Sidney P. Santos
Compression Service Contracts – When is it Worth it?
Sidney Pereira dos Santos,PETROBRAS – Gas Business Unit, Eduardo Saliby, COPPEAD / UFRJ
Copyright 2003, Pipeline Simulation Interest Group
This paper was prepared for presentation at the PSIG Annual Meeting held in Bern, Switzerland, 15 October – 17 October 2003.
This paper was selected for presentation by the PSIG Board of Directors following review of information contained in an abstract submitted by the author(s). The material, as presented, does not necessarily reflect any position of the Pipeline Simulation Interest Group, its officers, or members. Papers presented at PSIG meetings are subject to publication review by Editorial Committees of the Pipeline Simulation Interest Group. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of PSIG is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, Pipeline Simulation Interest Group, P.O. Box 22625, Houston, TX 77227, U.S.A., fax 01-713-586-5955.
It is not uncommon to face a situation when we need to make capital investment decisions to increase transportation capacity of a pipeline under uncertainties such as market development, project costs, schedule and so forth.
This was the situation we faced related to the energy shortage in Brazil that prompted for the need of alternative and reliable energy sources that could be put into operation in a short period of time based on many prospective Gas Fired Power Plant projects but without having the necessary agreements signed.
The Ministry of Mines and Energy of Brazil set a program addressing initially 55 thermo power plants totaling about 20,402 MW. From this total 18,263 MW of installed power was from 49 gas fired power plants demanding gas volumes in the range of 88 MMm3/d, most of this power were to be available from 2001 to 2003.
With this challenge, Petrobras has started to design a gas pipeline network expansion plan with investments of more than 1 billion US$ for its system alone, including new gas pipelines, new compressor and custody transfer stations and loop lines, in addition to expansion projects for the Bolivia-Brazil Gas Pipeline in Bolivia (0.2 billion US$) and in Brazil (0.35 billion US$), and the new gas pipeline from Argentina to Brazil (0.25 billion US$).
Under this scenario we considered the option of contracting compression service for some pipelines in our networks as an alternative for conventional and permanent compressor stations while discussing the gas and transportation contracts and others investments.
We did a feasibility analysis for two alternatives of compression service contract and permanent compressor station both using Monte Carlo Simulation Method. The results and methodology are presented in this paper.
Petrobras Gas Pipeline Expansion Project has faced a challenging situation: How to get prepared for a very high growth scenario in demand related to the program of The Ministry of Mines and Energy of Brazil that addressed the installation of 55 thermo power plants totaling about 20,402 MW? From this total, 49 gas fired power plants – GFPP (8.5 GW planned up to 2003) demanded gas volumes in the range of 88 MMm3/d that would more than double the installed gas pipeline transportation capacity and most of this power were to be available from 2001 to 2003. By the end of 2002, 4.6 GW of GFPP was already installed.
The energy shortage in Brazil prompted for the need of alternative and reliable energy sources that could be put into operation in a short period of time based on many prospective Gas Fired Power Plant projects but without having the necessary agreements signed.
Since that program was conceived in a changing scenario which also had undergone a market deregulation and the establishment of Brazil Regulatory Agency (ANEEL) and Energy Wholesale Market (MAE) and also considering that the prompt for all this program was the energy crisis in early 2001 due to a very low water level in the hydro plants reservoirs caused by a low rain precipitation, so fundamental to the hydro power generation, and also difficulties related to strategic decisions on the electric sector. Hydro power generation is more than 80 % of total of energy generation in Brazil, explaining how critical the crisis was.
To face that challenge we had to adopt a different approach with relation to how to expand transportation capacity and how to make investment decisions under uncertainties such as market growth, project costs, schedule and so forth.
We had started to design a gas pipeline network expansion plan with investments of more than 1 billion US$ for its system alone, including new gas pipelines, new compressor and custody transfer stations and loop lines, in addition to expansion projects for the Bolivia-Brazil Gas Pipeline in Bolivia (0.2 billion US$) and in Brazil (0.35 billion US$), and the new gas pipeline from Argentina to Brazil (0.25 billion US$).
To provide additional transportation capacity at a lower cost and tight schedule we considered the option of contracting compression service for some pipelines in our gas pipeline networks as an alternative to conventional and permanent compressor stations, while negotiating gas supply and transportation agreements and others investment decisions.
We did a feasibility analysis for two alternatives: compression service contract and permanent compressor station both using Monte Carlo Risk Analysis Simulation. The results and methodology are presented in this paper.
Risk analysis simulation has been used with increasing frequency as a risk mitigation tool in replacement to sensitivity analysis based on scenarios (least, normal and most favorable) that do not cover all possible probabilistic events that a project is subject to (Hertz, 1964; Vose, 1996).
In this paper the Monte Carlo Risk Analysis Simulation is used and a practical example is presented to illustrate its applicability to projects, requiring the identification of probabilistic distributions of interest variables associated to the project in addition to the normally used information for investment decision analysis.
This practical example covers the economic feasibility analysis of two alternatives for the installation of a compressor station in a gas pipeline in Brazil to increase gas transport capacity:
• Fixed compressor station;
• Compression service contract from a specialized company – Service Co.
The construction and assembling of a compressor station by a transportation company – Transco – the company that holds right to transport or is proprietary of the gas pipeline, follows certain requirements which must be satisfied, a priori, for its implementation:
• The signature of a ship or pay gas transportation agreement between Transco and Shipper – the company that holds the selling right or has its property and wants to sell the gas to the market using Transco to provide gas transportation;
• A 24 months schedule required for the compressor station construction, assembling and commissioning;
• Capital resources for equipment acquisition and construction and assembling agreements;
• Qualified staff to operate and maintain the installation.
In the other hand Shipper will only sign the ship or pay agreement with Transco after having signed a take or pay agreement with the local gas distribution company – LDC, that company that holds the right for gas distribution to end users.
Making even more complex this gas business chain LDC will only sign the take or pay agreement with Shipper if also has take or pay agreements signed with major end users so as to have guaranties that will enable it to honor its contractual obligations.
We all know that all that contractual negotiations between companies, necessary to close the deal, requires months or even years to be accomplished relying on market maturity, gas price policies. This contributes for the uncertainties we face nowadays in Brazil gas market.
The agreement between Transco and Service Co. presents advantages that help the gas business chain. The shorter implementation schedule, in the range of 6 months, and the fact that this option do not require capital investment from Transco but only operational expense against compression service provided by Service Co. presents a very attractive option for Transco.
2.1 Methodological Steps
1. Define project configuration.
2. Identify project financing indexes to be achieved such as IRR, debt-equity ratio and debt interest rate.
3. Identify relevant items such as capital required for equipment acquisition, construction and assembling costs, revenues, taxes, general and administrative costs, depreciation criteria and other items.
4. Identify uncertain variables and define probability distribution and their parameters.
5. Create a project economic model using a spread sheet to evaluate NPV and IRR.
6. Use a risk analysis software (@Risk 4.5 in this case) linked to a spread sheet to carry out the simulations.
7. Define the configuration for the risk analysis model such as sampling method and parameters for the simulation (trials, simulation runs, random seeds, etc)
8. Statistic analysis of the results (tables and graphics).
9. If necessary, define new actions that may mitigate risk levels identified from the simulation.
10. If necessary, review and adjust the model so as to satisfy return rate expectation of the project sponsors.
3. Case Study
This case study is based on the Volta Redonda Compressor Station that was installed under a compression service contract on the Rio de Janeiro – Volta Redonda Gas Pipeline – GASPAL. The decision on signing a compression service contract instead of installing fixed compressor station has followed the methodological steps described on item 2 above and the details and results are presented on the next items.
3.1 Project History
Due to market development in Rio de Janeiro related to GFPPs installation we had to anticipate 3.23 MMm3/d of incremental transportation capacity for the GASVOL gas pipeline so as to make Bolivian gas available to Rio de Janeiro market. We had a very tight schedule for the installation so we made a bid for contracting compressor service while we also analyzed the option of installing a fixed compressor station to have a way to evaluate the proposals received.
3.2 Business Model
The natural gas business model that makes the environment under which we analyzed the case study has a configuration that involves different interdependent players that wants to be protected from risks and therefore want to have agreements signed to cover them up.
With regard to compression service whenever there is a market opportunity to place more gas volumes Shipper will start negotiating with Gas Producers and LDC to have the ends meet. At the same time you start negotiation with Transco to have transportation capacity for the new demand. All of these deals takes time and are not an easy task in a new and under development gas market such as in Brazil.
The figure below illustrates this model focusing on the opportunity of having a fixed compressor station installed or to have a compression service contract to increase transportation capacity by 3.23 MMm3/d for GASVOL gas pipeline.
3.3 Technical, Economic and Costs Data for Compressor Station
The following data were adopted for the technical and economic evaluation.
For Permanent Compressor Station we estimate de costs based on a previous acquisition of Araucária Compressor Station that is similar to Volta Redonda Station and for Temporary Station we get cost information from Service Co during the bidding process.
Transco return rate: 15%
Service Co. return rate: 23.87% (based on contractual charges, see figure 2)
Transportation rates: @NPV=0 and return rates for Transco and Service Co.
Contractual Schedule: 3, 6 and 9 years
Debt-equity ratio: 0% (without leverage)
Depreciation: 10% a year
3.3.1 Service Co. Return Rate Evaluation
We calculated the Service Co. return rate as 23.78% based on the contractual charges for a tree years contractual term and the station costs presented at the bid. The spread sheet below shows all the cost items adopted and also includes some specific taxes applied in Brazil for this type of activity.
4. Risk Analysis Modeling
The risk analysis modeling consists of selecting the uncertain variables of the project and its probabilistic distribution linked to the economic spread sheet and by using a risk analysis tool (@risk 4.0) permits to make a predefined number of runs that provide sampling results from a sorting process based on Monte Carlo Simulation. The aim was to verify how the project IRR and NPV would vary as a function of the simulation process.
4.1 Uncertainty VariablesFor this kind of project we select Goods & Service costs, Transported Volumes, C&A Schedule and Salvage Costs.
4.2 Basic Spread SheetWith the contractual revenues and Transco return rate evaluated as shown in item 3.3.1 we calculated what would be the transportation rate on reference year (year 0), based on the transported volumes that would give the same revenues. We also considered as escalation factor of 0.5% a year for the transportation rate. This approach was carried out for each alternative and each contractual term (3, 6 and 9 years) and the resulting rates were then kept unchangeable and become the basis for the risk analysis simulations performed for each alternative. Below we see, as an example, the basic spread sheet for the Compression Service Contract with 3 years contractual term.
4.3 Risk Simulator ConfigurationAfter having defined the basic spread sheet as item 4.2, we included in the spread sheet de uncertainties as shown in item 4.1 and set the probabilistic distribution type (normal and triangle) for each of then to be used by the risk analysis simulation tool (@Risk 4.0) with the following command equation lines:
=RiskNormal(U25, V25, RiskName("Goods & Services Cost"))
=RiskTrigen(U27, U28, U29, 10, 90, RiskName("C&A Schedule"))
=RiskTrigen(U30, U31, U32, 10, 90, RiskName("Salvage Cost"))
=RiskNormal(U33, V33, RiskName("Transported Volumes"))
We also set the simulator parameters such as:
Number of iterations: 500
Number of simulation: 1
Sampling type: Latin Hypercube
Random generator seed: 1
Standard recalculation: Expected Value
Collect Distribution Samples: All
5. Risk Simulation
Running the simulation risk analysis we can produce tabular and graphic results that will allow us to visualize how risky is the alternatives and what measures we need to take if we find necessary to adjust some model values such as return rate or even spend more effort to narrow down the project uncertainties variables.
5.1 Tabular Results
5.1.1 Year 3
5.1.2 Year 6
5.1.3 Year 9
5.2 Graphical Results
5.3 Results Comparative Analysis
As we can see from the tabular and graphic results presented in item 5.2 the alternative of Compression Service Contract was the one who presented the lowest risk or the lowest variation of IRR and NPV, for years 3, 6 and 9, around their mean values. This means that even if all uncertain variables reach their worst values this alternative will still hold itself profitable. As for the others alternatives they would expose the Transco to a negative or close to zero return. The table below presents a comparison between the alternatives.
5.4 Sensitivity Analysis
By running the sensitivity analysis we can see how the variation of one standard deviation of one uncertain variable will affect the objective function (IRR or NPV) and therefore we will visualize which of them worth concentrating additional effort to mitigate its influence on the total risk of the project. The tornado graphic type presents a regression for it uncertain variable and quantifies its influence for 3 and 9 years of contractual term.
5.4.1 Compression Service Contract - IRR
5.4.2 Compression Service Contract - NPV
5.4.3 Fixed Station C&A 12mo. - IRR
5.4.4 Fixed Station C&A 12mo. - NPV
5.4.5 Fixed Station C&A 24mo. - IRR
5.4.6 Fixed Station C&A 24mo. - NPV
6. Final Considerations
This case study illustrates how important is the risk simulation for a project feasibility analysis under uncertainties. For example what would appear to be an attractive return such as 15% for Fixed Compressor Station with C&A of 12 months, proved to have a probability of ___ to be between __ and __ % and of __ to be between __ and __ %. For the Compression Service Contract the figures are much better and we have a a probability of ___ to be between __ and __ % and of __ to be between __ and __ %.
Based on this practical experience we can see the usefulness of the Monte Carlo Risk Simulation for a project feasibility analysis and we also could overcome a prevailing paradigm which stated that a fixed compressor station was ever a better economic option to increase transportation capacity of a gas pipeline.
- ROSS, S.A.; WESTERFIELD, R.W.; JAFFE, J.F. - Corporate Finance, fifth edition: The McGraw-Hill Companies, Inc., 1999.
- COPELAND, TOM; KOLLER, TIM; MURRIN, JACK. Measuring and Managing the Value of Companies – 3rd edition. John Wiley & Sons, Inc., 2001.
- HERTZ, D. Risk analysis in capital investment, Harvest Business Review, 42(1): 95-106, Jan/Mar. 1964.
- VOSE, DAVID. Quantitative Risk Analysis: A Guide to Monte Carlo Simulation Modeling. New York: John Wiley & Sons, 1996.
- RAFTERY, JOHN. Risk Analysis in Project Management. New York: E & FN Spon / Chapman & Hall, 1994.
- WINSTON, WAYNE L. Simulation Modeling Using @Risk. Canada: Duxbury / Thomson Learning, 2001.
8. About the Authors
Sidney Pereira dos Santos, the author, is a technical consultant at GASPETRO, has a Mechanical Engineering degree and a MBA in Corporate Finance, has 13 years of experience in shipbuilding design, and also 13 years in the oil and gas pipeline design at Petrobras . Has participated in most gas pipeline projects in Brazil, as the Bolivia-Brazil project. Has been conducting technical-economic studies and basic/conceptual design for the upcoming projects. He also participated as Project Manager in the first two years of operation (1998-1999) of the Bolivia-Brazil transportation company in Brazil side.