[Federal Register Volume 67, Number 92 (Monday, May 13, 2002)]
[Pages 32043-32049]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 02-11785]




[Docket No. R-1122; Release No. 34-45879; File No. S7-15-02]
RIN 3235-AI48

Interagency White Paper on Structural Change in the Settlement of 
Government Securities: Issues and Options

AGENCIES: Board of Governors of the Federal Reserve System and 
Securities and Exchange Commission.

ACTION: Concept release; request for comment.


SUMMARY: The Board of Governors of the Federal Reserve System 
(``Board'') and the Securities and Exchange Commission (``Commission'') 
(collectively, the ``agencies'') are publishing for comment an 
interagency White Paper titled: Structural Change in the Settlement of 
Government Securities: Issues and Options (``White Paper''). The White 
Paper is designed to facilitate the discussion of possible structural 
changes in the settlement of government securities transactions. The 
White Paper is not intended to suggest that any of the approaches 
represent an improvement over current arrangements or that structural 
change is necessary. The goal of the White Paper is to provide a 
framework for discussion by identifying issues and questions that need 
to be further explored.

DATES: Comments should be received on or before August 12, 2002.

ADDRESSES: Comments should be sent to both agencies at the addresses 
listed below.
    Board: Comments should refer to Docket No. R-1122 and should be 
submitted in triplicate to Ms. Jennifer J. Johnson, Secretary, Board of 
Governors of the Federal Reserve System, 20th Street and Constitution 
Avenue, NW., Washington, DC 20551, or mailed electronically to 
regs.comments@federalreserve.gov. Comments addressed to Ms. Johnson may 
also be delivered to the Board's mail facility in the West Courtyard 
between 8:45 a.m. and 5:15 p.m., located on 21st Street between 
Constitution Avenue and C Street, NW. Members of the public may inspect 
comments in Room MP-500 of the Martin Building between 9:00 a.m. and 
5:00 p.m. on weekdays pursuant to Sec. 261.12, except as provided in 
Sec. 261.14, of the Board's Rules Regarding Availability of 
Information, 12 CFR 261.12 and 261.14.
    SEC: All comments concerning the White Paper should be submitted in 
triplicate to Jonathan G. Katz, Secretary, Securities and Exchange 
Commission, 450 5th Street, NW., Washington, DC 20549-0609. Comments 
can be submitted electronically at the following e-mail address: rule-comments@sec.gov. All comment letters should refer to File No. S7-15-
02; this file number should be included on the subject line if e-mail 
is used. All comments received will be available for public inspection 
and copying in the Commission's Public Reference Room, 450 5th Street, 
NW., Washington, DC 20549. Electronically submitted comment letters 
will be posted on the Commission's Internet Web site (http://www.sec.gov).\1\

    \1\ The Commission does not edit personal, identifying 
information, such as names or electronic mail addresses, from 
electronic submissions. Submit only information you wish to make 
publicly available.

FOR FURTHER INFORMATION CONTACT: Board: Patrick Parkinson, Associate 
Director, (202) 452-3526, and Patricia White, Assistant Director, (202) 
452-3620, Division of Research and Statistics; and Jeff Stehm, 
Assistant Director, (202) 452-2217, Division of Reserve Bank Operations 
and Payment Systems, Board of Governors of the Federal Reserve System, 
20th and C Streets, NW., Washington, DC 20551. Telecommunications 
Device for the Deaf (TDD) users may contact (202) 263-4869.
    SEC: Robert L.D. Colby, Deputy Director, at (202) 942-0094; Larry 
Bergmann, Senior Associate Director, at (202) 942-0770; Jerry 
Carpenter, Assistant Director, at (202) 942-4187; Jeffrey Mooney, 
Senior Special Counsel, at (202) 942-4174, and Jennifer Lucier, 
Attorney, at (202) 942-0173, Division of Market Regulation, Securities 
and Exchange Commission, 450 Fifth Street, NW., Washington, DC 20549-

    After the September 11, 2001, terrorist attacks, discussions were 
held with market participants to learn their perspectives on 
vulnerabilities in settlements of government securities. Three options 
for addressing vulnerabilities were explored: (1) The clearing banks 
and key market participants implementing more robust contingency 
arrangements; (2) each

[[Page 32044]]

primary dealer establishing a backup clearing arrangement at a bank 
other than its existing clearing bank; and (3) implementing structural 
change such as by establishing a utility to conduct settlement. The 
discussions revealed consensus on two points--contingency planning 
should be enhanced but market participants felt that a backup clearing 
account would be of little value. Market participants were interested 
in exploring structural changes in the provision of settlement services 
for government securities, including the concept of creating a utility, 
but the discussion was unfocused because of the absence of specific 
    The purpose of this White Paper is to facilitate discussion of 
issues relating to the settlement of government securities transactions 
by describing more concretely ways in which a utility might be 
organized. The staffs of the agencies believe that further discussion 
of a utility is warranted because enhanced contingency planning alone 
does not eliminate the vulnerabilities that have been identified in the 
settlement process for the government securities market. The White 
Paper identifies possible structural approaches for a utility and 
possible evaluation criteria for assessing the approaches. The White 
Paper also offers a preliminary assessment of the various approaches. 
The agencies request the views of market participants on the analysis 
in the White Paper and the next steps to be taken in evaluating 
structural change further.
    The White Paper in its entirety is set forth below.

