Large EU prop traders dump once-loved capital method

Industry lobbied for requirement based on margin haircuts, but it has provided little relief

Risk.net montage

Once lauded by European proprietary trading firms as a way to avert excessively onerous prudential rules, a method that uses the haircut charged by clearing members for determining regulatory capital requirements has fallen out of favour. More than two-thirds of Dutch and UK non-bank trading firms have snubbed the approach, Risk.net has learnt.

“The margin method was meant to be a boon for investment firms. They introduced this idea that you could set the regulatory capital at the margin that your prime brokerage was calculating for you,” says a regulatory expert at a non-bank market-maker. “We don’t use it because of the way [legislators] designed it – that was pretty much a failure.”

In June 2021, new European Union rules known as the Investment Firms Regulation came into effect, establishing capital requirements for non-banks authorised as investment firms under the second Markets in Financial Instruments Directive. The IFR provides proprietary trading firms with two ways to calculate their market risk capital requirements. 

The first, known as client margin given (K-CMG), is derived from the margin haircuts charged by investment firms’ clearing members. The second method, dubbed net position risk (K-NPR), is a slightly adapted version of the standardised approach to market risk devised by the Basel Committee on Banking Supervision. Prop traders had originally lobbied for K-CMG due to its close resemblance to the minimum margin requirements their clearing brokers charge them, so the compliance build-out would be much simpler and any change in capital requirements should be manageable.

© De Nederlandsche Bank N.V.

De Nederlandsche Bank regulates the largest group of European prop traders

“Listening to the market on how stuff actually works and sticking to a metric that everybody has to monitor almost real-time anyway – it made sense,” says a senior compliance expert at a second non-bank market-maker. “Then it became political.”

Changes to K-CMG during the negotiating process between EU lawmakers effectively undermined its appeal for prop traders. A UK version of IFR came into effect in January 2022. Since the IFR was drafted by EU legislators before Brexit, the UK government was part of negotiations on the regulation. Although the UK added the word ‘prudential’ to the name of their version (IFPR), its contents are largely the same.

A spokesperson for the Dutch central bank – the prudential supervisor for the European nation which is home to the largest number of prop traders – says “approximately” 10 of a total of 35 principal trading firms under their purview are using the K-CMG method. In the UK, only four investment firms have been granted permission to use the K-CMG, according to the Financial Conduct Authority’s website. Risk.net has identified 14 investment firms within the FCA’s register of authorised firms that describe themselves as proprietary trading outfits. 

Hell and high watermark

The chief executive of the second non-bank market-maker says they find capital requirements under the K-CMG to be “a bit” higher than those set by the K-NPR. The K-CMG method sets an investment firm’s capital requirement at 1.3 times the third highest daily margin figure – known as the high watermark – charged by their clearing member over the previous quarter. The calibration was a political compromise, following a push by French policy-makers to ensure investment firms’ capital requirements would be the highest of the K-CMG or K-NPR. Instead, in the final package, firms could choose freely between K-NPR and a toughened version of the K-CMG.

The high watermark compounds the existing add-ons built into the margin charged by clearing members to their clients. The European Market Infrastructure Regulation requires central counterparties (CCPs) to collect margin that would be sufficient to cover losses from at least 99% of price movements over “an appropriate time horizon”. CCPs also impose add-ons for risks including high concentration and illiquidity. 

Related

Special K-factors: new buy-side capital regime sized to fit 

EC official offers hope to prop traders on capital rules 

Unlucky for some: Europe’s war on 13 Dutch prop traders 

Clearing members perform their own analysis of how much margin to charge clients to ensure they meet the CCPs’ requirements. These are broadly similar to the margin models of CCPs, but depending on how volatile markets are, banks may also apply a multiplier of their own on top of the clearing house margin. 

The senior compliance expert at the second non-bank market-maker says taken together, the IFR multiplier and the high watermark have an “emphasising” effect on capital requirements, especially as the clearing member margin requirements are already meant to capture tail risk. 

“Those elements combined make the risk assessment completely out of whack,” says the senior compliance expert.

A regulatory expert at a third non-bank market-maker says the high watermark specifically is what puts them off using the K-CMG. Although the 1.3 multiplier is “arbitrary”, they understand why regulators want an element of conservativism, whereas the high watermark means the firms’ capital requirements become unpredictable. Since the K-CMG is calculated quarterly with a one-quarter lookback period, a brief spike in CCP margin then gets baked into the firm’s capital requirements for months at a time. 

“Your capital requirements can never go below that level even if volatility sharply drops off tomorrow and your actual K-CMG is lower,” says the regulatory expert at the third non-bank market-maker. “You will still be tied to that high watermark for as long as the rolling period works, which means a firm has absolutely no flexibility over managing their capital requirements in volatile versus non-volatile periods.”

There may be market developments or risk appetite reasons out of our control that make relying on a clearing model unpredictable

Senior compliance expert

The senior compliance expert at the second non-bank market-maker says a waiting period of two years before firms can switch between the methods also puts their firm off using the K-CMG. Regulatory technical standards developed by the European Banking Authority established the cooling-off period to prevent the risk of arbitrage between the two methods. However, since K-CMG is effectively set by the margin charged by clearing banks, K-NPR provides principal trading firms with a more stable capital requirement to plan around. 

“The clearing banks can of course change the margin model at their whim,” says the senior compliance expert. “There may be market developments or risk appetite reasons out of our control that make relying on a clearing model unpredictable.”

Blunt instrument

The shift away from K-CMG doesn’t mean prop traders feel any great love for K-NPR. Sources complain it is not risk-sensitive enough for their activities, because it provides far less capital relief for netting between offsetting exposures compared with K-CMG.

Basel standardised approaches typically only allow netting if the offsetting exposures fall within the same defined exposure buckets. The result is that cheaper hedges using closely correlated but not identical instruments tend not to be recognised as offsets within capital requirements. The rules also explicitly forbid netting between similar exposures denominated in different currencies. 

Sources did not disclose how much larger capital requirements were under the K-CMG than the K-NPR. They suggest K-CMG may still be appropriate for smaller firms, because it is simpler to operate than the Basel methodology – effectively, K-CMG outsources part of the calculation work to a prop trader’s clearing bank.

According to the FCA’s register of permissions given for firms to use the K-CMG, two entities belonging to Mako Trading have adopted this method, as well as Tibra Trading and XConnect Market Maker. None of these companies responded to Risk.net in time for the publication of this article.

Editing by Philip Alexander