One subject I’ve taught about in my commodity trading classes for over a decade is asset ownership by commodity trading firms. As my white paper for Trafigura shows, there is considerable diversity among commodity traders’ asset ownership. Some are “asset heavy” while others are “asset light.”

Historically the big oil traders–Vitol, Trafigura, Gunvor, Mercuria–have been in the asset light category. (Glencore transitioned to an asset heavy mining entity with a supersized trading arm before the white paper was published–more on that below). Recently, however, they have gone on something of an asset acquisition binge, buying refineries in particular, as this recent Bloomberg piece describes. Which raises the question: why?

The standard consultant story is that assets have optionality. This is a superficial and incomplete answer. First, the fact that an asset has optionality says nothing about who is best positioned to own the asset and exercise the options. In the case of a refinery, is it better for an integrated oil company to own it? An independent refiner? A commodity trader? Any of these can exercise the options. Which gets the most option value?

Second, it is often possible to acquire the right to exercise optionality via contract rather than ownership. A ship (tanker or bulk carrier) is the best example of this. Ships have extensive optionality in terms of route, timing, and sometimes cargo. Time chartering a ship gives the charterer the ability to tell it where to go when and carrying what: ownership isn’t necessary to exercise these options. Off-take agreements from a mine or well also confer rights to exercise options (e.g., whom to sell the output to) without requiring the purchase of the mine or well.

Transactions cost economics is one analytical tool to apply to these issues. That’s what I used to analyze contracting and ownership patterns in the ocean shipping industry oh-so-many years ago. The Trafigura white paper applies this to commodity trader ownership generally.

With respect to commodity traders as optimal owners of assets with optionality I’ve been playing around with a formal model, the intuition for which is fairly straightforward. Physical commodity traders operate in bilateral markets rather than centralized ones. Prices and values are not transparent. Traders are essentially price entrepreneurs who search for and discover cheap supplies and high value buyers. Their intimate market knowledge of profitable trade opportunities may allow them to exercise real options, like the real options inherent in a refinery (especially regarding choice of crude input and the product slate), more efficiently than someone with poorer information.

But that still doesn’t fully answer the ownership question. The trader could obtain the optionality through a tolling agreement, for example. (These are not unknown in the oil business. I recall that when JPMorgan was in the commodities business, it had tolling agreements with refiners).

Further, why is it more efficient for a trader to buy the refiner, rather than the refiner to buy or build trading expertise? BP and Shell in particular are renowned for their trading expertise, and even a laggard–Exxon–has moved into trading.

Since this is also a relatively new development, one must also ask “why now?” (And to its credit, the Bloomberg article does just that). Has traders’ comparative information advantage increased recently? That’s a hard proposition to test, but I am skeptical.

As Bloomberg points out, this buying spree comes in the wake of boom years in commodity trading, and the big traders are awash in cash. (As the billions in bonus payments to traders shows). This is most likely the proximate cause of the acquisitions.

What this says about the efficiency/optimality of the acquisitions is unclear. I can tell two stories with opposite conclusions.

The first story is that this is a classic free cash flow problem. Companies with cash burning a hole in their pockets are prone to blowing it on unprofitable investments, including acquisitions. This was a driver behind the leveraged buyout boom in the 80s: leveraging up gives cash to stockholders and precommits the company to use cash flows to service debt rather than fritter it away in bad investments. In the oil industry, this was T. Boone Pickens’ game. (Academic research has shown that in the 80s E&P companies’ well drilling was a negative NPV activity).

If this is the driver behind commodity trader refinery acquisitions, it would imply that these purchases are inefficient. (I note that some earlier trader ventures into refining, like Gunvor’s, turned out badly).

The other story is that trader firm ownership structures, adopted for value maximizing reasons, constrained their ability to invest in physical assets like refineries. The major independent oil traders are private firms. One of the downsides of private ownership is that a private firm cannot tap public equity markets to fund acquisition of durable assets. Relatedly, it may constrain the ability to secure long term debt to finance long term assets. As a consequence, long term assets (like refineries) must be funded with retained earnings. In this case, even if traders can exercise real options more efficiently than an oil major or independent refiner, the higher capital costs inherent in their ownership structures makes it uneconomical for them to acquire refineries.

Arguably the main reason for Glencore’s transition to a publicly traded firm was that its decision to go asset heavy into mining was incompatible with private ownership. It needed to rely on the capital markets to fund this new strategy.

Against this cost of private ownership (limited diversification is another) is the benefit: a better alignment of the interests of managers and owners, because the managers are the owners and the owners are the managers. One reason this structure has clearly had survival value in commodity trading is that many of the other risks–especially commodity price risks–can be transferred via derivatives markets rather than the equity markets.

The cash flow gusher of the last few years loosens the constraint imposed by ownership structure. Thus, in this story, traders can have their cake and eat it too. They can remain private, and reap the benefits of better incentives, but can invest in fixed assets that were previously unaffordable. In this story, the positive cash flow and balance sheet shock enhances efficiency rather than reduces it (as in the free cash flow story).

Which story is correct? I have no strong opinion, but I will say that the governance benefits of private ownership do mitigate the free cash flow problem. So I guess I would lean towards “efficient.”

One must also recognize that it takes two to tango. These transactions have a seller too, of course. Meaning that “why are traders buying?” is not the only question: “why are oil companies selling?” is also important.

About a decade ago traders acquired downstream marketing operations, especially in developing markets like Africa, from oil companies. In that case, the development of liquid markets for refined products reduced the transactions cost benefits of ownership by the oil companies: the development of these markets mitigated the bilateral monopoly problems that occurred in relatively small downstream markets served by local refineries. That reduced the benefits of ownership for the oil companies, creating gains from trade by selling to the traders.

Perhaps something similar is going on today in refining, especially for relatively “simple” (“non-complex“) refineries in Europe and Asia. When I have a chance, I will look into the complexity index for the refineries the traders have acquired. With the entry of substantial high-complexity refinery capacity in the Middle East and elsewhere, I would not be surprised if a major driver behind these transactions is that refiners want to shed their less complex operations because they have a comparative advantage in operating bigger, more expensive, and more complex refineries.

Which, ironically, have more optionality: complexity translates directly into optionality.

In sum, the trader refinery acquisition boomlet is an interesting development. It raises intriguing economic questions that I have just sketched out here. Could be interesting fodder for an aspiring researcher–like my Université de Genève students (hint hint, nudge nudge).