[00:00:03]
[Aurora Energy Research Report Workshop on December 17, 2025.]
EVERYONE.THIS IS GORDON DRAKE FROM ERCOT, AND WANTED TO WELCOME YOU TO OUR AURORA ENERGY RESEARCH REPORT WORKSHOP.
JUST WANTED TO BEGIN WITH A FEW HOUSEKEEPING ITEMS. UH, TODAY'S SESSION IS BEING RECORDED VIA THE, THE SWAG PLATFORM, SO WE'LL BE AVAILABLE FOR REVIEW AFTER THE MEETING TODAY.
UM, WE WILL BE MANAGING BOTH THE, THE PRESENTATIONS AS WELL AS THE DISCUSSION QUEUE THROUGH WEBEX.
SO ENCOURAGE YOU IF, IF YOU HAVE A COMMENT OR A QUESTION TO PLEASE, UH, REGISTER THAT IN THE, IN THE CHAT WINDOW, AND WE'LL BE MONITORING THAT AND, UM, MAKING SURE THAT WE'RE KEEPING AN EYE ON BOTH THE DISCUSSION IN THE ROOM AS WELL AS, UM, AS LOOKING ONLINE TO ENSURE THAT EVERYBODY HAS A CHANCE TO ASK THEIR QUESTIONS.
UM, GLAD THAT YOU'RE HERE TO JOIN US TODAY AND WANT TO TURN IT OVER TO KEITH COLLINS, UH, TO KICK US OFF.
THANK YOU GORD, AND THANK YOU EVERYBODY FOR COMING OUT TODAY.
I KNOW IT'S OBVIOUSLY A BUSY WEEK BEFORE THE HOLIDAYS, UH, START NEXT WEEK.
SO, UH, THANK YOU FOR BEING HERE.
AND, UM, WE JUST GOT A COUPLE THINGS THAT WE WANT TO COVER AT THE BEGINNING AND THEN BEFORE WE HAND IT OVER TO OLIVIA AND HIS, HIS TEAM AT AURORA.
UH, SO IF WE COULD GO THE NEXT SLIDE.
SO, LIKE ANY PROPER, UM, UH, ERCOT ENDEAVOR, WE START WITH OUR ANTITRUST ADMONITION.
UM, AGAIN, TO AVOID RAISING CONCERNS ABOUT ANTITRUST, UH, PARTICIPANTS IN ERCOT ACTIVITIES SHOULD REFRAIN FROM PROPOSING ANY ACTION OR MEASURE THAT WOULD EXCEED ERCOT AUTHORITY UNDER FEDERAL OR STATE LAW.
FOR ADDITIONAL INFORMATION, STAKEHOLDERS SHOULD CONSULT STATEMENT OF POSITION ON ANTITRUST ISSUES FOR MEMBERS, ERCOT COMMITTEES, SUBCOMMITTEES WORKING GROUPS, WHICH IS POSTED ON THE ERCOT WEBSITE.
SO, AGAIN, NOT A FORUM FOR, UH, INAPPROPRIATE BEHAVIOR, BUT THANK YOU FOR, UH, FOR BEING HERE.
ALRIGHT, SO JUST A LITTLE CONTEXT HERE.
UH, EARLIER THIS YEAR WE, WE AT ERCOT ENGAGED AURORA, UH, ENERGY RESEARCH.
UH, ULTIMATELY THEY WERE ABLE TO PERFORM THIS ANALYSIS ON, UH, DISPATCHABLE RELIABILITY RESERVE SERVICE DRRS, AND WE ASKED THEM TO ESTIMATE THE IMPACT OF DIFFERENT MARKET DESIGN OPTIONS, UH, INCLUDING THE DIFFERENT OPTIONS AND VARIABILITY, UH, DIFFERENT VARIATIONS OF DISPATCHABLE RELIABILITY RESERVE SERVICE, UH, INCLUDING WHAT WE CALL ANSLEY SERVICE AND ANSLEY SERVICE PLUS, UH, UH, OBVIOUSLY SOME OF THE KEY VARIABLES.
THERE WERE SOME THINGS THAT HAPPENED THIS YEAR, SENATE BILL SIX, UH, WHAT ARE THE EFFECTS ON ON THAT? THEY DID, THEY DID INCLUDE THAT IN THE REPORT.
UH, THE, UH, LARGE LOAD FORECAST THAT WE HAVE IN OUR, UH, LONG TERM LOAD FORECAST, UH, WAS A PART OF IT.
AND, UH, ALSO STUDYING THE IMPACT OF ENERGY STORAGE RE RESOURCES, UH, EITHER ELIGIBLE OR INELIGIBLE IN, IN DIFFERENT OPTIONS AND WHAT THAT MEANS IN TERMS OF OUTCOMES.
UH, WE REPORTED, UH, THE FINDINGS TO THE PUBLIC UTILITY COMMISSION ON NOVEMBER 10TH, AND, UH, SORRY, THE REPORT WAS MADE AVAILABLE ON THE 10TH, AND WE REPORTED AT THE OPEN MEETING ON THE 14TH.
AND TODAY'S WORKSHOP IS REALLY, UH, AS WE STATED BACK THEN, THIS IS REALLY YOUR OPPORTUNITY AS, AS STAKEHOLDERS TO HAVE ACCESS TO THE CONSULTANTS TO ASK REALLY ANY QUESTIONS THAT YOU HAVE.
I KNOW WE'VE, BOTH, GORDON AND I HAVE TALKED TO FOLKS ABOUT PAGE, PAGE THIS, FOOTNOTE THAT, AND IT'S A GREAT OPPORTUNITY FOR, FOR EVERYONE TO HEAR THE SAME THING AND TO, TO BE ABLE TO, TO DRILL DOWN INTO THOSE PARTICULAR QUESTIONS THAT YOU MAY HAVE.
SO, REALLY, THIS IS AN OPPORTUNITY TO ASK ANY AND ALL QUESTIONS THAT YOU HAVE.
UH, BUT BEFORE WE DO THAT, WE'RE GONNA HAVE, UH, OLIVIER, HE'S GONNA GET UP IN, IN A MOMENT TO THE PODIUM.
UH, THEY'RE GONNA GO THROUGH, UH, UH, ESSENTIALLY THE EXECUTIVE SUMMARY.
AND THEN FROM THERE, WE'LL, WE WILL OPEN IT UP TO ANY QUESTIONS ON ANY PA, ANY SLIDE, UH, AND AND REALLY DIG INTO THE, THE REPORT AND ANY QUESTIONS THAT THAT FOLKS HAVE ON THAT.
SO THAT'S WHERE WE'RE AT AND WE LOOK FORWARD TO A GREAT DISCUSSION TODAY.
SO, OLIVIER, WE WILL TURN IT OVER TO YOU.
I THINK OLIVIER IS GONNA GO UP TO THE PODIUM AND, UH, WE'LL GO FROM THERE.
THANK YOU, UH, KEITH FOR THIS INTRODUCTION.
LOOKING FORWARD TO THIS DISCUSSION.
UH, WE ARE GOING, WE'RE PLANNING TO GO THROUGH A LITTLE BIT OF CONTEXT AND THEN JUST THE EXECUTIVE SUMMARY SECTION OF THE REPORT TO SORT OF LEVEL SET IN THE ROOM.
I IMAGINE ALL OF YOU, OR MOST OF YOU HAVE READ IT, BUT HELPFUL TO HAVE A LITTLE REFRESHER BEFORE WE DIVE INTO, UH, FOOTNOTE THREE.
UM, SO QUICK, UH, INTRODUCTION.
[00:05:01]
I'M OLI BOFI, I LEAD OUR ERCOT MARKET TEAM HERE AT AURA.UH, MAYBE A FEW WORDS ABOUT AURA, IF YOU'RE NOT AS FAMILIAR.
UH, SO WE ARE A GLOBAL POWER MARKET ANALYTICS FIRM.
UM, WE HAVE 17 OFFICES ACROSS THE US, EUROPE, AND ASIA.
UH, BUT OUR US HEADQUARTERS ARE HERE, HERE IN AUSTIN, TEXAS, WHERE WE HAVE ABOUT 150 PEOPLE.
UM, WE WORK ACROSS THE MARKET ACROSS, UM, TECHNOLOGIES WITH GENERATORS, UTILITIES, RETAILERS, LARGE CONSUMERS, EVEN DATA CENTERS, UH, AND REALLY SUPPORT ALL TECHNOLOGIES, UM, SO SOLAR, WIND BATTERIES, THERMAL GENERATORS.
UM, AND SO WE'RE, WE'RE REALLY TRYING TO PROVIDE, UH, INFORMATION EQUALLY TO ALL MARKET PARTICIPANTS.
UH, OUR ROLE AT THE END OF THE DAY IS TO DO ANALYSIS AND MODELING ON POWER MARKETS TO INFORM THE DISCUSSIONS AND HELP STAKEHOLDERS MAKE DECISIONS, RIGHT? WE DON'T HAVE A STAKE IN THE MARKET.
UH, WE'RE REALLY JUST HERE TO INFORM.
UM, MAYBE ANOTHER QUICK POINT IS WE HAVE QUITE EXTENSIVE EXPERIENCE IN ERCOT.
UM, AT THIS POINT WE'VE SUPPORTED GIGAWATTS OF, UH, GENERATION THROUGH, UM, THE FINANCING TRANSACTIONS AND INTERCONNECTION PROCESS ACROSS TECHNOLOGIES.
UH, AND SO WE'VE, WE'VE SEEN FIRSTHAND THE CHALLENGES OF, OF GETTING A GENERATION BUILT IN ERCOT AS WELL AS, UM, LOAD, UH, AND, AND COST OF OF POWER.
IF WE GO TO THE NEXT SLIDE, JUST THAT BRINGS US TO THE CONTEXT FOR THIS STUDY.
SO KEITH HAS MENTIONED A FEW THINGS, BUT, UM, REALLY THE GOAL, UH, WE WORKED WITH ERCOT TO DO TWO THINGS.
THE FIRST ONE WAS AN ASSESSMENT OF RESOURCE ADEQUACY NEEDS IN ERCOT, AND THE SECOND WAS ANALYZING AND QUANTIFYING THE IMPACT OF SPECIFIC PROPOSED OR, OR, UM, YEAH, PROPOSED MARKET DESIGN CHANGES.
UM, SO THE FIRST PART WE EVALUATED RESOURCE ADEQUACY NEEDS UNDER A STATUS QUO SCENARIO.
SO MEANING REALLY CURRENT MARKET DESIGN WITH NO CHANGES, UM, AND UNDER BOTH NORMAL WEATHER CONDITIONS, UH, WHICH WE'LL GET INTO IT, BUT IS IT 2013, UH, WEATHER YEAR AND EXTREME QUOTE UNQUOTE, UH, WEATHER CONDITIONS, WHICH WE LOOKED TO BE THE, UH, 2022 WEATHER YEAR WITH STORM ELLIOT AND THE 2023 WEATHER YEAR WITH, UH, THE 2023 SUMMER WE HAD, WHICH WAS PARTICULARLY HOT.
UM, THEN WE MODELED THE IMPACT ON RELIABILITY AND COSTS OF THREE DIFFERENT PROPOSED MARKET DESIGN CHANGES.
UH, THE FIRST ONE BEING D-R-D-R-R-S, UM, THE WAY IT'S BEEN SORT OF INITIALLY PROPOSED, UM, AS, AS THE, AND, AND IS REFLECTED IN ONE OF THE NRR.
UM, THE SECOND ONE BEING THE DRRS PLUS, UH, SO THAT THIS SORT OF MORE EXTENSIVE VERSION THAT'S BEING DISCUSSED IN THE OTHER NPRR.
AND THE FINAL ONE WE ALSO EVALUATED WAS AN OR DC EXTENSION.
UH, SO SORT OF A LONGER RDC CURVE, KEEPING THAT $5,000 PER MEGAWATT HOUR CAP.
SO THIS IS A DETERMINISTIC ANALYSIS.
WE RUN SPECIFIC SCENARIOS UNDER SPECIFIC CONDITIONS.
THIS DOES NOT CAPTURE THE INFINITE NUMBER OF COMBINATIONS OF LOAD GENERATION AND OUTAGES THAT CAN HAPPEN.
THERE'S OBVIOUSLY OTHER CONDITIONS THAT COULD LEAD TO DIFFERENT, DIFFERENT OUTCOMES.
UM, AND THE SECOND PIECE, AND THAT'S PROBABLY MORE A DISCLAIMER ON OUR SIDE, BUT OBVIOUSLY OUR REPORT OR OURSELVES ARE NOT ACTIVELY ADVOCATING FOR ANY SPECIFIC PROPOSALS OR MARKET DESIGN, RIGHT? WE'RE HERE TO BRING THE NUMBERS IN THE ANALYSIS, UH, FOR DISCUSSION.
UM, QUICK EXECUTIVE SUMMARY, UM, SORT OF OVERVIEW BEFORE WE DIVE INTO MORE OF THE DETAILS.
UM, SO THE FIRST PART I THINK THAT'S IMPORTANT TO KEEP IN MIND IS OBVIOUSLY ERCOT IS QUITE UNIQUE IN THE UNITED STATES.
IT HAS A SPECIFIC ENERGY ONLY MARKET DESIGN THAT HAS BEEN VERY EFFECTIVE AT BRINGING ON NEW LOAD, UH, AND, AND NEW ECONOMIC DEVELOPMENTS AS WELL AS REALLY, UM, INCENTIVIZING STORAGE AND RENEWABLE GENERATION TO, TO CONNECT TO THE GRID.
UM, BUT IT ALSO HAS YIELDED RELATIVELY LOW HISTORICAL OPERATIONAL RESERVES, AND WE CAN CERTAINLY GET INTO THE DEFINITION OF THAT.
BUT, UH, ESSENTIALLY, UH, LESS SORT OF RESERVES THAN IN OTHER MARKETS UNDER SORT OF TIGHT SYSTEM CONDITIONS.
AND THAT'S EVEN WHEN EXCLUDING YURI, WHICH OBVIOUSLY WAS NOT, UH, WAS, WAS PARTICULARLY, UH, COMPLICATED.
THE SECOND POINT IS WHEN WE WERE ASSESSING RESOURCE ADEQUACY NEEDS, UH, AND THE IMPACT OF LOAD GROWTH, UM, WE LOOKED AT TWO MAIN SCENARIOS.
ONE, A HIGH LOAD GROWTH SCENARIO THAT CORRESPONDS
[00:10:01]
TO THE ERCOT 2025 LTLF, UH, WHICH INCLUDES 20 20, 22 GIGAWATTS OF DATA CENTER DEMAND, PLUS ADDITIONAL DEMAND GROWTH FROM OTHER SECTORS.UH, AND ANOTHER ONE, WHICH WE WILL CALL THIS SORT OF LOWER LOAD GROWTH SCENARIO, WHICH IS THE AURA CENTRAL, WHICH HAS A MORE MODERATE AMOUNT OF LOAD GROWTH, STILL A FAIRLY LARGE LOAD GROWTH ANYWAY, BUT JUST, JUST MORE MODERATE.
UM, THE FINDINGS FROM THOSE SCENARIOS WERE THAT UNDER THE HIGH LOAD GROWTH AIR CODE LTLF AND THE PROVISIONS IN SB SIX, THE DATA CENTER FLEXIBILITY, UH, ACTUALLY HELPED TO, UH, ALLEVIATE THE LOAD SHEDDING, UH, RISKS FROM EXTREME WEATHER.
AND THAT FLEXIBILITY IN PARTICULAR WAS THE KEY TO ALLEVIATING IT IN, IN THAT SCENARIO.
WHEREAS UNDER THE LOWER LOAD GROWTH SCENARIO, THAT FLEXIBILITY WAS NOT ENOUGH UNDER EXTREME WEATHER CONDITIONS TO, UH, ALLEVIATE THE LOAD GROWTH.
UH, ONE QUICK POINT TO NOTE IN THIS, IN BOTH SCENARIOS, THE SORT OF NORMAL WEATHER CONDITIONS DID NOT LEAD TO LOAD SHEDDING IN EITHER SCENARIO.
SO, SO THAT'S AN IMPORTANT POINT, RIGHT? IS, IS WHAT WE'RE FINDING IS UNDER NORMAL MARKET CONDITIONS, THE STATUS QUO IS LIKELY ENOUGH, THE PROBLEMS REALLY START TO APPEAR UNDER THE MORE EXTREME WEATHER CONDITIONS.
UM, FINALLY WHEN WE EVALUATED THE THREE DIFFERENT PROPOSED MARKET DESIGN CHANGES, UH, WHAT CAME OUT, AND WE'LL GO INTO A LOT OF DETAILS ON THAT, I IMAGINE IS THAT THE DRS PLUS PROPOSAL WAS THE ONE THAT PROVIDED THE MOST RELIABILITY BENEFITS IN TERMS OF REDUCING LOAD SHED UNDER SCENARIOS, AS WELL AS INCREASING THE ADDITIONAL, THE AMOUNT OF DISPATCHABLE CAPACITY BEING BUILT AT, AT THE LOWER NET COSTS.
AFTER THAT, THE ODC EXTENSION ALSO PROVIDED, UH, A SIGNIFICANT, A SIGNIFICANT AMOUNT OF ADDITIONAL CAPACITY AND LOAD CHECK REDUCTION, BUT AT A HIGHER COST.
UM, AND FINALLY, THE DRS SORT OF ORIGINAL PROPOSAL, UH, HAD FAIRLY LIMITED RELIABILITY BENEFITS FROM A RESOURCE ENCY STANDPOINT, UH, OF, OF, AND OBVIOUSLY HERE WE'RE NOT TALKING ABOUT SORT OF THE MORE OPERATIONAL BENEFITS THAT, THAT THAT SYSTEM CAN HAVE.
UH, AND THAT WAS SORT OF IN LINE WITH THE MORE LIMITED BUDGET THAT WAS ALLOCATED.
SO THAT CONCLUDES THE SORT OF VERY HIGH LEVEL OVERVIEW.
UH, NOW THE, UH, PLAN IS TO GO THROUGH THAT EXECUTIVE SUMMARY IN A LITTLE BIT MORE DETAILS, UH, WITH KEVIN AND THE TEAM.
UH, IF YOU WANNA COME UP HERE AND, UH, I PROMISE WE'LL KEEP IT TO REALLY UNDER 30 MINUTES.
WE WANT TO LEAVE PLENTY OF TIME FOR QUESTIONS, SO THIS IS JUST REALLY A QUICK RUN THROUGH.
UH, AND THEN WE WILL, WE'LL OPEN THE FLOOR, OPEN UP TO THE FLOOR.
ALL RIGHT, THANK YOU, OLIVIER.
SO BEFORE WE DIVE RIGHT INTO THE EXECUTIVE SUMMARY, UH, WE DID WANT TO, UH, FOR THE SAKE OF EFFICIENCY, MAKE SURE EVERYONE'S ON THE SAME PAGE WITH HOW AURORA, UH, AND OUR MODELING, UM, METHODOLOGY ACTUALLY WORKS, UH, SO THAT WE CAN MAKE SURE THAT THE, UH, THE SPECIFIC QUESTIONS ARE, ARE ADDRESSED, UH, IN ADVANCE OF KIND OF THE, THE BROADER REPORT DETAILS.
SO TO START, UH, AURORA HAS OUR OWN DEMAND FOR FORECAST METHODOLOGY.
WE BUILD OUR VIEWS ON LOAD GROWTH BOTTOM UP, UH, AND IT'S SOMETHING THAT WE BASICALLY UPDATE ON A RECURRING BASIS TO ALIGN WITH OUR INTERNAL VIEWS ON HOW LOAD GROWTH IS EXPECTED TO GROW.
AGAIN, OBVIOUSLY, UH, WE, UH, OUR, OUR POSITION WITHIN THE MARKET IS TO BE A NEUTRAL OBSERVER AND THEN ALSO INCORPORATE SEVERAL DIFFERENT, UH, VOICES ACROSS THE INDUSTRY.
AND SO WE TRY TO TAILOR OUR OWN VIEWS BASED OFF OF THAT, AND THAT'S WHY IT DIFFERS FROM ERCO T'S OWN LTLF.
IN TERMS OF THE BROADER METHODOLOGY, UH, WE USE THE CAPACITY EXPANSION MODEL.
SO BASICALLY IT INCORPORATES ASPECTS OF, OF, UH, HOURLY SUB HOURLY DISPATCH, UM, ITERATIVE MODELING AS WELL AS AN INVESTMENT DECISIONS MODULE TO BASICALLY MATCH SUPPLY WITH DEMAND, UH, USING THE UNDERLYING ASSUMPTIONS.
AND THOSE ASSUMPTIONS INCLUDE THINGS LIKE DEMAND, UH, COMMODITY PRICE ASSUMPTIONS, TECHNOLOGY COSTS, ET CETERA.
UM, AND THIS INVESTMENT DECISIONS MODULE IS EFFECTIVELY TRYING TO SOLVE FOR THE SUPPLY DEMAND EQUILIBRIUM THAT WILL MEET THE DEMAND OF THE SYSTEM AT THE LOWEST SYSTEM COSTS.
AND WE COULD GO INTO MUCH MORE DETAIL ON THIS LATER IN THE EXECUTIVE SUMMARY AS WELL.
UH, BUT WHEN LOOKING AT THE DIFFERENT LOAD GROWTH TRAJECTORIES, THE KEY HIGHLIGHT HERE IS THAT THERE'S A WIDE RANGE OF VIEWS ON LOAD GROWTH.
AND SO OUR BASIC VIEW WAS TO TRY TO INCORPORATE A, A BROAD RANGE OF LOAD GROWTH SCENARIOS SO THAT THIS STUDY, UH, CAN APPLY ACROSS DIFFERENT LOAD GROWTH SCENARIOS.
EVEN THOUGH, UH, EACH ONE OF THESE INDIVIDUALLY, IT MIGHT NOT BE THE EXACT
[00:15:01]
LOAD GROWTH FORECAST OR THAT THAT AC ACTUALLY MATERIALIZES.SO MOVING ON TO HOW THE DIFFERENT SCENARIOS WERE STRUCTURED, WE HAVE BASICALLY TWO DIFFERENT DEMAND SCENARIOS THAT WE MODELED, AND WE ALSO APPLIED DIFFERENT WEATHER YEARS ON TOP OF THAT TO STRESS TEST THE SYSTEM UNDER EXTREME WEATHER.
NOW, WE ARE VERY DELIBERATE ABOUT CHOOSING SPECIFIC WEATHER YEARS THAT STRESS THE SYSTEM UNDER A SUMMER HEAT WAVE TYPE CONDITION, UH, THAT WAS ALSO RECENT, BUT ALSO WHAT WE THINK IS REASONABLE AS A A WEATHER YEAR OUTCOME IN THE NEAR FUTURE.
AND SO FOR THE HOT SUMMER, WE USED THE 2023 WEATHER YEAR, UH, AND FOR THE, UH, FOR THE WINTER STORM WE USED 22, UH, TO, TO SIMULATE THE WINTER STORM ELLIOT.
SO JUST DIVING INTO A LITTLE BIT ABOUT HOW OUR CAPACITY EXPANSION WORKS IN THE REALLY SHORT TERM, WE'RE TAKING INCLUSION OF CAPACITY FROM THE ERCOT INTERCONNECTION QUEUE TO INFORM WHAT OR WHAT PORTION OF THAT CAPACITY ACTUALLY COMES ONLINE.
WE HAVE DIFFERENT SUCCESS FACTORS THAT HELPS US BASICALLY FILTER WHICH CAPACITY WILL REALISTICALLY, UH, COME ONLINE THAT'S IN THE INTERCONNECTION QUEUE.
AND THEN LONGER TERM, WE'RE USING THAT CAPACITY EXPANSION LOGIC TO BASICALLY ENSURE THAT THE EQUILIBRIUM IS MET, UH, BASED OFF OF OUR CENTRAL CASE DEMAND SCENARIO OR ALSO THE LTLF, UH, SCENARIO THAT WE MODELED.
AS PART OF THIS STUDY, WE ALSO INCORPORATE DIFFERENT RETIREMENT ASSUMPTIONS.
AND SO IT'S, IT'S NOT JUST ABOUT ADDITIONS.
UH, WE HAVE A VIEW ON BOTH THE ECONOMICS, BUT ALSO THE TECHNICAL LIMITATIONS OF AGING PLANTS TO MAKE SURE THAT WE'RE CAPTURING A GOOD REPRESENTATIVE VIEW ON WHAT THE SUPPLY MIX WILL ACTUALLY LOOK LIKE.
AND THIS IS JUST A VIEW ON HOW THAT LOOKS IN TERMS OF MODEL OUTPUTS.
SO ON THE LEFT SIDE, THIS IS JUST THE CAPACITY MIX THAT RESULTS FROM TAKING THAT AURORA CENTRAL LOAD GROWTH AND APPLYING THAT CAPACITY EXPANSION MODULE, UH, TO FORECAST DIFFERENT, UH, CAPACITY EXPANSION OVER THAT PERIOD ACROSS THE DIFFERENT TECHNOLOGY TYPES.
UH, I'LL BASICALLY SKIP OVER MOST OF THE DETAILS THAT ARE ON THE PAGE, BUT, UH, WHAT WE BASICALLY SEE IS THAT THE SYSTEM CONTINUES TO, UH, ABSORB MORE OF THE INTERMITTENT RENEWABLES AS WELL AS FLEXIBLE GENERATION, UH, WHEREAS THE CONVENTIONAL GENERATION TENDS TO STAY RELATIVELY FLAT TO SLIGHTLY DECREASING DUE TO RETIREMENTS, PROBABLY SKIP YEAH, AND WE COULD SKIP JUST SOME OF THE DIFFERENCES.
BUT, UH, BASICALLY THE, THE LAST THING THAT I'LL MENTION IS THAT ACROSS ALL THOSE DIFFERENT, UH, WEATHER YEAR SCENARIOS AS WELL AS THE DIFFERENT LOAD ASSUMPTIONS, WE THEN APPLIED THE DIFFERENT MARKET DESIGN CHANGES ACROSS THEM SO THAT WE COULD EVALUATE, UH, THE DIRECT IMPACTS FROM, UH, THOSE MARKET DESIGN CHANGES IN TERMS OF CAPACITY, OUTCOMES AND ALSO RELIABILITY OUTCOMES.
ALRIGHT, SO THAT'S BASICALLY IN TERMS OF KIND OF THE BROADER METHODOLOGY, UM, WE WILL KIND OF TURN OVER TO THE REGULAR DECK WHERE WE WILL GO THROUGH THE EXECUTIVE SUMMARY OF THE REPORT.
ALL RIGHT, SO WE START THE EXECUTIVE SUMMARY BY ILLUSTRATING WHAT WE'VE BEEN OBSERVING HISTORICALLY WITHIN ERCOT AS COMPARED TO DIFFERENT COMPETITIVE ISOS IN THE UNITED STATES.
AND WHAT WE'VE SEEN IS THAT ERCO T'S, UH, UNIQUE MARKET STRUCTURE AS WELL AS APPROACH TO INTERCONNECTION HAS YIELDED SOME INSIGHTFUL RESULTS.
UH, SO IN TERMS OF, UH, PEAK LOAD GROWTH FROM THE YEARS 2020 TO 2024, ERCOT HAS THAT HAD THE HIGHEST GROWTH.
UM, WE'VE ALSO SEEN THE HIGHEST AMOUNT OF RENEWABLES CAPACITY ADDITIONS IN THAT SAME TIMEFRAME.
AND, UH, A COMBINATION OF OF THOSE HAVE ALSO YIELDED THE LOWEST AVERAGE HISTORICAL OPERATIONAL RESERVES IN THAT TIME PERIOD AS WELL.
SO THIS AGAIN, UH, ILLUSTRATES THAT HISTORICALLY WITH THE CURRENT STATUS QUO MARKET DESIGN, THE ACTUAL HISTORICAL OPERATIONAL RESERVES IN ERCOT HAVE BEEN LOWER THAN ANY OTHER COMPETITIVE ISO IN THE UNITED STATES.
