Description
Systemic risk propagated through over-the-counter (OTC) derivatives can best be
managed by a public-private central counterparty clearing house. The purpose of this paper is to
outline the market microstructure necessary for such a clearing house.
Journal of Financial Economic Policy
Centralized clearing for over-the-counter derivatives
Gordon Rausser William Balson Reid Stevens
Article information:
To cite this document:
Gordon Rausser William Balson Reid Stevens, (2010),"Centralized clearing for over-the-counter
derivatives", J ournal of Financial Economic Policy, Vol. 2 Iss 4 pp. 346 - 359
Permanent link to this document:
http://dx.doi.org/10.1108/17576381011100865
Downloaded on: 24 January 2016, At: 21:39 (PT)
References: this document contains references to 11 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 430 times since 2010*
Users who downloaded this article also downloaded:
Donald J . Mullineaux, Mark K. Pyles, (2010),"Bank marketing investments and bank performance", J ournal
of Financial Economic Policy, Vol. 2 Iss 4 pp. 326-345 http://dx.doi.org/10.1108/17576381011100856
David G. Tarr, (2010),"The political, regulatory, and market failures that caused the US financial
crisis: What are the lessons?", J ournal of Financial Economic Policy, Vol. 2 Iss 2 pp. 163-186 http://
dx.doi.org/10.1108/17576381011070210
Sven Blank, J onas Dovern, (2010),"What macroeconomic shocks affect the German banking system?:
Analysis in an integrated micro-macro model", J ournal of Financial Economic Policy, Vol. 2 Iss 2 pp.
126-148 http://dx.doi.org/10.1108/17576381011070193
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Centralized clearing for
over-the-counter derivatives
Gordon Rausser
Department of Agricultural and Resource Economics, University of California,
Berkeley, Berkeley, California, USA
William Balson
RBS Acquisition Company, Palo Alto, California, USA, and
Reid Stevens
Department of Agricultural and Resource Economics, University of California,
Berkeley, Berkeley, California, USA
Abstract
Purpose – Systemic risk propagated through over-the-counter (OTC) derivatives can best be
managed by a public-private central counterparty clearing house. The purpose of this paper is to
outline the market microstructure necessary for such a clearing house.
Design/methodology/approach – The paper proposes using an request for quote platform with an
active permissioning system that uses analytic approximations based on Monte Carlo simulation to
estimate default risk and a two-part pricing scheme to ef?ciently price that risk.
Findings – It is found that comprehensive clearing for complex and standardized derivatives is
feasible using the clearing framework.
Research limitations/implications – This research is limited by the authors’ ability to give
empirical examples. The paper gives a short example with data, but given the constraints on length,
cannot go into more detail.
Practical implications – This comprehensive clearing structure, in contrast to current proposed
government regulations, will not drive out the “good” with the “bad” OTC derivative instruments.
Originality/value – This is the only paper the authors are aware of that outlines a detailed
framework for clearing all OTC derivatives.
Keywords United States of America, Legislation, Financial risk, Derivative markets, Risk management
Paper type Research paper
1. Introduction
Over the course of the most recent ?nancial crisis, the government lacked regulatory
mechanisms to deal with ?rms whose failure could trigger the failure of other ?rms
through the over-the-counter (OTC) derivatives market and was forced to manage
the systemic risk posed by large ?nancial institutions on an ad hoc and ex post
basis. To prevent cascading defaults, the government facilitated the sale of some
large ?nancial institutions (e.g. Bear Stearns), allowed others to declare bankruptcy
(e.g. Lehman Brothers), and injected capital into many through the Troubled Asset
Relief Program (TARP) (e.g. Bank of America). Even the treatment of ?rms that
struggled after receiving payments through TARP was unpredictable; Citigroup and
AIG received additional support through direct public investment while commercial
investment trust received no further assistance beyond the initial support of $2.33
billion. The ad hoc nature of this process is unlikely to have been the most economically
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JFEP
2,4
346
Journal of Financial Economic Policy
Vol. 2 No. 4, 2010
pp. 346-359
qEmerald Group Publishing Limited
1757-6385
DOI 10.1108/17576381011100865
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
ef?cient choice except in the sense that urgency required action. These events have
demonstrated the need to develop a clear regulatory framework to ef?ciently manage
systemic risk whether posed by the OTC market or otherwise.
In this paper, we propose that systemic risk be managed directly through the
creation of a central counterparty clearing house (CCP). A CCP for OTC derivatives can
be designed to decrease systemic risk by eliminating default risk between the
counterparties to an OTC contract and by moderating the ?nancial incentives to
accumulate “excessive risk” in OTC markets[1]. The lack of such a CCP for OTC
derivatives, particularly credit default swap (CDS)’s, has been labeled as a signi?cant
factor in the current ?nancial crisis (Acharya et al., 2009) and the OTC derivatives
market, given its growth in recent years (Figure 1), will likely pose a larger threat to
future systemic stability without a CCP.
1.1 Summary of recent legislative history
Both the Administration and the Congress have proposed legislation to reform the
OTC derivatives market. These proposals resulted in the Dodd-Frank Act of 2010. The
Act encourages migration of OTC contracts to be cleared onto listed exchanges.
Unfortunately, the Act leaves to each derivatives clearing organization (DCO) the
decision of which contracts it will clear. As a result, the vast bulk of OTC derivatives
are unlikely to be cleared by DCOs in the near future, making it likely that systemic
risk will accumulate due to the bilateral credits risks of uncleared derivatives.
Several alternatives have been proposed prior to the Dodd-Frank Act. The
Administration’s proposal (Department of the Treasury, 2009) for ?nancial regulatory
reformexpanded on commitments by the previous administration to establish a clearing
house for OTC derivatives, focusing on promoting the public good by managing the
systemic risk posed by OTC derivatives and promoting transparency in the OTC
market. This proposal called for clearing “standardized OTC derivative transactions”
and increasing “regulatory capital requirements” on non-standardized derivatives[2],
though it contained no detailed plans for accomplishing these objectives.
Figure 1.
Ten-year growth of the
global OTC market and
exchange-traded
derivatives market –
100,000
200,000
300,000
400,000
500,000
600,000
700,000
Dec-99 Dec-01 Dec-03 Dec-05 Dec-07 Dec-09
N
o
t
i
o
n
a
l
v
a
l
u
e
(
b
i
l
l
i
o
n
s
o
f
U
S
d
o
l
l
a
r
s
)
OTC derivatives
Global OTC and exchange traded derivatives market
Exchange traded derivatives
Centralized
clearing for
OTC derivatives
347
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The House of Representatives had passed legislation that consolidated oversight[3],
required stronger capital cushions, regulated swaps, encouraged the use of a CCP, and
increased the cost of capital on all counterparties. The Senate’s Restoring American
Financial Stability Act, which was proposed at the committee level, focused on
increasing oversight of derivatives traders, not comprehensive clearing. As with the
Administration’s proposal, neither the House nor the Senate provided detailed plans to
manage the systemic risk posed by the OTC market.
Avery recent proposal (Volcker Rule) by the Administration aimed to prevent banks
from using their deposits as collateral for their OTC trading by implementing the spirit
of the repealed Glass-Steagall Act, which prohibited depository institutions from
operating in securities markets. The Dodd Frank Act contains a similar limitation.
In summary, a wide variety of legislative proposals have been advanced related to
OTC derivatives clearing, but the ?nal legislation only requires rule making by
commodity futures trading commission (CFTC) and securities and exchange commission
to accomplish that objective. We believe one explanation for going slow is the lack of a
detailed design for howa DCOcould reliablyclear the diversityof OTCderivatives. Inthis
paper, we present such a design.
1.2 The clearing dilemma
The heart of the dilemma faced by policy makers in reforming the derivatives market
is its size and complexity. As of June 2008, the notional value of all outstanding
OTC ?nancial contracts was in excess of $680 trillion, according to the Bank for
International Settlements (2010) (Figure 1). In contrast, the value of all cleared
derivatives traded by private regulated exchanges was below $20 trillion in notional
value. Given the limitless variation among derivatives (i.e. underlying assets, terms,
conditions, etc.), establishing capital requirements and clearing for this market have
long been technologically infeasible. Only recently have technology and ?nancial
theory reached the point that centralized clearing for both vanilla and complex
derivatives is possible.
We advance a proposal to:
.
centrally clear not only standardized but also complex derivatives with real-time
permissioning;
.
explicitly recognize the public-private partnership (PPP) structure needed to
effectively manage systemic risk;
.
implement ef?cient pricing of default risk and thus clearing;
.
allow structured ?nance to continue its pursuit of a world of more complete
contracts (e.g. bonds and CDS’s); and
.
incorporate government policy instruments to allow the public sector to be
compensated for controlling systemic risk over the business cycle.
Instead of imposing excessive capital requirements suggested by some proposals,
which would increase the cost of capital of trading such instruments, driving out the
“good” with the “bad,” our proposed framework will design a clearing solution that
can accommodate all derivatives, but price the clearing according to their relative risk.
This would allow traders to bene?t from the “good” complex contracts, and, with
centralized clearing, limit the systemic risk from the “bad” complex contracts.
JFEP
2,4
348
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2. Private clearing
Private DCOs have recently begun clearing a subset of OTC derivatives that will expand
as the CCPs develop methods to standardize some contracts and establish ef?cient
margin requirements. The largest private initiatives are the clearing operations associated
with IntercontinentalExchange, Chicago Mercantile Exchange, Eurex, and Euronext.
