Clearing Services for Global Markets A Framework for the Future Development of the Clearing Industry_8 potx

271 What theory reveals – framework for analysis of network strategies
Efficiency
Gains
Profit Increase
LOW HIGH
LOW
HIGH
1
2
3
4
Efficiency
Losses
Profit Decrease
CM
PR-G
/CM
AR-G
1
2
3
4
Cross-Margining Agreements
Clearing Links
Mergers & Acquisitions
Single CCP
Focus Area
Figure 7.14 Business Model Impact Matrix for regionally-to-globally active clearing members
Source: Author’s own.
Cross-margining agreements lead to low efficiency gains and result in a
proportional profit increase.

42
In contrast, clearing links, M&A and Single
CCP initiatives translate into a disproportionately higher profit increase. This
increase results from the positive revenue impact brought about by these
network strategies, which, in addition to reducing costs, increase profits.
From a prop. perspective, reg ionally-to-globally focused clearers benefit from
the increased scale and scope of services available at a lower cost as well
as from positive network effects. Disintermediation and enhanced internal
efficiency now increase the attractiveness of self-clearing additional prod-
ucts and markets in-house, which in turn enables the clearer to le verage its
infrastructure – thus increasing the theoretical ‘revenues’ resulting from self-
clearing.
Similar considerations apply toregionally-to-globally active clearers with an
agency focus. In this case, the disproportionately higher profit increase results
from the clearer’s enhanced internal efficiency, which positively impacts its
profit margin. Additionally, the clearer is able to offer its clients an increased
scale and scope of services at lower cost. Finally, clearing links and Single
42
Cross-margining agreements are assumed to have little or no impact on revenues, because a significant
increase in revenues would require an increase in the number of cleared contracts spurred by this type
of network initiative. Whereas this could occur in reality, such a scenario is not included in this analysis.
272 Clearing Services for Global Markets
CCP initiatives, which enable the CCP to internalise GCM level network
effects, further strengthen and enhance the value proposition of regionally-
to-globally active clearers. Their revenues are thus increased thanks to a greater
profit margin and increased customer traffic.
43
7.3.3 BMIM for globally active clearing members
Finally, Figure 7.15 illustrates the Business Model Impact Matrix for globally
active clearing members. In contrast to the previously outlined matrices, the

impact of network strategies on the business model of g lobally active clearers
with a prop. or an agency focus is strikingly different.
For clearers with a prop. focus, the efficiency gains generated by a cer-
tain network strategy translate into a proportional profit increase. Network
strategies are not assumed to impact revenues, because the scale and scope of
services are not broadened. However, clearing services for different products
and markets become accessible throug h a more consolidated infrastructure.
44
This allows clearers to leverage their infrastructure and reduce costs, but in
contrast to regionally-to-globally active clearers, globally active clearing mem-
bers have to overcome additional internal hurdles to realise these internal
efficiency gains. Globally active clearers usually employ different legal entities
within their company structure to become members of regional CCPs. This
set-up can give rise to additional internal and external complexities as well
as potential areas of conflict with regard to consolidating the clearing house
interfaces.
43
Note that by definition, regionally-to-g lobally active clearing members and their NCMs and/or cus-
tomers have an interest in being active in all of the partnering clearing houses’ markets and products.
It could be argued that due to internal economies of scale, a globally active clearing member can offer
the same products and markets at a lower price, so NCMs and/or customers of regionally-to-globally
active clearers should in theory consider becoming a member of the higher volume (i.e. globally active)
clearer. If the NCMs and/or customers base their decision purely on the basis of commissions charged
by the clearers, network strategies do not result in NCMs and/or customers conducting more of their
business through their re gionally-to-globally active clearer. Instead, these NCMs and/or customers can
be expected to switch to a globally active clearer. However, globally active clearers already had a compet-
itive advantage in terms of greater internal economies of scale prior to the network initiative; it would
thus have already been attractive for NCMs and/or customers to choose the globally active clearer, had
their sole decision criterion been commissions. It is therefore assumed that NCMs and/or customers of
regionally-to-globally active clearing members do not base their decision strictly on commission levels,

but also take additional factors into account, such as regional proximity, services tailored to the home
market, long-term business relationship, etc.
44
Revenues would only increase to the extent that the enhanced internal efficiency spurs business growth.
This scenario, which would lead to a disproportionately higher profit increase, is not covered by this
analysis, however.
273 What theory reveals – framework for analysis of network strategies
Efficiency
Gains
Profit Increase
LOW HIGH
LOWHIGH
1
3
2
4
Efficiency
Losses
Profit Decrease
CM
PG
/CM
AG
1
2
3
4
CM
AG
CM

PG
1
2
3
4
Cross-Margining Agreements
Clearing Links
Mergers & Acquisitions
Single CCP
Focus Area
Figure 7.15 Business Model Impact Matrix for globally active clearing members
Source: Author’s own.
Internal complexities can emerge by virtue of the different legal enti-
ties working under different local budgets and local management. Salaries,
bonuses and hierarchies may depend on these budgets, which can make
the consolidation of clearing house interfaces a difficult and complex inter-
nal undertaking. External difficulties related to consolidating clearing house
interfaces can result from NCMs and other customers objecting to such an
initiative, i.e. for reasons of demanding local contacts, etc.
For globally active clearers with a prop. focus, all network st rategies
positively impact their business model. The same is not true for globally
active clearers with an agency focus, however. On the one hand, this type
of clearing member benefits from high efficiency gains, as outlined in the
Efficiency Impact Matrix, but at the same time, they are at risk of being
disintermediated.
45
As outlined above, the internalisation of GCM level net-
work effects increases the attractiveness of disintermediation.
46
45

