2/2/98

Perot Systems Corporation CONFIDENTIAL Policy Assessment Corporation

PROJECT TASKS AND DELIVERABLES

The project phases are designed as self-contained tasks whose completion verifies that step of the project.

1. Phase I:
* Determine the real-time optimization opportunities in California. The focus will primarily be on electricity trading, but gas trading will be included if appropriate. Trading strategy will be developed for both physical and non-physical assets. Real- time efforts (moves) can include extended periods strategies.
* Use existing software tools and convert the moves to operationally reflect the detailed California PX/ISO specifications and protocols. Select the subset of moves that corresponds to the way Enron would like to do business.
2. Phase II: incorporate the data streams that will be available from the California PX/ISO and other market participants as soon as they are known and verify that the system performs well under the expected and unexpected operational conditions of the California PX/ISO.
3. Phase III: fine tune the model parameters as soon as actual California PX/ISO data support such adjustments. Further, provide support to Enron staff during system start- up , and thereafter as needed, to ensure maximum profitability from the system.
4. Phase IV: repeat tasks listed in Phases I-RI for WSCC or other locales (These tasks could be started in parallel with the PX/ISO tasks.)

For all phases, great effort would be made to keep Enron staff fully cognizant of the model's technology and operation.

PSC/PAC do not claim to have expertise in all aspects of electricity trading. The intent of this proposal is to complement Enron's existing electricity trading team with a real-time market-responsive software system and PSC/PAC consulting to significantly enhance the overall value for Enron. The focus is on formalizing the processes to limit the impacts of constraints and to leverage the opportunities available within the rules of a transitional marketplace.

Phase I Tasks

Tasks

1.) Review PX and ISO protocols and business opportunities: This task involves a detailed review of PX and ISO business and operational protocols and particularly the interactions between these two entities. The main goal is to become well versed with the ongoing changes in PX and ISO business and operational protocols, and to become aware of which strategies are available to competitiors in the California market.

Deliverable: The strategies will be documented.

2.)Implement the reduced scale PX/ISO operational model (CPXISO): This task implements a reduced scale model that would closely approximate the operation of California's PX and ISO systems. This task starts with existing full scale models available to PSC and PAC. Included in this model will be the PX's energy auction model, the ISO Congestion Management model, and the interaction between these models. The full 6000 bus ISO linear programming model will be simulated using a much smaller set of buses and nodes (approximately 100 buses). This ISO model, even though it will be much smaller, will be much faster and will effectively simulate the results of the actual ISO simulation and dispatch. Data interfaces between PX, ISO, and the deregulation model (CIGMOD) will also be developed. Data models for external systems will also be developed in this task.

Deliverable: A working version of the CPXISO system.

3.) Incorporate strategies into system: Based on Task 1, the full spectrum of strategies will be developed for each of the participants in the market. This task will emphasize getting the "levers" into the model to allow for the development of ever more sophisticated strategies. The full range of strategies will be linked to the HYPERSENS subsystem and thereby allow the exhaustive testing and selection of strategy portfolios under uncertain future conditions.

Deliverable: Working HYPERSENS subsystem with implemented strategy levers.

4.) Adapt PAC's short-term load forecast (SLF) model to work for California's energy market: This task involves integrating and fine tuning the SLF model for California competitors and developing interfaces with the PX/ISO operational model. The system will forecast hourly demands for all players in and out of California. The SLF model produces hourly load forecasts for the next 7 days based on the historical loads, the historical weather, and the weather forecast. It re- estimates the relevant coefficients on a daily or hourly basis by live links to the weather service and existing conditions.

Deliverable: Existing system with SLF integration.

5.) Develop interfaces between CIGMOD and CPXISO: The system interface includes system-control, Enron data-entry, and Enron bidding strategy output. The interface between the model and the operators will automate the input of hourly data, will produce timely reports and charts which effectively communicate the relevant information, and will facilitate the operator's decision making process.

Deliverable: Existing System site tested interface.

6.) Add Enron Proprietary Data: Any proprietary data that Enron would like included can be added at this time. From this point forward, the databases of the model cannot be removed from Enron offices without written permission.

7.) Calibrate CIGMOD/CPXISO model to California and Enron behavior: Data for each company and plant in the region will be incorporated into the model. This data will be available for use in determining bidding and market-placement strategies (prices and quantities) and will be used to determine the operating income of Enron and each of its competitors. The data will include capacities, heat rates, fuel prices, variable and fixed O&M, and embedded capital costs. The CIGMOD model will be calibrated to competitor and Enron data. The calibration process will test the model and will estimate critical operating and financial parameters.

Deliverable: Existing model with tested simulation of all relevant companies.

8.) Test model with Enron staff and update algorithms/interface: This task will test the operation on the model to insure that the simulation of the ISO, PX, and company financials are accurate and that the interface is informative and easy to use. Strategies will be tested to insure that they are simulated correctly and that the results are reasonable. The algorithms and interface will be revised and enhanced as necessary.

Deliverable: Staff training and system with enhanced interface

9.) Test model and strategies against realistic conditions: The system will be used to test the basic strategies to determine the realistic impact of these strategies. The growing list of strategies will be tested in several sets of simulations. These simulations will test human strategy against human strategy and human strategy against machine strategy to determine the effectiveness of the strategies and to develop alternative strategies. As good strategies are developed, counter strategies will be developed, incorporated and tested for their robustness. The creativity of the human players will be very important in developing new strategies. After a new strategy is developed it will be automated and utilized by computer players.

Deliverable: Fully functional system except for real-time PX/ISO analysis.

CIGMOD and associated software: Deliverable: At the beginning of the project, the CIGMOD system will be implemented on the Enron computer system for this project.

Phase II Tasks

Tasks

Integration of California's market data into CIGMOD/CPXISO model: This task will determine the availability of the data from the PX and the ISO. The assessment will include the timing of the data, the company and node detail of the data, and the form and source of the data. Any expected or possible problems or inconsistencies of the data will be noted and solutions proposed. Interfaces to allow the rapid inclusion of changing protocols/conditions will be included. Implement Al/statistical code into CIGMOD/CPXISO model:. The data stream will be analyzed to determine its use in the forecasting of competitor actions, spot market prices, and other system values. As data allow, the system will be modified to automatically analyze the data using accepted statistical and artificial intelligence methods.

Interface to real-time systems: . A real-time data retrieval system will be developed to bring in the data from the PX, the ISO, and other relevant data sources. Because Perot intimately knows the system, this task will provide Enron with the best understanding of the data available.

Develop portfolio rules: Portfolio rules for real-time and over-all strategies will be developed by testing each strategy under a variety of market conditions and competitor strategies. Portfolios as used here are defined as the time-dependent packages of bids and demands/supply that maximize Enron goals. Initial testing and training of Enron staff: This task will test the operation of the system by creating a synthetic hour-by-hour data stream. Enron staff will conduct the test after being trained on the use of the system. The results of the test will be used to revise and enhance the system.

Market dry runs: The proposed strategies and portfolio rules will be tested in real-time to determine realistic use of strategies and the operation of the system. Multiple sample data streams will be developed to test all aspects of the system. The decision making process of the competitors will be split between human players and computerized decision rules. This test will be used to further enhance the strategies and counter strategies for Enron in the California market.

Phase III Tasks

Tasks

Monitor early use and fine tuning of market strategies and addition of new Al based strategies: During the first days of the deregulation, large amounts of new information will become available. This information will be used to immediately improve the performance of the system. The task will determine and update any simulation sub-systems needing tuning or enhancement. This task will also test and refime the fail-safe algorithm for bidding under unlikely conditions where the system warns users of an ambiguous situation.

