositories Regulation (CSDR) includes a set of measures to prevent and address failures in the settlement of securities transactions (settlement fails), commonly referred to as settlement discipline measures. They consist of reporting requirements, cash penalties for Central Securities Depositories’ (CSD) participants in case of settlement fails, and mandatory buy-ins where a CSD participant fails to deliver the security within a fixed extension period.

The objective of the cash penalties is to act as a deterrent for participants that cause settlement fails, by charging the failing party a daily penalty for each business day that a transaction fails to settle after the intended settlement date. Commission Delegated Regulation (EU) 2017/389 defines the parameters and methodology for the calculation of the level of cash penalties that CSDs will impose on and collect from the failing participants in their securities settlement systems. Specifically, Article 2 states that “…the level of cash penalties referred to in the third subparagraph of Article 7(2) of Regulation (EU) No 909/2014 for settlement fails of transactions in a given financial instrument shall be calculated by applying the relevant penalty rate set out in the Annex to this Regulation to the reference price of the transaction determined in accordance with Article 3 of this Regulation...”. Accordingly, the Annex to the Delegated Regulation specifies penalty rates applicable to settlement fails. In the case of settlement fails due to a lack of cash (point 8 of the Annex) the applicable rate should be the official interest

[33] Commission's Decision of 6.6.2001 establishing the European Securities Committee, OJ L 191, 17.7.2001, p. 45.

rate for overnight credit charged by the central bank issuing the settlement with a floor of 0 (“zero”).

In order to discourage settlement fails due to a lack of cash, it is appropriate to use the costs of borrowing cash as a basis for the penalty rate. The most appropriate penalty rate should be the official interest rate of the central bank issuing the settlement currency that should evidence the borrowing costs for that currency. For instance, in the case of Euro-settled transactions this would be the rate on the marginal lending facility, which is the interest rate banks pay when they borrow money overnight from the European Central Bank (ECB).

CSDR or the relevant Delegated Regulation do not provide a common definition of the overnight credit rate to be applied by CSDs or an alternative proxy interest rate for calculating the cost of borrowing in case a central bank overnight lending facility does not exist for the settlement currency. This makes it difficult to apply a penalty to a settlement fail caused by a lack of cash in the concerned currency as required under point 8 of the Annex to the Delegated Regulation (EU) 2017/389. Alternative calculation methodologies or rates have been proposed, as evidenced by the CSDR Penalties Framework 34 of the European Central Securities Depositories Association (ECSDA). Nevertheless, no common calculation method or agreement on the variables used to calculate the alternative rate has been developed. Currently different settlement currencies use different domestic rates or a combination of a domestic benchmark rate and spread of key ECB interest rates35. This leads to a situation where different calculation methodologies can lead to varying degrees of severity of the cash penalties regime applied to settlement fails caused by a lack of cash, entrenching fragmentation of the European capital market by making settlement fails relatively less costly in some markets.

In light of the above, the Commission kindly asks ESMA to suggest a possible amendment to Commission Delegated Regulation (EU) 2017/389. Such amendment could take the form of:

A comprehensive definition of an official interest rate for overnight credit charged by the central bank issuing the settlement currency as used in the Annex of the Delegated Regulation,
A methodology for calculating an appropriate cost of capital rate to be applied to cash penalties calculations in the absence of a short-term interest rate charged by central banks when extending short-term loans to commercial banks,

[34] ECSDA CSDR Penalties Framework, update October 2021, p.42-43. Please see: https://ecsda.eu/wp-content/uploads/2021/10/2021_10_05_ECSDA_CSDR_Penalties_Framework.pdf

[35] Even when an ECB spread is applied it is calculated differently, as evidenced by the proposals from DK (which uses the spread between the marginal lending facility and the ECB rate on deposit facility) or BG (spread between the marginal lending facility rate and interest rate on main refinancing operations).

Indicate alternative interest rates to be applied to failing transactions in currencies where the relevant central bank does not offer an overnight credit facility.

1.2 Principles that ESMA should take into account

On the working approach, ESMA is invited to take account of the following principles:

The principle of proportionality: the technical advice should not go beyond what is necessary to achieve the objective of the Regulation. It should be simple and avoid excessive financial, administrative or procedural burdens for counterparties and financial infrastructure providers, in particular CSDs.
When preparing its advice, ESMA should seek coherence within the regulatory framework of the Union.
In accordance with the Regulation of the European Parliament and the Council establishing a European Securities and Markets Authority (the "ESMA Regulation")36, ESMA should not feel confined in its reflection to elements that it considers should be addressed by the amendment to the delegated act but, if it finds it appropriate, it may indicate guidelines and recommendations which, in its view, could be appropriate to accompany the delegated act to better ensure its effectiveness.
ESMA will determine its own working methods depending on the content of the provisions being dealt with. Nevertheless, horizontal questions should be dealt with in such a way as to ensure coherence between different standards of work being carried out by the various expert groups.
In accordance with the ESMA Regulation, ESMA should, where relevant, involve the European Banking Authority and the European System of Central Banks in order to ensure cross-sectoral consistency. It should also cooperate, where relevant, with the European Systemic Risk Board on any issues related to systemic risk.
In accordance with the ESMA Regulation, ESMA is invited to widely consult market participants in an open and transparent manner, and take into account the resulting opinions in its advice. ESMA should provide a detailed feedback statement on the consultation, specifying when consultations took place, how many responses were received and from whom, as well as the main arguments for and against the issues raised. This feedback statement should be annexed to its technical advice. The technical advice should justify ESMA’s choices vis-à-vis the main arguments raised during the consultation.

[36] Regulation (EU) No 1095/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Securities and Markets Authority), OJ L 331, 15.12.2010, p. 84.

ESMA - 201-203 rue de Bercy - CS 80910 - 75589 Paris Cedex 12 - France - Tel. +33 (0) 1 58 36 43 21 - www.esma.europa.eu 91

ESMA is invited to justify its advice by providing a quantitative and qualitative cost-benefit analysis of all the options considered and proposed. ESMA should provide the Commission with a description of the problem, the objectives of the technical advice, possible options for consideration and a comparison of the main arguments for and against the considered options. The cost-benefit analysis should justify ESMA’s choices vis-à-vis the main considered options.
ESMA’s technical advice should not take the form of a legal text. However, ESMA should provide the Commission with a clear and structured ("articulated") text, accompanied by sufficient and detailed explanations. Furthermore, the technical advice should be presented in an easily understandable language respecting current terminology in the Union.
ESMA should provide comprehensive technical analysis on the subject matters described in section 3 below, where these are covered by the delegated powers included in:

o the relevant provision of the Regulation as amended;

o the corresponding recitals; or

o the relevant Commission's request included in this mandate.
ESMA should address to the Commission any question to clarify the text of the Regulation or the relevant Regulatory Technical Standard it considers of relevance to the preparation of its technical advice.

