Mobile Edition
Print
Get the Mag
Weekly Updates1. INTRODUCTION Natural gas has traditionally been traded in regional markets, like the European, the North American, and the Asia-Pacific markets. Most of the international gas trade to and within Europe as well as North America has been transported in pipelines, whereas the Asia-Pacific market has been dominated by LNG (Liquefied Natural Gas) transport. (1) Over the last decade the costs of LNG have been significantly reduced, more producers have entered the gas market in general and the LNG market in particular, and the trade between continents has increased (IEA 2006a, 2007). IEA (2007) states that LNG accounts for 70% of the growth in inter-regional trade since 2004, and LNG trade volumes are expected to more than double before 2015. This trend reflects the fact that gas resources to a larger extent are located far from the main consuming regions, increasing the attractiveness of LNG. For instance, whereas the ratio of reserves over annual consumption is 10 years in North America (the largest single gas market), the ratio is almost 300 years in the Middle East (BP, 2007). In the EU, the ratio has fallen from 9 to 6 years over the last decade.
The reduction of transport costs over the last 10-15 years, especially for LNG, is one of the main driving forces for the growing globalisation of natural gas markets. Brito and Hartley (2007) argue that the change in market structure promoted by transport cost reductions can be much accelerated by endogenous expectations about how the market evolves. In this paper we investigate how lower transport costs for natural gas may affect the international gas markets. We explore how the natural gas prices in different regions change when the costs of LNG or pipeline transport are reduced. Except in the case with only two regions, the effects are not straightforward. Using both theoretical and numerical tools, we show that counterintuitive effects may appear. That is, import (export) prices may possibly increase (decrease) when transport costs are reduced, and price differentials between two import regions may increase. The outcome depends largely on transport distances and choice of transport technology. As a result, the relative cost development between pipeline and LNG may be highly important for future relative natural gas prices in different regions.
The LNG chain consists of three separate cost components (besides extraction). The most expensive ones are liquefaction and shipping costs. (2) According to EIA (2004), liquefaction costs decreased by 35-50 per cent from the early 1990s to 2003. Greaker and Sagen (2008) find that the cost reduction may be more due to increased competition among liquefaction technology suppliers than to technological progress. On the other hand, Jensen (2003) emphasizes economies of scale from larger liquefaction trains. The observed cost increase for new liquefaction plants since 2003 is mainly a result of rising raw-material prices and tight contracting market, not least in the LNG industry (Petroleum Economist, 2007). Jensen Associates (2007) refers to the current situation as "aberrations resulting from a heavily overheated construction industry", and expects liquefaction costs to fall further in coming years (see also IEA, 2003).
The unit costs of LNG shipping have been reduced by 40 per cent over the last decade (Brito and Hartley, 2007). This is primarily due to larger average cargo volumes and reduced costs of input factors in the construction of LNG ships (EIA, 2004). The LNG tanker fleet increased by 75 per cent from 2000 to 2005 (IEA, 2006a). Offshore pipeline construction costs also fell substantially during the 1990s because of technological advances - by about 50 per cent in the United States (IEA, 2003). Onshore pipeline costs increased in the same period because of higher labour costs.
In our analysis we assume that the international gas markets become liberalised and integrated, so that transport cost reductions generate a new market equilibrium with modified trade pattern and price levels (the qualitative results also hold under other market conditions, see below). However, whereas the North American and the UK gas markets are deregulated and prices have been linked to liquid spot markets for almost two decades, most of the European and the North-East Asian gas markets are still indexing the price of almost all their gas volumes to various oil products through long-term contracts. As gas competes in the enduser markets with other energy goods such as oil and coal, which are globally traded, there are indirect links between regional gas markets even in the absence of gas trade. As seen in Figure 1, prices in North America and Europe have been highly volatile through the years, with no clear correlation between them in the short term. This can be explained partly by the rigidity of the fixed contracts and partly by the high transportation costs. Nevertheless, whereas Siliverstovs et al. (2005) find no sign of price integration between the North American market and the European/Japanese markets in the period 1994-2003, Neumann (2007) finds increasing convergence of spot prices in the US and the UK based on data for 1999-2007. (3)
[FIGURE 1 OMITTED]
Although several obstacles remain, the current trends are clearly directed towards globalisation of gas markets. International spot trade of gas is growing rapidly, by a factor of 10 since 1998 (IEA, 2006a). Although it accounted for just 11 per cent of international trade in 2004, it is expected to double in a few years. Spot trade makes it easier for gas sellers to enter markets where long-term contracts previously have dominated the trade volumes. Hence, short-term price differentials may be exploited, possibly leading to increased price convergence between regions. A moderate level of spot trade based on arbitrage may be sufficient to balance the markets in different regions (Jensen, 2004). So far, most arbitrage trading has occurred in the Atlantic Basin, but recently the Middle East has become a swing supplier to both North-East Asia and the Atlantic Basin. Alongside increased spot trade, gas markets are gradually becoming more competitive and deregulated, also outside North America and the UK. This is particularly so in Europe, where the EU has adopted two directives on gas liberalisation over the last decade (EU, 1998, 2003).