Develop benchmark saving protocols and models: To determine the efficacy of the system, it needs to be benchmarked against "reference" conditions. This task will provide a secondary simulation that has the market acting according to assumed economic considerations such as marginal costs pricing or the existing trends in the market. A continuous accounting record of wins and losses will be maintained in the system.

Ongoing support Enron staff: The PX/ISO protocols and rules will probably change regularly in the early period of deregulation. Further, new commission rulings must be expected. This task provides Enron staff assistance in updating the systems and its data bases to reflect new conditions. Further, it provides assistance in strategy development when the new rules change the nature of the market operation.

Moving to Competitive Utility Markets:
Parallels with the British Experience

by
Dr. George Backus
Policy Assessment Corporation
Denver, Colorado, USA
and
Susan Kleemann
Systematic Solutions, Inc.
Dayton, Ohio, USA

Introduction
Although the UK privatization and the US deregulation of the electric industry start out from different places, comparative historical and simulation analyses show they will tread similar economic paths and quickly develop into dynamic industries characterized by gaming, volatile prices, overbuilding and eventual consolidation.

The British Experience with Privatization The official UK deregulation is considered to have begun on March 1, 1991. As of summer 1996, privatization of the nuclear components and the National Grid Company (NGC) appeared imminent. With an average load of 50,000MW, the UK capacity mix includes both old and new, nuclear, hydro, pumped storage, combined cycle, gas, oil, waste, coal, and wind generation. At the initial break-up, the UK system was comprised of 12 Regional Energy Companies (RECs), three primary generators, independent power producers (IPPs), "special" entities such as pumped storage facilities owned by the NGC (National Grid Company), and two vertically integrated Scottish companies.

Unlike the US, the UK government owned the original Central Electricity Generating Board (CEGB) whose break-up and privatization led to the current system structure. The CEGB components were sold for less than 20% of their estimated book value. The RECs are regulated via prices that escalate with adjusted inflation. Under-priced assets along with the new flexibility to control costs, naturally led to impressive profits. Because most of the US system is privately held, this mechanism for profit-taking will not be part of the US market dynamic - other than possibly through the acquisition of municipal, cooperative and large federal PMA assets.

The British system is essentially a spot market pool where all participants must bid in the day before. The NGC then determines a least-cost dispatch based on anticipated load, merit order (bid prices), and plant availability. For the next day's dispatch all generators are paid the current system marginal price (the bid price of the most expensive generation unit needed to operate). System operations and load following may require some higher costs plants to be dispatched early (constrained-on). Some lower cost generation may not be compatible with current load conditions and could be constrained-off. All available generators, running or not, receive a capacity payment (called the LOLP) which is the LOLP*(VOLL-max(SMP,bid price)), where the LOLP is the loss of load probability and VOLL is the value of loss of load.

The Phases of Deregulation
Transient disequilibrium conditions are forced on US and UK electric markets going from a regulated to a deregulated environment and they last until existing power plants, transmission lines, and customer equipment convert to a configuration appropriate for the new environment. During transition, market distortions necessarily exist which provide opportunities to the clever and catastrophe to the unwary.

The six phases of deregulation are arbitrary, overlapping classifications. The last phase may begin before the first phase ends; some phases may dominate the market for years while others may be passed through before they are recognized. Growing evidence from both the British experience and simulation modeling suggests that the US system will pass through all the phases without exception and reach the final phase as its still dynamic endpoint. The six phases are:
Transitional Market, Massive Restructuring, System Divestiture, Market Gaming, Re-regulation, and Industry Consolidation.

Phase 1 - Transitional Market The transition phase corresponds to regulators' attempts at "giving" up control while controlling the consequences. The changes between the FERC Open Access Ruling and the original Mega-NOPR alone substantiate this premise. The Open Access Ruling really had little to do with transmission - transmission simply provided the last vestiges of control. This control is needed as regulators continue to respond to complainants who want a level playing field that tilts in their favor. As each rule affects a new set of constituents, more rules are invoked in the name of "fairness." Ultimately the rules increase to the point where compliance becomes an overwhelming task and a collective solution that removes direct responsibility must found - in this case usually an Independent System Operator (ISO). Although an ISO will work, secondary dynamics result in excess capacity, non-converging higher prices and gas market consolidation.

Volatdle, Non-convergingPrices
As the UK utilities have learned, "optimal" analyses don't work in a competitive environment with physical constraints, because any "gamed" action makes the optimal plan less than optimal, and the competitor's loss is the game player's win. Further, this "gaminge advantage provides economic motivation to avoid the "market clearing price" and to keep the market volatile - just like in any other commodity market. Evidence of this gaming behavior is found in UK bulk power markets - a highly volatile market where provisional pool purchases prices have varied from lp/kwh to 1 .11/kwh. Figure 1 displays the dynamics of the bid price for aggregations of plant-types that should have similar, consistent avoided costs. "The message from the UK is clear. It was incorrectly assumed that the new commercial entities would continue to operate by the intent of the rules, even if not formally stated, when the new structure began. But commercial markets are commercial markets, profits are profits and any commercial advantage will be taken." (Tabors 1996, p. 49)

Comnparison first incremiental bids

[GRAPHIC OMITTED]

Figure 1: Bid rice Dynamics By Pant Type (OFFER 1992)

An historical analysis of independent power production i the US indicates that deregulation will not produce generation marginal cost pricing either. In regions where IPPs are common prices were neither falling not converging to a single value; the prices from one of the operating PPs could be twice that of the least expensive operator. (Comnes 1996)

Fipgre 2: Declared Capacity Vaiation (OFFER 1992)

Excess Capacity
In the UK, the "books" show roughly a 20% reserve margin, but that neglects 8GW of disconnected capacity. If all this capacity were included, the reserve margin would approach 40%. If all planned capacity is considered, including 24.4GW of CCGT, then the reserve margin approaches 50% by 2000 and subsequently declines as more plants retire. (NGC 1995) Figure 2 illustrates the "hold-back" dynamics of capacity in the UK electricity market. In the US, electric utility analysts, competing in a competitive market simulation game (CIGMOD) routinely built CCGTs based on economic considerations, raising already high reserve margins.

Natural Gas Market Consolidation
IPP's in the UK see the gas spot market "as an attractive option to burning contracted gas for electricity generation, thus effectively raising the expected profits of an IPP with a gas
Total available capacities

[GRAPHIC OMITTED]

1991 1992

contract, as he can choose whether selling electricity or gas is more profitable." (Newbery 1995a, p. 15) While TransAlta in Canada must decide between building generation plant in the US and transporting gas or building plant in Canada and transporting electricity to the US, US utilities are busy deciding which gas companies would make good partners (such as Texas Utilities and Houston Light & Power mergers with large gas utilities.) Additional perspective of the US potential of multi-fuel arbitrage is provided by Vu and Denard. (Vui 1996) They show that the current "gaming" conditions in the NW are an "arbitrageurs dream." The price of gas and electricity fluctuate by factors of 6 as seen in Figr 3.

DAILY PRICE OF ELECTRC#TY AND GAS

[GRAPHIC OMITTED]

Figure 3: Arbitrage of Gas and Electricfy in the US Northwest (Vu 1996)

Phase 2- Massive Market Restructuring Within the context of partial deregulation, a number of problems take oot in the definition of who has what rights. Reallocating these ights then undermines the justification of previous positions. In speaking about the early days of the UK deregulation, Hunt notes that: "Despite its many virtues, there are many reasons to believe that the generally wholesale competition will be merely a way station and testing ground for fuldl retail competition. Probably the main reason is that both in the United Kingdom and in the United States, when authorities have tried to grant some types of customers open access to sellers while excluding others, the problems of definition have become acute." (Hunt 1996a, p. 22)

TIhe power of the deregulation dynamics indeed shows itself in US movement toward retail wheeling. One description of the process: "Deregulation in one state puts pressure on its neighbors to follow suit, often from major industrial end-users concerned about losing a competitive edge through having to pay higher prices." (Trader 1996, p.8)

Phase 3 - System Divestiture Once deregulation achieves retail wheeling, conflicts among generation, marketing, transmission and distribution widen. Generation struggles in the market place every moment of every day balancing high risks and high potential gains. It has strong cash flow requirements and must behave in a competitive fashion by targeting markets and restricting access to information. Because transmission is still regulated, it has "an obligation to serve with minimal chance to earn significant returns. If transmission expansion is required, the assets of the generation would be needed to s ecure investment finds and support the project putting further economic pressure on the generator. This conflict of interest requires the two to severe all ties.

