The surprise announcement by JP Morgan Chase (Morgan) of a $2 billion trading loss will clearly have a substantial effect on the politics surrounding the reform of financial regulation. Advocates of reform will have an easier time, particularly in regard to the Volcker Rule, the part of the Dodd-Frank financial reform act that is intended to eliminate “proprietary trading” by banks. Industry arguments about the practical aspects of implementing reform will lose much of the benefit of the forceful advocacy that Morgan’s CEO, Jamie Dimon, had been able to bring with his strong reputation and that of his firm.
Putting aside the politics, what public policy lessons can be drawn from the facts of the Morgan debacle? This paper will lay out what we have learned and not learned. It will not focus on the effects on Morgan, except as there are implications for public policy more broadly. Nonetheless, I would like to be clear that I am an ex-employee of Morgan and retain great respect for the firm and friendship with many of those still there. However, this has not prevented me from being a strong supporter of financial reform and of Dodd-Frank, with a few exceptions that include the Volcker Rule.
The following questions related to the trading losses are at the heart of the public policy debate:
- What happened?
- Who bears the loss?
- How big a loss is this? Did it represent a risk to the larger financial system?
- Would the Volcker Rule have affected these trades, if the rule had been in force? Should it?
- How is this relevant to other proposed structural changes to banks?
- What does this tell us about communications and transparency issues?
Morgan took a number of large positions in derivatives and securities, primarily related to the risks that the Euro Crisis would cause substantial credit losses. Morgan says these positions were intended to offset or “hedge” other positions that they held throughout the bank, which would have lost money if the Euro Crisis had developed in that manner. The positions were taken through the unit run by the firm’s Chief Investment Officer. This unit is responsible both for hedging against the large cumulative risks that can build up in the rest of the firm and for making a good return on excess funds that the bank is not employing in other areas. It reportedly managed a total portfolio of about $400 billion.
In the event, these particular positions lost large sums of money, currently about $2 billion, prior to the offsetting reduction in income taxes. This might have been acceptable if the positions had indeed acted as an effective hedge, since this would mean that other parts of the bank would have made corresponding gains from the same market movements that would have cancelled out the large losses. However, Morgan has indicated that the hedges were not effective because “egregious mistakes” were made, partly because the hedging strategy was too complicated and difficult to understand and evaluate. In fact, the mathematical models that Morgan uses to assess its Value at Risk (VAR), a quantification of the short-term market risk of its investments, were found to have errors which helped hide the true volatility and risk of these positions. Morgan has temporarily shifted back to an older version of its VAR model, which it believes to be more accurate in this regard, while working to correct the errors.
Further, Morgan says that the positions, which are proving difficult to unwind at a reasonable price, could produce another $1 billion of losses, depending on market movements. It is also possible that the volatility could work in Morgan’s favor, although the firm did not quantify the upside potential.
As a result of the losses, the Chief Investment Officer will be leaving the firm and it appears that some of the traders involved will also be exiting.
There will doubtless be many additional facts that come out in the weeks ahead, but these are the broad outlines of the events.
Who bears the loss?
The loss will be borne entirely by Morgan’s shareholders. As explained below, the loss is small compared to Morgan’s size and therefore there is no expectation that any external aid will be needed from taxpayers, depositors, creditors, customers or anyone else.
How big a loss is this? Did it represent a risk for the larger financial system?
Two billion dollars is a lot of money, but it is actually fairly small in relation to the size of Morgan. The firm earned about $25 billion pre-tax last year, so the $2 billion pre-tax loss is about one month of earnings. The loss is about a hundredth of Morgan’s $189 billion net worth, which represents the amount the shareholders have at stake, providing protection to depositors and creditors. It is roughly a thousandth of the firm’s $2.3 trillion of assets. Finally, the loss is roughly the same as one quarter’s worth of credit losses on Morgan’s loan business. Thus, the loss does not remotely present a threat to Morgan’s viability and even less so to the strength of the larger financial system.
Some have implied that the loss shows that little has been done to strengthen the U.S. financial system, but such ideas very seriously understate the extent of regulatory reforms and of market-driven changes. To take one very important example, Morgan alone added $52 billion in tangible common equity (a conservative measure of net worth that excludes the value of some assets that are hard to monetize) between the end of 2008 and last quarter, or 26 times the recent pre-tax loss and some higher multiple of the after-tax loss. This occurred despite maintaining nearly flat levels of total assets and risk-weighted assets over the period, meaning that the additional equity is available to cover roughly the same volume of risk.
Some of this increase in conservatism was due to voluntary choices by the firm and other parts reflected market pressure. However, some portion of this was doubtless due to anticipation of regulatory changes that are being phased in over time. For example, the US has agreed in principle to implement the provisions of the so-called Basel III global financial accord, which is to start taking effect in 2013. This accord very considerably increases the level of a bank’s net worth required to do the same volume of business as prior to this agreement. In particular, the largest banks, such as Morgan, will have to raise their levels of common equity by 2.5% of their risk-weighted assets purely as a function of their importance to the financial system, over and above the requirements common to all banks. In Morgan’s case, this stipulation adds almost $30 billion to their required net worth, enough to absorb perhaps twenty losses the size of the one just announced, after taking account of the reduction in income taxes.
