Article: Targeted clearing mandate exemptions could reduce systemic counterparty risk31 March 2015 | Jon Skinner
On systemic counterparty risk reduction, recent public discourse and lobbying has focused on adjusting minimum bilateral margin regulations and Basel III minimum capital rules to more reasonably reflect portfolio risk. Concerns have even been raised that central banks pushing for more bank capital is dangerously experimental in nature (e.g. this Risk article following on from a piece by ex-deputy chairman of the Bank of England Paul Tucker - subs. required for articles linked). However, regulators are hinting that Basel III is largely done and dusted according to this FT article (subs. required). Only a slight delay seems likely to the start of the roll out of the bilateral margin rules despite market participant protestations about the workload to be ready. Once the rules are risk sensitive, banks and market utilities will respond to the incentives and push forward with innovative business models, trading methods and post trade capital optimization techniques.
Meanwhile on clearing we do hear debate about improvements which may be needed to CCP default resources and default management processes and what happens at the end of the waterfall. But the mandate itself seems to be unquestioned. You might therefore infer that the mandate is successfully doing its work of reducing systemic counterparty risk through consolidation of trades in CCP portfolios and the associated risk netting and risk mitigation by margin. You might infer that there is no systemic counterparty risk downside and that ESMAs mandate should roll on as more of a good thing.
Exacerbating cleared risk and entrenching bilateral risk
I disagree. I believe the clearing mandate overuses clearing which both exacerbates residual cleared counterparty risk and significantly entrenches systemic counterparty exposures in bilateral portfolios, thus offsetting the benefit of incremental new trade clearing.
Yes, the clearing mandate does reduce new trade OTC counterparty risk looked at in isolation. As implemented so far in the US the mandate has ensured that new trades in mandated products between mandated parties are cleared. Often this will reduce counterparty risk through exposure netting and mitigation of most of the net risk through variation and initial margin. In this way systemic counterparty risk can be reduced but this must be put in context:
1. Secondary / opaque clearing risks are a concern - implying overuse should be avoided
There has been a lot of press about the risks inherent in the clearing model. What happens when the default fund is exhausted? Are levels of default fund contributions and CCP skin in the game large enough? Will there be too much FCM concentration? Will portability work in practice or will clients be closed out along with a member? It's hard to size these risks even conceptually but their existence argues that clearing shouldn't be artificially over-used.
2. The clearing mandate entrenches bilateral risk
In general, executing trades that mutually reduce bilateral portfolio risk will reduce counterparty risk, capital usage and IM funding costs and systemic counterparty risk in the process. The cheapest and easiset way to execute such trades is in the most liquid products. So mandating liquid products to clear effectively disallows this risk reduction approach.
Trading mutually hedging packages allows both counterparty and market risk to be limited. For example, executing a swaption delta hedged with a swap as a package provides swaption vol exposure without delta exposure (to both parties). If the swap hedge is mandated to clear, though the market risk delta hedge still works, the bilateral counterparty risk on the swaption is higher than would otherwise be the case if the swap were allowed to stay bilateral. So mandating the swap to clear degrades the counterparty risk efficiency and increases the costs of trading such packages.
The risk entrenchment problem is bigger than bilateral notional outstanding figures suggest. The notionals show that most notional is already cleared and therefore we could infer that most of the counterparty risk is in cleared portfolios. Its worth noting though that bilateral risk is much more fragmented / less netted and the CCP IM incentive in fact incentivizes buy side firms (especially those not subject to Basel III or bilateral IM) to flatten cleared portfolio risk while keeping most of the net trading risk (if not subject to IM) in bilateral portfolios. So bilateral portfolios are much more risky "per notional" and most of the counterparty portfolio risk is likely still in bilateral portfolios.
I conclude that there is a significant possibility that entrenchment of bilateral risk - by slowing down risk reduction - may be canceling out or outweighing the risk reduction gains on new trade clearing mandate. Surely this problem could be fixed?
The clearing mandate in context of bilateral margin mandate
Whilst D2D portfolios may dominate notional, D2C portfolios dominate aggregate portfolio risk. Therefore limiting clearing to D2D would leave the majority of counterparty risk unaffected. So D2C clearing is targeting the majority of the systemic counterparty risk and is therefore an important initiative in its own right. This is an argument for encouraging clearing of D2C portfolios.
