The banking institutions' liquidity reserves

This section assesses the extent of the Danish banking institutions' contingency measures to counter liquidity risk in the VP Settlement and the Sumclearing. In accordance with the usual hypothesis in international standards for payment and settlement systems, a large banking institution is assumed to be unable to meet its payment obligations and, thus, has to be excluded from the settlement. It is then analysed whether the other banking institutions have sufficient liquidity reserves to cover their recalculated net positions.

A banking institution's liquidity reserves for a given settlement day can be calculated as the sum of the following two items:

The balance of the banking institution's current account at Danmarks Nationalbank at the close of the preceding monetary-policy day, i.e. at 3.30 p.m.

Its holdings of eligible securities that have not already been pledged as collateral for a loan from Danmarks Nationalbank.

•  
• This measure expresses how much liquidity a banking institution can reserve for the settlement without borrowing from other parties besides Danmarks Nationalbank. It must be emphasised, however, that this by no means gives the complete picture of the banking institutions' access to provide liquidity for the settlement, since they may also borrow against simultaneously acquired securities as collateral via the automatic collateralisation arrangement.

The methodology of the calculations is described in more detail in Box 19. The analysis is based on all banking days in 2005 and covers the Danish banking institutions participating in the VP Settlement and/or the Sumclearing. For each day in the course of the year it is assumed that the banking institution with the largest net outgoing payment is unable to meet its payment obligation and has to be excluded from the settlement. The other participants' net positions are recalculated on the basis of their estimated bilateral net positions vis-à-vis the excluded participant, and the net positions are then compared with the banking institutions' liquidity reserves.

Table 12 summarises the results of the analysis for all banking institutions. The Table shows the number of days in 2005 when one or more banking institutions had a larger net outgoing payment in the VP Settlement or the Sumclearing than the liquidity reserves. The Table compares the actual pattern in 2005 with the situation where on each day the banking institution with the largest net outgoing payment has been excluded from the settlement.

On four days in 2005, one banking institution's total net outgoing payment exceeded its liquidity reserves. There were no days when this was the case for several banking institutions. If the banking institution with the largest payment obligation had to be excluded from the settlement on each day, there would be 17 days when the total net outgoing payment of one or more banking institutions exceeded their liquidity reserves. On a few days several banking institutions would be affected, but on most days only one.

In Table 12 especially the small banking institutions have a net outgoing payment that exceeds their liquidity reserves, which to some extent can be attributed to the methodology, cf. Box 19.

Chart 56 shows the 20 largest banking institutions' maximum net outgoing payments as a ratio of their liquidity reserves during 2005, where on each day the banking institution with the largest payment obligation is excluded from the settlement. It appears that only one banking institution, on one day, has a net outgoing payment that exceeds its liquidity reserves. On all days in 2005, the liquidity reserves of most of the other banking institutions considerably exceed their total net outgoing payments in the VP Settlement and the Sumclearing.

As mentioned above, the analysis to some extent underestimates the banking institutions' contingency measures to counter liquidity risk in the VP Settlement and the Sumclearing. Thus, in the VP Settlement the banking institutions may, in addition to the calculated liquidity reserves, pledge acquired securities as collateral for intraday credit already in the settlement cycle in which the securities are received. As a result, the banking institutions' real contingency liquidity is considerably higher than the calculated liquidity reserves. Furthermore, the banking institutions will rarely lack liquidity in the VP Settlement even if no payments are received from a large banking institution.

The analysis confirms that in overall terms the Danish banking institutions have ample contingency liquidity to counter the liquidity risk in VP Settlement and the Sumclearing. This applies especially to the largest banking institutions whose total net outgoing payments only rarely exceed their liquidity reserves, even when the banking institution with the largest payment obligation has to be excluded from the settlement. In addition, in the VP Settlement the banking institutions can pledge simultaneously acquired securities as collateral for intraday credit via the automatic collateralisation agreement, which in practice significantly increases their real contingency liquidity.

See BIS, Core Principles for Systemically Important Payment Systems, 2001, principles 3 and 5. BIS/IOSCO, Recommendations for Securities Settlement Systems, 2001, contains similar principles for securities settlement systems.

The VP Settlement and the Sumclearing are described in more detail in Danmarks Nationalbank, Payment Systems in Denmark, 2005, Chapters 5 and 6.

I.e. limits to amounts payable by one participant to other participants in the settlement cycle.

Payments above this limit are normally settled in a real-time gross settlement (RTGS) system. Danmarks Nationalbank's payment system, Kronos, is an example of an RTGS system.

The electronic clearing and truncation handles credit transfers, giro and cheque payments, as well as Dankort (debit card) payments using transaction vouchers. The PBS clearing processes payments related to PBS's own products, e.g. Betalingsservice (direct debit) and Dankort payments via payment terminals.

Directive 98/26/EC on settlement finality in payment and securities settlement systems, cf. Danmarks Nationalbank, Payment Systems in Denmark, 2005, Chapter 9.

The handling of an insolvent participant in the Sumclearing is described in detail in Niels C. Andersen, Clearing and Settlement in a Legal Perspective, Danmarks Nationalbank, Working Papers no. 20, 2004.

In electronic clearing and truncation, no single payment may exceed kr. 100 million, but most banking institutions have set their own considerably lower limits.

Some retail payment systems provide for settlement of payments already on the day they are effected, i.e. intraday settlement, implying even shorter exposure durations.

See also the detailed description of the automatic collateralisation arrangement in Danmarks Nationalbank, Payment Systems in Denmark, 2005, Chapter 5.

Danmarks Nationalbank, Financial stability 2002, contains a chapter on payment system risks, including an analysis of the consequences of having to exclude a banking institution from the Sumclearing settlement. The analysis shows that this in only very few cases entails that other banking institutions have inadequate cover for their recalculated net positions.

Furthermore, these measures increase the robustness of settlement in the Sumclearing to operational events (e.g. computer system failure) that may have significant liquidity consequences for the participants.

The analysis also includes banking institutions' payments in the CLS foreign-exchange settlement system. The settlement of payments in CLS is described in more detail in Danmarks Nationalbank, Payment Systems in Denmark, 2005, Chapter 5.

The monetary-policy day in kroner runs from 4.00 p.m. to 3.30 p.m. the next banking day and is the period during which the banking institutions have access to intraday credit at Danmarks Nationalbank.

Data for 20 December 2005 is disregarded, however, cf. Box 19.

Usually, a banking institution is still able to settle its payments even if it has a net outgoing payment that exceeds its liquidity reserve. This is due to the automatic collateralisation arrangement, which allows the participants to pledge securities not yet part of their liquidity reserves as collateral, cf. below.

See Danmarks Nationalbank, Financial stability 2005, Assessment of Settlement Risks in VP Securities Services, for an analysis of the liquidity risk in VP Settlement.