The Gateway to Algorithmic and Automated Trading

Remarks of chairman Timothy Massad before the EU Parliament, Committee on Economics, Brussels

First Published 20th May 2015

Remarks of chairman Timothy G. Massad before the European Union Parliament, Committee on Economics, Brussels, Belgium.

Timothy Massad, CFTC

Good afternoon, Chairman Gualtieri and members of the Economic and Monetary Affairs Committee. Thank you for inviting me to participate in today's exchange of views. I very much appreciate the opportunity to meet with you. I have wanted to come to Brussels for some time. Frankly, shortly after I took this job, I had expected we would solve this issue regarding equivalence and recognition of CCPs relatively quickly, and I would be coming over to finalize an agreement many months ago. Of course, it has taken us longer. However, I think we and the Commission continue to work together in good faith toward that goal and hopefully we can reach agreement soon.

Before I begin, I also want to thank Lord Hill whom I will be meeting with tomorrow. I have appreciated our cooperation and collaboration on a wide range of issues, and look forward to that continuing. As he and I have discussed, there are many areas of derivatives regulation outside of CCPs where it's important to try to harmonize our rules as much as we can. I appreciate that these are global markets, and the only way we can effectively regulate them, and avoid regulatory arbitrage, is by working together. I would be happy to discuss any of these issues with you. But I thought that given the current focus on the CCP recognition issues, I would say a few words about how we view these issues. I would be happy to take your questions and discuss whatever issues you wish.

Last year, we spent a lot of time on the subject of whether the U.S. had an effective equivalent system of recognition of EU CCPs. The European Commission felt strongly that the CFTC needed to implement a framework for substituted compliance in addition to the existing recognition program that provides market participants in the U.S. with access to European futures markets. We worked hard to come up with that substituted compliance framework, and I believe that, if we can work through the rest of our differences, we have a framework that is satisfactory to both the EC and the CFTC.

I believe you know that lately we have been discussing our respective rules on margin methodology, and the focus has been on futures. And as you probably know, there has been much attention on the fact that, under CFTC regulations, our CCPs use a one-day minimum liquidation period and the new European regulatory regime requires a minimum two-day period. Now I will come back in a moment to the fact that the liquidation period is just one factor of several, and we believe it is critical to look at margin methodologies as a whole. Indeed one must look at risk mitigation practices as a whole. But I want to discuss this specific issue with you given the attention it has received.

There is no question that a two-day period produces more margin than a one-day period, all other things being equal. It is a simple mathematical relationship: two-day is higher by 41% - because one multiplies by the square root of two. But all other things are not equal. As you probably also know, we follow a policy of gross collection and posting of customer margin. That is, the clearing members must pass on to the CCP the full amount of initial margin for each customer. The new European regime allows for netting: if one customer's exposures offset another's, then the clearing member can post initial margin netted across customers. For some time we and the European Commission debated whether one-day gross was higher than two-day net or vice versa. Then in late February, when Lord Hill visited the U.S., we agreed to each do an analysis and share our results.

For our analysis, we used actual data for seven days. We reconstructed what the required margin would be under each regime for the nine largest clearing members of one U.S. CCP. These clearing members represent about 80% of the total customer margin. And what we found was that one-day gross was substantially higher than twoday net for each clearing member, and for each day. That is, the total amount of customer margin under one-day gross was as high as 421% of the amount under twoday net, and was never less than 160% of that amount. We have since looked at two other clearinghouses, and found even larger percentage differences.

I have prepared a few charts to illustrate the relative effects of netting and increasing the liquidation period. The first two charts do not contain actual numbers due to the need to protect confidentiality, but they are quite representative of the issues. Chart One illustrates the netting effect by comparing one-day gross to one-day net assuming the customers and portfolios are the same. The diagram at the top of the page shows that with gross posting, the clearinghouse receives $3000. The diagram at the bottom of the page shows the effect of netting. In our hypothetical the CCP receives $800. The extent of the difference will of course vary depending on the customers' portfolios. But the purpose here is to show the relative effects.

Chart two then does the same comparison, but this time using two-day net. The diagram at the bottom of the page shows that the total to the clearinghouse is $1120. This illustrates that the effect of the 41% increase for a two-day liquidation period doesn't offset the negative effect of netting. Indeed, one would have to have a very extreme distribution of customer positions for the two day effect to be greater than the netting effect. These hypothetical numbers include a clearing member with an extremely directional position (clearing member C). One-day gross is more than 2½ times higher than two-day net overall.

