Streetwise Professor

European politicians are desperate, and desperate politicians thrash around and grab stupid ideas–any idea–that they think will alleviate the source of their desperation.

The cost of energy is extremely high now in Europe, and that is stoking political desperation. The politicians don’t like the price signals that the market is sending, and so they are exploring myriad ways of interfering with the price system. These policies (e.g., price controls) cannot solve the underlying problem: Europe is structurally short energy. Moreover, they will create their own problems, which will result in panicked reactions that will create new problems. Wash, rinse, repeat. The whole thing is doomed to collapse. In tears.

There is no better illustration of the European failure to come to grips with their real problems than some of the recent proposals surrounding LNG. One is to cap the price of imported LNG.

Brilliant! That way you’ll have even less gas, and the marginal value of power will go up! Yay!

Er, the LNG market is a world market. Yes, Europe collectively is large enough to have some monopsony power and thus can reduce price: but one exercises monopsony power by cutting purchases. That is, less gas will flow to Europe. Europeans will consume less electricity, and they will substitute higher cost ways of generating it. The shadow price (the opportunity cost, i.e., the real cost) of electricity and gas will increase, not fall.

Another proposal is in some ways even whackier: to replace the Dutch Title Transfer Facility (TTF) price with the Japan-Korea Marker (JKM) as the benchmark price for gas in Europe. Because the JKM price is lower. Or something:

As a last resort in case of supply disruption in Europe the EU could also explore temporarily pegging the TTF to the JKM Asian benchmark as a dynamic cap. Yet that would require the use of other hubs or mechanisms to allocate gas inside Europe, the commission said in the document on benchmarks for the wholesale gas market. “In this situation, JKM would become the world price for international gas for some time,” the commission said. “The wholesale market would be therefore determined by LNG supply/demand, and not by the EU’s internal bottlenecks. LNG would still be attracted by the fact transport costs are lower to the EU.”

In other words, the EC doesn’t like the basis between the price of pipeline gas in Europe (as measured by the TTF price in the Netherlands) and JKM. So, voilà! Make JKM the benchmark and tie the TTF price to the JKM price.

This begs the question of why the basis is what it is. The basis–the spread–reflects bottlenecks as well as shipping costs. If TTF is at a premium to JKM (which it is) even though shipping costs to Europe are lower, that means that there is some bottleneck to transform LNG on the European coast into gas in a pipeline on the continent. The most likely bottlenecks is gasification capacity. The Europeans are scrambling to get floating LNG gasification facilities operating, but the basis/spread is saying that a lot more capacity is needed.

The EC concedes that TTF is at a premium, and that the premium has increased: ““The price premium between the TTF and Europe’s LNG delivered ex-ship indices has widened significantly bringing up questions about its representativeness as an index for linking the contracts in the whole EU-27.”

It only brings up questions to fools. Non-fools get it.

If by “pegging” the Europeans impose some spread between TTF and JKM (adjusted for shipping cost differentials) that does not reflect these EU bottlenecks and their associated costs, the supply of LNG to Europe will be reduced. The only way to incentivize the flow of gas from overseas into European pipes is to have a price that covers the cost of that transformation. If there are bottlenecks, that transformation is expensive.

Thus, it is utterly delusional to think that a price that does not reflect “the EU’s internal bottlenecks” is a good thing: you want a price that does reflect them. The bottlenecks don’t go away because you don’t let the market price reflect them. If you don’t let the market price reflect them, the gas will go away. (Asia will send its regards, by the way.)

You may not like the message the price sends, but you cannot wish it away. And if you try, the message will be sent in an even more painful way. The Europeans (and the UK) seem hell bent on proving this very basic point the hard way, rather than learning from the hard experience of others dating back millennia.

When playing Spin the Bottleneck with my students I say to look at what lies between the price of a transformed and untransformed commodity to identify the bottleneck. In my earlier post on the gaping spread between European (and Asian) LNG prices and the price of US gas (which is on the margin for both destinations) I noted two possible constraints: shipping and liquefaction capacity.

