This post summarises the latest work looking at the integration of the Dutch and German gas markets. In the period 2007–2013, the absolute value of the differences in price levels between the Dutch and the German NCG market decreased, indicating more integration. There is evidence that the improved connections within the German networks as well as between the Dutch and the German network contributed to this.
The strengthening of the connections with the UK market and the Russian supply, however, had a negative effect on market integration between the Dutch and the German NCG market. The liquidity-enhancing changes within the Dutch market had a negligible effect.
Growth of Gas networks
Since the liberalisation of European gas markets in the 1990s, market places in various European countries have been developed, such as the National Balancing Point (NBP) in the United Kingdom, the Title Transfer Facility (TTF) in the Netherlands, and Net Connect Germany (NCG) in Germany. The liquidity of in particular NBP and TTF has grown significantly over the past years (Heather, 2012). For the creation of a European gas market, the national market places need to be connected, enabling traders to engage in international price arbitrage.
The available transport capacity, however, frequently formed a constraint for international trade and access to the transport infrastructure was limited as long-term access rights were granted to the existing firms on the basis of non-market mechanisms.
Analysis of the integration of regional gas markets is a much studied topic with the latest research by Kuper and Mulder building upon the work of Siliverstovs, L’Hégaret, Neumann and von Hirschhausen (2005), Cuddington and Wang (2006), Marmer et al. (2007) and Growitsch et al. (2013).
Kuper and Mulder focused on the Dutch market, as here a large domestic supply and demand coincide with a high degree of connection with its neighbouring countries (Germany, Belgium and the United Kingdom), while a number of institutional changes have also occurred in the recent past. In the period 2007– 2013, three major changes in the Dutch gas market affected the liquidity of the TTF (Heather, 2012); in 2009, the obligation of market parties to book quality-conversion capacity was abolished, in April 2011, two other changes were implemented: the introduction of a market-based balancing regime, and the new policy of the Dutch incumbent gas trader GasTerra to supply all gas for the domestic market on the TTF instead of factory gates or city gates. Moreover, the Dutch market became stronger connected to the UK market in 2010.
Infrastructure constraints and gas prices
The degree of gas market integration is often measured by how price differences evolve over time. This is based on the idea that in a fully integrated market, price differences quickly disappear as a result of traders using arbitrage opportunities (Stigler and Sherwin, 1985). In such a market, price differences between countries do not exceed the actual costs of transportation. If, however, constraints between regional markets do exist, prices in these markets are not directly related anymore and, as a result, prices may diverge for a period of time.
Various papers analysed the degree of integration of gas markets on the basis of price differences between countries or hubs. Several authors have found evidence for economic integration of markets. Siliverstovs et al. (2005) find, on the basis of a co-integration analysis on data from the early 1990s to 2004, that the European and Japanese gas markets were integrated in the long term, because of the presence of similar long-term contract structures and oil-price indexation. Although co-integrated, short-term price differences did exist as a result of fluctuations in transportation costs as well as the use of different types of reference oils applied in the oil-price indexation contracts.
Regarding the relationship between the European markets and the US gas markets, these markets were not integrated as arbitrage was hardly possible between these regions, while there were no common drivers behind the gas prices. In the US, gas prices were already to a large extent determined in competitive gas markets, while in Europe gas prices were linked to the oil price. Marmer et al. (2007), however, argue that the US gas market consists of three relatively isolated regional markets: the Northeast, Midwest and California. Demand shocks in one of these regional markets appeared not to result in sufficient price adjustments in other regions. Cuddington and Wang (2006) also find different regional markets within the US.
Using a co-integration and a time-varying coefficient approach, Growitsch et al. (2013) find that the two major trading hubs in Germany (NCG and GPL) and the Dutch TTF market are reasonably well integrated. Nevertheless, price differences do occur which cannot be explained by transportation costs, i.e. the exit and entry charges imposed in the entry–exit system of the gas networks. They conclude that capacity constraints between the two German markets initially hindered the realisation of permanently equal prices, but this impact has been reduced as a result of the improved connection of the German networks. In addition, they conclude that the German NCG market and the Dutch TTF have also become increasingly integrated: prices between NCG and TTF appear to adjust within one trading day.
The Dutch gas market and its cross-border connections
A characteristic phenomenon of the Dutch market is the presence of a huge swing field (Groningen), and a number of small fields, both onshore and offshore. Because of the Groningen field, the Dutch gas industry is able to export gas with a high seasonal profile to the neighbouring countries. The Dutch gas network is connected with the networks in Germany, Belgium and the United Kingdom. The connections with German and Belgium are used both for import and export, while the connection with the United Kingdom is only used for export.
The net flows to Germany and Belgium, defined as Dutch import minus Dutch export, as well as the exports to the United Kingdom have a strong seasonal pattern which is related to the unique characteristic of the Groningen field to offer flexible supply as shown below.