Interagency White Paper \2\ Structural Change in the Settlement of 
Government Securities: Issues and Options

    \2\ Prepared by staff of the Federal Reserve Board, the Federal 
Reserve Bank of New York, and the Securities and Exchange 
Commission. Staff at the U.S. Treasury Department were consulted and 
provided comments.
    The agencies whose staff contributed to the drafting of this 
paper have not concluded that structural change is necessary.


    Payment and securities settlement systems have been marked by 
increasing consolidation of the institutions providing those services. 
During the 1980s and early 1990s, for example, mergers and exits 
reduced the number of banks providing settlement services for 
government securities trades, and today only two banks--JPMorgan Chase 
(Chase) and The Bank of New York (BONY)--provide the full range of 
services required by major market participants. Though these changes in 
the settlement of government securities are only one aspect of broader 
developments in financial markets, they are of particular interest to 
makers of public policy because of the key role that government 
securities play in the monetary policy process and as collateral in a 
wide array of financial market transactions.\3\

    \3\ The Group of Ten, ``Report on Consolidation in the Financial 
Sector,'' January 2001, provides a detailed look at the ongoing 
consolidation of financial institutions and the potential effects on 
the contours of the financial system.

    The business of settling trades in government securities involves 
the provision of a range of services: the transfer of government 
securities against funds (settlement), the provision of intraday credit 
to facilitate these settlements, position management services for 
primary dealers (including the matching of settlement instructions with 
incoming securities, automated options for handling mismatches, and the 
real-time reporting of transactions), and overnight and term financing 
through triparty repurchase agreements (repos). Settling trades, 
providing intraday credit, and providing tools (software) for position 
management are sometimes referred to as ``core clearing.'' \4\ The 
financing provided through triparty repos also is critical to the 
functioning of the government securities market. Triparty services for 
government securities currently are provided by the same banks that 
provide core clearing, but different entities may be able to offer the 
two types of services, as is the case for other types of securities.\5\

    \4\ Some view the use of the term ``clearing'' for these 
activities as a misnomer. The Government Securities Clearing 
Corporation (GSCC) serves as the clearing utility and central 
counterparty for trade comparison and netting in the U.S. government 
securities market. Trade comparison and trade netting services are 
also traditionally referred to as ``clearing.'' Because of the 
prevalence of the use of the term ``clearing'' for settling trades 
in government securities, however, its use is continued here.
    \5\ Triparty programs cover various securities for which core 
clearing is provided by the Depository Trust Company (DTC) or 
Euroclear Bank.

    All the primary dealers depend critically on either Chase or BONY 
for core clearing services and triparty repo arrangements, which are 
integral to the dealers' financing, and institutional investors rely on 
these clearing banks to place large volumes of funds in the highly 
secure and liquid triparty repos. The Federal Reserve also is dependent 
upon the clearing banks' records for open market transactions conducted 
through triparty arrangements, and the U.S. Treasury relies on the 
clearing banks for the settlement of a major portion of its securities 
at issuance.
    This concentration in the provision of clearing services gives rise 
to operational, financial, and structural vulnerabilities.
    (1) Operational problems at either of the two clearing banks can 
significantly impede the settlement of dealers' trades and the 
reconciliation of their positions. Market participants settling through 
a clearing bank with operational problems could not easily move to 
another service provider because of differences in the technology used 
by the clearing banks. Even if a switch in banks were technologically 
feasible, firms would be hampered because they would not know their 
securities and funds positions or have access to them at the clearing 
bank with operational problems.
    (2) Financial vulnerabilities arise from the potential for a 
clearing bank's financial condition to become impaired, perhaps because 
of losses from activities unrelated to the clearing business. 
Involuntary exit because of financial problems could force regulators 
to transfer the clearing operations to a bridge bank.\6\ Moreover, 
market participants might be uncomfortable with the uncertainty 
associated with a bridge bank, particularly because the ability to 
fashion a permanent solution (through, for example, the sale of the 
business) may be limited.

    \6\ The use of a bridge bank might be consistent with a least-
cost resolution; but if it were not, authorities would need to 
consider a systemic-risk exception to least-cost resolution, with 
the attendant increased costs in terms of moral hazard and 
diminished market discipline on large complex banks. The Federal 
Deposit Insurance Corporation generally must resolve failed 
institutions using the least costly method that meets its 
obligations to insured depositors. It can employ a method that is 
not least cost only if the Federal Reserve Board and the board of 
the FDIC recommend that step and if the Secretary of the Treasury 
(in consultation with the President) makes an explicit determination 
that a least-cost resolution would have adverse effects on economic 
conditions or financial stability and that the more costly method 
for resolution would avoid or mitigate those adverse effects. This 
is known as the ``systemic-risk exception.''