WE'VE DISCUSSED THIS LOAD GROWTH PICTURE.
I THINK THAT THE KEY THING TO ILLUSTRATE IS THAT, UH, MOST OF THE CONVERSATION AND DIALOGUE HAS BEEN AROUND THE, UH, MASSIVE GROWTH OF AI DATA CENTER DRIVEN DEMAND, BUT THERE ARE ALSO OTHER LOAD GROWTH DRIVERS THAT WE'VE BEEN OBSERVING.
THINGS LIKE POPULATION GROWTH, GDP GROWTH AND OTHER INDUSTRIAL LOAD GROWTH THAT HAS BEEN FUELING THIS, UH, THIS DEMAND.
UM, AND WE'VE ALSO, UH, IN RECENT HISTORY SEEN WARMER TEMPERATURES AS WELL THAT HAS CONTRIBUTED TO THAT.
AND SO WE HAVE VIEWS ON LOAD GROWTH DRIVERS ACROSS ALL OF THOSE DIFFERENT CATEGORIES TO JUST ENSURE THAT WE'RE, WE'RE
[00:20:01]
CAPTURING ALL THOSE DYNAMICS, UH, SIMULTANEOUSLY.SO DIVING A LITTLE BIT DEEPER INTO THE JUSTIFICATION FOR WHY WE SELECTED SPECIFIC WEATHER YEARS, UH, WE FIRST WANTED TO CAPTURE A WINTER STORM THAT WAS REFLECTIVE OF SOMETHING THAT HAPPENED RECENTLY THAT WE ALSO VIEWED, UH, IS A GOOD REPRESENTATIVE EXTREME CASE THAT COULD ALSO, UH, HAPPEN IN THE FUTURE.
WINTER STORM ELLIOT, UH, BROUGHT EXTREME COLD WEATHER, WHICH HAS ONLY BEEN MATCHED ONCE SINCE, UH, THAT WAS WINTER STORM HEATHER IN JANUARY, 2024.
UH, TEMPERATURES WERE NOT EX AS EXTREME AS DURING WINTER STORM URI, BUT WE VIEWED THAT AS REALLY AN OUTLIER CASE THAT WOULDN'T, UH, WOULD, WE, WOULD, IT WOULD BE MORE EFFECTIVE FOR US TO SHOW MORE OF A WINTER STORM ELLIOT STYLE EVENT FOR THE PURPOSES OF THIS STUDY.
AND SO WHAT WE ESSENTIALLY DID WAS WE USED OUR DEMAND AND RENEWABLES, UH, GENERATION ASSUMPTIONS AND OUTAGE PROFILES TO RECREATE THE EFFECT OF THIS, UH, WINTER STORM EVENT, BUT UNDER 2030 SUPPLY AND DEMAND CONDITIONS AND FOR 2023, UH, SOMETHING THAT, UH, A A YEAR THAT'S RELATIVELY RECENT.
AND ALSO I THINK FRESH ON PEOPLE'S MINDS, WE WANTED TO CAPTURE THE EVENT THAT BROUGHT EXTREME HEAT FOR AN EXTENDED PERIOD OF TIME DURING THE SUMMER, UM, AND ALSO THE CORRESPONDING OUTAGES, UH, THAT THAT IMPACTED THE SYSTEM, UH, TO AGAIN, SIMULATE THOSE CONDITIONS IN A FUTURE YEAR.
AND WHAT WE SAW UNDER OUR AURORA CENTRAL DEMAND FORECAST IS THAT LOW GROWTH WAS, UH, BASICALLY OBSERVED DURING THE, UH, WINTER STORM EVENT AS WELL AS THE SUMMER HEAT WAVE, WHICH BASICALLY HELPED US SEE THAT BECAUSE OF THE RESOURCE COMPOSITION THAT ERCOT HAS BEEN ADOPTING IN RECENT YEARS AND IS EXPECTED TO CONTINUE, UM, BASICALLY HIGHLY CONCENTRATED IN INTERMITTENT RENEWABLES AS WELL AS SHORT DURATION ENERGY STORAGE THAT THE SYSTEM WAS SUSCEPTIBLE FOR LOAD SHED, UM, GIVEN THE CURRENT STATUS QUO MARKET DESIGN.
AND HERE ARE SOME OF THE RELIABILITY METRICS THAT SUMMARIZE WHAT WE'VE SEEN.
SO THIS IS JUST IN TERMS OF SINGLE EVENT METRICS, BUT THE MAX LOAD SHED UNDER A WINTER STORM AND ALSO A SUMMER HEAT WAVE UNDER BOTH DEMAND SCENARIOS.
UH, THAT WAS 8.4 GIGAWATTS, WAS THE MAX LOAD SHED, UH, FOR THE ERCOT LTLF CASE UNDER A WINTER STORM AND 9.7 GIGAWATTS UNDER A SUMMER HEAT WAVE.
AND BASED ON A VOL OF $35,000 PER MEGAWATT HOUR, WE TRIED TO QUANTIFY WHAT THE OVERALL COST WOULD BE, UH, TO, UH, FOR THAT SPECIFIC LOAD SHE IN THOSE SPECIFIC YEARS.
AND THIS TEMPLATE FOR SUMMARIZING THE RELIABILITY OUTCOMES IS ALSO GOING TO BE KIND OF, UH, MIRRORED IN ALL OF OUR OTHER, UH, SIMULATIONS ACROSS DIFFERENT MARKET DESIGN CHANGES.
ALL RIGHT, SO NOW I'LL PASS IT OVER TO PAYDEN TO TALK ABOUT THE DEMAND SIDE RESPONSE SCENARIO, UH, THAT WE CONDUCTED AS PART OF THIS STUDY.
SO MOVING INTO THIS PHASE OF THE EXECUTIVE SUMMARY, UH, WE REALLY NEEDED TO CONSIDER IN THE CONTEXT OF SENATE BILL SIX, THE LOAD GROWTH.
WE'RE SEEING HOW COULD DATA CENTER FLEXIBILITY POTENTIALLY IMPACT THESE RESULTS? AND SO WHAT THIS SLIDE IS REALLY HIGHLIGHTING, AS WE CAN SEE ON THE RIGHT HAND SIDE, ERCOT 2025, LONG-TERM LOAD FORECAST BY 2030, THERE ARE 22 GIGAWATTS OF DATA CENTERS.
UH, SO CLEARLY A VERY SIGNIFICANT AMOUNT OF CAPACITY, AND THIS PROVIDES A LARGE POOL OF RESOURCES THAT COULD CONTRIBUTE, UH, A BENEFIT, UH, SOME FLEXIBILITY TO THE SYSTEM.
AND SO WE WANTED TO CONSIDER THAT IN THE ANALYSIS.
SO THINKING ABOUT HOW WE WERE GOING TO INCORPORATE IN THIS, IN THE ANALYSIS, UH, FIRST WE, UH, DREW UP THREE BUCKETS OF POTENTIAL BEHAVIOR FOR THESE DATA CENTERS.
SO THE FIRST IS, UH, PRICE RESPONSIVE DATA CENTERS.
SO THESE WOULD BE ONES THAT ARE PARTICIPATING, UM, MUCH AS A CRYPTO MINE WOULD THESE DAYS.
UH, RESPONSIVE TO PRICE, ECONOMICALLY DRIVEN DECISIONS LEADS TO LOAD REDUCTIONS IN PERIODS OF HIGH PRICES.
UH, THE SECOND WOULD BE EMERGENCY OR ANCILLARY SERVICE PROVIDERS.
SO THESE WOULD BE MORE EEA TWO EVENTS, UH, POTENTIALLY SENATE BILL SIX PROVISIONS, UH, MORE OF A LAST CASE, UH, UM, RESORT OF REDUCING LOAD.
UH, THIS WOULD NOT BE, UH, AFFECTING PRICES OR SYSTEM SCARCITY.
AND THEN THE, THE THIRD BUCKET WOULD BE INFLEXIBLE.
THESE WOULD BE DATA CENTERS THAT HAVE, UH, NO INCENTIVE TO TURN DOWN LOAD AND THEY WILL, WILL NOT PARTICIPATE IN ANY OF THOSE PROGRAMS. AND SO WITHIN THE CONTEXT OF SENATE BILL SIX, THERE'D BE SOME THAT MIGHT BE DIRECTED TO, UH, SOME THAT MIGHT NOT.
AND SO PUTTING BOUNDS ON THE AMOUNT WE WOULD ASSUME IN EACH CATEGORY, UH, WE CONDUCTED SOME EXTENSIVE RESEARCH.
WE INTERVIEWED MANY MARKET PARTICIPANTS TO GET A GOOD SENSE OF HOW THIS MAY PLAY OUT IN THE FUTURE.
UH, AND ULTIMATELY WE ESTABLISHED A, A 10% ASSUMPTION FOR PRICE RESPONSIVE DATA CENTERS, UH, 50% FOR, UH, EMERGENCY, EMERGENCY
[00:25:01]
OR ANCILLARY SERVICE, AND THEN 40% THAT WOULD BE INFLEXIBLE.AND SO THAT WAS USED TO EVALUATE, UH, THE FORECAST WITH THOSE LEVELS OF, UH, FLEXIBILITY.
SO IN THE CONTEXT OF THE ERCOT LONG-TERM LOAD FORECAST DEMAND GROWTH SCENARIO, UH, WE CAN SEE THAT, UH, WHERE WE DID HAVE LOAD SHED, UH, WITH NO FLEXIBILITY FROM DATA CENTERS, UH, ASSUMING THAT 60% WAS FULLY SUFFICIENT TO ALLEVIATE LOAD SHED.
SO THIS IS REALLY LOOKING AT THE LEFT HAND SIDE OF THE GRAPH ON THE TOP WAS THE EEMAX LOAD SHED IN 2030 WITHOUT DATA CENTER FLEXIBILITY ON THE RIGHT HAND SIDE IS WITH DATA CENTER FLEXIBILITY, UH, AND THE BOTTOM ESTIMATES THE VALUE OF, UH, LOAD COST THAT WOULD BE SAVED ASSUMING A $35,000, UM, VALUE OF LOST LOAD.
UH, THE INTERESTING RESULT IS WHEN WE, UH, CONDUCTED THE SAME, UH, ANALYSIS UNDER THE, THE CENTRAL DEMAND GROWTH CONDITIONS.
SO AGAIN, THIS WAS A MORE MODERATE DEMAND GROWTH ASSUMPTION.
AND IN THESE CASES, WHEN THAT POOL OF LARGE LOAD FLEXIBILITY IS MUCH SMALLER, IT IS NO LONGER SUFFICIENT, UH, TO FULLY PREVENT LOAD SHED UNDER THE EXTREME WEATHER CONDITIONS OF A HOT SUMMER OR A WINTER STORM ELLIOT STYLE WEATHER EVENT.
SO THERE IS STILL RETAINED VALUE, UH, BUT JUST DO THE SMALLER AMOUNT OF POTENTIAL FLEXIBILITY.
SO NOW I'LL TURN IT OVER TO TIM.
SO COMING OUT OF THE PREVIOUS SECTION OF THE REPORT, UM, HIGHLIGHTING THAT DEMAND SIDE FLEXIBILITY UNDER CERTAIN SCENARIOS IS NOT SUFFICIENT TO FULLY ALLEVIATE LOAD.
THE IDEA IS THAT WE WOULD MODEL, UM, THE IMPACT OF ALTERNATIVE MARKET DESIGNS, UH, TO SEE HOW THEY IMPACT RELIABILITY.
SO AS ALREADY ALLUDED TO, WE MODELED THREE DIFFERENT CASES.
UH, ANCI, UH, DRS IS AN ANCILLARY SERVICE DRS ANCILLARY SERVICE PLUS AND AN EXTENSION OF THE ORDC.
UH, LOOKING AT DRS ANCILLARY SERVICE, THE PURPOSE OF THIS DESIGN IS REALLY TO IMPROVE OPERATIONAL FLEXIBILITY, REDUCE RUCKING, ET CETERA.
IN TERMS OF ELIGIBILITY, WE MODELED TWO DISTINCT CASES.
IN ONE CASE, BATTERIES ARE NOT ELIGIBLE TO PARTICIPATE.
IN THE OTHER CASE, BATTERIES WITH A FOUR PLUS HOUR DURATION NAMEPLATE RATING ARE ELIGIBLE TO PARTICIPATE.
IN TERMS OF PROCUREMENT SIZE, VERY SIMILAR TO, UH, WHAT WE'VE SEEN WITH TRADITIONAL RESERVE SERVICES.
SO ONE TO FOUR GIGAWATTS PER HOUR, UH, ON AVERAGE IN 2030 DRS ANCILLARY SERVICE.
PLUS THE PURPOSE OF, UH, THIS MARKET DESIGN IS REALLY TO BOLSTER LONG-TERM RESOURCE ADEQUACY.
UM, IN TERMS OF ELIGIBILITY, UH, SAME THING AS DRS, ANCILLARY SERVICE BATTERY INELIGIBLE CASE IN A BATTERY ELIGIBLE CASE IN TERMS OF PROCUREMENT SIZE, UH, MUCH MORE EXTENDED THAN D-R-S-A-S.
UM, SO PROCURING BETWEEN 80 AND 140 GIGAWATTS PER HOUR, UH, UNDER OUR DIFFERENT MODELED CASES.
UH, THE BIG THING TO NOTE HERE IS THAT IN OUR MODEL, AFTER PROCURING, UH, THE CAPACITY, IT'S THEN RELEASED BACK FOR PARTICIPATION IN THE ENERGY MARKET.
AND THEN FINALLY, THE EXTENDED ORDC CASE.
THIS IS REALLY JUST, UH, TAKING THE CURRENT RDC STRUCTURE AND DRAGGING IT OUT TO INCREASE, UH, THE INVESTMENT SIGNAL.
COMING FROM THE ORDC, UH, IN THE MODEL, THIS WORKS OUT TO ADDING ABOUT TWO TIMES AS MUCH SCARCITY VALUE TO THE SYSTEM.
ALRIGHT, SO STARTING OUT LOOKING AT CAPACITY AND COST OUTCOMES, AND THERE'S A COUPLE DIFFERENT PERMUTATIONS OF THIS SLIDE, BUT FOR THIS SPECIFIC ONE, WE'RE LOOKING AT COST OUTCOMES UNDER THE ERCOT LTLF DEMAND SCENARIO.
AND IN THIS CASE, BATTERIES ARE NOT ELIGIBLE FOR PARTICIPATION.
UH, SO MOVING LEFT TO RIGHT WITH THESE GRAPHS WE HAVE ALL IN SYSTEM COSTS, UH, ACROSS ENTIRE YEAR 2030 AND THE CENTER, WE HAVE THE COST DELTA TO THE STATUS QUO FOR EACH OF THESE, UH, MARKET DESIGN OPTIONS.
AND THEN ON THE FAR RIGHT HAND SIDE, WE HAVE THE CAPACITY DELTA, UH, TO THE STATUS QUO.
FOCUSING ON THE GRAPH IN THE CENTER, UM, YOU'LL SEE THESE BLACK DIAMONDS, WHICH REPRESENT THE TOTAL COST DELTA.
SO D-R-S-A-S INCREASES COSTS RELATIVE TO STATUS QUO BY 700 MILLION DRS AS PLUS INCREASES COST BY 400 MILLION.
AND I'LL PAUSE THERE, JUST TO NOTE THAT THE D-R-S-A-S UH, PAYMENT ITSELF IS SLIGHTLY OVER 4 BILLION.
YOU CAN SEE THAT IN THAT, UH, LIGHTER BLUE SHADE, BUT YOU ALSO SEE THAT THERE'S A SIGNIFICANT OFFSET IN ENERGY COSTS THAT REDUCE TOTAL COSTS DOWN TO 400 MILLION.
AND THEN IN THE EXTENDED ORDC CASE, UH, COSTS
[00:30:01]
INCREASE RELATIVE TO THE STATUS QUO BY 2.2 BILLION.ON THE CAPACITY DELTA SIDE, UH, WE HAVE SOME GRAY DIAMONDS ON THOSE BARS.
THOSE ARE REPRESENTING, UH, RELATIVE TO THE STATUS QUO, THE INCREASE IN LONG DURATION DISPATCHABLE CAPACITY.
AND HERE WE'RE DEFINING LONG DURATION AS INCLUDING, UH, BATTERY STORAGE WITH A FOUR HOUR NAMEPLATE DURATION.
SO THE DRS AS WE'RE INCREASING CAPACITY BY 900 MEGAWATTS, DRS AS PLUS INCREASING BY FIVE GIGAWATTS.
AND THEN THE EXTENDED ORDC INCREASING BY 1.7 GIGAWATTS, SAME VERSION OF LAST SLIDE.
SO ERCOT, LTLF DEMAND GROWTH SCENARIO.
BUT IN THIS, IN THIS CASE, WE'RE MEASURING OUTCOMES WHEN BATTERIES ARE ELIGIBLE FOR, FOR PARTICIPATION.
UM, COST OUTCOMES ARE, ARE VERY SIMILAR WITH A SLIGHT EFFICIENCY GAIN, UH, OR REDUCTION IN COSTS, UH, FROM ALLOWING BATTERIES TO PARTICIPATE.
UH, WHERE THERE'S A BIT MORE OF A DIFFERENCE IS ON THAT CAPACITY DELTA STACK.
WE NOW HAVE THESE, UH, MORE PROMINENT, UH, RED SEGMENTS ON THE BAR STACK.
THOSE ARE FOUR HOUR PLUS DURATION BATTERIES THAT HAVE BEEN INCENTIVIZED, UH, BY THE DRS AS PLUS AND DRS AS DESIGN.
AND SO YOU, YOU CAN SEE THAT ALLOWING FOR BATTERIES TO PARTICIPATE DOES INCREASE TOTAL CAPACITY.
ON THE PREVIOUS SLIDE, IT WAS FIVE GIGAWATTS.
ON THIS SLIDE WE SEE, UH, 6.3 GIGAWATTS AND THEN FINAL SLIDE ON COST AND CAPACITY.
BUT NOW WE'RE LOOKING UNDER THE AURORA CENTRAL, UH, DEMAND VIEW AND JUST LOOKING AT DRRS AS PLUS AND HOW IT COMPARES, UH, TO THE STATUS QUO.
SO AGAIN, LOOKING AT THAT CENTER GRAPH, WHICH IS THE COST DELTA TO THE STATUS QUO HERE, WE SEE THAT THE, THE TOTAL COST IS HIGHER, UH, THAN UNDER THE LTLF CASE.
REMEMBER, UNDER THE LTLF CASE, WE ARE AROUND THREE TO 400 MILLION.
HERE, WE'RE AT 1.7 TO 1.8 BILLION.
UH, THE MAIN DRIVER OF THAT IS THAT IN THE LTLF CASE WITH HEAVY DEMAND GROWTH, WE HAVE HIGHER ENERGY COSTS AND MORE SCARCITY COSTS TO BE OFFSET, UH, THAN IN THE LOWER, UH, OR MORE MODERATE DEMAND GROWTH CASE THAT IS, UH, AURORA CENTRAL.
AND THEN FINALLY, ON THE RIGHT HAND SIDE, THE CAPACITY DELTA TO THE STATUS QUO.
UH, SIMILAR OUTCOMES, UH, AS UNDER THE LTLF DEMAND SCENARIO.
BUT YOU WILL NOTICE THERE IS A BIT MORE CAPACITY THAT'S INCENTIVIZED.
UH, THE MAIN DRIVER FOR THAT IS, UH, THE AURORA CENTRAL DEMAND ASSUMES A SMALLER SYSTEM IN 2030, FEWER RESOURCES TO SPLIT THAT DRS BUDGET ACROSS.
SO THE BUDGET GOES A BIT FURTHER AND WE GET A BIT MORE CAPACITY INCENTIVIZED.
TURNING OVER TO RELIABILITY OUTCOMES.
AND WE'LL LOOK AT A COUPLE PERMUTATIONS, UM, OF THIS SLIDE.
UM, BUT THE PURPOSES OF THESE SLIDES IS TO COMPARE THE RELIABILITY OUTCOMES, UH, FROM THESE NEW MARKET DESIGN OPTIONS AGAINST THE STATUS QUO.
AND WE'RE COMPARING AGAINST THOSE THREE, UH, BIG METRICS FOR THESE EVENTS.
HOW LONG DID THE EVENT LAST? WHAT WAS THE MAXIMUM LOAD SHED DURING THE EVENT? AND WHAT WAS THE TOTAL LOAD SHED DURING THE EVENT? SO STARTING HERE WITH DRS ANCILLARY SERVICE, THE BLUE BARS ARE REPRESENTING OUTCOMES FROM THE STATUS QUO CASE.
AND THEN THE, THE YELLOW AND TAN BARS ARE OUTCOMES FROM DRS AS WHERE BATTERIES ARE ELIGIBLE AND INELIGIBLE RESPECTIVELY.
NOW WE SEE THAT THERE IS AN IMPROVEMENT IN RELIABILITY IN THE DRS, UH, AS CASE.
UM, HOWEVER, RELATIVE TO THE OTHER MARKET DESIGNS THAT WE'LL SEE SHORTLY, THE IMPROVEMENT IS FAIRLY SMALL.
IF YOU LOOK JUST AT MAXIMUM, UH, LOAD, SHE, UH, THE REDUCTION IN LOAD, SHE FROM THE STATUS QUO IS UNDER A GIGAWATT ACROSS ALL THESE CASES.
CONTRASTING THAT WITH DRS AS PLUS HERE, WE HAVE MORE CAPACITY THAT'S BEEN INCENTIVIZED.
AND SO, UM, THERE IS MORE OF AN IMPROVEMENT IN RELIABILITY, UH, RELATIVE TO D-R-S-A-S.
UH, IN THIS CASE WE SEE AROUND FOUR GIGAWATTS OF PEAK LOAD REDUCTION UNDER BOTH THE WINTER STORM AND SUMMER HEATWAVE CASES, WHETHER BATTERIES ARE ELIGIBLE OR INELIGIBLE.
AND THEN IN TERMS OF TOTAL LOAD SHED ACROSS ALL CASES, UH, WE'RE REDUCING UNSERVED LOAD BY OVER HALF.
AND CORRESPONDINGLY YOU CAN SEE IN THE COST OF LOST LOAD, UM, WHICH IS MEASURED BY USING, UH, AVOL VALUE OF 35,000, UM, COSTS ARE REDUCED BY OVER 50% IN ALL THESE CASES.
UH, BUT NOW LOOKING AT THE AURORA CENTRAL CASE, UH, WE DO SEE SIMILAR OUTCOMES AS
[00:35:01]
FROM THE LTLF UH, DEMAND SCENARIO.IN THE LTLF CASE, WE SAW ABOUT FOUR GIGAWATTS OF PEAK LOAD, UH, LOAD SHED REDUCTION.
HERE WE'RE SEEING ABOUT FIVE GIGAWATTS.
AGAIN, THAT'S THAT LITTLE BIT OF ADDITIONAL CAPACITY THAT'S, UH, BEING INCENTIVIZED, UH, UNDER THE AURORA D-R-S-A-S, UH, OR THE AURORA CENTRAL DEMAND, UH, FORECAST.
UH, ONE THING YOU WILL KNOW IS THAT WHILE THERE IS UH, UH, AN IMPROVEMENT, UH, THERE'S STILL NOT A FULL, UH, REDUCTION, UH, IN LOAD SHED IN IN ANY OF THESE CASES.
SO THERE IS STILL SOME, UH, LOAD SHED OCCURRING.
AND THESE ARE WITHOUT THE DATA CENTER FLEXIBILITY, RIGHT? YES.
AND THIS IS IN, THAT'S OLIER JUST MENTIONED THAT THIS IS INDEPENDENT OF DATA CENTER FLEXIBILITY.
SO IF YOU WERE TO HAVE DATA CENTER FLEXIBILITY ON TOP OF THIS, THEN YOU COULD REACH A POINT WHERE, UH, YOU ARE FULLY REMOVING, UH, THE NEED TO SHED LOAD FORCIBLY.
UH, AND THE FINAL SLIDE HERE, UH, THAT I'LL, THAT I'LL WALK THROUGH IS JUST, UH, LOOKING AT THE EXTENDED ORDC AND WE'RE BACK TO THE ERCOT LTLF DEMAND SCENARIO HERE.
AND JUST AS WITH DRRS AS AN EXTENSION OF THE ORDC, UH, DOES IMPROVE RELIABILITY, UH, RELATIVE TO THE STATUS QUO.
SO LOOKING AT PEAK LOAD SHED, UH, UNDER WINTER STORM CONDITIONS, WE REDUCE PEAK LOAD SHED BY 1.5 GIGAWATTS, UH, UNDER A SUMMER HEAT WAVE, UH, STYLE, UH, CONDITION, WE REDUCE LOAD BY 1.7 GIGAWATTS.
UM, SO THE, UH, THE OUTCOME OF EXTENDED ORDC FALLING IN BETWEEN DRS AS IMPROVING RELIABILITY RELATIVE TO THAT CASE, UH, BUT NOT IMPROVING RELIABILITY AS MUCH, UH, AS THE DRS AS PLUS CASE.
UM, SO THIS CONCLUDES SORT OF THE EXECUTIVE SUMMARY, UH, PRESENTATION.
WE REALLY WANTED TO LEAVE THE MOST TIME FOR QUESTIONS.
I THINK GORDON HAS A QUEUE AND SO GORDON, I'LL LET, I'LL PASS IT ON TO YOU.
ALRIGHT, FIRST IN OUR QUEUE WE HAVE BILL BARNES.
A LOT OF INTERESTING OBSERVATIONS, UH, FOR SURE FROM THE REPORT.
I JUST WANTED TO CONFIRM ONE THING AND THEN I HAD A QUESTION.
IT, SO YOU GUYS BUILT YOUR OWN INDEPENDENT MODEL OR HAVE ONE YOU DIDN'T UTILIZE OR LEVERAGE ANY OF THE S SERVA MODEL THAT ERCOT HAS? THAT'S, THAT, THAT'S CORRECT.
SO WE, WE HAVE OUR OWN MODELING AND I'M CURIOUS, WAS THERE ANY BACK TESTING TO KIND OF VERIFY RESULTS FROM ERCOT SER MODEL TO THE RESULTS FOR, FROM YOUR STUDY TO SEE IF THEY'RE COMPARABLE? UM, THAT'S A GOOD QUESTION.
SO WE DID NOT RUN ANY COMPARISON TO ERCOT MODEL SPECIFICALLY.
NOW WE ARE CONSTANTLY DOING BACK CASTING AGAINST HISTORICAL DATA.
SO WE'LL RUN OUR MODEL VERSUS HISTORICAL CONDITIONS, COMPARE PRICING, MAKE SURE IT'S CONSISTENT.
UM, SO THE BACK CASTING IS DONE SORT OF ON A YEARLY BASIS TO MAKE SURE THERE'S ALIGNMENT.
WE HAVEN'T COMPARED IT SPECIFIC TO THE AIR CARD MODEL.
AND THEN QUESTION ON SLIDE SIX, WHICH IS, UH, LOAD FORECAST.
SO, UM, ONE THING THAT WE THOUGHT WAS VERY INTERESTING ABOUT YOUR ANALYSIS IS YOU HAVE A VERY CONSERVATIVE LOAD FORECAST, CONSERVATIVE MEANING VERY LOW, LIKE ON THE LOW END OF EVEN THE MOST PESSIMISTIC, UH, PROJECTIONS THAT WE'VE SEEN, WHICH I THINK HELPS FRAME THE, THE GOALPOSTS OF THE RESULTS OF THIS STUDY.
LIKE WE'RE STILL SEEING PRETTY SIGNIFICANT, UM, RELIABILITY ISSUES IN FIVE YEARS, EVEN WITH A VERY LOW FORECAST.
UM, WE WERE CURIOUS WHY ONE IS YOU GET TO ABOUT 105 MEGAWATTS OF PEAK DEMAND IN 2030, WHICH IS ACTUALLY BELOW THE AMOUNT OF SIGNED CONTRACTS AND, UH, LARGE LOADS THAT HAVE APPROVED PLANNING STUDIES TODAY, WHICH IS AROUND 22 GIGAWATTS.
