hub Pricing in Europe - An Argument For Gasage

Transcription

1 Dear Conference Participants, Dear Mr. Stern and Mr. Rogers! The first thing I would like to do is to apologize for calling you consultants in the research paper I issued on January 11, In calling you consultants - a reference that you vigorously oppose - instead of independent academic researchers, I had no intention of challenging your objectivity. But to be frank, I have not known of any consultant who claims that his research is not original, that it is dependent on a client s opinion, and that it is not insightful at all. Also, I have never considered consultant to be a rude term because, in my opinion, only the job of consultant opens up legitimate access to proprietary commercial information such as the terms and provisions of long-term gas contracts. Having access to this proprietary information serves as an antidote against the purely scholastic and blind application of textbook theories to the more complicated realities of today s gas market. To continue, I apologize for the improper use of the word consultant, a mistake which can be explained in part by the fact that I am not a native English speaker and may not well understand all the nuances of the language. I promise to use the term academics instead of consultants in the updated version of my paper on pricing that I plan to release soon. Let me also make another remark. If you proudly call yourselves independent academics, it does not decorate you with a tag of excellence or make your research superior to that delivered by the modest consultants or the analysts that come from the corporate world like me. The fact that I represent Gazprom does not mean that my conclusions are not objective or that they are inherently biased. I hope you will agree with that statement of mine. A discussion based on uncovering the vested interests that lie behind our research conclusions is not supportive of any form of constructive dialogue. Rather, the exclusive intention of my paper was to provoke a constructive discussion of the quality of the price signals that originate on European gas hubs. 1

2 Any doubts concerning the quality of price signals originating at gas hubs typically have been due to questionable price data collection procedures. The quality of price signals stemming from real-world transactions is clearly greater than market price quotations derived from bids and offers by way of an expert s assessment. There is no disagreement over that. We fully support the practice of hub price reporting based on real-world transactions rather than assessments by market participants, which can lead to price manipulation. Hubs play an important supplementary role in today s European gas market. In the near absence of make-up gas or carry-forward arrangements in the end-user market, they play a major role in balancing supply and demand. Hubs also support arbitrage operations of various kinds. But there are other, less visible aspects of hub price quality that need to be examined in detail. My major disagreement with Mr. Stern and Mr. Rogers is over what forces set the baseline trend for hub prices in Europe. Until recently, the Oxford Institute for Energy Studies showed no interest in discussing this subject because the answer was evident to the academics that the forces of supply and demand set the baseline trend for hub pricing in Europe. I consider this question to be extremely important to the European gas industry. From the viewpoint of the gas industry, it is as important as that pivotal question posed by astronomers many centuries ago: What is the orbital center of all celestial bodies the earth or the sun? Obviously, the correct answer to that question was an important achievement for all of mankind. Although the combination of forces that set hub prices is extremely complex, this should not prevent us from indentifying the main driver of hub prices. When specified and ranked by their dominant position, the choice is between only two options: The oil-indexed prices of long-term contracts; and Supply and demand for gas on the European market. Mr Stern and Mr Rogers, as I have already mentioned, claim that supply and demand fundamentals largely govern hub price behavior in Europe. As they consider hub prices to be self-contained, competitive, and largely free from influence by the fundamentals of other markets - and in fact ready to take over from oil-indexed prices - any influence that contract prices might have on hub-based pricing was considered out of the scope of their research. The intention of my paper was to point to the collective failure by industry participants - including some producers, politicians, journalists and the public at large and even the academics - in understanding the unique nature of hub prices within the European hybrid pricing system. If you look at these hubs, you will see a lot of activity, volumes, and reasonable price alignment. 2