    (3) The current concentration in clearing has resulted in part from 
voluntary decisions by banks to exit the clearing business. A business 
decision to exit by either clearing bank would concentrate risk and 
market power in a single, full-purpose, commercial bank. This 
concentration of risk would likely be unacceptable to market 
participants and public policy makers and might be unacceptable to the 
remaining clearing bank.
    As part of the stocktaking after September 11, staff from the 
Federal Reserve, the Commission, and the Treasury held discussions with 
market participants to learn their perspective on the vulnerabilities 
of the government

[[Page 32045]]

securities market. The discussants explored three options for 
addressing vulnerabilities: improving the operational resiliency of 
clearing banks, establishing backup securities accounts with the second 
clearing bank, and instituting structural change, for example, by 
creating an industry utility to conduct settlement.
    These discussions indicated consensus on two points. First, 
clearing banks as well as other market participants needed to improve 
their contingency backup arrangements.\7\ Second, backup securities 
accounts would be difficult to arrange and likely would be of little 
value. The technology used by the two clearing banks is sufficiently 
different to make it difficult and costly to establish and maintain 
such accounts. More important, quickly moving activity to another 
account would likely be difficult because of the need to determine 
positions at the bank with problems, transfer these positions to the 
backup account, and alter standing settlement instructions with 
counterparties to direct new transactions to the backup account.

    \7\ A complementary interagency group is working with private-
sector firms and utilities to improve the resiliency of financial 
market participants' backup arrangements. Goals of the group include 
developing guidance on business continuity issues, organizing 
industry testing, and addressing telecommunications issues, 
particularly switching and routing diversity.

    Market participants were interested in exploring structural change, 
including the concept of an industry utility. Discussions of such a 
utility were hampered, however, by different conceptions of how it 
might be organized and lack of systematic consideration of the concept 
on the part of most market participants.
    This paper facilitates exploration of structural change in 
settlements of government securities by describing more concretely some 
approaches to organizing an industry utility. The agencies believe 
further discussion of structural change is warranted because enhanced 
contingency backup arrangements alone do not eliminate the financial 
and structural vulnerabilities that the market faces. Indeed, the cost 
of improved contingency arrangements could exacerbate structural 
vulnerability by reducing the profitability of core clearing. This 
paper also identifies possible criteria for assessing the approaches 
and, to encourage further discussion, offers a preliminary evaluation 
of the various approaches using the assessment criteria.
    The agencies whose staff have contributed to the drafting of this 
paper have not concluded that any of the approaches described below 
represents an improvement over current arrangements. Nor is this paper 
intended to resolve that issue. Rather, the agencies believe that a 
broad industry discussion of these issues is timely and that such a 
discussion would benefit from a document that furnishes it with a 
framework. This paper, therefore, provides that framework and 
identifies issues and questions that need to be explored further.


    One of the difficulties in discussing the establishment of a 
utility is the wide variety of forms that such an entity could take. 
The structure of the utility determines how risks will be shared and 
costs will be borne. An important dimension along which utilities often 
differ is their ownership and governance. A utility can be organized as 
a private-sector entity, perhaps owned and governed by market 
participants but subject to oversight by a public-sector body. 
Alternatively, clearing and settlement functions might be performed by 
a governmental entity. Other important characteristics of a utility 
include how credit is supplied in the clearing process (by individual 
banks, by the utility itself, by the central bank) and how the 
operational infrastructure is supplied (by competing service providers, 
by a single private utility, by the central bank). To focus discussion 
on the specific characteristics that meet market needs and address 
market vulnerabilities, this analysis is limited to only three of the 
many ways in which a utility might be structured.

Old Euroclear Model

    A utility can be structured as an industry-owned depository and 
settlement entity that contracts with a commercial bank for the 
provision of most services.\8\ This model for a utility is similar to 
the original Euroclear model in which, until 2001, an industry-owned 
company contracted with Morgan Guaranty Trust Company for operational 
and credit services. Shareholders of a utility organized along these 
lines would largely be securities and banking industry participants. 
The governing body typically would be elected by shareholders, and it 
would establish membership criteria, prices, operating budgets, and 
investment priorities. The utility would contract with a bank (or banks 
or other service providers) for the operation of the settlement and 
depository services. Settlements would take place on the books of this 
bank, which would furnish securities and cash accounts to dealers. It 
would also furnish intraday financing, subject to risk controls it 
would establish. Overnight financing, including triparty repo services, 
would be provided either by the bank supplying the operational support 
or perhaps by other banks.

