Remarks at the ISDA Annual Legal Forum in New York
Gary Barnett, Deputy Director, Division of Trading and Markets
June 16, 2016
Thank you Katherine for that kind introduction. Good morning everyone, it’s a pleasure to be here with you today at ISDA’s Annual Legal Forum in New York. As we have just heard, you have a great line up of topics, panels and speakers today, and I am honored to be a part of it. In that regard, I do have to note that the Securities and Exchange Commission, as a matter of policy, disclaims responsibility for any publication or statement by any of its employees. The views expressed herein are mine and do not necessarily reflect the views of the Commission or of my colleagues on the staff of the Commission.
I will focus my remarks today on three main areas:
First, I will talk about risk governance, and in particular on risk management, culture and conduct, and board effectiveness. I will start that discussion with a bit of history which I think will lay an interesting foundation for your consideration of the efforts and progress that have been and are being made in these areas. Also, I will put more emphasis on board effectiveness, as it is now emerging as an area of practical attention by the regulatory community and I am sure you will be hearing much more about it in the near future.
Second, I will turn to a discussion about the deposit bank business model and the broker dealer business model to facilitate discussion in the area of WGMR and margin for uncleared derivatives transactions, which I see will be discussed in both morning and afternoon sessions today.
Third and last, I will talk a bit about mandatory clearing in the area of single-name CDS. Because the single-name CDS market can be so starkly different from much of the swaps — non-SBS — space, hopefully the contrast it provides can help illuminate many of the considerations in making mandatory clearing determinations, which will be discussed on your clearing panel today.
With that, let’s get started.
As I mentioned, I think some history can really set the context for our current efforts in risk management, culture and board effectiveness. So let me roll the clock back. Let’s go back to the late ‘80s and the ‘90s and the process of “globalization” in which we experienced a massive integration of people, companies and markets in a way that enabled people to reach each other and transact business on a “real time,” basis faster and cheaper than ever before. This whole process was enabled by many things, but some of the most important factors were:
- the end of the Cold War, and the failure of communism, which not only opened up borders but most importantly for our purposes today, from a cultural perspective, was taken as an affirmation of free market capitalism (in other words, the belief that the more market forces rule, the more the economy is open to free trade and competition, the more efficient and flourishing the economy will be);
- the development of new technologies around computers, digitalization, satellite communications, fiber optics and the internet, creating a new information based infrastructure;
- market capitalism powered by the new technologies enabled integration into real time competitive markets. The competition was so intense that those running the race wholeheartedly agreed with the economist Joseph Schumpeter’s idea about “creative destruction” and Intel’s Andy Grove’s idea that “only the paranoid survive”; and
- the faith in the effects of market forces, plus the necessity of doing so in order to survive the unremitting competition, led to additional corollary beliefs that greatly influenced management and governance theories. For instance, many believed that risk would largely be managed by the self-interest of market participants, i.e., that such would generate sufficient private market regulation. It also became widely believed that strong “top down” management or governance was problematic and just not possible in the environment of the times — where real time decisions had to be made. So business organizations were concluding that it was necessary to de-concentrate power, spread out information and decision making. This decentralization was seen as the only way to respond quickly enough to compete, survive, profit and win.
In the aftermath of the crisis, there was broad consensus that there had been a massive failure by institutions in addressing risk, and that there were very significant shortcomings in corporate governance generally. The assumed, largely unconscious, “invisible hand” of self-interest that would manage risk as and when needed had not materialized sufficiently. Other things also came to light— like the LIBOR and foreign exchange market scandals — which revealed massive problems with institutional culture. This is essentially the beginning baseline, and it set the stage for the efforts that have been ongoing in the areas of risk management, culture and now board effectiveness.
In terms of risk management today, we should note that it is still about self-interest and self-regulation. The key difference is that conscious intentionality is called for in managing one’s risks, and accountability is required. But it is still the company that defines its strategy and risk appetite. It is still the company that identifies its risks, chooses its risk tolerance limits, manages within those risks limits, including by establishing controls and policies and procedures around risk activities, and generally maintains monitoring, surveils for and remediates breaches of risk tolerance limits, all through the traditional 3 lines of defense. Also, because risk strategy and appetite is so important, as is the effectiveness of the risk management program generally, clearly this has become an important area of focus for the board.