1. Procedure

The Commission is requesting ESMA’s technical advice in view of the preparation of an amendment of Commission Delegated Regulation (EU) 2017/389 on the calculation of penalties in particular regarding the questions referred to in section 3 of this mandate.

The mandate takes into account the CSDR (Articles 7(5) and 67), the ESMA Regulation, the Communication on the implementation of Article 290 of the Treaty on the Functioning of the European Union 37 and the Framework Agreement on relations between the European Parliament and the European Commission38 .

The Commission reserves the right to revise and/or supplement this mandate. The technical advice received on the basis of this mandate will not prejudge the Commission's final decision.

In accordance with established practice, the Commission may continue to consult experts appointed by the Member States in the preparation of the amendment to the delegated act.

[37] Communication from the Commission to the European Parliament and the Council - Implementation of Article 290 of the Treaty on the Functioning of the European Union, COM(2009) 673

[38] OJ L 304, 20.11.2010, p. 47.

The Commission shall duly inform the European Parliament and the Council about this mandate. As soon as the Commission adopts the delegated act, it will notify it simultaneously to the European Parliament and the Council.

2. ESMA is invited to provide technical advice on the following issues

ESMA is invited to provide technical advice to assist the Commission in amending Delegated Regulation (EU) 2017/389. In particular, this advice should specify which alternative rate or methodology should be applied in the calculation of cash penalties for settlement fails caused by a lack of cash where no overnight credit rate charged by the central bank issuing the settlement currency exists.

In order to ensure a deterrent effect of cash penalties and incentivise timely settlement by failing participants, the penalty rate should reflect the borrowing costs for that currency. ESMA should ensure that the applicable interest rate is set such that the level of cash penalties provides incentives to failing participants to promptly settle failed transactions, without endangering the integrity of the EU capital market. Simultaneously when defining the alternatives their impact on the level of penalties and on the market should be considered. In particular, the proposed rate should not lead to further fragmentation of the single market for capital. Moreover, considering the automation of calculation of cash penalties the proposed alternative rate should be easy to source and compute.

The Delegated Regulation notes that the most appropriate benchmark of borrowing costs in the calculation of a penalty rate is the official interest rate of the central bank issuing the settlement currency39. Other potential substitute interest rates40 exist on the national and EU capital markets. Although some of them are set without the involvement of a central bank, they reflect the borrowing costs on the commercial inter-bank market and are used in several securities settlement systems41. The technical advice should reflect upon the relevance of these proxy rates for the calculation of cash penalties in case of settlement fails caused by a lack of cash in light of the requirements of the Delegated Regulation (in particular Recital 12 and point 8 of the Annex).

3. Indicative timetable

[39] Commission Delegated Regulation (EU) 2017/389, Recital 12.

[40] An example is the EURIBOR, the Euro Interbank Offered Rate, based on the averaged interest rates at which Eurozone banks offer to lend unsecured funds to other banks in the euro wholesale money market. It is published daily by the European Money Market Institutes. Similar rates exist for inter-bank markets loans in non-Euro currencies and are frequently compiled and published by national central banks.

[41] For instance, CREST used LIBOR, now SONIA, in the calculation of cash penalties.

object to a delegated act within a period of 3 months, extendible by 3 further months at the initiative of the European Parliament or of the Council. The delegated act will only enter into force if neither European Parliament nor the Council have objected on expiry of that period or if both institutions have informed the Commission of their intention not to raise objections.

The obligation by CSDs to calculate and collect cash penalties on participants to their securities settlement systems that cause settlement fails will enter into force on 01 February 2022. Although industry-led alternatives are in place, the amendment to the delegated act should be in place as soon as possible to ensure a coherent application of the measures to monitor and prevent settlement fails across the EU capital market. It is therefore of outmost importance to start the work on this issue as soon as possible.

The deadline set to ESMA to deliver the technical advice is therefore 30 September 202442 .

[42] According to the Commission request for ESMA technical advice on amendments to Commission Delegated Regulation (EU) 2017/389 ref. Ares(2023)5817200 – 28/08/2023.

6.3 Annex III – EC Mandate regarding Technical Advice on the treatment of historical reference data for the calculation of late matching fail penalties (LMFPs)

REQUEST TO THE EUROPEAN SECURITIES AND MARKETS AUTHORITY (ESMA) FOR TECHNICAL ADVICE ON A POSSIBLE AMENDMENT TO THE DELEGATED ACT SPECIFYING THE TREATMENT OF HISTORICAL REFERENCE DATA FOR THE CALCULATION OF LATE MATCHING FAIL PENALTIES (LMFPs) (Ref: Ares(2023)3379353-15/05/2023)

(Commission Delegated Regulation (EU) 2017/389)

With this mandate the European Commission seeks ESMA's technical advice on a possible amendment to the delegated act43 specifying the retention process for the parameters used in the calculation of Late Matching Fail Penalties under the Central Securities Depositories Regulation (CSDR)44. This amendment to the delegated act should be adopted in accordance with Article 290 of the Treaty on the Functioning of the European Union (TFEU).

The Commission reserves the right to revise and/or supplement this mandate. The technical advice received on the basis of this mandate should not prejudge the Commission's final decision.

For reasons of work planning this technical advice should be combined with the technical advice request on alternative interest rates to be applied to settlement fails caused by a lack of cash, sent to ESMA on 14 December 202245 .

The mandate follows the CSDR, the Communication from the Commission to the European Parliament and the Council – Implementation of Article 290 of the Treaty on the Functioning of the European Union (the "290 Communication")46, and the Framework Agreement on Relations between the European Parliament and the European Commission (the "Framework Agreement")47 .

According to Article 7(5) of the CSDR, the Commission is empowered to adopt delegated acts in accordance with Article 67 of the CSDR to specify parameters for the calculation of a deterrent

[43] Commission Delegated Regulation (EU) 2017/389 of 11 November 2016 supplementing Regulation (EU) No 909/2014 of the European Parliament and of the Council as regards the parameters for the calculation of cash penalties for settlement fails and the operations of CSDs in host Member States.

[44] Regulation (EU) No 909/2014 of the European Parliament and of the Council of 23 July 2014 on improving securities settlement in the European Union and on central securities depositories.

[45] Request for ESMA technical advice on amendments to Commission Delegated Regulation (EU) 2017/389 relating to setting appropriate penalty rates in case of settlement fails caused by a lack of cash under the Central Securities Depositories Regulation (CSDR), Letter from John BERRIGAN to Verena ROSS, 14 December 2022, ARES Ref: 9591987.