SO, UM, A LITTLE BIT LOW, BUT IN THAT SAME VICINITY, BUT WHY IT FLATTENS AFTER 2030.
SO IT, THERE'S NOT A WHOLE LOT OF GROWTH BETWEEN 2030 AND 20, YOU KNOW, 33, 35.
WE WERE KIND OF CURIOUS WHAT PATTERNS YOU'RE SEEING THAT, UH, LED TO INFORM YOU THAT YOU, YOU THINK THE LOAD GROWTH IS GONNA GONNA DIMINISH AFTER THAT PERIOD OF TIME, WHICH IS PRESUMABLY WHEN ALL THIS NEW TRANSMISSION IS GONNA BE ENERGIZED AND LIKE IN N G'S VIEW, LIKE THAT'S WHEN THE LOAD, THE LOAD GROWTH IS REALLY GONNA START TICKING UP FASTER.
'CAUSE YOU'RE GONNA HAVE INFRASTRUCTURE IN PLACE, BUT I'M JUST CURIOUS IF YOU GUYS CAN COMMENT ON WHY YOU HAVE MORE OF A FLATTISH LOAD GROWTH AFTER 2030.
YEAH, THAT, THAT'S A GREAT QUESTION.
SO IT, SO IT, IT DOES KEEP GROWING AFTER THAT, RIGHT? JUST TO, TO BE FAIR, WE CUT IT OUT AT 2023, AT 2033.
[00:40:01]
I THINK THIS IS A BIT OF A LULL, BUT IT, OUR, OUR LOAD GROWTH FORECAST DOES GO UP TO, UH, FROM MEMORY ABOUT 160 GIGAWATT BY 2050.SO, SO THERE'S GROWTH AFTER THAT.
WE'RE, WE'RE NOT SEEING THE AIRCRAFT MARKET JUST STOPS.
UM, I THINK THE, THE DYNAMIC HERE YOU'RE SEEING IS WE'RE SEEING THE BIG SURGE OF DATA CENTER DEMAND REALLY COMING IN IN THE NEXT FOUR OR FIVE YEARS AND THEN SLOWING DOWN A LITTLE BIT FROM THAT.
UH, AND, AND I GUESS OBVIOUSLY THAT, THAT'S A POINT THAT WE CAN DEBATE, RIGHT? UH, IN TERMS OF HOW MUCH IS SHORT-TERM GROWTH VERSUS, YOU KNOW, HOW MUCH LONG-TERM GROWTH WE'RE GONNA SEE ADDED TO THAT.
UH, BUT THERE IS LONG-TERM GROWTH AFTER THE SORT OF 2033 TIMEFRAME AS WELL.
I THINK THERE'S MORE OF A, OF A SORT OF SMALL LULL FROM THE MODELING MYTHOLOGY THAT HAPPENS OVER THOSE TWO YEARS, BUT IT, IT CONTINUES AFTER THAT.
AND THEN ON THE DATA CENTER FLEXIBILITY, YOU GUYS GOT AROUND 60%, IS, IS WHAT YOU'VE DETERMINED REQUIRED IN TERMS OF CURTAIL ABILITY, UM, ARE HOW MUCH OF THAT, OR IS THE ASSUMPTION THE VAST MAJORITY OF THAT CURTAIL CURTAIL ABILITY IS DUE TO ONSITE GENERATION OR BACKUP GENERATION AT THE LOAD OR VERSUS THE LOAD JUST BEING ABLE TO SHUT OFF? YEAH, I THINK, I MEAN, WE CAN, SOMEONE CAN COMMENT IN MORE DETAIL.
I THINK THE, OUR ASSUMPTION IS THAT THE VAST MAJORITY, IF NOT ALL DATA CENTERS WILL HAVE SOME LEVEL OF BACKUP GENERATION, AND FOR THE MOST PART WE'LL HAVE BACKUP GENERATION ON SITE.
SO THE ASSUMPTION IS THAT THAT FLEXIBILITY WOULD COME FROM THE BACKUP GENERATION TURNING ON RATHER THAN, UM, RATHER THAN THEM SHUTTING DOWN FULLY.
AND MAYBE DO YOU WANT TO COMMENT MORE ON THAT? NO, THAT, THAT'S ABSOLUTELY RIGHT.
JUST TO ADD ON, THERE ARE DIFFERENT WAYS THAT THEY COULD REDUCE LOAD, SO TEMPORAL OR GEOSPATIAL LOAD SHIFTING, UH, THROUGH ALL OUR INTERVIEWS, WE THINK THAT THE BACKUP GENERATION IS GONNA BE THE MOST LIKELY SOURCE THOUGH.
BLAKE HOLT, LCRA, UH, FIRST OF ALL, WANT TO RECOGNIZE THE GOOD WORK, AURORA, VERY DETAILED, VERY THOROUGH REPORT.
BUT I DO WANT TO TOUCH ON SOME COMMENTS I HAD AT PRS.
AND THIS IS KIND OF AT A, A HIGH LEVEL AND IN TERMS OF THE RESOURCE ADEQUACY SOLUTION OR, UH, DESIGN THAT AURORA PRESENTED VERSUS NPR 1310 THAT WE'VE SEEN COME FROM ERCOT, UH, NUMBER ONE, CONFUSED WHY BOTH OF THEM ARE BEING CALLED DRS PLUS BECAUSE THE DESIGNS ARE FUNDAMENTALLY DIFFERENT.
AURORA'S OPTION IS, HAS REVENUE BASED ON A BUDGET WITH LINKAGE TO CONE AND IS A STANDBY PAYMENT NPR 1310 BASED ON MY UNDERSTANDING, REVENUE IS DERIVED FROM COMPETITIVELY, UH, SUBMITTED OFFERS.
AND AS I SEE THAT, AS THOSE DESIGNS IS FUNDAMENTALLY DIFFERENT, I I WANT TO BE CONVINCED OF THE USEFULNESS OF, OF LOOKING AT THE RESULTS OF DRS PLUS IN, IN THIS CONTEXT, UH, SO THAT TO THE EXTENT THAT THAT CLARITY CAN BE GIVEN THROUGHOUT THIS CONVERSATION, THAT WOULD BE HELPFUL IN DETERMINING OUR OPINION ON, UH, A PATH FORWARD.
SO LOOKING FORWARD TO HEARING ABOUT THAT.
UH, I DO UNDERSTAND THE CONSTRAINTS THAT AURORA WAS UNDER IN, IN PUTTING TOGETHER RE THE REPORT AHEAD OF NPR 1310 PUT TO, UH, SUBMITTAL.
BUT THAT'S, UH, THE QUESTIONS AND COMMENTS THAT I HAVE AT THIS POINT.
UM, I THINK I, MAYBE I'LL ADDRESS THE PART ON THE MODELING SORT OF DIRECTIONALLY, AND THEN KEITH, I'LL, I'LL LEAVE THE NPR SPECIFICS TO YOU, UM, SINCE WE DIDN'T WRITE IT.
UM, SO ON, ON THE MODELING DIRECTIONALLY RIGHT, I THINK FROM A MODELING PERSPECTIVE, IT IS ALSO ALWAYS A LITTLE DIFFICULT TO CAPTURE ALL THE NUANCES OF, OF WHAT IS GOING ON, WHAT IS GOING INTO AN NPR.
AND I WOULD SAY THE MOST IMPORTANT PART IN THE MODELING AT THE END OF THE DAY IN TERMS OF HOW MUCH CAPACITY YOU BUILD IS THE AMOUNT OF REVENUE THAT THOSE PLANTS WILL BE SEEING, RIGHT? AND THE AMOUNT OF CERTAINTY THAT THIS REVENUE HAS FOR DIFFERENT TECHNOLOGIES, RIGHT? SO FROM LOOKING AT A PURE RELIABILITY EFFECT OF THESE DIFFERENT MARKET MECHANISMS, AND OBVIOUSLY THEN THE EFFECT ON SYSTEM COSTS, I THINK DIRECTIONALLY, UM, THEY'RE, THEY'RE FAIRLY ALIGNED.
AND SO OBVI YOU POINT OUT SOME DIFFERENCES THAT ARE, I THINK ARE NOT INSIGNIFICANT, BUT, BUT I THINK IN TERMS OF THINKING ABOUT THE SORT OF IMPACT ON INCENTIVIZING ADDITIONAL GENERATION AND THAT IMPACT OF THAT THIS ADDITIONAL GENERATION WILL HAVE ON RELIABILITY, THERE IS CONSISTENCY BETWEEN, BETWEEN THOSE TWO THINGS.
UM, NOW MAYBE I'LL LET KEITH COMMENT ON THE NPR SPECIFICALLY.
YEAH, I I THINK OBVIOUSLY THERE'LL BE MORE DISCUSSION ON, UH, 1309 AND 1310, UM, NEXT MONTH.
BUT JUST TO, TO COMMENT ON, ON BLAKE'S POINT HERE, UM, I THINK THE, THE BIG CHALLENGE WE HAD
[00:45:01]
WAS THE NPR DESIGN WAS NOT NAILED DOWN WHEN, WHEN THEY STARTED THE ANALYSIS.UM, BUT I THINK IT'S IMPORTANT TO NOTE THAT THAT AS, AS OLIVIER WAS SAYING, THE, THE, SOME OF WHAT YOU'RE TALKING ABOUT IS THE 1310 DOESN'T CREATE A REQUIREMENT, RIGHT? SO IT HAS THE RELEASE FACTOR.
AND SO IN ORDER TO STUDY THE EFFECTS ON THE NEW GENER ON THE GENERATION AND ITS IMPACT ON RELIABILITY, THEY HAD TO BUILD THAT WITHIN THE MODEL ITSELF.
AND THAT'S WHY YOU SEE DEMAND CURVE AND BUDGET AND ALL THOSE OTHER THINGS AND, AND ALL THESE OTHER LIMITATIONS THAT THEY'VE, THEY'VE BUILT INTO IT BECAUSE, UM, THEY HAD TO STUDY THAT.
WHEREAS THE RELEASE FACTOR IS NOT THE, THE FUNCTION OF THE RELEASE FACTOR IS, IS HOW IT MECHANICALLY WORKS WITHIN THE MARKET SYSTEMS ITSELF, RIGHT? SO THERE WILL BE A REQUIREMENT THAT COMES SUBSEQUENT, BUT IN ORDER TO DO THE STUDY, THEY NEEDED TO ASSESS WHAT THE RELIABILITY WAS, HOW TO IMPROVE THAT RELIABILITY, AND HOW TO ACHIEVE THAT THROUGH THE MODEL.
SO THERE IS A PARTICULAR MODELING DIFFERENCE THERE THAT WE ACKNOWLEDGE.
AND, YOU KNOW, JUST AS A, AS A QUICK ANECDOTE, ONE OF THE, THE THINGS THAT S AND I DID VERY EARLY ON IN, IN OUR ANALYSIS OF, OF DRS PERHAPS EVEN A YEAR AGO WAS WE, WE WANTED TO TRY TO UNDERSTAND THE, THE IMPACT ON PRICING, PARTICULARLY THE PRICING SHAPE OF, OF A DRS PRODUCT.
AND THE ONLY WAY TO DO THAT WAS WE TOOK THE DAY AHEAD MODEL AND WE SAID, WELL, HOW DO WE MODIFY THE DAY AHEAD MODEL TO GET WHAT WE WANT? AND WE SAID, WELL, LET'S INCREASE THE NONS SPIND REQUIREMENT.
AND SO IF YOU WERE TO COME TO ME AND SAY, WELL, KEITH, YOU'RE NOT INCREASING THE NONS SPIND REQUIREMENT.
AND THE ANSWER IS, WELL, YES, WE'RE NOT INCREASING THE NONS SPIND REQUIREMENT, BUT WE NEEDED A WAY TO GET THE MODELING TO WORK TO APPRECIATE THE RESULTS OF, OF THE IMPACT.
SO THAT'S REALLY WHAT WE'VE DONE HERE, IS TO UNDERSTAND WHAT IS THE IMPACT ON RELIABILITY, WHAT IS THE IMPACT ON THAT WITHOUT HAVING THOSE MECHANISMS THAT, UH, PARTICULARLY THE REQUIREMENT WHICH WE HAVE NOT BEEN ABLE TO ASSESS YET, BUT WILL OCCUR WITH THE RELIABILITY STUDY AS AS THAT MOVES FORWARD.
SO THAT'S WHY I THINK YOU'RE DOING IT.
PLUS, I THINK THE ONE OTHER THING THAT THIS STUDY DOES THAT'S DIFFERENT FROM THE, THE FINAL DESIGN ON, ON 1310 IS THAT 1310 IS, IS BOTH THE ANCILLARY SERVICE SCENARIO AND THE AS PLUS SIMULTANEOUSLY.
AND WHAT WE ASKED AURORA TO DO WAS TO STUDY THEM INDEPENDENTLY, RIGHT? AND SO THAT'S WHAT THAT RELEASE FACTOR ALLOWS US TO DO, IS TO ACHIEVE BOTH THE OPERATIONAL AND THE RESOURCE MECHANISM AT THE SAME TIME, WHEREAS THE STUDY SUFFERED, SEPARATED THE EFFECTS TO IDENTIFY AND ISOLATE WHAT THOSE IMPACTS ARE.
SO THAT'S WHY SOME OF THE REASONS WHY THAT THEY'RE, THEY'RE NOT IDENTICAL, UM, IN, IN THE STUDY AND IN THE NPR.
AND LAST NPRR RELATED QUESTION, I PROMISE.
UM, ARE THERE ANY PLANS AT THIS TIME TO PIVOT 1310 TO MORE OF AN AURORA DESIGN APPROACH? IS THAT SOMETHING THAT COULD POTENTIALLY HAPPEN DURING RESOURCE ADEQUACY OR RELIABILITY STANDARD REVIEW? UM, OR IS 1310 ENVISIONED TO STAY AS DESIGNED A COMPETITIVELY OFFERED AND CLEARED PRODUCT? UH, WE LIKE, WE LIKE THE DESIGN ON, ON 1310.
UH, WE'LL BE TALKING ABOUT THE REASONS FOR THAT, OBVIOUSLY DURING THE STAKEHOLDER DISCUSSION, BUT WE DO THINK IT ACHIEVES THE SAME, UH, THE SAME LEVEL OF RESULTS AS, AS YOU GET IN, IN THE AURORA STUDY.
SO WE, WE THINK IT'S REALLY GOOD AND, UM, WE, WE INTEND TO PROMOTE THAT.
I SAW YOU HAD A, A QUICK QUESTION ON THE, UH, RELEASE FACTOR.
I JUST WANNA MAKE SURE THAT, THAT WE, OUR CONVERSATION TODAY IS LARGELY FOCUSED ON THE AURORA REPORT RATHER THAN DELVING INTO THE DETAILS OF, OF 1310.
SO WITH THAT IN MIND, HAPPY TO YEP.
SO ANDREW REER, IMM, UH, JUST TO MAKE SURE I UNDERSTAND WHAT KEITH WAS JUST SAYING, I MEAN, SO IS IT, ARE WE EFFECTIVELY SAYING THAT AURORA'S VERSION OF DRS HAS A RELEASE FACTOR OF A HUNDRED PERCENT, IT'S A COMPLETELY DISTINCT THING THAT CLEARS ON ITS OWN AND ALL OF THAT CAPACITY IS ABLE TO PROVIDE OPERATING RESERVES AND THERE'S NO CONSTRAINT BETWEEN THOSE TWO THINGS? THAT THAT'S HOW WE MODELED IT.
ALRIGHT, NEXT IN THE QUEUE IS NED.
AND, AND ONCE AGAIN, UM, APPRECIATE ALL THE WORK THAT YOU'VE PUT INTO THIS.
IT'S CLEARLY, UH, WAS A, WAS A PRETTY BIG AND THOROUGH LIFT.
UM, OTHERS HAVE, HAVE TOUCHED ON SOME OF THE QUESTIONS THAT I HAD, SO I'LL TRY NOT TO REPEAT THEM EXPLICITLY.
[00:50:01]
BUT JUST TO BE, UM, GO DOWN THE LIST.UM, FIRST OFF IS IN YOUR EXTENDED OADC SCENARIO, DID YOU LOOK ONLY AT WHAT WE'LL CALL THE CLASSIC ORDC VERSUS THE NOW DISAGGREGATED ANCILLARY SERVICE DEMAND CURVES? UH, FURTHER, UH, REAL-TIME CO OPTIMIZATIONS, UH, PRODUCTION? THAT'S A GREAT POINT.
UM, IT WAS MODELED STILL UNDER THE ORDC CONSTRUCT, UM, WHICH WOULD LIKELY WILL FUNCTION FAIRLY SIMILARLY UNDER THE A SDC.
UH, BUT WE DID, I DON'T KNOW, TIM, IF WE WANT TO ADD ANYTHING, WE, WE DID MODEL IT UNDER THE OR DC CONSTRUCT IN TERMS OF THE, THE SCAR STATE PRICING.
THAT, THAT, THAT'S WHAT I ASSUMED AND LIKE AGREE SHOULD PROBABLY GET SOMEWHERE IN THE SAME, SAME BALLPARK.
BUT WANTED TO, TO CONFIRM, UM, ANOTHER QUESTION I HAD AND IS, SO YOU LOOKED AT THE DIFFERENT WEATHER YEARS TO KIND OF STRESS TEST, UH, AGAINST THE, YOU HAD THE ERCOT LONG TERM LOAD FORECAST, YOU HAD YOUR BASE ONE AND YOU DID THE STRESS TESTING, BUT IN YOUR BASE FORECAST, IS THERE A SPECIFIC MODEL YEAR THAT OR, UH, WEATHER YEAR THAT THAT IS ANCHORED TO, OR IS IT A MIXTURE OF FACTORS? HOW DID YOU KIND OF NORMALIZE THAT? YEAH, GREAT, GREAT POINT.
MAYBE YOU WEREN'T CLEAR ENOUGH.
SO THE, OUR BASE FORECAST IS BASED ON THE 2013 WEATHER YEAR, WHICH I BELIEVE IS ALSO THE YEAR ERCOT USES FOR, FOR THEIR PLANNING STUDY.
UM, MAYBE GORDON, I DUNNO IF YOU CAN CONFIRM, BUT UM, BUT YEAH, THAT'S A 20 20 13 WEATHER YEAR.
THANK YOU FOR CLARIFYING THAT.
UM, THEN ON THE, UH, FLEXIBLE LOAD, YOU HAD A LOT OF FOCUS ON DATA CENTERS AS BEING FLEXIBLE, BUT DID, UH, WAS THE, WAS THAT THE ONLY LOAD THAT YOU CONSIDERED AS FLEXIBLE OR DID YOU ALSO BAKE IN LIKE, UH, CRYPTO LOADS, MAYBE OIL AND GAS OR, UM, OTHER INDUSTRIAL LOADS THAT, THAT MIGHT HAVE, UH, FLEXIBLE CHARACTERISTICS? YEAH, GREAT QUESTION.
SO INHERENTLY, ANYTHING THAT ALREADY, UM, EXISTS IN ERCOT, YOU KNOW, SO SUCH AS, UH, CRYPTO MINES, UM, OIL AND GAS ELECTRIFICATION IS INCLUDED IN KIND OF OUR BASE LOAD SHAPE, UM, AND UNDERLYING ASSUMPTIONS.
UH, AND FOR THE FORECAST, UH, MOST OF THAT LOAD FLEXIBILITY WOULD BE WHAT IS CONSIDERED LARGE FLEXIBLE LOADS IN THAT LONG-TERM LOAD FORECAST.
SO, UH, I BELIEVE THE, THE UNDERLYING ASSUMPTION IS THAT MOST OF THAT IS DATA CENTERS, BUT THE EXISTING INDUSTRIAL CUSTOMERS, CRYPTO MINES WOULD ALREADY BE INCLUDED AND THE UNDERLYING MODEL ASSUMPTIONS.
AND SO THEY'RE, THEY'RE BUILT INTO THE, THE BASELINE LOAD, BUT IN THE EXTREME SCENARIOS WHERE THEN YOU'RE, YOU KNOW, LOOKING AT THE FLEXIBILITY OF DATA CENTERS TO, YOU KNOW, HOW MUCH THAT ADDRESSES THE, YOU KNOW, THE, THE RESOURCE ADEQUACY, UH, OUTCOME THERE.
UM, WERE YOU ALSO TAKING INTO ACCOUNT THE FLEXIBILITY OF THOSE? SO LIKE I KNOW ERCOT HAS SOME ASSUMPTIONS THAT, THAT THEY USE FOR, CERTAINLY FOR, UH, CRYPTO LOADS AND I, I, I'D BE LYING IF I COULD RATTLE OFF ALL OF THE ASSUMPTIONS THAT WERE IN THE MOST RECENT CDR, BUT UM, DO YOU HAVE DIFFERENT FACTORS FOR EACH OF THOSE? SO, SO I GUESS MAYBE A HELPFUL FRAMEWORK IS THIS SORT OF TWO CATEGORIES.
THE, THE FIRST CATEGORY WAS PEYTON WAS MENTIONING IS, YOU KNOW, IN, IN THE HISTORICAL DATA WE DO SEE ALREADY SOME DEMAND RESPONSE BAKED IN AROUND, YOU KNOW, KNOW RESPONSES TO FOUR CB THINGS LIKE THAT THAT WE DO TAKE INTO ACCOUNT BECAUSE WE CALIBRATE EVERYTHING AGAINST HISTORICAL DYNAMICS.
UM, THE OTHER PART WE'VE CARVED OUT IS WHAT I WOULD CALL SORT OF EMERGENCY RESPONSE, YOU KNOW, AND SO ERS ANCILLARY SERVICE RESPONSE, SO WE DO THE MODEL DO DOES, YOU KNOW, CALL ONTO THESE RESOURCES WHEN WE'RE LOOKING AT LOAD CHAINING.
UH, AND SO, SO THERE IS A FEW GIGAWATTS OF DEMAND RESPONSE THAT IS ALSO TRIGGERED BEFORE EVEN THE DATA CENTER DEMAND RESPONSE COMES IN.
BUT IT, IT'S SORT OF MORE STATUS QUO, RIGHT? SO WE LOOKED AT WHAT IS ACTUALLY HAPPENING IN THE
AND SO WE TOOK THIS SORT OF THIS AMOUNT OF DEMAND RESPONSE THAT THE MARKET ALREADY PRODUCES AND THEN PROJECTED IT FORWARD, OBVIOUSLY SCALE IT TO SORT OF THE, UH, GROWTH IN LOAD THAT WE'RE SEEING AND THEN LOOKED AT THE DATA CENTER FLEXIBILITY ON TOP OF THAT.
UM, BUT OBVIOUSLY IN TERMS OF MAGNITUDE, YOU'RE LOOKING AT MUCH MORE RESPONSE FROM DATA CENTERS POTENTIALLY THAN FROM THIS SORT OF HISTORICAL DEMAND RESPONSE YOU ALREADY SEE IN ERCOT.
UM, ANYTHING, UM, AND THE ANOTHER QUESTION IS ON UM, RETIREMENTS, I KNOW THIS WASN'T ON, IT MIGHT'VE BEEN ONE OF THOSE PAGE 71 SLIDE, UH, OR SLIDE 71 FOOTNOTE THREE KIND OF QUESTIONS, BUT I KNOW YOU HAD A, UH, A LIST OF, UM, YOU KNOW, ASSUMED RETIREMENTS IN THERE.
YOU KNOW, SORRY, DON'T NEED TO, IT'S GIVING EVERYONE SPECIFICS.
I DON'T, I DON'T EVEN KNOW IF THAT WAS THE RIGHT ONE.
[00:55:01]
UM, SO IN THE DIFFERENT SCENARIOS, WERE THOSE TAKEN AS GIVEN OR DID THOSE ADJUST TO THE CHANGE CHANGING MARKET SIGNALS THAT THAT SHOWED UP? YEAH, IT'S A GOOD QUESTION.SO NO, WE KEPT THOSE FLAT ACROSS SCENARIOS, UH, WHICH I ACKNOWLEDGE, RIGHT? THERE'S A WORLD WHERE IF YOU SEE A LOT OF LOAD GROWTH, ERCOT WOULD, MIGHT, MIGHT WANT TO KEEP SOME OF THESE PLANTS ONLINE TO ALLEVIATE SOME OF THE RELIABILITY CONCERNS MM-HMM
AND SO IN THAT CASE THAT CAN HELP MAYBE BRIDGE SOME OF THE GAP THAT WE'RE SEEING IN SOME OF THE SCENARIOS.
BUT, BUT THOSE WERE IDENTICAL ACROSS SCENARIOS.
UM, SO YOU AS WE'RE KIND OF MENTALLY HANDICAPPING AS WE'RE WE'RE WALKING THROUGH THIS, THAT'S ONE OF THE DYNAMICS TO KIND OF THINK ABOUT IS, YOU KNOW, YOU MAY, YOU MAY IN EACH ONE GET SOME INCREMENTAL, UH, GENERATION, BUT YOU ALSO MAY RETAIN EXISTING GENERATION FOR FOR LONGER AND YOU'LL GET A MIX OF OUTCOMES.
BUT I, I RECOGNIZE THAT IN ORDER TO SEPARATE THE EFFECTS THAT YOU KIND OF HAVE TO MAKE A CUT AND, AND MOVE FORWARD IN A MODEL.
AND THEN, UM, ON THE, WELL I'LL MAKE THIS THE LAST ONE, WAS THERE AN ANCILLARY SERVICE DEMAND CURVE FOR DR FOR DRRS, UM, CALLED PLAIN DRRS THAT Y'ALL USED? 'CAUSE I KNOW THAT WAS ONE OF THE THINGS THAT WAS INCLUDED IN NPR 1309 AND IT LOOKED LIKE THE DEMAND CURVE Y'ALL HAD FOR DRS PLUS WAS MAYBE DIFFERENT FROM WHAT'S IN 1310.
WE DON'T NEED TO, YOU KNOW, WALK THROUGH ALL THOSE DIFFERENCES NOW.
I WAS JUST CURIOUS IF YOU HAD A DEMAND CURVE IN, UH, THE REGULAR DRRS.
UH, SO IN, YOU MEAN IN THE SORT OF CLASSIC PLAIN ORIGINAL YEAH, IT'S HARD TO FIND THE NAMES.
UM, WE DID NOT HAVE A DEMAND CURVE FOR THAT SERVICE.
IT WAS, IT WAS CLEARED MORE THE, THERE'S A CLASSIC ANCILLARY SERVICE, SO MORE MODELED AFTER LIKE WHAT A NONS SPIN TYPICAL SERVICE WOULD BE.
AND IN THE DS PLUS SCENARIO, IS THAT DEMAND CURVE CO OPTIMIZED AT ALL WITH, UH, ENERGY AND ANCILLARY SERVICES IN REAL TIME OR IS IT YES.
ALRIGHT, I THINK I'VE GOT EVERYTHING OUT.
UM, I'M GONNA STAND UP JUST 'CAUSE I'M IN THE BACK ROW.
UH, I WANNA GO BACK TO ONE OF THE QUESTIONS NED ASKED ON THE FLEXIBILITY 'CAUSE I DIDN'T UNDERSTAND YOUR ANSWER, SO, UH, I APOLOGIZE FOR NOT UNDERSTANDING.
UM, UM, IS, AS I LOOK AT SLIDE OR THE PICTURE ON 10, WHICH SHOWS THE COMPONENTS OF LARGE LOADS, I SEE EIGHT GIGAWATTS OF CRYPTO.
AND MY QUESTION IS, IS THOSE EIGHT GIGAWATT OR ALL THOSE EIGHT GIGAWATTS ALREADY ASSUMED TO BE CURTAILED UNDER THESE STRESSFUL CONDITIONS SO THAT, SO AND SORRY, THAT'S PROBABLY US NOT BEING CLEAR ENOUGH.
UM, SO CRYPTO LOAD IN OUR MODEL IS ACTUALLY TREATED VERY SPECIFICALLY AND IS FULLY PRICED RESPONSIVE.
