The deregulation of the natural gas market in the United States: Effects on efficient pipeline utilization
---GAS on the MOVE---
J.T. (Jurian) Bootsma Studentnumber: 1581252 M: +316 11 22 48 54 E: jurianbootsma@hotmail.com Rijksuniversiteit Groningen N.V. Nederlandse Gasunie Groningen, September 2008
Supervisor N.V. Nederlandse Gasunie
Drs. A.J. (Arjan) Wesseling RC Finance Economics department
Master-thesis
International Business & Management, specialization
International Financial Management.
Faculty of Economics & Business
Thesis supervisor
Prof. Dr. L.J.R. (Bert) Scholtens
Professor International Finance
Faculty of Economics & Business
Preface
“Een pad ontstaat door er op te lopen1.”
“Ik hoor en ik vergeet, ik zie en ik onthoud, ik doe en ik begrijp2.”
After a research process of seven months, I present my Master-thesis about the regulatory influences on efficient interstate natural gas pipeline utilization in the United States. Besides the influence of an increased price on the spot market for natural gas on efficient interstate natural gas pipeline utilization is investigated.
This research is conducted on behalf of the Financial Economics department of Dutch “Gasunie N.V.3” in Groningen and is exercised to conclude the Msc-study International Financial Management at the “Rijksuniversiteit Groningen” in the Netherlands and at “Uppsala Universitet” in Sweden. This thesis would not have been succeeded without the support of a number of people, whom I want to say: “Thank you very much”.
Firstly, I want to thank Professor L.J.R (Bert) Scholtens, for his support in name of “Rijksuniversiteit Groningen”. His support and suggestions were indispensable for succeeding this Master-thesis. Furthermore, I would like to thank Drs. Arjan Wesseling, my supervisor at “Gasunie N.V”, for his “second look”, support, suggestions, time, knowledge sharing and his very useful ideas for this research. Thanks to my colleagues of “Gasunie N.V.” from the FE-department, especially to Dr. Martien Visser for his support and ideas during my period of research.
Especially, I would like to thank my girlfriend Marloes. Thanks a lot for your love, support, patience and confidence in a successful result. Your optimism and knowledge has been very stimulating. Also thanks to my friends, especially Thijs, for the necessary relaxation by being my opponent in playing soccer matches on the computer after a day of research activities.
At last, but certainly not least, thanks a lot to my parents and sisters. Their support and confidence were indispensable during my research, but also during my previous years of study in Deventer, Uppsala and Groningen. Without their ever listening ears and unconditional support I never could have reached what I reached now.
Groningen, August 31, 2008
Jurian Teake Bootsma
1 Tsui, T. Chinese Wijsheden. English: A path can only be created, if you start walking on it,
Abstract
The regulation of gas transmission networks has undergone major changes since the early 1990s in both the United States and Europe. Whereas the European market integration is still a continuing discussion within the European Parliament, the United States complemented their long-standing regime of cost-plus regulation by increasing pipeline competition.
In this research, I studied what the impacts of three events -two important regulatory decisions and of one striking increase in prices- are on efficient pipeline utilization, measured by analyzing the annually changed contributions of six selected independent variables on the dependent variable: Efficient pipeline utilization. Results are presented as outcomes from seventeen multiple regression analyses, wherein independent variables’ contributions to the dependent variable are measured in the years before, during and after the event mentioned.
Contents
Preface 2
Abstract 3
Contents 4
List of figures and tables 5
Abbreviations 6
1. Introduction 7
2. Background 12
2.1. History of regulation gas transmission in the United States 13
2.2. History of regulation gas transmission in Europe 18
2.3. U.S. gas market structure 20
2.4. European gas market structure 22
3. Model and variable selection 25
3.1. Interstate natural gas transmission: Literature overview 27
3.2. Variable selection 29
4. Data and measurement 37
5. Results 45
5.1. Interpretation results 50
5.2. Hypotheses and actual multiple regression results 57
6. Discussion and conclusion 60
References 63
Websites 67
Appendices 68
Appendix A: Interstate natural gas transport pipelines 69
Appendix B: Multiple regression techniques 73
Appendix C: Multiple regression analyses results and coefficient tables 75
Appendix D: Contributions independent variables in graph and table 84
List of figures and tables
Figures Title Page
1. U.S. natural gas consumption, production & net imports 20
Table Title Page
1. Important milestones in natural gas regulation United States 18
2. Important milestones in natural gas regulation Europe 19
3. US-Europe comparison of industry structure 24
4. Summary of the literature 28-29
5. Dependent and independent variable selection 32
6. Hypotheses: Contributions of independent variables 35
7. Events 38
8. Variable description and measurement notation 39
9. Descriptive statistics 1990-2006 40-44
10. Multiple Regression results; model summaries & actual contributions 47
11. Conclusions multiple regression analysis: Model significances 49-50
12. Hypothesized and actual independent variable predictions 58
13. Appendix C: Multiple regression analysis results 1990-2006 75-83
14. Appendix D: Annual beta-values independent variables 84
15. Appendix E: Average annual development independent variables 85
Graph Title Page
1. Operational development average U.S. interstate pipeline company 56
2. Efficient pipeline utilization and financial development pipelines 56
Abbreviations
Abbreviation DescriptionAVDAYn Average daily flow, year n
Bcf/day Billion cubic feet a day (European notation)
Beta Coefficient of independent variable ni prediction of dependent variable CAPn Capacity Pipeline, year n
Co. Company
DEA Data Envelopment Analysis
df Degrees of freedom
DSO Distribution System Operator
e Residuals
EBITDA-margin
Earnings Before Interest, Taxes, Depreciation and Amortization divided by Revenues
EBITDAmn EBITDA-margin, year n
EIA Energy Information Association
Enter-method Multiple regression method used; inclusion of all independent variables FEA Federal Energy Association
FERC Federal Energy Regulatory Commission (United States)
FPC Federal Power Commission
F-test Test for significance
G Event selected
IEA International Energy Association
Inc. Incorporated
InvRatn Investment-ratio, year n
k Number of terms in equation not counting the constant
LDC Local Distribution Company
LENGTHn Length of pipelines, year n
MMcf Million cubic feet a day (U.S. Notation)
MSE Mean Squared Error
N Number of interstate pipelines selected
N.V. Naamloze vennootschap (Dutch)
n Year(s) of research
NGA Natural Gas Act
NGPA Natural Gas Policy Act
NGSA Natural Gas Supply Association
p Confidence interval (level of significance)
PL Pipeline
R² Variance of dependent variable predicted by independent variable
ROCE Return On Capital Employed
ROCEn ROCE, year n
Sig(p) Significance interval SPSS Statistical software St.dev Standard Deviation
Std.Error Standard error in multiple regression analysis t Number of independent variables selected Tcf Trillion cubic feet
TSO Transmission System Operator
t-test Test for significance, derived from F-value Uy Average efficient pipeline utilization Y Efficient pipeline utilization
Ŷ Observed values in multiple regression model Ŷit Values fit by the multiple regression model
б Variance
б² Variance
1. Introduction
“Engineering is the professional and systematic application of science to the efficient utilization
of natural resources to produce wealth.” -T.J. Hoover and J.C.L. Fish (1941)4-
“Gas on the Move” is the carefully chosen title of this Master-thesis. The title consists of four easy words, while the meaning and importance of these words are endless. This slogan is applicable because it perfectly characterizes the spectrum “Gasunie N.V.” and many other natural gas transmission companies are operating in on a worldwide level.
