And I’m back! Here’s hoping that my prolonged absence has not cost me my entire audience.

The last substantive update on the Volcker Rule came from the WSJ a few weeks ago. The article contains an interesting — though not altogether surprising — nugget, which highlights one of the key issues in the proposed Volcker Rule. From the WSJ:

The SEC and a trio of banking regulators are butting heads over how to define the buying and selling of securities on behalf of clients, known as market-making, as well as over banks’ ability to invest in outside investment vehicles such as hedge funds, according to officials close to the discussions. Since brokers, which are overseen by the SEC, conduct market-making activities, the SEC is pushing for more influence over the issue, these people said.
So it’s the SEC vs. the banking regulators, and apparently on multiple fronts.

This divide makes sense with regard to the market-making exemption — the proposed Volcker Rule’s market-making exemption leans heavily on the SEC’s existing definition of “market maker.” In order to claim the market-making exemption, the bank’s trading activity must meet the proposed rule’s definition of “bona fide market making,” and the proposed rule explicitly states that “the Agencies expect to take an approach similar to that used by the SEC in the context of assessing whether a person is engaging in bona fide market making.” So it’s not hard to see why the SEC would be pushing for more influence over this issue.

However, the Volcker Rule’s market-making exemption will need to apply to a much larger range of financial products and markets than the SEC has ever had to apply its “market maker” definition to. Because all these markets have varying levels of liquidity, different trading infrastructures, etc., bona fide market making will look different in some markets than it does in others.

To address this, the proposed rule divides the universe of markets into (1) “relatively liquid” markets and (2) “less liquid” markets, and then broadly describes what legitimate market-making should look like in each.

Whether a particular market is put in the “relatively liquid” or “less liquid” bucket will matter a great deal, as it will be much easier for banks to claim the market-making exemption in less liquid markets. For example, in less liquid markets, banks will not have to demonstrate that their trading in that market “includes both purchases and sales in roughly comparable amounts” in order to claim the market-making exemption, which they will have to demonstrate in more liquid markets. In addition, in less liquid markets banks will not be required to “mak[e] continuous quotations that are at or near the market on both sides,” which, again, they will be required to do in more liquid markets.

So what is required for banks to claim the market-making exemption in less liquid markets? Essentially, banks just need to hold themselves out as willing to provide two-way quotes in that market on a regular basis, and be consistently active in that market. It’s fair to say that this is not hugely demanding.

Moreover, the proposed rule also states that bona fide market making in any market can include block trading when it’s done “for the purpose of intermediating customer trading.” Here, again, the proposed rule explicitly points to the SEC’s existing definition of “qualified block positioner” for guidance.

Now, it’s important to note that even within the “more liquid” and “less liquid” buckets, the regulators will almost certainly apply these criteria to each specific market differently — “as the facts and circumstances warrant,” as they say. But who gets to decide which markets go in which buckets, and how stringently to apply the market-making criteria to each market? Will the final rule add separate criteria for all the major markets, or will regulators apply the broad criteria on a case-by-case basis as questions arise.

My guess is that this is what the SEC and the banking regulators are fighting over.

In the wake of JPMorgan’s $2 billion trading loss, there’s been lots of talk about whether “portfolio hedging” is allowed under the Volcker Rule, and whether that should be changed. As I wrote in my previous post, the statutory language of the Volcker Rule very clearly allows portfolio hedging, and anyone who claims otherwise is lying to you. That’s just an objective fact, inconvenient though it may be for some people.

But the focus on portfolio hedging in the wake of JPMorgan’s trading loss is entirely misplaced. Portfolio hedging is only one of the seven criteria that a bank must meet in order to rely on the hedging exemption in the proposed Volcker Rule, and far from the most important. Commentators and certain politicians seem to believe that if JPMorgan’s trades met the definition of a “portfolio hedge,” then they would necessarily be allowed under the proposed Volcker Rule. That’s simply not true. Even if JPMorgan’s failed trades qualified as portfolio hedges, they would still have to meet other, more stringent requirements in order to qualify for the Volcker Rule’s hedging exemption.