UH, SO THAT'S SORT OF A CARVE OUT OF THE DATA CENTER PIECE THAT YOU'VE SEEN.
CRYPTO LOAD ESSENTIALLY STOPS, YOU KNOW, TAKING DEMAND ONCE PRICES ARE ABOVE, I CAN'T REMEMBER THE STRIKE PRICE, A HUNDRED BUCKS, MAYBE A LITTLE BIT LESS, WE CAN PULL THE STRIKE PRICE, BUT ESSENTIALLY WHEN MINERS ARE GONNA LOSE MONEY, THEY STOP.
AND SO IT'S FULLY IR RESPONSIVE.
THAT'S THE ONLY ONE THAT'S ACTUALLY FULLY RESPONSIVE.
UM, AND THEN THE, I GUESS THE OTHER TWO CATEGORIES WOULD BE LIKE THE DEMAND RESPONSE WE GET FROM, YOU KNOW, RESIDENTIAL INDUSTRIAL LOAD, MOST OF IT COMING FROM MAYBE FOUR CP AVOIDANCE, UH, TYPE BEHAVIOR.
AND THEN THE DATA CENTER, WHICH IS COVERED UNDER SENATE BILL SIX.
SO I'M GONNA TRY TO REPEAT BACK, JUST TO CONFIRM, UM, I THINK WHAT I HEARD YOU SAY IS, ALL RIGHT, SO DATA CENTER'S BIGGEST THING, WE CARVE THEM OUT.
WE HAVE A DIFFERENT SET OF, OF, OF, UH, OF EXPECTED CHARACTERISTICS FOR THEM.
EVERYBODY CRYPTO HAS ITS OWN SET OF ASSUMPTIONS AND THEY'RE GONNA BE GONE AT SOME RELATIVELY LOW PRICE.
AND THEN THE OTHER BLOCKS THAT ARE SHOWN HERE, OTHER INDUS OR SORT OF LARGE LOAD BLOCKS SHOWN HERE ARE THEN WRAPPED UP INTO THE ASSUMPTIONS YOU MADE LOOKING AT CURRENT, UH, DEMAND RESPONSE AND THEN SORT OF GREW THEM THROUGH TIME PROPORTIONALLY, I'M SEEING THREE HEAD NODS, WHICH I, I ASSUME I GOT IT RIGHT.
YEAH, I THINK IT'S A GOOD SUMMARY.
THANK YOU FOR MAKING IT CLEAR FOR EVERYONE.
ALRIGHT, ERIC, OFF, UM, I'VE GOT A FEW QUESTIONS, I APOLOGIZE, BUT I THINK THAT'S WHAT IT'S FOR.
SO I WANNA APOLOGIZE TOO MUCH.
I, SO I GUESS FIRST OFF, THERE WAS THE DISCUSSION EARLIER
[01:00:01]
ABOUT HOW, UM, WHAT WAS MODELED AS EXACTLY REFLECTED IN THE NBRR.UM, AND SO TO, TO ERCOT AND POTENTIALLY TO AURORA IF YOU FEEL LIKE IT, BUT I THINK IT'S PRIMARILY TO ERCOT.
UM, DOES THE AURORA OF MODELING 100% REFLECT WHAT YOU THINK 1310S IMPACTS WILL BE? OR 90% WHAT'S, WHAT ARE THE GAPS IN THE MODELING BETWEEN WHAT WAS MODELED AND WHAT YOU THINK WILL BE IN THE NPRR? YEAH, WE, WE THINK IT'S, WE THINK THE RESULTS ARE ARE PRETTY, PRETTY SOLID.
UM, SO, UM, ON SLIDE, UM, 14, UM, IF YOU CAN GO TO THAT, IS IT THIS ONE? YES.
SO WILL 1310 PROCURE THE 80 TO 140 GIGAWATTS? SO THAT, THAT'S A LITTLE HARDER FOR US TO SAY ONLY BECAUSE WE DO NOT HAVE, AT LEAST IN HOW WE'RE LOOKING AT IT, THIS ASSUMES A LEVEL OF OTHER SUPPLY GROWTH IN THERE IN TERMS OF, SO WHEN THEY'RE SHOWING THEIR NUMBERS, THERE'S THE STATUS QUO GROWTH, RIGHT? AND THEN THERE'S THE, THE ADDITIONAL GROWTH.
SO, SO THE POINT BEING IS WE DON'T HAVE AN INDEPENDENT MODELING OF THAT GROWTH.
SO, SO TO SAY THAT'S THE PARTICULAR NUMBER OR NOT IS NOT ENTIRELY CLEAR TO ME AT THIS POINT.
SO I GUESS THEN FOLLOW UP BASED ON THAT ANSWER TO AURORA, IS 80 TO 140 SUPPOSED TO REPRESENT 100% OF THE ELIGIBLE CAPACITY? THE UM, SO YES.
AND, AND MAYBE IT'S HELPFUL TO CLARIFY THE MODELING HERE A LITTLE BIT.
SO IN THE STATUS QUO SCENARIO, WITH NO MARKET DESIGN CHANGES, WE STILL THINK PEOPLE WILL INVEST IN NEW GENERATION, RIGHT? UM, PEOPLE HAVE BEEN INVESTING IN NEW GENERATION IN ERCOT FOR A LONG TIME, UH, AND THERE ARE CURRENT MARKET SIGNAL WITH LOAD GROWTH THAT ARE THERE FOR PEOPLE TO INVEST IN THAT GENERATION.
SO WE SEE THAT POOL OF GENERATION GROWING IN BOTH CASES AND OBVIOUSLY IN THE LTLF CASE MUCH MORE THAN IN THE OTHER CASE.
AND THE MODELING ON THE IMPACT OF DRS PLUS IS ON SORT OF THE ADDITIONAL GENERATION THAT GETS INCENTIVIZED BY JUST THAT MECHANISM.
AND SO THAT 80 TO 100 GIGAWATT PROCUREMENT IS MEANT TO LOOK AT SORT OF THE POOL OF GENERATION THAT EXISTS IN THAT GIVEN YEAR AND MODULATE ACROSS THE HOURS OF HIGHEST SYSTEM RISK, WHERE THAT PROCUREMENT WOULD BE SORT OF HIGHEST VERSUS THE HOURS OF LOWEST SYSTEM RISK WHERE PROCUREMENT IS ESSENTIALLY, YOU KNOW, THAT SORT OF 80 GIGAWATT NUMBER.
UM, THAT RANGE OBVIOUSLY IS BEING MORE, IS, IS FAIRLY BROAD AND YOU HAVE TO LOOK AT HOURLY PROCUREMENT.
BUT JUST TO REPEAT MY QUESTION, DOES THAT REPRESENT 100% OF ELIGIBLE CAPACITY? UM, I GUESS WAS THE PROCUREMENT DERATED OR WAS IT YEAH, IT'S NOT AVAILABLE THAT PORT IF IT, IT CONSIDERS OUTAGES AND THINGS LIKE THIS.
DO YOU WANNA SAY IN THE MICROPHONE? YEAH, ERIC, SO, SO BASICALLY IT'S LOOKING AT WHAT WOULD BE ELIGIBLE OR AVAILABLE TO THE SYSTEM IN THAT HOUR, RIGHT? SO IF, IF, YOU KNOW, THE SYSTEM HAS A HUNDRED GIGAWATTS OF THERMAL, YOU KNOW, CAPACITY UHHUH AND 15 OF IT IS ON OUTAGE AND THE OTHER, UH, 85 IS AVAILABLE, YOU KNOW THAT 85 WOULD BE REMUNERATED UNDER THE WAY.
SO I THINK THE ANSWER TO THAT IS YES, IT'S 100% OF ELIGIBLE CAPACITY.
IS THAT WHAT Y'ALL PLAN TO PROCURE UNDER 1310? WE DON'T KNOW WHAT THE REQUIREMENT IS YET UNTIL THE, UM, RELIABILITY STUDY IS DONE.
HOW CAN YOU BE CONFIDENT THAT THIS REPRESENTS WHAT YOU EXACTLY, WHAT YOU INTEND TO PROCURE AT 1310? WE, WE FEEL CONFIDENT THAT IT IS, IT IS GONNA GET WHAT WE NEED.
SO IF YOU DON'T KNOW IF IT'LL PAY EVERY ELIGIBLE, UM, FACILITY, UM, YET HOW CAN YOU COMPARE COSTS AT ALL OR, OR BE CONFIDENT IN YOUR ANSWER? TO MY EARLIER QUESTION, WE HAVE STRONG INDICATION OF WHAT WE THINK IS LIKELY TO BE THE CASE, BUT, OKAY.
I'M NOT SAYING SO IT, IT SEEMS LIKE, I'M SORRY, I'M NOT, I CAN'T HELP BUT BE COMBATIVE IN THE CONVERSATION, BUT I APOLOGIZE FOR THAT.
SO, UM, IT'S, IT SEEMS LIKE IF YOU BELIEVE THAT THIS ACCURATELY REPRESENTS 1310 MM-HMM
[01:05:01]
THEN YOU THINK IT'S LIKELY THAT YOU'LL PROCURE A HUNDRED PERCENT OF AVAILABLE CAPACITY, ELIGIBLE CAPACITY? IT'S, IT'S LIKELY FOR SURE.SO YOU WOULDN'T PROCURE 100% OF ELIGIBLE CAPACITY UNDER THE, WHAT YOU CALLED EARLIER, THE ORIGINAL DRRS YOU WOULD PROCURE BASED ON THE ENTRY SERVICES PLAN? IT'S BASED ON THE REQUIREMENT.
SO IS DRRS ANCILLARY SERVICE PLUS AND ANCILLARY SERVICE? YES.
AND SO I DON'T UNDERSTAND HOW AN ANCILLARY SERVICE COULD EVERY PROCURE 100% OF ELIGIBLE CAPACITY.
IT SEEMS LIKE THAT'S A STRETCH TO CALL IT A ARY SERVICE.
NO, WE VIEW IT AS A ARY SERVICE.
I'M LOOKING FORWARD TO FINDING OUT MORE ABOUT THAT.
BUT 13TH 10 IS NOT BEING DISCUSSED RIGHT NOW.
UM, SO, UM, COULD WE GO TO SLIDE FIVE? SO ON THIS SLIDE YOU POINTED OUT THAT ERCOT IS SEEING, UH, FASTER LOAD GROWTH THAN ANY OTHER MARKET AND HAS HAD HISTORICAL OPERATIONAL RESERVES THAT BEEN LOWER THAN OTHER ISOS.
DO YOU THINK THERE'S A RELATIONSHIP BETWEEN THOSE TWO THINGS OR IS THAT A COINCIDENCE? I MEAN, YEAH, THERE'S CERTAINLY A RELATIONSHIP, RIGHT? UM, WHAT DO YOU THINK IT IS
UHHUH A LITTLE BIT, RIGHT? UHHUH
UH, MAYBE THINGS THAT, UM, THAT YOU CAN SEE IN OTHER MARKETS WHERE THERE, THERE'S PRICE SIGNALS THAT CAN INCENTIVIZE ADDITIONAL GENERATION TO, TO BE THERE UHHUH, UH, TO PROVIDE ADDITIONAL RELIABILITY BECAUSE, BECAUSE IN ERCOT YOU DON'T GET PAID FOR IT, RIGHT? RIGHT.
AND, AND SO THAT, THAT'S THE BIG DIFFERENCE.
UM, NOW CERTAINLY IF YOU HAVE HIGHER PEAK LOAD GROWTH, THEN YOUR RESERVES MIGHT, MIGHT END UP BEING LOWER.
UH, THE ONE POINT I WOULD MAKE IS, IS OBVIOUSLY, UM, EVEN GOING BACK TO, TO 2021, RIGHT? LIKE BEFORE ALL OF THAT LOAD GROWTH MATERIALIZED, YOU, YOU STILL HAD A RELIABILITY EVENT IN ERCOT.
AND WE ARE ALSO HAVING HIGHER THAN OTHER, WE ARE HAVING HIGHER LOAD GROWTH IN OTHER ISOS AT THAT TIME AS WELL, RIGHT? YEP.
UM, I, I WOULD ARGUE THAT PERHAPS, UM, LARGE LOADS LIKE THE ONES IN THE TEXAS ENERGY BUYERS ALLIANCE, WHO I'M REPRESENTING TODAY, I DIDN'T MENTION THAT EARLIER, UM, ARE INTERESTED IN ERCOT BECAUSE IT'S AN ATTRACTIVE PLACE TO BUILD, YOU KNOW, NEW DEMAND FOR ELECTRICITY, WHICH IS MAYBE POTENTIALLY CORRELATED WITH THE COST OF ELECTRICITY.
AND THIS IS EVIDENCE THAT PEOPLE FIND TEXAS AN ATTRACTIVE PLACE TO DO BUSINESS.
UH, BUT I'M NOT ASKING YOU TO COMMENT THAT UNLESS YOU WOULD LIKE TO
UM, YOU, YOU MENTIONED THAT YOU DID SURVEYS.
COULD YOU DESCRIBE THE SURVEY PROCESS? SO I, I THINK JUST, JUST TO BE TRANSPARENT WITH, WITH EVERYONE HERE, RIGHT? THIS WAS NOT SORT OF A, A FORMAL SURVEY WHERE WE COLLECTED A HUNDRED DIF DIFFERENT RESPONSES.
UH, I THINK, I THINK THERE'S MORE OF SORT OF, WE, WE TALK REGULARLY TO DATA CENTERS, TO COMPANIES, TO TECH COMPANIES THAT WE WORK WITH DIRECTLY.
AND SO WE SORT OF SURVEYED RESPONSES AND HAD INTERVIEWS, UH, WITH SOME OF THESE COMPANIES AS WE'VE BEEN DOING STUDIES ON THIS TOPIC.
SO THIS WAS BASED OFF OF CONVERSATIONS AND INTERVIEWS WE'VE HAD WITH THOSE RATHER THAN A SORT OF FORMAL QUANTIFIED SURVEY TYPE OF ANALYSIS.
AND HOW DID THOSE CONVERSATIONS AND INTERVIEWS IMPACT YOUR ANALYSIS? SO GENERALLY, UM, THERE'S TWO ASPECTS.
ONE IS, IS TO RECOGNIZE THAT THERE'S A WIDE RANGE OF RESPONSES, RIGHT? AND IT'S, IT'S, IT'S DIFFICULT TO JUST PUT DATA CENTERS IN A BUCKET AND PRETEND THEY'RE ALL THE SAME AND THEY'RE ALL GOING TO DO THE SAME THING.
UM, YOU KNOW, ON ONE END OF THE SPECTRUM, WE'VE HEARD FROM SOME TECH COMPANIES THAT, YOU KNOW, THEY'RE LOOKING INTO MAKING VERY FLEXIBLE LOAD, UM, THE ABILITY TO HAVE THE LOAD BEING QUITE FLEXIBLE BY SHIFTING ACROSS DATA CENTERS.
AND, AND, YOU KNOW, WE'VE ALL SEEN, I THINK THE ARTICLES AROUND THAT.
UH, AND WE'VE HAD THAT CONFIRMED MORE DIRECTLY, UH,
[01:10:01]
ON THE OTHER END OF THE SPECTRUM, WE'VE ALSO HEARD FROM SOME DATA CENTERS THAT THEY'RE NOT REALLY INTERESTED IN TURNING OFF.UH, AND THAT, YOU KNOW, ULTIMATELY THAT'S THE JOB OF THE GRID TO PROVIDE RELIABLE POWER AND, AND THEIR BACKUP GENERATION IS, IS REALLY THERE IN CASE OF, UH, SORT OF THEY HAVE NO OTHER CHOICE RATHER THAN ACTIVE PARTICIPATION IN SOME OF THESE MARKET DYNAMICS.
AND SO OUR ASSUMPTIONS AT THE END OF THE DAY TRIED TO REFLECT THAT DIVERSITY OF OPINIONS AND OF OPEN, WELL, OPINIONS NOT THE RIGHT WORD, BUT THAT DIVERSITY OF SITUATIONS AND, AND ARRANGEMENTS THAT, THAT WE, WE HEARD.
AND SO THIS IS WHY YOU'RE SEEING ALSO THESE BREAKDOWNS RIGHT ON, ON THIS SLIDE OF SORT OF 10 50 AND 40%, UH, MEANT TO REFLECT THE FACT THAT WE THINK A LOT OF DATA CENTERS ARE GONNA BE VERY MUCH WILLING AND ABLE TO PROVIDE THAT FLEXIBILITY.
BUT FROM A RELIABILITY STUDY STANDPOINT, IT ALSO SEEMED OVERLY OPTIMISTIC TO ASSUME A HUNDRED PERCENT OF ALL DATA CENTER CAPACITY WOULD BE WILLING AND ABLE TO PROVIDE THAT SAME, UH, LEVEL OF FLEXIBILITY.
AND, AND SO OBVIOUSLY COMING UP WITH A NUMBER IS, IS, IS, IS COMPLICATED.
UH, BUT, BUT THE, THE ASSESSMENT WAS TO, TO BRING IT UP, BREAK IT DOWN BETWEEN THOSE NUMBERS, SO YOU GOT TO 60% BASED ON THAT, YOUR BEST, YOU KNOW, JUDGMENT BASED ON YOUR EXPERIENCE.
UM, HOW DOES THAT SQUARE WITH SB SIX'S REQUIREMENT TO, UM, HAVE UTILITIES INSTALLED THIS CONNECT DEVICES? RIGHT.
SO SB SIX HAS A PROVISION THAT, UM, IF YOU HAVE MORE THAN 50% BACKUP GENERATION, UH, YOU WILL BE SUBJECT TO, YOU KNOW, RIGHT INVOLUNTARY CURTAILMENT FROM ERCOT, UH, WORK, YOU KNOW, ASKED TO TURN ON THIS BACKUP GENERATION TO GET, GET OFF THE GRID.
UH, AND SO ASSUMING EVERY SINGLE DATA CENTER OPERATIONS IN TEXAS WOULD HAVE A HUNDRED PERCENT BACKUP GENERATION, YOU COULD THINK ABOUT OBVIOUSLY A HUNDRED PERCENT OF THAT LOAD, UH, BEING FLEXIBLE.
I THINK WHAT WE'VE SEEN ALSO IS WHEN YOU PUT THRESHOLDS OF A CERTAIN PERCENTAGE AFTER WHICH PEOPLE HAVE TO, TO ABIDE BY CERTAIN RULES, YOU ALSO CREATE A SITUATION WHERE PEOPLE MIGHT, MIGHT LOOK AT GETTING RIGHT BELOW THAT THRESHOLD.
AND SO JUST, UH, OUR APPROACH WAS TO, TO ASSUME NOT EVERYONE WOULD BE LOOKING TO BE SUBJECT TO SB SIX REQUIREMENTS, AND WE'LL LOOK AT, FIND WAYS TO MAYBE AVOID HAVING THAT REQUIREMENT.
AND SO, SO TAKING INTO ACCOUNT A 40%, UH, SEEMED, SEEMED LIKE A REASONABLE APPROACH.
UM, BUT YOU'RE NOT, UM, LOOKING AT WHETHER OR NOT THE UTILITY WILL HAVE A DISCONNECT DEVICE IN ORDER TO DEAL WITH MERCHANT SITUATIONS FOR ALL THESE NEW DATA CENTERS.
UM, SO IF I CAN JUST MAKE A, A COMMENT, I THINK THAT THERE IS A NOTE HERE THAT SAYS, YOU KNOW, THERE IS THE EEA THREE, RIGHT? SO THAT 40% CAN STILL BE SWITCHED OFF IN AN EEA THREE, WHICH I THINK IS KIND OF WHAT YOU'RE GETTING AT, RIGHT? IS IF YOU HAVE A KILL SWITCH, IF YOU WILL, YEAH, YOU'RE GONNA BE CURTAILED.
BUT I'M SURPRISED TO HEAR THAT THE KILL SWITCH WILL BE RESERVED TILL EEA THREE, I THINK.
SO, SO MAYBE THERE'S TWO, THERE'S SORT OF TWO PARTS HERE, RIGHT? THE FIRST PART IS LOOKING AT THE SORT OF MAYBE WHAT WE CAN CALL THE SB SIX FLEXIBILITY MM-HMM
RIGHT? WHICH IS YOU WILL START BY CURTAILING OR ASKING DATA CENTERS TO, TO FIRE UP THEIR BACKUP GENERATION IN THEN EEA TWO CONDITIONS.
SO BEFORE YOU HAVE TO ACTUALLY START S******G LOAD FOR THE REST OF US.
UM, AND, AND THOSE ARE THE ONES THAT ARE HAVE BACKUP GENERATION AND ARE SUBJECT TO THOSE PROVISIONS.
ONCE YOU GET TO EEA THREE CONDITIONS, THEN OBVIOUSLY YOU GET INTO THE REAL REALM OF UTILITIES WILL START S******G LOAD PROPORTIONALLY TO, YOU KNOW, THE LOAD THEY HAVE ON THE SYSTEM AND AIR CUT WILL START DIRECTING LOAD SHEDDING.
WE'RE MAKING AN ASSESSMENT HERE OF HOW MUCH OF THAT LOAD SHEDDING HAPPENS AS OPPOSED TO LIKE WHO IT IS DIRECTED TO.
AND SO IT'S SORT OF IN THE STUDY, AS SOON AS WE GET INTO THOSE EEA THREE CONDITIONS IS THOSE NUMBERS OF ROTATING YOU SEE HERE.
AND SO THEN THOSE NINE GIGAWATTS WOULD THEN HAVE TO BE SPREAD OUT OVER THE ENTIRE AIRCRAFT LOAD THAT'S REMAINING IN INCLUDING THE REMAINING DATA CENTERS? CORRECT.
UM, SO DID ANY OF THE, UM, DISCUSSIONS AND INTERVIEWS, UM, DID THEY OCCUR WITH ANY LOADS THAT CURRENTLY DO ECONOMIC PRICE-BASED DEMAND RESPONSE? I SEE NODS, SO YES.
UM, THIS IS MORE OF A COMMENTARY, BUT I WOULD WELCOME AND COMMENT ON IT IF YOU WANT TO RESPOND, BUT I WANNA BE FAIR, THIS IS NOT A QUESTION.
[01:15:01]
JUST, JUST LIKE ANY OTHER INVESTMENT, UM, WHETHER PEOPLE CHOOSE TO PARTICIPATE IN ECO ECONOMIC DEMAND RESPONSE AND HAVE A PLAN TO BE ABLE TO DO SO IN ADVANCE, WHICH IS CRITICAL FOR IT TO HAPPEN, WOULD BE BASED ON THE FREQUENCY, MAGNITUDE AND DURATION OF HIGH PRICES.UM, AND SO IF, UH, THOSE WERE TO OCCUR MORE OFTEN THAN YOU MIGHT SEE MORE ATTEMPTS TO AVOID PAYING THOSE HIGH PRICES.
UM, SO, UM, I THINK THIS WAS EITHER IN SLIDE 14 OR MAYBE SLIDE 18, BUT IT WAS, THERE'S ALSO THE COMPARISON FOR THE D-R-S-A-S QUANTITY.
UM, WAS THAT BASED ON YOUR OWN ESTIMATION? BASED ON THE NON SPEND PLAN? YEAH.
SO ESSENTIALLY, AND, AND I MEAN THIS WAS, THIS WAS SORT OF A, UM, INITIAL ASSUMPTION THAT, THAT WE VALIDATED WITH ERCOT.
OBVIOUSLY THIS IS AN INITIAL ESTIMATE.
I'M SURE THE FINAL PROCUREMENT WOULD BE SLIGHTLY DIFFERENT, BUT THE ORDER OF MAGNITUDE SHOULD BE THE SAME.
AND, UM, ONE OF THE THINGS THAT I'M STRUGGLING TO SQUARE WITH THE STORY IS, UM, HOW A DRS PLUS IS MERELY JUST MODIFYING THE DEPLOYMENT FACTOR IF THE QUANTITIES ARE SO DIFFERENT.
AND MAYBE THAT'S A QUESTION FOR A LATER TIME WHEN WE'RE TALKING ABOUT NPRS, OR MAYBE IT'S NOT.
YEAH, I THINK THAT MIGHT BE A BETTER QUESTION FOR OKAY.
UM, AND THEN, UM, ONE LAST QUESTION THAT MIGHT BE RHETORICAL ALONG THE LINES OF MY QUESTION ABOUT THE TEXAS ECONOMY ON SLIDE 16.
UM, WHY DO YOU THINK THAT YOU GET LOWER INVESTMENT IN ONE AND TWO HOUR BATTERIES, UM, UNDER THE MIDDLE, UH, BAR ON THE RIGHT? YEAH, I MEAN THAT, IT, IT, IT'S A VERY GOOD POINT.
SO OBVIOUSLY WHAT HAPPENS IS IN THE TEXAS MARKET, IF YOU BUILD ADDITIONAL GENERATION, RIGHT, YOU HAVE MORE RESERVES, AND SO THEREFORE YOU WILL SUPPRESS SOME OF THE SCARCITY PRICING SIGNAL.
UM, AND, AND THAT'S WHAT WE'RE SEEING IN THE MODELING, RIGHT? BATTERIES IN PARTICULAR WILL RELY ON THAT, UH, PRICING SIGNAL, PARTICULARLY SHORT DURATION BATTERIES.
UM, AND SO THEY HAVE LESS OF AN INCENTIVE TO BUILD, RIGHT? AND SO YOU SEE LESS OF THOSE SHORT DURATION BATTERIES BUILDING.
UM, I APPRECIATE THAT POINT, UH, ANSWERED SO CLEARLY, SINCERELY.
UM, AND, UM, THAT'S I THINK A CORE OF MY CONCERN, NOT WITH THE MODELING, BUT WITH THE SUGGESTED POLICY CHAINS, WHICH I APPRECIATE.
YOU SAID YOU'RE JUST DOING THE MODELING.
UM, IT, A CENTRAL PART OF OUR MARKET DESIGN IS BASED AROUND THE OCCURRENCE OF THIS PRICES AND THE DRRS PLUS IS, I FEELS DESIGNED TO MINIMIZE THOSE PRICES, UM, THROUGH PROCUREMENT.
AND I THINK YOUR MODELING SHOWED THAT THERE'D BE SIGNIFICANTLY LESS REVENUE IN THE REAL TIME MARKET UNDER THE DRS PLUS SCENARIO.
ANOTHER WAY OF SAYING IT IS THAT REVENUE IS SHIFTED FROM THE POLICY.
IT IS SHIFTED INTO LONGER DURATION RESOURCES, RIGHT? AND SO FOUR HOUR BATTERIES IN THIS CASE, IF THEY'RE ELIGIBLE OR THERMAL RESOURCES, THEN CAPTURE MORE OF THAT REVENUE.
BUT I THINK THAT'S FACTUAL, RIGHT? THAT'S WHAT HAPPENS.
AND WITH MY SPIN ON IT, WHICH I'LL ACKNOWLEDGE, UM, IT'S SHIFTING IT FROM ANYONE THAT CAN RESPOND TO A PRICE TO ONLY A SUBSET OF THOSE THAT CAN RESPOND TO A PRICE, UH, YES, ONLY ELIGIBLE RESOURCES, RIGHT? AS OPPOSED TO, TO THE FOUR ORDC.
UM, WELL THAT'S THE SUM OF MY QUESTIONS, ERIC, JUST TO MAKE A POINT THERE, THERE'S, THERE'S A DISTINCTION BETWEEN THE ENERGY MARKET AND THE, UM, AN SERVICE MARKET, RIGHT? SO YOU'RE SHIFTING THE MONEY INTO THE ANCI SERVICE MARKET, I RATHER THAN THE, THE ENERGY MARKET.
AND THAT'S, THAT'S WHAT THIS IS SHOWING.
I WOULD, UM, I APPRECIATE THE POINT YOU'RE MAKING.
I I, I WAS TRYING TO MAKE THE POINT EARLIER THAT I DON'T UNDERSTAND HOW AN ANCILLARY SERVICE CAN REPRESENT 100% OF THE ONLINE CAPACITY.
