From its treatment of portfolio hedges to correlation tests and inventory limits, the Volcker rule lacks detail on key elements, lawyers say. That is an attempt to give the industry – and supervisors – some flexibility, but it could make compliance a game of chance. Peter Madigan reports Banks will not be allowed to engage in portfolio hedging under the terms of the Volcker rule, agreed on December 10 by five US regulators – but they will be allowed to hedge risks on a portfolio basis. That awkward distinction makes the final rule tougher than the version proposed in October 2011, regulators have claimed, but lawyers are dismayed at the lack of detail added to the statute after three years of work and 18,000 comment letters – the final rule adds just four words on the topic beyond the language that was provided in the text of the Dodd-Frank Act. Similar problems arise elsewhere, lawyers say, obscuring concepts vital to the rule, such as correlation analysis and limits on market-making inventory (see boxes, “Reasonably expected near-term demand...” and Correlation analysis: what makes a hedge a hedge?). “No-one wants prescriptive regulation that is just flat wrong but, equally, no-one wants to be in a position where the regulators give no more guidance than Congress provided in the statute. That really does not give you much to hang your hat on in terms of what you can and can’t do,” says Micah Green, co-chairman of the financial services and tax practice at law firm Patton Boggs in Washington, DC. What banks cannot do is engage in proprietary trading; what they can do is risk-mitigating hedging and market-making (Risk December 2013, pages 22–25). Drawing a line between the good stuff and the bad is what has taken regulators so long and, from the industry’s point of view, portfolio hedging was on the right side of the line in the 2011 proposals – it was mentioned specifically as an acceptable form of risk-mitigating hedging. The big question is what its absence from the final rule means. When the issue came up during the open meeting at which the Federal Reserve Board approved the rule, the agency’s general counsel, Scott Alvarez, said regulators stuck to the language of the act for the sake of clarity. “Portfolio hedging means something different to just about everybody,” he said. But the hundreds of pages of preamble accompanying the 71-page final rule offer some extra guidance on the regulators’ thoughts, he added. “To be a risk-mitigating hedge, it has to be related to identifiable risks of identified positions at the organisation, not to general revenues or general expectations of losses, which some consider to be portfolio hedging. The rule allows aggregated positions to be hedged. That may fit some people’s opinions of portfolio hedging, but it is a narrower thing than the general view people have of portfolio hedging,” Alvarez explained. It may not make a huge difference. The rule states that the prop trading ban “does not apply to the risk-mitigating hedging activities of a banking entity in connection with and related to individual or aggregated positions, contracts or other holdings of the banking entity and designed to reduce the specific risks to the banking entity in connection with and related to such positions, contracts or other holdings”. Banks should still be able to put in place a short credit position as a hedge against a portfolio of corporate loans, for example, or a long interest rate volatility position against a bond portfolio, lawyers say. What the agencies appear to have clamped down on are vague, general-purpose hedges that are not attached to a specific business or portfolio. The bogeyman is the so-called London Whale, a trader who helped rack up an estimated $6.2 billion loss at JP Morgan’s chief investment office while engaging in what the bank originally claimed was a hedge. Investigations have since revealed the strategy changed dramatically in the early months of 2011 as the chief investment office tried to reduce its capital needs – it stopped being a hedge, in other words (Risk October 2012, pages 23–26). But regulators have continued to invoke the episode when arguing for a tougher version of the Volcker rule. “A lot of lawyers spent a lot of time trying to figure out where the agencies had authority to expand on the statutory language and where they didn’t. Clearly, the agencies could have said more here, and in fact the proposed rule did say more, but certainly what everyone was hearing from the Federal Reserve was that – between the proposed rule and today – the London Whale trades occurred and that renewed the focus on banks’ capacity to be able to take a huge risk position and call it portfolio hedging,” says Robin Maxwell, head of the US financial regulatory group at law firm Linklaters in New York. Elsewhere in the rule, there was good and bad news for dealers in the treatment of hedges executed by market-making desks. The proposed rule required these positions to comply with the terms of the market-making exemption, as well as those of the risk-mitigating hedging exemption, but the final version takes a softer stance because of fears the more