I think it’s helpful to consider where we started and the progress that has been made. We’ve seen a focus on stronger internal audit, a more proactive compliance function, a strong CCO or CRO with appropriate seniority, independence and reporting line, the risk management of all risks of the business (including legal, operational and technology risks), the efforts to coordinate the efforts of the three lines of defense while each fulfills its specific function, and so on. The efforts continue but a tremendous amount of progress has been made.
In terms of the culture effort, I think it’s helpful to think about “culture” as part of risk management — a sort of mitigant to all risks in that it should tend to bolster a compliance culture generally, but more specifically as a support for conduct risk.
Why is that? Because achieving and maintaining a good culture is intended to affect how everyone in the culture thinks and makes decisions. Take it down to the most elemental level in our decision making: each of us stands in the present moment in time, projecting and imagining our options, thinking about probabilities and potential consequences, how best to chart a course, and then … we choose! But hopefully our choices are also being evaluated and made in light of our values. The basic idea about culture is that we hope that bad choices will be much less likely when everyone in the institution is on the same page about the values, ethics, business culture beliefs and the overall narrative of the institution that it desires. Of course it won’t rule out all bad choices, but it is believed to be one important support, one significant mitigant, to conduct risk.
In engaging in the culture exercise, an institution needs to think hard about how it wants to define itself, what it wants its customers and employees to think and believe about it, and how it wants its employees to behave. And of course to make the culture “real,” the institution not only has to define its culture and propagate it internally, it has to make that culture real, which it can only do if the institution “walks the walk” and “talks the talk.” This means that the institution’s practices in hiring, compensation, promotion, and discipline have to be consistent through and through, and the institution has to monitor and surveille for, among other things, behavior that is inconsistent with culture and which must be dealt with appropriately to support consistency with the culture.
Let’s move then to boards and board effectiveness. There is a widely held view that preceding the crisis and contributing to it, there was a pervasive failure of governance at financial institutions. And now, boards are being pressed to be more active, to own the strategic decisions and direction of the company, including its risk strategy, to oversee management, and to function with independence.
When you put those demands in context you realize that difficult pressures can exist: making strategic decisions for complex, fast moving businesses, puts a premium on expertise from those who really know the business, i.e., management and others in the company — yet the board must still oversee and operate with independence from management.
In any event, in this context, directors are finding it necessary to meet more frequently, take in more information, and be more proactive than before the crisis. Not surprisingly many directors are concerned about time and work pressures, accountability, and are asking for more guidance on the parameters of their responsibility.
A discussion of these issues has begun and is springing up in various places. Some market observers or participants have been very helpful in fostering communication and conversation about the issues and questions. Others are taking more of an advocacy position, seemingly seeking to hold back change and roll-back higher expectations. I’ve seen some very good whitepapers but I have also seen some holding on to the past. But given the state of things before the crisis, the identified need for improvement and the positions taken by various standard setters and regulatory bodies, all of them being pretty consistent, it’s hard to imagine a lasting move back to the past, and efforts to do so may only delay resolution of issues and the ability to provide answers that directors very much want and need.
I think areas of importance whose timing may be affected by these discussions include the ability to determine what is material to the board, which in turn could impact agendas, as well as decision making about what information the directors should receive and how to shape it efficiently to both inform and facilitate decision making. This could also impact the ability to take a view on appropriate time commitments, number of meetings, whether additional support services are necessary, and last but certainly not least, ways to enable and help the NEDs — the nonexecutive directors — be more effective, or to think of additional ways to help balance the need for independence and expertise.
I turn now to my next topic: the business model of the deposit bank and that of the broker dealer. By “business model,” I mean a description of the core or chief business activities of companies of that type, how those activities tend to drive financing, and what issues and fragilities exist around those activities and financing, and how all of that may affect key issues like capital, liquidity and leverage.
You will see from these business models that the deposit bank and the broker dealer can have very different pressure points, which can have implications in the decisions one might make about margin for uncleared derivatives transactions.
So let’s start with a look at the traditional deposit bank business model…
First off, it’s important to note that the deposit bank serves a very important function in society: it collects savings in the form of short-term deposits, and puts those funds to “work” by using them to make loans to borrowers. And of course these borrowers use their loan proceeds in ways that can be very good for the economy. In this model, the bank builds an asset base largely consisting of term loans and “lives” on the spread between the interest paid by borrowers on these loans minus the interest the bank pays to savers on their deposits.
