Executive summary

We question the relevance of/ added value derived from all but the largest building societies’ data to the Bank of England’s stated objectives in this exercise.  Secondly, any building society without a separate in-house treasury function – most of the sector - may struggle if required to report daily, without significant investment.  

Introduction

The Bank of England intends to collect transaction-level data from building societies, banks and PRA-designated investment firms on their secured and unsecured sterling money market borrowing activity.  The aim is to provide the Bank with a better understanding of developments in short-term interest rates.

To do this, firms that are most active in sterling money markets – up to 100 - will be required to submit a daily return on secured and unsecured transactions.  The rest will submit an annual return.  The daily reporting population will be confirmed once the first annual survey has been completed.

Feedback is also sought on the Bank’s plans to reform SONIA, a key overnight unsecured interest rate benchmark in sterling markets.  We have not commented on this aspect of the consultation as several large societies have already contributed to discussions led by a clearing bank.

We are most grateful to the Bank of England officials who took the time and trouble to discuss their plans with the BSA.

Questions

Q2 Is the Bank’s approach to identifying daily reporters based on covering 90% of market activity as identified by the annual return appropriate?

We agree that an annual return is a straightforward way of assessing initially which firms are most active in secured and unsecured borrowing.   All building societies should be able to comply with the annual return requirement as set out in the consultation.  But we have reservations on the proposed scope of the daily data collection exercise.  Such a high percentage figure could catch smaller firms such as building societies which do not have the capability to provide the daily transaction reports required by the BoE.  

One of the aims of the new return is to provide a better assessment of overall market effectiveness.  To determine this, we consider that the BoE does not need to set a threshold at 90% of market activity – indeed, it says in the consultation it wants to look at the firms that are “most active”.  90% is far more than “most active” – a better threshold might be 75%.  The “most active” in secured and unsecured sterling money market borrowing are generally speaking the largest firms.  Twenty to thirty firms (rather than the estimated 50-100) might well be sufficient to help the BoE.

Roll out of the data collection starts with “early adopters” – presumably greater in number than the six banks that took part in the pilot - in December; the rest come on stream four months later.   We suggest that full implementation is delayed for at least a year until the results of the early adopters have been analysed.   It is possible that the information the early adopters yield provides sufficient information for BoE purposes, or at least confirms that a smaller daily reporting set would be adequate.

Q3 The Bank plans to allow for a three month preparation period between daily data contributors being notified and the commencement of data submissions. Is this sufficient time to prepare for the daily collection? What would prevent you from meeting this timeline?

This depends on the individual firm’s systems capabilities.  Only the very largest building societies have separate in-house treasury functions that could report daily without significant outlay (even so, there are costs attached to those that can supply the information).  The others would face major expenditure to generate daily transactional reports that play no part in the day to day operations of their business.  Should such expenditure be required, a three month period is too short.  Projects would have to be scoped, expenditure agreed by boards, suppliers found and finally systems tested.  For some, this cycle could take up to a year.   

Q6 Does setting a threshold for the minimum transaction size make reporting more or less straightforward?

A minimum transaction size is an appropriate threshold and should be easy to capture.  But we question the level.  Secured sterling money market borrowing for many building societies – and probably banks - is usually done in larger tranches - a minimum say of £5 million or even £10 million.  We note for instance that the Bank’s own market operations, such as the ILTR, specify a minimum transaction size of £5 million. We therefore suggest that the threshold for a reportable transaction is raised to £5 million.

Q13 Would providing a contact for the Bank between 07.00 and 08.30 each reporting day be problematic?

Yet again, this is a question of size.  Front offices of large banks will be staffed at this time (and probably 24 hours due to their global nature).  This is not the case for building societies, even those with an in-house treasury function.  We therefore suggest that the contact requirement is either dropped or made more flexible.

Q17 Will the proposed timetable for the commencement of reporting present any major issues?

As explained in our answer to question 3, the handful of building societies with in-house treasury functions should be able to comply with the timetable but most others will not.  They do not have full capability to generate, verify and submit daily reports with minimal effort.  Arguably, their data will not enhance the BoE’s ability to understand developments in short-term interest rates.

Click here for the consultation paper
Read more in the policy brief (BSA members and associates only)