For commercial reasons D2C counterparty risk is only partially mitigated by elective IM even under Basel III. Whilst Basel III incentivizes banks to receive IM on these portfolios this may not always be commercially viable. Hedge funds usually pay IM on their portfolios. However, in the current world these may be more favorable than cleared IM e.g. 1-day VaR is not unusual. Other buy side segments may put up IM but larger traditional asset managers or GSEs may have used the leverage associated with their large dealing volumes to negotiate away the need to put up initial margin. Not being subject to Basel III buy side firms have much less incentive either to clear or receive bilateral margin.
By instigating a bilateral initial margin mandate calibrated at a minimum of 10-day VaR, regulators are restricting the leeway of banks to incur systemic risk on large clients as a commercial quid pro quo. Once in place the bilateral IM mandate will act as a strong funding incentive for the buy side to trade using cleared products. Looking at relative timing we have the US clearing mandate in place with EU to follow in 2016 and later. Meanwhile the bilateral IM mandate goes live with G-SIFI banks and a small number of the largest buy side firms from September 2016 and rolling out to smaller financial firms in annual waves after that.
At this point a hypothetical question is interesting: how much clearing would there have been after the bilateral initial margin mandate if there was no clearing mandate? The answer in notional terms is a little less. However, that may be no bad thing for the following reasons.
A trade which offsets CCP portfolios but increases risk if left bilateral would be incentivized to clear. A trade which doesn't offset either cleared or bilateral portfolios would also clear given 5-day VaR IM levels and cheaper capital costs for banks. The only trade incentivized to stay outside clearing would be one which offsets the bilateral portfolio risk and therefore reduces the more expensive 10-day VaR bilateral IM or more expensive bilateral capital costs for banks. Such an outcome would be a win for everyone. Counterparties get reduced risk, funding and capital costs. Regulators get lower systemic counterparty risk. So the bilateral IM mandate would encourage clearing just when it actually reduces aggregate counterparty risk compared with bilateral.
Comparing this hypothetical scenario with where we are actually heading indicates that, once the bilateral margin mandate is in place, the clearing mandate prevents risk reduction trades like the one described above if the trade is mandated to clear. Both cleared and bilateral counterparty risk will be higher as a result.
So why in this context would there be a need for a clearing mandate? There is an argument that the mandate has moved financial firms outside the major dealer banks to a position where they are able to clear trades faster than incentives might have got them there. That may be true but in any case it is of little value to talk about what might have been. More importanly the hypothetical hints at a solution for the problems of exacerbation of clearing risk and entrenchment of bilateral risk.
Can we fix the clearing mandate?
A. Abolish it? No
There is a certain theoretical elegance to the idea of abolishing the clearing mandate altogether. Instead regulators could rely solely on the bilateral initial margin as an incentive to clear. This should engender clearing just when it made systemic risk sense to do so, and not otherwise. Enabling buy side clearing would arguably still have required a considerable amount of clearing regulation aside from the mandate e.g. LSOC, open access, etc. Such a grand approach might involve pushing "a very large rock up a very large hill" with regulators considering the current regulatory consensus for clearing mandates. A lot of re-analysis would be required considering the incentives in various scenarios and various categories of counterparties and factoring in relative timing of the onset of the EU clearing mandate and the global bilateral margin mandate.
Given the bilateral IM mandate is upon us, I suggest we discount this approach.
B. Exempt bilaterally margin mandated trades from clearing? Yes
It would be a lot simpler to build consensus for this approach and to implement it. If a trade would be subject to the bilateral margin mandate, i.e. both parties are subject to the bilateral margin mandate, then it is exempted from clearing.
Now the two parties can trade liquid products against their bilateral margin mandated portfolios and if they are smart will do so only when this reduces bilateral IM and bilateral capital charges for banks. If they trade to increase the risk of that portfolio they will pay increased bilateral IM and capital charges. They are unlikely to do the latter for any good reason.
If evidence is required for exempted trades, firms need only show that both they and the counterparty are subject to the bilateral IM mandate. I imagine this will be needed anyway for the bilateral margin mandate itself and therefore I am guessing the compliance burden for this approach will be light.
Some may have concerns that this might impact the SEF mandate. Actually this is a red herring as it is perfectly possible to SEF trade uncleared trades already. SEFs typically route the trade to an electronic affirmation vendor which automates the confirmation and confirmation matching of the trade. This is efficient. Perhaps some buy side firms might need to connect to one or two more electronic affirmation services to enable this.