UM, IT'S, IT'S WHATEVER IS, IS OF THAT PARTICULAR REQUIREMENT, RIGHT? SO IF YOU HAD A PARTICULAR SET OF RESOURCES, AND WE SEE THIS WITH NONS SPEND, IS NOT EVERY RESOURCE IS ELIGIBLE FOR NONS SPEND, BUT IT DEPENDS ON HOW MANY RESOURCES YOU HAVE AVAILABLE TO MEET THAT REQUIREMENT.
[01:20:01]
UM, OKAY.UM, WHAT IS, I, I APOLOGIZE FOR FOR THIS, BUT I THINK THE WORD ANCILLARY HAS A MEANING.
UM, WHICH DOESN'T MEAN THE SUM OF EVERYTHING.
IT'S SOMETHING THAT'S IN ADDITION TO EVERYTHING.
AND, AND AGAIN, I THINK IF WE WERE, UH, IT, IT DEPENDS ON THE SET OF RESOURCES, THEY'RE CAPABLE OF MEETING THAT REQUIREMENT, AND THE WAY THE REQUIREMENT WILL BE SET IS TO ACHIEVE THE RESOURCE ADEQUACY.
UH, MANY OF MY QUESTIONS WERE ANSWERED.
UH, I HAVE A FEW ADDITIONAL ONES.
UH, CAN YOU, UH, EXPLAIN A LITTLE BIT MORE, UH, HOW YOU DID THE DETERMINISTIC ANALYSIS? SO IT'S NOT A MONTE CARLO SIMULATION, YOU JUST TAKE ONE YEAR AND THEN YOU, UH, RUN IT AND THEN YOU SEE WHAT ADDITIONAL RESOURCES CAN SUSTAIN BASED ON THE CONE FOR JUST THAT YEAR, AND THEN YOU RUN IT AGAIN? IS THAT HOW IT IS DONE? UM, SO NOT EXACTLY, BUT I'M, I'M HAPPY TO EXPLAIN IT IN MORE DETAIL.
SO IN THIS CASE, WE DID NOT RUN A MONTE CARLO SIMULATION UNDER DIFFERENT CONDITIONS, WHICH YOU CAN ABSOLUTELY DO AND IS, IS A USEFUL EXERCISE AS WELL.
UM, THERE'S TWO REASONS WE WE DIDN'T DO THAT.
ONE IS, IS SOMETIMES THEY GET COMPLICATED TO EXPLAIN, UH, AND COMMUNICATE.
AND, AND THE, THE PURPOSE HERE WAS TO ALSO HAVE A STRAIGHTFORWARD VIEW OF UNDER THIS TYPE OF EVENT IN THIS HERE, THIS, THESE ARE THE RESULTS IN TERMS OF RELIABILITY OF, OF SOMETHING THAT HAS HAPPENED IS A HIGH LIKELIHOOD OF HAPPENING AGAIN IN TERMS OF ELLIOT IN 2023.
UM, SO THAT, THAT WAS A DELIBERATE CHOICE, BUT OBVIOUSLY HAS LIMITATIONS AS WELL IN TERMS OF HOW MUCH IT TELLS YOU.
UM, THE SECOND PART IN TERMS OF THE MODELING, SO THE WAY WE DO WEATHER YOU'RE MODELING ESSENTIALLY IS WE GET THE GENERATION PROFILES FOR ALL, UH, ALL RENEWABLE RESOURCES IN PARTICULAR, UH, WE GET AVAILABILITY INFORMATION FROM, UH, DISPATCHABLE RESOURCES IN TERMS OF WHAT WE EFFECTIVELY OBSERVED AT DIFFERENT TIMES OF THE YEAR.
UM, SO THAT GIVES US A BASIS FOR, YOU KNOW, PERCENTAGE AVAILABILITY AND, AND GENERATION PROFILES.
WE ALSO HAVE THE LOAD PROFILES, RIGHT FROM, UM, LET ME JUST LOAD THAT YEAR.
AND THEN EVERYTHING IS SCALED UP AND OBVIOUSLY THERE'S SOME ADJUSTMENTS TO BE, MAKE SURE THIS IS CONSISTENT ACROSS DIFFERENT LOAD CATEGORIES.
UH, BUT EVERYTHING IS THEN SCALED UP TO THE DEMAND AND THE SUPPLY OF THAT YEAR AND THEN RUN THROUGH A PRODUCTION COST MODEL, UH, TO SIMULATE THE HOURLY RESULTS FOR EACH HOUR OF THAT, UH, OF SORT OF THE 2030 FORECAST.
BUT UNDER WHETHER YOUR CONDITIONS OBSERVED IN 2022 OR 2023 SCALED UP TO THE SUPPLY AND DEMAND OF THAT YEAR.
SO MAYBE THE EXTREME EVENTS AND STUFF THAT COULD SIMULATE THE HIGHER PRICES, WHICH COULD THEN TRIGGER MORE INVESTMENT, THAT PART IS KIND OF MISSING THEN.
IS THAT CORRECT OR NO? SO, UM, SO YOU MEAN AS PART OF THE CAPACITY EXPANSION MODEL TO DETERMINES HOW MUCH YEAH, YEAH.
SO THE 2013 WEATHER YEAR, TO BE TOTALLY FAIR, HAS A WINTER STORM AND HAS SOME HOT DAYS IN THE SUMMER.
SO WE DO OBSERVE HIGH PRICES IN THE 2013 WEATHER YEAR, RIGHT? SO THIS IS NOT A, WE'RE NOT TAKING A YEAR THAT'S TOTALLY FLAT AND HAS NO SCARCITY.
UH, SO THERE IS SOME INVESTMENT SIGNALS THAT WE OBSERVE FROM THAT, BUT OBVIOUSLY IT DOES NOT INCLUDE THE EXPECTATION OF LIKE A YURI OR ELLIOT OR THE EXPECTATION OF A 2011 OR A 2023 SUMMER IN TERMS OF THE INVESTMENT SOLVE THAT HELPFUL.
SO IT, IT'S KIND OF LIKE NOT REACHING AN, UH, AN ECONOMICALLY OPTIMAL LEVEL FOR YOU TO COMPARE BETWEEN DIFFERENT CASES THEN.
BUT THAT, I MEAN, TO BE TOTALLY FAIR AND MAYBE THAT THAT'S GOING BACK TO OUR EXPERIENCE PROVIDING, UH, FINANCING STUDIES FOR, FOR GENERATION, RIGHT? LIKE, UM, BANKS ARE TYPICALLY FAIRLY CONSERVATIVE AND BANKS ASSIGN A FAIRLY LOW AMOUNT OF VALUE TO SORT OF HIGH PRICE, YOU KNOW, HIGH VOLATILITY YEARS IN TERMS OF, FOR EXAMPLE, PROVIDING DEBT TO PROJECTS, RIGHT? GENERALLY THEY WANT TO SEE WHAT IS THE EXPECT AVERAGE EXPECTED CONDITIONS AND THEN MAYBE A DOWNSIDE CASE TO TO, TO BE SAFE ON THE DEAD SIDE.
UH, BUT THEY ASSIGN FAIRLY LOW VALUE TO THOSE UPSIDE, RIGHT? THAT CAN BE PROVIDED BY, BY EXTREME WEATHER.
AND SO GENERALLY WE FIND THAT INVESTMENT DECISIONS ARE TEND TO BE MORE BASED ON SORT OF A, A CENTRAL, EVEN SOMETIMES A LOW CASE VIEW AS OPPOSED TO LIKE A VERY HIGHLY VOLATILE VIEW.
BUT THAT MAY BE COUNTER TO THE ENERGY ONLY MARKET, WHICH TRIGGERS INVESTMENT BASED ON THOSE HIGH PRICES, BUT NOT NECESSARILY, RIGHT? LIKE WHAT I'M TRYING TO SAY IS THE BANK WILL ASSIGN SORT OF AN AVERAGE AMOUNT OF SCARCITY THAT THEY SEE FAIRLY CONSISTENTLY ACROSS
[01:25:01]
THE FORECAST AS OPPOSED TO LIKE A HIGHLY VOLATILE AMOUNT OF SCARCITY.SO IN OTHER WORDS, RIGHT, THE BANKS WILL TAKE SCARCITY INTO ACCOUNT, BUT THEY'LL TAKE SORT OF THE AVERAGE THEY SEE OVER A TENURE PERIOD, WHICH TENDS TO MATCH PRETTY WELL WITH WHAT WE'RE SEEING IN 2013.
THEY WON'T TAKE INTO ACCOUNT THIS SCARCITY PRICING WE'VE SEEN IN 2023 TO DECIDE HOW MUCH MONEY THEY'RE GONNA LEND YOU.
UM, SO, UH, THE DRS EARLIER, THOSE WAS DISCUSSION THAT, UH, THE DRS AMOUNT IS NONS SPIN AMOUNT, SO YOU REDUCE THE NONS SPIN BY THAT AMOUNT AND THEN SHIPPED IT TO NO, IT'S ADDITIONAL, IT'S, IT'S ON TOP OF THAT.
AND THEN THERE IS NO AS, UH, DEMAND CURVE THAT IS USED FOR THAT CLEARING? IT IS, YES.
ALTHOUGH, UM, IN BOTH CASES, UH, WE HAD A RELEASE FACTOR OF A HUNDRED PERCENT.
UM, TIM, JUST IN THE, JUST IN THE RESOURCE, NO.
SO TIM, DO YOU WANNA CLARIFY EXACTLY THE IT'S A HUNDRED PERCENT AS NO RELEASING RIGHT, EXACTLY.
IN THE ANCILLARY SERVICE CASE.
AND THEN, UM, FOR THE DEMAND RESPONSE FOR THAT, UH, AURORA CASE, IT'S, IS IT CORRECT THAT IT IS SEVEN GIGAWATTS OF DATA CENTERS BY 2030 TOTAL, OR IS IT SEVEN GIGAWATTS ON TOP OF WHAT WE HAVE RIGHT NOW? TOTAL.
TOTAL TO TOTAL, BUT WE ALREADY HAVE FIVE AND SO IT'S ONLY THE TWO GIGAWATTS THAT WE CAN YOU KIND OF LIKE EXPLAIN, UM, THE RATIONALE.
LIKE WE ALL ARE TRYING TO UNDERSTAND HOW MUCH THE LOAD GROWTH IS GOING TO HAPPEN.
SO LIKE WHATEVER SURVEY OR OTHER INFORMATION THAT YOU HAVE.
YEAH, SO I MEAN, OBVIOUSLY THE SITUATION WITH DATA CENTERS IS PRETTY DYNAMIC.
SO IN THE FORECASTS FOR THE CONSERVATIVE CASE OR THE, THE, IN THIS CASE, THE AURA CENTRAL CASE, WE HAVE SEVEN GIGAWATTS OF DATA CENTERS AND THEN ADDITIONAL FOUR GIGAWATTS OF BITCOIN, RIGHT? AND I KNOW SOMETIMES THE, THE, THE DIFFERENCE BETWEEN THE TWO IS A LITTLE BLURRY, BUT, BUT ESSENTIALLY THAT'S THE TOTAL AMOUNTS, RIGHT? SO 11 GIGAWATTS BETWEEN THE TWO, UM, VERSUS THE AIRCRAFT CASE, WHICH HAS 20, 20, 22 GIGAWATTS OF DATA CENTERS AND THEN COMPARABLE AMOUNT OF BITCOIN.
SO WAS THAT BASED ON SOME SURVEY OR SOMETHING OR? YEAH, SO WE LOOKED AT ESSENTIALLY THE CAPACITY THAT WAS UNDER CONSTRUCTION.
UM, I MEAN, JUST TO ACKNOWLEDGE, RIGHT? LIKE I THINK AT THIS POINT THERE'S INVESTMENT ANNOUNCEMENTS ABOUT DATA CENTER CAPACITY PRETTY MUCH EVERY OTHER WEEK.
SO IT'S PRETTY DYNAMIC, RIGHT? WE DIDN'T RUN THIS STUDY LAST WEEK, SO, UM, YOU KNOW, A FEW MONTHS AGO THERE WAS A, THERE WAS LESS ANNOUNCEMENTS.
UM, I THINK OUR FORECAST ALSO TRIES TO CAPTURE THE FACT THAT NOT ALL ANNOUNCEMENTS WILL COME TO FRUITION, RIGHT? IT'S EASY TO ANNOUNCE PROJECTS AND TO PUT OUT A PRESS RELEASE.
IT'S ANOTHER ONE TO GET IT FINANCED AND, AND EFFECTIVELY BUILT.
AND WE'VE SEEN A FEW EXAMPLES OF ANNOUNCEMENTS THAT ARE NOW BEING MAYBE PUT INTO QUESTIONS, RIGHT? AND I DON'T NEED TO PUT ANY NAMES, BUT THERE'S PLENTY OF EXAMPLES AROUND, UH, AND SO FULLY ACKNOWLEDGED IT'S A CONSERVATIVE CASE, FULLY ACKNOWLEDGED THERE IS UPSIDE POTENTIAL TO THAT, BUT IT FELT LIKE AN INTERESTING CASE TO USE AS A CONSERVATIVE SORT OF FLOOR OKAY.
UM, SO I ALSO WANNA UNDERSTAND A LITTLE BIT ON THAT STUDY RESULT.
UH, SO I THINK ONE OF THE SLIDES THAT, UH, THERE'S A REDUCTION, UH, FROM THE SEVEN GIGAWATT AC CENTER, THE, UM, LOAD SHARE REDUCED FROM 8.7 TO 5.7.
SO IS HOW IS THAT COMING? LIKE IF I, I'M TRYING TO UNDERSTAND HOW IS IT IF 60% IS, UM, FLEXIBLE? YEAH, THAT'S A GOOD QUESTION.
SO OUR ASSUMPTION IS THAT, YOU KNOW, DATA CENTERS OF COURSE FOLLOW A DEMAND PROFILE.
UM, SOME OF THAT IS BASED ON UTILIZATION, BUT A LOT OF THAT IS ALSO, UH, ON COOLING DEMAND, RIGHT? SO IF YOU HAVE A ONE GIGAWATT DATA CENTER, IT'S ALMOST NEVER GOING TO BE RUNNING AT ONE GIGAWATT PULLING FROM THE GRID, RIGHT? IT'S PROBABLY GONNA BE FLUCTUATING FROM A BASE OF AROUND SIX 70 OR SO AND ABOVE BASED ON HEATING DEMAND.
SO THE REDUCTION THAT YOU'RE GETTING AT THAT MOMENT OF, OF THE EVENT IS BASED ON THE PROPORTION OF DATA CENTERS THAT ARE RESPONSIVE.
PLUS YOU HAVE TO DE-RATE FURTHER LOAD FACTOR AT THAT TIME.
'CAUSE IF YOU'RE A ONE GIGAWATT NAMEPLATE, BUT YOU'RE ONLY CONSUMING 700 MEGAWATTS, YOU'RE EFFECTIVELY ONLY TURNING DOWN, UH, 700 MEGAWATTS.
THE ONE OTHER PIECE ON THAT IS THAT THERE IS SOME OFFSET TO CAPACITY THAT WE HAVE.
WE HAVE A SLIDE IN THERE ON THAT AS WELL.
UM, SO SOME OF THAT PRICE RESPONSIVE, THAT 10% OF DATA CENTERS THAT'S PRICE RESPONSIVE, IT DOES REDUCE A BIT THE PRICE SIGNAL AND THERE'S A BIT OF OFFSET BUILD AS WELL.
AND THAT'S WHY ULTIMATELY THAT REDUCTION IN MAXIMUM LOAD SHED IS ONLY, YOU KNOW, LIKE YOU SAID, TWO TO THREE GIGAWATTS IN THAT CASE.
[01:30:01]
THEN THE ANOTHER SLIDE, IT SAYS THE, UH, MANY OF THOSE DATA CENTERS ARE NOT FLEXIBLE BECAUSE THE WALL IS TWO TO FOUR, UH, K, BUT OUR WALL IS 5K.SO WHAT, WHAT WAS THE RATIONALE BEHIND THAT THEN? SO THAT'S THE VA, SORRY,
SO THAT'S BASICALLY SAYING AS A DATA CENTER, MY BREAK EVEN PRICE IS TWO TO 4K.
UM, SO IT'S EXCEEDINGLY RARE ESSENTIALLY THAT I WOULD, UH, TURN DOWN.
UH, BUT THEN WHEN DURING THIS EMERGENCY CONDITIONS, THE PRICES WILL BE THERE.
SO YOU WOULD SEE THEM DURING AN EMERGENCY CONDITION, YOU WOULD YEAH, I THINK OUR ASSUMPTION ON INFLEXIBILITY IS THAT A LOT OF DATA CENTERS WILL HAVE, YOU KNOW, HEDGED PRICES, RIGHT? THEY'LL HAVE BOUGHT, UH, THEIR POWER ON FUTURES MARKETS, ET CETERA, AND THEY WON'T BE FULLY EXPOSED TO THOSE SPOT PRICES.
AND, UH, IN THE, UM, THE
IN IN THE ERCOT LTLF CASE, THE SYSTEM IS ESSENTIALLY GETTING LARGER, THAT PUTS MORE UPWARDS PRESSURE ON PRICES AND CAN SUPPORT MORE CAPACITY.
ANDREW? MAY, SORRY, MAY, MAYBE I COULD JUST CLARIFY AS WELL.
THERE WERE SOME ESTIMATES THAT WERE PUT INTO THE REPORT, UH, INDICATING WHAT, UH, A REPRESENTATIVE VALUE OF LOSS LOAD COULD BE FOR CERTAIN DATA CENTERS, BUT IT'S MEANT TO BE AN ESTIMATE FOR A CERTAIN TYPE.
WE DON'T NECESSARILY VIEW THAT TO BE THE VOL ACROSS ALL DIFFERENT DATA CENTER PROFILES.
SO IS IT WAS ONE PIECE OF INFORMATION THAT WAS MEANT TO INFORM WHAT PROPORTION OF THE ENTIRE DATA CENTER FLEET WOULD BE PRICE RESPONSIVE AND WHAT PROPORTION WOULD NOT? YEAH, IT'S SIMILAR TO WHAT WE ARE HEARING OF 2000.
I'M GONNA GO BACK TO THE PEDANTIC RELEASE FACTOR COMMENTARY BECAUSE I THINK THE LACK OF CLARITY BETWEEN THE AURORA STUDY AND THE DESIGN IN 1310 IS JUST A LITTLE CONFUSING.
UH, I THINK WE HAD SAID THAT THE RELEASE FACTOR IS ACTUALLY A HUNDRED PERCENT, IT'S ACTUALLY PROBABLY VERY HIGH, BUT NOT A HUNDRED PERCENT.
SO JUST TO USE THE EXAMPLES AND THE NUMBERS THAT YOU GAVE, UH, YOU KNOW, IF YOU'RE PROCURING 80 GIGAWATTS OF DRS PLUS AND SAY TWO GIGAWATTS OF DRS ASS OPERATING RESERVE NONPLUS, UH, THAT'S TWO, TWO GIGAWATTS IS LIKE 2.5% OF 80 GIGAWATTS.
AND IF I'M INTERPRETING 1310 CORRECTLY, THAT WOULD MEAN THAT'S A RELEASE FACTOR OF 97.5%.
SO, YOU KNOW, I, IF I'M UNDERSTANDING THAT CORRECTLY, MAYBE A NOD FROM SITE.
SO YEAH, AND JUST TO CLARIFY, RIGHT, SO WE MODELED THEM INDEPENDENTLY, YOU'RE MODELING THEM COMPLETELY INDEPENDENTLY.
SO IN OUR, IN OUR STUDY IN A WAY YOU COULD THINK ABOUT IT AS ZERO AND A HUNDRED, BUT THEY WOULD BE BLENDED IN, IN THE PROPOSAL.
SO NOW I THINK THAT MAKES SENSE.
UH, YOU KNOW, FOLLOWING ON WHAT REMI WAS ASKING ABOUT, UH, AND YOU MENTIONED, SO THIS IS A DETERMINISTIC MODEL.
UH, I THINK ONE OF THE CONCERNS THAT WE WOULD HAVE, SO WE ARE FANS OF THE, UH, EXTENDED RRDC VERSION OF THE STUDY THAT YOU DID.
BUT EVEN WITH THAT, IF YOU'RE RUNNING AN HOURLY DETERMINISTIC MODEL, IT'S GONNA BE EXCEEDINGLY RARE THAT YOU'RE ACTUALLY GOING TO GET THOSE SHORTAGE PRICES THAT WOULD MAKE A MECHANISM LIKE THAT EFFECTIVE.
PART OF THE REASON FOR THAT IS A LOT OF THE REASON THAT WE GET SCARCITY PRICING IS BECAUSE OF STOCHASTIC FORECAST ERROR AND THINGS LIKE THAT.
SO LIKE THE SPREAD BETWEEN DAY AHEAD AND REAL TIME WHERE WE THINK WE GET A COMMITMENT SOLUTION OUT OF DAY AHEAD, AND THEN THINGS CHANGE IN REAL TIME, AND NOW WE'RE A LITTLE UPSIDE DOWN.
UH, DO YOU DO ANYTHING IN YOUR MODELING TO KIND OF ACCOUNT FOR THAT OPERATIONAL REALITY? YEAH, SO YES, THE, THE ANSWER TO THAT IS WE DEFINITELY TAKE TIME TO CALIBRATE THE SCARCITY PRICING WE SEE TO HISTORICAL CONDITIONS.
SO, YOU KNOW, UNDER SIMILAR HISTORICAL CONDITIONS, WE SEE SIMILAR SCARCITY PRICING.
AND FOR THAT, YOU HAVE TO DO A LITTLE ADJUSTMENT, RIGHT? IF YOU TAKE STRAIGHT UP THE, THE ORDC CURVE AND YOU PUT IT INTO A DETERMINISTIC MODEL, YOU'RE NOT GONNA FIND THE SAME THING THAT YOU FIND IN REALITY.
SO OUR SCARCITY PRICING IS CALIBRATED TO ACCOUNT FOR SOME OF THAT JUST TO MAKE SURE WE ARE SEEING THE SAME BEHAVIOR.
AND SO THERE'S A BIT OF CROSS PROCESSING IN THE MODEL THAT IS DOING EXACTLY THAT TO BE ABLE TO, TO GENERATE SIMILAR CONDITIONS.
'CAUSE FUNDAMENTALLY, AND MAYBE THAT'S GETTING INTO TOO MUCH MODELING NERD, UM, SORT OF, YOU'RE NOT GETTING OFF THAT EASY REAL, BUT I KNOW I WON'T, I WON'T IMPRESS YOU, ANDREW, WITH THAT
WHEN YOU RUN A PRODUCTION COST MODEL, YOU'RE MUCH MORE REFLECTIVE OF
[01:35:01]
DAY HEAD CONDITIONS, RIGHT? YOU SORT OF, YOU KNOW, WHAT, WHAT THE WEATHER IS GONNA BE, YOU KNOW, WHAT THE LOAD IS GOING TO BE, YOU KNOW WHAT YOUR GENERATION FLEET LOOKS LIKE, YOU DON'T HAVE ANY UNEXPECTED OUTAGE JUST BEING THROWN IN THE MIDDLE.SO YOU'RE REALLY MODELING A DAY HEAD PRICE.
AND WE ALL KNOW THIS CAR C PRICING IN ERCOT IS IN REAL TIME, YOU KNOW, NOW IS IN ANCILLARY SERVICES, BUT IT'S STILL IN REAL TIME.
AND SO OBVIOUSLY YOU ARE NOT GONNA GET THE SAME KIND OF RESPONSES.
SO THERE IS SOME CALIBRATION WORK THAT NEEDS TO HAPPEN TO TAKE INTO ACCOUNT THAT CAR C PRICING IN YOUR MODELING.
SO, BUT TO BE CLEAR, SO WHEN YOU SAY CALIBRATING, ARE YOU REFERRING, FOR EXAMPLE, TO THE SHAPE OF THE ORDC OR THIS IS STRICTLY A POST-PROCESSING CALIBRATION? UH, THIS IS STRICTLY POST-PROCESSING.
DO YOU WANT TO GO INTO A LOT OF DETAILS? UH, TIM, NO, UH,
AND WHEN WE SEE THE MARKET GETTING TO A CERTAIN TIGHTNESS AND, YOU KNOW, IN TERMS OF OPERATIONAL RESERVES, THEN THOSE CARC PRICING GET KICKED UP.
UH, AND THE PROBABILITY OF A HIGH PRICE IS ESSENTIALLY DEPENDENT ON HOW LOW THE RESERVES ARE, BUT IT IS NOT MATCHING THE EXACT RRDC CURVE BECAUSE IF WE DO, THEN WE JUST DON'T GET THE RESULTS THAT MAKE SENSE? SO, SO WE HAVE OUR OWN CURVE THAT'S MEANT TO REFLECT THE RRDC CURVE, AND WHEN WE RUN IT AGAINST THE HISTORICAL YEAR, IT ESSENTIALLY MATCHES, UM, THAT'S THE CALIBRATION PART.
UH, BUT TO SO THAT IN THE FORECAST, IT GIVES US SENSIBLE RESULTS.
THAT'S, THAT SEEMS REASONABLE.
UH, BUT JUST TO BE CLEAR, SO THE, THE OPTIMIZATION ITSELF, YOU ARE CO-OP OPTIMIZING WHERE THE ORDC IS LIKE A PENALTY FUNCTION BASED ON YOUR RESERVES AS, AS CLEARED THROUGH LIKE THE HOURLY PRODUCTION COST MODEL.
SO LIKE IF I'M RUNNING THE CAPACITY EXPANSION STEP OR, OR WHATEVER IT IS, WHEN WE'RE SHORT ON RESERVES, THAT'S A PENALTY FUNCTION EFFECT EFFECTIVELY, RIGHT? SO IT IS CO-OPING ACROSS ANCILLARIES AND, AND ENERGY.
UH, NOW OBVIOUSLY, YEAH, AGAIN, THE MODEL SOLVES IN SORT OF A DAY AHEAD WORLD, RIGHT? SO THERE'S NOT THIS SORT OF WHOLE REAL TIME OPERATIONS DYNAMIC THAT HAPPENS, AND I DON'T THINK THERE'S ANY MODEL THAT REALLY CAN DO THAT IN THE FUTURE REALISTICALLY.
UH, BUT YEAH, IN THAT DAY HEAD WORLD, IT'S CO OPTIMIZED.
UH, SO THE ANOTHER POINT THAT WAS KIND OF BROUGHT UP WITH SOME OF THESE OTHER, UH, QUESTIONS WAS THE, YOU KNOW, EVEN IN THE DRS MINUS SCENARIO WHERE WE ARE HAVING, WE'RE ADDING THE VOLUME OF NONS SPIND TO THE MARKET, AGAIN, EFFECTIVELY, LIKE DRS WOULD BE AS BIG AS THE NONS SPIND PLAN AND ADDITIVE TO THE NONS SPIN PLAN.
SO, UH, MY OFFICE HAS DONE A LOT OF WORK ON THIS TOPIC, AND, UH, WE WOULD TEND TO ARGUE THAT WE ALREADY SEE A LOT OF PRICE SUPPRESSION FROM THE VOLUME OF NONS SPEND THAT WE'RE PROCURING.
IF, IF YOU HAVE THAT MANY OPERATING RESERVES PROCURED AND IN THE MARKET ALL THE TIME, I DON'T SEE HOW YOU ARE GOING TO GET ANY SCARCITY PRICING AND THAT, THAT'S GOING TO TEND TO MAKE THE DRS PLUS OPTION LOOK MORE COST EFFECTIVE AT GETTING AT INCENTIVIZING NEW GENERATION, I THINK.
DOES THAT SOUND PLAUSIBLE? SO YOU MEAN THE, THAT THE, THIS PRICE SUPPRESSION MEANS THE DS PLUS IS ALWAYS GONNA LOOK BETTER THAN THE DRS? YEAH.
BECAUSE YOU NOW HAVE, YOU CAN HAVE REVENUE COMING THROUGH THE CAP, THE CAPACITY MECHANISM AS OPPOSED TO THROUGH THE ENERGY MARKET.
