The LNG Market’s Transformation Continues Apace–and Right On Schedule

When he wrote about the Liquefied Natural Gas (LNG) market “racing towards an inflection point” in our Winter 2017 issue (See “Liquefying a Market: The Transition of LNG to a Traded Commodity” in Volume 29, Issue 1), University of Houston Professor Craig Pirrong made some specific but out-of-the mainstream predictions.

Those predictions turned out rather well.

As he noted in a recent blog post, he had made the following points:

  • First, the traditional linkage in long term LNG contracts to the price of oil (Brent in particular) was an atavism–a “barbarous relic” (echoing Keynes’ characterization of the gold standard) as I phrased it more provocatively in some talks I gave on the subject. The connection between oil values and gas values had become attenuated, and often broken altogether, due in large part to the virtual disappearance of oil as a fuel for electricity generation, and the rise in natural gas in generation. Oil linked contracts were sending the wrong price signals. Bad price signals lead to inefficient allocations of resources.
  • Second, the increasing diversity in LNG production and consumption was mitigating the temporal specificities that impeded the development of spot markets. The sector was evolving to the stage in which participants could rely on markets to provide security of demand and supply. Buyers were not locked into a small number of sellers, and vice versa.
  • Third, a virtuous liquidity cycle would provide a further impetus to development of shorter term trading. Liquidity begets liquidity, and reinforces the willingness of market participants to rely on markets for security of demand and supply, which in turn frees up more volumes for shorter term trading, which enhances liquidity, and so forth.
  • Fourth, development of more liquid spot markets will make market participants willing to enter into contracts indexed to prices from those markets, in lieu of oil-linkages.
  • Fifth, the development of spot markets and gas-on-gas pricing will encourage the development of paper hedging markets, and vice versa.
  • Sixth, the emergence of the US as a supplier would also accelerate these trends. There was already a well-developed and transparent market for natural gas in the US, and a broad and deep hedging market. With US gas able to swing between Asia and Europe and South America depending on supply and demand conditions in these various regions, it was likely to be the marginal source of supply around the world and would hence set price around the world.
  • Moreover, the potential for geographic arbitrages creates short term trading opportunities.

    When pressed about timing, I was reluctant to make a firm forecast because it is always hard to predict when positive feedback mechanisms will take off. But my best guess was in the five year range.

    Those predictions, including the time horizon, are turning out pretty well. There have been a spate of articles recently about the evolution of LNG as a traded commodity, with trading firms like Vitol, Trafigura, and Gunvor, and majors with a trading emphasis like Shell and Total, taking the lead. Here’s a recent example from the FT, and here’s one from Bloomberg. Industry group GIIGNL reports that spot volumes rose from 27 percent of total volumes in 2017 to 32 percent in 2018.

    There are also developments on the contractual front. Last year Trafigura signed a 15 year offtake deal with US exporter Cheniere linked to Henry Hub. In December, Vitol signed a deal with newcomer Tellurian linked to Henry Hub, and last week Tellurian inked heads of agreement with Total for volumes linked to the Platts JKM (Japan-Korea-Marker).* Shell even entered into a deal linked with coal. There was one oil-linked deal signed recently (between NextDecade and Shell), but to give an idea of how things have changed, this met with puzzlement in the industry:

    The pricing mechanism that raised eyebrows this week in Shanghai was NextDecade’s Brent-linked deal with Shell. NextDecade CEO Matt Schatzman said he wanted to sell against Brent because his Rio Grande LNG venture will rely on gas that’s a byproduct of oil drilling in the Permian Basin, where output will likely increase along with oil prices.

    Total CEO Patrick Pouyanne said he didn’t understand that logic.

    “Continuing to price gas linked to oil is somewhat the old world,” Pouyanne said on Wednesday. “I was most surprised to see new contracts linked to Brent, especially from the U.S. Someone will have to explain this to me.”

    I agree! In fact, the NextDecade logic is daft. High oil prices that stimulate oil production will lead to lower gas prices due to the linkage that Schatzman outlines. If you have doubts about that, look at the price of natural gas in the Permian right now–it has been negative, often by $6.00/mmbtu or more. This joint-production aspect will tend to make oil and gas prices less correlated, or even negatively correlated.

    But it’s hard to believe how much the conventional wisdom has changed in 5 years. The whitepaper was released in time for the LNG Asia Summit in Singapore, and I gave a keynote speech at the event to coincide with its release. The speech was in front of the shark tank at the Singapore Aquarium, and from the reception I got I was worried that I might get the same treatment from the audience as Hans Blix did from Kim Jung Il in Team America.

    To say the least, the overwhelming sentiment was that oil links were here to stay, and that any major changes to the industry were decades, rather than a handful of years, away. Fortunately, the sharks went hungry and I’m around to say I told you so.

    I surmise that the main reason that the conventional wisdom was that the old contracting and pricing mechanisms would be sticky was an insufficient appreciation for the nature of liquidity, and how this could induce tipping to a new market organization and new contract and trading norms. These were ideas that I brought from my work in the industrial organization of financial trading markets (“market macrostructure” as I called it), and they were no doubt alien to most people in the LNG industry. Just as ideas about spot trading of oil were alien to most people in the oil industry when Marc Rich and others introduced it in the 1970s.

    Given the self-reinforcing nature of these developments, I believe that the trend will continue, and likely accelerate. Other factors will feed this process. I’ve written in the past about how some traditional contract terms, notably destination clauses, are falling by the wayside due to regulatory pressure in Japan and elsewhere. The number of sources and sinks is increasing, which makes the market thicker and mitigates further temporal specificities. The achievement of scale and greater trading opportunities will encourage investment in infrastructure, notably storage, that facilitates trading. Right now most LNG trading involves only one of the transformations I’ve written about (transformation in space): investment in storage infrastructure will facilitate another (transformation in time)

    It’s been kind of cool (no pun intended, given that LNG is supercooled) to watch this happen in real time. It is particularly interesting to me, as an industrial organization economist, given that many issues that I’ve studied over the years (transactions cost economics, the economics of commodity trading, the nature and dynamics of market liquidity) are all present. I’m sure that the next several years will provide more material for what has already proved to be a fascinating case study in the evolution of contracting and markets.

    *Full disclosure: My elder daughter works for Tellurian, and formerly worked for Cheniere. I have profited from many conversations with her over the last several years. One of my former PhD students is now at Cheniere.

    Leave a Reply

    Your email address will not be published. Required fields are marked *

    You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>