3 However, for many years visual observations also supported the idea that Earth was the center of the universe. But there turned out to be a significant difference between these visual illusions and cosmic reality. It is the job of academics to differentiate illusions from realities. But these respected academics have failed to accomplish their mission because they have taken without due verification the assumption that hub prices in Europe represent a true market in nature and accurately reflect changing supply and demand conditions. Our analysis shows that the prices of oil-indexed long-term contracts determine the baseline trend for hub prices. NBP and TTF prices are not only reasonably aligned with each other but they also have a strong positive correlation with Gazprom s oil-indexed prices, with coefficients of 0.75 and 0.79, respectively. In real-life terms, this means that the baseline curve for spot prices is determined not by supply and demand dynamics at the hubs but by the oil-indexed contracts themselves. Supply and demand only mutate this price. The higher the contract prices due to oil-price escalation, the higher the hub prices. It would be correct to say that existing hub prices are a double derivative. While our contracts are derivatives of the oil product price, hub prices in their turn are derivatives of that derivative. The validity of a price signals originating at these hubs is therefore subject to reasonable doubts. Indeed what driver other than rising oil prices caused the growth in hub prices after October 2009? According to the conventional logic of mainstream industry analysts, hub prices that accurately reflect changing supply and demand conditions should be moving in the opposite direction (i.e., downward) as a result of the slump in demand triggered by the economic crisis and the massive inflow of LNG redirected from the U.S. Let me repeat it once again: in the existing hub pricing model there is only one benchmark: oil-indexed gas. Other gas prices are merely derivatives of this one benchmark. Once the midstream entities who typically hold long-term contracts with multiple suppliers exercise their arbitrage options, they set a soft price ceiling for the entire hub market. That soft ceiling, in fact, is a solid price structure. Hub prices in the mature hybrid pricing system that emerged in Europe after 2007 behave in a structured way and settle at a discount to contract prices, only occasionally equaling the contract price. The discovery of horse meat in products labeled as beef in the EU was considered to be a massive fraud. Yet the re-discovery of oil indexation as the major driver behind the pricing of gas on hubs that are perceived to be purely market-based and independent brings no reaction from the European authorities. This can be explained by the fact that hub prices tend to be lower than contract prices (I will come back to this issue later on). And as our clients say to us: we don t care about the origin of the hub prices, but we like them because they are cheaper. 3

4 To call European hub prices market-based is a gross inaccuracy; for such hubs do not perform at all like hubs in the U.S., which are a true indication of supply and demand fundamentals. Hub prices in Europe are not a function of total supply and demand or even a large segment of it. Although hub prices are reflective of supply and demand conditions, they play the role of balancing the residual volumes that remain after long-term oil-indexed contacts have met the bulk of demand. Therefore, European hubs function not to produce an equilibrium price for the whole market, but rather to perform a function of balancing and arbitrage of all kinds, between different contract pricing structures, between contract and spot prices, between hubs, between the UK and the Continent (see Exhibit on the slide for a schematic description of how the European hub hybrid pricing system operates). Therefore those producers who offer hub-based pricing may feel comfortable with existing price levels, mistakenly thinking that they are indexing to the equilibrium price for the gas market. In fact, however, they are crossreferencing their own oil-indexed prices or what remains of them. Advocates of supply-and-demand based gas pricing may try to counter my arguments by pointing to a simple and effective way of improving the quality of price signals coming from European hubs. By regulatory and other means (for instance, customer demands exclusively for hub gas), it is theoretically possible to displace oil-indexation from its dominant position in contracts by increasing the share of gas-indexation in pricing formulas. 4

5 This may not even be necessary if you take seriously the research done by Société Générale on the role of hublinked pricing, which concludes that oil-indexation as a dominant pricing formula has disappeared once and forever. According to Société Générale analyst Thierry Bros, a major revolution has already taken place in the European gas market. This revolution is exemplified by the fact that, as per Mr. Bros, oil-indexed gas as a percentage of the total gas market slipped from 58% in 2011 to 55% in 2012 and that by 2014 oil-indexed pricing should represent the minority of European gas supply. However, I would caution you not to hurry to the conclusion that European gas hubs have begun originating market-equilibrium prices. On several occasions we brought to Mr. Stern and Mr. Rogers attention the fact that the gas industry has two tiers with different pricing structures, and that it is incorrect to count the gas that comes in under import contracts along with the gas that is bought under downstream end-user contracts. IGU made a similar mistake which also resulted in overblown numbers for spot-indexed gas deliveries in Europe. Nevertheless, these respected academics continue to promote the results of this study as if there were no other studies on this subject. In my opinion, the results of one possibly inaccurate study should not be taken as the gospel truth. Based on data accumulated from gas supply companies, energy consultancies, and in-house research, Reuters published a report in March 2013 demonstrating that only 34.8% to 37.7% of all major European gas supplies are now priced off traded hubs. The American research company PIRA Energy Group came to the same conclusion as Reuters in its own February 2013 study, which estimated that roughly 2/3 of European gas consumption is still oil indexed. The game is not over. Three out of the four major suppliers of gas to Europe still support oil indexation and would prefer to sacrifice value rather than give on their principles and long-term interests. The term Potemkin village was originally used to describe a fake village that was built only to impress. The phrase is now often used, typically in politics and economics, to describe any construction (literal or figurative) built solely to deceive others into thinking that a situation is better than it really is. I recently came across the speech of one high-ranking European politician who presented as a major political victory the fact that spot gas now accounts for more than half of total gas deliveries. Believing in and basing market conclusions on such unverified numbers is an untenable strategy. 5