In addition, the distribution portion of the utility experiences regulatory pressures to minimize cost. If marketing is part of distribution, it must find low cost suppliers to maximize its market shares. The associated generator would then need to negotiate contracts under competitive bids with other generators. Further, the distribution company may find the generation supply the transmission topology allows is more expensive than that from a CCGT plant it could build nearby on its own. The pressures to divest become acute.

As mentioned before, the creation of an ISO critical. The abundance of operations problems and conflict-of-interest transactions find generic solutions only within the confines of a unbiased third party such as an ISO. But the market-forces creation of the ISO also adds pressure to separate distribution from transmission.

In the US, the temptation to game the transmission system is strong. The multiple connection points of unequal capacity can generate loop flows on a line with limited capability. Most NERC reliability regions require line load relief even if only the contingency is threatened. Generation representing as little as 5% of the line flow will be forced to reduce power or come off-line. In general, 100MW of required line load relief typically causes 400 to 600 MW to come off-line. (Hogan 1995, p. 35) The gaming of a small amount of generation in one area can have major impacts in another; there is no static competitive circle or a local generation market in the US One has only to remember July 2nd and August 12th 1996, when huge power outages occurred on the robustly designed Pacific Intertie transmission lines. This massive loss of power in the Northwest indicates the far reaching impacts of local phenomena.

With a leverage as high as 7 to 1, the monetary value of such transactions becomes hard to resist. Excessive operational rules to prevent such possibilities would limit flexibility and, therefore, cause added system reliability and stability problems. New legal methods just create new pressures (loop holes) at the limits of the new law.

Phase 4 - Market Gaming The'physical constraints on the system as well as payment rules, cause market distortions that can be gamed. Additional rules designed to avoid excessive profit taking are merely new distortions causing new, albeit different, gaming tactics. This gaming option makes optimal, equilibrium-pricing an outmoded concept because any gaming causes a loss from the competitor's optimal plan to benefit the game player. Volatile pricing and rewards for "surprise" tactics constitute the preferred market condition.

Gaming Generation Plant Availability
With variable costs dominating the spot market, capacity costs need to be included elsewhere in the price to make generation financially viable. In the UK, capacity payments are made to all available plants running or not. The capacity charge (LOLP) is a strong function of available capacity versus expected demand. By making plants "unavailable," generators can dramatically affect the value of this capacity charge. The capacity charge plus the System Marginal Price (SMP) determines the pool purchase price (PPP). Adding the average cost of transmission, including losses, plus payments to constrained-off generation, gives the pool selling price (PSP) that RECs would experience. Newbery shows the

relationship between peak reserve margin versus LOLP/SMP and indicates the dramatic exponential growth needed in capacity charge to "clear the market" for capacity. A 20% variance in reserve margin produces a thousand-fold increase in capacity charges! This dramatic price response can cause generators to remove plants as the price increases, knowing that remaining plants will be worth even more. (Newbery 1995, p. 50)

Early into the UK deregulation, the generator PowerGen, learned to make plants unavailable for maintenance and then make them suddenly available when its own unavailability had caused the capacity charges to rise substantially. Because of this practice, new rules require a plant to be out seven days before its absence affects capacity charges. With new rules, come new games. Now a plant that goes off-line because of failure may be better off to wait until its outage has driven up the capacity charge before coming back online. (Newbery 1995, p. 57) Further, when a significant plant does become unavailable for cause it may signal other plants to go off-line to increase the capacity charges. In fact when the LOLP is almost unity, the reward for making capacity available may be zero or negative because of its impacts on reducing capacity charges.

Gaming Transmission Constraints
Generators remain successful at "gaming the transmission constraints to increase profits (via increased uplift payments).." (Newbery 1996, p. 63) The UK is winter peaking and had abundant transmission capacity prior to deregulation. Generators appear to have learn how to improve their ability to produce constraints even in shoulder periods as shown in Figure 4. Note also that both low and high voltage lines undergo constraint activities.

Cost of transmission constraints

[GRAPHIC OMITTED]

Figure 4: Learning to Constrain (OFFER 1992)

Because of local transmission constraints, plants that would minimize overall system generating costs may have to be "constrained-off" and plants on the other side of the

constraint "constrained-on." The constrained-off plant obtains the revenue equal to the difference between the SMP and the bid price. Thus, if the plant expects to be constrained- off, it will bid in a low price. (Helm 1995, p. 5) Conversely, a plant constrained-on receives its bid price. (If it had bid lower, it would be already running and not constrained-on.) If a generator expects to be constrained-on, it naturally bids higher than it would otherwise. In 1991, a 95 MW load-following plant received L60M for playing the "constrained-on" game. Although its normal bid was C25/MWhr, its bid for this period was Ll20MWhr. It also increased its start-up and no-load charges plus increased its minimum stable load requirement to 95MW.

Constrained-on and constrained-off charges are added to the price of electricity as uplift. Note that by this definition, "all transmission constraints therefore contribute to uplift [and price]." (Hunt 1996, p.175) If a generator bids low to come on-line in a way that creates a transmission constraint that forces another plant to be constrained-off, the uplift prevents a legal issue by rewarding both parties. Figure 5 shows how dramatic of an effect these constraints (noted as operational outrun) can be. The tallest peaks are due to declaring an unscheduled unavailability and forcing several small-capacity increasing-cost units to replace the generation from a large low-cost unit.

Figmre 5: Components of Uplift Dynamics (POOL 1996)

[GRAPHIC OMITTED]

Figure 6 shows the "ups and downs" of uplift as generators utilize their capacity options. "Many of the RECs argue that the generators are attempting to recover the cost for constrained plant twice: once through Vesting CfDs and again via uplift revenues." (OFFER 1992a, p. 94)

Average daily Uplift payments April - September, 1992

[GRAPHIC OMITTED]

01-Apr 15Apr 29-Apr 3-May 7-MAy a-J 24-Jun 08-Jul fl-Jul 05-Aug 19-Aug 02-Sep 6SEp 30-Sep U Other components [l Operational Outturn

FRpgure 6: Uplift Volatility Under Non-Peak Conditions (OFFER 1992)

Blocking Entry
Free market entry is supposed to discipline competitive markets by removing incentives to raise price. One way to limit entry is to add price volatility to the market and make IPP entry riskier and more expensive. Although price volatility is certainly present, in the long term a stable average price may appear that provides a solid price signal fordecisions..j-owever there is a second entry limiting tactic: bringing higher cost, mid-merit plants into play. Baseload is bid low to make the average price of baseload power below the entry price. High-cost mid-merit plants are then bid in above cost to compensate because generators know that few if any IPP would risk entering that market. The added profits in the mid- merit market then defray any losses in the baseload market.

Examples of gaming can already be found in the US Although the actual prices used in the example below are only approximate, both involved parties have verified the phenomena happened. Suppose BPA sells energy to a municipal or large industrial customer for say 25 mills/kWh. Now suppose PacifiCorp goes to the customer with a promise to supply energy at 22 mills. The customer is aware that PacifiCorp may not have this power, but PacifiCorp insures that the deal would be honored. Abruptly, BPA loses possibly a few hundred MW of load. The waters still run, so BPA puts the excess hydro power on the spot/economy market for say 18 mills. Any guess on who buys the power for 18 mills and sells it for 22 mills?