[46] 46Communication of 9.12.2009. COM (2009) 673 final.

[47] OJ L 304, 20.11.2010, p. 47.

and proportionate level of cash penalties based on asset type and liquidity of the financial instrument and type of transaction that shall ensure a high degree of settlement discipline and the smooth and orderly functioning of the financial markets concerned.

The European Parliament and the Council shall be duly informed about this mandate.

In accordance with point 15 of the Framework Agreement, the Commission will provide full information and documentation on its meetings with experts appointed by the Member States within the framework of its work on the preparation and implementation of Union legislation, including soft law and delegated acts. Upon request by the Parliament, the Commission may also invite Parliament's experts to attend those meetings.

The powers of the Commission to adopt delegated acts are subject to Article 67 of the CSDR. As soon as the Commission adopts a possible delegated act, the Commission will notify it simultaneously to the European Parliament and the Council.

2. Context

1.3 Scope

The CSDR includes a set of measures to prevent and address failures in the settlement of securities transactions (settlement fails), commonly referred to as settlement discipline measures. They consist of reporting requirements, cash penalties for Central Securities Depositories’ (CSD) participants in case of settlement fails, and mandatory buy-ins where a CSD participant fails to deliver the security within a fixed extension period. Cash penalties are being applied to all failing transactions on the EU capital market as of 1 February 2022.

The objective of the cash penalties is to act as a deterrent for participants that cause settlement fails, by charging the failing party a daily penalty for each business day that a transaction fails to settle after the intended settlement date (ISD). Cash penalties are calculated as from the intended settlement date until the actual settlement or (bilateral) cancellation date of the instruction. Cash penalties also apply to settlement fails due to matching of settlement instructions after their ISD (late matching).

Commission Delegated Regulation (EU) 2017/389 defines further the parameters and methodology for the calculation of the level of cash penalties that CSDs are required to impose

[48] Commission's Decision of 6.6.2001 establishing the European Securities Committee, OJ L 191, 17.7.2001, p. 45.

on and collect from the failing participants in their securities settlement systems. Specifically, Article 2 states that “…the level of cash penalties referred to in the third subparagraph of Article 7(2) of Regulation (EU) No 909/2014 for settlement fails of transactions in a given financial instrument shall be calculated by applying the relevant penalty rate set out in the Annex to this Regulation to the reference price of the transaction determined in accordance with Article 3 of this Regulation...”.

If the trade is matched after the ISD, the trade is subject to a late matching fail penalty (LMFP) imposed on the participant that has submitted the instruction last. The penalties are levied for each day between the ISD and until the instruction is settled or (bilaterally) cancelled. The CSDR settlement discipline provisions imply that LMFPs must be calculated for settlement fails with an intended settlement date for any point in time as of 1 February 2022 onwards. This poses a challenge for any IT system, i.e. to calculate settlement fails for any given day in the past means that the related historical reference data must be kept available in the system in case a late matching settlement instruction is submitted in the system. This means that the amount of reference data is gradually increasing every business day.

This accumulation of historical reference data may have an impact on all EU CSDs as well as, most notably, on TARGET2-Securities (T2S) where the accumulation of past data over time will degrade the functioning of the system, even if the number and share of LMFPs represent a small proportion of penalties in the T2S.

In light of the above, the Commission kindly asks ESMA to suggest a possible amendment to Commission Delegated Regulation (EU) 2017/389. Such amendment should clarify the calculation method for LMFPs that prevents the accumulation of reference date over time and ensures the efficient operation of securities settlement systems.

1.4 Principles that ESMA should take into account

On the working approach, ESMA is invited to take account of the following principles:

The principle of proportionality: the technical advice should not go beyond what is necessary to achieve the objective of the Regulation. It should be simple and avoid excessive financial, administrative or procedural burdens for counterparties and financial infrastructure providers, in particular CSDs.
When preparing its advice, ESMA should seek coherence within the regulatory framework of the Union.
In accordance with the Regulation of the European Parliament and the Council establishing a European Securities and Markets Authority (the "ESMA Regulation")49, ESMA should not feel confined in its reflection to elements that it considers should be addressed by the amendment to the delegated act but, if it finds it appropriate, it may indicate guidelines and recommendations which, in its view, could be appropriate to accompany the delegated act to better ensure its effectiveness.
ESMA will determine its own working methods depending on the content of the provisions being dealt with. Nevertheless, horizontal questions should be dealt with in such a way as to ensure coherence between different standards of work being carried out by the various expert groups.
In accordance with the ESMA Regulation, ESMA should, where relevant, involve the European Banking Authority and the European System of Central Banks in order to ensure cross-sectoral consistency. It should also cooperate, where relevant, with the European Systemic Risk Board on any issues related to systemic risk.
In accordance with the ESMA Regulation, ESMA is invited to widely consult market participants in an open and transparent manner and take into account the resulting opinions in its advice. ESMA should provide a detailed feedback statement on the consultation, specifying when consultations took place, how many responses were received and from whom, as well as the main arguments for and against the issues raised. This feedback statement should be annexed to its technical advice. The technical advice should justify ESMA’s choices vis-à-vis the main arguments raised during the consultation.
ESMA is invited to justify its advice by providing a quantitative and qualitative cost-benefit analysis of all the options considered and proposed. ESMA should provide the Commission with a description of the problem, the objectives of the technical advice, possible options for consideration and a comparison of the main arguments for and against the considered options. The cost-benefit analysis should justify ESMA’s choices vis-à-vis the main considered options.
ESMA’s technical advice should not take the form of a legal text. However, ESMA should provide the Commission with a clear and structured ("articulated") text, accompanied by sufficient and detailed explanations. Furthermore, the technical advice should be presented in an easily understandable language respecting current terminology in the Union.
ESMA should provide comprehensive technical analysis on the subject matters described in section 3 below, where these are covered by the delegated powers included in:

o the relevant provision of the Regulation as amended;

[49] Regulation (EU) No 1095/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Securities and Markets Authority), OJ L 331, 15.12.2010, p. 84.

o the corresponding recitals; or

o the relevant Commission's request included in this mandate.

ESMA should address to the Commission any question to clarify the text of the Regulation or the relevant Regulatory Technical Standard it considers of relevance to the preparation of its technical advice.

3. Procedure

The mandate takes into account the CSDR (Articles 7(5) and 67), the ESMA Regulation, the Communication on the implementation of Article 290 of the Treaty on the Functioning of the European Union 50 and the Framework Agreement on relations between the European Parliament and the European Commission51 .