At first glance, these measures of the small relative size of Morgan’s loss seem at odds with the drop in Morgan’s share price of over 9% as a result of the announcement of the $2 billion loss. However, the stock market was reacting to considerably more than just the direct loss. First, investors tend to be conservative and to pull back after any negative surprise, in part because there could be more surprises coming. Second, Morgan has been viewed by many in the market as the best-managed of the large U.S. banks, which was reflected in a premium market price relative to many other banks. Investors are doubtless less confident now of the degree to which Morgan is really better and therefore have reduced that premium. Indeed that same perception of excellence was helpful with customers and in the recruitment and retention of employees, so any diminishment of reputation creates harm. Third, there will be investigations and possibly lawsuits related to the loss, which will do further damage to Morgan. Fourth, rating agencies and creditors are likely to take a less positive view of Morgan going forward, increasing its funding costs and hampering its operations at least modestly. Fifth, Morgan under Jamie Dimon was viewed as the best placed of the large banks to make the industry’s case on regulatory reforms related to a bank’s trading activities, since the firm came out of the financial crisis in much better shape than many other banks. This level of credibility will take considerable time to recover, if it is ever fully recovered. The perceived scandal of the loss, combined with Morgan’s direct reduction in lobbying clout, significantly increases the chance of regulatory reforms that harm industry profitability.
Would the Volcker Rule have affected these trades, if the rule had been in force? Should it?
The Volcker Rule is the portion of Dodd-Frank that requires regulators to forbid banks from engaging in “proprietary trading.” The classic example of proprietary trading is when a bank has set up an in-house hedge fund that operates separately from its other trading operations with the intent purely to use the bank’s money to make profitable trading bets. However, the exact limits of proprietary trading are very ill-defined and overlap with a great deal of activity that is conducted as part of servicing clients, conducting hedging operations, ensuring an adequate liquidity reserve of high-quality securities in case of cash needs, and other activities that Congress did not intend to discourage. This inherent ambiguity and subjectivity is a major reason why I have personally opposed the Volcker Rule, since it is likely to considerably restrain desirable activities, unintentionally, raising the price of credit and other financial services. (See http://www.brookings.edu/research/testimony/2012/01/18-volcker-rule-elliott for testimony I gave on this topic. I prefer other approaches that deal directly with the risk of investment positions.)
The practical difficulties of operationalizing the Volcker Rule necessitated serious delays in implementation and a proposed rule of great length that included hundreds of questions for comment by interested parties. It is likely that the rule will be reworked very substantially based on the extensive comments that were received.
We cannot know whether the Volcker Rule would have applied to these transactions in any way, had it been already in force. First, we do not know what the rule will say in the end. Second, we do not know the details of the transactions by Morgan. Some of this will come out over time, but much of the detail may remain confidential between Morgan and its regulators, for competitive reasons.
The key conceptual point is probably whether the transactions would have been viewed as true hedges under the final version of the Volcker Rule. The rule is not intended to prevent banks from taking actions to mitigate their risk. Morgan has indicated that a flare-up of the Euro Crisis could have cost it a substantial amount of money in credit losses and that the money-losing positions were taken to offset that. In theory, this was an intelligent risk-reducing activity, if executed sensibly, as was apparently not the case here. Incompetence of execution would presumably not trigger the Volcker Rule, if the intent were truly to hedge.
However, the final version of the Volcker Rule might limit the ability of banks to act as Morgan did in one of two ways. First, there is the possibility that “portfolio hedges” will be forbidden. These are hedges which attempt to offset the aggregate risks of a whole portfolio of investment positions, as opposed to taking a series of specific hedges related to each position within those portfolios. Portfolio hedges can be substantially less expensive, but can run more risk of imperfectly offsetting the risks. Senators Merkley and Levin, two of the early proponents of what became the Volcker Rule, have proposed not allowing an exemption from the proprietary trading rules for portfolio hedges because they believe they would create major loopholes in the Volcker Rule, since any definition of them would be necessarily broad.
Second, the rules might be set to allow hedges only if there is a high probability that they will be very effective. For example, it might make economic sense to enter a hedging transaction that might not fully offset the potential losses or that has some probability of failing to work. This kind of partial protection may be substantially cheaper than paying up for a higher level of insurance. Banks do not, and should not, try to eliminate all the risks they take. They conduct a cost-benefit analysis to decide what risks can be cost-effectively offset, so price is always an issue. The final Volcker Rule might set standards that would fail to exempt such incomplete hedges, perhaps out of a fear of allowing loopholes in the trading rules if excessive flexibility is allowed.