These ?nancial institutions began clearing OTC derivatives in response to pressure
from both the public and the private sectors following the most recent ?nancial crisis.
The clearing microstructure at these regulated exchanges is based on novation in
which the DCO becomes the counterparty (Appendix 1). Novation eliminates bilateral
counterparty risks but can cause the accumulation of ?nancial risk in the DCO.
Contracts are novated using a single fee for clearing along with initial and variation
margin requirements. The relationship between margin determination and the volatility
of underlyings is typically set by each exchange’s committee of experts. Contracts are
traded on a double-sided auction (i.e. a separate order queue for bids and offers) in which
the exchange determines set of tradable contracts with regulatory approval (e.g. CFTC).
Though each exchange has developed a platform that will ef?ciently
facilitate transactions for standardized OTC contracts with liquid markets, they are
not designed to provide complete public transparency or systemic risk management for
the broader OTC market. The exchanges lack suf?cient incentives to invest optimally
in transparency and systemic risk management because the bene?ts of managing
systemic risk generated cannot be appropriated by private CDOs. By only clearing
standardized, liquid contracts (i.e. CDS indices and interest rate swaps), the public will
only have price discovery on those segments of the OTC market, and the OTC market
for complex, custom derivatives will likely remain opaque. The private DCOs also lack
the capitalization necessary to remain solvent in the face of a ?nancial crisis with
cascading defaults in the OTC markets. Without adequate capitalization, the DCOs
cannot economically manage systemic risk and will be forced to rely on government
assistance in the event of a systemic crisis (despite legislative prohibitions on doing so).
As a result, DCOs have become another potential source of systemic risk, particularly
given the rapid consolidation that has taken place over the course of the last decade[4].
3. The need for a PPP
To correct the weaknesses of the private clearing initiatives, it is essential to
understand the nature of goods generated through centralized clearing: transparency,
systemic risk management, and the facilitation of trades. The pure facilitation of
liquid standardized trades is a private good, both rival and excludable. Transparency is
a public good, both non-rival and non-excludable, the bene?ts of which cannot be
entirely captured by a private DCO. The management of systemic risk is an impure
good[5]. Systemic risk management is non-excludable because the contracts of all
systemically important ?rms that default must be underwritten, either by the DCOor by
the government, but is rival because each defaulting ?rm requires a separate
incremental investment. Private DCOs lack suf?cient incentives to provide the socially
optimal level of clearing’s impure good.
3.1 Control rights and PPPs
A PPP can be formed to manage OTC derivatives’ CCP to ensure the optimal level of
the public, private, and impure goods are produced. A PPP could provide adequate
Centralized
clearing for
OTC derivatives
349
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
capitalization and ensure transparency while maintaining incentives to facilitate trades
through the assignment of control rights. The PPP literature has shown that ownership
should be determined by the type of good produced, and in the case of impure goods,
joint management by the public and private sectors can be optimal (Grossman and
Hart, 1986; Besley and Ghatak, 2001; Francesconi and Muthoo, 2006). Joint ownership,
which is determined by the contractual assignment of control rights, can provide both
sectors’ incentives to invest their resources and each suf?cient control to ensure
socially optimal levels of production (Rausser and Stevens, 2009).
A public-private OTC clearing partnership is inevitable, given the systemic
importance of OTC clearing and historic public sector support for failing ?nancial
institutions to prevent cascading defaults[6]. If, during a ?nancial crisis, a DCObecomes
insolvent, the government will be expected to act as lender of last resort. In the academic
literature, models of private OTC clearing implicitly assume that the government would
bailout a DCO to prevent widespread default, though there is no formal public
involvement or any compensation for the public sector to provide such services ( Jones
and Perignon, 2009).
If the partnership is explicitly recognized ex ante by forming a public-private CCP,
the government can manage system risk over the business cycle by choosing to be
compensated for the services it provides to stem systemic risk. If the partnership is
designed ex post, the implicit insurance provided by the public sector will only be
compensated on an ad hoc basis. But, the point of ex ante management is to moderate
the forces that lead to excess in advance rather than ex post. Beyond compensating
the government and strengthening their control over systemic risk, the creation of an
explicit partnership clari?es the “rules of the game” for derivatives markets and reduces
uncertainty over the government’s role during a ?nancial crisis.
4. Market microstructure
To improve the provision of public and impure goods, we propose the public-private
CCP use a request for quote (RFQ) platform along with the double-sided auction
platform currently used by private CCPs. Most OTC trading is currently conducted as
an informal RFQ auction in which telephone, fax, and electronic bulletin boards are
used to disseminate interest on either the buy or the sell side, and responses are made
that may or may not see competing bids or offers. Our proposal is to systematize this
informal process that already occurs.
4.1 RFQ platform
The RFQ execution platform we propose has four principle advantages:
(1) An RFQ platform allows traders to determine the set of complex contracts that
will be traded and cleared by combining standardized contract elements.
(2) An RFQ platform promotes transparency by publishing price feeds for all
traded complex contracts rather than just the standardized contracts.
(3) An RFQ platform can stem systemic risk by facilitating clearing for any
composite derivative, which would allow the government to require that
virtually all derivatives be cleared.
(4) An RFQ platform’s multi-party negotiations disseminate offer data to all
market participants, reducing the informational asymmetry in the OTC market.
JFEP
2,4
350
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
An RFQ market can coexist alongside existing double-sided auction markets and
indeed is the only auction mechanism capable of expanding clearing to almost all OTC
derivatives. In principle, double-sided auction markets are incapable of handling all
complex OTC derivatives due to the exponentially large problem of combinatorics.
For example, suppose an exchange lists and clears 10,000 names for standardized
option contacts with ten delivery dates and ten strike prices. Then there are at least a
trillion simple spread contracts with two legs. Clearly, a double-sided auction would be
unable to sustain such exponential growth. Fortunately, most transactions are focused
on a small fraction of such possibilities, but still there are too many. We believe that
any complex derivative composed of two or more standardized legs can reliably be
cleared using an RFQ transaction facility with the proposed pre-trade permissioning.
The transaction process begins on an RFQ trading platform when a subscriber
creates and posts the terms for a derivative contract. For example, if a subscriber
chooses to create a calendar spread on an RFQplatform, they determine the underlying,
strike price, class, quantity, and buy and sell expiration dates. Once posted, respondents
offer quotes and sizes for the contract, which are aggregated and disseminated to
all subscribers. Because the information processors for exchange-traded options
(i.e. Options Clearing Corporation and Options Price Reporting Authority) take complex
option structures as individual legs for clearing and reporting purposes, after a contract
is agreed on at its net debit/credit price, users can negotiate the actual leg prices prior to
?nal trade acceptance. At the end of a short RFQperiod, subscribers can “Post to Block”
for an RFQ they generated and move the request into the Block Trade facility to meet a
second, pre-de?ned party and affect a cross betweenthe best bid and offer prices[7] at the
end of the associated RFQ period. While the parties ?nalize the terms of a contract, the
CCP evaluates the contract’s risk (Sections 4.3 and 4.4) and sets the clearing fees
(Section 4.5).
4.1.1 Liquidity. To function, a double-sided auction requires market makers who
pro?t on the bid-ask spread, which is the spread between the price that clears buy orders
and the price that clears sell orders. The bid-ask spread compensates the market makers
for the costs of running its business as well as its retained risk (i.e. sometimes the ?owof
orders is imbalanced and the entity is expected to warehouse risk until the balance is
restored). It is not always true that the bid-ask spread compensates for retained risk,
especially in markets with few orders or in markets where the ?ow of orders is highly
imbalanced for a protracted period. In those cases, the observed market response is
typically that bid-ask spreads become very wide, liquidity dries up, and order ?ow
eventually ceases. In extreme cases, bids to buy can be entered as so-called “stub quotes”
and offers to sell can be entered at multiples of recent execution prices. The result is a
failed auction market in which trading is impeded by the auction design.
Since there is no bid-ask spread in an RFQ platform, it is not possible for the market
to collapse due to a lack of liquidity. In fact, the phone/fax methodology of the current
OTC market adjusts well to periods of varying liquidity. We expect that the order ?ow
will vary periodically and order executions would also vary. But market prices would
still be based on the most recent executions and quotes on the underlyings. Moreover,
restarting an RFQ market can begin as soon as order ?ow begins to become balanced,
unlike a double-sided auction in which the risk-reward relationship of a market maker
must also be re?ected in prices. Dysfunctional market prices, such as the ?ash crash,
are less likely in an RFQ market; since a buy and a sell order must be matched with no
Centralized
clearing for
OTC derivatives
351
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
intermediary, the concept of stub quotes does not occur. However, during disruptive
market periods, there can be a time variation of executed transactions.
Even in illiquid markets, an RFQ clearing platform can pro?tably clear contracts
given the typical size of such contracts and the variable risk-based pricing component.
Though we would not expect liquidity in all contracts, we anticipate that the
market would respond to improved transparency of prices, which cannot be achieved
on a double-sided auction platform burdened with inadequate liquidity. Moreover,
such transparency will decrease the informational advantage of ?nancial institutions
that have direct observation of order ?ows.
4.2 Clearing and active permissioning
An RFQ platform with clearing requires sub-second pre-trade permissioning to avoid
impeding the ?ow of executions. Since OTC products can be composed of multiple
underlyings, they can be decomposed into their elemental risks and the price feeds
from the standardized components can be employed to determine mark-to-market
value, even in illiquid markets. The decomposition procedure allows our approach to
utilize price feeds for the elements of an OTC contract and compute the risk properties
of an overall portfolio. OTC-customized instruments can and are priced on a real-time
basis, because multiple underlyings have mark-to-market forward price curves.