The Commission reserves the right to revise and/or supplement this mandate. The technical advice received on the basis of this mandate will not prejudge the Commission's final decision.

In accordance with established practice, the Commission may continue to consult experts appointed by the Member States in the preparation of the amendment to the delegated act.

4. ESMA is invited to provide technical advice on the following issues

ESMA is invited to provide technical advice to assist the Commission in amending Delegated Regulation (EU) 2017/389. This advice should specify how to deal with reference data accumulation caused by the need to calculate LMFPs, in particular by suggesting appropriate methods to calculate settlement fails penalties and handle reference data underlying transactions that are matched after the ISD.

5. Indicative timetable

[50] Communication from the Commission to the European Parliament and the Council - Implementation of Article 290 of the Treaty on the Functioning of the European Union, COM(2009) 673.

[51] OJ L 304, 20.11.2010, p. 47.

are subject to Article 67 of the CSDR that allows the European Parliament and the Council to object to a delegated act within a period of 3 months, extendible by 3 further months at the initiative of the European Parliament or of the Council. The delegated act will only enter into force if neither European Parliament nor the Council have objected on expiry of that period or if both institutions have informed the Commission of their intention not to raise objections.

It is of outmost importance to start the work on this issue as soon as possible. For reasons of work planning this technical advice should be combined with the technical advice request on alternative interest rates to be applied to settlement fails caused by a lack of cash, sent to ESMA on 14 December 202252. The deadline set to ESMA to deliver the technical advice is therefore 30 September 202453 .

[52] Request for ESMA technical advice on amendments to Commission Delegated Regulation (EU) 2017/389 relating to setting appropriate penalty rates in case of settlement fails caused by a lack of cash under the Central Securities Depositories Regulation (CSDR), Letter from John BERRIGAN to Verena ROSS, 14 December 2022, ARES Ref: 9591987

[53] According to the Commission request for ESMA technical advice on amendments to Commission Delegated Regulation (EU) 2017/389 ref. Ares(2023)5817200 – 28/08/2023.

6.4 Annex IV – EC Mandate regarding Technical Advice on the potential calibration of the structure and severity of cash penalties to discourage settlement fails, incentivise their rapid resolution and improve settlement efficiency

REQUEST TO THE EUROPEAN SECURITIES AND MARKETS AUTHORITY (ESMA) FOR TECHNICAL ADVICE ON POSSIBLE AMENDMENTS TO THE DELEGATED ACT SPECIFYING THE POTENTIAL CALIBRATION OF THE STRUCTURE AND SEVERITY OF CASH PENALTIES TO DISCOURAGE SETTLEMENT FAILS, INCENTIVISE THEIR RAPID RESOLUTION AND IMPROVE SETTLEMENT EFFICIENCY Ref: Ares(2023)5817200- 28/08/2023)

(Commission Delegated Regulation (EU) 2017/389)

With this provisional mandate, the European Commission seeks ESMA's technical advice on a possible amendment to the delegated act54 specifying the calculation method and penalty rates under the Central Securities Depositories Regulation (CSDR) 55 . This amendment to the delegated act should be adopted in accordance with Article 290 of the Treaty on the Functioning of the European Union (TFEU).

The Commission reserves the right to revise and/or supplement this mandate. The technical advice received on the basis of this mandate should not prejudge the Commission's final decision.

For reasons of work planning this technical advice may be combined with the earlier technical advice requests on (i) alternative interest rates to be applied to settlement fails caused by a lack of cash and (ii) late matching fails penalties, sent to ESMA on 14 December 202256 and 15 May 202357 respectively.

The provisional mandate follows the CSDR, the Communication from the Commission to the European Parliament and the Council – Implementation of Article 290 of the Treaty on the Functioning of the European Union (the "290 Communication"), 58 and the Framework

[54] Commission Delegated Regulation (EU) 2017/389 of 11 November 2016 supplementing Regulation (EU) No 909/2014 of the European Parliament and of the Council as regards the parameters for the calculation of cash penalties for settlement fails and the operations of CSDs in host Member States.

[55] Regulation (EU) No 909/2014 of the European Parliament and of the Council of 23 July 2014 on improving securities settlement in the European Union and on central securities depositories.

[56] Request for ESMA technical advice on amendments to Commission Delegated Regulation (EU) 2017/389 relating to setting appropriate penalty rates in case of settlement fails caused by a lack of cash under the Central Securities Depositories Regulation (CSDR), Letter from John BERRIGAN to Verena ROSS, 14 December 2022, ARES Ref: 9591987.

[57] Request for ESMA technical advice on amendments to Commission Delegated Regulation (EU) 2017/389 relating to the treatment of historical reference data for the calculation of LMFPs under the CSDR, Letter from John BERRIGAN to Verena ROSS, 15 May 2023, ARES Ref: Ares(2023)3379353

[58] Communication of 9.12.2009. COM (2009) 673 final.

Agreement on Relations between the European Parliament and the European Commission (the "Framework Agreement").59

According to Article 7(5) CSDR, the Commission is empowered to adopt delegated acts in accordance with Article 67 CSDR to specify parameters for the calculation of a deterrent and proportionate level of cash penalties based on asset type and liquidity of the financial instrument and type of transaction that shall ensure a high degree of settlement discipline and the smooth and orderly functioning of the financial markets concerned. Despite the application of cash penalties since 1 February 2022, settlement efficiency has not improved noticeably. As such, further actions must be explored to improve settlement efficiency on the EU capital market, including a reshaping of the structure and severity of cash penalties.

The European Parliament and the Council shall be duly informed about this mandate.

The powers of the Commission to adopt delegated acts are subject to Article 67 CSDR. As soon as the Commission adopts a possible delegated act, the Commission will notify it simultaneously to the European Parliament and the Council.

1. Context

1.1 Scope

The Central Securities Depositories Regulation (CSDR) includes a set of measures to prevent and address failures in the settlement of securities transactions (settlement fails), commonly referred to as settlement discipline measures. They consist of reporting requirements, cash penalties for central securities depositories’ (CSD) participants in case of settlement fails, and mandatory buy-ins where a CSD participant fails to deliver the security within a fixed extension period.

Although settlement fails cannot be totally eliminated, persistent settlement fails negatively affect the functioning and competitiveness of the capital market. It is understood that the European capital market is characterised by higher settlement fails than in other developed

[59] OJ L 304, 20.11.2010, p. 47.

[60] Commission's Decision of 6.6.2001 establishing the European Securities Committee, OJ L 191, 17.7.2001, p. 45.

financial markets. Furthermore, EU capital markets persistently affected by high settlement fail rates contradict the objectives of the Capital Markets Union61, which aims to improve the functioning of market infrastructures across the EU. A fully functioning and integrated market for capital will allow the EU’s economy to grow in a sustainable way and be more competitive.

