Oil Daily, a publication of the Energy Intelligence group, had a story yesterday authored by Gregory Gethard about a deal between the refining company Alon USA and Goldman Sachs’ commodities trading arm, J. Aron. The headline reads: “New Goldman Sachs US Refinery Deal Sets Alarm Bells Ringing”.
What is the structure of the deal, and what are the alarm bells about?
First, to the structure… According to the story, the refiner, Alon, agrees to purchase from Goldman 70,000 barrels of crude oil per day for use in its Big Spring Refinery in Texas. Goldman, in turn, purchases from Alon the refined products it produces. Goldman also pays to lease the storage facilities at the refinery. The sales are all made at market prices. The deal runs at least through May 2013, and potentially into 2016. These general terms of the deal were outlined by the refiner in an 8K filing with the SEC. A similar deal had been executed between Alon and Goldman in mid-2010 for product at Alon’s Krotz Spring refinery in Louisiana, as reported in a company 8K filing with the SEC. An article on Risk.net by Pauline McCallion describes that earlier deal and the motivations for it.
What are the results of the deal? How does it restructure risk?
As far as I can see, the obvious terms of the deal do not restructure risk at all. Since all of the purchases are done at market prices, the margins earned by the refiner are exactly what they would have been without the contract. If the crack spread widens, the refiner captures that profit. If it shrinks, the refiner’s margin is squeezed. There is no hedging built into this agreement.
On its face, this is essentially an inventory financing deal. The business of refining oil of necessity involves managing an inventory of the crude oil input, of intermediate product, and of refined products. Financing that inventory is a major headache for a refiner. Absent this deal, the inventory would have been financed with a letter of credit. This deal accomplishes the same financing by other means.
Why structure the inventory financing this way? The Risk.net article suggests that the structure “was designed to protect Alon’s balance sheet from crude oil price increases at a lower cost than traditional bank revolver financing.” Well, the deal doesn’t protect Alon’s balance sheet, at least the terms as disclosed do not obviously do that. The Risk.net article also points out that the deal locks in financing for inventory, saving Alon from liquidity risk: “Alon no longer needs to post a letter of credit to access crude for its refinery. When oil prices spike, liquidity risk can increase if credit requirements exceed the letter of credit capacity under a refiner’s bank revolver.” But in theory a letter of credit could also have been negotiated with sufficient flexibility to give Alon the liquidity it required for the same term as this deal. The Risk.net article also focuses on how this deal matches the refiner’s purchase of crude against the refiner’s delivery of refined products, achieving lower cost by netting these two. But the same can theoretically be done within a letter of credit arrangement.*
The real source of value to the deal is in how it optimizes the value produced with Alon’s assets. How?
In both deals the refiner “retains operating control of the refinery, controlling crude grades and run rates”, so changes in those operations are not the source of value. The key difference I see is in the management of each refinery’s storage facilities. Here Goldman gets some control. Explaining the Louisiana refinery deal in the Risk.net article, Goldman Sachs Vice-President Simon Collier is quoted saying that “The price risk we have is associated with the inventory sitting in the tank at any point in time and we manage that as part of our overall commodities book.” The Risk.net article continues, “Goldman Sachs can adjust inventory across the refinery system through its crude purchases into the refinery, based on its view of the current market structure and outlook. To build a barrel in the crude oil inventory, for example, it would simply buy the number of barrels required by Alon plus one. As such, Goldman Sachs can participate in the inventory swings, but Alon can run as many barrels as it wants and enjoy the margin on those barrels.”
Storage facilities, even when co-located with a specific refinery, can often provide storage services to the larger system. So the optimal management of those storage facilities is determined by market prices and factors beyond the specifics of the refinery’s own operation. If Goldman can optimize on the management of crude and product inventories stored at the refinery, they can add value. Optimizing the use of storage facilities is a complicated business that requires a very sophisticated modeling of both spot and futures prices in crude and products and the relationships across the entire complex. It is easy to believe that Goldman’s commodity trading arm, J. Aron, brings superior management skills to this task. This is the reason for this structure, I suspect.