WE'RE, WE'RE NOT GONNA CALL IT A CAPACITY MECHANISM.
BUT, UM, I THINK IT, IT IS DESIGNED, UH, TO BE PROCURED AS AN ANCILLARY SERVICE.
UH, BUT, UM, YEAH, SO I, I, I HEAR YOUR POINT ON THE, ON THE PRICE SUPPRESSION, THAT WOULD BE COMING FROM DRS DEPENDING ON THE RESOURCES THAT ARE ELIGIBLE TO PROCURE THAT SPECIFIC MECHANISM.
UH, I THINK IT IS IN OUR STUDY, YOU KNOW, THE LESS TECHNOLOGIES ELIGIBLE TO PROCURE EACH SERVICE, THE MORE MAYBE THAT EFFECT COMES THROUGH VERSUS HAVING MORE TECHNOLOGIES ELIGIBLE TO PROCURE EACH OF THE SERVICES.
CAN, CAN I ASK ANDREW A QUESTION HERE? HUH? AND I THINK I'M GOING BACK TO WHAT I HEARD DR. PATTON TALK ABOUT A COUPLE YEARS AGO NOW WHEN HE WAS VERY CONCERNED THAT E-R-S-E-C-R-S WAS A WITHHOLDING OF CAPACITY FROM THE MARKET AND HAD THE POTENTIAL TO ACTUALLY RAISE BECAUSE YOU'RE WITHHOLDING CAPACITY.
AND SO, BUT I THINK WHAT, WHAT YOU JUST SAID IS THAT IT'S THE OPPOSITE OF THAT IS THAT IT WAS ACTUALLY LEADING TO PRICE SUPPRESSION IS JUST TO CLARIFY, I THINK YOUR COMMENT, RIGHT? SO THE
[01:40:01]
ECRS, UH, IN BLIO WAS CAUSED BY, UH, ANCILLARY SERVICES THAT WERE ESSENTIALLY KEPT OUT OF THE ENERGY MARKET BECAUSE WE DIDN'T HAVE CO-OP OPTIMIZATION OF ENERGY AND ANCILLARY SERVICES.SO YOU HAVE ALL OF THIS CAPACITY BEHIND HASSLE, SCED SEES A ENERGY IMBALANCE, AND SO THEN YOU GET ALL THIS CRAZY PRICING THROUGH ENERGY SHORTAGES.
AND THE, THE BIG REASON THAT WAS SO EXPENSIVE AND COSTLY IS BECAUSE THE, THE RATE BASE FOR ELEVATED ENERGY PRICES IS JUST A LOT MORE SIGNIFICANT THAN WHATEVER REVENUE IS COMING THROUGH THE ANCILLARY SERVICE PLAN.
WE'RE TALKING ABOUT 10,000 MEGAWATTS OF ANCILLARY SERVICES VERSUS, YOU KNOW, 80 GIGAWATTS OF LOAD.
AND SO YOU JUST END UP WITH MUCH BIGGER COST OVERRUNS WHEN YOU HAVE CAPACITY HELD OUT OF THE MARKET IN A WAY THAT ISN'T REALLY, UH, THE WAY THE MARKET'S GOING TO WORK WITH REAL, REAL TIME CO OPTIMIZATION.
I MEAN, THERE YOU'RE PLUGGING IN DEMAND CURVES AND JUST GOING IN SHORT, GOING SHORT ON RESERVES WHEN THEY GET PROHIBITIVELY EXPENSIVE.
AND SO A LOT OF THE, THE ECRS STORY WOULD'VE BEEN KIND OF RESOLVED WITH REAL TIME CO OPTIMIZATION, WHICH CONGRATULATIONS GUYS.
I MEAN, YEAH, WE GOT ALMOST TWO WEEKS UNDER OUR BELTS
SO YOUR, YOUR POINT ANDREW, IS WITH REAL TIME CO-OP OPTIMIZATION, YOU DON'T HAVE CONCERNS OVER THIS? WELL, IT, IT ISN'T A PROBLEM WITH WHAT HAPPENED WITH ECRS, WHERE CAPACITY IS WITHHELD FROM THE SCARED'S ABILITY TO DISPATCH IT FOR ENERGY BECAUSE IT CAN BE CO OPTIMIZED BECAUSE IT CAN BE CO OPTIMIZED.
NOW WHERE THE, THE SUPPRESSION COMES FROM IS NOW YOU HAVE A LOT OF CAPACITY THAT'S KIND OF SUBOPTIMALLY ONLINE AND SPINNING AND SO, AND PROBABLY HAS LOW MARGINAL PRICES WHEN MORE OF THAT CAPACITY WOULD TEND TO BE OFFLINE, YOU KNOW, IF WE WEREN'T TRYING TO CLEAR SUCH A LARGE RANGE OF OPERATING RESERVES.
BUT DRRS CAN ALSO BE AWARDED TO OFFLINE, SO, RIGHT.
BUT WE ALL KNOW THAT THE ACTUAL STACK OF OFFLINE RESERVES THAT CAN COMMIT QUICKLY ENOUGH TO PROVIDE DRRS IS, IS RELATIVELY SMALL OR ONLINE ALREADY.
SO LIKE A LOT OF THE TWO HOUR COMBINED CYCLE CAPACITY IS ALREADY GOING TO BE ONLINE.
THEN YOU HAVE A SUBSET OF LIKE 30 MINUTE TO 60 MINUTE QUICK START STUFF, AND THEN YOU HAVE A LOT OF FOUR AND SIX HOUR STUFF THAT THE ONLY WAY FOR IT TO EARN DRS REVENUE IS GOING TO BE FOR IT TO BE ONLINE.
THE SAME ARGUMENTS WERE MADE WHEN, OR DC WAS FIRST INTRODUCED, THEY SAID THAT, HEY, YOU'RE GONNA HAVE ALL THESE OLD CLUNKERS COME ONLINE, TRY TO CUT, YOU KNOW, CAPTURE THE ONLINE RESERVES.
YEAH, I JUST THINK THAT THE, THE REVENUE WE'RE TALKING ABOUT WITH THE, UH, TRS PLUS, I MEAN THE VOLUMES ARE REALLY BIG.
WE'RE TALKING ABOUT 80 TO 140 GIGAWATTS.
I GUESS I'M KIND OF FLIPPING BETWEEN THE ANCILLARY SERVICE YEAH.
AND THE, YEAH, I MEAN, WHERE DO YOU SEE THE SAME THING RIGHT? ONLINE CAPACITY HEADROOM.
UH, SO FAIR ENOUGH TO THE POINT ABOUT THE, YOU KNOW, THE VOLUMES ON DRS PLUS, I JUST WANT TO MAKE SURE I UNDERSTAND SOMETHING.
SO I LOOKED AT THE CAPACITY FORECAST FROM THE AURORA CENTRAL CASE, AND THEN I COMPARE THAT TO THE FIVE TO 10 GIGAWATTS OF CAPACITY ADDITIONS, AND I'M SEEING LESS THAN A HUNDRED GIGAWATTS OF DISPATCHABLE GENERATION UNDER EITHER CASE.
AND THEN, BUT YOU'RE PUTTING IN A, A VOLUME FOR DRS PLUS AT 140, THERE'S A DEMAND CURVE I ASSUME FOR, UH, THE DRS PLUS.
SO IF YOU'RE GOING SHORT ON IT, YOU'RE HITTING A PRICE ACCORDING TO THAT DEMAND CURVE, RIGHT? SO THAT'S A PRETTY BIG DIAL TO TURN THAT DEPENDING ON HOW THAT DEMAND CURVE IS, YOU, YOU CAN TURN THE VOLUME AND YOU CAN TURN THE DEMAND CURVE AND WHICH ONE OF THOSE PRODUCES THE REVENUE OUTPUT YOU'RE TRYING TO GET.
YOU KNOW, IT'S KIND OF OVERDETERMINED, IF YOU SEE MY POINT.
LIKE WE DON'T REALLY KNOW WHAT THAT DEMAND CURVE LOOKS LIKE, BUT DEPENDING ON WHAT IT LOOKS LIKE, WE'RE HITTING IT SHORT SOMEWHERE, YOU KNOW? YEAH.
AND SO IN THE MODELING, AND OBVIOUSLY HERE, I THINK THAT'S WHERE THERE'S MAYBE A SLIGHT DISTINCTION TO MAKE BETWEEN THE MODELING AND NPR SINCE WE MODELED WHAT, YOU KNOW, WE THOUGHT, AND AT THE TIME WHERE THE NPR WASN'T DRAFTED TO BE SORT OF THE, THE MOST, UM, REASONABLE, REALISTIC CASE TO MODEL TO, TO ACHIEVE THOSE OUTCOMES, RIGHT? AND SO THERE'S, THERE'S TWO THINGS.
THERE'S A PROCUREMENT CURVE MM-HMM
RIGHT? AND THEN THERE WE HAD A MULTIPLIER TO THAT PROCUREMENT CURVE TO SORT OF REALLY EMPHASIZE THE HOURS.
SO THERE'S A PROCUREMENT CURVE THAT SAYS WE'RE GONNA PROCURE X AMOUNT OF GIGAWATTS FOR EACH HOUR, RIGHT? AND THAT'S WHAT VARIES BETWEEN THE 80 AND ONE 40 NUMBERS.
AND THEN BEHIND THAT THERE WAS A MULTIPLIER, WHICH IS ESSENTIALLY YOUR DEMAND CURVE THAT SAID THE BUDGET'S GONNA BE ALLOCATED MORE TO THE HOURS OF HIGHEST RELIABILITY RISK VERSUS THE HOUR OF LOWEST RELIABILITY RISK.
AND THEN FOR EACH OF THOSE HOURS THAT PAYMENT
[01:45:01]
WAS SPREAD ACROSS ELIGIBLE RESOURCES, RIGHT? UM, SO IN A WAY, IN, IN THE MODELING, THIS WAS NOT DONE THROUGH COMPETITIVE BIDDING, RIGHT? WHERE YOU COULD SORT OF BID YOURSELF DOWN TO A CERTAIN VIEW.THIS WAS SORT OF A SET PAYMENT BASED ON THE, WHAT THE DEMAND CURVE WOULD TELL YOU IN TERMS OF THE AMOUNT OF PROCUREMENT THAT DAY SLASH YOUR MULTIPLIER.
UH, THAT ALL SEEMS LIKE IT MAKES SENSE MECHANICALLY, I JUST DON'T, I'M HAVING A HARD TIME MAKING SENSE OF, OF SETTING THE PROCUREMENT VOLUME AT ONE 40, BUT THEN IN EQUILIBRIUM WE END UP WITH A HUNDRED GIGAWATTS OF CAPACITY, AND SO THAT MEANS THAT THE PRICE IS BEING SET BY GOING 40 GIGAWATTS SHORT ON THE PROCUREMENT TARGET.
AND SO THAT ALL THAT JUST SEEMS PRETTY STRANGE.
SO I, I THINK ONE THING TO CLARIFY FIRST IS THAT, YOU KNOW, THE BUDGET ULTIMATELY IS BEING SET AT THE ANNUAL LEVEL, RIGHT? SO REGARDLESS OF WHERE, YOU KNOW, THE PROCUREMENT VOLUMES ARE, THOSE WILL CONTROL LIKE THE HOURLY PRICE OUTCOMES, RIGHT? BUT IN TERMS OF HOW MUCH MONEY'S GETTING INJECTED INTO THE SYSTEM, ULTIMATELY, LIKE IF YOU HAVE A HUNDRED GIGAWATT PROCUREMENT, IF YOU HAVE 120 GIGAWATT PROCUREMENT, THAT WON'T CHANGE THE AMOUNT OF INJECTION FOR THE YEAR, RIGHT? UM, AND THEN THE SECOND THING TO TO, TO NOTE IS THAT ON THE HOURLY BASIS, UM, IT'S REALLY ABOUT TRYING TO CONSTRUCT A PROCUREMENT, UM, A, A RANGE OF PROCUREMENT TARGETS, UH, THAT BASICALLY CONCENTRATE THE PRICES, UH, AT THE TIMES OF HIGHEST NEED, RIGHT? AND THE WAY THAT WE CONSTRUCTED THAT WITHIN THE MODEL WAS TO BE, UH, SIMILAR UH, TO, YOU KNOW, WHERE THE SYSTEM GETS TIGHT, RIGHT? SO SUMMER EVENINGS, ET CETERA, WINTER MORNINGS, UH, SO ON AND SO FORTH.
UM, BUT THE PROCUREMENT VOLUMES THAT WE'RE USING FOR THOSE, FOR THOSE INSTANCES ARE BASICALLY BASED ON SORT OF WHAT YOUR MAXIMUM AMOUNT OF PROCUREMENT WOULD BE OR WHAT YOU WOULD NEED, RIGHT? SO IT COULD GO ABOVE, UH, THE TOTAL CAPACITY REQUIREMENT.
UH, I MIGHT HAVE MISUNDERSTOOD SOMETHING.
SO IS THERE A, A BUDGET THAT GOES INTO DRS PLUS THAT WILL NECESSARILY BE SPENT OR IS IT A CAP BASED ON LIKE NET CONE BASED ON WHAT HAPPENS IN THE ENERGY MARKET? YEAH, THAT'S A, THAT'S AN IMPORTANT POINT.
MAYBE WE SHOULD GO TO THAT SORT OF, THE WAY WE MODELED IT, RIGHT? JUST TO, TO BE CLEAR, SO THERE IS A BUDGET THAT ENDS UP BEING SPENT, UH, ACROSS THE YEAR THAT IT IS BASED OF LOOKING, THE BUDGET IS DETERMINED BY LOOKING AT THE PREVIOUS YEAR NET C CONE, SORRY, BY TIMES THEY RATED ELIGIBLE CAPACITY OVER A YEARLY.
SO THAT GIVES YOU A BASE ANNUAL BUDGET AND THAT BUDGET IS THEN SPREAD OUT ACROSS THE HOURS WHERE THE ANCILLARY SERVICE IS PROCURED.
SO EFFECTIVELY, RIGHT? BASED ON THE, THE WAY WE'VE MODELED IT, YOU, YOU'LL NEVER EXCEED YOUR ANNUAL BUDGET BECAUSE THAT, THAT IS HOW THE PAYMENT GETS DISTRIBUTED.
NOW, UM, THERE MAY BE A DIFFERENCE WITH HOW EXACTLY THE IMPLEMENTATION CALCULATES THE BUDGET AND DOES THE EFFECTIVELY THE, THE SPREAD OVER THE, ACROSS THE YEAR.
BUT JUST TO BE CLEAR, IN THE, THE, IN THE WAY WE'VE MODELED IT, THERE IS A BUDGET AND THEN THERE'S A MULTIPLIER TO THAT BUDGET THAT DETERMINES THE ALLOCATION.
SO THEN IF YOU PROCURE DIFFERENT VOLUMES, YOU'RE JUST DIVIDING UP THE GIGAWATTS THAT ARE GETTING PAID BY THE BUCKET OF MONEY.
UH, I THINK THAT'S ALL OF MY QUESTIONS.
YOU HAD A, A DIRECT FOLLOW UP ON THAT ONE.
SO YEAH, IF IT'S OKAY IF I JUMP THE CUTEST ON THIS ONE QUESTION.
SO, UM, I WANT, I WAS TRYING TO UNDERSTAND WHAT ANDREW WAS ASKING.
SO YOU'RE GONNA LOOK AT THE DEFICIT RELATIVE TO NET CONE FROM THE PRIOR YEAR, AND THEN YOU JUST DIVIDE THAT UP OVER, EXPLAIN HOW YOU THEN PAY THAT OUT IN THE CURRENT YEAR.
YEAH, AND SO IN THE MODELING, RIGHT, SO YOU HAVE THIS BUDGET THAT WE HAD A MULTIPLIER AS WELL THAT'S LOOKING AT DIRECTING THE BUDGET.
IF YOU ALREADY HAVE PLENTY OF RESERVE CAPACITY, RIGHT? YOU DON'T WANT TO BUDGET THAT, JUST GO UP TO INFINITY BECAUSE FOR WHATEVER REASON YOU HAVE TONS AND TONS OF CAPACITY IN RESERVE.
SO THIS IS MEANT TO, UH, AND SO THAT'S WHAT'S EXPLAINED ON THIS SLIDE, UM, IN TERMS OF HOW THIS GETS PROCURED, THAT'S THE NEXT SLIDE.
UM, SO IN THE MODELING, WE LOOKED AT BASICALLY THERE'S AN HOURLY PROCUREMENT TARGET THAT'S LOCKED IN AND THAT IS THE ONE, THE ONE THAT'S SORT OF VARYING BETWEEN THE HOURS, UM, BETWEEN THE, THE TWO NUMBERS.
AND THEN THERE'S A DEMAND CURVE THAT'S LOOKING AT PROCURING MORE OR LESS.
UH, SO WITH THE PROCUREMENT TARGET THAT'S MULTIPLYING THE BUDGET, SO ALLOCATING THAT BUDGET TO EACH HOUR BASED ON THE RELIABILITY SORT OF VIEW.
SO THE AMOUNT YOU'RE GONNA PAY OUT IS BASICALLY FIXED FROM THE PRIOR YEAR, BUT THEN YOU'RE GONNA AWARD MORE OF
[01:50:01]
IT IN CERTAIN HOURS THAN OTHERS.THAT THAT'S THE WAY WE MODELED IT AGAIN.
BUT IT'S NOT BASED ON NECESSARILY LIKE DIFFERENT HOURLY.
I MEAN, CAN YOU EXPLAIN HOW YOU WOULD MAKE THAT DECISION? SO ULTIMATELY, RIGHT, LIKE THE, THE WAY, THE WAY WE MODELED IT EACH HOUR WILL GET A CERTAIN POT OF MONEY THAT THEN GETS DISTRIBUTED ACROSS ALL THE ELIGIBLE RESOURCES HERE IN THIS SPECIFIC MODELING, THERE'S NO COMPETITIVE BIDDING WHERE, YOU KNOW, YOU COULD COME IN AND, AND BID MUCH LOWER AND SET THE PRICE MUCH LOWER.
THAT WOULD REDUCE THE BUDGET, RIGHT? THE PRICE IS FIXED, IT'S JUST LIKE A, HERE THE PRICE IS FIXED IN THE MODELING.
AND THAT'S ALSO PARTLY BECAUSE WE WERE REALLY FOCUSED ON SORT OF THE RELIABILITY AND THE COST, YOU KNOW, SORT OF EQUATION.
OBVIOUSLY, AGAIN, IN THE IMPLEMENTATION, IT, I THINK IT MIGHT LOOK A LITTLE DIFFERENT, BUT THAT'S THE WAY WE, WE, WE MODELED IT.
I HAVE MORE QUESTIONS, BUT I'LL GET BACK IN MY SPOT.
HEY, UH, TREVOR, SCA, SO I'M ACTUALLY, MY QUESTIONS ARE ON THIS SLIDE TOO, SO, UH, HAPPY WE'RE, WE'RE UNPACKING THE, THE ANNUAL BUDGET.
UM, I THINK I'LL, I'LL SAVE, UH, THE, THE QUESTIONS AROUND HOW THIS RELATES TO, TO 1310 UNTIL TILL NEXT MONTH.
BUT, UH, JUST WONDERING IF AURORA DID ANY ANALYSIS LOOKING AT DRS PLUS OR DRS WITH, UH, THE VOL BASED SHORTAGE PRICING IN COMBINATION, OR IF YOU ONLY LOOKED AT THESE MARKET DESIGNS KIND OF INDEPENDENTLY.
SO WE, WE LOOKED AT EACH OF THE MARKET DESIGNS INDEPENDENTLY AND MODELED THEM INDEPENDENTLY.
WHEN YOU SAY THE VOL BASED, UH, PRICING, YOU MEAN LIKE THE ORDC CURVE SETTING IT BEING SET AT VOL, RIGHT? RIGHT.
SO THAT, THAT'S SOMETHING WE HAD DISCUSSED.
UH, I THINK WE ENDED UP GOING WITH THE EXTENDED ORDC AT THE 5,000 CAP, MOSTLY BECAUSE, YOU KNOW, WE'RE NOT SURE THERE'S APPETITE, OR AT LEAST WE HEARD THERE MIGHT NOT BE APPETITE TO PUT AN ORDC CURVE AT $35,000.
I THINK THAT MAKES SENSE, RIGHT? NO CAPPING IT AT 5,000 MAKES SENSE.
I GUESS I'M, I'M THINKING THAT THE DRRS IS A STATUTORY REQUIREMENT, YOU KNOW, WE'RE GOING TO IMPLEMENT THAT AND THAT'S, THAT'S GONNA HELP US WITH SOME OF THE, THE LONGER DURATION RISK ASSOCIATED WITH, UH, FORECAST ERRORS OCCURRING OVER MULTIPLE HOURS.
WHEREAS THE SHORTAGE PRICING MECHANISM, YOU KNOW, IS, IS HELPFUL INTERVAL BY INTERVAL TO SIGNAL A NEED FOR CAPACITY, KIND OF AN ENERGY RISK, CAPACITY RISK DYNAMIC IN SOME, IN SOME RESPECTS IS, IS HOW I'M THINKING ABOUT IT.
UM, AND WHEN I LOOK AT, YOU KNOW, THIS BASE, THIS TYPE OF PROCUREMENT, UH, FOR THE DS PLUS, YOU KNOW, I, I'M KIND OF STRUCK THAT THE COSTS FOR THIS, THIS WERE, WERE CHEAPER THAN THE OPERATIONAL DRS, BUT I THINK THAT GETS TO THIS.
I I GUESS IT'S THE PRIOR SLIDE ACTUALLY, IF YOU DON'T MIND BACKING UP ONE SLIDE.
UH, SO THIS, THIS COMPARISON TO THE HISTORIC CONE, WELL KIND OF NATURALLY DEPRESS, I GUESS, OR SUPPRESS THESE PRICES WILL MAKE THE DRS PLUS VERSION OF THE MARKET DESIGN LOOK MORE AFFORDABLE.
IF YOU'RE LOOKING BACKWARDS AND SAY, YOU KNOW, WE'RE GONNA ADJUST THIS DOWN BASED ON HISTORICAL ENERGY AND ANCILLARY SERVICE PRICE OUTCOMES TO MAKE SURE WE'RE SENDING THE RIGHT SIGNAL IN THE UPCOMING YEAR.
I THINK IF YOU COMPARE THAT JUST TO A, A SEPARATE, A SEPARATE MARKET WHERE YOU HAVE HIGHER, UH, A MORE, UH, HIGHER SHORTAGE PRICING OR A HIGHER ORDC CURVE, THE ORDC ADJUSTED SHORTAGE PRICING MECHANISM, UM, IT'S NOT REALLY YOU'RE, YOU'RE COMPARING TWO DIFFERENT, THAT'S HOW I SAY THIS.
UM, THE, I THINK IF YOU WERE TO COMBINE, IT WOULD BE INTERESTING TO SEE THE DYNAMICS OF A SHORTAGE, UH, YOU KNOW, TAKING THOSE RDC BASED SHORTAGE PRICING ADJUSTMENTS AND COMBINING IT WITH THIS DESIGN TO SEE HOW THAT CHANGES SOME OF THE INCREMENTAL SUPPLY ADDITIONS.
BECAUSE, YOU KNOW, THESE, THESE, THESE PROGRAMS ARE NOT, THESE DESIGNS AREN'T ACTUALLY OPERATING INDEPENDENTLY, RIGHT? THEY'RE ALL, EVERYTHING'S KIND OF INTERDEPENDENT HERE.
AND SO YOU'RE, THESE SIGNALS ARE BEING EXPERIENCED BY EVERY RESOURCE AND OVER MANY INTERVALS IN THE YEAR.
SO I THINK SOME OF THOSE SYNERGISTIC EFFECTS MIGHT, YOU MIGHT BE ABLE TO TEASE OUT SOME OF THOSE SYNERGISTIC EFFECTS BY LOOKING AT THE COMBINATION.
SO, SO A COM WHAT, JUST, JUST TO MAKE SURE I UNDERSTAND RIGHT, A COMBINATION OF THE EXTENSION OF THE RDC CURVE WITH, WITH THE DS PLUS MARKET DESIGN, WITH THE DRS PLUS OR WITH THE DRS YEAH.
AND TO SEE HOW, YOU KNOW, HOW I THINK THAT WOULD, THAT WOULD PROVIDE A, ANOTHER INTERESTING DATA POINT.
AND OBVIOUSLY A LOT OF WORK WENT INTO THIS AS IS JUST LOOKING AT THE INDIVIDUAL CASES.
THAT'S A, A COMPLICATED EXERCISE, BUT, UH, JUST JUST THINKING ABOUT HOW THIS ALL KIND OF WORKS HOLISTICALLY, UM, AS, AS WE THINK ABOUT THE RELIABILITY ASSESSMENT COMING UP AND RESOURCE ADEQUACY, WE WANNA MAKE SURE THAT WE'RE THINKING ABOUT THESE SIGNALS KIND OF MORE THREE DIMENSIONALLY.
MAYBE KEITH, AN IDEA FOR FUTURE WORK, UH, AT THIS POINT.
WE'LL, WE'LL SEE, BUT I, I THINK AS, AS YOU SAID THAT, YOU KNOW, WE TRIED TO ISOLATE
[01:55:01]
THE INDIVIDUAL EFFECTS AND, YOU KNOW, TO THE EXTENT THERE ARE, YOU KNOW, COMBINED EFFECTS, THAT'S, THAT'S SOMETHING WE, WE DIDN'T HAVE THE, THE ABILITY TO DO, UM, AT THIS TIME.THERE'S THE QUICK QUESTION IN, IN THE QUEUE, I THINK FROM, FROM ANDREW, I THINK WE CAN ADDRESS QUICKLY.
IF YOU LOOK AT SLIDE 69 OLIVIER, THE QUESTION IS THE UNITS ON THE Y AXIS, OH, UM, FOR THE, UH, DEMAND CURVE, BUT THE DEMAND CURVE, I THINK THIS IS A MULTIPLIER, UH, CORRECT TIM? YEAH.
DO YOU WANNA EXPLAIN MAYBE A LITTLE BIT MORE AROUND THE DEMAND CURVE MULTIPLIER? YEAH, EXACTLY.
SO, SO IF YOU JUST MAP, IF YOU WERE TO LOOK AT THE LEFT HAND SIDE GRAPH, RIGHT, AND YOU WERE TO JUST MAP THAT TO THE COST ALLOCATION, WE'D BE PUTTING, YOU KNOW, BASICALLY HALF OF THE VALUE FOR SHOWING UP AT NOON, 12:00 PM AS WE WOULD BE AT 18 OR AT 6:00 PM RIGHT? SO WE PASSED THAT THROUGH A CURVE THAT'S BASICALLY CONCENTRATING IT, UM, TO A LEVEL THAT WE FELT WAS A BALANCE RIGHT BETWEEN NOT MAKING A PRICE SIGNAL THAT'S SUPER NARROW SUCH THAT IT ONLY SHOWS UP A HANDFUL OF TIMES A YEAR, YOU KNOW, COMPARABLE TO THE RDC, UM, BUT IS STILL SUFFICIENTLY INCENTIVIZING PERFORMANCE AT THE TIMES WHEN IT MATTERS.
SO OBVIOUSLY, RIGHT, LIKE WE SET A DEMAND CURVE FOR THE STUDY, THAT DEMAND CURVE I'M SURE WOULD BE SUBJECT TO IMPLEMENTATION IN THE STAKEHOLDER PROCESS AND MIGHT LOOK DIFFERENT, BUT THE IDEA IS TO CONCENTRATE PAYMENTS ON THE HOURS WHERE IT MAKES SENSE AND THAT DEMAND CURVE WOULD CHANGE LIKELY ON A SORT OF MONTHLY OR SEASONAL BASIS.
YEAH, MINE, I, SORRY, MINE I THINK IS PRETTY QUICK.
UM, UH, I'M THROUGH WITH THE FLU, BUT I'M JUST BEING SAFE BY THE WAY.
