November 4, 2015
As Prepared for Delivery
Good Morning. And thank you, Walt for that introduction. I’m very pleased to be here.
I’d like to thank the Futures Industry Association for inviting me to speak at this 31st Futures and Options Expo. For over 50 years, FIA has been a key voice for the derivatives industry, and I applaud them for putting together yet another impressive event.
There’s a lot going on at the Commission today. And I’d like to discuss a few of the items that are on our agenda for the next few months. But then I’d like to talk about something I haven’t discussed much lately. And that’s the issue of reporting of data on swaps. This is a critical component of the overall reform of the swaps market, and I believe today is a good opportunity to discuss this in some detail, for reasons I’ll explain in a moment.
So, let me first note a few issues where I expect us to act in the next few months.
Margin for Uncleared Swaps. When I spoke here last year, as well as at the FIA Boca gathering, I discussed my commitment to working with the prudential regulators, as well as with our international colleagues in Europe and Japan, to harmonize our rules on margin for uncleared swaps. We have been working very hard to achieve that goal, and I am pleased to say that much progress has been made. As you may know, the prudential regulators recently finalized their versions of the margin rules. On many issues, there is far greater similarity with the current European and Japanese proposals – than compared to a year ago.
I expect that the CFTC will act before the end of the year. And I support making sure that that these rules are similar on issues like the material swaps exposure threshold – which triggers margin requirements; the timetable of implementation; the acceptable types of collateral, including for variation margin; the currency of payment; and other matters. This reflects the work of our respective staffs to achieve greater similarity internationally on core issues. There will still be differences, of course. And even domestically, there may be differences – because there are differences between us and the prudential regulators in mission and regulatory framework. But we remain committed to harmonizing with them, as well as with our international colleagues, as much as possible.
Automated Trading. Recently, I noted that we will soon consider some proposals related to automated trading. These would be designed primarily to address the potential for market disruptions, not market structure issues, and they would be principles-based.
For example, we are considering requiring further pre-trade controls, such as message throttles and maximum order size limits. But we will not prescribe how those limits should be set. We may also propose requirements pertaining to the design, testing and supervision of automated trading systems. We’re considering measures such as “kill switches,” which facilitate emergency intervention in the case of malfunctioning algorithms.
We are considering requirements at the exchange level as well as at the clearing member and trading firm levels. We are considering whether to require proprietary traders who access the market directly and who are using automated trading – to register with the CFTC. This would ensure that all those with so-called “direct electronic access” to our markets are complying with pre-trade risk controls, testing and other requirements.
I anticipate that our automated trading proposals will be largely consistent with the best practices followed by many firms already. The input of the FIA has been helpful in that regard. And as always, we look forward to your feedback on this proposal.
Swaps Trading. In the area of swaps trading, our staff is working hard to make swap execution facility registrations permanent. Twenty-two platforms are temporarily registered. For the platforms that have provided complete information to us, we expect to make determinations of permanent registrations by early 2016.
And let me just note some of the other swaps trading issues are working on. Over the last 18 months, Commission staff has provided flexibility in several areas of our rules through “no-action” letters and guidance. Next year, I expect us to formalize some of these staff actions through rulemaking proposals. We will also consider some additional issues, such as the “made available for trade” determination process.
And now that ESMA has published the MiFiD II technical standards, we are also working to understand differences in our respective rules on swaps trading. This is an important first step as we look at ways to harmonize and achieve mutual recognition.
Cybersecurity. Another area where I expect us to take action soon is cybersecurity. We are very focused on the risk of cyberattacks, particularly by those who may be looking to intentionally disrupt our markets.
The Commission recently approved the National Futures Association’s cybersecurity guidance that will require members to adopt – and enforce –policies and procedures to secure customer data and protect their electronic systems.
We are also considering some proposals to ensure that the major exchanges, clearinghouses, and swap data repositories are doing adequate evaluation and testing of their own cybersecurity and operational risk protections.
As with automated trading, I expect these will be principles-based standards.
These are just a few of the things on our plate. There is work going on in many other areas, of course, including position limits and clearinghouse equivalence and recognition. But now I want to turn to the subject of swap data reporting.
I want to focus on this subject for a couple reasons. First, it is fundamentally important to the overall work of creating a sensible framework for regulating swaps. And second, there have been questions about the quality and usefulness of the swaps data.
And I believe that speaking about this subject to this audience, in this city, is the right time and place to discuss this issue in some detail. I often compare the effort to build the data reporting system to a big infrastructure project. It is a huge undertaking, one that will take considerable time. And Chicago is the place where much of the infrastructure of the futures industry was created. That infrastructure, created over many years, transformed the industry. It transformed how businesses of all kinds can hedge commercial risk. And let’s not forget, Chicago also played a leading role in another transformative infrastructure project in our nation’s history—the building of the national railroad system. Building an international data reporting system has some similarity to laying the tracks for a transcontinental railroad.