We propose combining an analytical value at risk (VaR) methodology with Monte
Carlo simulation to implement real-time OTC contract permissioning[8]. With this
approach, portfolios of hundreds thousands of non-linear contracts can be evaluated
in sub-second time intervals. Even aggressive trading behavior can, in principle, be
monitored in real time on a pre-execution basis without impeding the ?ow of
negotiations in the OTC market. The key design trade-off is the proper evaluation of
functionality designed for speed (simplicity) and accuracy (complexity).
Risk can be accurately assessed with relatively time-intensive Monte Carlo
methods. Monte Carlo methods are the only accurate method of assessing ?nancial
risk in markets characterized by irregular probability distributions, rapidly changing
volatility and correlation matrices, and highly non-linear payoff functions. Risk can
be quickly approximated using analytic VaR. Analytic value at-risk methods allow for
sub-second evaluation of risk for very large portfolios. Analytic methods are
amendable to very rapid calculations and analytic approximations are available for
almost all elemental derivatives.
Our essential insight is to use analytic methods as an extrapolation function that is
calibrated to the Monte Carlo simulations. It is possible to relax any restrictions on
mathematical form of the distribution of price changes by treating the analytic
expression as an extrapolation function with appropriate modi?cation enabling it to be
periodically calibrated to the results of Monte Carlo simulations (Figure 2).
4.3 Measuring VaR
VaR for a portfolio[9] is de?ned as a one-sided con?dence interval on portfolio losses:
Probability ðDV
A
0 ðT
Var
; DxÞ $ 2VaR
A
Þ ¼ 1 2a ð1Þ
where V
A
is the value of a portfolio of deal elements of interest, T
Var
is the risk horizon
of interest, x is a vector of random state variables, VaR
A
is the VaR of portfolio A, and
a is the level of con?dence. This formulation is applicable to any portfolio, any set of
JFEP
2,4
352
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
state variables, and any process governing the stochastic evolution of the value of the
portfolio. The portfolio may include deal elements that are securities, equities, bonds,
options, futures, derivatives, or other assets. The state variables may be prices on deal
elements, events that affect prices, external events, credit ratings, or other risk factors.
The price process may be a named stochastic process or may have jumps, reversion,
non-Markovian state evolution, stochastic volatility, discontinuities, or other features.
To use Monte Carlo simulation, VaR is rewritten as:
Z
VaR
*
A
21
Probability½ðV
A
ðT
Var
; xÞ?dx ¼ 1 2a ð2Þ
where VaR
*
A
denotes the result of calculating VaR using the most appropriate methods
for that portfolio (typically a combination of simulation, decision tree, historical, or
parametric methods).
For many portfolios of interest, accurate estimation of VaR
*
A
requires very large
Monte Carlo simulations and, when early exercise of options is considered, a stochastic
dynamic programming approach. The advantage of Monte Carlo simulation is that it
allows for an arbitrary level of accuracy depending on models and scenarios. The
disadvantage of Monte Carlo simulation is that calculations typically take 6-24 hours,
allowing for skewness, fat tails, etc.
The analytic delta-normal method can be used to rapidly approximate VaR, denoted
by VaR
0
, by restricting the distribution of V
A
by assuming: V
A
has derivatives with
respect to each argument, the state variables are the prices of the deal elements, the
periodic changes in value of V
A
with respect to each argument are jointly normally
distributed with mean zero. VaR
0
can be expressed as:
VaR
0
¼ 2
dV
A
dx
T
Var
þ ZðaÞ
????????????????
g
T
X
g
q
ð3Þ
The advantage of analytical approximation is the speed of approximation: millions of
portfolios can be calculated per second. This speed comes at the cost of accuracy since
the set of solutions is restricted by assumptions on the distribution of V
A
.
Figure 2.
Active permissioning
structure
Monte Carlo simulations
(24 hour turnaround)
Approximation analytic VaR
(sub-second turnaround)
Fast approximation to counterparty portfolio VaR
Accurate counterparty portfolio VaR
Price data
Portfolio
positions
Calibration procedure
Calibration variables
Centralized
clearing for
OTC derivatives
353
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The analytical approximation of VaR can be calibrated on the simulated VaR to
improve the approximation. This improved approximation is called position risk, PR
0
A
,
and may be written as:
PR
0
A
ðT
var
; Dx; a; CÞ ð4Þ
where C is calibration variable inserted into VaR
0
to calculate PR
0
A
. Then these
calibration variables are assigned values to minimize the difference:
C ¼ argmin
c
PR
0
A
ðT
var
; Dx; a; CÞ 2VaR
*
A
ðT
var
; Dx; aÞ
n o
ð5Þ
To support a rapid throughput of processing transactions, the calibration variables are
chosen with the viewpoint that VaR
*
A
can generally be computed within a 24-hour
period, while PR
0
A
is generally constrained to much tighter time limits. The calculation
of PR
0
A
is performed in real-time position risk system, while calculation of VaR
*
A
is
performed by a simulation-based position risk supervision system. The essential
feature of the combined systems is that PR
0
A
is calculated in real time, while VaR
*
A
is
calculated periodically.
4.4 Two-part pricing
To price clearing services, it is ?rst important to specify what quantity is exchanged in
the clearing market and how that quantity is priced. The quantity exchanged is default
risk which is calculated by our VaR method. There are many methods used to price
default risk which share a common goal: the price of risk is set to compensate the CCP
for the probability that a default occurs.
Currently, private CCPs charge a single fee for clearing with variation margin
requirements. Given the high level of uncertainty regarding default risk at the outset of
a contract, these CCPs typically overcharge for their clearing services. To most
accurately compensate the CCP for the probability of default, we propose using a
two-part pricing schedule:
fee ¼ a þ bxðM; f ; ZÞ ð6Þ
where a is a ?xed fee and b is the price of risk. The vector x is the portfolio risk of the
trader, which is based on the margin policy requirement, M, a vector of market price
feeds, f, and a vector of contract positions, Z. This non-uniform pricing schedule is
based on a simple two-part tariff (Tirole, 2004) composed of a ?xed access fee, a, and a
variable fee, b[10]. The functional form of xðM; f ; ZÞ is a many-to-one map isolating the
forces that determine the quantity of risk. Appendix 2 contains a detailed example of
two-part pricing.
Two-part pricing, which allows collateral or margins to be tailored to market
conditions, is more ef?cient than a one-part price with, or without, variable collateral
requirements, under a variety of assumptions because it allows for a risk-sharing
equilibrium. Ef?cient pricing is essential to adequately produce both the public and the
impure goods, since constrained pricing for clearing services necessarily generates an
equilibrium off the ef?cient frontier. Over, or under, pricing of clearing services would
lead to a sub-optimal amount of risk being borne by the CCP and would distort the
prices of OTC contracts.
JFEP
2,4
354
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
4.5 Exchange external reporting
A critical microstructure design issue is the effect of offsetting positions on the
determination of the price of risk. The CCP will only have a lens on the portfolio of
contracts held on the PPP exchange. If large offsetting positions are held off the exchange
and unknown to the CCP, the price of risk will be distorted. Systemically, large ?rms
might be required to report all of their capital structure continuously so the public
clearing component could be based on full knowledge of the ?nancial risks. As the price
of risk for that portion of portfolio on the CCP platform for a particular counterparty
begins to rise, there will be incentives for that counterparty to post additional collateral
and/or offsetting positions. This voluntary conduct will be in the self-interest of a
transactor if their overall portfolio includes offsetting positions or hedged transactions.
5. Conclusion
The bene?ts of the past year’s ex post and haphazard intervention have been
concentrated among ?nancial market participants who exploited the government’s
guarantee. The next cycle can only be more extreme as a consequence. A continued
failure to allocate the costs and bene?ts of the implicit, ad hoc public guarantee could
well continue to generate periodic catastrophic results. Regulations that inhibit ?nancial
innovation are not the answer, whether those regulations restrict speci?c forms of
contracts or restrict the allocation of economic rents among the producers of ?nancial
products.
A government-private partnership engaged in providing clearing services for
OTC derivative markets is feasible and does not require a technology leap. We have
argued here that one dimension of the technology leap is enabled by the technological
innovation of real-time permissioning and novation or guaranteeing of OTC ?nancial
contracts. Equally necessary is the real-time monitoring by the centralized clearing
organization during the negotiation of OTC ?nancial contracts. This requires that all
derivative exchanges and dealer networks be integrated into a cohesive uniform
communication and permissioning network using existing software communication
protocol routines[11].
The motivation for our proposed PPP is no less than the survival of the ?nancially
interdependent world that has been created over the last 20 years. Increasingly
sophisticated ?nancial market participants have learned how to maximize the value
they extract from the implicit guarantees provided by the world’s central banks or
“lenders of last resort.” In the most recent ?nancial crisis, exercising that guarantee has
pressed the ?nancial capacity of the global economy to an extreme not previously
witnessed even in the great depression. The bene?ts of the ex post and haphazard
intervention have been concentrated among ?nancial market participants who exploited
the under priced guarantee creating a system that is fraught with moral hazard. The
next cycle can only be more extreme as a consequence and a continued failure to allocate
the costs and bene?ts of the implicit, ad hoc public guarantee could well continue to
generate periodic catastrophic results.
