Reading Philip Stafford’s recent article in the Financial Times, Deutsche Borse makes ground in UK derivatives push, I was struck by the paragraph:
Seven times! Impressive, indeed.
Is this the start of a ratcheting up in Eurex volumes and gain in market share vs LCH SwapClear?
Lets look at what the data shows in detail.
EUR IRS Volumes
January 2018 with €60 billion gross notional, higher than any other month except for March, June and September (IMM Future roll months) and significantly higher than the €33 billion in January 2017.
However that is an increase of 82% year-on-year, a very good increase, but certainly not 600% (the seven times in the FT article). What explains the difference?
IRD not just IRS
Expanding our selection to include other EUR IRD products.
We see a huge spike in EUR FRAs in Jan 2018, with €325 billion cleared in the month versus €5 billion in the preceding 12 months. This explains the seven times surge in volume. (Of note, but of lesser significance in the surge is an increase in EUR OIS to €20 billion from €4 billion in Jan 2017).
Our total volume in Jan 2018 is €405.5 billion of single-sided notional.
Many clearing houses use what we term as double-sided notional, as a trade executed between two parties is novated in clearing into two trades, each between one party and the clearing house.
So doubling the €405.5 billion and dividing by 23 business days, gives the ADV of €35 billion stated by Eurex.
Mystery solved and the statement of an ADV of €35b and seven times increase is factually correct.
However like many marketing statistics, there is devil in the detail, which may or may not be relevant depending on the question we are seeking to answer.
EUR FRA Volumes
Drilling down into the €325 billion of EUR FRA gross notional, we see that this volume results from €118b on 16th Jan, €118b on 23rd Jan and €89b on 30th Jan.
The fact that these are consecutive Tuesdays and such large size, suggests an automated FRA program such as ones run by NEX Reset or TP-ICAP tpMATCH, to manage the EURIBOR rate reset risk that arises in IRS portfolios.
And given that outstanding notional in EUR IRS at Eurex has reached €880 billion from €387 billion a year earlier, there is certainly a need to manage the rate reset risk in these swaps.
While I cannot find any news item on the Eurex, NEX or TP-ICAP sites that mention this, I can only assume it is this activity causing a surge in FRA volumes and it is definitely a good sign that this important rate reset risk is now happening in large size at Eurex.
However including these FRA volumes in the January 2018 figures and comparing with 2017 monthly volumes, which had tiny FRA activity, does not help at all in answering the question I posed myself in this blog, namely ‘is Eurex gaining share in IRS clearing?”.
To do that we need more data.
Overall EUR IRS Volumes
We know from above that Eurex Jan 2018 volume was up 82% year-on-year, increasing from €33 billion to €60 billion gross notional.
However we need to establish whether Jan 2018 was a higher volume month than Jan 2017, due to more market volatility or other external factors causing higher volumes and if so, did Eurex capture a higher share of this volume than LCH SwapClear, the dominant market leader?
Lets chart monthly volumes for both.
This lays bare the mountain to climb for Eurex in challenging LCH SwapClear.
LCH SwapClear Jan 2018 volume was €4.5 trillion, which is up 34% from the €3.36 trillion in Jan 2017.
So Eurex has indeed grown more with 82% growth vs 34% growth at LCH SwapClear.
However Eurex’s market share was only 1% in Jan 2017 and 1.3% in Jan 2018.
And this is the change in a single month, year-on-year (Jan 17 to Jan 18), so subject to sampling error.
Digging deeper we see that Eurex’s monthly share in 2017 varies from a low of 1% to a high of 1.8%, with an average of 1.4%. Against this the 1.3% share in Jan 2018 does not stand out at all.
Meaning that the data does not support the view that Jan 2018 volumes show Eurex gaining market share in EUR IRS Clearing.
Average Daily Volume is a metric much quoted for Futures and one that we are keen to see used for Swaps, where there is sufficient data to calculate.
More useful values for ADV than the €35 billion quoted above are:
- ADV Jan 2018 EUR IRS at Eurex of €2.7 billion (single-sided)
- ADV Jan 2018 EUR OIS at Eurex of €900 million
While for EUR FRA it is meaningless to calculate ADV as there is volume only on 5 days in the month.
Before we end, it is worth taking a look at LCH-EUX CCP Basis Spreads.
Showing that 5Y is 0.55 bps, 10Y is 1.40 bps and 30Y 3.55 bps.
So these have narrowed from 28 April 2017, when 5Y was 1 bps and 10Y was 2 bps, though 30Y has widened from 3.25 bps. Narrower spreads will help level the choice of whether to clear at Eurex or not, but these spreads are much wider than market bid-offers.
They are however tighter than the USD LCH-CME Basis, which on 31-Jan-2018 was 1.35 bps, 2.85 bps and 3.85bps for 5Y, 10Y, 30Y and CME has 8.4% share in 2017 in USD Swaps (see 2017 CCP Market Share Statistics). So higher share is certainly possible with current levels of the LCH-EUX Basis.
That is not to say that there are not other factors that may drive an increase in Eurex share.
- the profit incentive scheme that Eurex have introduced for dealers,
- portfolio margining with Bund Futures,
- connectivity with MTFs and market structure,
- regulatory incentive or requirements
- … I could go on.
But so far the data does not show an increase in Eurex market share.
We will watch out for change in the lead up to March 2019.
For an earlier heads-up, can you afford not to try CCPView?
Remember the devil is in the detail.
Or as Mark Twain attributed to Benjamin Disraeli:
“There are three kinds of lies; lies, damned lies, and statistics”.