The FIA Expo brings together people who have worked in this industry for a long time, who understand the incredible innovation that this industry has experienced. You understand the operational and infrastructure issues that have created the foundation on which this industry grew. And so I think you will appreciate both the progress we have made, and the challenges we still face, when it comes to data reporting.
Important Progress on Data Reporting
Let me start by stepping back for a moment. Let’s remember where we were just a few years ago. In the fall of 2008, there was virtually no reporting of swap positions or transactions. That made it very tough for regulators to act in the midst of the crisis. Neither regulators nor market participants could assess the exposures of major institutions, or the interconnectedness of those exposures. The opaque nature of this market likely contributed to excessive risk-taking in the first place.
Today, the landscape is quite different. Reporting was a key goal of the G-20 leaders, and one of the most significant reforms of the Dodd-Frank Act. Now, all swaps, whether cleared or uncleared, must be reported to swap data repositories (SDRs). There are four SDRs operating in the United States, and two dozen internationally. We have requirements for who must report, what they must report, when they must report – and how the SDRs operate.
Regulators and market participants are seeing the benefits.
For example, you can now go to public websites of the SDRs and see the price and volume for individual swap transactions in real-time. This post-trade transparency facilitates efficient price discovery for all market participants, including end-users.
You can go to the CFTC website for a Weekly Swaps Report that gives a snapshot of the market, sliced and diced in various ways. And you can even sign up to receive it by email. Here’s our most recent report, showing outstanding notional amount of swaps in a few categories.
Users of swaps have access to data through SEF platforms and vendors that facilitate price discovery. Here’s an example of the daily post-trade data you can find for all swap transactions executed on one SEF. And of course, a key feature of SEFs is to provide pre-trade price information as well.
The transparency has fostered private sector innovation. One good example is right here: the FIA publishes a “SEF tracker” that provides a very good picture of what’s happening on swap execution facilities. Here’s your most recent report. And others are enhancing public data in additional ways, such as by aggregating information from the SDRs in real-time.
The swaps data is also improving our ability to oversee this market. As with futures and options data, swaps data is critical in helping us fulfill our core responsibilities.
Consider risk surveillance, for example. Over the years we have built a sophisticated surveillance system for futures that relies on a variety of inputs from clearinghouses, clearing members and large traders. Among other things, we receive data on clearing member positions by house and customer account, as well as by individual customer; large trader reporting; amounts of initial margin held and variation margin paid and received; as well as the financial resources of firms. With these and other inputs, we can look at market risk, liquidity risk, credit risk and concentration risk on a daily basis – at the clearinghouse, clearing member, and trader level. We can stress test exposures and back test the adequacy of margin coverage, for example.
Today, we are building swaps into risk surveillance systems so that we can identify and monitor swaps activities and exposures. We want to be able to stress test those exposures, compare them to available margin, and look at potential systemic issues. We want to look at activity and risk at the clearinghouse, clearing member, swap dealer and large customer level. And with uncleared exposures, we want to look at activity between particular counterparty pairs, and the interconnectedness of large institutions, among other things.
Risk surveillance is just one function. We need swaps data for enforcement, market surveillance, economic analysis of market trends, and evaluation of new products. In all these functions, we face similar challenges in creating an efficient and accurate reporting system.
Building this system is complex. Our futures reporting rules are just a few pages, in large part because we rely on a relatively small number of reporting entities. In addition, contracts are highly standardized. In swaps, there are thousands of entities reporting on an infinite variety of transactions. Because swaps are not tied to a single platform, and can be traded on a variety of platforms or bilaterally, we must have the ability to analyze and aggregate swaps data from across all these execution and clearing venues. Because there is much greater variety in swap transactions and terms than in futures contracts, we must have a reporting system that recognizes that variation but still enables us to aggregate where appropriate.
And because a swap can go through many stages and changes —through the clearing process, then through compression or reconciliation, and perhaps through other amendments—we must have ways to track that swap through its life cycle.
As I noted earlier, the process of building this system is like building a big infrastructure project. It is complex, and it takes time. It is iterative: as we understand the data better, and as the market evolves, we are refining the system. It is a process that requires collaboration with market participants. It requires collaboration across borders. We cannot do this alone. For one thing, we cannot do any analysis unless market participants provide us with data that is clean, consistent, accurate and timely in the first place. But we recognize that also requires working with market participants to refine exactly what data should be reported, how it should be reported, by whom and when. Only in this way can we create an efficient and effective process.
Improving Data Reporting
So let me talk about some of the challenges we face. I’d like to discuss a few areas where improvements are necessary – and what needs to be done.