Notes
1. This idea is not new: in 2004, Tim Geithner, then president of the Federal Reserve Bank of
New York, warned the Federal Open Market Committee that the $5 trillion CDS market
Centralized
clearing for
OTC derivatives
355
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
needed a CCP to control risk. At the time, the idea of a CCP lacked support from the ?nancial
sector and derivatives continued to be traded without centralized clearing.
2. Here, we draw the usual distinction between standardized, or vanilla, OTC derivatives that
have ?xed terms and conditions and custom, or complex, OTC derivatives that have variable
terms and conditions.
3. A problem with the current regulatory environment is the existence of overlapping
regulatory bodies, including Securities Exchange Commission, Commodity Futures Trading
Commission, Federal Deposit Insurance Corporation, and the Federal Reserve.
4. The systemic risk posed by consolidating clearing operations is discussed in Jones (2009).
5. Impure goods are either non-rival or non-excludable but not both. These goods lie on the
spectrum between public and private goods.
6. The most recent bailouts of ?nancial institutions are the last in a long line of government
interventions (i.e. CBOT silver in 1980 and NYME potatoes in 1976) that demonstrate its
willingness to save private clearing ?rms have become “too big to fail.”
7. Best bid and offer prices are the best available ask prices, when buying contracts, and the
best available bid prices, when selling contracts.
8. The proposed methodology for evaluating risk has been extensively tested and validated by
The Clearing Corporation (formerly the Board of Trade Clearing Corporation) supporting
documentation is available by request from the authors.
9. The value at risk answers the question, “What is the most portfolio A can lose – with a
1 2 a level of con?dence – over the next T
Var
days?” For example, an overnight, 99 percent
portfolio VaR of $1 million means that, under current conditions, 99 percent of the time, the
daily loss in the portfolio will not exceed $1 million.
10. The method we propose to quantify and price risk is quite common. Consider two-part risk
pricing used in car insurance: risk is quanti?ed using the purchaser’s driving record and that
risk is priced to compensate the insurance company for its expected loss. The risk is regularly
reevaluated and the price of risk changes in response to changes in the quantity of risk.
11. FIX stands for the Financial Information Exchange Protocol, which is an industry supported
standard for electronic communication of information about ?nancial contracts. It was ?rst
developed in 1992, is currently in Version 5.0, and is supported by most large participants in
?nancial markets.
12. Jones and Perignon (2008) discuss the effects of a CCP that provides a guarantee with no
explicit government involvement on systemic risk management.
References
Acharya, V., Engle, R., Figlewski, S., Lynch, A. and Subahmanyam, M. (2009), “Centralized
clearing for credit derivatives”, in Acharya, V. and Richardson, M. (Eds), Restoring
Financial Stability: How to Repair a Failed System, Wiley, New York, NY, pp. 251-68.
Bank for International Settlements (2010), “Semiannual over-the-counter (OTC) derivatives markets
statistics”, available at: www.bis.org/statistics/derstats.htm (accessed August 21, 2010).
Besley, T. and Ghatak, M. (2001), “Government versus private ownership of public goods”,
Quarterly Journal of Economics, Vol. 116 No. 4, pp. 1343-72.
Department of the Treasury (2009), “Financial regulatory reform”, available at: www.?nancials
tability.gov/docs/regs/FinalReport_web.pdf (accessed August 21, 2010).
Francesconi, M. and Muthoo, A. (2006), “Control rights in public-private partnerships”, IZA
Discussion Papers 2143, Institute for the Study of Labor (IZA), Bonn.
JFEP
2,4
356
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Grossman, S. and Hart, O. (1986), “The costs and bene?ts of ownership: a theory of vertical and
lateral integration”, Journal of Political Economics, Vol. 94 No. 4, pp. 691-719.
Jones, L. (2009), “Current issues affecting the OTC derivatives market and its importance to
London”, p. 21, available at: www.bourse-consult.com/wp-content/uploads/OTCDerivati
vesReportv21
Rausser, G. and Stevens, R. (2009), “Public-private partnerships: goods and the structure of
contracts”, Annual Review of Resource Economics, Vol. 1 No. 1, pp. 75-98.
Tirole, J. (2004), Theory of Industrial Organization, MIT Press, Cambridge, MA.
Williams, J. (2001), “Commodity futures and options”, in Gardner, B. and Rausser, G. (Eds),
Handbook of Agricultural Economics, North Holland, Amsterdam, pp. 746-816.
Further reading
Jones, R. and Perignon, C. (2008), “Derivatives clearing and systemic risk”, paper presented at the
Finance International Meeting AFFI – EUROFIDAI, December, Paris, available at:
http://ssrn.com/abstract¼1095695 (accessed August 21, 2010).
Appendix 1. Novation vs guarantees
In the microstructure for the proposed public-private CCP, a critical question arises in regard to
whether novation or a third-party guarantee for contracts is provided in case of default. Novation
has become the ?nancial guarantee of choice in regulated contract markets (Williams, 2001).
Novation means, literally, the remaking of the contract so that each original obligor (i.e. the
parties to the derivatives contract) is entirely removed and is directly replaced by novator
(i.e. the CCP). Though each party is replaced in the remade contract by the CCP, the contract has
the same terms and conditions as the original contract. At the time of settlement or default, all
enforcement and collection actions are taken directly against the novator and there are no direct
transfers between counterparties. Because the CCP becomes a direct counterparty to each side of
the trade, once novation occurs, the credit worthiness of the original counterparties is irrelevant
to each trader as the traders only have a contractual obligation to the CCP. Novation completely
isolates each party from the effects of a default by its counterparty, and indeed the counterparties
may be anonymous.
Unlike novation, a guarantee is a contingent, secondary form of obligation that supplements,
but does not replace, the original obligor. A CCP that only provides a third-party guarantee is
only involved if one or more parties default. In case of default, demand must be made on the
original obligator ?rst and that obligator must fail to perform before the CCP can be obligated to
ful?ll the contract. The CCP would typically only partially ful?ll the contract and a haircut would
be expected to be applied to the payments to each counterparty. A guarantee does not isolate the
parties from the effects of a default by a counterparty, but it does cap the losses of each party.
An additional difference between novation and guarantees is the level of anonymity among
traders. Novation allows for complete anonymity between trading parties and permits the CCP to
set universal standards for determining credit worthiness. Since guarantees place much of the
burden of determining creditworthiness on the trading parties, there can be no anonymity.
Anonymity combined with an elimination of counterparty credit risk between buyers and sellers
is largely responsible for the rapid growth in the volume of standardized ?nancial contracts over
the past 30 years. This growth has produced substantial bene?ts to the economy by making
prices of ?nancial products public information.
The choice between novation and guarantee determines the degree of active involvement by
the public sector. A CCP that selects novation for all OTC transactions would require regular,
direct involvement of the public sector in active management of the partnership. The partnership
would determine the credit standards for participation in the OTC trades and the government
Centralized
clearing for
OTC derivatives
357
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
would be compensated for all systemic risk insurance. If a CCP only provides guarantees for
OTC derivatives, the government would be involved when defaults exceed the CCP’s capital[12].
Clearly, the PPP must make a determination of whether they will implement the microstructure
for a novation or third-party guarantee process.
Appendix 2. An illustrative example of two-part pricing
Consider an equity contract between two counterparties, Aand B: party Abelieves that large value
caps will outperform growth and enters a $100 million contract that is long the Dow Jones
Industrial index (DJI), long the Standard and Poor 500 index (OEX), and short the NASDAQ 100
index (NDX). Party B, believing growth will outperform large value caps, enters a $100 million
contract that is short DJI, short OEX, and long NDX. Both parties post collateral of $50 million
re?ecting a 2:1 leverage ratio, with the remainder being a loan against the position. The duration of
the contract is one year and the data used in this analysis cover March 25, 2007 to March 24, 2008.
The daily volatility (exponentially weighted, l ¼ 0.94) of each index grew dramatically over
the course of the contract (Figure A1). The variation in party A’s position value was about
^10 percent over the period. The position risk for party A, measured as the 99 percent 10-day
VaR, grew approximately sixfold in response to the increasing daily volatility.
Using a single fee to price risk when facing such volatile risk (Figure A2) presents
an insurmountable ?nancial challenge: the CCP must either set a single fee, paid ex ante, to
compensate for the variation in risk over the life of the contract or require prohibitively large
cash collateral as default risk varies. However, as the “lender of last resort,” the government
implicitly does guarantee at least some portion of such contracts, without receiving compensation.
Clearing this contract with a two-part fee is simple: the price of risk (Figure A3) increases as
default risk increases. This price of risk can be calculated daily and applied to the VaR for the
transactors. It is true that by increasing the price of risk, the parties will likely ?nd their position
less attractive economically and may choose to hedge or liquidate. But that is an economic
decision and not constrained by inef?cient pricing of risk .
Figure A1.
Exponentially weighted
volatility
0
0.000
0.005
0.010
0.015
0.020
Index exponentially weighted volatility
50 100 150
Number of Days
200 250
JFEP
2,4
358
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Corresponding author
Reid Stevens can be contacted at: [email protected]
Figure A2.
Position risk
0
0.0e+00
2.0e+06
4.0e+06
6.0e+06
8.0e+06
P
o
s
i
t
i
o
n
r
i
s
k
1.0e+07
1.2e+07
1.4e+07
50 100 150
Index
200 250
Figure A3.
Price of risk
0
0e+00
2e+04
4e+04
6e+04
8e+04
1e+05
50 100
Index
150 200 250
Centralized
clearing for
OTC derivatives
359
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_129345043.pdf
Systemic risk propagated through over-the-counter (OTC) derivatives can best be
managed by a public-private central counterparty clearing house. The purpose of this paper is to
outline the market microstructure necessary for such a clearing house.