But this model has some inherent risks that could be very serious but, because of the great social utility of the bank’s function, the risks are supported by society. What are these potential risks? First, having an asset base that consists to a large degree of term loans or other assets that are held to maturity, can create liquidity risks. The bank could have plenty of net worth but not have sufficient cash on hand for current needs. But the bank typically has access to government-funded liquidity, which allows the bank to go to the discount window, pledge qualifying assets and receive liquid funds. Second, the bank is, to a significant degree typically financing its term portfolio with short-term deposits. What if large depositors grow concerned about bank credit risk and want to withdraw their funds? For this risk, depositors are provided with deposit insurance up to some limit, which has proved to work well in helping retain short-term deposits. In addition, the government will also help resolve failed banks with a view to protecting depositors.
These protections and resources help absorb much of the risk of building a longer term asset base and financing it with short-term funding. But we should remember that risk, of course, doesn’t disappear. Instead, in this case, it is borne by the government or, ultimately, the taxpayer for the benefit intended to be obtained by society.
With this kind of activity and assets and liabilities, where would you first think about putting capital to provide a cushion for unexpected losses? Not surprisingly the first pillar of capital is about the credit risks embedded in the bank’s assets and, given the relative stability of the portfolio of assets, the amount of that capital is determined by reference to the amount of risk embedded in those assets, which in turn may be approximated by looking at each asset and deriving an amount from a fixed percentage applicable to the relevant type of asset or, for some, determining the risk on a modeling basis.
Of course we shouldn’t stop with the traditional model because a lot of change has occurred: During the ‘80s and ‘90s various nonbank institutions were coming to market with products and services that competed with those of the commercial banks, and the banks were allowed to expand their remit in order to compete and survive. Also “disintermediation” was occurring — borrowers were turning more and more to the capital markets for funding — from junk bonds, to commercial paper to securitizations and so on — which also put pressure on banks to expand their remit and try their hand at new ways of doing things.
As it turned out, given their lower cost of deposit funding, the banks won the competition, with the result that various types of risk-taking moved into the banks, which severely tested the strength of the traditional bank model.
In the aftermath of the financial crisis, bank risk and risk taking are being addressed in various ways, from limitations on proprietary trading, increased capital, leverage limits, liquidity requirements, requirements to better align incentives and risk taking, and better assuring an appropriate time horizon for senior decision making, and so on. In fact, I am sure that other requirements affecting banks will come up in discussions today such as the impact of mutualized liability as a CCP member, whether and how to place a limit on such mutualized liability, and what that means for the CCPs and market participants; and perhaps the ISDA stay protocol and beyond, and other changes as well.
In any event, despite these various recent changes, the salient features of the bank model remain: deposit based funding of a balance sheet that consists in large part of loans; public supported funding (deposit insurance), public supported liquidity (fed discount window), and public assisted resolution.
Now let’s turn to the broker-dealer business model. A traditional broker-dealer (BD) is a person or company that is in the business of buying and selling securities on behalf of customers (as broker), for its own account (as dealer), or both.
Unlike the bank, the BD cannot solicit cash deposits similar to a bank and cannot use customer assets to grow or finance its proprietary trading activity. This means generally that a BD must either raise capital or access the short-term secured lending market (repo and stock loan) to finance its proprietary positions and source its liquidity. In addition, the traditional nonbank BD facilitates financing for securities through the repo or stock loan/stock borrow markets.
Given its business of buying and selling securities, the broker-dealer typically has a balance sheet comprised in significant part by liquid securities which are being bought and sold, margin loans, and reverse repos and stock borrow, which represent financing provided against such securities, all financed by similar short-term financing like repos and stock loans. Both sides of the balance sheet are short-term and typically rolled or refinanced as needed to support a business that involves the buying and selling of securities. To underscore this point, note that while a secured lending market may be quite liquid, lenders in this space often have a contractual right to unwind their secured loan on relatively short notice. To meet the short-term liability, a BD must maintain a highly liquid balance sheet so that these short-term obligations can be timely met, such as finding another secured lender or by selling the underlying collateral.
How does the secured funding model work for BDs during a stress event? For example, what happens if short-term debt cannot be refinanced or rolled? The BD has to unwind corresponding reverse repos by delivering pledged securities back for cash which can be used to repay short-term financing that isn’t being rolled. Or the BD can sell liquid securities and pay off maturing short-term loans. But what happens if repo borrowers aren’t able to accept their pledged collateral back and deliver cash? The BD will need to realize on the pledged securities by selling them in the market. But what if the BD can’t sell those securities or its own securities at that time and/or the value of those securities has plunged so they won’t generate sufficient proceeds to take out the loans that are due and payable? If the BD is not able to raise sufficient cash to repay its maturing short-term loans, it could well default! So in light of that, what protections should be built around this structure?