Chart Three is based on actual data. The chart illustrates actual customer initial margin for futures at one clearinghouse under both one-day gross and two-day net, involving positions held with the nine largest clearing members. This data was collected as part of our regular oversight and was used in the analysis performed by CFTC staff and provided to EC staff last month. These results are an average from a sampling period of seven different days over a six week period.

Chart Three shows that under one-day gross, the clearinghouse held $58 billion. Under two-day net, the clearinghouse would have held only $23 billion. The difference is $35 billion in initial margin held at the clearinghouse.

The results from our analysis of the nine largest clearing members based on one-day gross ranged from 2.4 times to 2.6 times the amount of two-day net , depending on the day.

Let me turn to house margin - the amount due from the clearing firm itself, which as you know is kept separate and cannot be commingled with customer margin. Two-day produced more than one-day. That is to be expected because if there is no netting effect, then two-day will be 41% higher.

But the size of house positions in futures markets is much smaller than the size of customer positions. Therefore, house margin is much lower than customer margin. Generally, house margin is around 14% of the total.

The first diagram in Chart Four shows actual house margin, again using real data. As you see, if we simply went to a two-day house requirement, the increase here is $4 billion dollars - $13 billion held under two-day compared to $9 billion held under oneday.

However, there is a difference in the treatment of house affiliates in the two regimes. We require that affiliate transactions be in the house account so that they don't jeopardize customers. Europe allows affiliates to be treated as customers.

The second diagram on Chart Four shows the difference between total house margin under U.S. rules today, and total house margin if the U.S. followed European rules - that is, excluding affiliates from the house account, but requiring margin for a two-day period. As you can see, the difference is very small - approximately $1.5 billion.

Chart Five summarizes the relative impact of our two regulatory regimes using the data we analyzed. The diagram on top shows the difference on the customer side - $35 billion. The difference in house margin, assuming we exclude affiliates, as is done in the EU, is only $1.5 billion.

The European Commission has nevertheless been concerned that the lower house margin is a risk. Doesn't this mean less margin is available in a default? We would agree that it does, but the far greater risk is the difference in customer margin. If a clearing member defaults, the question is how do we protect the customers of that member, the clearinghouse, and the other clearing members and their customers? The amount of total customer margin is what will largely drive the answers. It will determine whether the customers of the defaulted member can be transferred to another clearing member, or whether their positions must be liquidated, potentially destabilizing markets. It will determine whether the guaranty fund will be tapped. So if we wish to look at overall financial stability, there is far more to be gained by requiring gross posting for all customers than by increasing the minimum liquidation period for house accounts.

As for house margin, it has been suggested that U.S. CCPs could collect customer margin calculated using a one-day liquidation period, and could collect house margin calculated using a two-day liquidation period. The liquidation period for a product is typically based on its risk characteristics. We are concerned that charging different margin depending on who holds the product will create incentives to evade the regulation, and it would be especially problematic if Europe and the United States continue to treat house affiliates differently. In addition, while the amount of house margin is certainly relevant in a default, so is the house member's guaranty fund contribution, and the amount of one will affect the amount of the other.

Now I noted at the outset the importance of looking at margin methodologies, and indeed risk mitigation overall, holistically. There are many other issues one must consider to assess the overall outcome of a legal and supervisory framework. There are other differences in our respective frameworks. There are some areas where European rules are more conservative, and many where our rules are more conservative. I would be happy to discuss those with you.

But if our goal is to equalize key differences, and our concern is financial stability, we should start with customer margin - with requiring gross posting and mandatory collection from customers. Those are the big differences that will enhance financial stability and customer protection. As you may know, one-day gross posting is followed in other jurisdictions, including ones that have been granted equivalence already.

With respect to house accounts, we should reconcile differences in the treatment of affiliates. We should also look at guaranty fund contributions, and if we feel that a clearing member's own initial margin, together with its guaranty fund contribution, are not sufficient in the event of a default, we should look at ways of changing that. But we should avoid creating incentives for evasive behavior.

So with that let me stop and say again, thank you for inviting me to discuss these issues with you today. I welcome your thoughts and questions.

Click here for the full speech and charts