Phase 5 - Reregulation

The gaming of phase 4 quickly separates the weak from the strong. The weak demand "fairness" through regulatory changes. The inevitable regulatory response forces the strong to act collusively, which further excludes the weak from market participation.

Although electric prices fell in the early part of the UK deregulation, they fell at a rate lower than the decline in fuel costs. Regulatory review led to price caps. These in turn forced collusion among generators to simulate compliance while maintaining profit margins and keeping new entrants out.

Phase 6 - Industry Consolidation

Once the market starts to stabilize, participants will attempt to lock-in advantageous situations. Generators will find themselves regularly selling to the same distribution companies. The increased profits from further reducing uncertainty force reintegration creating economies of scope. The impetus for re-regulation implies that many participants have reached the "end" of their gaming days, therefore, sweep-up acquisitions surge. A handful of national vertically integrated utilities form accompanied by small market niche utilities. Transmission assets lose market value and a single national transmission company, probably with quasi-public ownership takes shape. The storm of deregulation has finally subsided.

In the UK, over 800 miles of transmission line is needed to connect the north to the south. In the US, BPA can and has transmitted power across the continental US to TVA. These two facts make any argument that a small group of trans-national utilities could not compete in the national market indefensible.

The reduction in coipanies does not signal the end of competition. In the UK where there are two large conventional generators, a nuclear utility, (soon) only one Scottish utility and minor IPP players, the analysis of the market indicates functioning competitive conditions.
(Newbery 1994)

Analyses with CIGMOD indicate that existing large US utilities would not necessarily be the survivors after consolidation. IPPs could easily win the game because accumulating large cash reserves is a winning strategy. Thus, companies such as Enron and banks involved with mergers and acquisitions appear more likely to succeed than US utilities.

Conclusion
The dynamics of deregulation appear inevitable. The final market, although bearing little resemblance to regulated utility markets, appears indistinguishable from many other commodity markets. The transition from a regulated to a deregulated market is one where the former relationships of generation, demand, markets, delivery, and pricing become burdens, yet slow to change because of the constraints of past long-lived investments. For a period of time the system is out of balance with the new forces of supply and demand, however, at the end of the transition, the "equilibrium" condition that dominates mature commodity markets comes to pass. The system is still dynamic, but those dynamics are no longer its defining characteristic. Critical to US utility thinking is the rather poignant evidence that indicates this tumultuous time period spans only five to seven years. Given

the changes that occurred between the Mega-NOPR and the Final Open Access Ruling, the US transition appears to be running right on schedule.

References

Comnes 1996: Comnes, Alan, et. al., The Performance of the U.S. Market for Independent Electricity Generation, The Energy Journal, Volume 17, No. 3, pp. 23-39.

CONGRESS 1996: Electric Consumers' Power to Choose Act of 1996, HR 3790 in 104th Congress, 2nd Session; July 11, 1996.

EPACT 1992: Energy Policy Act, Pub. L No. 102-486, 106 Stat. 2776, Title VII (1992).

FERC 1995: Federal Energy Regulatory Commission, Promoting Wholesale Competition Through Open Access Non-discriminatory Transmission Services by Public Utilities, Docket No. RM95-8-000 and Recovery of Stranded Costs by Public Utilities and Transmitting Utilities, Docket No. RM94-7-001, Notice Of Proposed Rulemaking And Supplemental Notice Of Proposed Rulemaking, March 29, 1995.

FERC 1996: Federal Energy Regulatory Commission, Promoting Wholesale Competition Through Open Access Non-discriminatory Transmission Services by Public Utilities, Docket No. RM95-8-000 and Recovery of Stranded Costs by Public Utilities and Transmitting Utilities, Docket No. RM94-7-001, Notice Of Proposed Rulemaking And Supplemental Notice Of Proposed Rulemaking, April 24,1996.

FERC 1996a: Open Access Same-Time Information System and Standards of Conduct, Docket RM 95-9-000. 18 CFR Part 37 April 24, 1996 Open Access Same-Time Information System and Standards of Conduct.

FERC 1996b: Capacity Reservation Open-Access Transmission Tariffs, Docket RM96-11- 000, 18 CFR Part 35, April 24, 1996.

Green 1992: R. Green, Contracts and the Pool the British Electricity Market, Department of Applied Economics, Cambridge University, August 1992 "Long term Contracts and Imperfectly Competitive Spot markets: A study of the UK Electricity Industry, Department of Economic Memorandum 15/1992, University of Oslo, 1992.

Helm 1995: Dieter Helm, et. al., "Competition versus Regulation in British Electricity Generation," in The UK Experience: A Model or a Warning, BIEE/Warwick University Energy Economics Conference, Warwick University, 11-12 Dec 1995.

Hogan 1995: William W. Hogan, "Electric Transmission and Emerging Competition," Public Utilities Fortnight/y, July 1, 1995.

Hunt 1996: Sally Hunt and Graham Shuttleworth, Competition and Choice in Electricity, John Wiley and Sons, 1996.

Hunt 1996a: Unlocking the Grid, Sally Hunt and Graham Shuttleworth, IEEE Spectrum, July 1996, pp. 20-25.

Newbery 1992: R. J. Green and D. M. Newbery, "Competition in the British Electricity Spot Market," Journal of PoliticalEconomy, Volume 100, Number 5, 1992, pp. 929-953.

Newbery 1994: David M. Newbery and Michael G. Pollitt, The Restructuring and Privati-ation of the CEGB- Was it Worth it?, Department of Applied Economics, University of Cambridge, 1994.

Newbery 1995: David M. Newbery, 'Tower Markets and Market Power," The Energy Journal, International Association of Energy Economists, Volume 16, Number 3, 1995, pp. 39-66.

Newbery 1995a: "Restructuring the UK Energy Industries: What have we learned?," David M Newbery. The UK Energy Experience: A Model or Warning, BIEE/Warwick University Energy Economics Conference, Warwick University, UK, 11-12 December, 1995.

Newbery 1996: David M. Newbery, "Regulation, Public Ownership and Privatisation of the English Electricity Industry" in International Comparison of Electriciy Regulation, Richard Gilbert, and Edward Kahn, ed., Cambridge University Press, 1996.

NGC 1995: National Grid Company plc, Seven Year Statement, London, March 1995.

OFFER 1991: Pool Price Inquiry, Office of Electricity Regulation, Birmingham, 1991.

OFFER 1992: Review of Pool Prices, Office of Electricity Regulation, London, December 1992.

OFFER 1992a: Report on Constrained-On Plant October 1992, Office of Electric Regulation, London.

Pool 1996: The Electricity Pool of England and Wales, Statistical Digest, London, Issue No. 67, March 1996 and earlier.

Pool Rules 1995: A Users' Guide to the Pool Rules. Issue 2.00, The Electric Pool of England an Wales., London, 1995.

PUHCA 1932: Public Utility Holding Company Act 15 U.S.C. %Section 79 et seq.

PURPA 1978: Pub. L. No. 95-617,92 Stat. 3117 (codified in U.S.C. sections 15,16,26, 30, 42, and 43).

Tabors 1996: "Lessons from the UK and Norway", Richard D. Tabors, IEEE Spectrum, New York, pp. 45-49.

Trader 1996: Energy Trader, August 9, 1996, Petroleum Argus, London.

Vu 1996: Mia T Vu and Darlene Denard, The Emerging World of Multifuel Marketing, USAEE Dialogue, Volume 4, Number 1, April 1996, pp. 2-4.

Advance Work Prior to Meeting

ECT to provide Perot with brief list of definitions of trading terms (e.g. "position", long, short, arbitrage, preschedule, etc.