The BD is not provided with deposit insurance, it doesn’t get publicly funded liquidity as a backup when securities markets are not available. In a stress event, a BD has to resolve itself (although SIPC provides custodial type protection to investors in respect of their investments held in the BD). Thus, the SEC’s rules require that the company retain sufficient liquid capital, in an amount substantially more than all the unsubordinated liabilities of the company. This protects customers and creditors by enabling the BD to pay them off quickly and in full from available funds if the BD needs to be liquidated. In valuing its securities assets for net capital purposes, the rule requires BDs to take haircuts related to the risk of the securities. Furthermore, fully paid securities of customers are held in custody and net amounts owed by the BD to customers are held in segregation.
To compare and contrast these two business models: The deposit bank, I think of as being built on a term structure with state supports for certain of its significant risks. In contrast, the BD is all about liquidity. Its business is about buying and selling securities and helping to facilitate the financing of the same. In a business-terminating scenario, the BD needs to have sufficient liquid assets to pay off all creditors and customers so it can self-liquidate and cease business. Maintaining and supporting liquidity is a key concern for the health of nonbank BDs and things that put pressure on liquidity deserve special attention.
These two business models are strikingly different, and it is in this context that I come to a few questions for consideration with respect to WGMR: WGMR proposes to have dealers posting initial margin (IM) to each other and as adopted so far in the US, to have such IM held, otherwise unavailable for use. What are the potential impacts of these requirements on BD liquidity and capital? How does this compare to the treatment that deposit banks receive under these requirements? Could there be potentially different outcomes for customers and creditors of each? Again, I leave these questions for the panels and your own consideration.
Finally, I want to say a few words about a single-name CDS clearing mandate. The CDS industry, including the CCPs, dealers, and some other firms, have been suggesting that mandatory clearing of certain single-name CDS will increase the liquidity of the single-name CDS market and reduce systemic risk.
Now why is there a desire to increase liquidity in the single-name CDS market? As industry participants are keenly aware, since the financial crisis in 2008-2009, single-name CDS trading has declined significantly. One commenter wrote to us saying that by the end of 2014, the outstanding notional amount of single-name CDS had decreased by over 60% since 2008. A recent ISDA Quarterly reported the notional outstanding volume to have fallen by more than 75% since June 2008. From what we have heard, a variety of factors may have contributed to the decline, including (1) changes in bank capital rules, (2) the significant decline in the synthetic CDO market, (3) uncertainty surrounding portfolio margining, (3) generally lesser demand whether for hedging or speculative trading perhaps responsive to lower default rates post financial crisis, and (4) perhaps the ban on naked sovereign CDS in certain jurisdictions. Also, with the decline in single-name trading volume, there has also been a decrease in the number of single-name CDS market participants and it’s been reported this may have also affected the number of market makers, which can put additional pressure on liquidity. Also, single-name CDS client clearing between 2012 and 2015 was low.
In response to the decline in single-name CDS market liquidity, the industry participants, including CCPs and buy-side firms, have established a variety of initiatives in an attempt to revive the single-name CDS market, including a reduction of single-name CDS roll frequency, a single-name back-loading incentive program by a CCP; and a buy-side commitment to voluntary clearing. And maybe these have helped as single-name CDS clearing volume in 2016 so far has increased significantly compared to the volume of the entire year of 2015. We’ll have to see how that plays out.
In any event, I understand the desire to get trading volumes and numbers of participants up in this market — that’s good business. But, as we all know, sufficient relevant liquidity is important in the course of addressing a clearing member’s default — to be able to move or replace trades of that defaulted member in order to get the CCP back into a matched book position. Thus liquidity is an important consideration in the analysis of any mandatory clearing determination. And for this very reason, whether mandatory clearing should be a solution to a lack of liquidity in the single-name CDS market needs to be carefully considered. We all bear the risk if the hoped for liquidity doesn’t materialize by the time it’s needed. Also, let us not lose sight of other unintended consequences: in my view, if mandatory clearing is forced into an area where the market is not convinced that the risks can or will be adequately managed — so that the only way to not be at risk is not to play the game — the result could be even less liquidity than where it started from.