Making Sure the Data is Complete. First, we must work towards obtaining complete data. There are several challenges here. Some required fields are not reported by participants. SDRs don’t believe they have the authority to reject data if it is incomplete. Shortly after I began as Chairman, I met with one SDR to discuss this issue. They showed me a chart called a “heat map.” It listed the top swap dealers, and contained columns for each of several reporting fields. And it indicated the frequency with which each swap dealer provided data for each field. Green indicated higher completion rates; red indicated low completion rates. There was a lot of red.
So I asked them to start doing this regularly, and to share it with both us and the swap dealers, so that those who were deficient would know that the SDR was keeping track. They did so. And completion rates have gone up significantly. A lot of red turned to green.
Overall, completeness has made a dramatic improvement across a number of data fields.
I encourage the SDRs to continue using heat maps and similar tools to improve the quality of the reporting. In addition, I believe we should change our rules so that SDRs have a greater ability to improve the quality of data before it arrives at the CFTC. We should empower the SDRs to validate not only the completeness, but also the accuracy of data before it comes to us. If the SDR rejects a participant’s submission, then I believe the participant should be considered out of compliance with our requirements. We will also hold the SDRs accountable for the manner in which they collect, compile and report the data they receive.
I expect we may propose these changes next year.
Making Sure the Data is Consistent and High Quality. In addition, the data we collect must be more consistent. It must be good quality data. It must be reliable and usable, without us having to check or clean it. But currently, that is not the case. For example, there is a lack of standardization in how many fields are reported. There is considerable variation in how the same information is reported to the SDRs, and in how the SDRs themselves transmit the same information to us. And this occurs even with relatively simple pieces of information. Here’s an example of the different ways that the WM Reuters foreign exchange benchmark is reported. Here’s an example of the different ways a specific CDS index is reported. In other cases, the differences may reflect legal distinctions, but if they aren’t critical to a particular analysis, we need to be able to treat them as identical.
In our original rules, we purposely didn’t prescribe exactly how each field should be reported – for a number of reasons. First, when the agency issued the reporting rules, we didn’t yet have any data to inform our views. And second, we expected the industry to develop standardized terms. That, unfortunately, did not happen.
So we are currently developing proposals to refine what it is we want and how we think it should be reported. The CFTC staff has made a dedicated effort to identify high priority fields where we believe standardization or clarification is needed. We are considering in particular what fields are critical to our uses of the data as well as where there have been reporting challenges.
The proposals will specify the form, manner and the allowable values that each data element can have. We will publish proposals on about 100 fields before the end of the year, and invite public comment. And we encourage market participants to get together and come up with common suggestions and other ideas.
We are also tackling this issue internationally. The CFTC co-chairs an international task force that is leading the effort to harmonize data reporting standards. This has been formed under the auspices of Committee on Payments and Market Infrastructures and the International Organization of Securities Commissions or IOSCO, and it involves many representatives from regulators in the G-20 countries.
One of its projects is to standardize the reporting of data fields by proposing definitions and formats for each. They recently published their first consultative document containing a batch of data fields. We are coordinating the in-house standardization efforts I just described with this international work.
Proposing standardized fields may require market participants to make changes in the way they report. However, we believe that this will ultimately reduce reporting burdens and enhance quality of data for everyone.
Refining Swap Identifiers. We are also working to develop effective means to identify swaps and swap activity by participant, transaction and product type throughout the swap life cycle. These include the Legal Entity Identifier – or LEI – as well as the Unique Transaction Identifier and Unique Product Identifier—the UTI and UPI. The LEI is the most advanced. It is a critical way to identify a specific entity and its activities. There are more than 400,000 LEIs today.
We need to expand the usefulness of the LEI so that it can be used to identify related entities – and aggregate positions or transactions among them, something that cannot be done today.
Aggregation is particularly important given that many market participants have a number of affiliates. We can aggregate entities manually by name, but that is time consuming and not always accurate. We are working closely with the LEI Regulatory Oversight Committee and other regulators to develop solutions that will address this challenge.
In addition, the international task force on data that I described earlier is developing a standardized unique transaction identifier, which is similar to our unique swap identifier. This will enable regulators to track a particular swap through its life cycle.
They are also developing a standardized unique product identifier, which will enable regulators to classify swaps by product type. They expect to issue guidance on both the UTI and UPI in 2016.
This work will enable us to track swaps and aggregate data much more effectively. I encourage you to comment on these international consultations.
I also want to take a moment here to thank the Office of Financial Research for leading the LEI effort. In addition, OFR has provided us with significant assistance on a number of data issues, such as the standardization and harmonization of swap data. And we are very grateful for their help.