Perot to provide list of definition of ISO/PX tariff terms and references describing rules that lead to market opportunities. References could be ISO or PX tariffs or other documents/procedures provided by ISO and PX.

Perot to provide initial list/description of 5-10 candidate games to be reviewed in seminar.

ECT to review games for potential implementation issues and identify games to be studied in detail during seminar.

8:00 - 9:00am  Introductions
                Overview of terms

9:00 - 12:00am  Overview of gaming opportunities
                Major types of games
                Resource characteristics to implement gaming strategies
                (flexible generation, load, etc.)

1:00 - 4:00pm  Overview of tariff and protocol issues
                Detailed discussion of mismatches in tariffs and computer
                software Review of specific examples of strategies to exploit
                mismatches

4:00 - 5:00pm  Review of day
                Refine Day Two Agenda

8:00 - 12:00am  Review top gaming strategies
                Work through details of implementation strategies

1:00 - 3:00am  Discuss resources needed to implement gaming strategies
                Identify weaknesses in Enron position/capabilities Discuss next
                steps

0

        po~lte                  Policy Assessment Corporation
        I0LJC_'               14604 West 62nd Place
~____         __ ___            Denver, Colorado 80004
        _                    Office: (303) 467-3566 Fax: (303) 467-3576
                                email: gbackus~boulder.earthnet.net
                                Energy, Environmental and Economic Planning

FAX TRANSMITTAL SHEET

To: Rich Davis Telephone:

From: George Backus Reference: The Microsoft Curse

Date Sent: Jun 1, 002

Number of Total Pages: 6

Rich,

Sorry for the delay. I could not actually edit the file that you see on Word 7. (It looks fine in WORD 97.) Hence, I am faxing this to you. There really is the PSC cost problem of Paul Gribik having such a high value right now. He is the only person I know who brilliantly understands both the "gaming" issues and understands the details of the ISO/PX. I have searched years, close to world-wide, to find people like him. I can't even find anyone else who understands the gaming process. (Although,.your Tom Delaney can understand the concepts pretty well.) I know you find it hard to believe, but if your mind can handle what Paul throws at you, he is well worth the money on this topic.) Thus, we do need to be creative to make this happen. I do believe that the project would make money for you but I also understand that you have KNOW that it will before moving on to the much more significant $ of the project. Some possibilities come to mind; maybe you can think of others. Short of a seminar with Paul, would any written materials provide you confidence, for example, a document that went through a detailed "ganlp" scenario? I would think that something may have been written up for the ISO so that it could provide data to FERC. Or you, your Paul, and Tim (sorry but I forgot the

woman's name (Valerie?) and I could find a CA ISO/PX "problem" of interest. I would take a shot at it in the language you need. We would then get Paul Gribik on a conference call to go over details where you needed more than my generic discussions. Third, we (my group) could provide a CIGMOD game for WSCC with your staff -- although that is already $50K for two days and still $20K if it is for a generic region. (I might be able to get you and some of your staff invited to the next WSCC CIGMOD 'game at BPA! Would that work?) Lastly, and here I am unclear of a specific proposal, can we think of a way to limit your first cost for the initial seminar with Paul Gribik, but compensate by a follow-on commitment (based on criteria) the recovers all the cost for Perot at a later (although relatively soon) date?

I am sympathetic with you that this is not a good foot to start off a business relationship, but (from being in the middle of all this) I still see Enron and Perot as a powerful and effective team to maximize Enron market value. Perot has been very professional, competent, and "human" with me and I would expect the same from them if they worked with you. I hope we can find a means to get this back on track.

Please contact me if you have any ideas or would like me to do some foot work or "check out a possibility." Even a strong generic statement to me from you of what you need from Perot and me to move forward, would help me grab the ball and run with it for you. (Remember, I am great at generic and can work without details! :-)

Thanks

George

DRAFT
(February 16, 1998)

PROPOSAL TO ENRON
Prepared by
Perot Systems Corporation (PSC)
In partnership with
Policy Assessment Corporation (PAC)

Obiective:
Deregulation of the electric & gas industries combihed with opening of the power markets to competition have created an opportunity for skilled market participants to optimize their bidding and asset positions for profit maximization. The California market is in the early days of these changes.

Enron Capital & Trade Corporation have expressed an interest in exploring and developing with PSC and PAC appropriate strategies and tools for seizing these market opportunities.

Approach
It is proposed that Enron's optimization strategies be developed through a multi-phased effort. This approach offers maximum opportunity to explore concepts; refine the market scenarios; test the validity of supporting systems & tools; and, to develop related performance metrics. Decisions to progress from phase to phase will be determined on a valuation of benefits achieved and an estimation of expected future return on investments. Each successive phase is designed to build upon the other and, independently produce value to Enron. This approach limits Enron's financial risk while affording a foundation for refining design requirements as work is progressed.

Phase 0: Initial Seminar
Phase 0 is designed to test the concepts employed in developing optimization strategies. The framework is a 1 1/2 day seminar used for identifying "gaps" in the market protocols; designing scenarios and optimization strategies; and, then framing these scenarios for testing within Enron's trading environment.

Employees of Enron, PSC and PAC will use specific examples of gaps in California market protocols seen to offer opportunities for market optimization. This joint team effort led by Paul Gribik, Hemant Lall and Ed Smith of PSC, and George Backus of PAC, will closely examine 3 concrete California market examples previously verified by PSC with the California Power Exchange/Independent Service Operator as valid illustrations of market "gaps".

The seminar will design detailed strategies and plans around these "gaps" to reach agreement on the range of market opportunities they pose; develop appropriate strategies for optimization and, prove the reality and soundness of such tactics. The approach will also illustrate how existing rules can be re-evaluated to produce new market

Confidential: Covered by Non-disclosure @ Perot Systems Corporation 1998

.. . . .

opportunities. One of the on-going benefits to be derived by Enron's staff is education in the process used for examining these situations

Enron employees will share insights on existing capabilities and asset portfolios for the purpose of plotting positions against which these scenarios will be tested. Testing will also include analysis of any trading floor procedural constraints that may be seen to exist. . The success of this phase in validating the existence of "gaps" against which optimization strategies can be mapped, as well as the realistic ability to act on such opportunities, will form the basis for moving to Phase 1.

Phase 1
Phase 1 will be split into two parts.

Phase la greatly expands on Phase 0 by conducting a detailed review of the PX and ISO business and operational protocols for the purpose of cataloging the perceived "gaps" within and between these two entities. These findings will then be prioritized in terms of their relevant impact and, once compared with Enron's capabilities and trading portfolio, corresponding electric and gas tactics/strategies will be designed/developed. The range of game/tactic categories will be extensive enough to consider games associated with generation, trading, transmission, and customers. Moves that are made independently or in combination with other market players (across function and across companies) will also be considered. This consideration is essential, as moves limited to trading only may not take adequate advantage of market opportunities. Combined moves, across market functions, provide more flexibility and higher pay-offs. Countermoves will also be addressed to examine plans for protecting Enron from competitor actions.

Enron staff will decide on areas where Enron
* would like to focus on immediate trading-function efforts,
* would like to focus on building up alliances for combined functional or non- Enron company efforts, and
* does not want to become involved.

Dependent upon the level of interest expressed, Phase la can also include interactive team "war" efforts using PAC's CIGMOD software simulation models to formally test and review the generic gaming options within the simulated frame work of the California market place and Enron environment.

Because the rules and the market are evolving, this phase will continue to see changes. However, the development of an exhaustive list of categories, along with a selection of those key areas of most interest to Enron, will constitute the completion and success of this phase.