Now maybe the market initiatives will bear enough fruit so this problem becomes a non-issue…the effort is definitely positive news. But having sufficient market liquidity is only one factor. The Commission is also required to consider a variety of other factors in connection with a mandatory clearing determination, and I will mention two of them now for your consideration: including factors like: the effect on competition; and the existence of reasonable legal certainty in the event of the insolvency of the relevant clearing house or 1 or more of its clearing members with respect to the treatment of customer and counterparty positions, funds and property.
These raise serious questions for consideration. Consider first, the factor about reasonable legal certainty in the event of the insolvency of the relevant clearing house, which is interesting. Keep it in mind as you listen to the clearing session today. Two years ago, CCP recovery, including loss allocation and return to a matchbook, was one of the most frequently discussed topics. Now discussions on CCP recovery are coordinated with the discussion of CCP resolution for good reason. Is there currently sufficient clarity and certainty with respect to CCP resolution? Also, for that very reason, the buy-side has commented that they should not be forced to clear if they are not protected from CCP failures or if they are worse off in clearing than in bilateral trading in the event of CCP recovery or resolution.
Last, in connection with the consideration of the effect on competition, the buy-side firms have argued strongly that mandatory clearing should not be implemented unless and until more than one CCP offers clearing in the same swap. For them, the issue is not only a matter of assuring appropriate fees and charges but also having an alternative CCP may increase the likelihood of preserving continuity of critical services, minimizing the destabilizing effect of a CCP failure and protect clearing participants, including customers.
That concludes my prepared remarks. I am happy to take questions. Thank you.
* * *
(In Q&A, the following was provided to explain how business models might help facilitate analysis and discussion of differences…)
Let me illustrate the point with a short story. This is a story about a furniture guild in a far-away land and a disagreement between two groups of guild members over a rule that one of those two groups had proposed for application by all.
This story really finds its beginning when the members of the guild who made their living building dressers came under pressure because of injuries to consumers after several of their high top dressers tipped over when several top drawers were heavily packed and were opened at the same time.
The dresser makers carefully studied the problem and came up with what they thought was a good solution: looking at height, width and depth, they concluded that the largest dimension should never exceed the other two dimensions by more than a certain ratio.
The rule seemed so easy and useful that the dresser makers passionately sought to have the rule adopted by the entire guild. But there was one group in the guild that expressed concern and requested to be excused from application of the rule or, at a minimum, that the rule be modified to fit their situation. Not understanding the reason for the concern, the dresser makers rejected the request, arguing that a message of guild wide “harmonization” could send an assuring and powerful message to the public.
Finally the guild members all got in a room to hash it out and the dresser makers who were the proponents of the rule went first. “Why don’t you like this rule?” they demanded of the other group. “It addresses the problem perfectly. Not to mention that solidarity creates great optics and messaging and, frankly, don’t you realize that if you aren’t on board people may think there’s something wrong with your furniture?”
The respondents shook their head and replied, “Look, we think the idea behind your rule is great but we think we have to have some adjustments to make it work reasonably for us. You make high top dressers. We make twin beds. The longest dimension of the furniture we make is length. We’ve studied the risks and, frankly, length really isn’t a problem until the length of the bed is REALLY, REALLY long. But in any event, if we apply your rule without any modification, the ratio would limit the length of our twin beds to 5’3” long….
Now the book in which this story is told is quite old and some of its pages have been lost so I can’t tell you how the story ends. But I do think it has some application to the discussion of margin in the context of two different kinds of businesses. To just state our respective rules doesn’t necessarily tell us why the rules are as they are or how well they align to the fragilities of the respective business. Also, for two or more regimes to try to mix and match rules or adopt one without considering outcomes in the context of the respective business types may not get an optimal result either.
 See, for instance the Federal Reserve System’s Enhanced Prudential Standards from 2014 or the Basel Committee on Banking Supervision’s Guidelines on Corporate governance principles for banks or the OECD’s Principles of Corporate Governance both from 2015.
 Generally, the BD does not have to take a capital charge for reverse repo or stock borrow where the collateral is liquid. On the other hand, if a BD enters into a reverse repo or stock borrow for an illiquid security, the BD must deduct the value of the asset in full when computing net capital.
 Generally, if a BD repos a security, it must still take a haircut on the position as if it still owned it.