Discussion Papers

Marcus Stronzik, Margarethe Rammerstorfer, Anne Neumann

Competitiveness of the storage market

No. 305 / March 2008

Summary

Storage facilities for natural gas play an increasing role in the overall European gas market as they allow for intertemporal arbitrage in order to match the supply with seasonal demand patterns. Moreover, storage might also be used for short-term arbitrage. The aim of the study is to develop a monitoring tool to evaluate the performance of the European gas market. One cornerstone of the analysis is the intertemporal balancing function of storage facilities.

The study is based on the findings of the theory of storage which provides for noarbitrage conditions under effective competition and implications concerning the relation of market prices and storage behavior. Our analysis is based on the indirect test of Fama and French (1987). This test allows us to analyze the performance of the three major European trading points – the National Balancing Point (NBP,U.K.), the Zeebrugge hub (Belgium) and the Title Transfer Facility Hub (TTF, Netherlands) – in relation to storage with lacking inventory data. We assess the no-arbitrage condition using ordinary least squares (OLS) estimators. We introduce seasonal dummies to map storage levels controlling for seasonalities in the convenience yield and the risk-free interest rate to capture capital lockup. As dependent variable we use the difference of spot and futures prices – the so-called basis – for six and twelve months maturities of contracts.

Partly, the results are counter-intuitive and contradicting expectations. While the explanatory variables show significant influence on the basis in most of the considered cases, estimations led to interest rate coefficients far away from one. This deviation from the competitive benchmark indicates a fairly high arbitrage potential not being exploited by market participants hinting at market imperfections. Surprisingly, though NBP is much more developed than the other two hubs, no significant differences across markets could be observed.

Further research seems advisable in order to get a clearer picture on the interrelations between price developments and storage behavior, e.g. by incorporating direct information on storage levels as well as price data spanning longer time periods. At the moment, it is too early to apply the tool to the German market for natural gas directly due to the lack of sound price data. As soon as this becomes available, our approach will be ready to give some first insights into the performance of the German market.

[only a german version available]

Discussion Paper is available for download.

To top  |  Print