Although it is unlikely that a globally active clearer would be put out of business completely, it is
also unlike ly that the remaining NCMs would do more business (thus making up for the lost NCM
business), unless lower internal costs were to be translated into reduced commissions, which would in
turn translate into an increase of the number of cleared contracts.
46
Regionally-to-globally active clearers could, of course, also run the risk of being disintermediated by
NCMs that now consider it cost-efficient to self-clear their business. This scenario is not scrutinised here,
274 Clearing Services for Global Markets
Figure 7.7 showed that clearing links have a high to very high potential
to internalise GCM level network effects,
47
whereas Single CCP and M&A
strategies possess a high and medium potential for internalising GCM level
network effects, respectively. It is consequently assumed that in the context
of M&A initiatives, efficiency gains are counteracted by a medium strong
risk of disintermediation, which results in lost revenues. The overall effect is
consequently assumed to be a very low profit increase.
Because clearing link and Single CCP initiatives significantly increase the
attractiveness of disintermediation, it is assumed that revenue losses ultimately
outweigh the efficiency gains. The only way for globally active clearers with an
agency focus to circumvent the risk of disintermediation is to restrict access
to the CCP level networ k.
48
Influencing the level of access to the CCP network is assumed to be possi-
ble when clearing houses are user-owned and/or user-governed. Figure 7.16
illustrates the potential efficiency gains for globally active clearers with an
agency focus that succeed in restricting access to the CCP network; doing
so enables them to circumvent disintermediation and the associated rev-
enue losses. These clearers stand to gain from restricting access to the CCP
level network under clearing links, M&A and Single CCP initiatives. However,

regionally-to-globally active clearers suffer from restricted access, because they
lose their disintermediation benefits and are prevented from leveraging their
internal infrastructure. A scenario of restr icted access would thus serve to rein-
force the structural particularities of the European Value Provision Network
by sustaining the competitive advantage held by high volume clearers over
lower volume clearers. Globally active clearers would consequently maintain
their dominant position within the European VPN, i.e. a very high percentage
of the European market share in derivatives clearing would continue to be
concentrated in the hands of a few very high volume clearers.
as this analysis is concerned with the disintermediation of globally active GCMsby regionally-to-globally
active clearers.
47
Note that the risk of disintermediation through clearing links is minimal, unless the CCPs are able to
convince clearing members to expect positive future network effects – i.e. by persuading them that the
clearing link will be extended inscope, and that the clearing houses will make up for the lost intermediary
level by providing these services themselves, thus successfully internalising GCM level network effects.
Only then can the utility derived from the network outweig h the costs of alternation in the long run,
and the starting problem will be overcome. For the purpose of this analysis, it is assumed that clearing
link initiatives overcome the starting problem. If, in reality, the link initiative fails to convince clearing
members to expect positive future network effects, then a Single CCP initiative has by far the greatest
potential for successful disintermediation.
48
Restricted access in this case refers to scenarios in which only globally active clearers are granted full
access to a CCP level network resulting from a clearing link, M&A or Single CCP initiative.
275 What theory reveals – framework for analysis of network strategies
Efficiency
Gains
Profit Increase
LOW HIGH
LOWHIGH

1
3
2
4
Efficiency
Losses
Profit Decrease
CM
PG
/CM
AG
1
2
3
4
Only
achievable if
CM
AG
can
implement
restricted
access
CM
AG
CM
PG
1
2
3

4
Cross-Margining Agreements
Clearing Links
Mergers & Acquisitions
Single CCP
Focus Area
Figure 7.16 Business Model Impact Matrix for globally active clearers that succeed in implementing restricted
access to the CCP level network
Source: Author’s own.
To summarise, although some network strategies are cost efficient, they are
not necessarily profit-maximising for all clearing member types. Regionally
active clearers with a prop. or an agency focus suffer from a negative impact
on their business model when M&A and Single CCP initiatives are in force.
Globally active clearers with an agency focus do not fare we ll under Single
CCP initiatives and clearing links (when these replicate the size of the Single
CCP network) that allow unrestricted access to the CCP level network.
7.4 Preliminary findings – impact of network strategies on efficiency
Building on the findings of Chapters 5 and 6, Chapter 7 establishes a frame-
work for analysing the impact of network st rategies on the efficiency of Euro-
pean clearing. Four mat rices were used to assess the impact of cross-margining
agreements, clearing links, M&A and Single CCP initiatives and allowed pre-
liminary conclusions to be drawn about the impact of these network strategies
on the efficiency of clearing. The preliminary conclusions drawn from these
276 Clearing Services for Global Markets
matrices w ill be compared to real-world case studies in Chapter 8 , which are
analysed according to the above-mentioned framework. The findings of the
matrices are br iefly summarised below, compared to findings of the European
Central Bank, and used to further clarify the insights provided by the so-called
‘McP Curve’ and to evaluate the claim that the clearing industry exhibits
‘natural monopoly’ characteristics.

7.4.1 Scale Impact Matrix
r
The Scale Impact Matrix serves to classify the magnitude of p otential
demand- and supply-side scale effects related to a particular network strat-
egy.
r
The findings from the Scale Impact Matrix suggest that clearing links hold
the potential for the greatest ‘net’ scale economies. Whereas M&A strateg ies
have higher supply-side and lower demand-side scale effects than the Single
CCP scenario, the relative magnitude of their respective net scale effects is
comparable.
7.4.2 Transaction Cost Impact Matrix
r
The Transaction Cost Impact Matrix analysed whether or not the demand-
and supply-side scale effects of each network strategy translate into a propor-
tional or disproportional impact on transaction costs for different clearing
member types.
r
The analysis differentiated the following customer groups: CM
PR
/CM
AR
(regionally active clearers with a prop. or an agency focus), CM
PR-G
/CM
AR-G
(regionally-to-globally active clearers with a prop. or an agency focus) and
CM
PG
/CM

AG
(globally active clearers with a prop. or an agency focus).
r
For regionally active clearers with a prop. or an agency focus, indirect costs
are the core cost driver. Cross-margining agreements have no cost impact for
this ty pe of clearer, while M&A and Sing le CCP initiatives actually increase
indirect costs. Clearing links, on the other hand, do not imply additional
indirect costs, and could potentially lead to cost reductions. Nonetheless,
clearing links are not suited to reduce significantly these clearers’ indirect
costs.
r
For regionally-to-globally active clearers with a prop. or an agency focus,
indirect costs are also the core cost driver. These clearers benefit from
network strategies enabling them to clear many markets through their
domestic home clearing house, thus disintermediating the clearer(s) they
277 What theory reveals – framework for analysis of network strategies
employ as intermediary (ies) to other markets. Cross-margining agree-
ments serve slightly to lower indirect costs, but do not enable disinterme-
diation. While clearing links, M&A and Single CCP initiatives are all suited
to reduce significantly these clearers’ indirect costs, clearing links (followed
by Single CCP and M&A initiatives) turn out to have the greatest positive
impact.
r
For globally active clearers with a prop. focus, direct costs are the core cost
driver. These clearers’ main benefit from network strategies comes from the
potential to reduce clearing house charges. Cross-margining agreements
serve slightly to lower costs, but have no potential significantly to reduce
direct costs. Clearing links, M&A and Single CCP initiatives, on the other
hand, are all suited to impact significantly these clearers’ core cost driver.
For globally active clearers with a prop. focus, the Single CCP initiative