6 Gazprom Export s own estimates are pretty in line with Reuters and PIRA conclusions. Hub indexed gas accounts for 30.5% of the final consumption, and 27.3% of imported gas. 6

7 An important question to ask is: Why are hub prices generally lower than contract prices? Hub prices are not independent. They are derivatives of the contract prices that set a baseline trend for their behavior. The principal difference between the two prices is the quality of the products offered. Gas that comes under long-term arrangements represents a more valuable product than hub gas because it offers a combination of supply security and delivery flexibility. In fact, what Gazprom brings to its clients is more than a commodity it is gas plus the related services necessary to deliver a secure and flexible source of supply. That is why our weaknesses higher prices are merely indicators of our strengths. The major reason for price divergence as it is singled out by the market is a combined value of security and the flexibility provided by longterm pipeline suppliers. Hubs offer no flexibility but a commodity in fixed sizes that requires time and effort to structure its delivery to the end-user. In many cases, European hubs are not liquid enough and cannot offer security of supply comparable to these long-term contracts. In situations of short-term under-supply, it is more convenient and cheaper (when prices do not differ much) for wholesalers to rely on existing long-term contractual arrangements to secure additional deliveries. That is why hub prices do not cross the contract price line for significant periods of time. That line becomes a real line on the sand. When hub prices approach contract levels, consumers cease buying gas at the hub and instead switch to contract deliveries. When prices exceed the contract level buyers immediately react by arbitraging their product. Arbitraging brings spot prices below the contract level. 7

8 Another important question to ask: What is a combined premium for flexibility and security embedded in the oil indexed price compared to the hub price? Our estimates shows that the size of this premium fully covers the average gap between the two prices. A major difficulty comes in trying to evaluate the premium for security of supply that is embedded in the oil-indexed price. One method for calculating this premium is to assume it is equal to the fine Gazprom must pay if it fails to meet its clients obligations. But the real value of contract s security is higher due to other factors, such as the use of pricing arbitrage on hubs. This is possible when the volume of gas delivered under long-term contracts is reduced in favor of purchasing spot gas when hub prices are depressed. Offering security of supply in combination with contract flexibility gives special benefits to the gas buyer. Let us compare two types of contracts. One contract is flexible with a minimum annual quantity ranging from 75% to 85%. That contract fully covers the demand of a given client. This means that 15% to 25% of volumes can be bought on the spot market when prices there are lower or taken from the long-term contract if hub prices are high. Compare this with a flat contract without flexibility that covers 75% to 85% of demand but forces the buyer to purchase from the only hub for the remainder of its requirements. Contracts offering a delivery guarantee allow the buyer to take full advantage of any available arbitrage opportunities. In fact, based on historical data from the past two years, these arbitrage opportunities can save the buyer anywhere from US$7-9/mcm. 8

9 We used the cost of one full-cycle of underground gas storage under a typical one-year contract as an estimation of the premium built into the long-term contract price to account for seasonal supply flexibility. This cost covers working gas capacity, injection, and send-out. One full cycle of gas storage is defined as the sequence of injection and withdrawal of the maximum volume of working gas. The cost of one storage cycle was defined as the annual cost divided by the number of cycles per year in order to minimize the frequency of cycle factor. Our analysis shows that the average cost of one storage cycle in Continental Europe for 29 tariff plans in 26 storage locations is approximately US$64/mcm. Obviously, not all gas bought under flat-take contracts needs storage: a portion of the volume is cycled through gas storage while another portion is consumed directly. Based on seasonal consumption patterns for several European countries and assuming that over the year the volume of gas pumped into underground storage equals to the volume of withdrawals, minimum savings on seasonal storage equals to roughly US$20/mcm. In case of a flat contract, the gas buyer has to have access not only to seasonal flexibility but to short-term flexibility too. It therefore has to pay for flexible transportation capacity. Based on tariffs published by Gasunie, we estimate the additional fee for the underutilized transportation capacity to be about US$14/mcm. As you add all these numbers together you will cover fully the gap between contract and hub prices. We assume that this flexibility premium should be growing in importance with time as demand for gas becomes increasingly volatile, due to more unpredictable weather conditions, the increased role of renewables in power generation, and temperature-driven demand for gas in the residential sector. These do not represent an exhaustive list of all the factors that explain the positive difference between long-term contract and hub prices. But they should provide some much-needed context to the pricing debate, which until now has not fully reflected all of the market dynamics at work. Gazprom offers a premium quality product compared to hub gas. If we make contract and spot prices comparable simply by lowering our contract prices and making no adjustments to the contract, we are giving up value. In this case Gazprom still must incur the substantial additional costs of providing supply flexibility in our long-term contracts without any reciprocal benefits or tangible rewards. Moreover, by giving our approval to further decreases in the oil-indexed contract price, we initiate a new cycle of downward adjustment in the hub prices, which are derivatives of contract prices, due to the feedback loop shown on this slide. These adjustments in turn trigger new requests for contract price revisions from buyers. 9