It is still too early to judge what actually happened at Morgan, but there is the possibility that some of those doing the actual trading tried to use the excuse of hedging to take positions that would profit from expected market movements related to the Euro Crisis. This would be an example of the type of activity by traders or their managers that opponents of portfolio hedging fear would be used to circumvent the Volcker Rule. (Of course, the Volcker Rule would not have been the motivation in this instance, but rather working around internal management guidelines or constraints.)
The last several paragraphs of speculation assume that the relevant trades would not have fallen outside the Volcker Rule for some other reason. Certain types of proprietary trading are excluded from the Volcker Rule, such as those involving U.S. government bonds, and medium- and long-term transactions are also exempted. Some of these exemptions may be expanded, such as to include transactions involving highly creditworthy foreign government securities, which might conceivably have been used as part of the hedging against developments in the Euro Crisis.
In my view, it would be undesirable to have the Volcker Rule implemented in a way that made it excessively difficult to offset firmwide risks, such as those from the Euro Crisis. We would not want U.S. banks to be stuck with excessive exposures to certain major events simply because implementing the hedges became too difficult for regulatory reasons. At the same time, if we are serious about the Volcker Rule, despite concerns by myself and many others about its flaws, it will likely be necessary to put rules around hedging activities to avoid creating loopholes that allow large amounts of proprietary trading.
How is this relevant to other proposed structural changes to banks?
The Morgan debacle has given further life to other proposals to reduce the trading activities of banks, such as the proposal by Thomas Hoenig, former President of the Federal Reserve Bank of Kansas City and current FDIC director. That proposal would force banks almost entirely out of the business of trading financial instruments, producing a split something like the original Glass-Steagall Act did. The trading losses are also being used as ammunition by those who would like to break up the largest U.S. banks to reduce their size. A similar dynamic will occur in Europe’s debate as well, giving increased importance to the Liikanen commission that is studying potential structural changes to the banking industry there.
The $2 billion of trading losses are argued to show that (a) big banks can lose very large sums of money and (b) they do not understand the risks they are taking. The first point is self-evidently true, but is not necessarily important and should certainly not be news after the financial crisis. As described earlier for Morgan, sheer size means that it is not difficult for the largest banks to generate multi-billion dollar losses, although these more usually come from the loan book in a recession than from hedging operations. The real key is the level of difficulty in generating losses that are large enough to threaten the capital base of these immense organizations, which the $2 billion loss at Morgan did not remotely do. It is to defend against that possibility that the banking system, and big banks in particular, are being required to have much more net worth than they used to have. As noted for Morgan, one key regulatory provision alone will require that firm to increase its common equity by about $30 billion. Many other regulatory changes are embedded in Dodd-Frank and other measures in order to further bolster the banking system.
It is harder to measure the increased concern that we should have about whether the big banks know what they are doing and what level of risk they are carrying at any given point in time. The ability for a generally well-run bank to mess up by $2 billion does not necessarily mean that banks are vulnerable to a level of losses that would have systemic significance. However, it is certainly concerning to see these kind of losses result from misunderstandings and mistakes at a good bank. This leaves an unquantifiable, although likely low, probability that systemically threatening mistakes are being made at firms that are generally less well run. At a minimum, this debacle will doubtless lead to a useful review by banks and their regulators of trading and hedging activity across the board.
Were policymakers or politicians to conclude that the large banks do not understand the risks they take, it would certainly add impetus to consideration of structural measures to reduce bank complexity and/or size. I, personally, believe that there is no need for such more radical remedies and that the costs to the wider economy would be large, in terms of decreased credit availability and increased costs. In addition, the transition to such new arrangements would be difficult and painful, with a high probability of a period of reduced credit availability while things sorted out. That would be particularly painful if undertaken while our economy remains in a vulnerable state.
What does this tell us about communications and transparency issues?
In addition to the losses themselves, the Morgan debacle raises issues of communications and transparency. News stories began focusing on trading activities associated with the botched hedging operations roughly a month prior to the announcement of the $2 billion in losses. Morgan initially downplayed the stories and there is now reportedly an informal SEC investigation into whether Morgan failed to properly disclose what it knew in a timely manner. There may not be any policy implications about the communications pattern, since there are well-established laws and rules that require firms to disclose material information to shareholders about losses and risks. If there were violations, they can presumably be dealt with under current law.
The more interesting policy questions may be about transparency. As we learn more of the facts, it may become clear that the accounting and reporting for certain types of trading and hedging activities need to be done more effectively. However, it is difficult to draw any conclusions at this early stage.
In sum, if the facts were clear to Morgan, or should have been clear save for negligence, and should have been reported earlier, current law and regulation presumably provide the necessary remedies. If, however, the facts were unclear even to management because current accounting or reporting is insufficient, then we need to find better ways for both internal and external parties to track what is actually happening.