Journal of Financial Economic Policy
Centralized clearing for over-the-counter derivatives
Gordon Rausser William Balson Reid Stevens
Article information:
To cite this document:
Gordon Rausser William Balson Reid Stevens, (2010),"Centralized clearing for over-the-counter
derivatives", J ournal of Financial Economic Policy, Vol. 2 Iss 4 pp. 346 - 359
Permanent link to this document:
http://dx.doi.org/10.1108/17576381011100865
Downloaded on: 24 January 2016, At: 21:39 (PT)
References: this document contains references to 11 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 430 times since 2010*
Users who downloaded this article also downloaded:
Donald J . Mullineaux, Mark K. Pyles, (2010),"Bank marketing investments and bank performance", J ournal
of Financial Economic Policy, Vol. 2 Iss 4 pp. 326-345 http://dx.doi.org/10.1108/17576381011100856
David G. Tarr, (2010),"The political, regulatory, and market failures that caused the US financial
crisis: What are the lessons?", J ournal of Financial Economic Policy, Vol. 2 Iss 2 pp. 163-186 http://
dx.doi.org/10.1108/17576381011070210
Sven Blank, J onas Dovern, (2010),"What macroeconomic shocks affect the German banking system?:
Analysis in an integrated micro-macro model", J ournal of Financial Economic Policy, Vol. 2 Iss 2 pp.
126-148 http://dx.doi.org/10.1108/17576381011070193
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Centralized clearing for
over-the-counter derivatives
Gordon Rausser
Department of Agricultural and Resource Economics, University of California,
Berkeley, Berkeley, California, USA
William Balson
RBS Acquisition Company, Palo Alto, California, USA, and
Reid Stevens
Department of Agricultural and Resource Economics, University of California,
Berkeley, Berkeley, California, USA
Abstract
Purpose – Systemic risk propagated through over-the-counter (OTC) derivatives can best be
managed by a public-private central counterparty clearing house. The purpose of this paper is to
outline the market microstructure necessary for such a clearing house.
Design/methodology/approach – The paper proposes using an request for quote platform with an
active permissioning system that uses analytic approximations based on Monte Carlo simulation to
estimate default risk and a two-part pricing scheme to ef?ciently price that risk.
Findings – It is found that comprehensive clearing for complex and standardized derivatives is
feasible using the clearing framework.
Research limitations/implications – This research is limited by the authors’ ability to give
empirical examples. The paper gives a short example with data, but given the constraints on length,
cannot go into more detail.
Practical implications – This comprehensive clearing structure, in contrast to current proposed
government regulations, will not drive out the “good” with the “bad” OTC derivative instruments.
Originality/value – This is the only paper the authors are aware of that outlines a detailed
framework for clearing all OTC derivatives.
Keywords United States of America, Legislation, Financial risk, Derivative markets, Risk management
Paper type Research paper
1. Introduction
Over the course of the most recent ?nancial crisis, the government lacked regulatory
mechanisms to deal with ?rms whose failure could trigger the failure of other ?rms
through the over-the-counter (OTC) derivatives market and was forced to manage
the systemic risk posed by large ?nancial institutions on an ad hoc and ex post
basis. To prevent cascading defaults, the government facilitated the sale of some
large ?nancial institutions (e.g. Bear Stearns), allowed others to declare bankruptcy
(e.g. Lehman Brothers), and injected capital into many through the Troubled Asset
Relief Program (TARP) (e.g. Bank of America). Even the treatment of ?rms that
struggled after receiving payments through TARP was unpredictable; Citigroup and
AIG received additional support through direct public investment while commercial
investment trust received no further assistance beyond the initial support of $2.33
billion. The ad hoc nature of this process is unlikely to have been the most economically
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JFEP
2,4
346
Journal of Financial Economic Policy
Vol. 2 No. 4, 2010
pp. 346-359
qEmerald Group Publishing Limited
1757-6385
DOI 10.1108/17576381011100865
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
ef?cient choice except in the sense that urgency required action. These events have
demonstrated the need to develop a clear regulatory framework to ef?ciently manage
systemic risk whether posed by the OTC market or otherwise.
In this paper, we propose that systemic risk be managed directly through the
creation of a central counterparty clearing house (CCP). A CCP for OTC derivatives can
be designed to decrease systemic risk by eliminating default risk between the
counterparties to an OTC contract and by moderating the ?nancial incentives to
accumulate “excessive risk” in OTC markets[1]. The lack of such a CCP for OTC
derivatives, particularly credit default swap (CDS)’s, has been labeled as a signi?cant
factor in the current ?nancial crisis (Acharya et al., 2009) and the OTC derivatives
market, given its growth in recent years (Figure 1), will likely pose a larger threat to
future systemic stability without a CCP.
1.1 Summary of recent legislative history
Both the Administration and the Congress have proposed legislation to reform the
OTC derivatives market. These proposals resulted in the Dodd-Frank Act of 2010. The
Act encourages migration of OTC contracts to be cleared onto listed exchanges.
Unfortunately, the Act leaves to each derivatives clearing organization (DCO) the
decision of which contracts it will clear. As a result, the vast bulk of OTC derivatives
are unlikely to be cleared by DCOs in the near future, making it likely that systemic
risk will accumulate due to the bilateral credits risks of uncleared derivatives.
Several alternatives have been proposed prior to the Dodd-Frank Act. The
Administration’s proposal (Department of the Treasury, 2009) for ?nancial regulatory
reformexpanded on commitments by the previous administration to establish a clearing
house for OTC derivatives, focusing on promoting the public good by managing the
systemic risk posed by OTC derivatives and promoting transparency in the OTC
market. This proposal called for clearing “standardized OTC derivative transactions”
and increasing “regulatory capital requirements” on non-standardized derivatives[2],
though it contained no detailed plans for accomplishing these objectives.
Figure 1.
Ten-year growth of the
global OTC market and
exchange-traded
derivatives market –
100,000
200,000
300,000
400,000
500,000
600,000
700,000
Dec-99 Dec-01 Dec-03 Dec-05 Dec-07 Dec-09
N
o
t
i
o
n
a
l
v
a
l
u
e
(
b
i
l
l
i
o
n
s
o
f
U
S
d
o
l
l
a
r
s
)
OTC derivatives
Global OTC and exchange traded derivatives market
Exchange traded derivatives
Centralized
clearing for
OTC derivatives
347
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The House of Representatives had passed legislation that consolidated oversight[3],
required stronger capital cushions, regulated swaps, encouraged the use of a CCP, and
increased the cost of capital on all counterparties. The Senate’s Restoring American
Financial Stability Act, which was proposed at the committee level, focused on
increasing oversight of derivatives traders, not comprehensive clearing. As with the
Administration’s proposal, neither the House nor the Senate provided detailed plans to
manage the systemic risk posed by the OTC market.
Avery recent proposal (Volcker Rule) by the Administration aimed to prevent banks
from using their deposits as collateral for their OTC trading by implementing the spirit
of the repealed Glass-Steagall Act, which prohibited depository institutions from
operating in securities markets. The Dodd Frank Act contains a similar limitation.
In summary, a wide variety of legislative proposals have been advanced related to
OTC derivatives clearing, but the ?nal legislation only requires rule making by
commodity futures trading commission (CFTC) and securities and exchange commission
to accomplish that objective. We believe one explanation for going slow is the lack of a
detailed design for howa DCOcould reliablyclear the diversityof OTCderivatives. Inthis
paper, we present such a design.
1.2 The clearing dilemma
The heart of the dilemma faced by policy makers in reforming the derivatives market
is its size and complexity. As of June 2008, the notional value of all outstanding
OTC ?nancial contracts was in excess of $680 trillion, according to the Bank for
International Settlements (2010) (Figure 1). In contrast, the value of all cleared
derivatives traded by private regulated exchanges was below $20 trillion in notional
value. Given the limitless variation among derivatives (i.e. underlying assets, terms,
conditions, etc.), establishing capital requirements and clearing for this market have
long been technologically infeasible. Only recently have technology and ?nancial
theory reached the point that centralized clearing for both vanilla and complex
derivatives is possible.
We advance a proposal to:
.
centrally clear not only standardized but also complex derivatives with real-time
permissioning;
.
explicitly recognize the public-private partnership (PPP) structure needed to
effectively manage systemic risk;
.
implement ef?cient pricing of default risk and thus clearing;
.
allow structured ?nance to continue its pursuit of a world of more complete
contracts (e.g. bonds and CDS’s); and
.
incorporate government policy instruments to allow the public sector to be
compensated for controlling systemic risk over the business cycle.
Instead of imposing excessive capital requirements suggested by some proposals,
which would increase the cost of capital of trading such instruments, driving out the
“good” with the “bad,” our proposed framework will design a clearing solution that
can accommodate all derivatives, but price the clearing according to their relative risk.
This would allow traders to bene?t from the “good” complex contracts, and, with
centralized clearing, limit the systemic risk from the “bad” complex contracts.
JFEP
2,4
348
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2. Private clearing
Private DCOs have recently begun clearing a subset of OTC derivatives that will expand
as the CCPs develop methods to standardize some contracts and establish ef?cient
margin requirements. The largest private initiatives are the clearing operations associated
with IntercontinentalExchange, Chicago Mercantile Exchange, Eurex, and Euronext.