Cash penalties should deter participants from causing settlement fails: the failing party is charged a daily penalty for each business day that a transaction fails to settle after the intended settlement date (ISD).

Article 7(5) CSDR specifies that cash penalties must ensure a high degree of settlement discipline and the smooth and orderly functioning of the financial markets concerned. In order to achieve this aim, the same Article empowers the Commission to adopt delegated acts in accordance with Article 67 CSDR to specify parameters for the calculation of a deterrent and proportionate level of cash penalties.

Commission Delegated Regulation (EU) 2017/389 defines further the parameters and methodology for the calculation of the level of cash penalties that CSDs will impose on and collect from the failing participants in their securities settlement systems. In particular, Article 2 states that [t]he level of cash penalties referred to in the third subparagraph of Article 7(2) of Regulation (EU) No 909/2014 for settlement fails of transactions in a given financial instrument shall be calculated by applying the relevant penalty rate set out in the Annex to this Regulation to the reference price of the transaction determined in accordance with Article 3 of this Regulation.

Cash penalties are being applied to all failing settlement instruction in EU CSDs as of 1 February 2022. Unfortunately, the effect of cash penalties on settlement rates on the EU capital market does not seem to show a clear improvement of settlement efficiency62. While the settlement fail rate for equities seems to improve slowly, settlement fails rates for corporate and government bonds alike appear to have been deteriorating since February 2022.

The co-legislators recently concluded negotiations on the review of the CSDR. The provisional agreement on CSDR maintains mandatory buy-ins as part of the settlement discipline toolkit. However, they will only apply as a measure of last resort where the rate of settlement fails in the EU is not improving and is presenting a threat to financial stability. Hence, to ensure that mandatory buy-ins are a necessary, appropriate and proportionate means to address the level of settlement fails on the EU capital market the full potential of other measures, in particular cash penalties, to address settlement fails must be explored. This indicates that cash penalties will play an even greater role in ensuring settlement discipline in the future and points to the need to reassess the current framework.

In light of the above, the Commission asks ESMA to suggest a possible amendment to Commission Delegated Regulation (EU) 2017/389. The Agency should assess the

[61] Communication from the Commission, A Capital Markets Union for people and businesses – New Action Plan, COM(2020) 590 final

[62] “Report on Trends, Risks and Vulnerabilities”, European Securities and Markets Authority, ESMA50-165-2438, No. 1, 2023, Graph 45

effectiveness and proportionality of the current penalty mechanism and propose, if justified, changes to the structure or severity of the mechanism and consider alternative methods for calculating cash penalties, including by introducing progressive penalty rates. In drafting its technical advice, ESMA should consider how the changing interest rate environment, including negative interest rates, affect a participant’s incentive to fail and how this could be mitigated. Furthermore, ESMA should reflect on the need for further flexibility with regards to penalties for settlement fails imposed on illiquid financial instruments. The proposed amendments to the structure and severity of the mechanism should effectively discourage settlement fails, incentivise their rapid resolution and improve settlement efficiency.

1.2 Principles that ESMA should take into account

On the working approach, ESMA is invited to take account of the following principles:

The principle of proportionality: the technical advice should not go beyond what is necessary to achieve the objective of the Regulation. It should be simple and avoid excessive financial, administrative or procedural burdens for counterparties and financial infrastructure providers, in particular CSDs.
When preparing its advice, ESMA should seek coherence within the regulatory framework of the Union.
In accordance with the Regulation of the European Parliament and the Council establishing a European Securities and Markets Authority (the "ESMA Regulation")63 , ESMA should not feel confined in its reflection to elements that it considers should be addressed by the amendment to the delegated act but, if it finds it appropriate, it may indicate guidelines and recommendations which, in its view, could be appropriate to accompany the delegated act to better ensure its effectiveness.
ESMA will determine its own working methods depending on the content of the provisions being dealt with. Nevertheless, horizontal questions should be dealt with in such a way as to ensure coherence between different standards of work being carried out by the various expert groups.
In accordance with the ESMA Regulation, ESMA should, where relevant, involve the European Banking Authority and the European System of Central Banks in order to ensure cross-sectoral consistency. It should also cooperate, where relevant, with the European Systemic Risk Board on any issues related to systemic risk.
In accordance with the ESMA Regulation, ESMA is invited to widely consult market participants in an open and transparent manner, and take into account the resulting opinions in its advice. ESMA should provide a detailed feedback statement on the consultation, specifying when consultations took place, how many responses were received and from whom, as well as the main arguments for and against the issues raised. This feedback statement should be annexed to its technical advice. The technical advice should justify ESMA’s choices vis-à-vis the main arguments raised during the consultation.

[63] Regulation (EU) No 1095/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Securities and Markets Authority), OJ L 331, 15.12.2010, p. 84.

ESMA is invited to justify its advice by providing a quantitative and qualitative cost-benefit analysis of all the options considered and proposed. ESMA should provide the Commission with a description of the problem, the objectives of the technical advice, possible options for consideration and a comparison of the main arguments for and against the considered options. The cost-benefit analysis should justify ESMA’s choices vis-à-vis the main considered options.
ESMA’s technical advice should not take the form of a legal text. However, ESMA should provide the Commission with a clear and structured ("articulated") text, accompanied by sufficient and detailed explanations. Furthermore, the technical advice should be presented in an easily understandable language respecting current terminology in the Union.
ESMA should provide comprehensive technical analysis on the subject matters described in section 3 below, where these are covered by the delegated powers included in:

o the relevant provision of the Regulation as amended;

o the corresponding recitals; or

o the relevant Commission's request included in this mandate.
ESMA should address to the Commission any question to clarify the text of the Regulation or the relevant Regulatory Technical Standard it considers of relevance to the preparation of its technical advice.

2 Procedure

The mandate takes into account the CSDR (Articles 7(5) and 67), the ESMA Regulation, the Communication on the implementation of Article 290 of the Treaty on the Functioning of the European Union (TFEU)64 and the Framework Agreement on relations between the European Parliament and the European Commission65 .

The Commission reserves the right to revise and/or supplement this mandate. The technical advice received on the basis of this mandate will not prejudge the Commission's final decision.

In accordance with established practice, the Commission may continue to consult experts appointed by the Member States in the preparation of the amendment to the delegated act.

[64] Communication from the Commission to the European Parliament and the Council - Implementation of Article 290 of the Treaty on the Functioning of the European Union, COM(2009) 673.