    \8\ Contracting with multiple banks for these services may be 

A Private Limited-Purpose Bank

    A private limited-purpose bank (like the Depository Trust Company 
[DTC] or the new Euroclear Bank) is an alternative type of industry-
owned depository and settlement mechanism.\9\ Key features 
distinguishing this model from the old Euroclear model are the means of 
providing depository and settlement services and the sources of 
liquidity support. Rather than contracting with a commercial bank, the 
utility itself would furnish the operational support. Settlements of 
government securities currently require aggregate extensions of 
hundreds of billions of dollars of intraday credit to dealers, and a 
private limited-purpose bank would need to arrange a backup liquidity 
facility to ensure final settlement in the event one of its 
participants failed to cover an overdraft. Based on the experience of 
other utilities in arranging facilities a fraction of that size, a 
private limited-purpose bank might find arranging sufficient backup 
liquidity support difficult, other than possibly from the Federal 
Reserve. Overnight funding, including triparty repo services, could be 
provided by the limited-purpose bank or perhaps by other commercial 

    \9\ This option could be implemented by expanding an existing 
depository such as DTC or by creating an organization de novo.

Enhancement of Federal Reserve Services

    A third alternative is a public utility in which the Federal 
Reserve provides depository and settlement services. The Federal 
Reserve and the Commission generally prefer private-sector solutions to 
policy problems unless a market failure suggests a clear need for 
government intervention. In evaluating potential structural changes, 
however, it is important to discuss the widest possible set of ways to 
address the vulnerabilities for the government securities market, which 
includes enhancing Federal Reserve services.
    In a simple version of this model, the Federal Reserve would need 
to provide nonbank securities dealers, as well as the GSCC (and 
possibly interdealer brokers), direct access to securities

[[Page 32046]]

accounts, funds accounts, and secured credit. As noted earlier, dealers 
routinely use substantial intraday credit, which would need to be 
supplied by the Federal Reserve. A dealer also might find itself unable 
to fund its holdings of government securities in a financial crisis, 
and in that event, the Federal Reserve would need to provide liquidity 
support in the form of overnight credit. For this model to be 
effective, the Federal Reserve would have to furnish operational 
support by developing products that replicate at least some of the 
position management and information services currently provided to the 
dealers by the clearing banks. Dealers would continue to need the 
overnight funding supplied by triparty repo services. These services 
might be provided by commercial banks. Alternatively, the Federal 
Reserve could develop the product. In this case, the Federal Reserve 
would need to consider how triparty services might be offered without 
also extending accounts to nonbank institutional investors, perhaps by 
using these investors' accounts at their custodian banks.
    Variants on this simple model of enhanced Federal Reserve services 
also might be explored. For example, the Federal Reserve could provide 
direct operational interfaces with the dealers, but the dealers' 
transactions could settle through accounts held at depository 
institutions. In this way, depository institutions would intermediate 
the intraday credit used in the settlement process.

Evaluation Criteria

Operational, Financial, and Structural Vulnerabilities

    Any proposal to restructure government securities settlements must 
address the operational, financial, and structural vulnerabilities that 
are inherent in the current arrangements. Arguably, no utility could be 
designed to eliminate all these vulnerabilities. Rather, the relevant 
criteria for evaluating options are the extent to which the utility can 
reduce existing vulnerabilities. Proposals thus should be evaluated on 
their ability to improve the operational resiliency of government 
securities clearing, to better insulate the clearing process from the 
risks of financial problems at a key service provider, and to reduce 
the vulnerability of the clearing process to voluntary exit by firms 
that provide critical services. In the addressing of these 
vulnerabilities, however, it is equally important that new ones not be 
introduced, and evaluations of structural change should take this 
concern into account.

Efficiency and Innovation

    Other criteria that are critical for evaluating any restructuring 
proposal are the proposal's implications for the efficiency and 
innovativeness of the settlement process and related financing 
transactions. Existing arrangements in which the clearing banks compete 
in the provision of services to dealers create a mechanism both for 
holding down costs and for fostering innovation. The development of 
triparty repo services illustrates how clearing banks, in responding 
innovatively to market demands, have reduced dealers' financing costs 
and benefited investors. Proposals for structural change, therefore, 
should be evaluated on their ability to replicate the strengths of the 
existing system, encourage ongoing innovation, and deliver services in 
a cost-effective manner.
    Fully evaluating proposals may be difficult because the evaluations 
will depend on the governance structures adopted, which will determine 
pricing and investment decisions. In general, various proposals' 
governance structures (and the transparency of those structures) will 
have implications for a range of important issues, from the robustness 
of the risk-management system to the fairness (particularly with 
respect to access) of the system. Furthermore, assumptions about the 
initial investment required and the potential for savings on operating 
costs over time are necessary for making judgments about the efficiency 
of proposals.

