Streetwise Professor

Nasdaq shut down for 3 hours yesterday, the second electronic trading SNAFU of the week (the first being the Goldman option fiasco).  The problem?  Another example of the effects of a SLOB (simulacrum of a limit order book).  One of the links in the US SLOB is the SIP (Security Information Processor) connecting NYSEArca and Nasdaq.  This sends information about quotes from one exchange to the other.  This information is necessary to link the markets: in order to prevent locks and crosses, and to direct orders to the best priced market, each venue needs to know the quotes at the others.  According to Nasdaq, irregularities in the SIP were causing problems with its system, so it shut down to prevent a more catastrophic failure.

Moral of the story: links are major points of failure in a SLOB.  (Though it is interesting to note that on the same day Globex was displaying crossed markets in the bean complex, and Globex is a true CLOB not a SLOB.  But trading continued.)

My first thought on hearing of the shutdown was a SLOB problem.  My second was hacking.  There has been a spate of stories of late about exchanges being major potential targets of hackers/cybercriminals.

This is one of the reasons why I take an interest in Appelbaum, Snowden, Anonymous, etc.  Exchanges, and financial infrastructure generally, are major targets of these psychopaths for at least a couple of reasons.  First, some are just criminals, who like Willie Sutton target banks and the like because that’s where the money is.  Second, the hacker movement is a hotbed of anti-capitalist ideology that motivates attacks on capitalist institutions.

Meaning that the defenses of our financial infrastructure are vitally important.  And this is another source of major concern, which the Snowden affair puts front and center.  Infosec workers are recruited from the same milieu as the hackers.  The membrane between hackers on the outside and supposed defenders on the inside is very permeable.  There is lots of talk about the color of hacker hats, but ethically and ideologically the distinctions between black, white, and grey hats are quite blurry, and the boundaries are transgressed often.

Snowden is a classic example of that.  Former head of the NSA, General Hayden, said that perforce NSA and the government generally has to recruit from this group, but find ways to protect from those with a “romantic attachment to absolute transparency at all costs.”  Yes, the transparency romantics are a potential problem, but I think they’re the least of our worries: transparency romanticism is often a public pose for those motivated by malign anti-US and anti-capitalist beliefs.

The narcissism and grandiosity that Snowden exhibits are also all too common in this community, both inside government and financial institutions, and outside.  Indeed, there is a very strong and widespread tendency of these people to view themselves as superior in some way, and justified in acting as judge, jury, and executioner when it comes to the release of information, and operations against targets that violate their highly tuned-and self-righteous-moral senses.  Again, Snowden is a perfect example: he arrogated to himself the ability to decide what was illegal and immoral.  So did Bradley Manning, though he (and yes, he is still a he and he was a he when he made this statement, so I’m passing on the Chelsea thing) repudiated that in his pre-sentencing statement.

Indeed, this is a common theme in the infosec community.  Take a look at the attached video by a Pied Piper of this community, ex-priest (and, alas, University of Chicago grad) Richard Thieme.  (H/t @libertylynx, who also raised many of the points below.) This guy is lionized in this community.  He matters.  He is viewed as the voice of many. Check out the fawning, personality cultish comments to the video.

Thieme basically asserts that infosec specialists answer to a higher law, in part because there is a nexus between law enforcement and criminality, and in part because traditional legal distinctions (e.g., foreign vs. domestic) have been obliterated by technology and shadowy, scary supra-state actors.  Check out the rants around minute 32:00.  And for a dose of virulent anti-bank rhetoric, which really has to be heard to be believed, fast forward to minute 43 or so.

You know that there are infosec people in banks-and exchanges-who idolize this guy, and buy into his message.  The message is banks are criminal.  The government is criminal.  So it’s not criminal to attack them. Indeed, one can be justified in doing so.  Yes, many of the would-be attackers are on the outside, but as Snowden shows, there are inevitably many on the inside. And some on the inside have no problem in dealing with Appelbaum and the like. Case in point, an infosec guy I interacted with on Twitter until he blocked me. (Thin skin seems to be another occupational hazard.) He would countenance no criticism of Appelbaum.

Meaning that the most dangerous threats to exchanges and financial infrastructure may be those who are hired to defend them.

So this is one of the reasons I write about Snowden, and Appelbaum, and Assange, and Greenwald, and the rest of this crowd. Their agendas and connections need to be understood-and broadcast. For these avatars of transparency are notoriously protective of their own activities and connections. Given the inevitability that some of those to whom they are connected are on the inside of vital financial and government institutions, this is a matter of grave concern. And a matter that mainstream journalism has been shockingly negligent in investigating and reporting.

Appelbaum and Poitras are in Berlin now, and Berlin-land of the Pirate Party-is arguably the epicenter of this community. But it is an international community, and what happens in Berlin doesn’t stay in Berlin. It could be coming to a bank or exchange near you. If it hasn’t happened yet, it’s not for lack of trying, or the lack of potential acolytes within these very institutions.

The CME has seen increased volumes in its cash settled Brent contract in the aftermath of its launch of a program that pays rewards to those who trade the contract.  This is a standard way to attract volume to a contract competing against an incumbent/dominant contract.  Indeed, it’s probably the only way to do it.

Eurex tipped the Bund market away from LIFFE in 1998 after implementing a fee holiday, which LIFFE smugly refused to match, so confident was it of its liquidity advantage.  But Eurex had enough organic order flow that it was fairly liquid already, and the fee cut made it cheaper to trade Eurex Bunds, as I document in my paper “Bund for Glory: Or, It’s a Long Way to Tip a Market” (which will be a chapter in a book I’m currently writing).  The market tipped within a period of mere months.