I, UM, UH, I THINK IT'S SLIDE 16 WHERE YOU COMPARE THE COST OF THE THREE COST OUT TO THE STATUS QUO.
DID YOU SAY 16? YEAH, WITH, WITH BATTERIES.
SO THIS IS AN ANNUAL COST, ASSUMING A 2013 WEATHER YEAR.
AND SO YOU DIDN'T RUN THESE SAME COSTS FOR THE 2023 EXTREMES, AND, AND I GUESS THE QUESTION IS WHY NOT? SO THAT'S QUESTION NUMBER ONE.
AND THEN QUESTION NUMBER TWO, AND THIS MAY HAVE ALREADY BEEN ASKED, BUT WE DIDN'T DO SORT OF A, UM, LOOKING AT ALL THE YEARS WE'VE HAD IN THE LAST 10 YEARS, AND IF WE COMPARED THESE THREE, YOU KNOW WHAT I MEAN, LIKE IT, ONE OF THEM MIGHT MAKE A LOT OF SENSE IN A CERTAIN YEAR, BUT NOT IN ANOTHER YEAR.
BUT SEEING KIND OF THE MORE LIKE MONTE CARLO SEEING THE PROPORTION MIGHT BE HELPFUL.
AND SO I'M WONDERING IF YOU DID ANYTHING LIKE THAT, IF THAT'S SOMEWHERE IN THE STUDY.
AND THEN ALSO WHY DIDN'T YOU SHOW THIS JUST FOR THE, USING THE 2023, WHETHER YEAR TO COMPARE THE THREE? UM, GREAT QUESTION.
UM, SO WE DID LOOK AT A LOT OF THESE METRICS AS PART OF THE STUDY PROCESS TO SINCE CHECK EVERYTHING.
UH, AT SOME POINT THERE WAS A, JUST A DETERMINATION THAT WE NEEDED TO KEEP THE SLIDES AND THE NUMBER OF THE AMOUNT OF INFORMATION WE'RE PUTTING THIS UNDER CONTROL.
AND I, I SAW A COMMENT ON TWITTER THAT WERE BEING PAID BY THE PAGE, SO I GUESS THIS WOULD'VE BEEN EVEN LONGER.
UH, SO, SO WE TRIED TO FOCUS ON LIKE PROVIDING THE RESULTS THAT WERE THE MOST HELPFUL AND MOST IMPORTANT.
AND, AND SO NOT EVERYTHING THAT WE MODELED IN, IN THE ENTIRE PROCESS, UH, BUT WE DID, DID LOOK AT THESE COSTS AND DIRECTIONALLY, AS YOU WOULD IMAGINE, THEY'RE, THEY'RE HIGHER RIGHT IN, IN THOSE, UH, MORE EXTREME WEATHER YEARS, RIGHT? WHERE YOU SEE HIGHER SCARCITY, HIGHER PRICING IN THE SYSTEM.
UM, WITH YOUR QUESTIONS AROUND, UH, YOUR QUESTION AROUND DIFFERENT WEATHER YEAR, UM, SO WE HAVE, WE HAVE MODELED IN, IN PAST EXERCISE, NOT FOR THIS, BUT YOU KNOW, SORT OF EVERY SINGLE YEAR IN THE LAST 15 YEARS AND SORT OF LOOKED AT THE DIFFERENCES BETWEEN YOURS.
UM, I DON'T KNOW IF WE HAVE THE, THE SLIDE IN THERE AROUND LIKE THE, THE DIFFERENT WEATHER YEARS.
UM, I THINK IT MIGHT BE, IT MIGHT BE IN THE APPENDIX.
MAYBE I'LL, I'LL TRY TO SHIFT THIS.
UH, AND SO GENERALLY IF I FIND IT, YEAH, SO THIS IS, THIS IS WHAT WE TYPICALLY DO.
UH, AND SO LOOKING AT BASICALLY THE NUMBER OF HEATING DEGREE DAYS, THERE'S NOT AN EXACT H-D-D-C-D-D METRIC.
IT'S, IT'S A BIT OF A HYBRID, SO BEAR WITH ME.
BUT, UH, IF YOU LOOK, THE BIGGER THE, THE BLUE BAR, THE MORE THE MOST COLD DAYS YOU HAD AND THE BIGGER THE RED BAR, THE, THE MOST NUMBER OF, OF HOT DAYS YOU HAD, RIGHT? IF YOU LOOK AT 22 AND 23, THEY'RE REASONABLY IN THE RANGE OF LIKE A YEAR THAT SAW FAIRLY
[02:00:01]
HOT SUMMER, RIGHT? 2011 IS STILL THE, THE TOP ONE.UH, BUT 2023 IS RIGHT AFTER, UH, AND THEN 2022 SIMILARLY SAW A FAIRLY HIGH NUMBER OF COLD DAYS.
AND THEN IN PARTICULAR THE STORM, UH, ELLIOT DAY.
SO IT IS, IT IS IN THE RANGE OF SORT OF LIKE A, A FAIRLY, FAIRLY COLD WINTER.
UH, BUT FOR THE 2022 WEATHER YEAR, WE REALLY FOCUSED ON THE, THE EVENT, RIGHT? AS OPPOSED TO LIKE THE WHOLE WINTER.
AND SO IT WAS REALLY A STORM ELLIOT TYPE OF MODELING, BUT WE DID COMPARE WITH OTHER YEARS TO SEE SORT OF WHERE THEY FIT.
AND AGAIN, I THINK YOU ALREADY ANSWERED THIS QUESTION.
YOU GUYS WERE NOT HIRED TO LOOK AT SOME COMBINATION.
SO DRS AS AN ANCILLARY SERVICE AND SOME EXTENSION OF ALREADY C THAT'S NOT SOMETHING YOU LOOKED AT TO SEE IF THAT WOULD, WE, WE DIDN'T DO IT AS PART OF THIS STUDY.
I THINK THERE'S THE CHOICES WE HAD TO MAKE ON THE NUMBER OF THINGS WE WERE GOING TO MODEL ULTIMATELY.
AND, AND WE, WE'VE, I THINK THE, THE DECISION WAS THAT ISOLATING THE EFFECT WAS GONNA BE MOST HELPFUL FOR THE DEBATE.
UH, THAT SAID, I TOTALLY RECOGNIZE THAT, THAT ALL THOSE DIFFERENT COMBINATIONS COULD BE HELPFUL TO MODEL.
SO, YOU KNOW, IF, IF WE GET TO THAT AT SOME POINT, WE'RE HAPPY TO DO IT.
BUT THAT WASN'T PART OF THE STUDY.
KATIE, KIND OF MAYBE RELATED TO WHAT CYRUS JUST ASKED, I'M JUST WONDERING FOR CONTEXT IF YOU COULD EXPLAIN LIKE WHAT THE ASSIGNMENT WAS LIKE, WHAT YOU GUYS WERE ASKED TO DO SPECIFICALLY.
SO, SO ESSENTIALLY TWO THINGS.
THE FIRST THING WAS ASSESS THE RELA THE RESOURCE ADEQUACY SITUATION IN ERCOT IN THE SORT OF 2030 TIMEFRAME.
SO LOOK AT UNDER THE CURRENT MARKET DESIGN, HOW ARE WE DOING IN A 2030 YEAR IN TERMS OF, YOU KNOW, SORT OF NORMAL WEATHER CONDITION VERSUS MORE EXTREME WEATHER CONDITIONS.
SO DO 20 22, 20 23 AND DETERMINE LIKE, DO WE HAVE A PROBLEM? SO THAT WAS THE FIRST PART OF THE ASSIGNMENTS IS DO WE EVEN HAVE A PROBLEM THAT WE NEED TO SOLVE? I GUESS THE ANSWER TO THAT IS WE DO SEE LOAD SHE IN THOSE SORT OF MORE EXTREME WEATHER UNDER THE STATUS QUO MARKET DESIGN.
AND THE SECOND PART OF THE ASSIGNMENT WAS LOOKING AT THE DIFFERENT MARKET DESIGN PROPOSALS THAT ARE OUT THERE, RIGHT? UM, AND OBVIOUSLY THESE THREE THAT WE SELECTED, UH, WITH THE AIRCO TEAM, WHAT ARE THE IMPLICATIONS FOR RELIABILITY AND SYSTEM COSTS? SO THEN RUNNING UNDER THESE DIFFERENT MARKET DESIGN, LOOKING AT WHAT KIND OF GENERATION GETS INCENTIVIZED IN THE MARKET, AND THEN RERUNNING THOSE DIFFERENT SCENARIOS TO SEE WHETHER OR NOT THERE IS AN IMPROVEMENT IN TERMS OF THE RELIABILITY, IN TERMS OF HOW MUCH LOAD SHED WE SEE AS WELL AS IN TERMS OF SYSTEM COSTS, YOU KNOW, BASED ON PRICING, OBVIOUSLY THAT COMES OUT OF THE MODEL, UM, WITH THE TWIST THAT WE ALSO EXAMINE THE IMPACT OF LOAD FLEXIBILITY RIGHT UNDER SB SIX OR MORE GENERALLY UNDER ALL OF THE INITIATIVES WE'RE DOING.
AND SORT OF HOW DO, UH, A LARGE AMOUNT, A LARGE NUMBER OF DATA CENTERS COMING ONLINE IMPACT THIS WHOLE STORY.
AND SO OBVIOUSLY THE MODELING WITH DATA CENTER FLEXIBILITY, HOPE THAT'S, SO, SO THE CONCEPT, I GUESS WHAT I'M TRYING TO UNDERSTAND IS THE CONCEPT OF DRRS PLUS, IS THAT SOMETHING YOU GUYS CAME UP WITH OR IS THAT SOMETHING THAT ERCOT DESCRIBED TO YOU? YEAH, SO I MEAN, TO BE TRANSPARENT AND, AND WE'VE DISCUSSED IT WITH KEITH, RIGHT? LIKE, SO WE WE'RE NOT NECESSARILY ADVOCATING FOR ANY OF THESE MARKET MECHANISMS. SO WE SAT DOWN WITH ERCOT AND MADE A SCOPE OF WHAT WE WERE GOING TO MODEL.
AND ERCOT TOLD US WE WOULD LIKE YOU TO MODEL THESE THREE MARKET DESIGNS.
AND SO WE WENT AND MODELED THE THREE MARKET DESIGNS.
WOULD YOU EXPECT THAT IF YOU, UM, COME UP WITH ANY WAY OF PAYING OUT PAYMENTS TO ACHIEVE NET CONE, WHETHER IT'S THROUGH DRS PLUS OR ANY OTHER MECHANISM, THAT THE RESULTS WOULD BE THE SAME? I MEAN, ISN'T THE OUTCOME OF YOUR STUDY JUST BASED ON PAYING NET CONE, NOT HOW IT'S PAID, NOT THE SPECIFIC MARKET DESIGN, BUT JUST THE FACT THAT YOU'RE PREDETERMINING THE MARKET REVENUE? YEAH, THAT'S A FAIR QUESTION.
SO I, I DO THINK THERE IS A DIFFERENCE.
SO THERE'S, I GUESS THERE'S TWO THINGS.
THERE'S THE DIFFERENCE IN THE MODELING AND THERE'S PROBABLY THE DIFFERENCE IN REALITY.
AND OBVIOUSLY THE DIFFERENCE IN REALITY IS NOT COMING THROUGH THE MODELING PROBABLY.
BUT UM, BUT THE DIFFERENCE IN MODELING IS, IS THE WAY YOU ALLOCATE THOSE PAYMENTS, RIGHT? SO THERE IS A DIFFERENCE IN TERMS OF ALLOCATING THOSE PAYMENTS ON AN HOURLY BASIS TO ELIGIBLE RESOURCES BECAUSE YOU'RE GONNA HAVE CONCERNS AND, AND QUESTIONS OF AVAILABILITY FOR BATTERY STATE OF
[02:05:01]
CHARGE.UM, AND SO IT'S NOT THE SAME TO DO IT IN THIS WAY THAN TO SAY, RUN A CAPACITY MARKET WITH AN ANNUAL AUCTION WHERE EVERYONE GETS PAID ON ANY LCC DERIDED BASIS, THOSE PAYMENTS WOULD COME THROUGH DIFFERENTLY, RIGHT? AND WITH, WITH SORT OF DIFFERENT, UH, DIFFERENT ALLOCATIONS.
SO YOU'RE USING WHATEVER THE DRRS ELIGIBILITY REQUIREMENTS ARE, ULTIMATELY, WHETHER IT INCLUDES BATTERIES OR NOT, YOU'RE JUST USING THOSE ELIGIBILITY REQUIREMENTS TO DETERMINE ESSENTIALLY WHAT CAPACITY CAN RECEIVE THESE PAYMENTS WELL AND MODULATED WITH SORT OF WHAT WE OBSERVE IN TERMS OF AVAILABILITY OF RESOURCES IN DIFFERENT SEASONS, IN DIFFERENT TIMES, RIGHT? SO LIKE SPRING AVAILABILITY FOR THERMAL GENERATIONS NOT THE SAME AS SUMMER AVAILABILITY, RIGHT? SO THERE'S, THERE'S LIKE, THERE'S A COMPONENT OF WHO IS AVAILABLE AT SPECIFIC TIMES THAT IS INCLUDED IN THE MODELING, WHICH IS DIFFERENT FROM LIKE SPREADING IT, SAY LIKE PAYING EVERYONE ON JANUARY 1ST JUST BASED ON THEIR RATED CAPACITY WOULD NOT YIELD THE SAME RESULT AS LOOKING AT EXACTLY HOW MUCH YOU GET PAID AT SPECIFIC HOURS, RIGHT? SO THAT, I THINK THAT'S A DIFFERENCE, I WOULD SAY.
UM, SO THERE MAY BE AN, THERE MAY BE DIFFERENCES IN HOW EXISTING RESOURCES ARE PAID IN A FORWARD CAPACITY MARKET AS COMPARED TO THIS, BUT THE PREDICTIONS THAT YOU'RE MAKING ABOUT NEW BUILD ARE FUNDAMENTALLY IN BOTH SCENARIOS WOULD BE BASED ON JUST ADDING THE ADDITIONAL CONE BASED REVENUES TO THE MARKET.
I MEAN, THERE'S A HIGH CORRELATION, I MEAN, FUNDAMENTALLY, RIGHT? JUST TO BE TRANSPARENT, THERE'S A HIGH CORRELATION BETWEEN THE AMOUNT OF MONEY YOU DISTRIBUTE TO GENERATORS AND THE BUILD YOU GET, RIGHT? YOU CAN'T ESCAPE THAT.
AND I THINK THAT IS PROBABLY VALID REGARDLESS OF SORT OF THE MECHANISM IN WHICH YOU DISTRIBUTE THOSE FUNDS.
UM, SO I THINK THAT YEAH, THAT'S A FAIR STATEMENT.
UM, THE OTHER PART I WOULD SAY IS THERE IS A DIFFERENCE FUNDAMENTALLY FOR, UM, WHETHER RESOURCES GET BUILT IN TERMS OF THE PREDICTABILITY AND THE, THE SORT OF TRANSPARENCY AND HOW A MECHANISM WORKS, RIGHT? SO FOR A MECHANISM TO HAVE SOME SORT OF IMPACT, PEOPLE NEED TO BE ABLE TO VALUE THOSE REVENUE STREAMS WITH SOME DEGREE OF CERTAINTY, OTHERWISE, THOSE REVENUE STREAMS ARE SORT OF CONSIDERED TOO RISKY AND THEY'RE EITHER HEAVILY DISCOUNTED OR THEY'RE JUST ZEROED OUT IN A FINANCING MODEL, RIGHT? AND SO, AND MAYBE THAT IS A KEY DIFFERENCE WITH PCM TO, TO SAY A, A, A, A WORD THAT'S, THAT'S SOMEWHAT SENSITIVE HERE.
BUT, UM, PCM THERE WAS A LOT OF CONCERNS AROUND HOW PREDICTABLE AND HOW, UM, BANKABLE THE REVENUE STREAMS WOULD BE, RIGHT? THE IDEA HERE IS THAT THERE IS A CERTAIN DEGREE OF PREDICTABILITY THAT WOULD ALLOW YOU TO GO RAISE DEBT AGAINST THOSE REVENUE STREAMS TO FINANCE YOUR PROJECTS.
OBVIOUSLY THAT IS NOT A DIFFERENCE NECESSARILY COMES IN THE MODELING, UM, SORT OF THE BANKABILITY OF, UH, OF THE SPECIFIC MECHANISM.
BUT I WOULD SAY THAT'S, THAT'S ANOTHER DIFFERENCE IN HOW YOU DISTRIBUTE THOSE FUNDS.
SO YOU GUYS ARE FAMILIAR WITH ALL THE PCM THE STUDY AND ALL OF THAT.
YOU GUYS HAVE LOOKED AT THAT BEFORE? I MEAN, WE, I'VE READ IT, YEAH.
UM, I WAS JUST CURIOUS IF YOU REMEMBER WHAT THE NET COST OF THE PCM WAS SUPPOSED TO BE AND HOW THAT WOULD COMPARE TO WHAT YOU'RE PREDICTING HERE? UM, TO BE FRANK, I DON'T REMEMBER.
UM, SO I, I COULDN'T TELL YOU, BUT I'M SURE WE CAN LOOK IT UP.
UM,
SO, UM, AND SOMEBODY MENTIONED EARLIER THAT, YOU KNOW, IF YOU'RE ON OUTAGE, YOU DON'T GET THE PAYMENTS, BUT OTHERWISE, UM, IF IT'S A, I GUESS GOING BACK TO THE RELEASE FACTOR CONCEPT, IF IT'S A RELEASE FACTOR OF A HUNDRED AND ALL ELIGIBLE RESOURCES GET PAID THE PAYMENTS, HOW WOULD YOU DETERMINE WHO'S AVAILABLE FOR PURPOSES OF GETTING THE PAYMENTS? IF YOU COULD JUST ANSWER THAT SPECIFICALLY AND MAYBE YOU DON'T KNOW.
'CAUSE MAYBE THAT'S A 1310 DISCUSSION, BUT I'M JUST CURIOUS HOW YOU GUYS ARE CONSIDERING THAT.
WELL, SO AGAIN, I MEAN, NOT, NOT TO HIDE UNDER THE SORT OF THIS IS HOW WE MODELED IT, BUT YOU KNOW, AT THE END OF THE DAY THIS, THAT'S WHAT WE DID, RIGHT? LIKE, I MEAN, 1310 MIGHT DETERMINE SPECIFIC ELIGIBILITY CRITERIA FROM OUR PERSPECTIVE IS SORT OF RATED CAPACITY.
SO, YOU KNOW, CAPACITY ON OUTAGE IS NOT ELIGIBLE CAPACITY THAT'S ELIGIBLE IS OBVIOUSLY LOOKING AT LIKE CAPACITY THAT'S AVAILABLE, RIGHT? SO IF THERE'S CONSIDERATION AS TO IF YOU'RE A BATTERY, IF YOU DON'T HAVE ENOUGH STATE OF CHARGE, YOU WOULDN'T BE ELIGIBLE FOR THE FULL PAYMENT, EVEN IF YOU'RE FOUR HOUR BATTERY VERSUS, UM, I WOULD SAY A THERMAL ASSET THAT, YOU KNOW, MAY NOT BE ABLE TO PERFORM AT NAMEPLATE, BUT BELOW, RIGHT? LIKE IT WOULD ONLY BE ELIGIBLE TO THE CAPACITY IT CAN PROVIDE TO THE MARKET.
BUT YOU DON'T HAVE TO HAVE LIKE OFFERED IN OR NOT FOR THIS MECHANISM, NOT FOR THE DRS PLUS,
[02:10:01]
YOU JUST HAVE TO BE AVAILABLE TO GENERATE EXISTS AND NOT BE ON OUTAGE.UM, HANG ON, LEMME SEE IF I HAVE ANY OTHER QUESTIONS.
AND SO I'M JUST SEEING THE ANDREW THING POP UP, BUT, UH, YEAH, SO THERE'S ELIGIBILITY REQUIREMENTS, RIGHT? SO IF YOU CAN'T START FAST ENOUGH, THEN NO, YOU'RE NOT ELIGIBLE.
BUT THAT'S DIFFERENT FROM AVAILABILITY, THAT'S DIFFERENT FROM AVAILABILITY.
IT'S BASICALLY TECHNICAL REQUIREMENTS BE ACCREDITED, YOU'RE, YES, EXACTLY.
UM, I WAS JUST WONDERING IF YOU, I MEAN, I THINK I'M KIND OF DISTILLING IN MY MIND THE DISTINCTION HERE, BUT YOU SAID EARLIER THAT DRRS, LIKE THE OPERATIONAL COMPONENT IS LIKE A CLASSIC ANCILLARY SERVICE AND DRRS PLUS IS A PLUS SERVICE.
UM, I'M JUST WONDERING IF YOU COULD IN YOUR MIND, EXPLAIN WHAT YOU THINK THE DIFFERENCE IS? YEAH, THAT'S A GOOD QUESTION.
SO ULTIMATELY, THE REASON THAT SORT OF DRS PLUS IS CALLED AN ANCILLARY SERVICE, RIGHT? IS IT IT IS BEING PROCURED ON AN HOURLY BASIS, LIKE AN ANCILLARY SERVICE AND IT AIMS TO SERVE RELIABILITY GOALS.
NOW, OBVIOUSLY IT'S NOT A TRADITIONAL ANCILLARY SERVICE IN THE SENSE OF LIKE THE PURPOSES THAT A REGULATION SERVICE OR A SPINNING RESERVE OR NON SPINNING RESERVE SERVICE SERVE, WHICH IS MORE SORT OF LIKE IMMEDIATE OPERATIONAL, UH, NEEDS.
AND IT'S, IT'S NOT, IT'S, IF I'M NOT TRYING TO PUT WORDS IN YOUR MOUTH, BUT IT'S NOT TO MEET AN OPERATIONAL NEED OR A CONTINGENCY RISK, IT'S ABOUT MEETING A, UH, LONG-TERM PLANNING RESERVE REQUIREMENT, RIGHT? I THINK THAT'S FAIR TO SAY THAT, UH, THE ANCILLARY SERVICE WOULD BE DESIGNED TO PROVIDE A RESOURCE ADEQUACY SIGNAL AS OPPOSED TO LIKE RAMP UP A BUNCH OF CAPACITY IF A NUCLEAR PLANT TRIPS, RIGHT? IT'S A, IT'S A DIFFERENT USE CASE.
NOW, I THINK IT'S FAIR TO ALSO RECOGNIZE THAT WE HAVE FIVE ANCILLARY SERVICE, THEY'RE ALL SERVING DIFFERENT USE CASE.
WE CAN ALL BUCKET THEM INTO ANCILLARY SERVICE, BUT OBVIOUSLY WHAT REGULATION DOES IS VERY DIFFERENT THAN WHAT RS OR ECRS DO.
UH, AND SO IN THIS CASE, I THINK THAT SORT OF DRS PLUS IS SERVING A DIFFERENT PURPOSE WITH A DIFFERENT PROCUREMENT SIZE, UH, BUT IT IS PROCURED WITH THE SORT OF SAME LOGIC IN MIND OF AN HOURLY PROCUREMENT THAT'S DETERMINED BY ERCOT.
UM, SO IN THAT SENSE, IT IS AN ANCILLARY SERVICE IN SORT OF THE BUDGET SIZE OF THE PROCUREMENT SIZE.
YOU KNOW, IT LOOKS OBVIOUSLY A LITTLE DIFFERENT.
IT'S MEANT TO ACHIEVE A REVENUE GOAL, NOT AN OPERATIONAL GOAL.
I THINK, WHERE'S THE LIE, KEITH, DO YOU WANNA COMMENT? YEAH, I THINK THERE'S, THERE IS SOME DIFFERENCE HERE.
I THINK 1310, UM, ACCOMPLISHES BOTH SIMULTANEOUSLY IN TERMS OF OPERATIONAL AND RESOURCE ADVOCACY.
AND YOU KNOW, I THINK WHEN, WHEN YOU THINK OF SOMETHING LIKE THE ORDC THAT WE TALKED ABOUT EARLIER, OR ADCS, THE, THE MECHANISM THERE IS, YOU KNOW, WE DON'T, YOU KNOW, WE DON'T GET BENT OUTTA SHAPE THAT THAT'S DESIGNED TO ENHANCE RELIABILITY, RIGHT? AND SO WHAT WE'VE DONE IS USED THE ANSLEY SERVICE TO ENHANCE RELIABILITY AS OPPOSED TO AN ORDC, WHICH USES ENERGY TO ENHANCE RELIABILITY.
AND THE PRIMARY REASON FOR THIS, AS, AS OLIVIER MENTIONED EARLIER, IS, IS THE TARGETING OF THE RELIABILITY OF THE RESOURCES.
AND THAT'S WHERE THE ORDC AND I THINK IN THE EXAMPLE THAT THEY SHOW FAILS, IS THAT YOU'RE GOING TO GET MORE OF EVERYTHING, BUT EVERYTHING DOES NOT ACHIEVE RELIABILITY THE WAY THAT THE TARGETED DRS DOES.
AND SO WE, WE LOOK AT THE MECHANISM OF DRS, THE ELIGIBILITY REQUIREMENTS THAT COME WITH IT, WHICH ARE UNIQUE TO DRS, AND WE ALLOW THAT TO, UM, ENHANCE THE RELIABILITY LIKE THE ORDC, BUT THROUGH ANCILLARY SERVICES INSTEAD TO TARGET WHAT WE NEED TO ACHIEVE THAT OUTCOME.
SO IN OUR MIND, IT'S, IT'S USING THE ANCILLARY SERVICES TO ACHIEVE RELIABILITY.
SO IT ISN'T AN ANCILLARY SERVICE PLUS, AND I'M NOT AWARE OF ANY OTHER MARKET THAT HAS A TOOL QUITE LIKE THAT, AND THAT'S FINE.
UM, UH, AND THAT'S UNIQUE, BUT, UH, IT, IT, BASED ON THEIR STUDY, WE'LL ACHIEVE THAT.
SO THAT'S MY, MY 2 CENTS ON THIS ITEM.
I HAVE ONE MORE MECHANICAL QUESTION.
SO, UM, IF YOU'RE, AND THIS MAY BE THAT I JUST DON'T UNDERSTAND THIS MULTIPLIER THING STILL, BUT IF YOU'RE FIGURING OUT HOW MUCH REVENUE YOU'RE GONNA PAY BASED ON NET CONE FROM THE PRIOR YEAR, IT MAY NOT ALIGN WITH, IT'S NOT PROBABLY GOING TO ALIGN WITH, UM, PROGRESS TOWARD ACHIEVING NET CONE IN THE ENERGY MARKET IN THE CURRENT
[02:15:01]
YEAR.SO YOU COULD, YOU COULD HAVE SCENARIOS WHERE YOU DIDN'T HAVE MUCH ENERGY MARKET PRICING IN ONE YEAR.
SO THE DELTA TO ACHIEVE NET CONE IS VERY BIG.
AND THEN IF YOU'RE TRYING TO MAKE THAT UP IN THE CURRENT YEAR, BUT YOU DO HAVE A LOT OF ENERGY MARKET REVENUE IN THE CURRENT YEAR, YOU COULD HAVE BIG SWINGS IN REVENUE FROM YEAR TO YEAR.
YOU'RE NOT GONNA HAVE NET CONE ON AN EACH YEAR BASIS, RIGHT? YOU COULD HAVE VERY HIGH ENERGY PRICES AND BASICALLY A BACK PAYMENT FOR NET CONE FROM A MILD YEAR FROM THE PRIOR YEAR THAT THAT'S RIGHT.
SO IT'S A LAGGING, IT'S A LAGGING INCENTIVE, RIGHT? IT'S NOT FORWARD LOCKING, IT'S, IT'S LAGGING FROM THE PREVIOUS YEAR.
UM, JUST TO PUT THIS INTO CONTEXT, TO BE TOTALLY FRANK, I THINK THAT THAT'S JUST THE NATURE OF THE
SO, SO THE AIRCRAFT MARKET BY NATURE IS VERY VOLATILE.