The title covers both literally and figurative the area of research in this thesis. Literally because “Gasunie N.V.” is a gas infrastructure company, which owns and manages one of the largest high pressure gas pipeline grids within Europe, consisting of 12,000 kilometers of pipeline, dozens of installations and approximately 1,100 gas receiving stations5. Figurative, the slogan covers the area of research, because worldwide energy markets were, are or become in transition. Worldwide natural gas reserves are decreasing, while demand for natural gas increases. Companies that want to play a significant role within the natural gas transmission industry in the future need inventive, innovative, creative and focused strategic action to keep their natural gas transmission activities “on the move”.
Gas transmission company-networks are regulated rather differently in Europe and in the United States6. Whereas the European regulation focuses on a gradual liberalization of the industry and the creation of a competitive integrated internal market, with security of supply ensured (SEC, 2006), the US regulation shifts its focus from cost to value by complementing cost-of-service regulation with institutions fostering competition and market integration (O’Neill, 2005). An important similarity between the US and European regulation is that they are both based on the notion of natural monopoly (Jamasb et al., 2008).
Unlike gas distribution or electricity networks, gas transmission networks are not necessarily natural monopolies in Europe and the United States (Jamasb et al., 2008). Mattias Hamberg (2001) defines a natural monopoly as a situation where one company has total control over a product, service or resource that has no close substitutes. According to Jamasb et al. (2008), natural monopoly relates to efficient service provision for a given market and not to economies of scale at the level of a single pipeline.
Arano and Blair (2006) argue in their article that prior to the mid 1970s the structure of the interstate natural gas transmission industry could be characterized as a natural monopoly in
the United States. The influence of technology requirements in the transmission sector from the mid 1970s on is researched by Sickles and Streitwieser (1998) and MacAvoy (2000). They conclude that these requirements promoted the development of monopolistic buyers and sellers of natural gas. To protect consumers against the effects of this monopolistic situation, FERC7 introduced Orders to control rates and set regulations to abuse market power (Arano & Blair, 2006).
Kahn (1971) argues that natural gas markets easily can be served by several pipelines, although the existence of economies of scale in relation to pipeline diameter. In 2002, Kahn summarized deregulation studies in a comprehensive survey of the existing theoretical literature about deregulation of industries in the United States from 1970 to 2000 (Arano & Blair, 2006). Kahn (2002) concludes that competition is superior to comprehensive regulation as demonstrated by the experiences of other industries8. This claim is supported by O’Neill (2005), stressing that the non-traditional approach to regulation might provide superior results and that natural gas pipelines are oligopolies rather than monopolies.
Whereas Kahn, Arano & Blair and O’Neill focus on the structural changes as an effect of (de)regulation9 in the United States, I focus on changes in efficient pipeline utilization as a result of (de)regulation in the United States. The benchmarking-based model as introduced in the articles of Jamasb et al. (2007; 2008) is according to them a viable short to mid-term solution for the regulation of gas transmission in Europe10. Furthermore, Jamasb et al. (2008) argue that “the performances of the US natural gas companies under the new regulatory framework are high, which indicates that in the long run a focus on market integration might provide additional company benefits11.” Recent discussions with Dutch market specialists of “Gasunie N.V.” suggest that the European (de)regulation can lead to these similar company-benefits as occurred in the United States.
Jamasb et al. (2008) refer in their article to Makholms’ (2007) detailed description of the institutional framework FERC puts in place to complement its traditional rate-of-return regulation with competition between pipelines. This reference is very important for this research, because Makholm (2007) argues that both Europe and the United States could achieve security of supply as a result of increased competition and market integration.
7 FERC is an abbreviation of the US Federal Energy Regulatory Commission, 8 Airlines, telecommunications and electric utility industries (Kahn, 2002),
9 The terms “regulation and “deregulation” are used one through another; both terms present the
regulatory process,
10 In their report for the Council of European Regulators, they benchmarked several gas transmission
operators against a sample of US interstate transmission companies. Unfortunately, they were not able to include any data for European operators in this work. The often stunning differences in transparency between the US and Europe are discussed by Makholm (2007). The US approach on transparency is discussed for instance by Olsen (2005). Footnote literally derived from Jamasb et al. (2008),
11 These “company benefits” are: larger markets to serve; opportunities for acquisitions, collaborations,
In this research, I insert interstate efficient pipeline utilization as the dependent (criterion) variable and three operational interstate pipeline indicators (length of pipeline, maximum daily capacity and average daily flows through the pipeline) and three financial performance indicators (EBITDA-margin, ROCE and investment-ratio’s) as independent (predictor) variables. Efficient pipeline utilization can be defined as “operating a natural gas pipeline system as close to full capacity as possible to maximize revenues” (EIA, 200812). A report by the Energy Information Association (2008) mentions that the average efficient utilization rate of a natural gas pipeline system seldom reaches 100 percent. They state that these outages are mainly the result of three important factors13:
- Temporary decreases in market demand - Scheduled or unscheduled maintenance - Weather-related limitations to operations
Despite of these three factors, efficient pipeline utilization is an important target for natural gas transmission companies on longer term. On short term, pipeline utilization can be affected by one of the three factors mentioned, however over a longer term (n>3 years), substantial changed developments in pipeline utilization can be observed. Sickles and Streitwieser (1991) state that “increased competition between pipelines and the efficiency incentives inherent in the lag between rate cases would allow for maximizing behavior”. After discussions with specialists from Dutch “Gasunie N.V.”, I decided to use interstate efficient pipeline utilization as the criterion variable to represent company efficiency.
According to EIA (2008), there are several ways to estimate efficient pipeline utilization:
• As a measure of the average-day natural gas throughput relative to estimates of system
capacity at State and regional boundaries
• The system-wide pipeline flow rate, which highlights variations in system usage relative
to an estimated system peak throughput level
• A system peak-day usage rate, which generally reflects peak system deliveries
I have chosen to measure efficient pipeline utilization by analyzing the pipeline flow rates on a system-wide level, highlighting variations in system usage relative to the maximum throughput level capacity. The aim of research is to investigate if and how six independent variables predict the dependent variable before, during and after regulations and market circumstances. The dataset for three operational variables is derived from the Energy Information Association (EIA), specifically from the EIA-176 database. Financial data are
derived from the OSIRIS-database14, provided by “Bureau van Dijk”. A multiple regression test will be used to test what the changes in contributions of the independent variables on the dependent variable are15 before, during and after the event chosen.
“Market structure” can be defined in this case as “the portioning and positioning of inter- and/or intra-state/region/country specific companies in and around gas transmission”. The changes in structure on European and North American gas transmission pipeline markets are investigated in literature by Freytag, Schiereck & Thomas (2005). They analyzed the impact of takeovers on market power by studying 70 takeovers of US-American and 69 takeovers of German energy utilities by applying event study methodology. They conclude that companies’ potential to increase market power is an important motive for takeovers within the German market; however they could not find any significance within the United States market. As a result of takeovers, the potential for market power still exists in the United States, despite of the intention of the federal government to develop a more competitive market, abusing market power (Gabriel et al., 2005).
Murry and Zhu (2006) investigated price behaviors among various hubs in the United States in their article about market performance. They found that the presence of market power was a very plausible explanation for price behavior in some hubs vice verse. Therefore, the influences of prices in a competitive market with hubs are not taken into account in this research. There is also literature on the effect of regulatory change on productivity efficiency of US transmission companies. Examples are Sickles and Streitwieser (1991; 1998). They make clear that productivity, -measured by analyzing a set of 24 US interstate pipeline companies from 1978 until 1985- decreased in the first eight years after the deregulation of well-head prices in the United States in 1978.
In this thesis, two sets of questions are asked. First, does average interstate pipeline company efficiency change after regulatory policy? And second, how do company performance indicators contribute to efficient pipeline utilization before, during and after the year of important regulatory implementations?
Efficiency (Jamasb et al., 2008) and productivity (Sickles & Streitwieser, 1991) fell after the deregulation of well-head prices in 1978. Furthermore, takeovers, mergers and acquisitions led to fewer companies within the industry.
14 OSIRIS is s a database containing financial information on globally listed public companies, including
banks and insurance firms from over 190 countries,
15 Unfortunately, I am not able to include any data for European interstate transmission companies in this
Already in 1987, about one third of city gate markets received services from multiple pipelines (Kalt and Schuller, 1987). In 2001, 14 large interstate conglomerates owned about 85 percent of the interstate natural gas pipelines in the United States. (Johnson et al.,1999).
Given that the sample in this research (1990-2006) starts three years before the latest regulatory push for more competition in 1993, when FERC implemented Order 636, I expect to observe first increased efficient pipeline utilization because of increased competition and an increased attention towards cost-reduction and efficiency. From the mid-1990s on, I expect to observe a higher (efficient) pipeline utilization as in previous years (1990-1992), because of the implementation of FERC Order 636 in 1993, which focus on more competition and higher efficiency. A control group can not be included; however I presume that regulations as introduced in chapter two provides only weak incentives for performance improvements. Therefore, the results of this research can give an indication of how the development of company performance indicators (operational and financial) are affected by regulations (1993; 2000) and rise in spot-prices (2005) during the sample period (1990-2006) in the United States.
I will make absolutely clear that all the results and conclusions apply to interstate gas transmission only. Both in Europe and the United States several industry networks are regulated in different ways and with varying levels of success.
2. Background
The main objective of European energy policy in the area of natural gas has been the gradual liberalization of the sector and the creation of a competitive integrated internal market, with security of supply ensured (SEC, 2006). Despite of this European objective, the European Commission has to deal with many barriers towards a competitive single European gas market, which are also being impediments to liberalization. A very important condition to realize a competitive integrated European market for natural gas is ownership unbundling. Ownership unbundling means that the transport network grid (operational) activities of natural gas transmission companies separate from the trading activities into two independent operating companies.
A first step towards unbundling dates from 1998, when the European Union stated in their first Directive that administrative unbundling of activities was possible. In 2003, the second EU Directive stated legal unbundling of TSO’s16, however it was a voluntary agreement. In 2008, the European model of liberalization still leads to a lot of discussion. The European regulatory approach aims at achieving competition and security of supply under the conditions of non discriminatory access to infrastructure, transparency of market information, integration of formerly national markets and full market opening since 2007. The instruments provided in the legislative framework to reach these goals are unbundling of TSOs and DSO17s, non-discriminatory regulated access and a strong role for regulatory authorities (European Commission, 2007). Despite of this regulatory approach, there exist many uncertainties about the development of an integrated European market for natural gas transmission. Transmission companies are not sure about their current18 position in the European natural gas transmission market. Besides, they are wondering what position they have to obtain towards the future. Furthermore, it is not clear how the market structure will change as a result of the transition- and liberalization-process.
The European Commission stated in its Energy Sector Inquiry that the US gas market is much more developed than its European counterpart (Jamasb et al., 2008). The US gas market already developed from a vertically integrated, geographically fragmented, and heavily regulated industry towards an integrated and lightly regulate industry (Jamasb et al. 2008), however the process has been one of a long tale of trial and error (Makholm, 2007).
2.1. History regulation gas transmission in the United States
Together with the natural gas transmission industry, the complexity of maintaining regulation developed fast in the early 1900s, when natural gas was shipped between regions in the United States. The first intrastate pipelines began carrying gas from city to city, resulting in the first US interstate pipeline in 1891. This pipeline was 120 miles in length and carried natural gas from wells in central Indiana to the city of Chicago.19 Already in 1872, the first long distance pipeline was finished, however it did not cross a state-border.20
Due to the state-crossing pipelines, a new discussion existed between individual states of how to regulate these pipelines. In response to this new situation, state level governments intervened by regulating the new interstate natural gas market, and determining rates for transmission that could be charged by gas distributors. Between 1911 and 1928, several states attempted to assert regulatory oversight of the increasing number of these interstate pipelines. However, an interstate commerce clause of the U.S. Constitution21, named “Supreme Court Commerce Clause” stated that interstate transmission companies were beyond the regulatory power of the state-level government, resulting in another new situation.
In 1935, over a quarter of the interstate natural gas pipeline network was owned by only 11 holding companies.22 After a report of the Federal Trade Commission in 1935, the US Congress passed the Public Utility Holding Company Act to limit the ability of holding companies to gain undue influence over a public utility market (Gordon, 1992).
In 1942, the NGA23 specified that no new interstate pipeline could be built to deliver natural gas into a market already served by another pipeline. The high concentration of the natural gas industry and the monopolistic tendencies of interstate pipelines were the main rationales for this passage of the NGA.
In the 1940s and early 1950s, wellhead24 prices were unregulated, however the Federal Power Commission contended that if the natural gas producer and pipeline were unaffiliated, natural market forces existed that would keep wellhead prices competitive. This changed in 1954 when the Supreme Court stated that natural gas producers who sold natural gas into
19http://www.naturalgas.org/overview/history.asp,
20 Information derived from website “the city of Mesa”: www.cityofmesa.net,
21 The United States Constitution is the supreme law of the USA, adopted on September 17, 1787 by the
Consitutional Convention in Philadelphia, Pennsylvania. (Source: US government report),
22 These companies also controlled a significant portion of gas production and distribution. Sources:
www.naturalgas.org; www.ferc.gov,
23 The Natural Gas Act (NGA), developed in 1938 by the federal government gave the Federal Power
Commission (FPC (Created in 1920 with the passage of the Federal Water Power Act)) jurisdiction over regulation if interstate natural gas sales,
interstate pipelines fell under the classification of ‘natural gas companies’ in the NGA, and thus were subject to regulatory oversight and control by the FPC.25
Between 1954 and 1960, the FPC attempted to deal with producers and their rates on an individual basis, however due to unfeasible administration at the FPC, they decided in 1950 to set rates based on five separate geographical areas. The problem with applying rates for a particular area based on cost-of-service methodologies exceeded that there existed many wells in each area, with vastly different production costs. In 1974, The FPC adopted national price ceilings for the sale of natural gas into interstate pipelines, which was in place until the passage of the Natural Gas Policy Act (NGPA) in 1978. The mistake FPC made was that they only regulated producer wellhead prices for natural market gas destined for the interstate market, while the intrastate market was relatively free of regulation. Because of the relatively low prices, producers got for their natural gas on the interstate markets, it became much more interesting to sell their gas against a higher price to intrastate bidders26.
As a result of these developments, many factories in the Midwest were forced to close, due to a shortage of natural gas to run their facilities in 1976 and 1977. Meanwhile, in the producing states, virtually no shortage was felt, due to the thriving intrastate market satisfying natural gas demand in these states.
The US Congress reacted by enacting the Natural Gas Policy Act in 1978, which served three main goals:
• Creating a single national natural gas market. • Equalizing supply with demand
• Allowing market forces to establish the wellhead price of natural gas.
In 1977, the FEA was replaced by the Federal Energy Regulatory Commission (FERC), under the Department of Energy Organization Act.
The NGPA was a fundamental and important first step in deconstructing the regulatory problems created by the NGA. The market response to the provisions of the NGPA included: 27
• Pipelines signed up for many long-term natural gas contracts, due to gas shortages in the past years.
• Producers expanded exploration and production, drilling new wells and using the long-term sales contracts with pipelines to recover their investment.
• Average wellhead prices rose dramatically in the years following the NGPA.
25 Information derived from ‘Natural Gas Supply Association’. www.ngsa.org, 26 See footnote 19,
27 Source: The Natural Gas Policy Act of 1978, US government website;
• Prices for end-users increased, but were mitigated by the pipelines, which blended the cost of gas under new contracts with regulated gas under old contracts when selling their bundled product to their customers.
• Price increases led to decreased demand, due to these regulatory issues.
A reaction of pipelines-owners on the regulations of the NGPA was to sign up for long-term ‘take or pay’ contracts. These contracts required the pipelines to pay for a certain amount of the contracted gas, whether or not they can take the full contracted amount.
In the 60s and 70s, it was necessary to curtail natural gas deliveries due to high demand and low supply, however this situation reversed in the period from 1980-1985, when rising natural gas prices resulted in the dropping off of some of the demand that had been built up when the price for natural gas was held below its market value28.
The natural gas market in the United States has seen significant changes in the past twenty years (1985-200529) due to deregulation and restructuring (Zhuang & Kiet, 2005).
The first steps towards deregulation of the natural gas market were taken by the Natural Gas Policy Act, resulting in the concluding FERC Order No. 436.30 In 1985, FERC issued Order No. 436, which changed how interstate pipelines were regulated. The unbundling of pipelines31 product must give customers the ability to purchase their own natural gas and to transport it via pipelines. A voluntary framework was established under which interstate pipelines could act solely as transporters of natural gas, rather than filling the role of a natural gas merchant. The effects of the order were both immediate as well as on longer term visible, with offering transmission service to all customers and the immediate unbundling of the activities as most important results.
While FERC Order No. 436 and 50032 made the unbundling of pipeline services possible, the establishment of transportation services by only one pipeline continued to be voluntary. FERC Order No. 636 completed the final steps towards unbundling by making pipeline unbundling a requirement. Because of this, Order 636 is often referred to as the “Final Restructuring Rule”, there it was seen as the culmination of all of the unbundling and deregulation that had taken place in the past 20 years.33
28 See footnote 19,
29 Zhuang & Kiet (2005) focus on the 1985-2005 period,
30 Order also known as “Open Access Order”, because of the movement towards allowing pipeline
customers the choice in the purchase of their natural gas and their transportation arrangements,
31 “Under the NGA and the NGPA, pipelines purchased natural gas from producers, transported it to its
customers (mostly LDCs), and sold the bundled product for a regulated price. Instead of being able to purchase the natural gas as one product, and the transportation as a separate service, pipeline customers were offered no option to purchase the natural gas and arrange for its transportation separately.” (www.naturalgas.org),
Issued in 1992, the Order states that pipelines must separate their transportation and sales services, so that all pipeline customers have a choice in selecting their gas sales, transport and storage services from any provider, in any quantity.
The results of this Order were that pipelines could no longer engage in merchant gas sales, or sell any product as a bundled service. The Order allows the complete unbundling of transportation, storage, and marketing. The customer now chooses the most efficient method of obtaining its gas. Furthermore, it required that interstate pipelines offer services that allow for the efficient and reliable delivery of natural gas to end users, such as the institution of ‘no-notice’ transportation service, access to storage facilities, increased flexibility in receipt and delivery points, and “capacity release34” programs.
At last, Order 636 requires interstate pipelines to set up electronic bulletin boards, accessible by all customers on an equal basis, which show the available and released capacity on any particular pipeline. A customer requiring pipeline transportation can refer to these bulletin boards, and find out if there is any available capacity on the pipeline, or if there is any released capacity available for purchase or lease from one who has already purchased capacity but does not need it35.
The 2008 regulation of gas transmission pipelines by the Federal Energy Regulatory Commission (FERC) has designated that interstate pipelines can serve only as transporters of natural gas. In the past, as becomes clear from the previous paragraph, interstate pipelines acted as both a transport-company of natural gas, as well as a seller of the commodity, both of which were rolled up into a bundled product and sold for one price. Now, the transport companies can only sell the transportation component, and never take ownership of the natural gas themselves. The regulation of the natural gas industry is the responsibility of the FERC, which has the jurisdiction over the regulation of interstate pipelines and is concerned with overseeing the implementation and operation of the natural gas transportation infrastructure. An overview of the most important and influential Orders as issued by FERC can be found in table 1.
FERC stated that: “Although FERC is a government agency, FERC is designed to be independent from any political influence or (company) affiliation, as well as it is independent from any influence from the executive or legislative branches of government, and industry participants, including the energy companies over which it has oversight. The decision-process of FERC operates by majority rule, whereby the five commissioners vote when decisions
regarding Orders have to be taken. In charge with regulations, FERC must ensure that companies do not abuse their natural monopoly positions.”
In the natural gas industry, FERC regulates the rates and services offered by interstate pipeline companies, as well as certifying and permitting new pipeline construction and some closely related environmental issues.
There are three important dilemmas that companies have to keep in mind before building new pipelines or expanding existing ones:
• How a new pipeline serves public interest. • Degree of economic feasibility.
• How to deal with possible environmental issues.
There are essentially two types of issues faced by FERC: making company specific decisions, and making industry-wide decisions. Company specific issues can include applications for rate changes for one company’s transmission services, applications for changes in terms or conditions of transportation contracts, and complaints filed by another industry member, including utilities, project sponsors, trade associations, or any other interested party.36
The second types of issues faced by FERC are the industry wide issues and decisions. These may be much more complicated than company specific issues. Because issues and regulations that affect the entire industry are being contemplated, the number of interested parties can be very high, and countless opposing viewpoints may exist. It is the task of FERC to consider all different points of view, and issue a decision based on what they believe is the best course of action for the industry in general37.
36 Definition company specific decisions by FERC, derived from website: www.ferc.gov &
www.naturalgas.org,
37 Definition industry specific decisions by FERC, derived from website: www.ferc.gov &
Table 1: Important milestones in natural gas regulation United States38 Year of
issuance
FERC Order
Title Aim & Effects
1978 Natural
Gas Policy Act
Consumer protection against natural gas shortages as an effect of monopoly prices
Creating a single national market for natural gas; balance supply and demand; market forces allowed establishing wellhead prices of natural gas.
1984 Order
380
Elimination of variable costs from certain natural gas pipeline minimum bill provisions. (start of process of natural gas restructuring39
Enforceability of contract provisions on minimum payments between local gas distributors and the pipeline that served them.
1985 Order
436
The open access rule A pipeline could self-activate discounted service for any customer.
1987 Order
500
Take-or-Pay cost recovery Mechanism for pipeline companies to shift more of their liability under long-term contracts to producers,
consumers and downstream
pipelines.
1993 Order
636
The final restructuring rule Pipelines are required to separate their transportation and sales services, to give consumers the choice in selecting their transportation, storage and sale services from any provider.
2000 Order
637
Update final restructuring rule
Implementation of daily auctions for pipeline capacity. Increase of market transparency and efficiency.
2.2. History regulation gas transmission in Europe
The EU gas market is officially liberalized since July 1st 2005; however the implementation of the liberalization process is marked by many setbacks, often as a result of different attitudes of countries towards liberalization and unbundling. Unlike Europe which has national markets with national regulators for natural gas, the United States had a common market with a single federal regulator for all interstate commerce. O’Neill (2005) categorizes non-traditional tariff
models in the United States into flexible short-term rates (efficient allocation of capacity (Jamasb et al., 2008)) and long-term negotiated (fixed) rates (to meet security of supply).
Jamasb et al. (2007) summarize the difference between EU and US regulation as follows:
“Whereas European regulators aim at incentive regulation for monopolies, US FERC aims at complementing traditional rate-of-return regulation with competition through encouraging (or mandating) the development of the necessary market institutions”.
Table 2 highlights the important milestones in European gas regulation.
Table 2: Important milestones in natural gas regulation Europe40 Year of
issuance
Order / Directive
Title Aim & Effects
1998 1st Gas
directive
Regulations for an internal natural gas market
Guarantee free transport of natural gas, to secure the supply of natural gas and to improve competitiveness of the industry
2003 2nd Gas directive Second step towards an internal natural gas market
Legal unbundling of transmission system operators (TSO’s), voluntary.
2008 Twists in natural gas regulation Required unbundling or independent transmission operators
EU commission prefers full ownership unbundling as in the Netherlands since July 1st 2005. France and Germany launce the idea of initiating independent transmission operators. Still no decisions taken about which option is preferable.
Despite of the differences in regulation between the United States and Europe, there are many parallels with current efforts in Europe to unbundle (Jamasb et al., 2008). Whereas the regulation in the United States aimed at competition through a combination of controlling abuse of market power, unbundling and flexible short-rate setting, the European regulators moved towards more incentive regulation of pipelines within a value-chain. Furthermore, the US market is both physically (structure) as well as economic more embedded in a larger structure. The following paragraph discusses the structures of both the European and the US gas market, beginning with the most experienced one: The United States.
2.3. U.S. gas market structure
The U.S. natural gas production and consumption were nearly in balance until 1986. After this year, consumption began to outpace production, and imports of natural gas rose to meet U.S. requirements. In 2007, production stood at 19.3 trillion cubic feet (Tcf), net imports at 3.8 Tcf, and consumption at 23.1 Tcf, as can be shown in figure 141.
Figure 1: U.S. natural gas consumption, production and net imports
As figure 1 presents, the expansion in consumption from 1985 might be a first indication of the success of regulations since 1985 in the United States. Where consumption outpaces production, shortages exist. Therefore, Canada and Mexico are the most important foreign sources of natural gas in the United States. The United States are a net importer of natural gas from these areas of production.
In Texas (57,160 miles), Oklahoma (18,494) and Louisiana (18,358) in the southwest region42 and Kansas (15,284) in the central region, most natural gas pipelines were estimated in 2006. The total US Pipeline Mileage is 300,281 miles; 213,409 miles are interstate pipelines and 86,882 are non-interstate43 ones44. The U.S. natural gas pipeline grid includes more than 210 mainline natural gas pipeline systems45, from which 118 are interstate pipelines. The combined natural gas pipeline (both intrastate and interstate) capacity on mainline intrastate systems is only about 22 percent as much as interstate natural gas pipeline capacity, or about
41 Energy Information Administration (EIA). 2008. Annual Energy Review 2007, Energy Prospectives, 42 In the Gulf of Mexico some large-scale gathering systems are FERC jurisdictional and are therefore
counted as interstate,
43 Includes intrastate transmission and non-FERC jurisdictional large diameter gathering systems or
headers. Local distribution company (LDC) mileage excluded,
44 Data derived from website Energy information administration: EIA-176 database,
45 That is, they are engaged in the transportation of natural gas from production area to market area.
33 Bcf46/d.47 Interstate natural gas pipeline systems account for about 148 Bcf/d of total U.S. natural gas transportation capacity and as mentioned before, together they are 213,000 miles in length. The thirty largest interstate natural gas transport companies account for more than 78 percent of the capacity of 148 Bcf/d48 in 2001.
Until the late 70s, the structure of the natural gas industry in the United States was rather simple (Bongers et al., 2003). Natural gas producers explored and produced natural gas, which was sold to natural gas pipeline companies. The federal government regulated the price for which producers could sell their natural gas to interstate pipelines. Also the prices from pipelines to utilities to customers were regulated by the federal government.
Corporate ownership of interstate natural gas pipeline companies has changed substantially since the Federal Energy Regulatory Commission’s (FERC) issuance of Order 636 in 1992. In a report of the Energy Information Administration (EIA, 2001), it is concluded that corporations with strong ties to the electric power industry have gained significant ownership of natural gas pipelines while corporations heavily engaged in natural gas exploration and production have made some of the largest divestitures of pipeline assets. In 2001, fourteen corporations49 accounted for over 85 percent of interstate natural gas pipelines and associated facilities (Tobin, 1999). According to the core of this research, these fourteen corporations will be the source of a substantial portion of the future capital expenditures for interstate natural gas pipelines and associated facilities.
In the Central- and Southwest region in the United States most natural gas producing companies are located, because of the presence of natural gas in these regions. Consequently, these two regions are also a starting point for many of the major interstate pipeline companies that transport natural gas to the major markets located in the Northeast, Southeast, Midwest and Western regions. To overcome the effects of geographic scope when gas is transported (Leitzinger & Collette, 2002) in a country-wide natural gas market, new types of country-wide oriented companies developed by mergers, alliances and acquisitions. According to Park et al. (2008), these new corporations can flatter industry performances geographically by applying economies of scale and scope.
46 Billion cubic feet,
47 Energy Information Administration, Natural Gas Pipeline Affiliations Database, 2006; EIA-176 Database, 4848 Energy Information Administration, Natural Gas Pipeline Affiliations Database, 2006; EIA-176
Database,
49 Coastal Corporation; Columbia Energy Group; Consolidated Natural Gas; Duke Energy Corporation; El
The natural gas market in the United States consists of over 8,000 independent producers, from which 24 are major producers (Huijsen, 2006). There are 140 natural gas pipelines, from which 118 are selected as officially registered interstate50 natural gas pipelines. The total length of the natural gas pipelines is 285,000 miles (interstate pipelines 221,075 miles). There are 260 unregulated natural gas marketers and 1,400 local gas utility participants. The total length of utility pipelines in the United States is 833,000 miles, which are regulated by the state utility commissions. Interstate pipeline companies, providing transmission services, are privately owned and regulated. Local distribution companies (local gas utilities or LDCs) are usually privately owned (Huijsen, 2006). Appendix A presents the 118 interstate pipelines, pipeline lengths and the ownership structures. Several interstate transmission companies are integrated with electricity companies which form together large energy conglomerates.
2.4. European gas market structure
The continental European natural gas market exists of a number of countries. The European Union has the responsibility in ensuring the supply of energy to all consumers at reliable prices with respect for the environment and striving for healthy competition on the European energy market (Bongers et al., 2003).
The common European gas market differs from the US gas market. Until the First European Directive51 98/30 (1998), every single country has its own policy around natural gas. Besides, there were and are differences between European countries with respect to access to gas resources. The Netherlands, Norway, United Kingdom, Russia and Algeria were or still are the most important producers of natural gas, which gave them a powerful competitive position towards other EU-member-states. The three targets to liberalize the European gas market in 199852 were:
• Freedom of choice by customers (where to obtain natural gas from) • Increase of efficiency on industry level
• Improved services and innovations
50The absolute focus of this investigation lies on the interstate natural gas pipeline companies. The
reason for this is that European politics try to transform several national country-oriented policies around liberalization into a fully integrated European market for natural gas (transportation). This integration already took place in the United States, where single states keep their own intrastate law- and decisionmaking, however the interstate natural gas transportations are regulated by the powerful federal body FERC,
50. This development the European policy-makers want to occur also in the European Union. The influence
of state (United States) and national (Europe) fall outside the scope of this study, because these pipelines do not cross state- or country borders. Besides, the rapid emergence of Liquefied Natural Gas (LNG) is not included in the investigation, since there are too less historical data available,
51 European approach based on Directives. This means mandatory with respect to objectives, but free with
respect to the means to achieve the objectives,
52 Snijder R., 2007. Liberalisering Europese Gasmarkt. Presentation “Regulatory Affairs”, Energy Delta
The European gas market structure for midstream companies in the 1980s is characterized as an industry with “monopolistic” national companies with wholesale activities and gas transmission bundled. In 2007, market opening in phrases started with freedom of choice for every single customer.
Despite of the good intentions of the European Union to create a liberalized European market for natural gas, many difficulties exist. Many differences in national rules and systems exist between the 27 EU member countries53. These differences lead to lack of market integration, dominant national companies and difficulties in cross-border investment, because of protectionist measurements by national governments. Furthermore, they suggest that unbundling of networks is not effective, which is also an effect of different country policies. Unequal access to information, illiquid markets and non-transparent price formation -and network access- are signs that markets are not transparent. At last, the still regulated prices distort national markets and a common European market for natural gas and access to gas storage is insufficient (Stiphout v., M., 2007). In their presentation, they conclude that further measures are needed in order to achieve an internal European gas market54.
Table 3 provides a comparison between the US and EU natural gas transmission industries55. From this table, the following can be observed: The total number of interstate pipelines is higher in the United States. Besides, the physical characteristics of interstate pipelines in the United States show higher values of lengths, mean values and pipeline capacities. The US network has more interconnection points, however Europe does not fare worse according to
these interconnection points due to total pipeline lengths56.
53 According to the Directorate General for Energy and Transport, unit electricity and gas, 54 Conclusion based on European Commission reports, published January 10th 2007;
http://www.ear.eu.int/publications/main/pub-reports.htm,
Table 3: US – Europe comparison of industry structure
Variable \ Industry United States Europe (EU-25)
Number of inter-state pipelines
118 pipelines
(EIA-176 Query System, 2006)57
40 national, 38 regional (European Commission, 2005) Length of interstate
pipeline (in miles)
221,075 miles58 (2006) Mean: 2,028 miles (2006) St. Dev: 3,221 (2006)
(EIA-176 Query System, 2006)
18,542 miles (2007) Mean: 515 miles (2007) St. Dev: 608 miles (2007) (Jamaasb et al., 2008; Makholm, 2007)
Average annual flow59 (Year: Total(bcf/d); Mean; St.Dev)
1989: 263,372; 2,232; 5,326 1990: 263,372; 2,232; 5,326 1991: 270,672; 2,294; 5,407 1992: 277,973; 2,356; 5,491 1993: 292,573; 2,479; 5,668 1994: 292,573; 2,479; 5,668 1995: 298,115; 2,526; 5,703 1996: 301,475; 2,555; 5,737 1997: 310,065; 2,628; 5,853 1998: 319,289; 2,706; 5,950 1999: 324,637; 2,751; 5,975 2000: 335,535; 2,844; 6,009 2001: 340,955; 2,889; 6,060 2002: 354,917; 3,008; 6,193 2003: 363,439; 3,080; 6,284 2004: 367,211; 3,112; 6,315 2005: 369,136; 3,128; 6,317 2006: 372,115; 3,180; 6,346 (EIA-176 Query System, 2006)60
57,300 bcf/d in 2005 (European Commission, 2007)
Interconnection points 308 (2008)61 79 (Gas Infrastructure Europe website,
2008)62
Market hubs 14 (Energy Information Agency, 2007) 13 (European Commission, 2007)
57 http://www.eia.doe.gov/oil_gas/natural_gas/applications/eia176query.html , 58 Average of 110 pipelines. Lengths of 8 interstate Pipelines were not available,
59 The average annual capacity is measured by calculating the total realized average daily flow through all
interstate natural gas pipelines together in the year selected,
60 http://www.eia.doe.gov/oil_gas/natural_gas/applications/eia176query.html,
3. Model and variable selection
Experiences from Jamasb et al. (2008)63 learns that applying company’s financial and operational performance and efficiency measurements are likely to be sensitive to the variables selected and the model chosen. Jamasb et al. (2008) model efficiency “for a given unit in a given year as the distance to an input-oriented, constant returns to scale, non-parametric frontier”. They use a Data Envelopment Analysis (DEA) for the static efficiency scores and a Malmquist Productivity Index for productivity change and its decomposition into technical efficiency change and technical change (Jamasb et al., 2008).
Jamasb and Pollitt (2003) state that “the central aims of regulatory reforms are the introduction of market-oriented measures into natural gas transmission and distribution, and to improve the efficiency of the activities of transmission through regulatory reforms.” A central issue in their article is how to set the efficiency requirements. They benchmark the most efficient firms in the electricity sector and measure the relative performance of less efficient firms against these. The electricity sector is part of the wide utility-sector and is therefore comparable with the natural gas sector for this area of research (Jamasb et al. 2008). This study differs from Jamasb et al (2003), because it does not make a distinction between companies’ efficiency performances.
There are several different approaches to the measurement of efficient pipeline utilization of interstate pipeline companies. In this research, I make use of the six-stage model-building process as presented by Hair et al. (2005). The application of the multiple regressions used can be categorized as explanatory, because I examine the regression coefficients for each independent variable and the substantive or theoretical reasons for the effects of these independent variables on the dependent variable. The selection of the independent variables is based on their statistical64 relationships to the dependent variable, and the choice is also based on conversations with specialized employees from Dutch “Gasunie N.V.”. The hypothesized directions (positive, neutral or negative) of the independent variable relations with the dependent variable are presented in table 6. The outcomes of this research are sensitive to choices made, and are only reliable if the same variables and research model are used.
The aim of research is to explore how two selected regulatory FERC orders65 and a rise in prices for natural gas in the United States66 have had an influence on efficient pipeline
63 Jamasb et al., 2008,
64 Whereas a functional relationship calculates an exact value, a statistical relationship estimates an
average value (Hair et al., 2005),
65 FERC Order 636 in 1993 and FERC Order 500 in 2000 respectively, 66
utilization. Increasing efficiency is an important driving force when a market liberalizes (Rider, 1999; Muir, 2000). Efficient pipeline utilization is a direct derivative of company efficiency; therefore pipeline utilization is the dependent variable to declare in this research. Because multiple regression is a dependence technique, there are three issues that can affect any decision when selecting independent variables: strong theory, measurement error, and specification error. In the following paragraph the choice of independent variables will be motivated.
The research consists of more than one independent variable; therefore the statistical technology multiple regression is used to test the influences and contributions of the independent variables (or explanatory variables or predictors) on the dependent variable (or criterion variable or response variable) by fitting a linear equation to the observed dataset. Every value of an independent variable x is associated with a value of the dependent variable y.
The multiple regression model used for the explanatory variables x1, x2, …, xt is defined as:
where t represents the number of independent variables selected and n represents the year to investigate.
The assumption is made that the observed values for Y (t=118) vary about their individual means Uy. This multiple regression model includes a term for this variation, which represents the deviations of the observed values y from their means Uy, which are normally distributed with mean “0” and variance ơ. The notation for this model deviation is ε.
The values fit by the model above are denoted as Ŷit and the residuals ei are equal to Yi - Ŷit, the difference between the observed (Yi) and fitted values (Ŷit). The sum of the residuals is equal to zero.
The variance ơ2 may be estimated by s² =∑ei2/n-p-1, also known as the mean-squared error (or MSE). The estimate of the standard error s is the square root of the MSE. More technical information about the multiple regression model, the parameters, equations and assumptions made are presented in Appendix B.
In this research, the multiple regression models are used annually from 1990 until 2006, with the independent and dependent variables constant, however with different values for these
variables dependent from the year67 of research. The outcomes of the multiple regression tests will be compared to analyze the independent variable annual mean values and to analyze the changes in independent variables’ contribution to predict the dependent variable, which are probably changed due to FERC regulations68 or spot price increases69. Dependent from the changed contributions of the independent variables on the dependent variable, the influence of regulations and price effects on efficient interstate pipeline utilization is analyzed. The selection of independent variables is based on strong theories and discussions with specialists from “Gasunie N.V.”. There is no measurement error, because the dependent variable “efficient pipeline utilization” is an accurate and consistent measure of the concept studied: “efficiency”. I am aware of the specification error that exists because of the inclusion of irrelevant variables in the set of independent variables. However, the independent variables selected are all included to find out what their changed contributions towards predicting the dependent variable are. As pointed out before, efficient pipeline utilization could be inherently affected by many other external factors, and therefore it is not possible to make accurate predictions. Multiple regression allows to identify a set of predictor variables which together provide a useful estimate of an interstate pipelines efficiency score on the criterion variable “efficient pipeline utilization”.
3.1. Interstate natural gas transmission: Literature
An article about transmission of natural gas focus on production functions where the prime output is gas delivered and inputs are capital and labor is written by Jamasb et al., (2008). Where Callen (1978) focuses on the technical aspect of natural gas transmission, Sickles and Streitwieser (1991) highlight the distances in gas transport as a function of delivery volumes. Ellig (1993) researches the O&M expenses as an effect of total pipeline throughput, length of pipelines, gas purchasing costs, third-party delivery volumes and annual sales. Ellig uses a sample size of fifty Texan gas transmission companies in 1989. The recent article of Jamasb et al. (2008) researches the impact of US regulatory reform by using a Malmquist-based productivity analysis for a panel of 85 US interstate pipeline companies. They conclude that if one takes productivity and convergence as performance indicators, regulation has been rather successful, in particular during a period of demand stability. In their article, they highlight a research of Granderson (2000), who analyzes the effect of horsepower weight of pipeline steel and labor on compressor fuel by researching 20 US interstate pipeline companies from 1977 until 1987. These studies all focus on the US interstate gas transmission, however they do not all include important conclusions about efficiency. Therefore, from the highlighted articles above only the article of Jamasb et al. is useful as an important addition to this research.
67 Years of research: 1990; 1991; 1992; 1993; 1994; 1995; 1996; 1997; 1998; 1999; 2000; 2001; 2002;
2003; 2004; 2005; 2006,
Juris (1998) discusses the changed competitive environment in the United States as a result of regulations. He concludes that if companies focus on efficiency, markets become more transparent and diversified: “The focus on efficiency by interstate transmission companies lowered natural gas prices and broadened the range of services offered by gas companies in the United States”. I do not agree with this assumption, because natural gas prices increased, while companies still focus on efficiency. Besides, I argue that external factors70 could have an influence on the relation between natural gas prices and companies’ efficiency.
I use a model that for the purpose of this research is input-oriented and therefore treats output as a declaration of the dependent variable. The input consists of six independent variables, which will be discussed in the following section.
Table 4: Summary of the literature about U.S. interstate pipeline performances
Authors Data Dependent
Variable Independent Variables Aim Callen (1978) US inter-state gas transmission companies in 1965
Delivery volume Horsepower; weight of pipeline steel (technical) Analyzing social welfare impact of rate of return regulation on this industry Sickles and Streitwieser (1991) 14 US inter-state pipeline companies in 1977-1985
Delivery volume Transport distances Impact of the NGPA on the technical efficiency and productivity of interstate pipeline companies Ellig (1993) 50 Texan gas
transmission companies, 1989
O & M Expense Sales; total throughput; third-party delivery volume most firms operate at substantial decreasing returns to scale, and the largest firms suffer the biggest diseconomies of scale Juris (1998) Effects of 15 years of regulation in the United States Functioning of the natural gas market in US
Regulatory changes
Lower natural gas prices and broadened range of services provided by gas companies
Granderson (2000) 20 US interstate pipeline companies in 1977-1987
Compressor Fuel Labor; horsepower weight of pipeline steel
Open access on gas markets lead to a small reduction in transportation cost and small increases in firm efficiency.
Micola & Bunn (2006)
3 sequential, oligopolistic energy markets, 2005
Profit increase Price;
interdependency higher prices relate to the nature of energy markets, which facilitate the emergence of financial netback effects. Jamasb et al. (2008) 85 US interstate pipeline companies, 1996-2004 Productivity Efficiency; efficiency change; technical change; convergence; total factor productivity change In the long-run, market integration and competition are alternatives to the current European model 3.2. Variable selection
First, I turn to the dependent variable in the multiple regression analysis: efficient pipeline utilization. The affiliate rules from the Natural Gas Supply Association (NGSA)71 state that there is a significant integration between the natural gas- and electric power industries. Despite of this integration, energy security of supply, energy independence and environmental sustainability remain important drivers from a customer point of view within the natural gas market (Zabors, 2008). He concludes that “Energy efficiency can create incremental capacity, energy and ancillary resources”. Tubb (2002) concludes differently by suggesting that as interstate pipeline systems expand, load duration curves or efficient pipeline utilization curves will have to respond accordingly. The relationship between degree of capacity utilization of interstate pipelines and the natural gas price is researched by Micola & Bunn (2006). They conclude that there is a relation between natural gas prices and pipeline utilization; however this relation is weak due to the degree of information available, uncertainty or risk aversion and because of the increased utilization of interconnector72 points in natural gas transmission. Based on the article of Micola & Bunn (2006), I argue that it is not necessary to include natural gas prices in the multiple regression models used.
71 http://www.ngsa.org/docs/affiliate_rules.pdf,