Conclusion: Exemption B enables risk reduction trading using clearable products against bilateral portfolios where the two parties are subject to the bilateral margin mandate. The exemption has the advantage that it can relatively easy be evidenced for compliance purposes by showing both parties are subject to the bilateral margin mandate.
C. Exempt mutually legacy risk reducing trades from both clearing and bilateral margin? Yes
Exemption B seems simple. However, it has a disadvantage.
First let's outline the portfolio structure after the bilateral margin mandate goes live. Presuming ISDA's proposed netting set and CSA structure is adopted, bilateral portfolios will be split into a CSA for bilateral margin mandated trades and a CSA for bilateral legacy portfolios. Both portfolios sit together in a single netting set. Initial margin is calculated separately per CSA (and the legacy CSA may even have no IM) but risk and capital are calculated across the netting set. This means that the bilateral IM in total offsets the net portfolio risk across the portfolio.
In this context there could be many cases where bilateral counterparties largely trade cleared product going forward. Suppose now they use exemption B to trade mutually risk reducing bilateral trades. They would reduce bilateral risk and capital charges but would incur bilateral IM at 10-day VaR. This is an unnecessary disincentive to do this trade as everyone is better off if they do that trade due to reduced risk, funding, capital and systemic counterparty risk. In particular the buy side firm may prefer to clear the trade than pay that IM.
This could be fixed if an exemption was also available from the bilateral margin mandate for trades reducing risk on pre-bilateral margin mandate legacy portfolios.
To make this work counterparties would need to stipulate at the point of trade which exemption they were taking and would actually put the trade in the legacy CSA rather than the bilateral IM CSA. Once attached to the right portfolio calculation of initial margin would work as before.
The question though is how to evidence the exemption i.e. show that the trade reduces bilateral legacy portfolio risk. Two challenges need to be overcome:
- Risk metric calculations are complex. They are often based on proprietary analytics whose results differ between trading participants. Risk metrics may be difficult to agree between parties.
- A possible solution is to use the agreed bilateral IM model as a risk metric (though IM would not be put up on the legacy portfolio). Whilst not a perfect risk metric it has the advantage of being available (at least once both parties are ready for the bilateral IM mandate).
- Legacy portfolios are a moving target. If the two parties have not yet become subject to the bilateral margin mandate the legacy portfolio continues to grow as non-clearable trades are added. At worst a trade done in the middle of the day may be risk reducing at the time but as the portfolio moves may look risk increasing at the end of day.
- A possible solution is to limit the exemption to parties subject to the bilateral IM mandate. Now the legacy portfolio is a fixed set of trades executed prior to the mandate commencement date plus any trades under exemption C. Demonstrating that trades reduce the legacy portfolio risk is a more straightforward comparison of legacy portfolio hypothecal IM before / after the exempted trades.
Conclusion: Exemption C complements exemption B by appropriately incentivizing legacy portfolio risk reduction. Practically evidencing the exemption may be easiest if both parties are subject to the bilateral margin mandate and use the bilateral IM calculation as a risk metric (not to calculate margin) for the legacy portfolio and the legacy portfolio is a fixed set of trades.
Can regulatory appetite be mustered?
If these ideas are not considered, counterparty portfolio risk - in bilateral legacy, bilaterally margin mandated and cleared portfolios - will be materially higher than otherwise might be.
No doubt regulators will want numerical evidence which I hope banks can provide on an ad-hoc basis. I fear that similar ideas may have already been articulated to and rejected by regulators.
Given the likely scale of otherwise entrenched bilateral legacy risk, I suggest the above two exemptions be considered, even if this reverses the outcomes of prior discussion.
Bank regulators' minimum capital, liquidity reserve and margin levels are undoubtedly a major weapon in incentivizing counterparty risk reduction.
Participants need to be able to reduce counterparty risk by all rational means in response.
Unintentionally the clearing mandate inhibits the reduction of bilateral risk and this effect may at worst negate or outweigh the counterparty risk reduction benefits of clearing new trades.
Targeted exemptions described above offer the prospect of eliminating these unintended consequences and there is hope of a lighter compliance burden through reuse of bilateral IM compliance tools.
I suggest the CFTC is well placed to take a global lead in addressing this important issue.