The Efficiency of Cross Border Settlement Systems

1
The Efficiency of Cross Border Settlement Systems

• May 29-June 2, 2006
• Shanghai, China
• Noritaka Akamatsu
• The World Bank

2
Contents

• Background
• Alternative channels of cross border settlement
• Existing networks of global custodians in the
  region,
• Institutional networks and system linkages among
  existing clearing houses and CSDs in the region,
• Risk management issues,
• Legal and regulatory infrastructure for ownership
  security and settlement finality,
• Multi-currency fund settlement arrangements,
• Business, political and technical feasibility of
  a regional clearing house.
• Capital account liberalization requirements,
• Strategy

3
Background

• Efficient and secure settlement is critical to
  promoting cross border trading of bonds in the
  region.
• Unless the settlement is possible, secure and
  efficient, investors will not trade across
  national borders.
• How can a trade between, e.g., a Korean investor
  and a Thai counterpart in Malaysian bonds kept in
  custody in Hong Kong be settled?
• National CSDs are now being linked up. But their
  settlement modes differ because each of them is
  designed to serve for domestic needs.
• Banking services are weak and different DVP modes
  prevail in different national markets.
• Bilateral links have not been a popular channel
  of cross-border settlement except when it
  involves an ICSD.

4
Background continued

• International investors without direct access to
  national CSDs rely on global custodians and their
  local agents.
• Liquidity pressure on local agents as repos (a
  popular instruments in cross border trading)
  and/or back-to-back transactions grow.
• Any room for use of an ICSD or creation of a
  regional CSD?

5
Trading needs and patterns

• Investors have a need and interest to manage
  exposures to regional risks and currencies.
• Often seek liquid government securities market
  where risk-adjusted interest rate is attractive
  and exit is easy to manage exchange rate risk.
• OECD experience
• Government securities repos and back-to-back
  trades by dealers and market makers have grown
  rapidly.
• Those created a pressure on intermediaries in the
  settlement chain (e.g., local agent
  sub-custodians).

6
Alternative channels
National CSD
Local agent
Local agent
Local agent
Country of issue
Other countries
Global custodian
ICSD
National CSD
Non-resident counterparty
Non-resident counterparty
Non-resident counterparty
Non-resident counterparty
Non-resident counterparty
Direct access
Local agent
Global custodian
ICSD
CSD-to-CSD
Source Cross-Border Securities Settlement, March
1995, CPSS, BIS
7
Alternative channels ofcross border settlement
1

• Direct access to local CSD.
• Non-resident often not permitted.
• Local money settlement arrangement necessary.
• Use of a local agent (custodian).
• Often provides pre-matching services and even
  settlement on its own book. But custody risks
  are involved.
• Use of bilateral links among national CSDs.
• A national CSD is often exclusively linked to a
  national stock exchange, which limits
  non-resident investors access to non-listed
  securities.
• Links are created primarily to allow domestic
  parties to trade in foreign securities rather
  than to trade across the border.
• Banking and cash management services are weak,
  and different DVP models prevail.

8
Bilateral network of national CSD
Central Securities Depository
CSD
CSD
CSD
Local Central Depository
Local Central Depository
CSD
CSD
9
Alternative Channels 2Use of global custodian

• Offer a single entry point for investors in
  multiple markets.
• Have a network of local agent sub-custodians
  (some of which may be their own branches or
  subsidiaries).
• Inexpensive service compared to separately
  accessing local custodians and other service
  providers thanks to economies of scale and scope
  despite IT investment required.
• Provide multi-currency banking services and
  management of cash and securities positions.
• Internal settlement over its own book.
• Can handle settlement of back-to-back trades.

10
Alternative Channels 3Use of ICSD

• Provide a single entry gateway to multiple
  markets as does a global custodian.
• Able to settle internally over its own book with
  direct links to national CSDs.
• I.e., can handle settlement of back-to-back
  trades.
• In Europe, those links were exploited to settle
  European government securities over ICSDs.
• Inexpensive service for core clearing and
  management of cash and securities positions.
• Use existing ICSDs or create a new regional ICSD
  in Asia??

11
Regional Custody Architecture - 1
Local Central Depository
LCD
LCD
Regional Hub
LCD
Local Central Depository
Local Central Depository
LCD
LCD
12
Alternative architectures in Europe

• Networking CSD model.
• Bilateral links among national CSDs, making each
  other a sub-depository (e.g., European Financial
  Superhighway by ECSDA)
• Hub and spoke model (ICSD)
• Euroclear
• Single CSD model (ICSD)
• Creation of Clearstream by merger between
  Deutsche Borse clearing house and Cedel.
• Controlling ownership of Clearstream by Deutsche
  Borse may be an impediment.

13
Regional networks of global custodians

• Global custodians active in emerging markets
  cross-border transactions
• Bank of New York
• JP Morgan (including Jardine Fleming)
• State Street
• Citibank
• Deutsche Bank
• PNB Paribas Those in bold Italic have
• Northern Trust strong local presence and
• ABN-AMRO Mellon client base in Asia.
• HSBC
• Morgan Stanley Global Custody
• MeesPierson

14
Regional networks of global custodians continued

• But there are differences among them. E.g.,
• State Street presence in regional hubs only but
  not in each individual country nor does it
  directly participate in local CSDs.
• Citibank presence in each individual country
  (i.e., local as well as global custodian) and
  directly participate in local CSDs.
• Their business strategies differ.
• State Street serving for global investors and
  interested in serving as a regional hub for
  international and intra-regional settlement.
• Citibank not interested in being a regional hub
  because it directly participate in local CSDs and
  can directly serve local clients (i.e., able to
  internalize regional settlement).