granular compliance and documentation requirements for hedging would hurt market liquidity. Under the proposed rule, dealers would have been bound by the hedging exemption’s requirement to use only reasonably correlated hedges, for example, which some banks argued would deter market-making in cases where there may not be a direct hedge, or where a partial hedge is needed to reduce costs. Instead, only the market-making exemptions will apply, with the preamble explaining regulators decided it was best to allow banks to “determine how best to manage the risks of trading desks’ market-making-related activities through reasonable policies and procedures, internal controls, independent testing, and analysis, rather than requiring compliance with the specific requirements of the hedging exemption”. But this only applies where a desk is hedging its own exposures. If a second desk is responsible for hedging the market-making positions – as is the case for the centralised functions that manage credit valuation adjustment at many dealers, for example – then it must comply with the risk-mitigating hedging exemption, the preamble states (see box, CVA desks face correlation tests). Regulators accept this will increase the burden on these desks but insist the extra controls are necessary. The final rule also requires that US banks with more than $50 billion in assets worldwide, or foreign banks that have $50 billion in assets in their US operations, compile seven metrics for each trading desk to help management and regulators monitor compliance with the Volcker rule – including inventory turnover, the ratio of customer-facing trades and comprehensive profit and loss attribution (Risk August 2011, pages 22–25). The idea is to create an early-warning system, rather than a set of limits, with reporting due to begin in July this year. “The metrics are designed to be a tool for triggering further scrutiny by banking entities and examiners in whether a banking entity is engaging in prohibited proprietary trading, high-risk trading strategies or maintaining exposure to high-risk assets. The final rule does not propose specific thresholds or other bright lines that the individual metrics may not breach,” said Sean Campbell, deputy associate director at the Federal Reserve, speaking at the regulator’s open meeting. The agencies will start collecting this data from July 2014, and will decide whether the rules need to be amended by September 30, 2015. After the reporting regime beds down, the asset threshold beyond which compliance is required will drop to $25 billion on April 30, 2016 and then to $10 billion on December 31, 2016. As first reported by Risk on November 15, implementation of the prop trading ban will also be delayed. The final rule pushes the date back until July 2015 at the earliest, with the possibility left open that full implementation could be delayed until July 2017. “The rule we have presented today is going to require a lot of adjustment by the banking entities subject to the rule, including the development of pretty detailed compliance programmes for the largest institutions. The statute allows the board to allow three one-year extensions, one at a time. Because of the new requirements, the staff of the board and the staff at the other agencies recommend that the board grant a one-year extension of the conformance period until July 21, 2015,” said Alvarez. The decision to extend the conformance period by an additional 12 months was passed by a unanimous vote of the Federal Reserve governors a short time later. “Reasonably expected near-term demand...” The Volcker rule gives trading desks much of the flexibility they wanted, but it also sets boundaries. Or, rather, it tells banks how to set their own boundaries. And it doesn’t do a very good job of it, lawyers claim, making it difficult for banks to know they are complying. One example comes from the market-making exemption. It allows banks to hold positions in anticipation of client demand – good news for dealers, which had argued they would be reduced to brokers without it – but states they can only hold enough inventory to meet “reasonably expected near-term demand”. That sounds sensible enough, but although this clunky phrase appears 77 times in the Volcker rule and its preamble, lawyers say it’s not clear how to apply it. According to the rule, expectations of demand must be based on “the liquidity, maturity and depth of the market for the relevant types of financial instruments; and demonstrable analysis of historical customer demand, current inventory of financial instruments, and market and other factors regarding the amount, types and risks, of or associated with financial instruments in which the trading desk makes a market, including through block trades”. The preamble seeks to flesh this out, but derivatives attorneys say its guidelines are just as vague. For example, the “demonstrable analysis” must be both backward- and forward-looking, “taking into account relevant historical trends in customer demand and any events that are reasonably expected to occur in the near term that would likely impact demand”. But that is not always