These ?nancial institutions began clearing OTC derivatives in response to pressure
from both the public and the private sectors following the most recent ?nancial crisis.
The clearing microstructure at these regulated exchanges is based on novation in
which the DCO becomes the counterparty (Appendix 1). Novation eliminates bilateral
counterparty risks but can cause the accumulation of ?nancial risk in the DCO.
Contracts are novated using a single fee for clearing along with initial and variation
margin requirements. The relationship between margin determination and the volatility
of underlyings is typically set by each exchange’s committee of experts. Contracts are
traded on a double-sided auction (i.e. a separate order queue for bids and offers) in which
the exchange determines set of tradable contracts with regulatory approval (e.g. CFTC).
Though each exchange has developed a platform that will ef?ciently
facilitate transactions for standardized OTC contracts with liquid markets, they are
not designed to provide complete public transparency or systemic risk management for
the broader OTC market. The exchanges lack suf?cient incentives to invest optimally
in transparency and systemic risk management because the bene?ts of managing
systemic risk generated cannot be appropriated by private CDOs. By only clearing
standardized, liquid contracts (i.e. CDS indices and interest rate swaps), the public will
only have price discovery on those segments of the OTC market, and the OTC market
for complex, custom derivatives will likely remain opaque. The private DCOs also lack
the capitalization necessary to remain solvent in the face of a ?nancial crisis with
cascading defaults in the OTC markets. Without adequate capitalization, the DCOs
cannot economically manage systemic risk and will be forced to rely on government
assistance in the event of a systemic crisis (despite legislative prohibitions on doing so).
As a result, DCOs have become another potential source of systemic risk, particularly
given the rapid consolidation that has taken place over the course of the last decade[4].
3. The need for a PPP
To correct the weaknesses of the private clearing initiatives, it is essential to
understand the nature of goods generated through centralized clearing: transparency,
systemic risk management, and the facilitation of trades. The pure facilitation of
liquid standardized trades is a private good, both rival and excludable. Transparency is
a public good, both non-rival and non-excludable, the bene?ts of which cannot be
entirely captured by a private DCO. The management of systemic risk is an impure
good[5]. Systemic risk management is non-excludable because the contracts of all
systemically important ?rms that default must be underwritten, either by the DCOor by
the government, but is rival because each defaulting ?rm requires a separate
incremental investment. Private DCOs lack suf?cient incentives to provide the socially
optimal level of clearing’s impure good.
3.1 Control rights and PPPs
A PPP can be formed to manage OTC derivatives’ CCP to ensure the optimal level of
the public, private, and impure goods are produced. A PPP could provide adequate
Centralized
clearing for
OTC derivatives
349
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
capitalization and ensure transparency while maintaining incentives to facilitate trades
through the assignment of control rights. The PPP literature has shown that ownership
should be determined by the type of good produced, and in the case of impure goods,
joint management by the public and private sectors can be optimal (Grossman and
Hart, 1986; Besley and Ghatak, 2001; Francesconi and Muthoo, 2006). Joint ownership,
which is determined by the contractual assignment of control rights, can provide both
sectors’ incentives to invest their resources and each suf?cient control to ensure
socially optimal levels of production (Rausser and Stevens, 2009).
A public-private OTC clearing partnership is inevitable, given the systemic
importance of OTC clearing and historic public sector support for failing ?nancial
institutions to prevent cascading defaults[6]. If, during a ?nancial crisis, a DCObecomes
insolvent, the government will be expected to act as lender of last resort. In the academic
literature, models of private OTC clearing implicitly assume that the government would
bailout a DCO to prevent widespread default, though there is no formal public
involvement or any compensation for the public sector to provide such services ( Jones
and Perignon, 2009).
If the partnership is explicitly recognized ex ante by forming a public-private CCP,
the government can manage system risk over the business cycle by choosing to be
compensated for the services it provides to stem systemic risk. If the partnership is
designed ex post, the implicit insurance provided by the public sector will only be
compensated on an ad hoc basis. But, the point of ex ante management is to moderate
the forces that lead to excess in advance rather than ex post. Beyond compensating
the government and strengthening their control over systemic risk, the creation of an
explicit partnership clari?es the “rules of the game” for derivatives markets and reduces
uncertainty over the government’s role during a ?nancial crisis.
4. Market microstructure
To improve the provision of public and impure goods, we propose the public-private
CCP use a request for quote (RFQ) platform along with the double-sided auction
platform currently used by private CCPs. Most OTC trading is currently conducted as
an informal RFQ auction in which telephone, fax, and electronic bulletin boards are
used to disseminate interest on either the buy or the sell side, and responses are made
that may or may not see competing bids or offers. Our proposal is to systematize this
informal process that already occurs.
4.1 RFQ platform
The RFQ execution platform we propose has four principle advantages:
(1) An RFQ platform allows traders to determine the set of complex contracts that
will be traded and cleared by combining standardized contract elements.
(2) An RFQ platform promotes transparency by publishing price feeds for all
traded complex contracts rather than just the standardized contracts.
(3) An RFQ platform can stem systemic risk by facilitating clearing for any
composite derivative, which would allow the government to require that
virtually all derivatives be cleared.
(4) An RFQ platform’s multi-party negotiations disseminate offer data to all
market participants, reducing the informational asymmetry in the OTC market.
JFEP
2,4
350
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
An RFQ market can coexist alongside existing double-sided auction markets and
indeed is the only auction mechanism capable of expanding clearing to almost all OTC
derivatives. In principle, double-sided auction markets are incapable of handling all
complex OTC derivatives due to the exponentially large problem of combinatorics.
For example, suppose an exchange lists and clears 10,000 names for standardized
option contacts with ten delivery dates and ten strike prices. Then there are at least a
trillion simple spread contracts with two legs. Clearly, a double-sided auction would be
unable to sustain such exponential growth. Fortunately, most transactions are focused
on a small fraction of such possibilities, but still there are too many. We believe that
any complex derivative composed of two or more standardized legs can reliably be
cleared using an RFQ transaction facility with the proposed pre-trade permissioning.
The transaction process begins on an RFQ trading platform when a subscriber
creates and posts the terms for a derivative contract. For example, if a subscriber
chooses to create a calendar spread on an RFQplatform, they determine the underlying,
strike price, class, quantity, and buy and sell expiration dates. Once posted, respondents
offer quotes and sizes for the contract, which are aggregated and disseminated to
all subscribers. Because the information processors for exchange-traded options
(i.e. Options Clearing Corporation and Options Price Reporting Authority) take complex
option structures as individual legs for clearing and reporting purposes, after a contract
is agreed on at its net debit/credit price, users can negotiate the actual leg prices prior to
?nal trade acceptance. At the end of a short RFQperiod, subscribers can “Post to Block”
for an RFQ they generated and move the request into the Block Trade facility to meet a
second, pre-de?ned party and affect a cross betweenthe best bid and offer prices[7] at the
end of the associated RFQ period. While the parties ?nalize the terms of a contract, the
CCP evaluates the contract’s risk (Sections 4.3 and 4.4) and sets the clearing fees
(Section 4.5).
4.1.1 Liquidity. To function, a double-sided auction requires market makers who
pro?t on the bid-ask spread, which is the spread between the price that clears buy orders
and the price that clears sell orders. The bid-ask spread compensates the market makers
for the costs of running its business as well as its retained risk (i.e. sometimes the ?owof
orders is imbalanced and the entity is expected to warehouse risk until the balance is
restored). It is not always true that the bid-ask spread compensates for retained risk,
especially in markets with few orders or in markets where the ?ow of orders is highly
imbalanced for a protracted period. In those cases, the observed market response is
typically that bid-ask spreads become very wide, liquidity dries up, and order ?ow
eventually ceases. In extreme cases, bids to buy can be entered as so-called “stub quotes”
and offers to sell can be entered at multiples of recent execution prices. The result is a
failed auction market in which trading is impeded by the auction design.
Since there is no bid-ask spread in an RFQ platform, it is not possible for the market
to collapse due to a lack of liquidity. In fact, the phone/fax methodology of the current
OTC market adjusts well to periods of varying liquidity. We expect that the order ?ow
will vary periodically and order executions would also vary. But market prices would
still be based on the most recent executions and quotes on the underlyings. Moreover,
restarting an RFQ market can begin as soon as order ?ow begins to become balanced,
unlike a double-sided auction in which the risk-reward relationship of a market maker
must also be re?ected in prices. Dysfunctional market prices, such as the ?ash crash,
are less likely in an RFQ market; since a buy and a sell order must be matched with no
Centralized
clearing for
OTC derivatives
351
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
intermediary, the concept of stub quotes does not occur. However, during disruptive
market periods, there can be a time variation of executed transactions.
Even in illiquid markets, an RFQ clearing platform can pro?tably clear contracts
given the typical size of such contracts and the variable risk-based pricing component.
Though we would not expect liquidity in all contracts, we anticipate that the
market would respond to improved transparency of prices, which cannot be achieved
on a double-sided auction platform burdened with inadequate liquidity. Moreover,
such transparency will decrease the informational advantage of ?nancial institutions
that have direct observation of order ?ows.
4.2 Clearing and active permissioning
An RFQ platform with clearing requires sub-second pre-trade permissioning to avoid
impeding the ?ow of executions. Since OTC products can be composed of multiple
underlyings, they can be decomposed into their elemental risks and the price feeds
from the standardized components can be employed to determine mark-to-market
value, even in illiquid markets. The decomposition procedure allows our approach to
utilize price feeds for the elements of an OTC contract and compute the risk properties
of an overall portfolio. OTC-customized instruments can and are priced on a real-time
basis, because multiple underlyings have mark-to-market forward price curves.