What are the other criteria that a bank must meet in order to qualify for the proposed Volcker Rule’s hedging exemption? The first two have to do with the “programmatic compliance regime” that banks are required to establish under the Volcker Rule, so we can skip those for now. The third criterion deals with portfolio hedging — the trade has to mitigate specific risks, which can be done on a portfolio basis.

The fourth criterion is the most important, and it requires that the hedge “be reasonably correlated, based upon the facts and circumstances of the underlying and hedging positions ... to the risk or risks the transaction is intended to hedge or otherwise mitigate.”

Would JPMorgan’s trades have satisfied the requirement that they be “reasonably correlated” to the underlying risks being hedged? Maybe, maybe not. JPMorgan’s failed hedging strategy has been described several different ways in the press, so it’s impossible to say at this point. A lot depends on what specifically JPMorgan was trying to hedge in the first (original) leg of the hedging strategy, and what exposures the second leg of the strategy was intended to hedge. At first, the press was reporting that JPMorgan had been worried about weakness in the European economy, which led it — for whatever reason — to buy protection on the CDX.NA.IG.9 index. Now, I haven’t run the numbers, but I would question whether an index of investment-grade corporate credits is “reasonably correlated” to the European economy. So if those press reports are to true, then JPMorgan’s trade might have failed to qualify for the hedging exemption right there. But again, a lot of this depends on what JPMorgan was actually trying to hedge.

The fifth criterion, which is also crucially important, requires that the hedge “not give rise, at the inception of the hedge, to significant exposures that are not themselves hedged in a contemporaneous transaction.” JPMorgan’s trades could have failed to satisfy this requirement in two ways: first, when it initially bought protection on the IG.9; and second, when it later sold short-dated protection on the same index to hedge its original hedge. Why did JPMorgan have to hedge its original hedge? Was it because of changes in the economic outlook, or was it because they bought too much CDS protection initially? If it was the latter, then JPMorgan’s initial hedge may not have qualified for the Volcker Rule’s hedging exemption in the first place. The same analysis also applies to the second leg of the hedging strategy, in which JPMorgan reportedly ended up selling far too much CDS protection on the IG.9 index. If JPMorgan significantly over-hedged by selling too much CDS protection, then those trades also would not have been allowed under the Volcker Rule.

The sixth criterion requires that the hedge “be subject to continuing review, monitoring and management after the hedge position is established.” Jamie Dimon has all but admitted that the trades wouldn’t have satisfied this requirement.

Finally, the seventh criterion requires that “the compensation arrangements of persons performing the risk-mitigating hedging activities are designed not to reward proprietary risk-taking.” Now, I don’t know what Bruno Iksil and Achilles Macris’s compensation arrangements were, but I would actually be very surprised if their compensation arrangements didn’t reward proprietary risk-taking. Mostly this is because the Chief Investment Office (CIO) had something of a dual mandate — the CIO not only hedged the bank’s risks, but it also invested excess deposits, which is a risk-taking role. So I would suspect that compensation arrangements in the CIO accounted for their risk-taking mandate. If so, then the trades may have failed to qualify for the Volcker Rule’s hedging exemption for yet another reason.

The point of all this is that the debate over “portfolio hedging” is a complete sideshow — portfolio hedging is definitely allowed under the statutory language of the Volcker Rule, but this in no way means that the proposed hedging exemption is too weak, or has been “gutted.” JPMorgan’s trades may have qualified as “portfolio hedges,” but still may have failed to qualify for the Volcker Rule’s hedging exemption on at least four other grounds. The “reasonable correlation” and “no new exposures” requirements are far more important than whether portfolio hedging is allowed.

Commentators and journalists would do well to remember this.