Implications for Federal Reserve Policies

    Proposals to restructure the settlement process will have 
implications for Federal Reserve services and policies. The 
implementation of some proposals would require the Federal Reserve to 
broaden dramatically the scope of services that it provides market 
participants and, most important, change policies with respect to the 
types of firms that are granted access to accounts and to credit. 
Consideration of these proposals should entail an assessment of the 
Federal Reserve's legal and operational ability to deliver the required 
services. Proposals also should be evaluated to determine whether 
broader access and the provision of credit to nondepositories poses 
significant risk to the Federal Reserve or entails significant moral 
hazard. Other proposals raise the possibility that the Federal Reserve 
would greatly reduce its role in settling secondary market transactions 
for government securities, and the implications of that possible 
outcome also should be assessed.

Evaluations of Approaches

Old Euroclear Model

    This model's ability to address the vulnerabilities in the current 
system is mixed; though operational vulnerabilities could be addressed, 
financial vulnerabilities would not, and the effects on structural 
vulnerabilities would be unclear. Operationally, the utility would 
contract with one or more entities to provide support for depository 
and settlement activities, and its resiliency would depend upon the 
standards it set for firms providing the services. There is no reason 
to believe that the operational resiliency of this model would not be 
on par with that of the current system, and it might be possible to 
hold the banks providing services to higher standards because the costs 
would be more transparent and, therefore, dealers might be more willing 
to bear them.
    The ability of this model to address financial and structural 
vulnerabilities is much more limited, however. The utility would be 
exposed to the risk that a bank providing operational and credit 
services could involuntarily exit the business because of financial 
difficulties unrelated to clearing activities. This risk would be 
diversified if more than one firm provided these services. However, 
given the economies of scale and scope in clearing, the willingness of 
multiple banks to provide the critical services and, therefore, the 
potential for diversifying the risks may be limited. The extent to 
which the structural vulnerabilities are addressed depends on the 
ability of the utility to negotiate long-term contracts with suppliers 
of critical services at terms that the supplier will find sufficiently 
attractive to remain in the business.
    Because of the critical role of triparty activity in the financing 
of dealers, the market would be vulnerable to operational, financial, 
or structural problems if triparty services continued to be 
concentrated among only a few providers. Dealers might be able to 
manage these risks by requiring standardization of software that would 
enable them to move their accounts more easily in the event of 
operational problems or exit, but the challenges of reconciling 
positions in such events would remain.
    The ability of this model to deliver innovative services cost 
effectively will depend critically on the governance structure of the 
utility and the standards

[[Page 32047]]

it sets for banks supplying services. The utility may be able to foster 
competition similar to that of the current system by contracting with 
multiple banks for services. Product innovation would be dependent upon 
the utility's policies as well. The management and board of the utility 
clearly would have to be mindful of these issues if this model of 
utility were to retain these features of the current system.
    This model would not require any changes in Federal Reserve 
policies. The model continues to rely on private banks to provide 
operational and credit support for settlements; the utility itself 
would be a vehicle for administration and governance rather than a 
provider of services.

A Private Limited-Purpose Bank

    The creation of a limited-purpose bank to function as the utility 
would concentrate depository and settlement activity within one entity, 
thereby concentrating operational risk. The ability of this model of 
utility to improve the operational resiliency of government securities 
settlements thus will depend upon the resources it devotes to backup 
facilities. The current system requires each clearing bank to incur 
these costs; so conceivably, a limited-purpose bank could devote more 
resources to backup facilities than an individual clearing bank but 
would still offer a cost savings. A limited-purpose bank is, by 
construction, less exposed to financial problems from unrelated 
activities than a full-service bank because of limits on the scope of 
its activities. Similarly, it is unlikely to voluntarily exit the 
business of clearing, having been created solely for that purpose.
    The assessment of the ability of a limited-purpose bank to address 
financial and structural vulnerabilities in the government securities 
market is less sanguine if the utility does not provide triparty 
services and these services remain concentrated among a few banks. 
Triparty services are so integral to the financing of dealers in 
government securities markets that these markets will be operationally, 
financially, and structurally vulnerable to the banks that provide such 
services. These banks, which have broader business lines than a 
limited-purpose bank, will be vulnerable to losses in activities 
unrelated to clearing and triparty services. They are also free to make 
the business decision to voluntarily exit the triparty business. If the 
separation of core clearing from triparty services lowers the barriers 
to entry and attracts entrants to the triparty business, however, 
structural vulnerabilities would be ameliorated with a reduction in 
    The ability of this model to deliver services efficiently and 
innovatively will depend upon the governance structure of the limited-
purpose bank. Assuming that users own the bank and control the 
governance structure, these users will have incentives to monitor costs 
and to create mechanisms for developing new products. In recent years, 
triparty repo services have been the area of most innovation in the 
clearing of government securities. Undoubtedly, the competition between 
the clearing banks has spurred the innovation. Some of this pressure to 
innovate thus might be lost if triparty services were provided 
exclusively by the utility. Over time, a failure to innovate in the 
triparty area could have adverse implications for dealer financing.
    This model has several important implications for Federal Reserve 
policy. As was noted in the description of the limited-purpose bank, 
the Federal Reserve may be the only feasible entity to provide a backup 
liquidity facility of large enough size. Providing this facility to a 
limited-purpose bank would entail a change in policy with respect to 
discount window access for limited-purpose banks or trust companies. 
But it is not clear whether risk to the Federal Reserve or moral hazard 
would increase. With the current arrangements, the Federal Reserve 
effectively provides back-stop liquidity to the clearing banks. 
Providing the same liquidity to a utility might, in fact, entail less 
risk and moral hazard because of the restrictions on the utility's 
activities, more intense supervision of the utility, and greater 
transparency. The creation of this type of utility would also reduce 
(and might eliminate) the Federal Reserve's role in settling secondary 
market transactions for government securities. The vast majority of 
transactions would be settled on the books of the limited-purpose bank, 
particularly if it were providing triparty repo services as well as 
core clearing.