The net flow to Belgium is always negative, meaning that the export to this country continuously exceeds imports. The net flow to Germany is mostly negative as well, but it is positive during some summer days when imports exceed exports.
The import of gas consists only of gas from the GasunieDeutschland (GUD) network, which is a part of GASPOOL. This gas, coming from Norway and Russia, is partly used by industrial consumers, including electricity companies, while the other part is re-exported. The latter implies that the Dutch network is also used as a transit network that brings gas from for instance Russia to the United Kingdom. These transit flows are less temperature related than the domestic demand by residential users. The data show that import flows are fairly flat during a year.
The liquidity of the TTF has grown substantially. The share of the TTF in the Dutch gas market increased from 5% in 2007 to about 40% in 2011 and remained on that level. In addition, the number of active players increased strongly. The churn ratio between the volumes of traded and the volumes actually delivered on the TTF increased to above 20. Note, that a churn ratio above 10 indicates that a market is a mature market (GTS, 2012;Heather, 2012). The churn of the NCG, however, is about 3 (Growitsch et al., 2013), implying that the trade in this market is relatively strongly related to physical delivery and that this market is much less liquid than TTF and NBP.
The day-ahead price on the NCG correlates strongly with the day-ahead price on the TTF (the correlation coefficient is 0.994).
Institutional changes in the Dutch gas market
A key institutional change in the Dutch market was the abolishment of the obligation of market parties to book quality-conversion capacity as of July 1, 2009. Moreover, the connection with the UK market was strengthened through the introduction of interruptible reverse (backhaul) flow services on the BBL (Balgzand–Bacton Line), making it possible to book gas in the reverse direction. Other measures to improve the liquidity of the Dutch wholesale market were the implementation of a market-based balancing regime on April 1, 2011 and the policy of the Dutch incumbent gas supplier GasTerra, to deliver all gas for the domestic market on TTF instead of factory or city gates (Heather, 2012).
Two institutional changes affected the connection of the Dutch and the German networks. In 2008, the operator of the Dutch network (GTS) acquired a significant part of the German market, now called GasunieDeutschland (GUD). And in 2013, new mechanisms were introduced to improve the allocation of cross-border capacity.
Within the German market also several institutional changes occurred. After the introduction of an entry–exit system in October 2007, several networks were pooled resulting in two network areas. In addition, in 2012 the second part of the connection between the German network and the Russian gas fields (Nord Stream) was realised.
Integration of the Dutch and German Gas markets
Comparing the daily gas prices between the Dutch market (TTF) and the German market (NCG), we find that these markets have become more integrated over the past years. At the end of 2013, the absolute level of the difference in price levels was 0.276 euro/MWh while the highest level was 0.860 (from October 2008 to June 2009). Over the last 4 years of the period of analysis, the absolute level of price difference was, however, fairly stable.
Controlling for a number of external factors, there is evidence that several institutional factors have contributed to the increased integration between the Dutch and the NCG market in Germany.
In particular, the integration of networks within Germany had a positive effect on this market integration. This finding is in line with the conclusion of Growitsch et al. (2013) who found that the efficiency of the German market increased strongly after the introduction of the entry–exit system and the creation of the NCG market zone. We also find that the improved connections between the Netherlands and Germany contributed to market integration. Both had on average a joint effect of about −1.7 euro on the absolute value of the price difference.
For the measures increasing the liquidity within the Dutch market, the effect on market integration appears to be negligible. Although these measures likely have contributed to the strong increase in liquidity of the TTF, they did not have a measurable joint effect on market integration. The increased connections to the UK market and the supply from Russia, however, reduced the integration between the Dutch and the NCG market.
From these findings, we learn that measures directly reducing barriers within networks are most effective in realising more integrated markets.
Examples of such measures are the merging of networks into larger networks and introduction of market-based mechanisms to allocate capacity on primary or secondary markets. Although connecting networks to other markets appear to increase price differences within these networks, such connections can be beneficial for gas consumers if they increase the availability of gas and, hence, the security of supply. A similar remark must be made for measures which are directly meant to increase the liquidity of markets. Although these measures appear to have a negligible effect on market integration, they may reduce costs for traders which is beneficial for gas consumers at the end of the day.
- Cuddington, J.T.,Wang, Z., 2006. Assessing the degree of spot market integration for U.S. natural gas: evidence from daily price data. J. Regul. Econ. 29 (2), 195–210.
- Kuper & Mulder – Cross-border constraints, institutional changes and integration of the Dutch–German gas market.
- Neumann, A., Siliverstovs, B., 2005. Convergence of European spot market prices for natural gas? A real-time analysis of market integration using the Kalman filter. Discussion Paper in Economics 05/05. Dresden University of Technology.
- Siliverstovs, B., L’Hégaret, G., Neumann, A., von Hirschhausen, C., 2005. International market integration for natural gas? A cointegration analysis of prices in Europe, North America and Japan. Energy Econ. 27 (4), 603–615