Clarifying Reporting Obligations, and Eliminating Unnecessary Reporting Obligations. Another issue pertains to clarifying who has the obligation to report data, what data must they report, and eliminating reporting obligations that are not necessary. We’ve recently taken some steps in this regard.
For example, our staff made it clear that SEFs do not have an obligation to report confirmation data they do not possess—such as confirmation data that is incorporated from an underlying Master Agreement. The staff has said that SEFs need only report the primary economic terms and such other confirmation data to which they already have access. This relieves SEFs of any obligation to obtain an underlying Master Agreement or similar documentation.
We have also proposed modifications to the recordkeeping and reporting of cleared swaps. Under the current regime, if a swap is transacted on a SEF, it is reported to an SDR. If that “alpha” swap is then cleared, the so-called “beta” and “gamma” swaps that are created as a result are also reported. But those two new swaps might be reported to a different SDR than the one to which the original alpha swap was reported, and there might not be any record of the termination of the alpha swap.
Recently, we proposed to fix these issues by creating a simple, consistent process for the reporting of cleared swaps. That means clarifying the reporting obligations of the clearinghouse where the swap is cleared. The proposal will help ensure that there are not multiple records of a swap that can lead to erroneous double counting, and that accurate valuations of swaps are provided on an ongoing basis. It will eliminate unnecessary reporting requirements. It will help to reduce reporting costs and improve the quality of swap data. And it will improve the Commission’s ability to trace swaps from execution through clearing.
We should also eliminate reporting obligations that do not provide sufficient benefit. For example, we recently proposed to eliminate the obligation of commercial participants to report trade options to SDRs, as we determined that this was unnecessary. I expect the Commission will take final action on this very soon.
Further, the staff is recommending that we eliminate the Index Investment Data report, which we produce monthly. This is a survey of index-related holdings of certain traders and dealers. The report was started before our current swap reporting rules were implemented. It is not a comprehensive picture of index investing-related activity in the swaps markets. And I do not feel it is appropriate to continue to request data from dealers and traders for this report.
Enforcing Reporting Obligations. I’ve talked about the need to clarify some reporting obligations, and eliminate unnecessary ones. Equally important is the need to enforce reporting obligations. For those industry participants who do not make timely, complete and accurate reporting, we will not hesitate to carry out enforcement actions. We have done so already. Recently, we fined a major global bank $2.5 million dollars for repeated failures to comply with swap reporting obligations, including failing to report swaps and failing to correct errors in its reporting. And since the beginning of 2014, we have brought actions against six other institutions, including other major banks and an exchange, for various types of reporting violations. We will continue to promote compliance in recordkeeping and reporting – and hold those who are not in compliance accountable.
These are not the only steps we are taking to improve the swaps data reporting system. There are other areas of activity. For example, we are looking at the legal issues involved in cross-border reporting and sharing of data. We also recognize that there can sometimes be a conflict between public transparency and the ability of market participants to hedge effectively, such as in products that are extremely illiquid. The law directs us to take into account whether public disclosure will materially reduce liquidity, so we will continue to keep this issue in mind also.
The Upcoming Report on the De Minimis Exception: An Example of How Far We’ve Come. I’ve talked a lot about the challenges in the area of swap data reporting and the work ahead. But as I said at the outset, we’ve come a long way since 2008. The data today is extremely useful, not only for public transparency and price discovery, but also in the way it can inform policy and regulation. And you will see a practical application of this —I expect before the end of the month – when we will publish a preliminary report on the swap dealer de minimis exception. This is a provision of the CFTC and the SEC’s joint rule defining swap dealers that exempts certain entities from registration if they only engage in a small quantity of swap dealing. That threshold is currently $8 billion dollars – and will fall to $3 billion in about two years unless the Commission takes action.
The fact is that when the two agencies wrote the “de minimis exemption,” they did it without the benefit of much data.
Now we are taking a fresh look at the issue – this time with data to inform the public discussion. We are in a much better position today to estimate the impact of setting the level at different thresholds. This includes estimating the number of participants who might be subject to registration; and the percentage of the market that might be covered.
Our plan is to publish a preliminary version of this report and invite public comment on the data, the methodology and the issues discussed. As you will see, there are gaps in the data. But now we have a wealth of information that we can use to have a discussion about what is the appropriate level at which to set the de minimis threshold.
Before 2008 we couldn’t do that.
The data reporting project is a big, global undertaking. As with any good infrastructure project, building it right takes time.
But while there is much to do, there should be no doubt that we have made substantial progress. During the crisis, we were totally in the dark regarding this market. Today, we have much greater transparency, which benefits regulators as well as market participants. As we refine the data and reporting system over time, we will further enhance that transparency and the resiliency of our financial system.
Thank you for again for inviting me. I welcome your thoughts and questions.
Last Updated: November 4, 2015