Enhancement of Federal Reserve Services

    If the Federal Reserve provides accounts, credit, and services 
directly to dealers, the existing vulnerabilities in the government 
securities market would be reduced. Under this model, the Federal 
Reserve would be providing the operational support for the settlement 
process, and these enhanced products would be integrated into the 
existing backup contingency arrangements for the Fedwire system. The 
Federal Reserve's arrangements have been more robust than those of 
private-sector firms and other market utilities, and the Federal 
Reserve has spent appropriate amounts to meet contingency requirements. 
Federal Reserve services are not vulnerable to disruption because of 
financial difficulties.
    To address vulnerabilities fully, the Federal Reserve may need to 
develop triparty repo as well as core clearing services. Alternatively, 
if the Federal Reserve limits its enhanced services to core clearing, 
there may be opportunities for a wider set of firms to offer triparty 
services, reducing structural vulnerability in the triparty market. A 
separation between core clearing and triparty repos, however, would 
require an additional transfer of securities from the dealer to the 
triparty provider, as in the current process with DTC-eligible 
securities used for triparty repos. The number of additional transfers 
could be reduced through the creation of a facility to transfer 
securities in blocks (bulk transfers) rather than security by security.
    It is not clear whether this model could deliver services as cost 
effectively as the current system or how product innovation would be 
affected. Although the Federal Reserve is required to price services to 
cover its costs over the long run, the benefits of competition would be 
lost. Perhaps more significant in the long run, innovation would no 
longer be spurred by competition. Because the Federal Reserve is not 
subject to the same profitability constraints that a private-sector 
business is, some industry participants may view its assumption of the 
role of service provider for settlement services negatively.
    Providing direct access to dealers would be a marked departure from 
existing Federal Reserve policy. The Federal Reserve would need to 
provide accounts and hundreds of billions of dollars of credit to 
nondepository institutions routinely during the day and, in a crisis, 
overnight. From a risk-management perspective, however, credit 
extensions presumably would be collateralized with highly liquid 
securities, and government securities brokers and dealers would be 
subject to federal regulation by the Commission or the Treasury.
    Direct access to dealers could be perceived as providing dealers 
with broad access to liquidity support from the Federal Reserve. Any 
adverse effects on market discipline would be mitigated by federal 
regulation of the dealers, collateralization of the credit extensions, 
fees for intraday and overnight credit, and the potential for the 
Federal Reserve to impose quantity constraints on the amount of 

[[Page 32048]]

credit extensions. Still, expansion of access could raise concerns 
about moral hazard. Perception of a safety net extension might be 
further attenuated through some variant of this model that leaves the 
dealers' accounts in a depository institution. In addition, if the 
Federal Reserve were to provide triparty repo services, the issue of 
accounts for a broad set of institutional investors might arise unless 
market practices changed.

Questions for Further Discussion

    1. Have the vulnerabilities in the government securities market 
been identified correctly? Are there other vulnerabilities that should 
be considered in evaluating the need for structural change?
    2. Are there other structural approaches to a utility that should 
be given serious consideration besides the three basic options 
described in this paper? If so, what are they?
    3. Are the evaluation criteria set out in this paper the relevant 
ones for assessing the merits of an industry utility? If not, what 
other criteria are relevant?
    4. Can concerns about efficiency, innovation, and competition be 
addressed through governance? If so, how?
    5. Is it feasible to separate the provision of core clearing from 
the provision of triparty repo services? Would the separation of core 
clearing from triparty repo enable other banks to compete more 
effectively in the provision of triparty services? Can triparty repo 
services be provided by a utility?
    6. How much intraday credit would a utility need to provide in the 
settlement of government securities trades? Would a utility likely be 
able to arrange backup liquidity through committed lines of credit at 
commercial banks of the magnitude necessary to ensure timely settlement 
in the event a participant failed to cover an intraday credit 
    7. What is the likely size of the initial investment to create an 
industry utility? What factors determine the effects of a utility on 
costs generally? On costs to dealers of core clearing services? On 
financing costs to dealers?
    8. Who should own a private utility? How should its board of 
directors be chosen? What legal form should it take (for example, a 
bank, registered clearing agency, an Edge Act corporation)?
    9. What should be the next steps in evaluating alternative 
structures? What type of decisionmaking framework should be created, 
and which groups should be represented in that process?