We propose combining an analytical value at risk (VaR) methodology with Monte
Carlo simulation to implement real-time OTC contract permissioning[8]. With this
approach, portfolios of hundreds thousands of non-linear contracts can be evaluated
in sub-second time intervals. Even aggressive trading behavior can, in principle, be
monitored in real time on a pre-execution basis without impeding the ?ow of
negotiations in the OTC market. The key design trade-off is the proper evaluation of
functionality designed for speed (simplicity) and accuracy (complexity).
Risk can be accurately assessed with relatively time-intensive Monte Carlo
methods. Monte Carlo methods are the only accurate method of assessing ?nancial
risk in markets characterized by irregular probability distributions, rapidly changing
volatility and correlation matrices, and highly non-linear payoff functions. Risk can
be quickly approximated using analytic VaR. Analytic value at-risk methods allow for
sub-second evaluation of risk for very large portfolios. Analytic methods are
amendable to very rapid calculations and analytic approximations are available for
almost all elemental derivatives.
Our essential insight is to use analytic methods as an extrapolation function that is
calibrated to the Monte Carlo simulations. It is possible to relax any restrictions on
mathematical form of the distribution of price changes by treating the analytic
expression as an extrapolation function with appropriate modi?cation enabling it to be
periodically calibrated to the results of Monte Carlo simulations (Figure 2).
4.3 Measuring VaR
VaR for a portfolio[9] is de?ned as a one-sided con?dence interval on portfolio losses:
Probability ðDV
A
0 ðT
Var
; DxÞ $ 2VaR
A
Þ ¼ 1 2a ð1Þ
where V
A
is the value of a portfolio of deal elements of interest, T
Var
is the risk horizon
of interest, x is a vector of random state variables, VaR
A
is the VaR of portfolio A, and
a is the level of con?dence. This formulation is applicable to any portfolio, any set of
JFEP
2,4
352
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
state variables, and any process governing the stochastic evolution of the value of the
portfolio. The portfolio may include deal elements that are securities, equities, bonds,
options, futures, derivatives, or other assets. The state variables may be prices on deal
elements, events that affect prices, external events, credit ratings, or other risk factors.
The price process may be a named stochastic process or may have jumps, reversion,
non-Markovian state evolution, stochastic volatility, discontinuities, or other features.
To use Monte Carlo simulation, VaR is rewritten as:
Z
VaR
*
A
21
Probability½ðV
A
ðT
Var
; xÞ?dx ¼ 1 2a ð2Þ
where VaR
*
A
denotes the result of calculating VaR using the most appropriate methods
for that portfolio (typically a combination of simulation, decision tree, historical, or
parametric methods).
For many portfolios of interest, accurate estimation of VaR
*
A
requires very large
Monte Carlo simulations and, when early exercise of options is considered, a stochastic
dynamic programming approach. The advantage of Monte Carlo simulation is that it
allows for an arbitrary level of accuracy depending on models and scenarios. The
disadvantage of Monte Carlo simulation is that calculations typically take 6-24 hours,
allowing for skewness, fat tails, etc.
The analytic delta-normal method can be used to rapidly approximate VaR, denoted
by VaR
0
, by restricting the distribution of V
A
by assuming: V
A
has derivatives with
respect to each argument, the state variables are the prices of the deal elements, the
periodic changes in value of V
A
with respect to each argument are jointly normally
distributed with mean zero. VaR
0
can be expressed as:
VaR
0
¼ 2
dV
A
dx
T
Var
þ ZðaÞ
????????????????
g
T
X
g
q
ð3Þ
The advantage of analytical approximation is the speed of approximation: millions of
portfolios can be calculated per second. This speed comes at the cost of accuracy since
the set of solutions is restricted by assumptions on the distribution of V
A
.
Figure 2.
Active permissioning
structure
Monte Carlo simulations
(24 hour turnaround)
Approximation analytic VaR
(sub-second turnaround)
Fast approximation to counterparty portfolio VaR
Accurate counterparty portfolio VaR
Price data
Portfolio
positions
Calibration procedure
Calibration variables
Centralized
clearing for
OTC derivatives
353
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The analytical approximation of VaR can be calibrated on the simulated VaR to
improve the approximation. This improved approximation is called position risk, PR
0
A
,
and may be written as:
PR
0
A
ðT
var
; Dx; a; CÞ ð4Þ
where C is calibration variable inserted into VaR
0
to calculate PR
0
A
. Then these
calibration variables are assigned values to minimize the difference:
C ¼ argmin
c
PR
0
A
ðT
var
; Dx; a; CÞ 2VaR
*
A
ðT
var
; Dx; aÞ
n o
ð5Þ
To support a rapid throughput of processing transactions, the calibration variables are
chosen with the viewpoint that VaR
*
A
can generally be computed within a 24-hour
period, while PR
0
A
is generally constrained to much tighter time limits. The calculation
of PR
0
A
is performed in real-time position risk system, while calculation of VaR
*
A
is
performed by a simulation-based position risk supervision system. The essential
feature of the combined systems is that PR
0
A
is calculated in real time, while VaR
*
A
is
calculated periodically.
4.4 Two-part pricing
To price clearing services, it is ?rst important to specify what quantity is exchanged in
the clearing market and how that quantity is priced. The quantity exchanged is default
risk which is calculated by our VaR method. There are many methods used to price
default risk which share a common goal: the price of risk is set to compensate the CCP
for the probability that a default occurs.
Currently, private CCPs charge a single fee for clearing with variation margin
requirements. Given the high level of uncertainty regarding default risk at the outset of
a contract, these CCPs typically overcharge for their clearing services. To most
accurately compensate the CCP for the probability of default, we propose using a
two-part pricing schedule:
fee ¼ a þ bxðM; f ; ZÞ ð6Þ
where a is a ?xed fee and b is the price of risk. The vector x is the portfolio risk of the
trader, which is based on the margin policy requirement, M, a vector of market price
feeds, f, and a vector of contract positions, Z. This non-uniform pricing schedule is
based on a simple two-part tariff (Tirole, 2004) composed of a ?xed access fee, a, and a
variable fee, b[10]. The functional form of xðM; f ; ZÞ is a many-to-one map isolating the
forces that determine the quantity of risk. Appendix 2 contains a detailed example of
two-part pricing.
Two-part pricing, which allows collateral or margins to be tailored to market
conditions, is more ef?cient than a one-part price with, or without, variable collateral
requirements, under a variety of assumptions because it allows for a risk-sharing
equilibrium. Ef?cient pricing is essential to adequately produce both the public and the
impure goods, since constrained pricing for clearing services necessarily generates an
equilibrium off the ef?cient frontier. Over, or under, pricing of clearing services would
lead to a sub-optimal amount of risk being borne by the CCP and would distort the
prices of OTC contracts.
JFEP
2,4
354
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
4.5 Exchange external reporting
A critical microstructure design issue is the effect of offsetting positions on the
determination of the price of risk. The CCP will only have a lens on the portfolio of
contracts held on the PPP exchange. If large offsetting positions are held off the exchange
and unknown to the CCP, the price of risk will be distorted. Systemically, large ?rms
might be required to report all of their capital structure continuously so the public
clearing component could be based on full knowledge of the ?nancial risks. As the price
of risk for that portion of portfolio on the CCP platform for a particular counterparty
begins to rise, there will be incentives for that counterparty to post additional collateral
and/or offsetting positions. This voluntary conduct will be in the self-interest of a
transactor if their overall portfolio includes offsetting positions or hedged transactions.
5. Conclusion
The bene?ts of the past year’s ex post and haphazard intervention have been
concentrated among ?nancial market participants who exploited the government’s
guarantee. The next cycle can only be more extreme as a consequence. A continued
failure to allocate the costs and bene?ts of the implicit, ad hoc public guarantee could
well continue to generate periodic catastrophic results. Regulations that inhibit ?nancial
innovation are not the answer, whether those regulations restrict speci?c forms of
contracts or restrict the allocation of economic rents among the producers of ?nancial
products.
A government-private partnership engaged in providing clearing services for
OTC derivative markets is feasible and does not require a technology leap. We have
argued here that one dimension of the technology leap is enabled by the technological
innovation of real-time permissioning and novation or guaranteeing of OTC ?nancial
contracts. Equally necessary is the real-time monitoring by the centralized clearing
organization during the negotiation of OTC ?nancial contracts. This requires that all
derivative exchanges and dealer networks be integrated into a cohesive uniform
communication and permissioning network using existing software communication
protocol routines[11].
The motivation for our proposed PPP is no less than the survival of the ?nancially
interdependent world that has been created over the last 20 years. Increasingly
sophisticated ?nancial market participants have learned how to maximize the value
they extract from the implicit guarantees provided by the world’s central banks or
“lenders of last resort.” In the most recent ?nancial crisis, exercising that guarantee has
pressed the ?nancial capacity of the global economy to an extreme not previously
witnessed even in the great depression. The bene?ts of the ex post and haphazard
intervention have been concentrated among ?nancial market participants who exploited
the under priced guarantee creating a system that is fraught with moral hazard. The
next cycle can only be more extreme as a consequence and a continued failure to allocate
the costs and bene?ts of the implicit, ad hoc public guarantee could well continue to
generate periodic catastrophic results.