the case. If the market is immature, the bank would be expected to draw on experience in similar products. If the market is illiquid, regulators say that should also be taken into account. And if a trading desk expects a change in the pattern of customer needs, then historical trends can be downplayed. None of these qualifiers is defined separately – the bottom line seems to be that inventory levels ought to be more or less consistent with those seen in the past, and if they are different, the bank will need to show why that is the case, lawyers say. But that won’t be easy to comply with or police. “The expected near-term demand requirement is a very tough standard. A bank would have to look at the historical sales and volume the dealer has experienced, and if that desk is significantly outside of that experience, that would be enough to trigger the interest of the regulators – but in that instance, the dealer could always respond that it had been expecting some large orders to come in that never materialised,” says Matt Magidson, chair of the derivatives and structured products practice at law firm Lowenstein Sandler in New York. “If supervisors are going to look at this measure in hindsight and judge whether desks were holding too much inventory during a certain period and that those assets were not moved within a reasonable amount of time for the purposes of taking a proprietary position, it will be very hard for dealers to judge in real time whether they are holding an appropriate amount of inventory in the immediacy of any given moment,” he adds. In practice, supervisors will provide the missing detail, applying their own interpretation to the text, says Scott Cammarn, a Charlotte, North Carolina-based partner specialising in financial regulation at law firm Cadwalader, Wickersham & Taft. “Establishing a framework to monitor a regulated entity’s expected near-term customer demand will have to be worked out in a supervisory guidance document for bank examiners – it is almost impossible to define these matters in rule writing,” he says. Correlation analysis: what makes a hedge a hedge? The final version of the Volcker rule abandons two controversial elements of the proposed exemption for risk-mitigating hedging – that a hedge would only be permitted if it was “reasonably correlated” with the hedged item, and that ongoing testing would need to show the offset remained intact. In their place is a requirement for correlation analysis to be conducted at the outset of a trade – a lower hurdle, but one with problems of its own. Instead of the tough, prescriptive approach the industry had feared, banks now face a requirement with no clear boundaries, lawyers warn. “Correlation between hedges and hedged exposures is an area where the regulators want to encourage flexibility, but the industry still has the right to some certainty about what counts as a sufficient level of correlation, as well as to know what kind of guidance is being given to the supervisors about what constitutes permissible correlation,” says Kevin Petrasic, a partner in the global banking practice at law firm Paul Hastings in Washington, DC. The rule states that “correlation analysis [must] demonstrate that the hedging activity demonstrably reduces or otherwise significantly mitigates the specific, identifiable risks being hedged”, but then adds “it is important to recognise that the rule does not require the banking entity to prove correlation mathematically or by other specific methods. Rather, the nature and extent of the correlation analysis undertaken would be dependent on the facts and circumstances of the hedge and the underlying risks targeted.” Attorneys are struggling to work out how this would be applied in practice. “If you don’t have a mathematical bright line test for correlation, then you could end up with positions that are not actually correlated at all, but which theoretically could count as a hedge if the bank can justify its original thinking to the regulator. Without a test, though, you won’t know whether it is permitted or not,” says Julian Hammar, counsel at law firm Morrison & Foerster in Washington, DC and formerly assistant general counsel at the Commodity Futures Trading Commission. Similar criticisms are levelled at the sketchy rules on the amount of inventory a bank can hold and still qualify for the market-making exemption (see box, “Reasonably expected near-term demand...”). In both cases, attorneys recognise the regulators are trying to give the industry much-needed flexibility, but they warn the outcome will be a period of uncertainty as the rules bed down, with different banks taking different approaches. “I think the hedging rules will be expanded on and further addressed during the examination phase – but it seems like a trial-and-error process. Institutions are being asked to give regulators parameters for what they consider regular behaviour, but they will be subject to second guessing, and the examiners can come back and say the metrics will not suffice and that the bank has to do them over again. In the case of correlation, the agencies do not specify the type or level of correlation