10 What happens when demands to reduce the price of a superior quality service to a price of its lower quality substitute are implemented could be best illustrated by the example from the rental car business. If clients and/or regulators persistently require that the price of renting a car with a driver should not be higher than the price of renting a car without a driver, service providers could meet these absurd requests for a while. But at a certain point, they will take this superior quality service away from their offer list. If our clients, according to what they tell us, cannot recover the value of the higher priced product (gas with flexibility and security premium) from their own customers, this highly-demanded service will disappear from the market, despite its evident value. Our colleagues from Statoil have already made service quality adjustments part of their practice of contract modification. More spot gas means that we take that flexibility back, they say. Viable long-term sales contracts offer benefits to both buyers and sellers and are based on equitable risk-sharing between the two parties. Old and existing oil-indexed contracts that have effectively served the European gas market for nearly 40 years fully meet this viability definition. Shippers take responsibility to invest in upstream infrastructure in order to deliver natural gas for their clients needs in accordance with their daily nominations. If shippers fail to meet these obligations, they are subject to serious fines. On the buyer s side, there are volumetric risks (take-or-pay obligations) and price risks which they also share with the shipper. These arrangements give the shipper a guarantee that his investment will be paid off. The buyer also has a right to price revisions in cases where cheaper gas offered by other producers becomes available locally. In contrast to our Norwegian colleagues, we cannot accept the proposal to convert our long-term contracts to 100% hub indexation. This new model for long-term contracts suggests shifting the balance of risks in favor of the buyer. Although such risk-sharing schemes already exist in some contracts, we doubt the viability and stability of a pricing system in which such contracts would play a dominant role. In the case of 100% gas hub indexation, any take-orpay obligations on the buyer s side lose their function as a guarantee of demand security because buyers can dispose of excess volumes on hubs with no risk to their revenues. Producers from third countries will run the intolerable risk of gas price erosion because there is virtually no force in Europe interested in preserving the value of natural gas. 10

11 The specific nature of natural gas price signals in Europe is demonstrated by the low churn ratios at Continental hubs. In order to produce sustainable price signals, the churn ratio should be over 8, according to Mr. Stern and Mr. Rogers. In Europe, only the NBP meets this condition. Some analysts say that low churn ratios on the Continent are a reflection of the transition phase, and that, as hub markets mature, churn ratios will grow. We are pessimistic in this respect. It is not because European financial institutions are reluctant to play with the forward curve. Rather, it is extremely hard to predict what the price on a balancing market will be in two or three years time because these prices are not about supply and demand but about arbitrage opportunities. In this respect, it is a mature market already and yet quite distinct from the U.S. model. These arbitrage opportunities are largely the result of the hybrid system in which they arise, and they are by nature transitory, location-specific, and independent of aggregate supply and demand: and thus hardly the basis for rational longer-term price discovery. 11

12 Over the past ten years, despite their conflicting fundamentals, a broad range of exchange-traded commodity groups, including metals and agricultural and chemical products, has grown in price by 2.3 to 3.3 times in U.S. dollar terms. This is within the same range 2.9 times as the oil-indexed gas price as measured by the German BAFA import price index. In fact, when measured in calorific terms, over the past ten years oil-indexed gas prices in Continental Europe have grown at a lower rate than other energy commodities such as coal and electricity. What this tell us is that when a product s pricing dynamics match those of other raw commodities including those that are necessary inputs to the manufacture of that product it gives investors confidence that they will be able to successfully launch a new investment cycle. By contrast, negative growth in the Henry Hub gas price has put long-term investments in dry gas production at risk in the U.S. because producers must buy the commodities necessary for their investments at rising prices. 12

13 13