[65] OJ L 304, 20.11.2010, p. 47.

3 ESMA is invited to provide technical advice on the following issues

ESMA is invited to provide technical advice to assist the Commission in amending Delegated Regulation (EU) 2017/389. In particular, in order to ensure a continuing deterrent effect of cash penalties and incentivise timely settlement by failing participants, the technical advice should specify if and how to amend the structure and severity of cash penalties to discourage settlement fails, incentivise the timely resolution of fails and discourage keeping settlement instructions unsettled for extended periods of time.

The technical advice should contribute to lowering settlement fail rates on the EU capital market and in turn contribute to the objectives of the Savings and Investments Union. In its advice ESMA should take account of the different types of securities, their time to maturity and liquidity, where appropriate. In addition, ESMA should reflect if the changing interest rate environment affects the participants incentives to fail and how this should be accommodated in revised penalty rates, if justified. Lastly, the proposed methodology for calculating settlement fails penalties should not lead to further fragmentation of the single market for capital. Moreover, to support the automation of calculation of cash penalties, the proposed alternative rates should be easy to source and compute.

4. Indicative timetable

This mandate takes into consideration that ESMA requires sufficient time to prepare its technical advice and that the Commission needs to adopt the amended delegated act according to Article 290 TFEU. The powers of the Commission to adopt delegated acts are subject to Article 67 CSDR that allows the European Parliament and the Council to object to a delegated act within a period of 3 months, extendible by 3 further months at the initiative of the European Parliament or of the Council. The delegated act will only enter into force if neither European Parliament nor the Council have objected on expiry of that period or if both institutions have informed the Commission of their intention not to raise objections.

It is of outmost importance to start the work on this issue as soon as possible. For reasons of work planning and overlapping scope this technical advice may be combined with the earlier technical advice requests on alternative interest rates to be applied to settlement fails caused by a lack of cash and late matching fails penalties, sent to ESMA on 14 December 202266 and 15 May 2023 respectively 67.The deadline set to ESMA to deliver the technical advice is therefore 30 September 2024.

[66] Request for ESMA technical advice on amendments to Commission Delegated Regulation (EU) 2017/389 relating to setting appropriate penalty rates in case of settlement fails caused by a lack of cash under the Central Securities Depositories Regulation (CSDR), Letter from John BERRIGAN to Verena ROSS, 14 December 2022, ARES Ref: 9591987.

[67] Request for ESMA technical advice on amendments to Commission Delegated Regulation (EU) 2017/389 relating to the treatment of historical reference data for the calculation of LMFPs under the CSDR, Letter from John BERRIGAN to Verena ROSS, 15 May 2023, ARES Ref: Ares(2023)3379353

6.5 Annex V: Cost-benefit analysis

6.5.1 Alternative parameters, when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available

Problem identification

1. Propose a clear and reliable alternative methodology to calculate cash penalties for settlement fails caused by a lack of cash, where no overnight credit rate charged by the central bank issuing the settlement currency exists.

Policy objectives

2. The proposed methodologies to calculate cash penalties, when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available, should ensure that it is cheaper to borrow cash to settle the transaction than to pay the penalties and obtain interest on the unpaid cash.

Baseline scenario

3. No common calculation method or agreement on the variables used to calculate an alternative rate has been developed to date when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available. Currently, different settlement currencies use different domestic rates or a combination of a domestic benchmark rate and spread of key ECB interest rates. Bulgaria is expected to join the Eurozone in 2024/2025 (exact date to be confirmed), therefore it appears that the only EEA jurisdiction without an official interest rate for overnight credit for the national currency will be Denmark.

Options considered and preferred option

4. Four methodologies were considered:

Option 1

STEP 1: converting the respective cash amount subject to penalties into EUR (using the official exchange rate - the conversion rate should be from the same day as the reference data for the calculation of penalties for a given business day) STEP 2: applying the ECB marginal lending facility rate (interest rate for overnight credit from the Eurosystem) to the converted cash amount

STEP 3: where needed, converting the result back into the original currency (using the official exchange rate - the conversion rate should be from the same day as the reference data for the calculation of penalties for a given business day)

Option 2

STEP 1: calculating the spread on the shortest maturity (e.g. 1 month) market rates available for EUR and currency X (currency without an official interest rate for overnight credit)

STEP 2: adding the ECB marginal lending facility rate (interest rate for overnight credit from the Eurosystem) to calculate the penalty rate to be applied for the other currency

Option 3

STEP 1: Identification of comparable interest rates of the ECB and the central bank. In the example those interest rates are ECB’s “main refinancing operations” (MROs) and the corresponding interest rate for weekly credit of the central bank.

STEP 2: The comparable interest rate of the central bank (the interest rate for weekly credit) is added the spread between the ECBs deposit facility rate and the ECBs marginal lending rate. The spread is added because a weekly credit rate will typically be lower than an overnight credit rate. Adding the spread will compensate for this difference in order for the interest rate used to calculate the cash penalties is as close to the official interest rate for overnight credit of the ECB.

In the absence of an overnight interest credit rate due to the monetary policy of the central bank issuing the settlement currency, other comparable interest rates of the ECB and the relevant central bank could be used to calculate a proxy which a CSD can use to calculate the cash penalties due to lack of cash. It is a precondition for this alternative method that the interest rates of the ECB and the relevant central bank are comparable. The proxy is calculated by the central bank issuing the settlement currency and must be recalculated whenever either the interest rates of the ECB and/or the interest rates of the central bank issuing the settlement currency are subject to changes.

Option 4

Option 4 goes further than addressing the situation when the official interest rate for overnight credit charged by the central bank issuing the settlement currency is not available. Under this option, higher fixed rates for settlement fails due to lack of cash, irrespective of the currency, would apply depending on the length of the settlement fail:

Daily penalty rates for settlement fails due to lack of cash 10 basis points - 1st BD of fail 15 basis points – 2nd BD of fail 20 basis points – 3rd BD of fail 25 basis points – 4th BD of fail 30 basis points – starting with 5th BD of fail

5. Option 1, envisaging currency conversion, would require multiple IT developments making the related high costs disproportionate compared to other options. Option 2 and 3 do not entail high implementing costs.

6. Option 4 is the costliest option to implement, as it requires structural changes. Its methodology, envisaging progressive penalty rates, may lead to high rates which are deemed disproportionate.

Preferred option

7. Option 3 is considered the preferred option because it is easier to implement, being the one more closely aligned with the current methodology used in Denmark. Option 3 will not require functional changes to the T2S penalty mechanism, hence high costs to implement, while offering a viable alternative to calculate the penalties.