Well, it ain’t shipping.

There is a surfeit of LNG shipping capacity. So much that LNG shipping is effectively free between the US and Europe (down from $273K/day in December). Yet the spread remains very wide. So the binding constraint is definitely liquefaction capacity, in the US in particular. Those who have the rights to that capacity–notably firms that entered into contracts with the likes of Cheniere or Freeport that buy gas at the US price and pay a contractually fixed liquefaction/tolling fee–are coining it. They capture the bulk of the existing spread between TTF or UK Balancing Point prices and Henry Hub. (The LNG companies are benefitting only to the extent that they reserved some of their capacity for their own trading, which is rather de minimis).

So in the short run liquefaction capacity is quite valuable. The question is what will its value be over the longer term? Will current events convince enough financiers to provide capital for a large expansion of US capacity? Given the long gestation period of these projects it is a hard issue for banks and equity to analyze.

One thing to note. Another thing I discuss extensively in my classes is the importance of government/regulatory bottlenecks. Such bottlenecks may be a constraint on expansion of US LNG capacity. Many of the projects under development do not have the requisite federal permits. The Biden administration is unlikely to grant more. Thus, like taxicab medallions in NYC, existing permits likely have a substantial scarcity value–thanks to a government-created bottleneck.

This has interesting implications for financing of US LNG projects. Financiers of a given project face less risk of a glut of capacity coming online in a few years, and this should make them more willing to finance already permitted projects. But, of course, they are taking on political risk by doing so: might a new administration change course post-2024? Or might political pressure induce a change in course by the current administration? There are already a lot of political risks in investing in anything fossil-fuel related (attributable to climate hysteria). This is a US LNG-specific risk.

Although conditions have eased somewhat, Europe’s energy situation is still fraught. It has experienced a winter of extremely high power prices and natural gas prices. Some luck with weather in January brought some relief, but who knows what the next months have in store.

As if demand and non-Russia supply fundamentals didn’t pose enough difficulties, Russia has proved a gas supply riddle, mystery, and enigma. The Ukraine situation has only made the situation more fraught, with a prospect that things could get very cold in Europe if the conflict between Russia and Ukraine goes hot.

In anticipation of this, the Biden administration has played its by now familiar role of supplicant, appealing to major LNG suppliers–notably Qatar–to increase output and divert supplies to Europe, and to major buyers (e.g., Japan) to direct supplies to Europe.

Good luck with that. There is nothing that can be done to augment global supplies in the short run, let alone by enough to address a material loss of gas from Russia.

The world LNG industry is operating at effectively full capacity. Although supplies produced anywhere can physically go anywhere, there are no more additional supplies to be had. So if Russian gas exports decline, LNG cannot fill the gap.

This is illustrated by the US, which recently became the world’s largest supplier of LNG. Note how US exports have remained essentially constant (at about 300 bn cubic feet/month) since around March 2021, despite skyrocketing prices for gas:

There’s another way to see that liquefaction capacity is the bottleneck. This is the spread between the price of LNG in Asia (JKM) and Henry Hub NG:

My teaching mantra (repeated last night to my Energy Derivatives students!) is that spreads price bottlenecks in the transformation process. The spread depicted here captures two potential bottlenecks in transforming US produced gas into gas delivered in Asia (or Europe): US liquefaction capacity and shipping costs. Shipping costs have increased, but not nearly enough to explain the rise in the spread in 2021, especially the explosion starting in mid-September. This is a clear indication that US liquefaction capacity is a binding constraint. The extreme volatility in the spread also demonstrates this. Shocks to demand in Europe or Asia can’t be accommodated by adjustments in supply, so all the burden falls on the spread. Thus, even slight demand shifts can lead to big spread moves.