Other exchanges didn’t repeat LIFFE’s near fatal mistake, and responded to fee cutting entrants by cutting their own fees to match.  This is how CBOT saw off Eurex’s attempt to enter the US Treasury futures market in 2003, and how CME torpedoed EuronextLIFFE’s attempt to enter into the Eurodollar futures market a few years later.

Here’s the deal: incumbents have a huge liquidity advantage that entrants must overcome.  How can this be done?  The only way, really, is to cut fees, or actively reward volume to attract enough liquidity to compete with the incumbent.  And the odds are still long, especially if the incumbent follows the CBOT script, and matches the reduced fees.

But if you want to encourage inter-exchange competition, you have to look favorably on entrants using strategies that reward volume as a way of building liquidity to make themselves into viable competitors.

This basic fact makes this FT article very disappointing.   It criticizes the CME initiative, and only quotes those who are critical or dismissive of the endeavor.  Yeah, it’s probably true that the only reason why those who participate in the incentive program trade is to collect the rewards.  But their order flow can attract order flow from others: other traders see that there is two-way activity, and can be more confident that if they establish a position, it won’t be a Hotel California that they can check into, but can’t leave.  Meaning that it’s BS to say, as one anonymous source to the story says, “there is no economic benefit to the trades whatsoever.”  The economic benefit is that it can make the entrant a more viable competitor to the incumbent.

And quoting Bart Chilton.  Please.

The quote that claims that spread trades-which dominate the activity in the CME Brent contract-“make no loss because they buy and sell at the same price” is just a crock.  Yes, spread trades are lower risk, but calendar spreads do move.

The CME is engaging in a penetration pricing strategy.  The only real strategy that permits competition against an incumbent.  If you like competition, you should like a strategy that harnesses the greed of trading firms to generate liquidity and volume.  Because, quite frankly, that’s the only way to compete.

But if you like monopoly, go ahead and denigrate what the CME is doing.

It would also be interesting to see how ICE built market share in its WTI cash settled contract that competes with the CME’s flagship oil contract.  It is my recollection that ICE rebated fees for firms that brought volume to the ICE contract.  (The structure of the ICE inducement, and Eurex’s in Bunds, were interesting.  They basically made it a tournament, with rebates/bonus payment being limited to the top three-if memory serves- order flow providers.  That provides a very high powered incentive while controlling the cost of the program.)  If my recollection is correct, doesn’t that mean that sauce for the goose is sauce for the gander?

Competition in futures markets, where order flow is not socialized, is highly imperfect.  Incumbents have huge advantages.  Aggressive price competition-pentration pricing, specifically-is about the only way that an entrant can hope to dent the incumbent monopolist’s huge liquidity advantage, and even that is a highly dicey proposition.  The FT story would have been much better had it recognized this reality rather than taking an all too critical view, and containing only quotes that were no doubt music to Jeff Sprecher’s ears.

I just read Haim Bodek’s “The Problem of HFT.”  It reinforced my belief in what I’d long concluded, namely, that the problematic aspects of HFT result from the decision of the SEC to create a Simulacrum Limit Order Book-a SLOB-rather than a true Central Limit Order Book-a CLOB-when it promulgated RegNMS.  As I noted at the time, the SEC chose an “information and linkages” approach to create a National Market System that consisted of interconnected competing exchanges, rather than to create a single CLOB.  In achieving competition, it was hugely successful.  Whereas NYSE dominated trading in its listings pre-RegNMS (with a market share in the low-to-mid 80s), now executions are divided rather equally among a handful of exchanges.  There are few rents in securities execution in the US anymore, in contrast to the old floor days.  NYSE’s equity business contributed relatively little to the price paid by ICE for NYSEuronextLiffe.

Part of the decision to  was political economy: politically, the SEC couldn’t have realistically favored one venue, or displaced existing exchanges by creating a new national CLOB.  Part of this decision was predicated on a defensible and understandable desire to encourage competition in securities trading, in fees, rules, liquidity, and technology.  However, the SEC didn’t anticipate how the market structure would evolve, and in particular, how HFT would transform securities trading.  I can’t say that anybody did.  I certainly didn’t.  I thought that automated trading would link markets, but I didn’t see it playing out like it did.

I call what the SEC wrought a “simulacrum” order book because although it incorporates some aspects of a central limit order book, specifically price priority across exchanges for the top of book, it lacks several crucial features of a true CLOB.  Price priority is for top of book only.  Time priority (or other secondary priorities) are not enforced across trading venues.  Markets can cross or lock-which can’t happen in a CLOB, where locked or crossed orders execute automatically.  Moreover, latencies, particularly in the public feed of the NBBO create behaviors and incentives that diverge from a CLOB.

Because it links multiple execution venues, the SLOB has many seams.  And one thing we know is that traders traders look for and exploit seams like Tom Brady looks for and exploits the seams in a zone defense.  Bodek’s little collection of essays goes into considerable detail on how traders have worked the seams, and how their methods have changes in the last 6 years.

The most critical seam in the SLOB, in Bodek’s accounting, is the rule on locked and crossed markets.  The SEC rules state that an order on exchange A that locks or crosses an order on exchange B cannot be displayed.  This complicated the task of quoting aggressively: an HFT that submitted an order that locked or crossed a market would not be able to establish position at the head of the queue.  To address this, traders first used spam and cancel: repeatedly sending orders that were canceled if they locked or crossed, but which would go to the front of the queue as soon as the quote in the away market changed.  Exchanges had implemented algorithms that re-priced orders that locked or crossed away markets, and put them at the end of the price queue.  HFTs circumvented this using hidden order types that were not displayed if they locked or crossed, but which were displayed-and crucially went to the head of the queue at the new price-the instant their price would no longer lock or cross an away market.  HFTs also used sweep orders to establish priority.

The common features behind all of the things Bodek criticizes are that (a) they are all intended to establish priority, and facilitate aggressive quoting, and (b) they can happen in a SLOB, but not a CLOB.  They give an advantage to HFT firms that invest in the technology and in understanding the order types, to the detriment of those who don’t (including many institutional investors): this is Bodek’s main beef.  They are an artifact of the prohibition on locks and crosses, and this can only occur in a fragmented market structure that precludes locks and crosses.  If locking and crossing orders were matched against one another and executed, as would happen in a true CLOB, much of the gaming of order types Bodek laments would go away: I suspect, however, that latencies across a system of linked exchanges would make this impractical.

This relates to, and complicates, points I made in my writing on exchange competition in the late-90s and early-00s.  Competition between privately operating CLOBs with no obligation to route orders to markets displaying better prices is likely to result in the survival of a single dominant platform that exercises market power.  This is arguably the case in futures markets.  A CLOB utility that is regulated to mitigate this market power problem raises myriad difficulties.  Who will operate it?  How will rates be regulated?  How will access be regulated?  (One way to exercise market power is to limit access to the CLOB.)  How will the CLOB utility be governed?  How is capture (by major intermediaries, for instance) to be avoided?  How can the CLOB be incentivized to be technologically innovative, rather than a lazy monopolist (especially since some technological changes may be very disruptive to the business models of important market participants)?

Socializing order flows by requiring exchanges to route orders to other markets displaying better prices encourages competition, and seemingly avoids the problems posed by private CLOBs and utility CLOBs, but raises its own difficulties.  Indeed, these are the problems that we are experiencing today, and which Bodek identifies.  This raises the question: can we clean up SLOBs so they are more truly CLOB-like, while retaining interexchange competition that mitigates market power problems?  The answer to this question has both technological and political components.   A system of linked exchanges necessarily has seams and latencies and inter-market coordination issues absent in a true CLOB, and these are vulnerable to gaming and breakdown, and even during normal operation represent a more complicated and error-prone way of allowing all orders to interact with one another than a true CLOB.  The political components arise from the fact that interested parties, most notably exchanges, major sell side intermediaries, and big buy side institutions, will strive to shape the system’s design and rules to advance their interests.

The trade off is therefore pretty stark: it’s a pick-your-poison kind of thing.  Don’t socialize order flow and live with market power, or socialize order flow to facilitate competition, and grapple with the technological challenges of creating a virtual CLOB, and fighting the never ending battle against attempts to exploit the seams.  I don’t claim to know the right choice, but I think that framing the issues in this way will permit far more constructive debate than the one currently ongoing.  This debate has become Manichean: good vs. evil: black hats vs. white hats.  It is oddly ahistorical, in its failure to recognize the problems with the previous system.

Perfect is not an option here.  Which is why debates over securities and derivatives market structure have been raging for as long as securities and futures have been traded.  The debate would be more productive-less sterile-if the choices were framed more clearly.  I’ll keep trying.

Richard Epstein has written about “government by waiver,” and its pernicious effects.  We’ve had numerous examples of late, from the Obamacare waivers, to the suspension of the employer mandate, to yesterday’s decision to exempt Congressional staff from Obamacare.  Yesterday also brought news of another administrative version of Emily Litella’s (and Nirvana’s!) “Nevermind”: the EPA is easing renewable fuel standards that had wreaked havoc in the gasoline market, and driven prices for Renewable Information Numbers “RINs” from pennies to nearly a dollar per gallon:

The Environmental Protection Agency, which administers the mandate, has said it will reduce total required volumes in its forthcoming proposal for 2014. It also extended the compliance deadline for 2013 by four months to June 30 of next year.

We’ve seen this movie before.  Congress passes expansive feel-good legislation that is at complete odds with economic reality.  Those who have to deal with economic reality importune the executive branch and administrative agencies for relief, and some sort of relief is granted.  (As another example, CFTC has also granted various forms of relief for some of the Frankendodd craziness.)

In the case of RINs, Congress demanded that refiners utilize far more renewable fuels (such as ethanol) than is feasible, given the installed base of auto engines which can only utilize 10 percent ethanol fuels.  Refiners could purchase unused renewable fuel credits (RINs) accumulated in past years when renewables usage exceed the then-lower mandated levels, but these are running out, and the renewable requirement was increasing even more for 2014.  Thus, the price of RINs had rocketed higher and there was risk that there wouldn’t be near enough for 2014.  The regulations also favored exports of gasoline because exported gas does not count towards renewables requirements.

So EPA stepped in and gave relief.  Because Congress couldn’t be bothered to worry its little head about minor details like the ability of the engines in hundreds of millions of autos to use the amount of ethanol it, in its infinite wisdom, decreed.

This is perverse in many ways.  It encourages rent seeking, at times to the point of corruption, or at least highly dubious political favoritism and dealing.  It empowers unelected bureaucrats who are difficult enough to hold accountable as it is, and who are often pursuing ideological agendas, and who can use their leverage to advance those agendas.  It allows Congress to avoid being held responsible for, let alone accountable for, its careless legislating: why take care in attempting to anticipate perverse effects when there’s someone there to clean up your mess?  It creates substantial uncertainty that discourages investment and hiring.  We end up with a thicket of ad hoc responses to legislative pipe dreams that lead to further problems that lead to more ad hoc responses and on and on.

In some respects, the RIN fiasco is the least of the renewable mandate’s sins.  Ethanol is a disaster on many dimensions, economic and environmental.  These problems are bad enough, but the EPA’s action is emblematic of a broader problem with the mega-legislation of the Bush-Obama years.  The legislation is unworkable, and compels delegation of considerable discretion to largely unaccountable administrators.  It undermines the rule of law and makes the US more like a natural state-or banana republic.

It’s all so confusing.  During the last spasm of Congressional criticism of the “financialization” of commodity trading, mainly during the Frankendodd “debates” over position limits, the target of criticism was evil speculators who trade commodities for purely financial reasons, and who have no involvement in the physical markets.

In the latest incarnation, the alleged problem is that financial firms that trade commodities paper do have physical commodity operations:

What concerns Brown and Senator Jeff Merkley, Democrat-Oregon, is that banks have numerous physical commodity holdings as well as the means to deliver those commodities, including oil tankers, transmission rights and pipelines, while they at the same time trade on energy prices in futures and swaps markets.

Big banks can buy “vast quantities” of commodities and pipelines, tankers and warehouses, said Merkley. This gives them “a huge amount of market information that’s very advantageous in trading” and gives them a “thumb on the scale in the terms of supply and demand and being able to have some influence over the price,” he said.

“If you’re simultaneously allowed to bet on the price and you’re allowed to have your thumb on the scale in reflecting the price it’s a huge conflict of interest,” he said.

“Financial institutions own fleets of oil tankers and can withhold delivery, while wagering on the price of oil or they can speculate on the price of energy while controlling supply,” Brown said.

This, the senators said, is a clear conflict of interest that is driving up commodity prices.

Make up your minds, people (which generously assumes you have them).  Is the combination of physical and financial trading a good thing, or a bad thing?

As someone who has arguably written more about commodity market manipulation than anyone alive, I can say definitively that the most costly manipulative strategies-corners and squeezes-involve trading of “paper” (e.g., futures contracts) and the physical commodity.  These strategies involve either taking excessive deliveries, or buying in excessive quantities in the physical market.  Engaging in a market power manipulation requires some involvement in the physical market.

Historically, many of the biggest manipulations have been undertaken by hedgers.

And that’s the key point that escapes the Washington hand-wringers.  On one of their wringing hands, they lament the involvement of non-hedgers in markets (let’s leave aside how you can have hedgers without speculators to take the other side of trades).  On the other, they bemoan the fact that those involved in both the physical and financial commodity markets-who are often hedgers-can sometimes exploit market power to enhance profits.  But if you believe that a primary function of commodity derivatives markets is to facilitate risk management by those involved in the producing, marketing, storing, and processing of commodities, you have to permit the physical traders to trade derivatives.

One of the consequences of that is that there will be manipulations by physical players.  But severing physical and financial trading in order to prevent manipulation, which is what Brown and Merkley and others are suggesting, is crack-brained.  It throws the hedging baby out with the manipulative bath.

And it is particularly crack-brained to focus on just one category of market participants, like banks.  The potential for manipulation exists due to frictions in the physical market, and the ability of some market participants to trade both derivatives and the underlying commodities.  Someone will step into the banks’ role if they are precluded from the market, and the potential for manipulation will remain.  Taken to its logical conclusion, the Brown-Merkley “conflict of interest” standard would preclude any entity-not just banks-from trading both commodity derivatives and the underlying physicals.

So what’s the best way to attack manipulation?  Not by destroying the ability of the markets to perform their vital risk transfer function.  Instead, as I’ve argued for going on 20 years, the best way is to impose penalties after the fact for manipulative conduct.  Thus, at least conceptually, I support FERC and CFTC enforcement efforts.  The problem is that practically their efforts are often misdirected and intellectually confused.

Congress should economize on its limited comprehension of these issues by focusing on the key problems, rather than throw around vapid phrases like “conflict of interest” and recommend policies that undermine the ability of financial and physical commodity markets to function effectively.  Better anti-manipulation laws predicated on a clear understanding of how manipulation works would be the place to start.  And no, this does not involve the Xeroxing of 10(b)(5) of the Securities and Exchange Acts and inserting it into commodity trading laws, which has been Congress’s default response for about a decade.  There is a complete mis-match between the fraud-based thrust of 10(b)(5) and the market power problems that sometimes arise in commodity markets.  I’ve made recommendations over the years on the right regulatory/legislative approach that would target market power problems more precisely.  This approach would permit the markets to exploit the synergies between paper and physical trading while at the same time improving the efficiency of these markets.

But instead, we get Congressional schizophrenia and meat cleaver approaches that would, if taken to their logical conclusion, destroy the markets in order to save them.

A further thought.  Those who have the ability to trade the physical, and who have extensive information and developed logistical capabilities are usually the best positioned to break corners, and to mitigate the frictions that make corners possible.  They also have the strongest incentive when they have a derivatives position that would be hurt by a manipulation.

I’ll post a link when I have a chance, but P.D. Armor’s breaking of the Leiter wheat corner in 1898 is a great example of this.  There are many others.

No sooner than I had written a post that showed the potential comparative advantage that banks had in dealing in physical commodities did JP Morgan Chase announce that it was throwing in the towel, and exiting the business:

Wall Street’s biggest bank said an “internal review” had concluded it should pursue “strategic alternatives” for its physical commodities operations, which includes assets like its Henry Bath metals warehousing subsidiary and a vast global team trading everything from African crude to Canadian natural gas.

The firm will explore “a sale, spinoff or strategic partnership” for its physical arm, the statement said. It said the bank remained “fully committed” to its traditional financial commodity business, including trading derivatives and its activities in precious metals.

This decision was not the result of any single thing, but the cumulative result of many things.  Greater capital requirements for banks.  The maturation of commodity markets and the resulting flattening out of profits.  The reputational risks of commodities dealing, with multiple regulators on the hunt for manipulation cases, and asking for big dollars to settle them.  And, most recently, the political heat directed at banks’ commodity trading operations.  Morgan’s commodity trading operation was big absolutely, but relatively not as big as Morgan Stanley’s or Goldman’s, and hence more readily expendable.  Morgan decided to get out of the kitchen, rather than stand the heat.  Especially since there is political pressure on all aspects on the banking business, and since navigating the post-Frankendodd, Volcker Rule world will be sufficiently challenging to the core businesses that commodities risks, political and reputational more than financial, just aren’t worth it.

Morgan Stanley has already mooted selling off its commodity arm.  It will be interesting to see what Goldman decides to do, especially in light of the Metro melodrama.  The European banks (notably Deutsche and Barclays) also have some thinking to do.

The exit or downsizing of bank commodity operations will redound to the benefit of the trading houses, and to private equity shops and perhaps hedge funds, including potentially some new entrants.  I doubt it will make the commodities markets more competitive, and indeed, the opposite is more likely to be the case. This turns my minds to thoughts regarding the political economy of the pressure on the banks.  The exit and downsizing also run contrary to the meme that banks have it all their way inside the Beltway.

The banks have done a terrible job of making their case.  Perhaps it would be more accurate to say that they really haven’t made a case at all, likely as a result of hubris.  They know they are doing God’s work (to steal from an old commodities guy, Lloyd Blankfein, who started his career at J Aron), but really need to work on their evangelism.

There is indeed a karmic aspect to this.  Almost instant karma, in fact: following my musing about that by only a few days.  I don’t say that with the least tinge of schadenfreude.  Just stating the obvious.

I wonder what will happen to Blythe Masters, the head of Morgan’s commodity operation.  I know her some.  Most of our interactions were positive, until we butted heads over the report I wrote on commodity trading firms: in fact, she approached me to write the report because she’d liked what I’d written on the blog, and in comments to the CFTC.

I have a distinct recollection of sitting next to her at the JPM commodities conference in October, 2011, and watching the muscles in her jaw bulge as she ground her teeth furiously.  She was obviously very tightly wound.  If she was that intense before a relatively low-key conference presentation, I can only imagine how her jaw is working now.  My guess is that she will stay at Morgan for a decent interval, and then land somewhere else.  Like perhaps a private equity commodity trading shop.

Exit for economic reasons is part of the market process.  Exit driven by political or regulatory pressure is problematic.  Both factors are at work now, but unfortunately political and regulatory issues are predominating.  The result will be less efficient, less competitive commodity markets.  Which is exactly the opposite result that the politicians and regulators baying at banks and nipping at their heels claim to want.  Unintended consequences yet again.

In its 2003 Physical Commodities Trading Order, the Fed permitted bank holding companies (not the depository institutions under the BHC umbrella) to purchase and sell physical commodities because they are “meaningfully connected to a financial activity such that it complements the financial activity.”

I mentioned one example of this the other day in the “Get Physical” post (and don’t worry: I won’t post any Madonna videos!), but want to go  into more detail.  The example is offtake agreements (and relatedly, tolling deals).  In these arrangements, a bank (or other counterparty) agrees to supply inputs (e.g., crude to a refinery, gas to a power plant) and receives delivery of outputs (e.g., gasoline and heating oil, electricity).

Refineries, power plants, and the like typically need to pay for the inputs they process before they receive payment for their outputs.  This creates a need for working capital to finance the timing gap between cash outflows and inflows.

Providing funding for working capital is clearly a traditional banking activity.  One way to do this is for the bank to provide a loan or credit facility, and leave the refiner or power plant to acquire inputs and market outputs, and bear and perhaps manage the price and operational risks associated with those activities.

This exposes the bank providing the funds to risk: adverse movements in prices could put the refiner or generator into financial distress, and perhaps cause a default.  The bank could require the borrower to hedge, but there is a moral hazard: if it doesn’t, or doesn’t do it effectively, the lender bears risk.  The bank can monitor, but this is costly, and often imperfect.

The moral hazard problem can be eliminated by passing the risk on to the lender.  An uptake agreement or tolling deal does this.  These types of deal implicitly provide funding to bride the outflow-inflow gap, and pass the price risks back to the lender.  The lender can manage these risks, and the agency cost in this arrangement is lower: there’s no moral hazard, and no need to monitor.  Thus, bundling price risk management and funding can reduce the cost of funding working capital needs.  This is presumably more valuable for lower credit quality refiners and generators.

There are other potential benefits.  The lender may have a comparative advantage in managing risk due to specialization and expertise in this function: banks and commodity trading firms have a comparative advantage in risk management.  Moreover, they may able to be able to manage risk more cheaply because they run large books: there are economies of scope in risk management.  For instance, a lender doing an uptake deal with a refinery is short crude and long products, but it might have a long crude position based on a trade it executed with producer, and might have a short products position as the result of a swap with an airline or heating oil dealer.  These natural hedges reduce the amount of trading necessary to manage the risks.

Moreover, trading firms specialize in marketing and logistics, and there are scale economies and scope economies in these activities.  It may be cheaper for a big trading shop to provide marketing and logistical services, thereby eliminating the need for the refiner or the power plant to pay the overhead associated with such activities.

Thus, there are strong complementarities that make it beneficial to bundle financing, logistical, and marketing activities for some firms that process commodities.  Who should provide these bundles?  Banks have comparative advantages in funding and risk management, and can develop and staff physical trading operations.  Commodity trading firms are typically more expensive suppliers of financing, but they have strong competencies in marketing and logistics.  So both banks and commodity traders are reasonable suppliers of the bundled services, and thus reasonable counterparties in offtake and tolling deals.

What would be the effect of banning some market participants, like banks, from these activities?  Econ 101 gives the answer.  If these entities currently supply the bundled service-as is the case today-elminating them from the market will shift up the supply curve for this service, and in equilibrium it will be more costly.  Meaning that refiners and power plants will incur higher costs that will be passed upstream (lower prices for inputs) and downstream (higher prices for outputs).  To the extent that restricting participation gives (or enhances) the market power of those who remain in the market, these effects will be exacerbated.

What if some market participants-such as banks-get a subsidy of some sort that allows them to offer these services below cost? That is a source of inefficiency, to be sure: too much of the service is provided in equilibrium, and the beneficiaries of the subsidy have an excessively large market share.  But it does not follow that banning the subsidized from the market is the right policy response.  It’s usually better to eliminate the subsidy, especially in the case of something like a too big to fail subsidy, which has pervasive and perverse effects in all the markets in which the subsidized banks operate.

Evaluating the regulation of bank participation in any market, including the commodity markets, should proceed along similar lines.  The concept of complementarity is the right way to consider these issues, and that principle can be analyzed in particular instances using basic economic analysis.  The survival principle also suggests that if a firm gets market share in a particular business, it is because it is providing that service at lower cost, or delivering higher quality.  If the lower cost is an artifact of some sort of subsidy, eliminate the subsidy, especially when it is a pervasive on that cuts across myriad lines of business.  Populist venting about banks, especially venting based on some cramped understanding of what banking is and predicated on some belief that the involvement of financial institutions in commodity markets is somehow unnatural, is counterproductive.  It is also likely really a mask for some political economy agenda, an effort to benefit one group of firms by handicapping-or eliminating-others.

Earlier in the year I wrote about the obscure, but very important, regulation of futures commissions merchants that essentially requires them to hold enough of their own money to cover shortfalls in customer margin accounts.  This “residual interest” rule has caused a wholesale freakout among FCMs and market participants, and for good reason, as I pointed out in that post.  The rule would not have prevented an MF Global or Peregrine situation-those involved fraud and/or numerous rule violations.  Moreover, the rule puts further strains on the funding liquidity of the futures markets.  There are already new strains aplenty, and liquidity shortages are a major source of systemic risk.  A rule ostensibly designed to protects customers puts the entire system at greater risk.

The near universal opposition of the industry is leaving the CFTC unmoved.  But the CFTC’s response is purely legalistic, and does not even attempt to address the fundamental economic issues:

But on Thursday, at an open meeting of the Agricultural Advisory Committee to the CFTC, regulators were unmoved by the criticism.

The proposed rules, said Ananda Radhakrishnan, director of the CFTC’s division of clearing and risk, merely clarify what the law already says: funds from one customer must not be used to pay for the position or deficit of another customer

“Nobody’s been able to make the argument, with all due respect, that what we are suggesting is not what the law says it is,” Radhakrishnan said. “The arguments we’ve heard … (are) that it’s going to be expensive, the earth is going to fall and so on and so forth. But nobody has done, to my view, a legal analysis saying, ‘your analysis is wrong.'”

Translation: “The law may well be an ass, but we’re rolling with it.”

I have some sympathy, because Congress has handed the CFTC many legislative directives that are at odds with economic common sense, and which are actually contrary to the purposes of the legislation (notably, the reduction of systemic risk).  But where there’s a will, there’s a way.  I get the sense that the CFTC is just hiding behind the legalisms in order to force through a proposal that it-and most likely Gensler-wants.  If it had the will, it could likely find a way to adhere to the law as written but substantially mitigate the negative impacts of the regulation on liquidity, costs, and systemic risk.  The failure to do so, and the appeal to the “Congress made me do it” defense (seen so clearly in the position limits regulation, and repeated in its appeal briefs) strongly suggests that the CFTC doesn’t have a strong economic justification for its regulation, can’t rebut the arguments of the critics, and is playing legal rope-a-dope in order to force through one of Gensler’s (and perhaps the staff’s) bright ideas/obsessions.

Who woulda thunk that commodities-especially ones like aluminum-would become such a huge story.  Maybe it’s the summer news doldrums, but in the last week there have been two major, related stories using up loads of ink and pixels.  The stories are: (1) the Federal Reserve’s announcement that it is evaluating whether to continue permitting banks it regulates trade physical commodities, and own physical assets like storage facilities and power plants used to transform commodities in space, time, or form, and (2) the role of banks like Goldman and Morgan in the industrial metals storage business.  The latter story was the subject of an extensive piece in the NYT on Sunday.  The stories are related because the metals warehouse controversy is Exhibit A in the case against allowing banks to be involved in physical commodities.

First the warehouse story.  I wrote about this issue back in January.  There is smoke here.  The premiums in physical aluminum above the LME in-store price is an indication of a bottleneck in getting material out of warehouses.  The question is whether this bottleneck is being artificially exacerbated by game playing-manipulation, perhaps-by the warehouse operators.  The NYT article does cast some light on some mysterious practices, but it does not seal the deal in my mind.  Not to say that manipulation is not occurring, just that the NYT piece doesn’t convince me.

One puzzle is that the operators of warehouses, who cannot own physical metal in them, pay third parties a bonus over the LME price approximately equal to the cash market premium in order to put their material in store, and thereby put themselves at the mercy of the loadout queue.  Presumably, those taking these deals only do so because they are compensated for the higher storage expenses they incur once their metal gets stuck in the roach motels.  If Goldman (or any other warehouseman) paid everyone these premiums, Goldman couldn’t make any money by running the roach motel: it would pay in premium what it collects in storage fees.  It would compensate the roaches for the costs of getting stuck in the motel.  Which suggests that they only pay some customers the premium to attract metal.  What would be interesting to know is how this would permit the warehousemen to collect storage for a longer period from others whose metal is in the warehouses, but who don’t get paid the bonus.  Are there side deals with the storers who receive the premiums?   Does the mere fact that there is more metal in the warehouses increase the ability of the operators to slow down load-out, and thereby collect storage for a longer period?    This seems to be the most curious practice, but the articles I’ve read don’t shed enough light on how they could be part of a manipulative scheme.

Tom Maguire of Just One Minute does a good job at debunking the math in the Times piece: the profits accruing to Goldman and the costs imposed on soda and beer quaffing consumers alleged by the Times don’t withstand scrutiny.  To which I would add another point.  Namely, these calculations assume that the cash premiums reflect an inflation in price.  Any uneconomic act that exacerbates the bottleneck in the transformation of metal in store to metal outside LME warehouses has effects on prices on both sides of the bottleneck: it reduces the price of metal in store, and increases the price at the Coca Cola or Budweiser facility.  So even if the premium is inflated, this does not mean that the price paid by consumers (directly or indirectly) is inflated by the same amount because part of the inflation is due to a depression of the in-store price.  Those with metal in store, or producers who put their metal into store, bear some of the cost.

Now on to the bigger issue, which the hue and cry over bankers owning metals warehouses is intended to illustrate: namely, whether banks should be involved in physical commodity markets.  This is the subject of a Senate hearing today, where one of the main anti-bankers, Ohio Senator Sherrod Brown, is leading the charge to get the banker-squids’ tentacles off our commodities.

I find most of the reasons advanced for keeping banks out of physical commodities, and the ownership of commodity transformation assets, to be unpersuasive.  What’s so special about commodities?  Are they uniquely risky?

I think not, and believe that there are good economic reasons for banks to enter into the commodities business.

In addressing this issue, it’s important to be specific about the kinds of risk involved.  Price risk, spread risk, and operational and reputational risks are perhaps the most important to consider.

The conventional view is that commodity prices are extraordinarily volatile, and hence pose extraordinary risks on those who hold them.  One thing to note is that for a leveraged investor like a bank it is possible to transform a relatively low volatility flat price exposure into a much riskier exposure.  So one thing that matters is whether, taking leverage into account, bank positions in commodities are riskier than their positions in more traditional lines of business.  Moreover, it is risk at the portfolio level that matters, and the correlation between the price exposures and other elements of the banks’ portfolios.  A highly volatile position may not increase portfolio risk substantially if it is uncorrelated with other exposures, or negatively correlated.

But there’s a more important point.  Most commodity trading by commodity trading firms and banks does not consist of punts on the flat price, on whether the price of oil or copper is going to rise or fall.  Instead, most commodity trading is a margin business or a fee business.  Money is made on margins or fees in transportation, storage, and processing.  These margins-spreads-tend to be much more stable than flat prices.  Asset ownership, or contractual control of physical assets (of the type that JP Morgan has over California power plants that have attracted FERC’s ire), is a position on the value and cost of transformation, and spreads and margins measure these values and costs.  The riskiness in these spreads/margins, and the ability to manage spread/margin risk through paper trades, is what drives the risk of bank commodity plays.  My reading of the coverage has turned up virtually no understanding of this issue.

Moreover, the profitability of physical assets can provide a natural hedge for other bank businesses.  Most banking his highly pro-cyclical: it profits when the economy booms, and does badly when the economy tanks.  Some commodity assets are countercyclical.  Storage is a classic example.  The amount of oil or metal in storage-and revenues from storing these things-goes up when the economy weakens.  This is basic economics.  So metal or oil storage-a fee business-is a natural hedge against the strongly procyclical traditional banking and dealing businesses.

Taking these factors into consideration, it is wrong to believe that commodity businesses are unduly risky, compared to other banking activities.

Operational risks are perhaps more problematic.  A major oil spill, for instance, leads to substantial costs on the operator of the ship or pipeline or terminal that spills it.  A bank that owns or charters a tanker that ends up on a reef is potentially exposed to a substantial risk of loss.  But much of this risk can be passed on to insurance companies via the insurance market.  What’s more, the risk doesn’t go away if banks can’t touch physical oil: someone has to bear it.  The question is whether banks are less efficient bearers of this risk than others (where others include insurance companies).

Many of the stories that have come out in recent days focus on the reputational and legal risks associated with commodity trading.  For instance, JP Morgan’s impending settlement with FERC of allegations that it manipulated the California electricity markets has received considerable play.  Well, this is not a problem unique to commodities.  Traditional banking businesses, and other less traditional activities now widely considered to be legitimate activities for banks, have huge reputational and legal risks.  Consider Libor.  Or the Toure trial.  Or mis-selling.  Or mortgage servicing/foreclosures.

I could go on. And on. And on.  If you think that keeping banks out of commodities you’ll reduce their rep or legal risks, I have a bridge to sell you.