that has to be demonstrated, for example,” says Don Lamson, a partner specialising in derivatives at law firm Shearman & Sterling in Washington, DC. CVA desks face correlation tests The Volcker rule’s treatment of hedging will increase the workload for credit valuation adjustment (CVA) desks and, in a worst-case scenario, could prevent them putting in place capital-mitigating trades, dealers fear. Some banks are now consulting with in-house lawyers on implications for the desks. The head of CVA trading at one large European bank with operations in the US says the text looks problematic, but adds “the lawyers need to dig into this a bit more before we reach any kind of verdict”. According to the rule, bank trading desks can hedge risks arising from their own trades, under the terms of a market-making exemption. This allows banks some freedom over how and when to hedge, as well as what to hedge with, as long as they do so in line with clear internal policies. CVA desks, however, centrally manage counterparty risks accumulated by other traders, exposing them to the more demanding criteria laid out for what the rule calls risk-mitigating hedging. Trades falling under this second exemption have to meet tougher conditions, including requirements that hedges reduce “one or more specific, identifiable risks” and must not, at inception, give rise to additional risks that are not being hedged. Banks are required to show they expect to meet these conditions, in part by conducting correlation analysis (see box, Correlation analysis: what makes a hedge a hedge?). The problem is that some CVA desks put trades in place to reduce regulatory capital requirements, rather than to mitigate identifiable counterparty exposure. It’s not clear this counts as a hedge for the purposes of the Volcker rule, which lists identifiable risks as “market risk, counterparty or other credit risk, currency or foreign exchange risk, interest rate risk, commodity price risk, basis risk, or similar risks, arising in connection with and related to identified positions, contracts or other holdings of the banking entity”. In addition, regulators insist index hedges be used in many cases if a bank wants to claim capital relief. The resulting positions may not be well correlated with the underlying counterparty risk – as demonstrated by the €115 million loss racked up by Deutsche Bank’s capital mitigation programme in the first three-quarters of last year, which went hand in hand with a huge reduction in risk-weighted assets (Risk November 2013, pages 24–26). “You have to establish correlation, and that might become a problem in some cases when correlation between individual types of counterparty risk and the index may not be able to be established,” says Dmitry Pugachevsky, director of research at software vendor Quantifi in New York. In the long preamble to the Volcker rule, regulators state that, if correlation cannot be demonstrated, a bank will need to “explain why not and also how the proposed hedging position, technique or strategy is designed to reduce or significantly mitigate risk and how that reduction or mitigation can be demonstrated without correlation”. One CVA head says a lot depends on the regulators’ definition of correlation, as well as the risks that can be hedged, but he argues CVA trading ought to be safe. “I think they see the point in CVA desks – Volcker was aimed at things such as the London Whale. CVA would be collateral damage if it was affected,” he says. The other requirements of the exemption include the establishment of written hedging policies and ongoing monitoring to ensure hedges comply with those policies throughout their life, and also remain within the boundaries laid down by the rule. Along with the need for correlation analysis, this will add to the workload for CVA desks – an objection raised by JP Morgan, among others, during the comment period for the rule. Regulators acknowledge that in the preamble. “The agencies understand this rule will result in additional documentation or other potential burdens for market-making-related hedging activity that is not conducted by the trading desk responsible for the market-making positions being hedged.” But they conclude the risks of a CVA desk or other centralised hedging function being used to evade the Volcker rule are significant enough to warrant the extra burden. Quantifi’s Pugachevsky says this should not be too problematic – CVA desks tend to trade relatively infrequently, he claims, so should have time to tick all the new boxes, and are also no stranger to regulation. “CVA desks and people who support CVA desks have got used to this kind of control and validation. It’s not a new thing for them,” he says. Lukas Becker and Laurie Carver...
Start a FREE trial or subscribe to continue reading:
Start a 4 week free trial
Try Risk.net's premium content for a limited period. Register now for your FREE trial to one of our leading brands.
*not available to previous trialists or subscribers.
Log In or Subscribe Now
Subscribe to Risk.net Business now to access all our premium news & features content for 1 year.
Pay by Credit Card for immediate access.