As for reallocating supplies between Europe, Asia, and South America, Mr. Market can handle this, and is indeed handling it to a considerable degree. As the demand balance shifts between Asia and Europe, LNG carriers are changing destinations in mid-voyage. In an extreme case a ship that had already transited the Panama Canal en route to Hawaii reversed direction, transited the canal eastbound, and delivered its cargo in the UK. In short, what is going on is a classic example of how the price mechanism allocates resources to their highest value use, and how commodity traders optimize flows. (It also largely validates the predictions of my 2014 white paper on LNG.)

The main impediment appears to be contractual. Specifically, destination clauses in legacy contracts, specifically those for Qatar Gas. These clauses prevent the buyer from reselling the cargo to other markets.

Perhaps the administration was begging Qatar to relent on these, but if so, their appeals appear to have fallen on deaf ears (despite the US promoting Qatar to “major non-Nato ally). Qatar says it views its contractual commitments as sacrosanct:

“Keeping our contractual word is sacrosanct in Qatar,” said Kaabi, implying that it will not be possible to divert to Europe gas shipments already contracted for delivery to other countries without their consent.

This smacks of dishonesty. It is not the buyers‘ consent that matters here: if the price is right, Asian buyers would be happy to ship contracted gas to Europe if the price is right. And it appears that the price is right: the “arb” between Asia appears open (i.e., the TTF price is above JKM by more than the cost of shipment). So what’s stopping it from being closed? Almost certainly Qatar’s failure to consent to waiving the destination clauses (in its supersekret contracts). (Destination clauses limit the ability of the buyer to divert cargoes to other locations, and are basically a means of price discrimination.)

Relatedly, Qatar has said it is willing to sell gas to Europe . . . as long as the Europeans consent to destination clauses. Oh, and Qatar is exploiting Europe’s straits by demanding Europe “resolve a long-running EU probe into Qatar’s long-term gas contracts, for the EU to be less dependent on spot sales and more on long-term contracts to boost its energy security.”

Kind of Russian, when you think about it. When you have them by the short ones . . . . With “allies” like these!

In brief, Europe is SOL, especially if there is a material disruption in Russian gas supplies. The world capacity of LNG is maxed out and cannot fill any supply void. The market is working to allocate supplies efficiently, but that process is impeded by an opportunistic actor that really doesn’t want the spot market to work.

Meaning that the best that post-religious Europe can do right now is pray for balmy weather, high winds, and Russian forbearance. Good luck with that.

After languishing in the doldrums in the Covid era, and at times touching historic lows, the price of LNG delivered to Asia skyrocketed in recent weeks before plunging almost as precipitously:

As always happens with such big price moves, there has been an effort to round up suspects. Here, since the visible price increase occurred in the spot market, the leading culprit is the spot market–something that has been growing rapidly in recent years, after being largely non-existent prior to 2014 or so.

For example, Reuters’ Clyde Russell writes:

What is more likely is that some buyers misjudged the availability of spot cargoes, and when hit with a surge in demand found themselves unable to secure further supply, thus bidding up the prices massively for the few cargoes still available.

Frank Harris of Wood Mackenzie opines:

“Buyers are going to become aware that you may not always be physically able to source a cargo in the spot market regardless of price,” Mr Harris says. “The most likely outcome is it shatters some of the complacency that’s crept into the market over the last 12-18 months.”

It is incorrect to say that a shortage of spot cargoes per se is responsible for the price spike registered in the spot market. It is the supply of LNG in toto, relative to massive increase in demand due to frigid weather, that caused the price increase. How that supply was divided between spot and non-spot trades is a secondary issue, if that.

The total supply of LNG, and the spatial distribution of that supply, was largely fixed when the cold snap unexpectedly hit. So in the very short run relevant here (days or weeks), supply in Asia was extremely inelastic, and a demand increase would inevitably cause the value of the marginal molecule to rise dramatically. Price is determined at the margin, and the price of the marginal molecule would be determined in the spot market regardless of the fraction of supply traded in that market. Furthermore, the price of that marginal molecule would likely be the same regardless of whether 5 percent or 95 percent of volume traded spot.