Our study relates somewhat to the theoretical literature on pricing and regulation of gas transport. There exist a number of studies on this issue, see, e.g., Newbery (1999), Cremer and Laffont (2002) and Mizuno and Shinkai (2006). Cremer et al. (2003) derive a number of theoretical results regarding first- and second-best tariffs in a three-node network (denoted 'triopolistic' market). First-best tariffs equal marginal transportation costs with a mark-up when the line is congested, whereas second-best tariffs also involve Ramsey corrections.
In our paper we first derive a theoretical, triopolistic market model to study the effects of lower transport costs on natural gas prices. Although the analysis is quite simple, it offers new insight into the literature on gas transportation. So far, this literature has mainly focused on network regulation and tariff design, which are particularly relevant for pipeline transport. LNG offers more flexibility than pipelines in transport services, as trading is less dependent on site-specific capacities (regasification capacities are seldom fully utilised, cf. Jensen, 2003). Thus, LNG makes it easier to exploit arbitrage opportunities, increasing the relevance of this study.
Second, we apply a detailed numerical model of the international gas markets to investigate the effects of transport cost reductions for either LNG or pipelines. In line with the theoretical analysis, we focus on gas prices and particularly examine whether or not prices in the main import regions fall. We also look at the effects on price convergence between import regions. The model (FRISBEE) distinguishes between 13 geographical regions, and specifies bilateral trade between each pair of regions. It also has a detailed modelling of gas production, based on gas endowments in the different areas.
Earlier numerical analyses of natural gas markets have mainly focused on one specific region, which is not surprising given the separation of regional markets historically. Golombek et al. (1998), Boots et al. (2004) and Egging and Gabriel (2006) analyse different aspects of market power in the European gas market (including Russia and North Africa). Hirschhausen et al. (2005) and Sagen and Tsygankova (2008) examine how respectively Russian transport strategies and domestic price policy affect the export of gas from Russia to Europe. MacAvoy and Moshkin (2000) and Gabriel et al. (2005) simulate the North American gas market, focusing on further deregulation and the potential for market power, respectively. EMF (2007) presents future scenarios of the international gas markets based on several global and regional simulation models. (4) We are not aware of any other published papers that simulate the global natural gas markets. The numerical analysis we present are therefore timely given the globalisation process noted above. Moreover, no studies have looked at the effects of lower transport costs in a global context, except for Brito and Hartley (2007) mentioned above.
The theoretical analysis is presented in the following section. In Section 3 we briefly describe the numerical model FRISBEE, before we present the simulation results in Section 4. Finally we conclude.
2. THEORETICAL ANALYSIS OF TRANSPORTATION COSTS AND GAS PRICES
In this section we introduce a simple theoretical approach to study how transport cost reductions may affect gas prices in different regions of the world. We will take into account that gas may be transported in different ways (LNG or pipeline transport), so that cost reductions may or may not affect all gas trade. Moreover, some transport costs are increasing with distance (pipeline and shipping costs), and some are not (liquefaction and regasification). We use a classical three-node model. This is also used in e.g. Cremer et al. (2003), who discuss how the distances between the sub markets influence on the efficient transportation charges. Although our focus is not on tariffs, we show (among other things) that transport distances are very important for how regional gas prices may react to transport cost reductions. Although the global gas market consists of more than three regional markets that produce and consume natural gas, the mechanisms revealed in our theoretical model are of a more general character.
FEED