One note of caution for the refiner… Although the storage facilities at each refinery can be used to serve the larger system, they are also there to serve the associated refineries. True optimization requires comparing (i) the profit that can be achieved by looking beyond each refinery’s own requirement to (ii) the profit that can be achieved by serving the refinery. There is an inherent conflict between the use that maximizes Goldman’s profit on the storage facilities and the use that maximizes the refinery’s profit. It is of necessity an uneasy partnership. The Risk.net article suggests that the refinery’s needs will always be accommodated, but that is simply unrealistic and impossible. All storage facilities must be operated recognizing the possibility of stockouts which would temporarily impede the refinery’s operations. Reducing that probability to zero is costly and not sensible. Reducing that probability to zero would not be maximizing the value of the storage facility. So compromises must be made. So far the news reporting on the arrangements between Alon and Goldman hasn’t delved deeply enough to explain how these compromises are to be made. These are the details that a company like Alon must carefully negotiate when it chooses this arrangement. These are the details that will determine whether such an arrangement is truly a plus for refiners like Alon.
Second, what about the Oil Daily’s alarm bells? The article suggest that this deal “could potentially enable the investment bank to sidestep tighter regulations on speculative commodities trading” … “by allowing Goldman’s trading arm to claim physical ownership of oil and products.” End-users are exempt (under certain circumstances) from some of the requirements imposed by the new Dodd-Frank legislation. The articles suggest that this deal is a ruse to allow Goldman Sachs’ trading arm to be designated an end-user even though the physical assets are actually still being operated and managed by the other party to the transaction—hence the title of this post, a synthetic end-user. Should we be alarmed?
Goldman Sachs, like many other banks and derivative dealers, already does have some amount of ownership of physical assets. This particular transaction doesn’t change the basic equation much at all.
Sometimes a bank comes to own physical assets through the bank’s proprietary trading, which can involve buying and selling physical assets. Sometimes a bank owns physical assets in order to make possible its financial business. When trading in commodity futures, it can sometimes be advantageous to own physical assets that allow one to take or make delivery of the commodity. The objective is to facilitate the financial trading, but the bank’s hands do touch physical commodities. It is still a bank.
Sometimes a bank or derivative dealer also owns physical assets because they are actually running a physical business, pure and simple. Like any other conglomerate, a bank holding company can house both financial businesses and physical businesses. The holding company does not have to be only purely financial or purely physical. I’ve written in previous posts that companies well known as end-users with a clear physical business—like Shell and Cargill—can also get into the business of dealing in derivatives or otherwise running financial businesses—see here, here, and here. Well, it works the other way around, too. Companies well known as banks with a clear financial business—like Goldman Sachs and Morgan Stanley—can also get into physical businesses that look exactly like the businesses of other end-users. Storage facilities are a great example of a physical businesses that happen to oftentimes be well managed by companies with skills honed at financial trading desks. As indicated above, the optimal management of these physical facilities requires understanding complex price modeling and valuation exercises that are very analogous to the tools used by financial trading houses. So it should be no surprise that companies involved in commodity derivative dealing may also get involved in managing some physical assets such as storage facilties. It is widespread and works both ways. Financial firms that developed the tools to deal commodity derivatives have extended their operations to managing storage facilities and other activities related to commodity trading, and end-users that have developed the capabilities to manage storage facilities and other activities related to commodity trading have extended their operations to dealing commodity derivatives.
The Dodd-Frank legislation mandates that end-users are exempt from certain requirements (under certain conditions and limits). The CFTC is tasked with drafting the implementing regulations. The only way a transaction like the deals between Alon’s refineries and Goldman Sachs could be used to circumvent the law’s requirements is if the implementing regulations create unreasonably expansive exemptions: for example, if doing ANY business in the physicals somehow exempted ALL of a company’s activities. (Of course, in that case, this deal would be unnecessary because Goldman already does plenty of business in the physicals.) What the CFTC should do is acknowledge that underneath one corporate holding company there can be different divisions and desks, some of which should be treated as end-users and some of which should be treated as financial businesses. Divisions that are exclusively end-users should be exempt. If a company mixes the two types of activities in one division, then that division should not be exempt. Goldman Sachs’ commodities arm, J.Aron, would need to either organize its operations to keep its physical business clearly distinct, or it cannot claim any exemption. In any case, all of the rest of Goldman’s operations are clearly not exempt.
*There may be some aspect of financing inventories this way that is advantageous and which I am overlooking. Inventory financing is a peculiar business that varies by industry. For example, retailers in various goods deploy a wide array of arrangements, including factoring. Understanding the particular structure best suited to a specific industry—in this case refining—requires understanding many institutional details of the business. In particular, it is important to understand what happens when everything goes wrong, who gets what in bankruptcy events and other events of financial distress. Deals like this one which look like they replicate something as simple as a letter of credit usually differ on vital legal details during those crisis moments. I am not a specialist in the legal aspects of this business, and looking from the outside, it is impossible for someone like me to pinpoint the reason why this particular structure turns out to be better than a letter of credit.