Clarus Financial Technology

New: Kids on the Block (Sizes)

The new block sizes in SDR data make trends in the data more difficult to identify – because now only 25% of notional should be “dark” (i.e. above the capped thresholds), down from 50%. More transparency is good, but we need to be careful when looking at data from before and after the new block thresholds were implemented.

The briefest possible summary of the changes comes from my previous blog, stating;

With a couple months worth of data now, I’ll have a look at some of the changes.

Number of Block Trades

First up, a really nice chart showing the impact of regulatory changes!

Number of capped trades reported each month. Source: SDRView

Showing;

Notional Amounts

A health warning should be attached to the following chart. Look at the notional amounts reported each month:

Notional Amounts of OIS in $Bns. Source: SDRView

So October was a record month, right? WRONG! If this blog saves just one intern (or senior manager!) from making that mistake, it will be worth the 700 words. The calibration of the block thresholds means that 25% of total notional should now be above the capped thresholds, where-as previously it was 50%. Therefore if we scale up the pre-October amounts by 100%, and the post-October amounts by 33% we get a more accurate picture of total volumes:

September and January were most likely the largest notional months in 2024 (and by extension, the largest months we have seen reported to SDRs). November volumes were really nothing to write home about.

Nasty Data

Here’s a lesson kids – not all data can be displayed nicely in a chart! When trying to make sense of the new thresholds, I thought it would be good to look at the proportion of trades in each currency that is reported as a capped trade. Here’s a cut-out and keep data table:

USDGBPJPYEUR
% of Capped Trades before October8%5%5%16%
% of Capped Trades since October2%4%6%5%

Showing;

Unfortunately, the data only passes this sniff test for USD. The other currencies really don’t tie in with the increases (and decreases) that we have seen in block sizes. Weird.

The Joys of Data

None of this is to say that the new block sizes are calibrated “incorrectly”, or even that the old block sizes were the right ones!

We are dealing with data across multiple dimensions:

I won’t bore you with the details, but even a “model” based on volume weighted trade counts per maturity is awful at making sense of how many capped trades we expect to see as a result of the new thresholds.

CCPView

What is good (for us at Clarus at least!), is that these changes only impact SDR data. And that is not the only source of data that we have. We also have CCPView.

This gives us an opportunity to benchmark just how much of the USD Swaps market is reported to SDRs. Assuming that CCPView represents 100% of the total USD OIS market:

Notional amounts in USD swaps as seen in SDR data and CCPView Data. $Bns. Sources: SDRView and CCPView

It looks like 80% of the global cleared USD swaps market was reported to SDRs in November 2024. Pretty impressive transparency.

In Summary

Stay informed with our FREE newsletter, subscribe here.

Exit mobile version