Mobile Edition
Print
Get the Mag
Weekly Updates1. INTRODUCTION
Two main factors governing the policy debate on the European natural gas market are the decline of indigenous resources, combined with the growing dependence for a large share of gas supplies on a few foreign gas exporters, and the market power this may confer. In one to several decades, gas resources of major EU producers UK and the Netherlands are forecast to dry up. To meet growing demand, three large exporters to Europe--Russia, Norway and Algeria--will see their roles, and perhaps the associated pricing power, enhanced.
An important potential competing source of supplies will be imports of liquefied natural gas, LNG. LNG suppliers increasingly compete on a global market, shipping their production to the highest price bidders. Conditions in other consumer markets will therefore set prices for LNG imports, and such prices will largely be exogenous to regional European developments.
We analyze the relations between market power, resource depletion and LNG imports in Europe in a simulation model, NATGAS, that includes both the effects of resource constraints and of market power at the production level. Finite resources lead to interdependencies of current production decisions and future opportunities. Depletion decisions in turn interact with the potential for large producers to set market prices above marginal costs, where these costs themselves should be interpreted to include opportunity costs associated with conserving resources for future production.
Within this framework we study the impact of different assumptions on the availability of LNG. In the face of varying LNG imports, what substitution patterns do we observe from indigenous production, external pipeline supplies and demand? One key observation is that although total available indigenous resources are fixed and finite, the changing long-run price perspectives will lead resource constrained European producers to shift their production profiles over time. At any given moment, these intertemporal constraints may cause a player's production to function either as a complement or as a substitute to contemporaneous LNG imports.
The NATGAS model is a mixed complementarity model of Cournot producers in a network setting. It builds on a sequence of earlier simulation models of the EU energy markets. Golombek et al. (1995, 1998) used a Cournot model of the European gas market to assess the impact of liberalization on prices in Europe, while Mathiesen et al. (1987) compared Cournot, competitive and collusive equilibria in the European market. Haurie et al. (1987) studied an application to a market with stochastic demand. (1) In Boots et al. (2004), the impact of traders in a conjectural variations production market is investigated. More recent contributions focussing on producer market power in the European gas market include Egging and Gabriel (2006), Holz et al. (2008), Egging et al. (2008), Lise et al. (2008), and Zwart and Mulder (2006). Related recent Nash-Cournot models of the US market are studied, for example, in Gabriel et al. (2005a,b).
Modelling of (endogenous) longer-term dynamics in network models of natural gas markets is still underexplored. An early discussion on how Cournot network models could be applied to the European natural gas market was provided in Flam and Zaccour (1989), though here the idea was not yet implemented in practice. Dynamic investment in electricity market models was addressed by Denis et al. (2002), Pineau and Murto (2003) and Murphy and Smeers (2005).
These authors all focus on investment in production capacity. In NATGAS, we combine this approach to investment with the finiteness of resources (see e.g. Withagen, 1999, for an overview of literature on oligopoly power with exhaustible resources). An application of intertemporal optimization under Cournot competition in energy markets is given in Bushnell (2003), in a complementarity model of electricity markets including hydropower generation.
The structure of this paper is as follows. We first sketch the structure of the EU natural gas market and its suppliers. We then briefly describe the key characteristics of the NATGAS model. In section 4 we analyze the patterns of production and imports in various scenarios on LNG costs and market power. We close with some concluding remarks.
2. GAS SUPPLIES TO EUROPE
Currently, the major EU producers are the UK and the Netherlands. Denmark, Germany, Italy, Poland and Romania play smaller roles, both in terms of production and reserves. By the end of 2006, total indigenous remaining reserves were approximately equal to 2800 bcm (billion cubic meters, BP, 2008), or less than six times annual EU consumption. Total annual EU production was slightly over 200 bcm in 2006 (IEA, 2007), and EU consumers depend to a large extent on pipeline imports of natural gas from, in particular, Russia, Norway and Algeria.