Notes
1. This idea is not new: in 2004, Tim Geithner, then president of the Federal Reserve Bank of
New York, warned the Federal Open Market Committee that the $5 trillion CDS market
Centralized
clearing for
OTC derivatives
355
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
needed a CCP to control risk. At the time, the idea of a CCP lacked support from the ?nancial
sector and derivatives continued to be traded without centralized clearing.
2. Here, we draw the usual distinction between standardized, or vanilla, OTC derivatives that
have ?xed terms and conditions and custom, or complex, OTC derivatives that have variable
terms and conditions.
3. A problem with the current regulatory environment is the existence of overlapping
regulatory bodies, including Securities Exchange Commission, Commodity Futures Trading
Commission, Federal Deposit Insurance Corporation, and the Federal Reserve.
4. The systemic risk posed by consolidating clearing operations is discussed in Jones (2009).
5. Impure goods are either non-rival or non-excludable but not both. These goods lie on the
spectrum between public and private goods.
6. The most recent bailouts of ?nancial institutions are the last in a long line of government
interventions (i.e. CBOT silver in 1980 and NYME potatoes in 1976) that demonstrate its
willingness to save private clearing ?rms have become “too big to fail.”
7. Best bid and offer prices are the best available ask prices, when buying contracts, and the
best available bid prices, when selling contracts.
8. The proposed methodology for evaluating risk has been extensively tested and validated by
The Clearing Corporation (formerly the Board of Trade Clearing Corporation) supporting
documentation is available by request from the authors.
9. The value at risk answers the question, “What is the most portfolio A can lose – with a
1 2 a level of con?dence – over the next T
Var
days?” For example, an overnight, 99 percent
portfolio VaR of $1 million means that, under current conditions, 99 percent of the time, the
daily loss in the portfolio will not exceed $1 million.
10. The method we propose to quantify and price risk is quite common. Consider two-part risk
pricing used in car insurance: risk is quanti?ed using the purchaser’s driving record and that
risk is priced to compensate the insurance company for its expected loss. The risk is regularly
reevaluated and the price of risk changes in response to changes in the quantity of risk.
11. FIX stands for the Financial Information Exchange Protocol, which is an industry supported
standard for electronic communication of information about ?nancial contracts. It was ?rst
developed in 1992, is currently in Version 5.0, and is supported by most large participants in
?nancial markets.
12. Jones and Perignon (2008) discuss the effects of a CCP that provides a guarantee with no
explicit government involvement on systemic risk management.
References
Acharya, V., Engle, R., Figlewski, S., Lynch, A. and Subahmanyam, M. (2009), “Centralized
clearing for credit derivatives”, in Acharya, V. and Richardson, M. (Eds), Restoring
Financial Stability: How to Repair a Failed System, Wiley, New York, NY, pp. 251-68.
Bank for International Settlements (2010), “Semiannual over-the-counter (OTC) derivatives markets
statistics”, available at: www.bis.org/statistics/derstats.htm (accessed August 21, 2010).
Besley, T. and Ghatak, M. (2001), “Government versus private ownership of public goods”,
Quarterly Journal of Economics, Vol. 116 No. 4, pp. 1343-72.
Department of the Treasury (2009), “Financial regulatory reform”, available at: www.?nancials
tability.gov/docs/regs/FinalReport_web.pdf (accessed August 21, 2010).
Francesconi, M. and Muthoo, A. (2006), “Control rights in public-private partnerships”, IZA
Discussion Papers 2143, Institute for the Study of Labor (IZA), Bonn.
JFEP
2,4
356
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Grossman, S. and Hart, O. (1986), “The costs and bene?ts of ownership: a theory of vertical and
lateral integration”, Journal of Political Economics, Vol. 94 No. 4, pp. 691-719.
Jones, L. (2009), “Current issues affecting the OTC derivatives market and its importance to
London”, p. 21, available at: www.bourse-consult.com/wp-content/uploads/OTCDerivati
vesReportv21
Rausser, G. and Stevens, R. (2009), “Public-private partnerships: goods and the structure of
contracts”, Annual Review of Resource Economics, Vol. 1 No. 1, pp. 75-98.
Tirole, J. (2004), Theory of Industrial Organization, MIT Press, Cambridge, MA.
Williams, J. (2001), “Commodity futures and options”, in Gardner, B. and Rausser, G. (Eds),
Handbook of Agricultural Economics, North Holland, Amsterdam, pp. 746-816.
Further reading
Jones, R. and Perignon, C. (2008), “Derivatives clearing and systemic risk”, paper presented at the
Finance International Meeting AFFI – EUROFIDAI, December, Paris, available at:
http://ssrn.com/abstract¼1095695 (accessed August 21, 2010).
Appendix 1. Novation vs guarantees
In the microstructure for the proposed public-private CCP, a critical question arises in regard to
whether novation or a third-party guarantee for contracts is provided in case of default. Novation
has become the ?nancial guarantee of choice in regulated contract markets (Williams, 2001).
Novation means, literally, the remaking of the contract so that each original obligor (i.e. the
parties to the derivatives contract) is entirely removed and is directly replaced by novator
(i.e. the CCP). Though each party is replaced in the remade contract by the CCP, the contract has
the same terms and conditions as the original contract. At the time of settlement or default, all
enforcement and collection actions are taken directly against the novator and there are no direct
transfers between counterparties. Because the CCP becomes a direct counterparty to each side of
the trade, once novation occurs, the credit worthiness of the original counterparties is irrelevant
to each trader as the traders only have a contractual obligation to the CCP. Novation completely
isolates each party from the effects of a default by its counterparty, and indeed the counterparties
may be anonymous.
Unlike novation, a guarantee is a contingent, secondary form of obligation that supplements,
but does not replace, the original obligor. A CCP that only provides a third-party guarantee is
only involved if one or more parties default. In case of default, demand must be made on the
original obligator ?rst and that obligator must fail to perform before the CCP can be obligated to
ful?ll the contract. The CCP would typically only partially ful?ll the contract and a haircut would
be expected to be applied to the payments to each counterparty. A guarantee does not isolate the
parties from the effects of a default by a counterparty, but it does cap the losses of each party.
An additional difference between novation and guarantees is the level of anonymity among
traders. Novation allows for complete anonymity between trading parties and permits the CCP to
set universal standards for determining credit worthiness. Since guarantees place much of the
burden of determining creditworthiness on the trading parties, there can be no anonymity.
Anonymity combined with an elimination of counterparty credit risk between buyers and sellers
is largely responsible for the rapid growth in the volume of standardized ?nancial contracts over
the past 30 years. This growth has produced substantial bene?ts to the economy by making
prices of ?nancial products public information.
The choice between novation and guarantee determines the degree of active involvement by
the public sector. A CCP that selects novation for all OTC transactions would require regular,
direct involvement of the public sector in active management of the partnership. The partnership
would determine the credit standards for participation in the OTC trades and the government
Centralized
clearing for
OTC derivatives
357
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
would be compensated for all systemic risk insurance. If a CCP only provides guarantees for
OTC derivatives, the government would be involved when defaults exceed the CCP’s capital[12].
Clearly, the PPP must make a determination of whether they will implement the microstructure
for a novation or third-party guarantee process.
Appendix 2. An illustrative example of two-part pricing
Consider an equity contract between two counterparties, Aand B: party Abelieves that large value
caps will outperform growth and enters a $100 million contract that is long the Dow Jones
Industrial index (DJI), long the Standard and Poor 500 index (OEX), and short the NASDAQ 100
index (NDX). Party B, believing growth will outperform large value caps, enters a $100 million
contract that is short DJI, short OEX, and long NDX. Both parties post collateral of $50 million
re?ecting a 2:1 leverage ratio, with the remainder being a loan against the position. The duration of
the contract is one year and the data used in this analysis cover March 25, 2007 to March 24, 2008.
The daily volatility (exponentially weighted, l ¼ 0.94) of each index grew dramatically over
the course of the contract (Figure A1). The variation in party A’s position value was about
^10 percent over the period. The position risk for party A, measured as the 99 percent 10-day
VaR, grew approximately sixfold in response to the increasing daily volatility.
Using a single fee to price risk when facing such volatile risk (Figure A2) presents
an insurmountable ?nancial challenge: the CCP must either set a single fee, paid ex ante, to
compensate for the variation in risk over the life of the contract or require prohibitively large
cash collateral as default risk varies. However, as the “lender of last resort,” the government
implicitly does guarantee at least some portion of such contracts, without receiving compensation.
Clearing this contract with a two-part fee is simple: the price of risk (Figure A3) increases as
default risk increases. This price of risk can be calculated daily and applied to the VaR for the
transactors. It is true that by increasing the price of risk, the parties will likely ?nd their position
less attractive economically and may choose to hedge or liquidate. But that is an economic
decision and not constrained by inef?cient pricing of risk .
Figure A1.
Exponentially weighted
volatility
0
0.000
0.005
0.010
0.015
0.020
Index exponentially weighted volatility
50 100 150
Number of Days
200 250
JFEP
2,4
358
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Corresponding author
Reid Stevens can be contacted at: [email protected]
Figure A2.
Position risk
0
0.0e+00
2.0e+06
4.0e+06
6.0e+06
8.0e+06
P
o
s
i
t
i
o
n
r
i
s
k
1.0e+07
1.2e+07
1.4e+07
50 100 150
Index
200 250
Figure A3.
Price of risk
0
0e+00
2e+04
4e+04
6e+04
8e+04
1e+05
50 100
Index
150 200 250
Centralized
clearing for
OTC derivatives
359
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
3
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_129345043.pdf