6.5.2 Treatment of historical reference data for the calculation of late matching fail penalties

Problem identification

8. The current CSDR settlement discipline regime requires CSDs to deal with a considerable amount of reference data, as CSDs must calculate LMFPs for settlement fails with an ISD for any point in time. The consequent accumulation of data poses challenges to IT systems that store the data for the LMFPs calculations, especially considering that the amount of reference data gradually increases every business day, with no possibility for historical reference data deletion. This accumulation of historical reference data may have an impact on all EU CSDs as well as, most notably, on T2S where the accumulation of past data over time will degrade the performance of the system.

Policy objectives

9. This Technical Advice’s objective is to present a methodology that enables CSDs to calculate settlement fails penalties in the most accurate way possible, while being able to effectively handle reference data underlying transactions that are matched after the ISD.

Baseline scenario

10. The CSDR provisions on settlement discipline do not set a threshold on the use of actual daily reference data for LMFPs calculations. Under the current regime, actual daily reference data must be stored and made available by the system, without an envisaged cut-off time beyond which oldest available reference data may be used.

Options considered

11. ESMA considered a 92 business days threshold and a 40 business days threshold for the use of actual versus historical reference data.

12. The costs of extending the threshold for calculation of LMFP to 92 business days are considered disproportionate with respect to the related benefits. Such change would imply very high development and yearly costs for the T2S operator, while allowing to calculate only a very low percentage of cash penalties with reference data going beyond 40 business days68 .

Preferred option

13. On the basis of the proportionality principle, a 40 business days threshold appears to be adequate for the mandatory use of actual daily reference data. Beyond such

[68] The T2S operator has raised a change request to investigate the implementation efforts and costs of extending the threshold for calculation of LMFP from 40 to 92 days. As per the change request detailed assessment, its implementation would imply 925k€ of development costs and around 200k€ of yearly running & maintenance costs. These costs would only allow T2S to perform LMFP calculation with the correct (older) reference price for an additional 0.02% of the penalties in T2S platform, as per July-October 2023 data. On top of T2S operator costs, development, testing, and running costs for CSDs and their participants should also be considered.

threshold, the oldest available reference data may be used for the calculation of cash penalties. This proposed threshold would serve well the objectives of preventing the accumulation of reference data over time and ensuring the efficient operation of securities settlement systems, while also enabling the application of effective and deterrent penalties. ESMA also considers that CSDs that can keep and have access to older historical reference data should be able to use the respective data for the calculation of cash penalties for LMFPs going beyond 40 business days in the past.

14. Having regard to the above, ESMA proposes the following amendments to Article 3 of Commission Delegated Regulation (EU) 2017/389:

Article 3

Reference data price of the transaction

3. Where settlement instructions have been matched after the intended settlement date, and that intended settlement date is within 40 business days in the past from the matching date, the actual daily reference data, such as reference prices and exchange rates, shall be used for the calculation of the related cash penalties.

Where settlement instructions have been matched after the intended settlement date, and that intended settlement date is beyond 40 business days in the past from the matching date, the oldest available reference data may be used for the calculation of the related cash penalties.

In the cases mentioned in the second subparagraph, CSDs which are part of an interoperable link shall establish arrangements for determining the relevant reference data to be used for the related penalties calculation.

In the cases mentioned in the second subparagraph, CSDs shall communicate to their competent authorities the procedure determining the oldest available reference data they may use for the calculation of the related cash penalties.

6.5.3 Alternative methods for calculating cash penalties, including progressive penalty rates

Problem identification

15. Since their application to failing settlement instructions in EU CSDs as of 1 February 2022, cash penalties have not proven sufficiently deterrent to incentivise a sustained improvement of settlement efficiency.

16. The analysis of the ratio of settlement fails over all settlement instructions, both in terms of value and volume, demonstrates an overall decrease of settlement fails since February 2022. This decrease can however be considered subdued, and certain asset classes, in particular ETFs, are still associated with high levels of settlement fails.

Policy objectives

17. The primary goal of cash penalties is to lower settlement fail rates and attain a high degree of settlement discipline, in turn ensuring the smooth and orderly functioning of the financial markets and of market infrastructures across the EU, one of the core objectives of the Savings and Investments Union.

18. Cash penalties should not only deter participants from causing settlement fails, but also incentivise the failing party to rapidly resolve the settlement fail: according to the third subparagraph of Article 7(2) of CSDR, the failing party is charged a daily penalty for each business day that a transaction fails to settle after the ISD.Moreover, following the provisions of CSDR Refit that make the mandatory buy-in process a last resort measure, the full potential of other measures, in particular cash penalties, to address settlement fails must be explored.

19. Last but not least, a low level of settlement fails is essential in light of the ongoing discussions about a shortening of the settlement cycle in the EU. This indicates that cash penalties will play an even greater role in ensuring settlement discipline in the future and points to the need to reassess the current framework.

20. The calculation of cash penalties should be simple, i.e. easy to source and compute, and harmonised, to prevent further fragmentation of the single market for capital.

Baseline scenario

21. Overall, the application of cash penalties with the current structure and severity has led to a subdued improvement of settlement efficiency. In accordance with ESMA report on Trends, Risks and Vulnerabilities No.1 2024, “settlement fails are occurring less frequently, particularly for corporate bonds […] and for equity instruments […].

However, for equity instruments most of this impact occurred in 1H23 with no further noticeable declines in 2H23.” 69

22. While it is challenging to distinguish the impact on settlement efficiency of the application of cash penalties from the impact of the overall economic and interest rate environment, the levels of settlement fails, in particular for ETFs, remain high.

23. It can be concluded that the current structure and severity of the cash penalty mechanism does not lead to a continuous and durable improvement of settlement efficiency.

Options considered

24. ESMA has considered a variety of policy options, affecting both the structure and the severity of the cash penalty mechanism, to increase the deterrent effect of the current mechanism while maintaining its proportionality for CSDs and market participants.

25. The feedback from respondents to the ESMA CP does not allow to draw up a detailed quantitative analysis of the costs and benefits of the options considered. A qualitative assessment has nonetheless been included in ESMA’s analysis, based on the feedback from most respondents that an overhaul of (some aspects) of the penalty mechanism would lead to significant IT development costs for all EEA CSDs, and for market participants.

Option 1 : Do nothing (baseline)

26. ESMA has considered introducing no changes to the penalty mechanism, considering the feedback received on the limited period during which the current structure and severity has been applied.

27. Benefits in terms of increased settlement efficiency are expected to be limited, as data on settlement efficiency rates demonstrates that the effects of the application of the penalty mechanism have been subdued following an initial period of adjustment.

28. No additional costs related to the penalty mechanism can be expected both for CSDs and for market participants if the status quo is maintained.