Phase lb takes the prioritized list of opportunities converting them into detailed concrete rmoves. Each tactic is formally defined in terms of specific plant, transmission line, load, and system conditions for both gas and electricity. The formal definition not only allows

Confidential: Covered by Non-disclosure @ Perot Systems Corporation 1998

for the actual execution of the strategies but also provides the analytical formalization to operationally use the strategies in Phase 2. Market and operational information input requirements will also be identified and sources for obtaining this information will be investigated.

Phase 2
Phase 2 will also be split into two parts.

This phase will convert the detailed and formalized listing of strategies into computerized simulations of game moves in the real world. This phase will have two sub-phases.

Phase 2a involves the creation of "study mode" computer models simulating the California PX and ISO systems. Actual protocols and rules are carefully detailed. Simulations will include the PX's energy auction model, the ISO congestion management model and, define their interaction. Transmission simulation will allow accurate constraint analysis. The HYPERSENS software is activated as an umbrella overarching the energy-system simulation to capture the uncertainty and impacts of competitor counter moves. Data for each company and plant in the region will be incorporated in the model to be available for determining bidding strategies. It can also determine those moves seen to bring Enron the most advantage under noted operating conditions and a "pay-off matrix" examining values for the moves, or portfolio of moves.

Phase 2b develops "real time" optimization versions of the "study mode" models to test the system against real world, real-time data. Real PX, ISO and trading floor, OASIS, and SC data will be used in the model to compare modeled to actual positions. The value of transactions/positions from the simulation will also be compared to those that actually occurred. Performance benchmarks for evaluating future transactions will be established. This effort validates the model capabilities to test or recommend moves in real time. Most importantly, the list of strategies from Phase lb can be tested for value and potency within the real marketplace. The resolution of the model would minimally be hour-by- hour. Required interfaces for inputs of market and operational data will either be established, or alternative arrangements made at this point.

As part of Phase 3, these models will then be fully integrated with Enron's computer infrastructure and linked to real-time data systems. This linkage of real time data is-a task in and of itself. Data collection, transfer, reconciliation, timing and use-must be operationalized, coordinated and verified.

Phase 3
This phase will integrate the tools and knowledge developed during the previous phases into Enron's normal trading processes, systems and, operations. It is expected that the early use of these tools with real-time data will find areas where the system must be tuned or modified to capture unforeseen opportunities, cope with human or data-stream errors, recognize divergence of idealized from real operations, and prevent inappropriate responses in ambiguous conditions. The "fine tuning" of the system and the modification

Confidential: Covered by Non-disclosure @ Perot Systems Corporation 1998

of the system and its user interface to maximize profits/usability is a critical focus. Training of Enron staff will be integral to this transition. This phase will include a benchmark sub-system to compare the system-recommended performance to a "reference" baseline performance.
The changing market rules and data streams will require continued maintenance and refinement of the system. This ongoing maintenance, consulting and technical/software support could be considered as Phase 4 or as a separate project.

The design of the project allows for future extension of the system (for example, to include all of WSCC in detail) or cloned (for example, to maximize Enron profits in the UK, Brazil, US East Coast or US Midwest'markets). Efforts along these lines are considered outside the scope of this project. (Note that the system possesses the inherent capabilities to simulate the financial and strategic impacts of mergers, acquisitions, purchasing of facilities/generation and the building of generation, gas-storage or transmission.)

Professional Fees
The PSC/PAC professional fees for Phase O's "Proof-of-Concept" Seminar will be $40,000 plus all reasonable expenses. PSC will be the prime contractor to Enron, and PAC will sub-contract to PSC. It is proposed that Phase 0 be held on Feb 28 and March 1, 1998 at a location of Enron's choice.

Estimates and an approach to calculation of professional fees for future Phases will be provided after conclusion of Phase 0.

Other mutually agreeable contractual or partnering relationships are not precluded and could become part of this overall proposal.

Confidential: Covered by Non-disclosure @ Perot Systems Corporation 1998

DRAFT
(Febmary 16, 1998)

PROPOSAL TO ENRON
Prepared by
Perot Systems Corporation (PSC)
in partnership with
Policy Assessment Corporation (PAC)

tive-
Deregulation of the electric & gas industries combined with opening of the power markets to competition have created an opportunity for skilled market participants to optimize their bidding and asset positions for profit maximization. The California market is in the early days of these changes.

Enron Capital & Trade Corporation have expressed an interest in exploring and developing with PSC and PAC appropriate strategies and tools for seizing these market opportunities.

Aproach

It is proposed that Enron's optimization strategies be developed through a multi-plased effort. This approach offers maximum opportunity to explore concepts; refine the market scenarios; test the validity of supporting systems & tools; and, to develop related performance metrics. Decisions to progress from phase to phase, will be determined on a valuation of benefits achieved and an estimation of expected future return on investments. Each successive phase is designed to build upon the other and, independently produce value to Enron. This approach limits Enron's financial risk while affording a foundation for refining design requirements as work is progressed.

Phase 0: Initial Seminar

Phase 0 is designed to test the concepts employed in developing optimization strategies. The framework is a 1 1/2 day seminar used for identifying "gaps" in the market protocols; designing scenarios and optimization strategies; and, then framing these scenarios for testing within Enron's trading environment.

Employees of Enron, PSC and PAC will use specific examples of gaps in California market protocols seen to offer opportunities for market optimization. This joint team effort led by Paul Gribik, Hemant Lall and Ed Smith of PSC, and George Backus of PAC, will closely examine 3 concrete California market examples previously verified by PSC with the California Power Exchange/independent Service Operator as valid illustrations of market "gaps".

The seminar will design detailed strategies and plans around these "gaps" to reach agreement on the range of market opportunities they pose, develop appropriate strategies for optimization and, prove the reality and soundness of such tactics. The approach will

Confidential: Covered by Non-disclosure @ Perot Systems Corporation 1998

also illustrate how existing rules can be re-evaluated to produce new market opportunities. One of the on-going benefits to be derived by Enron's staff is education in the process used for examining these situations

Enron employees will share insights on existing capabilities and asset portfolios for the purpose of plotting positions against which these scenarios will be tested. Testing will also include analysis of any trading floor procedural constraints which may be seen to exist. . The success of this phase in validating the existence of "gaps" against which optimization strategies can be mapped, as well as the realistic ability to act on such opportunities, will form the basis for moving to Phase 1.

Phase 1
Phase 1 will be split into two parts.

Phase la greatly expands on Phase 0 by conducting a detailed review of the PX and ISO business and operational protocols for the purpose of cataloging the perceived "gaps" within and between these two entities. These findings will then be prioritized in terms of their relevant impact and, once compared with Enron's capabilities and trading portfolio, corresponding electric and gas tactics/strategies will be designed/developed. The range of game/tactic categories will be extensive enough to consider games associated with generation, trading, transmission, and customers. Moves that are made independently or in combination with other market players (across function and across companies) will also be considered. This consideration is essential as moves limited to trading only may not take adequate advantage of market opportunities. Combined moves, across market functions, provide more flexibility and higher pay-offs. Countermoves will also be addressed to examine plans for protecting Enron from competitor actions.

Enron staff will decide on areas where Enron

* would like to focus on immediate trading-function efforts,
* would like to focus on building up alliances for combined functional or non-Enron company efforts, and
* does not want to become involved.

Dependent upon the level of interest expressed, Phase la can also include interactive team "war" efforts using PAC's CIGMOD software simulation models to formally test and review the generic gaming options within the simulated frame work of the California market place and Enron environment.

Because the rules and the market are evolving, this phase will continue to see changes. However, the development of an exhaustive list of categories, along with a selection of those key areas of most interest to Enron, will constitute the completion and success of this phase.

'I.