possesses the strongest cost-reduction potential, followed by clearing links
and M&A initiatives.
r
For globally active clearers with an agency focus, indirect costs are the core
cost driver. These clearers’ main benefit from network strategies comes from
being able to consolidate their various clearing relationships, i.e. centralise
these into a single relationship. Cross-margining agreements serve slightly
to lower costs, but do not enable such a consolidation. Clearing links, M&A
and Single CCP initiatives are all suited to impact significantly these clearers’
core cost driver. For globally active clearers with an agency focus, clearing
links (followed by Single CCP and M&A initiatives) turn out to have the
greatest positive impact.
7.4.3 Efficiency Impact Matrix
r
The Efficiency Impact Matrix consolidated the results of the Transaction
Cost Impact Matrices and allowed for conclusions on the overall efficiency
impact of the various network strategies to be drawn.
r
The Efficiency Impact Matrix shows that besides cross-margining agree-
ments, which generally have little or no impact on efficiency, clearing links
are the only network strategy that results in efficiency gains for all clearing
member types.
r
Regionally active clearers potentially suffer from efficiency losses as a result
of M&A and Single CCP initiatives.
r
Nonetheless, while clearing links are suited to reduce significantly the core
cost driver of globally active clearers with a prop. focus, these clearers enjoy
the greatest efficiency gains from a Single CCP initiative.
278 Clearing Services for Global Markets

7.4.4 Business Model Impact Matrix
r
The Business Model Impact Matrix illustrated whether the efficiency
increase (decrease) corresponding to the different network strategies trans-
lates into a propor tional or disproportional profit increase or decrease for
different clearing member types.
r
It illustrated that the efficiency increase (decrease) resulting from different
network strategies translates into a proportional profit impact for regionally
active clearers; clearing link, M&A and Single CCP initiatives lead to a
disproportionately higher profit increase for regionally-to-globally active
clearing members.
r
For globally active clearers with a prop. focus, the efficiency gains of network
strategies translate into a proportional profit increase, whereas for globally
active clearers with an agency focus, Single CCP and clearing link initiatives
can lead to a disproportional decrease in profits.
r
The analysis shows that although some network strategies are cost efficient,
they are not necessarily profit-maximising for all clearing member types.
r
Regionally focused clearers with a prop. or an agency perspective suffer from
efficiency losses and a negative impact on their business model when M&A
and Single CCP initiatives are undertaken.
r
Although globally active clearers with an agency focus benefit from efficiency
gains, they simultaneously suffer from a negative impact on their business
model if Single CCP or clearing link initiatives (which replicate the size of
the Single CCP network) are set up.
r

It was outlined above that particularly large investment banks are active in
both proprietary and agency business. The Business Model Impact Matrix
illustrated the internal conflict that arises when these firms have to decide
whether or not to support a certain network strategy: the only network
strategy that has a positive impact on the business model of both the prop.
and the agency side relates to M&A initiatives between CCPs. These clearers
are thus likely to support M&A initiatives, but unlikely to support a clearing
link set-up. For globally active clearers with a prop. focus, a Single CCP ini-
tiative is preferable over clearing links, as it has the greatest positive potential
impact. For globally active clearers with an agency focus, clearing links actu-
ally have a significantly negative impact on their business model, which is
why they have an interest in circumventing such a scenario. Whether or
not large investment banks support the creation of a Single CCP will prob-
ably depend on whether the bank places greater emphasis on proprietary
279 What theory reveals – framework for analysis of network strategies
or agency business. From a prop. perspective, they should absolutely be in
favour of a Single CCP; from an agency perspective, however, the y can be
expected to oppose the implementation of a Single CCP.
7.4.5 Summary of impact
To summarise on the impact that network strategies have on the efficiency
of European clearing, the preliminary findings suggest that a clearing link
set-up appears to be the most attractive network strategy because it enables
all clearing member types to benefit from efficiency gains.
As outlined above, the services provided as well as the products and markets
processed by different European clearing houses are not identical, and due
to their installed bases, they are not necessarily interchangeable. Enlarging
the size of a CCP network through clearing links benefits European clearing
houses because it enables them to leverage their significant installed base
and helps them to sustain their unique services in cases where these services
and processes have been tailored to the specific demands of regional market

particularities and different regulatory environments. Additionally, it enables
clearing houses to internalise the GCM level network effects and to strengthen
their unique CCP level network effects. Clearers can additionally benefit from
the high growth potential afforded by links; they may entice other clearing
houses outside of the defined European markets to join the network. Links
are also not prone to give rise to negative network effects.
The European Central Bank (ECB) finds that ‘[i]t is clear that, in the
short-term, a single infrastructure would maximise network externalities and
economies of scale. However, these short-term advantages have to be balanced
against the inefficiencies that may be caused in the long run by the absence
of competition (e.g. a lack of dynamism and innovation).’
49
In accordance
with the doubts expressed by the ECB, the analysis suggests that consolidating
the European clearing industry through mergers and acquisitions to form a
single monopoly CCP runs the risk of depriving mar ket participants of the
competitive forces inherent to a link solution and denies them the benefit of
utilising clearing services that have been tailored to best fit regional market
particularities. Findings from the Scale Impact Matrix additionally suggest
that even in the short term, a single infrast ructure created through a Single
CCP initiative does not serve to maximise network externalities or economies
49
European Central Bank (ed.) (2001c), p. 3.
280 Clearing Services for Global Markets
of scale. Rather, clearing links incorporate greater potential for economies
of scale and scope as well as network effects. The Efficiency Impact Matrix
further illustrated that while a Single CCP could potentially lead to efficiency
gains for regionally-to-globally and globally active clearers, regionally active
clearers suffer efficiency losses. This analysis thus indicated that, even in the
short term, a Single CCP is not suited to enhance the efficiency for all European

clearing members. However, clearing members benefit from a link solution,
because it does not oblige participation and entails lower invest ments. The
link strategy thus translates into efficiency gains for all European clearing
members.
Nonetheless, the benefits of a link set-up can only be achieved if the clearing
houses endeavour to overcome the link-inherent starting problem.
50
The
starting problem will only be overcome and the utility derived from the CCP
level network can only offset the individual clearer’s costs of alternation in the
long run when the partnering CCPs:
r
convince their clearers to expect positive future network effects – implying
that the clearing link will be extended in its scope;
r
compensate for the lost intermediary (GCM) level by providing most of
these services themselves – thereby successfully internalising GCM level
network effects; and
r
make up for the lost participation in other CCP networks by providing a
similar service level and processing all products and markets.
Additionally, the benefits of a link set-up can only fully flourish when the
partnering clearing houses provide a choice of clearing location. If no choice
of clearing location is offered, network participants are locked into a partic-
ular clearing network.
51
However, again, this situation does not represent an
instance of strong lock-in: it is not a coordination failure among users that
prevents clearers from actually switching to the superior network, but the
legal restrictions established by the exchanges or clearing houses that prohibit

the choice of clearing location. T herefore, to era dicate non-network-effect-
related coordination problems on the CCP level, the provision of choice of
clearing location helps to prevent this particular scenario. Consequently, pro-
vided that opportunities for choice of clearing location exist through clearing
links, it appears unlikely that CCP level network effects lead to strong forms
of lock-in. If strong lock-in effects do not exist on the GCM and CCP levels,
first-mover-wins and winner-takes-all conditions are also unlikely to emerge.
50
Refer to section 7.1.2.2.2 for an explanation of the starting problem.
51
Note that clearing houses alone cannot offer choice of clearing location. Rather in a first instance, the
respective exchanges must give allowance to the use of various clearing houses.
281 What theory reveals – framework for analysis of network strategies
Market participants then benefit from the competitive forces between clearing
houses in the context of a clearing link set-up.
The analysis reveals that it is crucial for the success of clearing link initiatives
that the partnering CCPs endeavour to compensate for the lost intermediary
(GCM) level services by providing most of these services themselves. However,
it also showed that it is very difficult for clearing houses to internalise the entire
range of GCM level network effects. It can thus be assumed that there will
continue to be a justification for the value-added function of the GCM level,
despite the att ractiveness of disintermediation spurred by such a link initiative.
For some counterparties, it will simply remain more cost efficient to retain
an intermediary relationship with their GCM(s). Even so, it should be noted
that globally active clearers with an agency perspective can still be expected
to oppose the implementation of such a clearing link scenario, because their
business model is likely to suffer significantly from disintermediation.
52
The only way for globally active clearers with an agency focus to circum-
vent the risk of disintermediation is to restrict access to the newly created

CCP level network – thus circumventing disintermediation and the associ-
ated revenue losses. Influencing the level of access to the CCP network is
assumed to be possible only in the case of user ownership or user governance.
Despite the finding that a clearing link set-up offers greater efficiency gains for
these clear ing member types, globally active clearers with an agency focus are
likely to prefer the establishment of a Single CCP over the implementation of
clearing links. The reason for this is that it is presumably easier to succeed in
implementing user ownership or user governance (thus controlling the rules
of clearing) for one entity than it is to control the rules of clearing of various
entities.
The analysis also revealed that globally active clearers with a prop. focus
benefit most, in terms of efficiency gains, from the implementation of a
Single CCP. Nonetheless, these clearers will presumably only adopt a Single
CCP approach if they have the means to control the rules of clear ing, i.e. to
ensure that clearing fees are sufficiently reduced and that the consolidated
infrastructure does not abuse its monopoly position. Nonetheless, the r isk of
a user ownership/user governance structure that is dominated by high volume
clearers is that these clearing members will succeed in further strengthening
their already preferential position within the European VPN and that low and
medium clearers will consequently become further disadvantaged.
52
Note that in spite of this negative impact on the business model, the overall efficiency of the industry
can still be increased, because disintermediation reduces transaction costs.
282 Clearing Services for Global Markets
∞
∞
0
Number of CCPs
Private Benefits
2002

Public Benefits
Public/Private Benefits
Figure 7.17 McP curve – public versus private benefits in the context of CCP consolida tion
Source: McPartland (2002), p. 105.
These conclusions are in accord with and serve to clarify further the insights
provided by the ‘McP Curve’.
53
The curve represents the magnitude of private
and public benefits (vertical axis)
54
associated with a decreasing number of
CCPs, which is indicated on the horizontal axis and decreases from right
to left.
The upper line illustrates the financial benefits derived by clearing partici-
pants given the number of CCPs. The lower line represents the public benefits
associated with the number of Central Counterparties. From right to left, it
can be seen that it is in everyone’s best interest – including the p ublic interest –
to have fewer and fewer CCPs; this was at least the case until approximately
2002. McPartland observes that up to roughly this point in time, there were
increasing private financial benefits for clearing participants as well as con-
current public benefits associated with having fewer clearing houses. After
2002, however, the remaining benefits are private; the public benefits begin to
become negative at some future point in time. He argues that the real benefit of
decreasing the number of clearing houses is that the high volume clearers can
greatly simplify their internal operations and achieve significant internal cost
53
Cf. McPartland (2002), pp. 105–6.
54
Magnitude is represented on the vertical axis. It is primarily a measure of financial magnitude when
applicable to private benefits (top line), and intrinsic magnitude when applicable to public benefits

(lower line). Many, but not all, public benefits are financial in nature. Nonetheless, the relev ant public
benefits, given the number of CCPs, are expressed on the vertical axis as though it were a financial
measure (relative to the level of private benefits). Cf. McPartland (2002), p. 105.
283 What theory reveals – framework for analysis of network strategies
savings. McPartland (2002) concludes that the greatest cost savings beyond
this particular point are consequently private financial benefits, which accrue
to the largest clearing members as a result of their significantly streamlined
operating efficiencies.
The findings of the analysis in Chapter 7 reinforce the insights provided by
the McP Curve, but they also serve to clarify them further. They show that
the above conclusions are only true for a reduction in the number of clearing
house networks resulting from a Single CCP initiative or a clearing link set-up
in which high volume clearing members succeed in restricting access to the
CCP level network (and thus in impeding disintermediation). However, if the
consolidation of the CCP networks is achieved through a clearing link set-up
in which the partnering clearing houses engage to overcome the link-inherent
starting problem, choice of clearing location is provided and access to the
network is unrestricted, the findings of the McP Curve are unlikely to hold.
It is thus only when a Single CCP solution or link set-up that restricts access
to the CCP network is employed that the number of clearing house networks
becomes reduced beyond the point where high volume clearers (especially
clearing members with a prop. focus) benefit most, and public benefits begin
to become negative.
Clearing members’ ability to control the rules of clearing and influence the
structural set-up or leve l of access to the CCP level network should therefore
be minimised. Additionally, it is crucial that clearing houses understand the
benefits of engaging in network strategies that enable them to leverage their
installed base, strengthen their unique CCP level network effects, internalise
GCM level network effects and enlarge the scope of complementary products
and services provided by their CCP network.

Finally, when strong network effects and economies of scale exist in an
industry, it is often argued that the industry constitutes a naturalmonopoly.
55
Nonetheless, it is disputed whether clearing qualifies as a natural monopoly.
56
The literature both supports and refutes the asser tion; some sources advo-
cate classifying the securities clearing and settlement industry as a natural
55
‘A natural monopoly exists if a single supplier can serve the market in question more cost-efficiently
than several suppliers, meaning that the cost function in the relevant area of demand is subadditive.
Reviews of the cost side of networks focus primarily on the bundling advantages achieved through
economies of scale and economies of scope in service provision. These bundling advantages can imply
that a single network operator may be able to serve a given market at a lower cost than a number of
competing suppliers.’ Knieps (2006), p. 49. Also refer to Baumol (1977); and Baumol/Panzar/Willig
(1982).
56
Cf. European Central Bank (ed.) (2001b), p. 84.
284 Clearing Services for Global Markets
monopoly,
57
whereas others oppose this view.
58
Most of these contributions,
however, tend to refer to the clearing services provided by CSDs/ICSDs rather
than by CCPs.
The preliminary findings of this study provide no evidence supporting the
natural monopoly assumption of the European derivatives clearing industry.
First of all, the services provided by the different European clearing houses
are not necessarily interchangeable. Second, provided that opportunities for
choice of clearing location exist through clearing links, it appears u nlikely

that CCP leve l network effects lead to strong forms of lock-in. If strong lock-
in effects do not exist on the GCM and CCP leve ls, first-mover-wins and
winner-takes-all conditions are consequently also unlikely to exist. Third,
despite strong economies of scale and network effects, consolidation in the
form of a monopolistic Single CCP was shown potentially to induce highly
counteractive forces in the form of necessary investments, costs related to the
complexity of integ ration (possibly giving rise to diseconomies of scale) and
negative network effects. Finally, provided that choice of clearing location is
guaranteed by clearing houses through links, the European clearing industry
can be assumed to be contestable.
The second step of the analysis of the impact of network strategies on
the efficiency of clearing comes next: Chapter 8 compares the preliminary
findings of this chapter with conclusions from three case studies. This allows
final conclusions to be drawn.
57
See Competition Commission South Africa (2001); Cruickshank (2001), p. 325; European Securities
Forum (ed.) (2002), p. 6; Lee (2002), p. 12; Russo (2002), p. 237; European Financial Services Round
Table (ed.) (2003), p. 2; Heckinger/Lee/McPartland (2003), p. 15; Singapore Exchange (ed.) (2004), p.
8; CONSOB (ed.) (2004), p. 24; Office of Fair Trading (ed.) (2004), pp. 43–4; Branch/Griffiths (2005),
pp. 4–5; Rochet (2005), p. 9; Bliss/Papathanassiou (2006), p. 36; LIBA (ed.) (2006), p. 6; and Direction
G
´
en
´
erale du Tr
´
esor et de la Politique
´
Economique (ed.) (2006), p. 5.
58

See, e.g. Federation of European Securities Exchanges (ed.) (2004), p. 5; and Knieps (2006), p. 59. Milne
(2002), pp. 10–13, outlines that certain core functions provided by custodians have the characteristics
of a natural monopoly, whereas all other clearing and settlement-related services are competitive in
their nature. Also refer to the studies of Van Cayseele/Voor de Mededinging (2005); and Serifsoy/Weiß
(2005).
8
Checking theory against reality – case
studies of network strategies
The previous chapter delivers preliminary finding s regarding the impact of
network strategies on the efficiency of clearing; this chapter challenges these
conclusions with findings from the empirical study and real-world case stud-
ies. In the following, a case study analysis is performed for each of three types
of network strategy: clearing links (including cross-margining agreements),
1
mergers and acquisitions, and a Single CCP.
As the centreof the analysis is on the Europeanclearing industry,the selected
case studies focus on network strategies that involve European clearing houses.
The clearing link study (section 8.1) presents findings from the empirical study
concerning cross-margining agreements and clearing links and analyses the
link established in 2003 between Eurex Clearing (a Frankfurt-based clearing
house) and the Clearing Corporation (a Chicago-based clearing house). T he
mergers and acquisitions study (section 8.2) examines the merger between the
London Clearing House (LCH) and Clearnet (a Paris-based clearing house)
in 2003, which produced LCH.Clearnet. In contrast to the other two case
studies, the Single CCP study (section 8.3) deals with a hypothetical scenario:
the creation of a single clearing house for Europe.
Each case analysis mines the interviewees’ perceptions of the respective
network strategy in terms of its overall benefits and constraints. Additionally,
the interviewees’ feedback regarding the suitability of clearing links or a Single
CCP to integrate European clearing is presented.

2
Whilst all three case studies
are analysed according to the framework established in Chapter 7, the length
and detail of the respective studies varies: because the analysis conducted in
Chapter 7 identified clearing links as the most suitable network strategy to
1
Note that due to their very low efficiency impact, cross-marg ining agreements are not analysed separately,
but rather as part of the clearing link study.
2
The suitability of mergers and acquisitions as a means to integrate the European industry is not analysed
in the merger study (section 8.2), but in t he Single CCP study (section 8.3). The reason for this is that
consolidating the European clearing industry through M&A initiatives results in the creation of a Single
CCP.
286 Clearing Services for Global Markets
CHECKING THEORY AGAINST REALITY –
CASE STUDIES OF NETWORK STRATEGIES
Clearing Link Study
8.1
8
Challenges the preliminary conclusions on the
impact of network strategies on the efficiency of
clearing with insights from the empirical study and
case study findings.
Presents findings from the empirical study on
cross-margining agreements and clearing links.
Case study: The clearing link established in 2003
between The Clearing Corporation and Eurex
Clearing.
Case study: The merger between the London
Clearing House and Clearnet in 2003.

Presents findings from the empirical study on a
single CCP.
Case study: The creation of a single clearing
house for Europe (a hypothetical scenario).
Merger Study8.2
Single Central Counterparty Study8.3
Final Assessment – Impact of Network Strategies on Efficiency
8.4
PURPOSECHAPTER
Figure 8.1 Structure of Chapter 8
increase the efficiency of the European clearing industry, the clearing link
study receives more weight in the subsequent analysis and is conducted in
greater detail. Furthermore, due to the limited scope of this study and the lack
of detailed data available regarding the merger and Single CCP studies, these
two c ases are presented in abbreviated form.
In the final assessment (section 8.4), conclusions are drawn regarding the
impact of network strategies on the efficiency of European clearing; summaris-
ing whether the insights from the case studies support or refute the findings
from Chapter 7. This process also reveals the shortcomings of a purely the-
oretical assessment by outlining the difficulties associated with translating
theoretical concepts into reality.
8.1 The clearing link study
This chapter presents insig hts from the empirical study and a real-world case
study, the conclusions of which are compared to the findings from Chapter 7.
First, findings from the empirical study regarding cross-margining agreements
and clearing links are presented (section 8.1.1). Second, a more detailed
analysis of a concrete case study, i.e. the clearing link established between
287 Checking theory against reality – case studies of network strategies
Eurex Clearing and The Clearing Corp oration, is furnished (section 8.1.2).
The section concludes by summarising the findings regarding the impact of

clearing links on efficiency (section 8.1.3).
8.1.1 Findings from the empirical study
The empirical study furnished insights as to what role the different network
strategies play in reality by soliciting feedback from various stakeholders
on two critical issues. Interviewees were asked to critique cross-margining
agreements (section 8.1.1.1) and clearing links (section 8.1.1.2)intermsof
their ability to provide value-added; the interviews produced valuable details
about the major benefits and constraints of these initiatives. Furthermore, the
stakeholders were asked to comment on the suitability of clearing links as a
means to integrate the European clearing industry (section 8.1.1.3).
8.1.1.1 Cross-margining agreements: gener al benefits and constraints
The industry has hugely benefited [from cross-margining agreements], because the
cost of carrying offsetting positions is greatly diminished, and a lot more volume
ensues.
3
Theeconomicbenefitgivenwasneversufficienttomakeitworthwhile Ihavemore
money in my pocket now than the industry has ever saved in cross-margining.
4
Roughly 57 per cent (forty-five out of seventy-nine) of the interviewees issued
concre te opinions on cross-margining agreements.
5
At first sight, their over-
all assessment appears unambiguous: thirty-two interviewees felt that cross-
margining agreements provide value-added and increase the efficiency of
clearing, ten respondents saw no value-added and three interviewees were
conflicted (see Figure 8.2).
However, the picture becomes fuzzier upon closer analysis. Although the
great majority of the interviewed clearing house and exchange representatives,
market experts and NCMs acknowledged the value-added provided by such
agreements, the clearing members’ responses were more divided. Seven clear-

ers asserted that such network strategies provide no value-added; four clearers
made the opposite claim. On the other hand, three clearers saw general value
in cross-margining agreements, but felt that these initiatives could potentially
harm their business model. These clearers are thus classified as ‘conflicte d’.
3
Interview with William C. Floersch.
4
Statement made by interviewed market expert.
5
The remaining interviewees either said that they did not know, had no clear opinion, did not specify
their opinion or declined to answer.
288 Clearing Services for Global Markets
10
No Value-
Added
CROSS-MARGINING AGREEMENTS
Conflicted
3
Value-
Added
32
No Value-
Added
7 CM
2 ME
1 CH
Value-
Added
8 CH
4 EX

11 ME
4 CM
5 NCM
Conflicted
3 CM
No Value-
Added
4 LON
3 EU
3 US
Value-
Added
19 US
4 EU
9 LON
Conflicted
2 US
BY INTERVIEWEE GROUP
BY INTERVIEWEE LOCATION
1 LON
Figure 8.2 Interviewees’ assessment of cross-margining agreements
6
Source: Author’s own.
The fact that the US-based interviewees provided the bulk of the con-
crete assessments
7
is likely due to the significantly greater number of cross-
margining initiatives in the US as compared to Europe.
8
To understand the

different views expressed with regard to the value-added of cross-margining
agreements, the following presents the general benefits and constraints of such
initiatives as identified by the interviewees themselves.
Interviewees claimed that in terms of benefits, cross-margining agreements
generally enable a more efficient use of capital by minimising the costs of
carrying offsetting positions. When cross-margining opportunities extend
6
Interviewee groups: CM – clearing member; NCM – non-clearing member; ME – market expert; EX –
exchange; and CH – clearing house. Interviewee locations: US – United States; EU – Continental Europe;
and LON – London.
7
69 p er cent of all US-based inter viewees provided assessments, whereas only 52 and 41 per cent respec-
tively of all London- and Continental-Europe-based respondents gauged the relevance of this network
strategy.
8
Refer to section 3.3.2 for an overview of European and American cross-marg ining initiatives.
289 Checking theory against reality – case studies of network strategies
across different asset classes, clearing members can better leverage the same
amount of capital or financial resources, because they can assume a more
integrated risk profile. The higher the correlation between the products for
which cross-margining is offered, the greater the potential benefits in terms of
reduced cost of capital and risk management, as the need for risk monitoring
decreases. Reducing the aggregate level of margin that a clearing member is
required to post to collateralise the positions held through the participating
clearing houses can ultimately lead to higher volumes. This is especially true
when counterparties are collateral poor – and thus highly sensitive to the
cost of capital – because capital is thereby freed up that can then be allocated
for additional trading opportunities. The primar y benefit of cross-margining
thus lies in its potential to reduce the cost of capital. Consequently, whether or
not cross-margining agreements have appeal as stand-alone initiatives greatly

depends on individual clearing members’ sensitivity to the cost of capital.
9
In addition to the benefits that cross-margining agreements can provide,
the interviewees identified a variety of constr aints inherent to the utilisation
and implementation of this network strategy. First of all, clearing members
will gener ally consider the trade-off between the potential savings they expect
to realise and the back-office and IT development work that will be necessary
in order to be able to interact with the cross-margining schema. As outlined in
section 3.3.1.1, some initiatives require cumbersome processing. Additionally,
as collateral-rich clearing members generally tend to be less interested in cross-
margining agreements, the issue of required system changes is not so much
about costs, but rather the prioritisation of the multiple demands placed
on their IT resources. For globally active clearers with a prop. focus, the
implementation of cross-margining agreements is contingent on the internal
demand. If a clearer is able to pass on collateral savings to the individual trader
level (i.e. has an appropriate internal system in place for this purpose), traders
will clearly strive to benefit from such savings, which can affect their bottom-
line. The attractiveness of cross-margining agreements is consequently related
to whether the charge on capital is computed efficiently within firms.
The enforcement or auditing of those capital charges is still pretty primitive and
happens from time to time. It doesn’t happen daily, the way I would have thought;
it’s kind of monthly that people look at these things. I think if re gulators require a
more regular audit or measurement of how banks or intermediaries in the market are
covering the liabilities they have, then they will have to manage them more actively,
banks themselves will have to manage them more actively. That means that yesterday
9
Refer to the findings of section 5.1.4.2 for details on the interviewees’ assessments of the cost of capital.
290 Clearing Services for Global Markets
they didn’t bother to move a position into a clearing system, today they feel they need
to, because there will be an economic driver to do it. Banks just don’t care enough