If anything, the growing prevalence of spot contracting in recent years mitigated the magnitude of the price spike. Traditional long term contracts, especially those with destination clauses, limited the ability to reallocate supplies efficiently to meet regional demand shocks. The more LNG effectively unavailable to be reallocated to the buyers that experienced the biggest demand shocks, the less elastic supply in the spot market, and the bigger the price increase that occurs in response to a given demand shock. That is, having less gas contractually committed, especially under contracts that limited the ability of the buyers to sell on to those who value it more highly, mitigates price spikes.

That said, the fundamental factors that limit the total availability of physical gas, and constrain the ability to move it from low demand locations to high demand locations in the short time frames necessary to meet weather-driven demand changes (ships can’t magically and instantaneously move from the Atlantic Basin to the Far East), mean that regardless of the mix of spot vs. contract gas prices would have spiked.

Some have suggested that the price spike will lead to less spot contracting. Clyde Russell again:

The question is whether utilities, such as Japan’s JERA, continue with their long-term vision of moving more toward a spot and short-term market, or whether the old security blanket of oil-linked, but guaranteed, supplies regains some popularity.

It’s likely LNG buyers don’t want a repeat of the recent extreme volatility, but perhaps they also don’t want to return to the restrictive crude-linked contracts that largely favoured producers by guaranteeing volumes at relatively high prices.

The compromise may be the increasing popularity of short-term, flexible contracts, which can vary from a few months to a few years and be priced against different benchmarks.

Well, maybe, but color me skeptical. For one thing, contracts require a buyer and a seller. Yes, buyers who didn’t have long term contracts probably regretted paying high spot prices–but the sellers with uncommitted volumes really liked it. The spike may increase the appetite for buyers to enter long term contracts, but decrease the appetite of sellers to enter them. It’s not obvious how this will play out.

I note that the situation was reversed in 2020–buyers regretted long term contracts, but sellers were glad to have them. Ex post regret is likely to be experienced with equal frequency by buyers and sellers, so it’s hard to see how that tips contracting one way or the other.

This conjecture about the price spike leading to more long term contracting also presupposes that the only way of managing price risks is through fixed price contracts (or oil-indexed) contracts for physical supply. But that’s not true. Derivatives allow the separation of who bears price risk from the physical contracting decision. A firm buying spot (and who is hence short LNG) can hedge price risk by purchasing JKM swaps. This has the additional advantage of allowing the adjustment of the size of the hedge in response to more timely information regarding likely quantity requirements, price projections, and risk appetite than is possible with a long term contract. That is, derivatives permit unbundling of price risk from obtaining physical supplies, whereas long term contracts bundle those to a considerable degree. Moreover, derivatives plus short term/spot acquisition of physical supplies allows more flexible management of supply, and management of supply based on shorter term forecasts of need: these shorter term forecasts are inherently more accurate than forecasts over contracting horizons of years or even decades.

So rather than lead to more long term contracts, I predict that this recent price spike is more likely provide a fillip to the LNG derivatives market. Derivatives are a more flexible and cheaper way to manage price risk than long term contracts.

This is what happened in the pipe gas market in the US post-deregulation. Spot/short term volumes grew dramatically even though price spikes were a regular feature of the market: market participants used gas futures and swaps and options to manage these price risks, and benefited from the greater flexibility and precision of obtaining supplies on a shorter term basis. This shifted a lot of the price risk to the financial sector–which is the great benefit of the much bewailed “financialization” of commodity markets.

The same is likely to occur in LNG.

For 5+ years I have been calling oil-linked LNG contracts a “barbarous relic” (echoing Keynes and the gold standard). This opinion was widely castigated when I first voiced it, but I had a strong basis for it: oil and LNG prices are driven by totally different fundamentals, and as a result oil prices are almost always disconnected with gas market fundamentals.