Confidential: Covered by Non-disclosure @ Perot Systems Corporation 1998

Phase lb takes the prioritized list of opportunities converting them into detailed concrete moves. Each tactic is formally defined in terms of specific plant, transmission line, load, and system conditions for both gas and electricity. The formal definition not only allows for the actual execution of the strategies but provides the analytical formalization to operationally use the strategies in Phase 2. Market and operational information input requirements will also be identified and sources for obtaining this information will be investigated.

Phase 2
Phase 2 will also be split into two parts.

This phase will convert the detailed and formalized listing of strategies into computerized simulations of game moves in the real world. This phase will have two sub-phases.

Phase 2a involves the creation of "study mode" computer models simulating the California PX and ISO systems. Actual protocols and rules are carefully detailed. Simulations will include the PX's energy auction model, the ISO congestion management model and, define their interaction. Transmission simulation will allow accurate constraint analysis. The HYPERSENS software is activated as an umbrella overarching the energy-system simulation to capture the uncertainty and impacts of competitor counter moves. Data for each company and plant in the region will be incorporated in the model to be available for determining bidding strategies. It can also determine those moves seen to bring Enron the most advantage under noted operating conditions and a "pay-off matrix" examining values for the moves, or portfolio of moves.

Phase 2b develops "real time" optimization versions of the "study mode" models to test the system against real world, real-time data. Real PX, ISO and trading floor, OASIS, and SC data will be used in the model to compare modeled to actual positions. The value of transactions/positions from the simulation will also be compared to those that actually occurred. Performance benchmarks for evaluating future transactions will be established. This effort validates the model capabilities to test or recommend moves in real time. Most importantly, the list of strategies from Phase lb can be tested for value and potency within the real marketplace. The resolution of the model would minimally be hour-by- hour. Required interfaces for inputs of market and operational data will either be established, or alternative arrangements made at this point.

As part of Phase 3, these models will then be fully integrated with Enron's computer infrastructure and linked to real-time data systems. This linkage of real time data is a task in and of itself. Data collection, transfer, reconciliation, timing and use must be operationalized, coordinated and verified.

Phase 3
This phase will integrate the tools and knowledge developed during the previous phases ihto Enron's normal trading processes, systems and, operations. It is expected that the early use of these tools with real-time data will find areas where the system must be tuned

Confidential: Covered by Non-disclosure @ Perot Systems Corporation 1998

or modified to capture unforeseen opportunities, cope with human or data-stream errors, recognize divergence of idealized from real operations, and prevent inappropriate responses in ambiguous conditions. The "fine tuning" of the system and the modification of the system and its user interface to maximize profits/usability is a critical focus. Training of Enron staff will be integral to this transition. This phase will include a benchmark sub-system to compare the system-recommended performance to a "reference" baseline performance. The changing market rules and data streams will require continued maintenance and refinement of the system. This ongoing maintenance, consulting and technical/ software support could be considered as Phase 4 or as a separate project.

The design of the project allows for future extension of the system (for example, to include all of WSCC in detail) or cloned (for example, to maximize Enron profits in the UK, Brazil, US East Coast or US Midwest markets). Efforts along these lines are considered outside the scope of this project. (Note that the system possesses the inherent capabilities to simulate the financial and strategic impacts of mergers, acquisitions, purchasing of facilities/generation and the building of generation, gas-storage or transmission.)

Professional Fees

The PSC/PAC professional fees for Phase 0's "Proof-of-Concept" Seminar will be $40,000 plus all reasonable expenses. PSC will be the prime contractor to Enron, and PAC will sub-contract to PSC. It is proposed that Phase 0 be held on Feb 28 and March 1, 1998 at a location of Enron's choice.

Estimates and an approach to calculation of professional fees for future Phases will be provided after conclusion of Phase 0.

Other mutually agreeable contractual or partnering relationships are not precluded and could become part of this overall proposal.

Confidential: Covered by Non-disclosure @ Perot Systems Corporation 1998

PROPOSAL TO ENRON
Prepared by
Perot Systems Corporation (PSC)
in partnership with
Policy Assessment Corporation (PAC)

Approach

To ensure Enron comfort with and understanding of,-all aspects of the proposed effort, it is mutually agreed that this initiativee-effei4 should be pursued inaes a series of phases. Each proposed phase builds upon the previous phase and independently produces value to Enron. The phasing approach limits Enron's financial commitment tofe* each phase of the project, guarantees that a concrete deliverable is provided at each step, and allows for project adjustment in future phases as need dictates.

Phase 0

An initial phase, called Phase 0, will provide a 2 day seminar lead by Paul Gribik-(PSC),- George Baekus (PAC) and Hemant Laln Ed S of PSC and George Backus of PAC. This single task effort would closely examine 3 concrete examples (and more by references) with to examine the detailed steps of the strategy/tactic, prove the reality and soundness of such tactics, and most importantly, illustrate how existing rules can be re- evaluated to produce new market opportunities. This latter aspect will allow Enron staff to apply the "methods" of "gaming" to recognize and execute tactics made available by the ever evolving market rules associated with the dynamics of deregulation. To ensure realism and validity, the examples used would be those actually evaluated within the PX/ISO because of their potential or ready realization in the marketplace.

As part of Phase 0, PSC and PAC. would learn Enron's existing capabilities and operations to understand immediate areas where Enron has advantages, disadvantageous, opportunities, and limitation. The success of this phase in verifying the existence of the claimed moves. as well as the realistic ability to act on such opportunities. will form the basis for moving to Phase la.

Phase 1

Phase 1 will be split into two parts. Phase la, greatly expands on Phase 0 by exhaustively cataloging the generic electric and gas tactics/strategies with a strong emphasis on those associated with the NW and CA. The range of game/tactic categories would be extensive enough to include those games that are more common to other parts of the country or are common to other countries. This is done because the rules of CA can and will change. The impacts of the new rules on rest of the WSCC area, for example, could easily bring forth conditions that require moves far removed from those currently available in the CA market. Games associated with generation, trading, transmission, and customers will be considered. Moves that are made independently or in combination with other market players (across function and across companies) need to be considered. Moves limited to trading only do not take adequate advantage of market opportunities. Combined moves, across market functions, provide more flexibility and

Confidential: Covered by Non-disclosure Perot Systems Corporation 1998

higher pay-offs. Countermoves must also be addressed to protect Enron from competitor action and to prepare Enron for countermoves to any of its own actions.

A review of these moves with Enron staff would delineate those areas where Enron

* 1) where Efen-would like to focus on immediate trading-function efforts,
* 2) where Effon-would like to focus on building up alliances for combined functional or non-Enron company efforts. and
* wd e s not DOES-NOT-want to become involved. (This last area- would then require Enron to focus on countermoves to keep it out of these games yet protected form them.)

Part of Phase-la can include interactive team efforts (as proposed by Paul Gribik) to uncover -opportunities unique to Enron operations. Another approach is to use CIGMOD to formally test and review the generic gaming options within the simulated frame work of the CA/WSCC market place and Enron environment.

Because the rules and the market are evolvinge, this phase is never totally completed. However, thebutrhe development of an exhaustive list of categories, along with a selection of those key areas of most interest to Enron, will constitute the completion and success of this phase.

Phase lb takes the generic list and converts it into detailed concrete moves. Each tactic is formally defined in terms of specific plant, transmission line, load, and system conditions for both gas and electricity. The formal definition not only Wallows for the actual execution of the strategies but provides the analytical formalization to operationally use the strategies in Phase 2.

Phase 2

This phase -will alee-converts the detailed and formalized (and ever-changing) leflist of strategies into simulations of game moves in the real world. This phase can also have two sub-phases.

The development of the real-time system (using existing tools) constitutes Phase 2a. The simulations here include a an exact or near etrepresentation of the actual PX and ISO software. Actual protocols and rules are carefully detailed. Transmission simulation would allow accurate constraint analysis. NW and WSCC rules/limitations would be included as needed. The HYPERSENS software is activated as an umbrella over the energy-system simulation to capture the uncertainty and impacts of competitor counter moves. It can also determine those Enron moves with the most advantage under noted operating conditions along with the pay-off matrix for the moves, or portfolio of moves.