Appendix 1

Clearing and Settlement Arrangements for Government Securities

    1. Within the universe of about 1,700 dealers, the trading of U.S. 
government securities is concentrated largely among 22 primary dealers 
and a handful of interdealer brokers (IDBs).
     Interdealer brokers collect dealer quotes, post them to 
electronic screen services, and execute trades between dealers, thereby 
facilitating price discovery, liquid markets, and anonymity in the 
interdealer market; about one-third of dealer-to-dealer trades are 
executed through an IDB.
     Among the primary dealers, most trading activity is 
concentrated in five to ten dealers.
     Trading activity includes dealer financing (repo) 
transactions and outright purchases and sales on behalf of customers 
and for the dealer's own account.
    2. After a trade is executed, counterparties to the trade must 
compare trade details and determine settlement obligations (clearance).
     The Government Securities Clearing Corporation (GSCC) 
serves as the clearing utility and central counterparty for trade 
comparison and netting in the U.S. government securities market.
     GSCC is registered with and supervised by the Commission.
     Through trade comparison, netting, and central 
counterparty guarantees, GSCC decreases its participants' counterparty 
settlement risk and helps ensure orderly settlement in the marketplace.
     Each day, GSCC compares trades valued at more than $1.3 
trillion. About one-third of these trades are for outright purchases 
and sales, and the remaining two-thirds are repo transactions.
     GSCC has 122 direct participants--consisting of dealers, 
interdealer brokers, investment managers, and banks--one-quarter of 
which use trade comparison services only.
     GSCC participants also clear trades for another 468 
dealers, banks, and investment managers, through correspondent 
relationships. Generally, these correspondent relationships are for 
trade comparison services only.
    3. Following the clearance process, securities must be exchanged 
for funds (settlement) on either a gross or a net basis.
     Government securities are transferred against funds 
(settled) through depository institutions acting as agents for nonbank 
dealers. Interbank settlement occurs through the Fedwire securities 
transfer system.
     Settlement typically occurs one business day after the 
trade (T+1), either through transfers on the books of a depository 
institution or, if settlement must occur between two depository 
institutions, on the books of the Federal Reserve through the Fedwire 
securities transfer system. Repo transactions generally settle on a 
same-day (T) basis.
     More than $800 billion in securities is transferred 
through the Fedwire securities transfer system each day.
     The two banks, Chase and The Bank of New York (BONY), that 
provide settlement services to primary dealers account for more than 
three-quarters of the value of Fedwire settlement activity. On a 
typical day, these two banks settle more than $600 billion in 
government securities transactions through Fedwire. The clearing banks 
apparently settle another $200 billion to $300 billion per day 
internally, excluding triparty repo transactions.
     GSCC settles net obligations valued at about $415 billion 
per day through its accounts at the two clearing banks.
     Chase's and BONY's client bases consist of the primary 
dealers, other dealers and banks, and GSCC.
    4. The settlement of financing (repo) transactions occurs either 
through bilateral exchanges (delivery-versus-payment or DVP repos) of 
securities and funds between a dealer (borrower) and an investor 
(lender) or through the use of triparty repos on the books of the 
clearing banks.
     DVP repos are generally settled over Fedwire between 8:30 
a.m. and 12:00 p.m., Eastern time.
     Triparty repos are settled after the close of the Fedwire 
securities transfer system, generally between 5:00 p.m. and 7:00 p.m., 
Eastern time. The two clearing banks estimate that together they settle 
on their books between $600 billion and $1 trillion in triparty repos 
each day.

Appendix 2

Triparty Repo

The Market

    Understanding the role of the clearing banks in the clearance of 
U.S. government securities requires an appreciation of the triparty 
repo market and the critical role that such banks play in facilitating 
triparty repo transactions. Essentially, these transactions involve the 
secured financing of broker-dealer securities inventories by a large 
number of cash investors, with settlement occurring on the books of the 
clearing banks. Over the last decade, the importance of the triparty 
repo market grew significantly, so that now it is integral to the 
financing methods of all major broker-dealers and

[[Page 32049]]

involves nearly $1 trillion per day in transactions.
    The success of the triparty repo market is due to its ability to 
meet the needs of both the broker-dealers who need secured financing 
and the cash investor community, who desire highly secure and liquid 
outlets for the investment of cash on a short-term basis. The cash 
investors in triparty repo consist of money market mutual funds and 
other institutional money managers such as pension funds. Both the pool 
of funds that such institutional investors need to invest and the size 
of the broker-dealer securities inventories have grown significantly in 
recent years, with no signs of a slowdown yet apparent. The clearing 
banks also benefit from providing triparty repo services as a 
profitable line of business and as an opportunity to cross-sell other 
custody and banking services to cash investors.

Settlement: The Critical Role of the Clearing Banks

    In a typical triparty repo transaction, a broker-dealer contracts 
with a cash investor to provide a certain amount of securities in 
exchange for cash at the outset of the transaction, with the 
transaction to be unwound at the end of its term. All movements of cash 
and securities are to take place on the books of the broker-dealer's 
clearing bank. That is, both the broker-dealer and the cash investor 
will use cash and securities accounts at the clearing bank, and the 
clearing bank will play a critical role in settling the transaction. It 
is typical for the broker-dealer to pay for the setting up of accounts 
at its clearing bank on behalf of all its cash investors.
    Triparty transactions are typically arranged early in the morning 
so that dealers can be assured of meeting their financing requirements. 
Importantly, however, these transactions typically do not specify the 
individual securities that the broker-dealer will provide as 
collateral. Rather the transactions are based on broad categories of 
collateral, such as U.S. government or agency securities. Different 
qualities of collateral engender different financing rates, and the 
triparty market has been steadily expanding beyond U.S. government 
securities to encompass a wide range of mortgage-backed securities, 
corporate bonds, and non-U.S. securities. However, U.S. government and 
agency securities remain the dominant form of triparty collateral, 
accounting for more than two-thirds of the total market.
    The fact that triparty transactions do not uniquely specify 
individual securities is central to their appeal for the broker-dealer 
community. This flexibility allows the broker-dealers to trade their 
securities inventory during the normal business day, settling whatever 
transactions come due, without significant concern regarding their 
financing arrangements. For example, settlement of cash-market U.S. 
government and agency securities continues until 3:30 p.m. on a normal 
day, the time when the Fedwire book-entry transfer system closes. Soon 
after this point, the clearing banks begin to process the broker-
dealer's triparty repo transactions. This processing involves comparing 
the generic triparty transactions that the broker-dealers have 
submitted with the specific securities that now reside in their 
accounts at the clearing bank. The clearing banks have developed 
routines for optimizing the allocation of specific collateral to 
individual triparty transactions to minimize the financing costs for 
the broker-dealers.
    The collateral optimization and allocation routines run in the late 
afternoon, with settlement of the triparty transactions on the books of 
the clearing bank typically occurring in the early evening. The 
efficiency of these procedures, together with the familiarity of the 
broker-dealers with them, means that the need for residual financing 
(that is, securities to finance that cannot be financed through 
triparty repos) is generally only very small, on the order of 1 percent 
or less of their total eligible inventory.

Benefits to Investors and Dealers

    Triparty arrangements between a broker-dealer and a cash investor 
may be either on an overnight or on a term basis. Importantly, however, 
even if the transactions are done on a term basis, all collateral is 
typically unwound on a daily basis (early in the morning). This daily 
unwinding has two implications. First, the cash investors get access to 
their funds on the books of the clearing bank on an intraday basis. 
Second, the broker-dealers get access to their securities inventory and 
thus can effectively ``substitute'' other collateral into the 
agreements as their inventory shifts over the term of the agreement.
    From the cash investors' perspective, the triparty repo market 
provides a great deal of liquidity and safety for their cash holdings. 
During the day, the cash resides in deposit accounts at their clearing 
bank (or elsewhere if they choose to wire it back and forth, although 
most do not). Overnight, they are exposed to the credit risk of their 
broker-dealer counterparties but are protected by the presence of 
collateral held in their accounts at the relevant clearing bank. 
Moreover, the flexibility of the triparty arrangement allows them to 
frequently adjust the size of their cash investments as their pool of 
available funds fluctuates. For the broker-dealer, the triparty repo 
market obviously provides a highly flexible mechanism to minimize the 
costs of financing.

Triparty Repo an Important Source of Intraday Overdrafts

    For the clearing banks, the triparty repo mechanism is an important 
complementary service to their core clearance activities in the 
underlying securities. However, a major implication of the triparty 
mechanism as currently designed is the presence of extremely large 
intraday overdrafts in the deposit accounts of the broker-dealers at 
the clearing banks. That is, because all the cash is returned to the 
cash investors daily, the entirety of a dealer's inventory is 
effectively financed by the clearing bank on an intraday basis. Still, 
the clearing bank is secured to the extent that the broker-dealer's 
securities remain at the bank. These figures can approach $100 billion 
for the largest individual dealers on peak days.

    By order of the Board of Governors of the Federal Reserve 
System, May 7, 2002.
Jennifer J. Johnson,
Secretary of the Board.
    By the Securities and Exchange Commission.

    Dated: May 6, 2002.
Margaret H. McFarland,
Deputy Secretary.
[FR Doc. 02-11785 Filed 5-10-02; 8:45 am]
BILLING CODE 6210-01-P, 8010-01-P