Furthermore, even if it is shown that banks have engaged in dodgy activity in commodities, that doesn’t mean that getting banks out of commodities will eliminate the dodgy activities.  Consider the warehouse issue.  If LME rules and cash market frictions make it profitable for warehouse operators to manipulate markets, forcing banks to sell off these assets won’t reduce the manipulation: the new owners will take ownership of the strategies along with the storage sheds.  Note that Glencore-not a bank!-has been accused of playing similar games.   Keeping banks out of commodities may affect who plays dodgy games, and profits therefrom, but won’t have much of an impact on the amount of dodginess in the markets.  That dodginess ultimately derives from economic opportunity (which is driven by frictions in the market) and regulation (flawed rules and rule enforcement mean that market participants calculate there’s a substantial probability of escaping punishment for misdeeds).  Barring banks from the market won’t change either of these things.  If crime pays, someone will commit it.  If you care about reducing the deadweight losses from manipulative activities, you shouldn’t really care about who profits from them.  You need to fix the problem at the level of incentives and enforcement of rules: keeping banks out of the market doesn’t do that.  Period.

The foregoing is basically a rebuttal-or at least a skeptical questioning-of the case against bank ownership of commodity assets, or participation in physical commodity markets.  What about the affirmative case for such participation?

For several years, in response to shrieks about the evil effects of the “financialization” of commodity markets, I’ve made the following basic point: a good deal of commodity trading is about allocating commodities over time, and allocating risk.  Both of these are fundamental financial functions: finance is about trading off the future and the present, and allocating risks.  Banks are linchpins of the financial system, and engage in intermediation of risk and of resources over time.  Why should we exclude them from a sector where the allocation of resources over time and the allocation of risk are vitally important?  That is the comparative advantage of banks: why keep them out?

Indeed, banks have gained market share in commodities in large part because they can perform the time- and risk-allocation functions more efficiently than others.  If banks have the lowest funding costs, it makes sense that they fund some inventories.  Banks have expertise in risk management and hedging, making it sensible that they utilize this expertise in the management of commodity price risks.  In other words, the success of banks in commodities is a feature, not a bug, because it reflects their comparative advantages in allocating risks and resources over time.

Moreover, there is often a synergy between financing and risk management activities, and banks can exploit these synergies.  For instance, one physical market activity that banks engage in is offtake agreements, whereby they provide say crude oil to a refinery, and receive the refinery’s output which they market.  This bundle of transactions involves a funding element-the bank is basically providing working capital to the refinery.  It also involves a risk management element: the bank is managing the price risks on the crude and refined product side (as well as the logistical/operational risks).

Crucially, by taking on the risk, the bank is reducing the moral hazard that would be involved if it was providing working capital financing to the refinery.  Yes, if it just extended loans or credit lines to the refinery, the bank would presumably impose requirements on the refinery to hedge its risks, but there are inherent agency problems associated with such an arrangement that can be avoided by putting the price risks on the bank.   Moreover, the operational and logistical risks cannot be hedged, and pose a potential moral hazard.

The most telling objection that might be raised against these arguments is that banks have gained market share in commodities because they are subsidized as a result of too big to fail.  I would note that the investment banks that made the first and biggest forays into commodities-Morgan Stanley, Goldman, Bear-did so when they did not have access to insured deposits for funding or to the Fed window, though one could argue that they still were subsidized by the belief that they were too big for the Fed to allow to fail.

But even if you buy into the subsidy argument, that’s no reason to single out commodities.  The subsidy leads to an excessive expansion of big banks generally, across all lines of business.   Meaning that keeping them out of commodities will not materially reduce the perverse effect of the subsidies.  Implicitly or explicitly subsidized banks will exploit that subsidy to the hilt, and if they can’t do it in commodities, they’ll do it somewhere else.  TBTF has to be tackled at the roots, not the branches or the leaves-and particular lines of business like commodities are the latter.

One last point.  It is beyond ironic that the banks are under attack for their commodity dealings.  GFMA wanted to hamstring banks’ competitors in commodity trading-namely, the commodity trading firms.  Perhaps they should have paid more attention to making the affirmative case for their participation in these markets, than they did attempting to raise their non-bank rivals’ costs.

The word karma comes to mind.

Marco Dialosa of the Pension Insurance Corporation echoes some SWP clearing themes in an interview on Derivsource.com (sorry we didn’t connect in London last month, Julia):

Q. Given the liquidity squeeze that will result from CCP clearing, where do you see the liquidity coming from?

A. We believe that the amount of high quality liquid assets on balance sheets will increase. We see the repo market as a mean to provide liquidity for cash variation margins. What is less clear at this stage is whether the repo market is the right place for collateral transformation and how this market will behave in times of stress. One of the unintended consequences of the clearing initiative is that it may be transferring systemic risk away from the OTC world into the repo market.

Q. Does the movement of systemic risk away from OTC world and into the repo market an unintended consequence concern you or do you view it as just inevitability?

A. It is an unintended consequence in our view. The OTC market is akin to an ecosystem, it is in a stable equilibrium. A small change in a particular parameter (i.e. central clearing) will be corrected by some negative feedback that will bring the parameter back to its original point of balance. It is extremely difficult to eliminate systemic risk in practice without disrupting the balance of the whole system.

Just so.  Most of the effects of clearing mandate are to redistribute risk, not to reduce it.  Moreover, the equilibrium in the system will adjust in response to an exogenous policy shift: the “negative feedback” will attempt to find a way to replicate the old equilibrium as closely as possible, given the constraints imposed by the new rule or regulation.

In other words: you need to take a systemic approach to analyzing systemic risk.  You have to understand how a change imposed on one part of the system is going to be transmitted to other parts of the system, and how those other parts are going to respond to this impulse.   Systems achieve a particular configuration for particular reasons, and it is very difficult to change these configurations fundamentally.  The particular reasons that led to its particular configuration lead to the negative feedbacks that Dialosa mentions.  And a good thing too.  Negative feedback is a stabilizing force.