Phase 2b would test the system against real world, real-time data. Real PX, ISO and trapding floor, OASIS, SC, and WSCC data would be used in the model to compare what

Confidential: Covered by Non-disclosure i). Perot Systems Corporation 1998

it would have proposed to what actually happened. The value of transactions/positions from the simulation can be compared to those that actually occurred. This effort validates the model capabilities to test or recommend moves in real time.. Most importantly, the list of strategies from Phase lb can be tested for value and potency within the real marketplace. The resolution of the model would minimally be hour-by-hour. Most probably 5-minute real-time resolution would be needed for some processes.

Because, Phase 2 is the implementation of the Phase lb, they are inseparable from a value perspective. As such, if Enron should decide at the end of Phase lb to discontinue the proposed project, some of the value associated with Phase 2 would be attributed to Phase lb and duly billed as such. In other words, the intellectual and capital investment of PSC and PAC is shared with Enron at a discount in Phase lb under the understanding that Phase 2 will be pursued. If phase 2 is not pursued, PAC and PSC then would need to recover the potion of "expertise development' costs in Phase b.

As part of Phase 2b or as part of Phase 3, the model can be fully implemented on Enron computers and linked to real-time data systems. This linkage of real time data is a task in and of itself.. Data collection, transfer, reconciliation, timing and use must be operationalized, coordinated and verified.

Phase 3

This phase uses the tools and knowledge of the previous phases for actual Enron operations. The use of the systems developed systems use-must be monitored; Enron staff must become fluent in the systems' uses and PSC/PAC technical support must guarantee that the system maximizes Enron profitability. It is expected almsteeA that the early use of the tool with real-time data will find areas where the system must be tuned or modified to capture unforeseen opportunities, cope with human or data-stream errors, recognize divergence of idealized from real operations, and to prevent inappropriate responses in ambiguous conditions. The "fine tuning" of the system and the modification of the system and its user interface to maximize profits/usability becomes the focus of the phases. The success of this phase is denoted by increasing margins over what would have otherwise occurred. (This phase would include a benchmark sub-system to compare the system-recommended performance to a "reference" baseline performance.)

The changing market rules and data streams will require continued maintenance and refinement of the system. This ongoing maintenances consulting and technical/software support could be considered part of Phase 3 or as a separate project.

The design of the project allows the system to be extended (for example, to include all of WSCC in detail) or cloned (for example, to maximize Enron profits in the UK Brazilian, US East Coast or US Midwest markets). Efforts along these lines could be considered in other phases or other projects. They can be performeds in parallel with the CA-system- pmay or sequentially thereafter. (Note that the system can a-simulate the financial and

Confidential: Covered by Non-disclosure (Ei) Perot Systems Corporation 1998

strategic impacts of mergers, acquisitions, purchasing of facilities/generation and the building of generation, gas-storage or transmission.)

Professional Fees

The PSC/PAC professional fees for Phase 0 will be $20,000 plus all reasonable expenses. PSC will be the prime contractor to Enron, and PAC will sub-contract to PSC. It is proposed that Phase 0 be held on Feb 28 and March 1, 1998 at a location of Enron's choice. Other contractual or partnering relationships with PSC or PAC are not precluded by and could become part of this overall proposal.

The professional fees for Phases 1 &2 will be provided upon conclusion of Phase
0. - Because Phase 2 is implementation of Phase lb. they are inseparable from a value perspective. As such, if Enron should decide at the end of Phase lb to discontinue the proposed project. some of the value associated with Phase 2 would Ibe attributed to Phase lb and duly billed as such. In other words, the intellectual and capital investment of PSC and PAC is shared with Enron at a discount in Phase lb under the understanding that Phase 2 will be pursued. If phase 2 is not pursued, PAC and PSC then would need to recover the portion of "expertise development" costs in Phase lb.

Other mutually agreeable contractual or partnering relationships are not precluded and could become part of this overall proposal.

Confidential: Covered by Non-disclosure (H Perot Systems Corporation 1998

TO:          George Backus

FROM:        Phil Carver

DATE:        Monday,June 17, 200

SUBJECT:     DRAFT Survey on CIGMOD to CREPC

Thanks for your suggestions. Please edit or make suggestion to clarify or make more saleable.

TO:          Committee on Regional Electrice Power Cooperation (CREPC)

FROM:        Phil Carver, OR Dept. of Energy

DATE:        June 17, 2002

SUBJECT:     Survey on workshop using the CIGMOD Model

As discussed at the October CREPC meeting in San Diego, I propose an all day workshop on April 10 on electric restructuring. The workshop would use the CIGMOD model to simulate the impacts of restruturing in a multi-player environment. As a game-theory model, CIGMOD simulates the dynamics of a real market better than optimization or equilibrium models.

If there is sufficient interest from members of the committee, we will ask for the first two hours on the agenda for the CREPC meeting on the 11th of April. During this time the' consultants for Systematic Solutions would help debrief the results of the -modeling effort on the 10th. This would allow committee members who could not attend the workshop to share in the insights from the modeling effort.

CIGMOD also allows participants to gain experience and skills as if one were a player in the restructured electricity market. During the simulation, players build resources or buy power from long term contracts or from the spot market. Players set the terms, conditions and prices of the power they sell. Players can merge with other companies or may be forced to learn how to handle bankruptcies.

As we discussed in San Diego, WIEB would take possession of the CIGMOD model after the simulation. It would be availble for any CREPC member to borrow and use. The workshop and model cost $15,000. Systematic Solutions recommends we limit the number of actual players to 30. If so, the cost would be $500 per player There is no limit to the number of people who can watch. Each player could bring as many observers as he or she wishes. If there are not enough CREPC players, I will try to recruit enough players from Portland area utilities, but the CIGMOD can play with fewer players or have the computer act as a player(s)..

CIGMOD can simulate many possible futures. Below I have listed several choices. If you are interested in attending the workshop (either as a observer or player), please fill out the survey. Indicating you are interested does not committ you to attend. After this initial survey, we will indicate how the model might be configured and ask you if you plan to attend the April 10 workshop. Players will make the final decision on model configuration on April 10.

Please return this survey to Brad Wetstone via FAX by January 18. If you are not interested please indicate that on the first question arid return the survey anyway. This will help us distinguish between people who are not interested and people who forgot to return the survey. If you are not interested, you would not need to answer the other questions. Thanks for your help.

Please return this survey by JANUARY 18 by fax to:
Brad Wetstone at WIEB: FAX (303) 573-9107.

If there are other people in your organization who might be interested in attending the Systematic Solutions simulation workshop on electric restructuring, please copy the letter and survey and distribute.

NAME

PHONE# ( FAX# (
ORGANIZATION

Please check the appropriate blanks.

Are you interested in attending the April 10 workshop? YES NO_

(The costfor players is $500, observers are free)

If you answered "NO" above, you don't need to answer the rest of the survey, but please return the survey so we can get a complete count.

If you are interested, do you think you would attend as a(n)?

la.   Retail wheeling (i.e. direct access)?             YES     NO
lb.    la. with strong brand loyalty
         (i.e. lowprice responsefor retail customers)?  YES_    NO

2.    Transmission pricing by (choose one)
             ZONED POSTAGE STAMP PRICES?                     or
             MW-MILE PRICES?

3.    Protection of shareholders from stranded costs?   YES     NO
4.    Renewable resources priced near fossil-fueled?    YES     NO
5.    Volatile fuel prices?                             YES     NO
6.    Unstable load growth                              YES     NO _

If there are other features of the restructured market that you might lilk in he simulation model, please describe what you would like to see: