Networks of counterparties in the centrally cleared EU-wide interest rate derivatives market 1

Transcription

1 Networks of counterparties in the centrally cleared EU-wide interest rate derivatives market 1 Pawe l Fiedor Sarah Lapschies Lucia Országhová,, European Systemic Risk Board Secretariat Národná banka Slovenska University of Economics in Bratislava 1 The views expressed are not necessarily the views of the ESRB, or the ESCB.

2 Overview of the presentation Dataset Results Conclusions

3 What data do we use?

4 Dataset All open centrally cleared interest rate derivative contracts in the EU:

5 Dataset All open centrally cleared interest rate derivative contracts in the EU:

6 Dataset All open centrally cleared interest rate derivative contracts in the EU:

7 Dataset All open centrally cleared interest rate derivative contracts in the EU:

8 Dataset All open centrally cleared interest rate derivative contracts in the EU: From all 6 trade repositories

9 Dataset All open centrally cleared interest rate derivative contracts in the EU: From all 6 trade repositories Including both house & client clearing

10 Dataset All open centrally cleared interest rate derivative contracts in the EU: From all 6 trade repositories Including both house & client clearing Trade Repository # of contracts UnaVista 2,352,891 DTCC OTC 1,541,691 DTCC ETD 149,476 Regis-TR 102,349 KDPW 7,588 CME 3,076 ICE 268 Total 4,157,339

11 Dataset All open centrally cleared interest rate derivative contracts in the EU: From all 6 trade repositories Including both house & client clearing Trade Repository # of contracts UnaVista 2,352,891 DTCC OTC 1,541,691 DTCC ETD 149,476 Regis-TR 102,349 KDPW 7,588 CME 3,076 ICE 268 Total 4,157,339 Density e+05 1e+08 1e+11 Notional amount (EUR)

12 What kind of companies are in the network?

13 Institutions We have a variety of institutions present in the centrally cleared IRD market:

14 Institutions We have a variety of institutions present in the centrally cleared IRD market: Type of institution Full House Client CCP G Bank Other financial 5, ,246 Insurance & pension Non-financial Total 6, ,354

15 Institutions We have a variety of institutions present in the centrally cleared IRD market: Type of institution Full House Client CCP G Bank Other financial 5, ,246 Insurance & pension Non-financial Total 6, ,354

16 Institutions We have a variety of institutions present in the centrally cleared IRD market: Type of institution Full House Client CCP G Bank Other financial 5, ,246 Insurance & pension Non-financial Total 6, ,354

17 Institutions We have a variety of institutions present in the centrally cleared IRD market: Type of institution Full House Client CCP G Bank Other financial 5, ,246 Insurance & pension Non-financial Total 6, ,354

18 What does the network look like?

19 The network structure Around 90% of the counterparty relations are on the client clearing side.

20 The network structure The full network has three tiers: inner core, outer core, and periphery.

21 The network structure House clearing network (on the left) has a different structure: only two tiers.

22 The network structure The inner core, of mostly large dealers, creates the backbone of the network.

23 Is the network stable over time?

24 Persistence across time The network is relatively stable between Q3 & Q4 of 2016, including the counterparty relations.

25 AUD BRL CAD CHF CLP CZK DKK EUR GBP HKD HUF JPY MXN NOK NZD PLN SEK SGD USD ZAR common only in one period common only in one period Persistence across time The network is relatively stable between Q3 & Q4 of 2016, including the counterparty relations. % of edges CCP CM CM Client FULL CCP CM CM Client FULL Currency % of edges Maturity year

26 What can we find in the networks?

27 Average number of counterparty relations by type Difference between the full network and the house cleaning network by maturity BANK CCP G16 INSURANCE & PENSION NON FINANCIAL OTHER FINANCIAL Degree Maturity year BANK CCP G16 INSURANCE & PENSION NON FINANCIAL OTHER FINANCIAL Degree Maturity year

28 Average number of counterparty relations by type Difference between the full network and the house cleaning network by currency Degree AUD BRL CAD CHF CLP CZK DKK EUR GBP HKD HUF JPY KRW MXN MYR NOK NZD PLN SEK SGD USD ZAR Currency TYPE BANK CCP G16 INSURANCE & PENSION NON FINANCIAL OTHER FINANCIAL 60 Degree AUD BRL CAD CHF CLP CZK DKK EUR GBP HKD HUF JPY MXN MYR NOK NZD PLN SEK SGD USD ZAR Currency TYPE BANK CCP G16 INSURANCE & PENSION NON FINANCIAL OTHER FINANCIAL

29 Are the networks stable if we remove counterparties?

30 Random removal of counterparties by type We remove a number of institution at random and see how many additional ones are disconnected with the main component.

31 Random removal of counterparties by type For the full network G16 dealers have the most effect, followed by banks, and only then CCPs.

32 Random removal of counterparties by type For the house clearing network this information is lost, while it s useful for financial stability, e.g. question of the suspension of clearing obligation.

33 0 0 0 Removal of the largest clearing members We remove a number of the largest clearing members and see how many additional ones are disconnected with the main component. BANK INSURANCE & PENSION NON FINANCIAL OTHER FINANCIAL % of institutions losing access to the main market hub Number of removed clearing members

34 Access to liquidity We define access to liquidity of an institution as the total number of counterparty relations of its direct neighbours. Type of institution Total access to liquidity Average access to liquidity G16 16,899 1, Non-financial 9, Bank 15, Insurance & pension 9, Other financial 17,

35 Access to liquidity We define access to liquidity of an institution as the total number of counterparty relations of its direct neighbours. Type of institution Total access to liquidity Average access to liquidity G16 16,899 1, Non-financial 9, Bank 15, Insurance & pension 9, Other financial 17,

36 Access to liquidity We define access to liquidity of an institution as the total number of counterparty relations of its direct neighbours. Type of institution Total access to liquidity Average access to liquidity G16 16,899 1, Non-financial 9, Bank 15, Insurance & pension 9, Other financial 17,

37 Access to liquidity We define access to liquidity of an institution as the total number of counterparty relations of its direct neighbours. Type of institution Total access to liquidity Average access to liquidity G16 16,899 1, Non-financial 9, Bank 15, Insurance & pension 9, Other financial 17, When we remove clearing members at random the loss of access to liquidity is proportional to the starting access to liquidity, irrespective of the type of the institution.

38 1 1 Access to liquidity 1 1 However, it is still useful to see what the impact of the removal of the largest clearing members is on the access to liquidity of various types of institutions. BANK G16 INSURANCE & PENSION NON FINANCIAL OTHER FINANCIAL % loss in access to liquidity Number of removed clearing members

39 Conclusions Main findings:

40 Conclusions Main findings: Client clearing is important and changes the perception of the centrally cleared interest rate derivatives market dramatically:

41 Conclusions Main findings: Client clearing is important and changes the perception of the centrally cleared interest rate derivatives market dramatically: In particular, the insurance sector is only visible via client clearing;

42 Conclusions Main findings: Client clearing is important and changes the perception of the centrally cleared interest rate derivatives market dramatically: In particular, the insurance sector is only visible via client clearing; Importance of various types of institutions changes dramatically;

43 Conclusions Main findings: Client clearing is important and changes the perception of the centrally cleared interest rate derivatives market dramatically: In particular, the insurance sector is only visible via client clearing; Importance of various types of institutions changes dramatically; CCPs are not systemic in isolation, and should be analysed in conjunction with G16 dealers also for policy issues (e.g. R&R).

44 Thank you