about it, because they’re not under pressure to perform in that way.
10
High volume clearers with an agency focus are hesitant to support such
initiatives, because cross-margining agreements potentially counteract one of
the value propositions of their business model.
The job of the broker is to make order out of the chaos of different clearing systems
and different methodologies for the customer. To the extent that the exchanges and
clearing houses themselves do that, they disintermediate us, the broker. So, you know
it is probably good for the clients; it is probably not so good for us.
11
Additionally, as globally active clearers with an agency focus usually calculate
their customers’ risk position and the respective margin requirements by
taking into account all of their customers’ positions at a multitude of clearing
houses, the benefit created through a particular cross-margining agreement
might not be substantial from the customers’ point of view. In this case,
the customers of a GCM are unlikely to put pressure on their clearer to
participate in a particular cross-margining agreement. Finally, whether or not
the collateral offsets induced by a cross-margining initiative are convincingly
powerful highly depends on the respective trading strategy:
If you look at the Bund and the Treasury, they are highly correlated; if we are short in
Europe, we’re generally also short in the US, so we won’t get any margin offset. For a
trading house like us whenwemodelthecollateraloffset,itisnotaspowerfulas
one might imagine that it could be.
12
Finally, interviewees also identified several constraints in terms of implement-
ing cross-margining agreements on the clearing house level. On the one hand,
most clearing processes are electronic, so it is easier to quantify the risk as
well as the opportunities associated with correlated products. On the other
hand, it is difficult to find highly correlated products cleared at different CCPs
in the first place. This is particularly true for cross-margining in products

across different currencies.
13
When considering whether or not to engage in
10
Statement made by interviewed exchange representative.
11
Interview with Gary Alan DeWaal.
12
Statement made by interviewed clearing member representative.
13
Cf. interviews. This was also the reason that one of the interviewed clearing houses attributed no
value-added to cross-margining agreements: ‘Because you do not have total, full correlation between
products in different currencies, you would have to set the parameters so far away from each other
that I don’t think you would get much benefit from it.’ Statement made by interviewed clearing house
representative.
291 Checking theory against reality – case studies of network strategies
cross-margining agreements, clearing houses have to weigh up the benefits
versus the expected competitive impact.
8.1.1.2 Clearing links: general benefits and constraints
Now, is a clearing link going to be bad for the market? It is going to be bad for some
people in the market; is it going to be bad for the end-users that use those markets?
Well, it is probably going to be very good for them, because it will increase the amount
of competition, lower costs, and this will lead to an increase in volumes. So it will be
good for the market as a whole.
14
Links are largely irrelevant for us because we pretty much have established operations
in all the places we want to be, and as we go to new places, we expand that presence.
15
As for clearing links, roughly 71 per cent (fifty-six out of seventy-nine) of all
interviewees gave a concrete assessment (see Figure 8.3).

16
Again, akin to the
cross-margining analysis, the assessments were not monolithic.
The majorit y of interviewees assessed clearing link initiatives as provid-
ing value-added, but (as compared to cross-margining agreements) more
respondents denied the potential value-added or were conflicted. The assess-
ment provided by clearing members was again diverse. Ten clearers felt that
clearing link strategies provide no value-added, while six clearers held the
opposite view. Four clearing members were conflicted; on the one hand, they
noted benefits associated with link solutions, but also cited the potential harm
these initiatives could inflict on their business model. To elaborate on the dif-
ferent views vis-
`
a-vis the value-added of clearing links, the general benefits and
constraints of such initiatives (as identified by the interviewees themselves)
are presented next.
The interviewees emphasised a number of benefits. First of all, clearing
links can reduce transaction costs by providing a single point of entry to var-
ious markets. This reduces interface and system maintenance and hardware
costs, and leads to more streamlined management of open positions. Link
solutions can also generate collateral saving s
17
and potentially enhance liq-
uidity. In addition, clearing links can potentially improve upon stand-alone
cross-margining agreements: links are mechanically usually more efficient and
allow for greater STP than plain cross-margining arrangements. The latter are
more cumbersome by nature, because the cooperation between the clearing
14
Interview with Michael G. McErlean.
15

Interview with Paul J. Brody.
16
The remaining participants either felt that they did not know, had no clear opinion, did not specify
their opinion or declined to answer.
17
Provided the link agreement includes cross-margining arrangements between the partnering clearing
houses. Cf. interviews.
292 Clearing Services for Global Markets
17
No Value-
Added
CLEARING LINKS
Conflicted
7
Value-
Added
32
6 CH
4 EX
13 ME
Value-
Added
6 CM
3 NCM
Conflicted
4 CM
1 NCM
1 ME
1 CH
No Value-

Added
10 CM
3 ME
2 CH
2 EX
BY INTERVIEWEE GROUP BY INTERVIEWEE LOCATION
No Value-
Added
4 LON
4 EU
9 US
Value-
Added
13 US
6 EU
13 LON
Conflicted
1 US
3 LON
3 EU
Figure 8.3 Interviewees’ assessment of clearing links
18
Source: Author’s own.
houses is limited to a singular service. Due to the restricted scope of pure
cross-margining agreements, the part nering clearing houses have to limit the
risk sharing among them. These limits tend to present operational burdens
or incur additional technology costs to clear ing members.
The unique character of links enables clearers to gain local access to foreign
markets, thus permitting a choice of jurisdiction. Note that the interviewees’
opinions differed on the utility of this supposed benefit:

Jurisdiction is important! We would never clear, as an example, a Eurodollar in
Singapore, because we don’t understand the jurisdiction the way we do in the US. But
when you look at Europe and the US, I don’t think that it matters that much. I don’t
think that we wouldn’t clear in the US because of jurisdiction. As long as you are in
an OECD regulated capital market I think that it doesn’t matter.
19
18
Interviewee groups: CM – clearing member; NCM – non-clearing member; ME – market expert;
EX – exchange; and CH – clearing house. Interviewee locations: US – United States; EU – Continental
Europe; and LON – London.
19
Interview with Steve G. Mar tin.
293 Checking theory against reality – case studies of network strategies
I do not think that it is relevant anymore. I think once upon a time, when the regulatory
world was uneven, there was advantage to that. I do not see a material difference in
the level of regulation around the world anymore. This is the big difference of the last
10 years. I think everyone is sort of at a common playing field.
20
Whereas the ultimate benefit of jurisdictional choice within Europe and the
US offered through clearing links is debatable, interviewees acknowledged the
advantage that clearing links can give to smaller, less developed and possibly
niche markets and their local users. Links can provide these markets with
an opportunity to be part of bigger pools of liquidity. Regionally-to-globally
active clearers in particular can thus benefit from direct access to the away
market:
Many of the Asian brokerage firms tend to be smaller; they don’t have the capital of
a Morgan Stanley or a Goldman Sachs, so the Asian clearing houses can play a role
in representing the financial standing and credit standing of their smaller brokerage
firms. Just to give you an example, the Eurex clearing house would not take a lot of
Asian-based brokers as their clearing members, but they might take the Singapore

clearing house or someone with a stronger capital base. So, yes, I think these clearing
links are going to be very, very important in the future.
21
Accessing smaller and less developed markets through a clearer’s trusted and
well-established CCP relationship can spur the development of these markets,
particularly when market participants trust that the CCP’s level of expertise
can be transferred to the partnering clearing house, thus increasing the attrac-
tiveness of interacting with this CCP.
Besides these general benefits, interviewees identified a number of
constraints regarding the utilisation and implementation of clearing link
initiatives. As is the case with cross-margining initiatives, whether or not
a link set-up delivers benefits to clearers depends on individual cost-benefit
analyses as well as on the particular clearing houses’ requirements. In line
with the findings of the theoretical analysis, interviewees pointed out the lim-
ited att ractiveness and benefits of clearing links for globally active clearers.
Not only have most of them already implemented a relatively efficient global
infrastr u cture, but for clearers with an agency focus, links also expose them
to the threat of disintermediation.
I don’t think clearing links add much value. I real ly do believe that there is a limited
amount of value that a clearing link is able to bring to the table. I think that the value
a clearing link brings is largely geared to local players in the mar ket and it essentially
20
Interview with Gary Alan DeWaal.
21
Interview with Fred Grede.
294 Clearing Services for Global Markets
disintermediates the global intermediaries who have provided that service; and I’m
not too bothered by it, to be perfectly honest, but some of my competitors get very
worried.
22

I am slightly conflicted in that part of what we have as a competitive advantage is
that we can actually offer all these exchange accesses anyway. People wil l come to us,
because we can do it. If you narrow that down, then what I think will happen is that
you end up with the big institutions, the big hedge funds just becoming members
themselves, which gets rid of our competitive advantage. We are always a bit conflicted
about this one, I think!
23
Whereas interviewees observed that links can theoretically help to enhance
the efficiency of globally active clearers’ in-house operations, they also noted
that the final link set-up must provide great benefits and involve low costs of
implementation in order for it to be considered attractive.
But essentially, again, the brokers have found a way around this inefficiency and that’s
why you havetoreallyquestionwho isgoingto payforanyfurtherlinks, and I
think the incentive to pay is low.
24
The potential for cost savings for globally active clearers is taken down another
notch if an additional fee is charged for the utilisation of the link. Regionally-
to-globally active clearers will benchmark the link fee charged for the clearing
of positions at the away CCP with the commissions charged by their respect ive
GCM. However, globally active clearers only end up better off if their internal
efficiency gains exceed the link fee, or if the fee is lower than the internal costs
of utilising their global infrastructure. However, respondents stressed that the
efficiency gain must be fairly substantial in order to make a link solution
attractive to globally active clearers.
Even if they [the clearing house] were to charge a tenth of a cent for this facility, it is
going to exceed the IT and back-office savings. So, you can see our sensitivity to any
increase in fees is very high.
25
Interviewees also confirmed this study’s previous findings regarding the start-
ing problem inherent to clearing links. Unless their customers pressure them

to utilise the initiative, clearers with an agency focus are unlikely to adopt a
link initiative.
26
Globally active clearing members with a prop. perspective also
22
Statement made by interviewed clearing member representative.
23
Ibid.
24
Ibid.
25
Ibid.
26
Retail investors, for example, focus on investing in their home countries’ securities, and are unlikely to
put great pressure on the clearer to adopt this strategy. Cf. interviews. For details on investors’ market
choices, refer to Schiereck (1995); Schiereck (1996a); and Schiere ck (1996b).
295 Checking theory against reality – case studies of network strategies
have little incentive to participate, unless the initiative translates into signifi-
cant cost savings that can be passed on to the individual traders. Whereas the
regionally focused clearers interviewed did not express much interest in link
initiatives, the regionally-to-globally active clearers confirmed their general
interest in this network strategy.
In addition to these limitations concerning the adoption of links on the
clearing members’ part, interviewees identified several other obstacles that
might discourage clearing houses from voluntarily engaging in clearing link
arrangements.
I think the clearing houses, the providers in this space, are more likely to merge or
buy one another than they are to have this nice world where everybody allows systems
to be interlinked and cross-margin one another.
27

Implementing clearing links can be cumbersome for clearing houses, espe-
cially on the regulatory side. They have to establish default proceedings and
obtain regulatory approval. The process becomes even more unwieldy when
links involve clearing houses that are subject to different regulatory regimes.
28
Cross-border linkages can also be hampered by potential political hurdles if
a link is perceived as a threat to local businesses and markets. Additionally,
the implementation of links can generate overhead and additional costs at
the partnering clearing houses. On the one hand, clearing links benefit the
partnering clearing houses by enabling them to provide clearing services for
a broader range of products and markets. At the same time, clear ing houses
might be worried that the implementation of a particular clear ing link could
limit their business development responsiveness by requiring all of the part-
nering clearing houses to develop clearing facilities for that product. When
clearing houses expect to either lose business to the partnering CCP or suspect
that they are essentially being locked out of the other clearing houses’ markets
(which could also stifle business development) through a link, they will be
reluctant to enter into such an agreement. Finally, one of the most important
complicating factors in implementing clearing link agreements is rooted in
the underlying motivation of the link set-up. Interviewees explained that, to
date, no clearing link initiative exists that was created to enhance the effi-
ciency of clearing and thus positively impact the efficiency of capital markets.
Developments have usually been driven by the exchange side rather than the
clearing side. Therefore, the value of links generally depends on the objective
being pursued by the initiator of the link.
27
Statement made by anonymous referee.
28
Cf. interviews.