Even though the corresponding reserves to production ratio suggests that in about 14 years the EU will entirely depend on imports, the situation is slightly less gloomy if one takes into account the additions to reserves from gas exploration activity. Indeed, as Stern (2002) points out, proven reserves in the EU stayed more or less constant over the years 1981-2001, in spite of continuous production: new reserves were added at a similar pace. More important is the quantity of total resources, including those too expensive to exploit at current prices and technology, and those currently yet undiscovered. Though by their nature such resource estimates are substantially more uncertain than figures for proven reserves, both known but undeveloped finds and surveys of suitable geological structures do allow for some indications of their magnitudes. For example, the Department of Business, Enterprise and Regulatory Reform (BERR, 2007) provides a central estimate of total remaining UK resources of 1313 bcm of natural gas, compared to a proven reserves figure of 684 bcm and past cumulative production of just over 2000 bcm of natural gas. And similarly, the midpoint estimate for resources in the Dutch Continental Shelf small fields equals about twice their proven reserves (EZ, 2008). Seeliger (2004) lists values for other European producers.
Even with these figures of remaining resources, EU production is widely expected to decline substantially in the coming decades, in particular for the two major producers, the UK and the Netherlands. Because of still growing demand for natural gas, the need for increased imports from outside the EU is apparent. Fortunately, reserves and estimates for remaining resources in the EU's suppliers are huge: for instance, remaining Norwegian resources are estimated at almost 5000 bcm (NPD, 2008), and Russian proven reserves alone equal about ten times this figure (IEA, 2007). For the EU, a more pressing question is at what price these resources will be made available. Since the 2007 Norwegian Statoil--Norsk Hydro merger, state controlled monopolists have dominated production in all three major gas exporters to the EU: Norwegian StatoilHydro, Sonatrach for Algeria and Gazprom for Russia. It is natural to assume that these firms will, to some extent, use the market power they are endowed with. Also within Europe, entry into the producing sector is typically restricted, and producers are not necessarily price takers, though levels of competition in the various countries will differ. In this paper we will indeed assume that production can be described by an oligopoly model.
As indigenous resources dwindle, will the European gas market essentially become a triopoly, perhaps complemented by smaller pipeline supplies from other North-African or former Soviet Union states? Not necessarily: supplies from elsewhere in the form of liquefied natural gas, LNG, may contest this market, depending on prices in Europe and other consumer regions. The global market for LNG has experienced substantial growth over the past decade, with annual growth rates of over 7% (Elkins, 2008). Currently, LNG only caters for a small share of total gas demand in the European market, around 50 bcm out of total gas demand of over 500 bcm. Although this share is expected to increase, LNG consultant Jensen Associates (2007a) indicates that projections of LNG growth are very uncertain, and estimates a range for the global LNG market in 2020 of 400 bcm to 600 bcm (compared to some 190 bcm in 2005). His estimate for the share going to Europe displays even greater uncertainty, ranging from hardly any growth (if pipeline imports continue to be relatively cheap) to 170 bcm in the most positive scenario.
LNG is more flexible in destination than pipeline gas. Liquefaction facilities in source countries make up the larger part of costs in the LNG chain, while on the other hand, regasification terminals in destination countries account for a much smaller share of the costs in the chain. Indeed, in this latter segment there is overcapacity. Global regasification capacity is roughly twice as large as global liquefaction capacity.
Differences in shipping costs of LNG to various demand centers (Europe, the US, and the Far East) are, at one to a few cents per cubic meter, relatively minor compared to gas prices. Therefore, a producer such as Qatar or Nigeria may choose a destination for its gas based on the highest net-back price. As a result, provided LNG prices are competitive with costs of piped gas, LNG could give rise to arbitrage between these various global regions. Although currently, the LNG market is still dominated by relatively inflexible long-term contracts allowing little scope for interim renegotiation of the cargo's destination, the market for flexible LNG is growing. As Jensen Associates (2007a) documents, flexibility of destination might grow as spot markets for LNG grow in importance, but also as more and more LNG contractors become themselves active in multiple downstream markets, and can arbitrage within their own portfolios. Jensen Associates (2007b) estimates current flexible volumes in the Atlantic Basin LNG market at some 30% of total volumes.
FEED