Seriously, Jesse Eisinger needs to stop writing about the Volcker Rule, because he has absolutely no idea what he’s talking about. His latest article makes a number of egregious errors, but I want to focus on the one that’s most demonstrably untrue. Eisinger writes: (emphasis mine)

The Congressional authors of the Volcker Rule worried about this very thing [i.e., the potential to move prop trading into the bank’s treasury operation], and you can trace their concerns through their drafts. The original language of the rule had a broad exception: banks couldn’t trade for their own account, but they could hedge to mitigate their risks.

The authors quickly realized that the exemption was absurdly broad. After moving these businesses to other divisions, banks would then argue that their bets were either market-making activities or simply hedges that offset risks.
So Congress tightened the language. It wrote that the hedges had to be specific. When the Dodd-Frank financial reform law came out, the Volcker Rule provision defined “risk mitigating activities” as trades that were “designed to reduce the specific risks to the banking entity in connection with and related to such positions, contracts, or other holdings.” No macro-hedging, only micro-hedging. That is the will of Congress.
This is patently untrue on so many different levels. For one thing, Eisinger conveniently omits the first part of the hedging exemption, which explicitly allows portfolio hedging. But we’ll get to that. First, let’s do something that Eisinger clearly didn’t do — actually trace the hedging exemption through Congress’s drafts.

Here, in chronological order, are Congress’s main drafts of the hedging exemption in the Volcker Rule (with the key language that was added to each version in red):

Draft 1 — PROP Trading Act (S. 3098), introduced by Senators Merkley and Levin on March 10, 2010:
“(C) Risk-mitigating hedging activities.”
Draft 2SA 3931, introduced by Merkley and Levin on May 10, 2010:
“(C) Risk-mitigating hedging activities designed to reduce risks to the banking entity or nonbank financial company.”
Draft 3SA 4101, introduced by Merkley and Levin on May 18, 2010:
“(C) Risk-mitigating hedging activities designed to reduce the specific risks to a banking entity or nonbank financial company supervised by the Board.”
Draft 4 (Final Language) — Final conference report (H. Rept. 111-517), which was signed into law on July 21, 2010:
“(C) Risk-mitigating hedging activities in connection with and related to individual or aggregated positions, contracts, or other holdings of a banking entity that are designed to reduce the specific risks to the banking entity in connection with and related to such positions, contracts, or other holdings.”
Look at the difference between the last two drafts — as you can see, contra Eisinger, all of the changes in the final version substantially broadened the language of the hedging exemption.

The key is the language that Eisinger conveniently omitted in his article, which specifies that banks’ hedging can relate to “aggregated positions, contracts, or other holdings” — i.e., portfolio hedging. Note also that the final version clarifies that “specific risks” does not mean specific positions, because the “specific risks” that banks must be hedging can also relate to “aggregated positions, contracts, or other holdings.” The interest rate risk in a bank’s mortgage portfolio, for instance, is a “specific risk” that relates to aggregated positions. That’s what portfolio hedging is — it’s hedging the risks of the bank’s aggregate positions.

Eisinger clearly does not understand this distinction — he claims, falsely, that Congress “wrote that the hedges had to be specific.” Not true. Again, it’s the risks that have to be specific, and those risks can relate to aggregate positions.

This is exactly what Congress intended. All of this language about “aggregated positions” was added between Draft 3 and Draft 4 specifically to ensure that portfolio hedging would be allowed under the Volcker Rule. Indeed, there’s no other logical explanation for why Congress added the language about hedging “aggregated positions” between Draft 3 and Draft 4.

Thus, as you can see, it was clearly “the will of Congress” that portfolio hedging be allowed under the Volcker Rule.

So why did Eisinger claim that portfolio hedging is prohibited under the statute when it’s so clearly allowed? Probably because it allows him to self-righteously criticize both the banks and the regulators, and to present himself as far too knowledgeable and savvy to be fooled by all this fancy talk about “portfolio hedging.” (This kind of posturing is becoming increasingly common.)

But the reality is that portfolio hedging is 100% allowed by the Volcker Rule’s statutory language, and this is exactly what Congress intended — and anyone who tells you otherwise doesn’t know what they’re talking about.

In my previous post on the Volcker Rule, I admittedly glossed over the real issue in my second prediction. This issue is important, but it’s also reasonably complex (way too complex for some people, I suspect), and it requires some context and explanation to fully appreciate.

The statutory text of the Volcker Rule contained a glaring flaw: the statute prohibited proprietary trading by basically defining everything as “proprietary trading,” and then carving out exemptions for everything else. This was a colossal mistake for a variety of reasons, but the most important reason is that rather than having the regulators simply define “proprietary trading,” it put the regulators in the position of having to define every form of legitimate trading that banks do — underwriting, market-making, hedging, etc. Obviously, that’s a much, much more difficult task, and one that’s significantly more likely to lead to problems due to gaps — whether intended or unintended — in the proposed rule’s exemptions. It’s hardly a targeted solution to the problem of government-backed prop trading, to say the least. (I imagine the ultimate blame for this lies with someone in the Legislative Counsel’s office, although Merkley and Levin’s offices bear some blame here too, as they were clearly in way over their heads during this entire process.)

‘Flipping the Presumption’

The upshot of this is that it creates a presumption that all trades are prohibited prop trades, unless proven otherwise. What the banks want to do is to flip the presumption — instead of regulators scrutinizing whether each trade falls into one of the nine “permitted avtivities” exemptions, regulators would be scrutinzing whether each trade is a prohibited prop trade.

Whether flipping the presumption would dilute the strength of the ban on prop trading depends entirely on the quantitative and qualitative metrics that regulators ultimately use to identify prohibited prop trades. The right metrics would appropriately identify prohibited prop trades, while the wrong metrics could either identify too many trades as prop trades, or too few. But that’s where the debate would shift — or rather, should shift — if the regulators flipped the presumption. (The metrics described in the proposed Volcker Rule would, if anything, be overinclusive, and would require the regulators to apply some judgment to flagged trades — which I think is appropriate.)

Now, regulators only have so much discretion here. The statute is the statute, and flawed though it may be, regulators still have to work within its confines. But there are ways to effectively flip the presumption.

One way to do this is to define the main exemptions (market-making, hedging, and underwriting) very broadly, but include a carve-out for trades done for the “trading account” — which is, bizarrely, where the real definition of proprietary trading is located in the statute. The effect of this would be to allow regulators to focus on whether a bank’s trades exhibit the characteristics of trades done for the “trading account” (i.e., prop trades), based on the quantitative and qualitative metrics the regulators have identified, rather than focusing on whether each trade can fit into one of the defined exemptions. In other words, the presumption would be that a trade falls into the market-making or hedging exemptions, unless the regulators believe otherwise.

This is basically what I predicted the regulators would do in my previous post — although, crucially, I limited my prediction to the market-making exemption, and I said that the regulators would make this an “alternative” market-making test. A broader market-making exemption with a metrics-heavy carve-out for prohibited prop trading would go a long way toward: (a) alleviating concerns about the Volcker Rule’s impact on market-making without necessarily diluting the prop trading ban; and (b) making the regulators’ task a lot less daunting, and a lot less likely to cause unforeseen and unintended disruptions to the financial markets.

Since there's evidently a substantial amount of confusion surrounding Greece CDS procedures, I thought it would be helpful to post this:

Monday, February 20, 2012

Volcker Rule Predictions

The due date for comment letters on the proposed Volcker Rule has now passed. The comment letters are interesting for a variety of reasons. (I had the benefit of seeing draft versions of several of the comment letters, and I’ll just say that I’m amazed at the level of convergence they were able to achieve between the industry letters — both substantively and stylistically.)

Instead of trying to write a single, lengthy post with all my analysis of the competing arguments — an undertaking which I simply don’t have the time to complete — I’m going to sort of provide analysis as I go.

In my first post, I’m going to do something which, in my professional capacity, I never get to do: make predictions. I do this without any inside knowledge or the regulators’ thinking, and with full awareness that it’s far too early to know exactly what the regulators will do. But if you don’t make firm predictions, then how are you supposed to say “I told you so” when you end up being right? With that in mind, here’s what I think will ultimately happen to a few of the key portions of the Volcker Rule:

1. The requirement that market-making activities be “related to clear, demonstrable trading interest of clients” will be completely scrapped. The industry will win this one, and for good reason. The “clear, demonstrable trading interest” standard is simply inconsistent with the statutory text, which permits market-making activities that “are designed not to exceed the reasonably expected near term demands of clients.” Client demand can be “reasonably expected” well before the demand is “clear” and “demonstrable.” The statute permits market-making desks to trade in anticipation of “reasonably expected” near-term client demand, so applying a “clear, demonstrable trading interest” standard would be plainly inconsistent with the statute.

To be honest, I don’t think the regulators ever truly intended to apply this standard. The language about market-making activities being “related to clear, demonstrable trading interest of clients” was included in the criterion for bona fide market making, and NOT in the criterion for “reasonably expected near-term demands of clients,” which was far broader. It’s probably a fair bet that different people wrote those two sections, and they were never properly reconciled before the regulators published the proposed rule. However, I also think that the criterion for “reasonably expected near-term demands of clients” will be broadened as well. While that criterion is broader than the “clear, demonstrable trading interest” standard, there’s still a reasonably strong argument to be made that the language in that criterion is too narrow, and doesn’t reflect congressional intent.

2. Regulators will keep the market-making exemption in the proposed rule, but will add a broader “alternative” market-making exemption that relies more heavily on trading metrics to identify prohibited prop trading. Admittedly, this one is a longshot. The industry wants the regulators to replace their current market-making exemption with a much broader exemption, described in the main SIFMA/Clearing House letter:

“We believe a business should be viewed as customer-focused, and therefore engaged in market making, to the extent it is oriented to meeting customer demand throughout market cycles. This can be evidenced by, among other activity, a focus on offering execution to customers, building relationships with customers and providing sales coverage, providing research to customers and participating in the interdealer market in order to serve customer demand.”
Now, this is clearly way too broad. But it does have the benefit (and, from the regulators’ perspective, the attraction) of being simple and inclusive enough to serve as a uniform standard of market-making across all asset classes. If the regulators press on with their current definition of market-making, which contemplates different standards for each different asset class, then the regulators will be drawn into endless battles over what constitutes permitted market-making in every single asset class and market. That’s a daunting task, and I can’t imagine that the regulators are looking forward to writing 50 different, customized definitions of permitted market-making. I’m sure a simple, uniform definition of market-making for all asset classes will look pretty appealing.

However, because the definition of market-making that SIFMA et al. propose is clearly far too broad (sorry guys, but “providing research to customers” ≠ market-making), the regulators will still need some other, non-definitional way to identify prohibited prop trades. And that’s where the trading metrics come in. The proposed rule already identifies several quantitative metrics that can be used to reliably distinguish between market-making and prop trading, and can also be used across the different asset classes.

I think the bargain that the regulators will end up striking here is to add a broader, uniform market-making exemption that relies heavily on quantitative metrics as an alternative to the market-making exemption in the proposed rule. This alternative market-making exemption would obviously have to be narrower than the definition that SIFMA et al. propose, but would still be broad enough to encompass all legitimate market-making across different asset classes.

3. The “trading account” exemptions will stand. The proposed rule excludes repos, securities lending, and positions taken for bona fide liquidity management from the crucial definition of “trading account” — which effectively means that those positions are exempt from the Volcker Rule’s prop trading ban. While Merkley and Levin (amusingly) tried to argue in their comment letter that there is “no statutory basis” for these exclusions, they failed to offer any, you know, actual evidence for their claims. That’s usually fatal to an attempted legal argument. The group “Occupy the SEC1 even tried to describe various scenarios in which banks could exploit the “trading account” exemptions to put on prop trades, but their examples tended to be inaccurate (in that they would not have legitimately circumvented the prop trading ban), or ultimately irrelevant.

At best, I think the repo and securities lending exemptions might be slightly revised to include an explicit “anti-evasion” clause. But past that, I think all three “trading account” exemptions will stand.

4. The regulators will issue a re-proposed Volcker Rule rather than a final rule. I think the changes will ultimately be too significant to go straight to a final rule, and that regulators will want another notice-and-comment period to get feedback on any changes.


1 While “Occupy the SEC” should certainly be commended for engaging in the rulemaking process (which is where all the real action is), and for engaging in a substantive manner, I don’t think their Volcker Rule comment letter was terribly persuasive on any major point, and I don’t think it will ultimately result in many (or any) substantive changes to the final rule. Now, a lot of the letter’s problems were the result of this clearly being the authors’ first time writing a comment letter to banking regulators. So allow me to offer some constructive criticism (since I imagine there will be another notice-and-comment period for the Volcker Rule): For one thing, the letter contained far too many sweeping, conclusory statements, for instance about what did and did not contribute to the financial crisis, and these sweeping statements too often served as the sole basis for the group’s desired change. Regulators are not responsive to those kinds of arguments, to say the least. (In general, “a guy I read on Huffington Post said X contributed to the financial crisis” is not a winning argument at this level.)

Occupy the SEC’s letter also spent far too much time making tangential arguments that were often based on a simple misunderstanding of the proposed rule, which is a great way to kill a letter’s credibility and undermine its more legitimate arguments. Lehman’s use of Repo 105 was a scandal, but not one that was relevant to the proposed Volcker Rule. It’s absolutely imperative that you pick your battles in these comment letters — focus on the truly meaningful debates, and emphasize your strongest arguments. Also, your audience is the regulators, not other activists, so tone down the self-righteousness. Big-time. Regulators are professional civil servants, so they’re hardly responsive to letters that lecture them about their duty to the public.

Earlier this week, Joe Nocera wrote a puff piece on Karen Petrou of Federal Financial Analytics. In the piece, Nocera and Petrou repeat a frequently-heard — but nevertheless exceedingly superficial — argument that has always bothered me. From Nocera’s NYT column:

[Petrou] also points to a contradiction in the way the Too Big to Fail institutions are being dealt with. On the one hand, Dodd-Frank is very clear that if a big bank becomes insolvent, there can be no taxpayer bailout. It must be wound down, just like any other bank. Yet, at the same time, she says, the federal and international regulators are adding a host of special Too Big to Fail capital requirements and rules. “They are acting as if these institutions are still too big to fail. The two thrusts are incompatible.”
Oy. This is not a contradiction. Applying additional capital requirements and more stringent regulations to certain large financial institutions is absolutely not incompatible with ending Too Big to Fail — or with Dodd-Frank’s new resolution authority for large financial institutions.

The new resolution authority makes is easier for large financial institutions (known as SIFIs) to fail, which is the exact opposite of “acting as if these institutions are still too big to fail.” Moreover, the additional capital requirements and regulations for SIFIs in no way contradict the resolution authority. The purpose of the additional capital requirements and enhanced prudential regulations for SIFIs is to make it less likely that a SIFI will fail. The purpose of the resolution authority is to make sure that when a SIFI fails, it can do so without bringing down the entire financial system. These are not contradictory — they’re complementary.

These are pretty basic concepts of financial reform — I wrote a post way back in October 2009, before Congress even took up the financial reform bill, explaining how a SIFI regime and a new resolution authority would fit together. The fact that so many financially-focused pundits have yet to move beyond the “hey, if they’re ‘systemically important’ they must be TBTF!”-level of analysis is just sad.