[69] Report on Trends, Risks and Vulnerabilities Risk Monitor, European Securities and Markets Authority, ESMA50-524821-3107, No. 1, 2024, p. 22

Option 2 : Moderate increase of the severity of penalty rates

29. ESMA has considered two options for a moderate increase of the severity of penalty rates, without changes to the structure of the penalty mechanism.

(a) Option 2a: moderate increase with the current fail types

(b) Option 2b: moderate increase with the current fail types and a new fail type for ETFs

30. Benefits in terms of increased settlement efficiency are expected to be moderate, with a possibility to refine the impact on specific asset classes with different increases based on the fail type. In particular, the introduction of a new fail type for ETFs under option 2b could allow to better target the higher levels of settlement fails with a specific penalty rate for this asset class.

31. Option 2a is associated with low one-off costs for CSDs and market participants and low on-going costs for CSDs, as it would not require significant implementation costs to adapt the existing systems.

32. One-off costs for CSDs under option 2b are considered medium, as creating a new fail type for ETFs would represent a change to the existing systems requiring some implementation resources. Low on-going costs (if settlement fails decrease) or Medium on-going costs (if settlement fails don’t decrease) can also be expected for market participants, as increased penalty rates for certain asset classes will lead to higher penalties paid by market participants responsible for settlement fails.

Option 3 : Significant increase of the severity of penalty rates

33. As presented in the Consultation Paper, ESMA has considered several factors leading to a significant increase of the severity of penalty rates, which can be summarised in three options:

(a) Option 3a : significant increase without progressivity

(b) Option 3b: significant increase with progressivity

34. In all three options, benefits in terms of settlement efficiency would be significant, as the revised penalty rates would be a very strong deterrent for settlement fails. The adverse impacts of such prohibitive penalties on the smooth functioning of financial markets are however to be carefully considered.

35. On-going costs for market participants are expected to be high as participants responsible for settlement fails would be subject to significantly higher penalties.

36. Under option 3a, on-off costs for CSDs and market participants, and on-going costs for CSDs, would be low, as implementation would consist in increasing rates in the existing system. Options 3b and options 3c would however entail more complex implementation projects, with medium to high on-off costs, and more elaborate maintenance overtime, with medium to high on-going costs, as reference data for the penalty for each failed settlement instruction would evolve overtime in a non linear manner.

Option 4 : Structural changes to the penalty mechanism

37. ESMA’s Consultation Paper also includes a variety of suggestions for structural changes to the penalty mechanism, which can be categorised under three options:

(a) Option 4a: introduction of streamlined fail types (liquid financial instruments, illiquid financial instruments, cash)

(b) Option 4b: changes to the parameters defining the rate levels (e.g. explicit link between the penalty rates and interest rates)

(c) Option 4c: fundamental changes affecting the calculation of the value of the failed instructions, the setting of settlement efficiency targets and/or a specific focus on CSDs participants with high settlement fail rates

38. Across all these options, the benefits on settlement efficiency are unknown.

39. These options also entail very high on-off costs for both CSDs and market participants, due to the implementation of a major overhaul of existing systems. Some suggestions, in particular under options 4a and 4b, may however lead to medium to low on-going costs for CSDs and/or market participants, due to a simplification of the factors taken into account to calculate cash penalties.

Preferred option and next steps

40. In light of these considerations (see summary table below), ESMA believes option 2a to be the preferred option at this stage, and is advising the EC to reflect this option in Commission Delegated Regulation (EU) 2017/389 in the short term, ahead of a shortening of the settlement cycle in the EU.

41. The policy options described above will be further considered in the medium term, including subject to a potential request for technical advice ahead of the next review of Commission Delegated Regulation (EU) 2017/389, given that, according to the

ESMA European Securities and Markets Authority

last subparagraph of Article 7(5), the EC shall review the parameters for the calculation of the level of the cash penalties on a regular basis and at least every four years in order to reassess the appropriateness and effectiveness of the cash penalties in achieving a level of settlement fails in the Union deemed to be acceptable having regard to the impact on the financial stability of the Union.

Summary table of policy options considered (preferred option in bold)

Policy options	Benefits in terms of increased settlement efficiency	Costs for CSDs	Costs for market participants
Option 1 : Do nothing (baseline)		Limited	One-off costs: N/A On-going costs: N/A	One-off costs: N/A One-going costs: N/A
Option 2 : Moderate increase of the severity of penalty rates	Option 2a: moderate increase with current fail types	Moderate	One-off costs: Low On-going costs: Low	One-off costs: Low On-going costs: Low (if settlement fails decrease) or Medium (if settlement fails don’t decrease)
	Option 2b: moderate increase with new fail type for ETFs	Moderate	One-off costs: Medium On-going costs: Low	One-off costs: Low On-going costs: Low (if settlement fails decrease) or Medium (if settlement fails don’t decrease)

Policy options	Benefits in terms of increased settlement efficiency	Costs for CSDs	Costs for market participants
Option 3 : Significant increase of the severity of penalty rates	Option 3a : significant increase without progressivity	Significant	One-off costs: Low On-going costs: Low	One-off costs: Low On-going costs: Low (if settlement fails decrease) or High (if settlement fails don’t decrease)
	Option 3b: significant increase with progressivity	Significant	One-off costs: High On-going costs: Medium	One-off costs: Medium On-going costs: Low (if settlement fails decrease) or Medium (if settlement fails don’t decrease)
	Option 3c: significant increase with progressivity and convexity	Significant	One-off costs: Very high On-going costs: High	One-off costs: Medium On-going costs: Low (if settlement fails decrease) or Medium (if settlement fails don’t decrease)
Option 4 : Structural changes to the penalty mechanism	Option 4a: introduction of streamlined fail types (liquid financial	Unknown	One-off costs: Very high	One-off costs: Medium
instruments, illiquid financial instruments, cash)		On-going costs: Medium	On-going costs: Low (if settlement fails decrease) or Medium (if settlement fails don’t decrease)
Option 4b: changes to the parameters defining the rate levels (e.g. explicit link between the penalty rates and interest rates)	Unknown	One-off costs: Very high On-going costs: Medium	One-off costs: Medium On-going costs: Low (if settlement fails decrease) or Medium (if settlement fails don’t decrease)
Option 4c: fundamental changes affecting the calculation of the value of the failed instructions, the setting of settlement efficiency targets and/or a specific focus on CSDs participants with high settlement fail rates	Unknown	One-off costs: Very high On-going costs: High	One-off costs: Medium On-going costs: Low (if settlement fails decrease) or Medium (if settlement fails don’t decrease)