When the Volcker Rule went into effect in July 2015, it was a significant revolution in the trading practices of US banks. Millions were spent preparing for compliance, and lots more on the new business as usual. Despite all of the effort, however, there are still some aspects that aren’t running smoothly. This article looks at some of those lingering problems and asks whether the VR will help or hurt the safety of the banking business.
When the Volcker Rule (VR), one of the major parts of the Dodd-Frank Act, went into effect in July 2015, it was a significant revolution in the trading practices of US banks. Millions were spent on preparing for compliance, and lots more on the new business as usual; but it shouldn’t surprise us that, despite all that effort, there are still some aspects that aren’t running smoothly. Let’s take a look at some of those lingering problems, sorted by exemption.
This was always regarded as the most troubling aspect of the VR, so it was where banks spent most of their time and effort in preparation. Perhaps the most resources were applied to the tracking of trade volumes in comparison to reasonably expected near-term demand (RENTD). As it turns out, the tracking was much easier than determining RENTD in the first place.
It hasn’t helped that most markets have been more volatile than everyone had expected, or that one major market, fixed income, has been decidedly less volatile than expected. And the future is full of possible sea-change events, such as a Brexit, the long-awaited Fed tightening, or the possibility of a hard landing in China. To top it off, lower trading margins and profits have prompted banks to reduce staff and positions, and to rely more and more on algorithmic trading.
All those factors, taken together, lead to a very murky view of RENTD, to the point of calling into question the very idea of a “reasonable expectation.” Those market makers that expected the Fed to tighten by now, with the resulting paroxysm in the fixed income markets, have prepared for customer demand which, however reasonable to expect, never materialised. After one or two repetitions of that exercise, they probably backed off. Inevitably, somewhere down the line we will see those events come to pass, but by then most people’s RENTD won’t be anything close to reality.However, even if the VR had never happened, those market events would be most likely to occur. Long periods of low volatility and low spreads, with the resulting drop in liquidity, followed by sea changes in monetary policy, will invariably lead to market dislocations. If market makers feel constrained by RENTDs that are unrealistically low, based on past experience, they will be hanging back just when they are needed on the front lines. When that happens, critics may blame the VR, but it’s hard to see how it could play out any other way, even without the rule.
This exemption got much less press than market-making, but its lingering problems may be significantly worse. The culprit here is a combination of the hold-to-maturity basis of the exemption and the extraordinarily low short-term rates. With historically low loan demand, and suppressed earnings throughout the banking industry, this has led to banks’ reaching for yield in their liquidity portfolios.
The same impending market changes that bode ill for RENTD will have significant impacts on the liquidity exemption. If banks have been moving out on the yield curve and down the quality spectrum for yield, their liquidity portfolios are exposed to high degrees of volatility whenever rates start rising. Any attempts to liquidate those portfolios in the face of rising rates faces a double whammy – losses on the portfolio and questions about compliance with the held-to-maturity exemption rules.
There doesn’t appear to be much a bank can do about this conundrum, short of giving up on the idea of getting yield from the portfolio. Even repositioning the portfolio in advance of a rate rise gives the appearance of excessive trading as defined by the VR, and waiting for the inevitable will only make matters worse. Here, again, we can ask whether the projected financial consequences of reaching for yield and then having to liquidate have anything to do with the VR, or are simply the result of market forces. Either way, we would be explaining a bad situation.
This exemption was probably the most far-reaching of the VR reforms, and undoubtedly the least discussed – in the press, anyway. The hallmark of this exemption is metrics and math; the quantification of both the risk and the hedge, the tracking of the correlation, and making mid-course corrections to any hedge positions.
All of these requirements look like good business processes, bringing discipline to what was often a seat-of-the-pants practice. But we must remember that the exemption applies to all hedging, not just the hedging of market risk. In other words, risks associated with such basic banking practices as loan origination, asset-liability management, or even cross-border expansion must follow the same rigorous requirements.
In many of these cases, banks may not have good foundations for either the risk assessment or the hedge construction. Even where they have history to rely on, it may not be a good yardstick for an uncertain future. It may not only mispredict the risk, it may mispredict the hedge performance as well.
Of course, risk is inherently about the unknown, but there are two kinds of unknowns to be confronted here. The first is the path of what we can think of as the independent variables, the things that could cause the losses. The second is the relationship between the independent and dependent variables, particularly in the more arcane risks.
One possibility of these complications and higher rigor may be a decision to bypass the hedge entirely as too difficult to construct. If banks feel they may face criticism for attempting a hedge that didn’t work, taking on the risk itself may look like the easier path.
Is the Rule to Blame?
Given the ample opportunities for things to go wrong in banking, we should anticipate a certain amount of finger-pointing in the future. One discussion I fully expect to hear is about whether the VR helped or hurt the safety of the banking business in general. That discussion will likely focus on results, without taking into account the alternatives, but we need to understand both sides, and preferably before the big changes happen. Some questions that need to be asked:
To what extent does the rule encourage good processes while discouraging bad ones, and vice versa? The intentions of rules are always good, but the results aren’t always what we expected. Where rules complicate a good process to the point of making it unattractive, they can have exactly the opposite of their intended effect.
How much can any results be traced to the rule itself and how much are a function of market forces? Many of the changes in bank trading over the last five years have been due to those market forces, and not to rule changes. In the future, as we assess additional impacts on the structure of the banking industry, we need to separate out those things caused by rule changes from those things that would have happened anyway.
What is the difference between rules and good management? The VR drafters, and its enforcers, sincerely believe that they are fostering good risk management across the banking industry. But, if risk is about the unknown, then no rule, no matter how specific, will cover the waterfront. So to the extent that rules lock banks into specific behaviours, they may actually inhibit risk management.
The future of the banking sector in the US is a function of many things, among them the Fed’s actions as a central bank and management’s skill and flexibility in dealing with uncertainty. Since almost all the money in the country is made up of banks’ promises to pay, we all have a major stake in how well they do their jobs. To the extent that the VR prompts better risk management, it will have been worth it. To the extent that it is a burden, we may all live to regret it.
See full article by George Bollenbacher here