My presentations include a graphic and some statistics to make the point: correlations between oil and gas prices (measured by the Japan-Korea Marker, or JKM) are about zero, and so are the correlations between gold and gas prices. So it makes about as much sense to index LNG contracts to gold as it does to index them to oil.

Platts has released a report making the same point:

Asian utilities buying liquefied natural gas under rigid long-term contracts linked to oil prices risk paying an average of $20bn more each year up to 2022 than if they bought the superchilled fuel directly in the market.

S&P Global Platts, which sets the benchmark LNG price for the region, said there was a “huge disconnect” between the of cost oil-linked contracts and prices in the so-called “spot market” for LNG.

Nice of you to catch up, Platts. This “huge disconnect”–the exact phrase I’ve used in public presentations for years–has been evident in the data . . . for years. Indeed, since the beginning of the JKM index.

Platts spins this as utilities paying too much. It wouldn’t be any better if they paid too little. The point is that divergences between contract prices and product values cause misallocations of resources, including transactions costs associated with trying to circumvent contracts with prices that are misaligned with fundamentals.

There are still some who are in the thrall to the barbarous relic:

Jera, the world’s biggest buyer of LNG, said that oil-linked contracts still often made sense. The spot market was still small and vulnerable to volatile spikes as seen in early 2014, it said, while long-term contracts offered “stable procurement for buyers”, as well as providing a guarantee of demand needed by developers to launch new LNG projects.

This is a jumble of sloppy thinking.

Volatile spikes? If the spikes are driven by gas market supply and demand conditions, tying contract prices to gas prices rather than oil is a feature not a bug. You want prices that reflect fundamentals, and if fundamentals are jiggy, so be it. And, er, last time I checked, the oil market was subject to “volatile spikes” despite its greater size. And this is a bug, not a feature, when it comes to LNG because these oil price spikes are almost always unrelated to gas supply and demand conditions.

That is, it is better to tie LNG contract prices to markets with jumps that reflect gas fundamentals, than it is to tie prices to markets with jumps that don’t.

There can be a role for long term contracts, though the transactions cost (“opportunism”) related reasons for such contracts become less important as the short-term market becomes more liquid. Furthermore, even to the extent that long term contracts facilitate investment, it is a non sequitur to conclude that these long term contracts should be indexed to oil, rather than a gas price (be it JKM, or Henry Hub, or TTF, or whatever). The contracts will perform better–transactions costs will be lower–the lower the likelihood that pricing terms become misaligned with fundamentals. Gas indexing results in lower probability of misalignment than oil indexing.

For years, the mantras in the LNG market have been “security of supply” and “security of demand”, and that long term contracts are the only way to secure it:

David Thomas, an independent adviser with experience at Vitol and BP, said that Japanese utilities “value security of supply and there’s a strong relationship between buyers and sellers” but that was changing as the LNG market became more liquid and Japan’s gas and power markets liberalize.

The oil and refined product markets rely on markets for security of supply and demand. A big refinery or a major oil development are as, or more, capital intensive than an LNG product. But these things get created without long term contracts because liquid markets allow transactions at prices that reflect supply-demand fundamentals.

Long term contracts play a role when spot markets aren’t “thick” enough: under these conditions, opportunism (“holdup”) is a problem, and long term contracts can mitigate that problem. The evolution of spot markets in LNG will progressively mitigate the potential for holdup problems. And as I’ve noted, this is a virtuous cycle: as spot markets become more liquid, the need for long term contracts will decline, which will contribute to greater spot market liquidity, which will vitiate further the need for long term contracts.

As the producer of the JKM index, Platts is obviously talking its book here. But that’s not to say that it’s wrong. It isn’t. And economic reality will inevitably push contracting practices in the LNG market away from the barbarous past. Or perhaps I have the tense wrong there: is pushing would be a better way of putting it.