15
Agent sub-custodiansin East Asia

• 1999 2000 2001
• China (mainland) 5 4 5
• (Hong Kong) 6 7 7
• Indonesia 4 4 5
• Japan 6 5 9
• Korea 12 4 10
• Malaysia 10 5 9
• Philippines 5 4 5
• Singapore 6 5 8
• Thailand 9 5 9 .
• Q.How many of them are branches or subsidiaries
  of global custodians?

16
Networks linkages among clearing houses and CSDs

• Clearing houses and CSDs in the region.
• China CGSDTC and CSDCC
• Indonesia BI-SSSS, KSEI and KPEI
• Japan BOJ-Net and JASDEC
• Korea KSD
• Malaysia MCD, SCANS
• Philippines ROSS, SCCP and PCD
• Singapore Central Depository Ltd.
• Thailand Bhatnet 2 and TSD

17
Risk management

• Issues How to ensure cross border DVP?
• Systemic risks in links among settlement systems.
• Local sub-custodians under pressure to receive
  and redeliver securities on the same day.
• How to settle back-to-back trades?
• Links have been being built among CSDs and
  between CSDs and ICSDs. But
• Local CSDs built to meet local needs only.
• Different modes of DVP even when they are said to
  be achieved over electronic books.

18
Multi-currency cross-border fund settlement

• Are continuous linked settlement and CLS Bank
  needed to support the settlement in regional
  currencies?
• CLS bank has an account with each central bank in
  the region.
• Participating banks have an account with
  sub-accounts for different currencies at CLS bank
  (as well as accounts at the central bank).
• Participating bank pays to CLS bank account at
  the central bank by use of RTGS system, or
  instructs its nostro agent to pay to CLS bank
  account at the central bank of the agents
  economy.
• Is use of CLS Bank possible?
• CLS Bank does not provide settlement in Asian
  currencies other than Japanese yen.
• Create a regional CLS bank?? Does it make
  business sense??

19
Multi-currency cross-border fund settlement
continued

• If a regional CLS bank should be created,
• Who should own / govern / participate in it
  (I.e., governance)? Should it be another
  subsidiary of CLSS?
• Global custodians serving for the settlement in
  regional currencies.
• Local custodians
• Central banks as well?
• Where should it be located?
• Center of the regional time zone?
• In a jurisdiction whose legal framework is
  acceptable to as many participants as possible?

20
Multi-currency cross-border fund settlement
continued

• Cross-border DVP is essential.
• National CSD must achieve DVP domestically in
  order to enable cross-border DVP.
• DVP settlement instructions must be sent to the
  central bank and custodian.
• How to ensure settlement finality?
• Between banks and between a bank and CLS bank at
  each central bank.
• Under what law? Law of the jurisdiction in which
  CLS bank is located? Laws of the jurisdictions
  of the central banks involved in each
  transaction? English law?

21
Legal Regulatory Infrastructure

• Conflict of laws and legal risks in defining
  settlement finality and managing cross border
  collateral.
• Most government securities in the region are
  demeterialized and recorded as electronic
  entries.
• They are registered in domestic registries
  (central bank registries or CSDs), but where are
  they kept in custody?
• How can legal finality of settlement of offshore
  transactions in such securities be achieved and
  ensured?
• Need a common legal basis to which all the
  jurisdictions and market participants in them can
  agree?
• Article 9(2) of EU Settlement Finality Directive

22
Legal Regulatory Infrastructure

• Definition of settlement intermediaries.
• Legal status of securities differ across
  countries.
• Finality of delivery and payment
• Concept of bankruptcy

23
Capital account liberalization

• Exchange control on capital account transactions
• Role of central banks in implementing the
  exchange control and its implications for cross
  border securities trading.
• Taxation of cross-border securities transactions.

24
Business, political and technical feasibility

• Is it politically feasible to establish a new
  regional ICSD? Does it make business sense?
• Information technology (IT) enables
• Building bilateral links among national CSDs
• Building bilateral links between ICSDs and
  national CSDs.
• Global custodians to provide sophisticated
  customer services for institutional investors,
  alluring them to outsource not only backoffice
  but also some midoffice functions while global
  custodians themselves outsource core clearing
  to ICSDs.
• Substantial IT investment is required, and
  communication protocol and messaging standards
  need to be established.
• Requires training of staff.

25
Strategy

• Leave various channels of competition open.
• Policy support to encourage competition among
  national CSDs to become a regional ICSD.
• Ensure system connectivity.
• A CSD owned by and serving for a national stock
  exchange will be out and can only be a spoke.
• A CSD serving for government securities has
  competitive advantage from economies of scale and
  scope.
• A CSD operated by the central bank cannot compete
  for profit and beyond the national border.
• A CSD supported strong clearing banks and the
  Central Bank.
• A CSD open to direct participation by global
  custodians will have a greater chance to become a
  regional ICSD.
• A country that is to house a regional ICSD must
  have a liberal capital account.

26
Links with other initiatives

• Mutual acceptance of regional sovereign bonds as
  collateral for central bank intra-day credit.
• Use of ABC bonds as such collateral?

27
Thank you