YOU WOULD, YOU, YOU DON'T NECESSARILY DAMPEN THAT VOLATILITY, THAT VOLATILITY STILL EXISTS.
WELL, AND IT COULD IN, IT COULD IN FACT EXACERBATE IT IF YOU'RE COMPOUNDING IT ALL IN ONE YEAR, YOU, YEAH.
I MEAN, AND YOU COULD, YOU COULD DEFINITELY SEE CASES WHERE YOU'RE PROBABLY COMBINING A, UH, AN EXPENSIVE YEAR WITH A BIG PAYMENT, WHICH MEANS THEN THE FOLLOWING YEAR WOULD HAVE ALMOST NO DRS BUDGET, RIGHT? BECAUSE THE BUDGET'S BASED ON THE NET CON, IF YOU'VE MET YOUR NET CON AND BUDGET'S IS ZERO, AND THEN YOU COME BACK TO, AND THEN THE NEXT YEAR WOULD COME BACK TO WHATEVER THE, THE OR DC SIGNAL IS, OR A IS DC SIGNAL IS.
SO, SO YEAH, TO, TO YOUR POINT, THERE COULD BE YEARS WHERE IT'S COMBINED AND THEN YEARS WHERE IT'S, IT'S THE OPPOSITE.
UM, AND SO YOUR FOUR, I THINK THIS IS SIMILAR TO THE QUESTION ANDREW WAS ASKING.
SO THE 400 MILLION ESTIMATE THAT IS ON AVERAGE AT EQUILIBRIUM, RIGHT? AND WELL, SO THIS IS BASED ON OUR 2030 MODELING, RIGHT? SO THAT WE TOOK THAT SPECIFIC YEAR.
NOW THE WAY WE'RE MODELING BY THAT POINT IN 2030, JUST FAIRLY EQUILIBRIUM POINT.
SO IF YOU WERE TO TAKE 20, 29, 20, 31, THE ANSWER, THE AVERAGE INSERT DOESN'T CHANGE.
WE'VE ACTUALLY CHECKED, RIGHT? SO WE LOOKED AT OTHER YEARS AND THE ANSWER DOESN'T FUNDAMENTALLY CHANGE YEAR ON YEAR ON AN EQUILIBRIUM SITUATION.
HEY, UM, SO THIS IS WARREN LASHER, UM, SPEAKING FOR TGA.
UH, I'VE GOT A, UH, SIMILAR TO A LOT OF PEOPLE.
I HAVE SEVERAL QUESTIONS HERE.
FIRST I'D LIKE TO SAY I ALSO APPRECIATE THE MODELING.
UH, YOU KNOW, I APPRECIATE THE EFFORT.
UM, THERE'S A LOT OF REALLY GOOD USEFUL INFORMATION HERE.
UM, UH, IMPORTANTLY THOUGH, IT'S IMPORTANT TO KNOW KIND OF THE LIMITATIONS OF THE MODELING APPROACH IN ORDER TO UNDERSTAND WHAT YOU CAN ACTUALLY TAKE AWAY FROM THAT.
UM, SO, UH, WITH THAT BEING SAID, UH, GOING BACK A BIT, AND I APOLOGIZE, WE'RE STEPPING BACK.
THE FEW THINGS THAT ARE STILL NOT CLEAR TO ME, THE RETIREMENTS TO START WITH FIXED IN EVERY SCENARIO THAT YOU LOOKED AT, UM, WHAT EXACTLY IS THE HEURISTIC HERE? UM, SO YEAH, I THINK THIS WAS PROBABLY A DESIGN CHOICE IN THE STUDY TO KEEP THINGS.
I, I THINK MAYBE A CHALLENGE IN THE MODELING, RIGHT, IS IF EVERYTHING IS MOVING AROUND ALL THE TIME, IT MAKES CASES INCREASINGLY HARDER TO COMPARE.
UM, AND SO ONE DESIGN CHOICE WE MADE ON THE RETIREMENTS IS SORT OF HAVE A, A FIXED SET OF RETIREMENT AND THEN RUN DIFFERENT SCENARIOS ON IT.
UM, OBVIOUSLY RECOGNIZING THE LIMITATION THERE, RIGHT? AND IN A SCENARIO WHERE, UM, YOU HAVE REALLY HIGH PRICES, HIGH ECONOMIC INCENTIVES, YOU COULD SEE LESS RETIREMENTS, UH, PEOPLE KEEPING THEIR UNITS ONLINE, UH, LONGER.
UH, AND SO, SO OBVIOUSLY THAT WOULD IMPACT THE RETIREMENT STACK YOU SEE HERE.
UM, I, I MEAN, I JUST AS A COMMENT, I THINK IT'S INTERESTING THAT THERE ARE COMPANIES, UH, PROBABLY APOLOGIZE ABOUT THE MICROPHONE.
THERE ARE COMPANIES MAYBE REPRESENTED IN THIS ROOM.
I HAVEN'T SEEN EVERYBODY THAT ARE, UM, UH, MAKING EFFORTS TO RESTART CLOSED NUCLEAR REACTORS IN THE CONTEXT OF WE'RE SAYING WE'RE GONNA LOSE 10,000 MEGAWATTS OF RIGHT THERMAL CAPACITY IN THE NEXT FIVE YEARS HERE IN THE YORK SYSTEM TO, TO BE FAIR, THERE'S NONE TO RESTART IN TEXAS, RIGHT? MM-HMM
THERE ARE NONE TO RESTART IN TEXAS.
BUT I MEAN, THAT BEING SAID, THE, YOU KNOW, THE MARKET LOOKS A LOT DIFFERENT THAN WHAT WE'RE SEEING HERE RIGHT AT THE MOMENT.
I MEAN, WE ALSO HEAR CPSS IS DESPERATELY TRYING TO RETIRE THEIR UNITS, UM, AND WOULD REALLY MUCH LIKE TO RETIRE THEM, BUT POINT TAKEN, RIGHT? LIKE I, I DO THINK THERE
[02:20:01]
IS AN ECONOMY CASE FOR SOME OF THESE UNITS TO, TO STAY ONLINE, AND I THINK THAT'S TOTALLY FAIR.UM, I, I DO THINK THE SORT OF NARRATIVE OF NOW EVERYONE'S GONNA TRY, TRY TO KEEP ALL THEIR PLANS ONLINE FOREVER, PROBABLY HITS A WALL WHEN YOU THINK, WHEN YOU'RE THINKING ABOUT 60-YEAR-OLD PLANTS ABSOLUTELY.
THAT ARE SORT OF FALLING ABOUT.
SO I THINK THERE'S SORT OF A MIDDLE GROUND WHERE I THINK SOME OF THEM STAY ONLINE AND THEN SOME OF THEM RETIRE.
I MEAN, OLD PLANTS ARE LIKE OLD PICKUP TRUCKS.
YOU KEEP THEM UNTIL SOMETHING VERY, VERY EXPENSIVE NEEDS TO GET REPLACED.
AND THEN, BUT IT'S, THAT'S FAIR ENOUGH.
UM, SO NEXT QUESTION ON THE GENERATION EXPANSION, UH, THERE WAS A COMMENT BY SHAMS IN THE CHAT, YOU MAY NOT HAVE SEEN WHERE HE WANTED THOSE NUMBERS TO BE PROVIDED BY SCENARIO.
UM, UH, BUT I, MY QUESTION IS SOMEWHERE IN THE DECK IT SAYS THAT YOU BASICALLY INCLUDED EVERYTHING WITH AN INTERCONNECTION AGREEMENT, BUT THEN JUST AT THE BEGINNING OF THIS PRESENTATION, UM, YOU MADE IT SOUND LIKE AT SOME POINT IN THE QUEUE YOU HAVE A, A BIT OF YOUR OWN TECHNICAL REASONING THAT GOES IN THAT MAY DE DE DETRACT FROM THE NUMBER OF UNITS IN THE INTERCONNECTION AGREEMENT PROCESS.
SO CAN YOU EXPLAIN THAT TO ME? YEAH, THAT, THAT'S A GREAT POINT.
UH, AND BY THE WAY, IF, IF, IF SOMEONE WANTS THE FULL, UH, NUMBERS ON THE CAPACITY STACK, WE'RE, WE'RE HAPPY TO PROVIDE THAT.
THAT'S FAIRLY STRAIGHTFORWARD REQUEST.
UM, SO JUST TO BE TRANSPARENT, UM, SO WE DO NOT INCLUDE ALL GENERATION WITHIN INTERCONNECTION AGREEMENT IN THE STACK.
WE LOOK AT SUCCESS FACTORS OF DIFFERENT GENERATOR TYPES, RIGHT? UM, I THINK AT THIS POINT, EVERYONE IN THIS ROOM IS FAMILIAR WITH THE FACT THAT NOT EVERY SINGLE SOLAR OR WIND PLANT IN THE QUEUE IS COMING ONLINE, UH, OR IS EVER GOING TO COME ONLINE.
UM, AND SO WE LOOK AT SORT OF BIOTECHNOLOGY AND ERCOT IS ACTUALLY PUBLISHED VERY USEFUL SETS OF NUMBER ON THAT EXACT TOPIC OF SORT OF WHAT'S THE PERCENTAGE SUCCESS FACTOR WE SEE AT DIFFERENT PHASES IN THE QUEUE.
AND SO WE DO OUR OWN ANALYSIS TO DETERMINE SORT OF HOW MUCH CAPACITY THAT IS ON A GIVEN YEAR.
THERE'S ALSO A CURVE BECAUSE EVEN THOUGH EVERYONE WOULD LIKE TO BE CONNECTED IN 2026, THAT'S NOT ALWAYS THE CASE.
AND SO WE SORT OF COME UP WITH OUR OWN ESTIMATE OF INTERCONNECTION QUEUE PROJECTS THAT COME ONLINE IN THE NEXT THREE, FOUR YEARS.
UM, UH, YOU MENTIONED JUST IN YOUR PREVIOUS STATEMENT THE DELAY FACTOR.
I THINK THAT'S ALSO IN YOUR SLIDE DECK.
UM, HOW IS THAT DELAY FACTOR CALCULATED? WHAT, WHAT YOU'RE COMPARING THE COMMERCIAL OPERATION DATE TO WHAT EXACTLY WHEN YOU COME UP WITH THE DELAY FACTOR? YOU MEAN IN ADDING IT PROJECTS FROM THE INTERCONNECTION QUEUE? YEAH, SOMEWHERE IN THE, IN THE DECK IT TALKS ABOUT THE FACT THAT YOU DID A SCENARIO WITH A DELAY FACTOR OR, OH, SORRY, SO THAT THAT'S SEPARATE.
SO WE ALSO ANALYZED A, UM, SO SPECIFIC SUPPLY CHAIN ISSUES.
SO THAT WAS ONE SPECIFIC SENSITIVITY WE RAN.
KNOWING THAT RIGHT NOW, AND THIS WAS PARTICULARLY FOCUSED ON THERMAL, RIGHT? THERE'S A, THERE'S BEEN A LOT OF DEBATE AROUND THERMAL SUPPLY CHAIN CHALLENGES IN THE FACT THAT IT'S HARDER TO GET A TURBINE, UH, IN AND, AND A-A-C-O-D PLANT FOR A THERMAL PLANT QUICKLY.
AND SO WE LOOKED AT HOW, AND, AND MAYBE TIM YOU CAN SPEAK A LITTLE BIT ABOUT IT MORE, BUT BASICALLY WHAT WAS THE AVERAGE DELAY WE WERE SEEING IN TERMS OF COD BETWEEN WHAT WAS ANNOUNCED IN THE QUEUE AND, AND EFFECTIVE, YOU KNOW, PLANS COMING ONLINE AND THEN APPLIED THESE SAME DELAYS TO LIKE OUR VIEW TO SHOW SORT OF A DELAYED SUPPLY VIEW AND, AND THE IMPACT OF THAT ON RELIABILITY.
SO TO BE CLEAR, YOU WERE COMPARING THE ACTUAL COMMERCIAL OPERATION DATE WITH THE, UH, STATED WITH THE OPTIMISTIC COMMERCIAL OPERATION DATE THAT WAS INITIALLY ENTERED IN THE RIO SYSTEM BY THE GENERATION DEVELOPER HONOR THERE.
YEAH, SO WE WERE JUST LOOKING AT, YOU KNOW, HISTORICAL GIS REPORTS 2019 THROUGH 2024, YOU KNOW, AND IF YOU SAY YOU'RE GOING LIVE IN JULY, 2022, AND THEN WE LOOK AT THE JULY, 2022 GIS, AND YOU'RE STILL, YOU KNOW, YOU'RE NEW COMMERCIAL OPERATION DAYS, JULY, 2023, YOU KNOW, WE LOOK AT THE PROPORTION OF THOSE THAT ARE GETTING PUSHED BACK BY AT LEAST A YEAR.
ONE QUICK CLARIFYING POINT IS THAT, UH, THIS ANALYSIS HERE, UH, WAS SEPARATE FROM THE ANALYSIS THAT'S UNDERPINNING MOST OF THE RESULTS THAT WE'VE SEEN.
THIS WAS MORE SO TO STRESS TEST THE IMPACT OF SUPPLY LAGS AND HOW THAT COULD BE AN ADDITIONAL, UH, CHALLENGE, UH, ESPECIALLY UNDER THE HIGH DEMAND GROWTH SCENARIO.
EXCEPT FOR THE FACT THAT JUST A FEW MINUTES AGO YOU SAID THAT THOSE OUTAGES WERE ALSO, THE, THE DELAYS WERE ALSO SOMETHING THAT WAS CONSIDERED IN THE HOW YOU, UM, UH, REDUCED THE AMOUNT OF INTERCONNECTION AGREEMENT UNITS THAT COME ONLINE IN YOUR MAIN SCENARIOS.
[02:25:01]
SO JUST TO CLARIFY, IT WOULD BE ADDITIVE ON TOP OF THAT, CORRECT.YOU LOOKED AT A WINTER STORM ELLIOT CASE FOR THE, THE ANALYSIS OF CONSUMER IMPACT.
UH, WHAT WERE YOUR THERMAL RESOURCE OUTAGE ASSUMPTIONS IN THAT SCENARIO? YEAH, SO ERCOT PUT OUT A LOT OF NICE INFORMATION POST WINTER STORM ELLIOT, INCLUDING, UH, OUTAGES BY DIFFERENT FUEL TYPES.
UH, SO WE BASICALLY TOOK THE PORTION OF THE THERMAL FLEET THAT WAS ON OUTAGE, UH, DURING THAT EVENT IN 2022, AND WE SCALED THAT UP, UH, SO THAT IN 2030, UH, WE WERE ASSUMING THE SAME RATE OF OUTAGES BY, UH, TECHNOLOGY.
SO NO IMPROVEMENT FROM THE WEATHERIZATION EFFORT BY ERCOT AND NO IMPROVEMENT BASED ON THE FACT THAT YOU'RE GETTING NEW UNITS ONLINE AS A PART OF SOME OF THESE SCENARIOS.
OUR VIEW IS THAT THE THERMAL, UH, YOU KNOW, COMPARED TO YURI PARTICULARLY, UH, WEATHERIZATION IMPACTS WERE, I THINK AT THE TIME VIEWED AS A SUCCESS, UH, DURING WINTER STORM ELLIOT MM-HMM
UM, AND SO IT WAS OUR VIEW THAT IT WAS A REASONABLE ASSUMPTION TO KEEP IN LINE WITH, UH, THE OUTAGE RATES THAT WE SAW, ESPECIALLY FOR CONSISTENCY WITH THE OTHER INPUT ASSUMPTIONS.
AND TO BUILD ONTO THAT, UH, SUCCESS LEVEL, UH, THERE WERE NO SCARCITY CONDITIONS DURING ACTUAL WINTER STORM ELLIOT.
UM, WELL, THERE WERE VERY HIGH PRICES DURING WINTER STORM ELLIOT WELL OVER THAT DOESN'T, THAT DOESN'T 2000, YEAH, THAT'S A FEATURE.
SO, I MEAN, SOME OF THESE QUESTIONS ARE LEADING INTO THE FACT THAT I THINK THERE'S A LITTLE BIT OF, UM, MISMATCHING HERE.
AND YOU KNOW, THIS IS VERY SIMILAR TO WHAT WE SAW IN THE E THREE REPORT.
YOU KNOW, OBVIOUSLY THEY ARE DIFFERENT STUDIES, DIFFERENT PRODUCTS, UM, BUT THERE'S A MISMATCH BETWEEN KIND OF THIS CONCEPT OF LONG-TERM EQUILIBRIUM AND 2030.
UM, AND SO YOU'RE SAYING WE'RE GONNA LOSE ALL THESE UNITS, WE'RE NOT GONNA GET WHAT'S ACTUALLY IN THE INTERCONNECTION QUEUE BASED ON OUR ANALYSIS BASED ON MARKET DESIGN, AND ALL OF THIS IS GONNA COME TO FRUITION IN THE 2030 TIMEFRAME.
UM, SO KIND OF, I HAVE A QUESTION TO YOU.
DO YOU CONSIDER THIS TO BE A LONG-TERM EQUILIBRIUM STUDY, OR DO YOU CONSIDER THIS TO BE A FORECAST FOR WHAT WE ARE GOING TO SEE IN 2030 IF WE SEE WINTER STORM ELLIOT? BECAUSE I WOULD ARGUE THAT IF WE WANT A FORECAST OF WHAT WE'RE GONNA SEE IN 2030, WE SHOULD LOOK TO THE CDR.
UM, THAT'S AN INTERESTING QUESTION.
UM, I, I THINK THE CDR R HAS A SPECIFIC METHODOLOGY THAT'S THAT'S, THAT'S PART OF THE CDR.
AND SO I'M CERTAINLY NOT GONNA GONNA GO AND, AND COMMENT ON THE CDR R METHODOLOGY.
UM, I THINK THE FORECASTING, RIGHT, IN GENERAL, AND THAT'S PRETTY MUCH WHAT WE SPEND OUR TIME DOING, UH, FOR DEVELOPERS OF PROJECTS FOR BANKS, IS, IS, IS A TOUGH EXERCISE, RIGHT? JUST BECAUSE NO ONE HAS A CRYSTAL BALL.
AND FUNDAMENTALLY, YOU HAVE TO MAKE ASSUMPTIONS AS TO WHAT THE WORLD WILL LOOK LIKE, AND THEN ALL OF THOSE ASSUMPTIONS YOU COULD PICK APART AND, AND EVERYONE CAN PROBABLY HAVE A DIFFERENT WORLDVIEW ON, ON EACH OF THESE ASSUMPTIONS.
SO, SO WE SPEND A LOT OF TIME COMING UP WITH ASSUMPTIONS THAT ARE BOTH, UM, SORT OF REASONABLE GIVEN MARKET CONDITIONS AND WHAT WE'RE SEEING AS WELL AS ASSUMPTIONS THAT ARE SOMEWHAT CONSERVATIVE, RIGHT? UM, AND, AND THAT'S PROBABLY A PRODUCT OF WHAT OUR FORECASTING IS MEANT TO DO, RIGHT? IT'S MEANT TO SUPPORT INVESTMENT DECISION IN GENERATION.
IT'S MEANT TO SUPPORT PEOPLE LOOKING AT, AT INVESTING AND, AND WORKING IN THE TEXAS MARKET.
AND FUNDAMENTALLY, PEOPLE NEED TO KNOW WHAT THEY CAN COUNT ON FOR THESE INVESTMENTS AS OPPOSED TO SORT OF FANTASY FORECAST THAT WHERE THEY COULD MAKE A LOT OF MONEY, BUT THAT THE SORT OF PROBABILITY OF THAT HAPPENING IS, IS QUITE LOW.
AND SO FUNDAMENTALLY, WE TEND TO HAVE A SLIGHTLY CONSERVATIVE SPIN ON THIS FORECASTING, WHICH I THINK COMES THROUGH WHEN YOU LOOK AT, UH, RETIREMENT ASSUMPTION, UH, ALSO COMES THROUGH WHEN YOU LOOK AT, UH, THE AMOUNT OF GENERATION THAT COMES THROUGH FROM THE INTERCONNECTION QUEUE.
UM, YOU COULD BE A LOT MORE OPTIMISTIC, BUT WHAT WE ALSO TEND TO FIND IS IF YOU'RE TOO OPTIMISTIC, YOU'RE PROVEN WRONG PRETTY QUICKLY, UM, YOU KNOW, CODS TEND TO BE OPTIMISTIC.
UH, PEOPLE THINK THEY'RE GONNA, UM, THEY'RE GONNA BUILD PROJECTS A LOT FASTER THAN THEY ACTUALLY DO.
AND SO ALL OF THAT TOGETHER, RIGHT? WE END UP BUILDING A FORECAST.
I WOULD STILL QUALIFY A FORECAST THAT THAT IS LOOKING AT SORT OF THE MOST PROBABLE OUTCOME UNDER THE SET OF ASSUMPTIONS WE RUN.
THE PART THAT IS BASED ON THE EQUILIBRIUM, RIGHT? IS HOW MUCH NEW CAPACITY YOU INCENTIVIZE FROM THE PRICE SIGNALS YOU PUT IN THE MARKET, UM, OR SORT OF FUNDAMENTAL BELIEF, RIGHT? UH, IS THAT ULTIMATELY PEOPLE LOOK AT PRICE SIGNALS AND MAKE INVESTMENT DECISION BASED ON PRICE SIGNALS, OR AT LEAST THEIR, THEIR EXPECTATIONS OF PRICE SIGNALS,
[02:30:01]
RIGHT? SO FORECASTS ESSENTIALLY LIKE, LIKE OURS, AND THAT IS A REASONABLE METRIC AND A REASONABLE EXPECTATIONS OF WHAT KIND OF DECISIONS PEOPLE ARE GOING TO LOOK AT MAKING.UM, BECAUSE THEN, THEN ESSENTIALLY THE, THE NEXT STEP IS, YOU KNOW, YOU TAKE THOSE FORECASTS AND YOU GO RAISE, YOU KNOW, $500 MILLION IN FRONT OF A, A BANK IN WALL STREET TO, TO BUILD YOUR PROJECT, AND THEY'RE GONNA ASK THE EXACT SAME QUESTIONS.
AND SO ALL OF THESE FORECASTS ARE ESSENTIALLY SERVING THAT PURPOSE OF, YOU KNOW, WHAT IS THE EXPECTED REVENUE FOR MY PROJECT AND CAN I EXPECT TO BUILD IT? SO I DON'T KNOW IF THIS WAS A DIRECT ANSWER TO YOUR QUESTION, BUT TRYING TO SORT OF, YOU KNOW, PORTRAY THE, THE APPROACH AND THE PHILOSOPHY IN TERMS OF WHERE IT'S, WHAT IT'S TRYING TO DO.
YEAH, NO, I COMPLETELY AGREE AND I APPRECIATE THE ANSWER, EVEN THOUGH I'M NOT SURE IT WAS A DIRECT RESPONSE.
BUT, BUT AGAIN, I APPRECIATE THE ANSWER VERY MUCH.
I, AND I THINK FROM MY PERSPECTIVE, UM, UH, THE 2030 TIMEFRAME, UH, IS, YOU KNOW, IT'S A RELEVANT DATA POINT, IT'S A RELATIVE POINT TO KIND OF PUT DOWN A MARKER, BUT IT ALSO IS A POINT AT WHICH IT KIND OF HAS THE, THE, THE, THE, THE DUAL BENEFIT, UM, OF, UM, BEING CLOSE ENOUGH THAT PEOPLE WILL FEEL LIKE THEY NEED TO BE CONCERNED, BEING FAR ENOUGH THAT IT FEELS LIKE, OH, WE CAN MAKE A DIFFERENCE.
UM, AND SO IT DOES FEEL LIKE IT IS SPECIFICALLY CHOSEN TO INFLUENCE THE REGULATORY DECISION THAT'S GOING ON.
UH, I HAVE A COUPLE OTHER ISSUES.
UH, IF I, IF I CAN JUST COMMENT ON THAT.
I, I THINK IT WASN'T A QUESTION, BUT PLEASE GO AHEAD.
NO, NO, IT IS, BECAUSE IT IS, RIGHT? LIKE ULTIMATELY THE REASON WE CHOSE 2030 IS BECAUSE IT IS THE YEAR THAT HELPS INFORM THE REGULATORY PROCESS, RIGHT? LIKE, IF WE CHOOSE 2027, NO ONE REALLY CARES.
'CAUSE EVERYTHING'S ALREADY BAKED IN, RIGHT? AT THAT POINT, WE'RE JUST, YOU KNOW, MAKING A BET FOR THE FORWARD MARKET.
UH, IF WE CHOOSE 2035, IT'S PROBABLY TOO LATE REALLY FOR, UH, PEOPLE TO REALLY CARE.
AND SO THAT'S EXACTLY WHY WE CHOSE 2030 TO BE TRANSPARENT.
UM, AND I WOULD ALSO SAY THAT I, YOU KNOW, I, I FEEL LIKE THE TERM LONG-TERM EQUILIBRIUM, UM, IS KIND OF LIKE WHEN YOU SEE DR.
STRANGE TALK ABOUT THE MIRROR DIMENSION, IT DOESN'T REALLY EXIST.
UM, IN THE ERCOT MARKET, 22,000 TO 2010 WAS THE ERA OF COMBINED CYCLES RADICALLY CHANGING THE MARKET.
AFTER THAT, IT WAS THE ERA OF WIND COMPLETELY CHANGING THE MARKET AFTER THAT 20 16, 20 20 TIMEFRAME.
IT WAS THE ERA OF SOLAR COMPLETELY CHANGING HOW WE THINK ABOUT RESOURCE ADEQUACY ON A HOT SUMMER DAY.
I MEAN, IT USED TO BE IN THE DAY IT GOT TO FOUR 30 AND YOU WERE WATCHING PRC, AND IF YOU MADE IT TO FIVE 30, YOU WERE LIKE, WHOA, THANK GOODNESS.
THERE WERE DAYS WHERE WE WERE JUST STARING AT PRC AND NOWADAYS, FIVE O'CLOCK, SIX O'CLOCK DOESN'T MATTER.
UM, NOW WE'RE IN THE ERA OF BATTERIES, UH, AND WE'RE ENTERING THE ERA OF LARGE CUSTOMER DEMANDS, SOME OF WHOM WILL BE FLEXIBLE.
AND THAT'S A, THAT'S A QUESTION THAT'S BEING DISCUSSED.
UM, IF AT ANY POINT IN THOSE PERIODS YOU HAD DONE ONE OF THESE STUDIES AND LOOKED AT THE LONG-TERM EQUILIBRIUM, YOU WOULD NOT HAVE SEEN THOSE TECHNOLOGIES COMING AND YOU WOULD'VE COME UP WITH PERFECTLY GOOD REASONABLE OPTIONS THAT POINTED BACKWARDS.
AND THAT'S KIND OF WHY THE OP MARKET WITH THE FOCUS ON PUT THE MONEY WHERE THE NEED IS, UM, AND MOVE FORWARD FROM THERE, HAS BEEN SUCCESSFUL SO FAR TO DATE.
UM, I, SO I'LL LET YOU COMMENT ON THAT IF YOU'D LIKE TO.
I THINK IT'S A VERY FAIR POINT.
UM, JUST MAYBE, MAYBE TO REEMPHASIZE SOMETHING, RIGHT, IN TERMS OF PRICE SIGNALS, RIGHT? LIKE PEOPLE WILL MAKE INVESTMENT DECISIONS BASED ON THEIR EXPECTATIONS OF FUTURE REVENUE, AND THAT'S WHERE THE LONG-TERM EQUILIBRIUM COMES FROM, RIGHT? BECAUSE ULTIMATELY WE ARE SHOWING THE RESULTS FOR 2030, BUT WHEN WE RUN THE MODEL, WE RUN IT ALL THE WAY TO 2050 TO DETERMINE WHAT THOSE EXPECTATIONS OF FUTURE REVENUES ARE AND WHAT THE PRICE SIGNAL INCENTIVIZE, BECAUSE NO ONE MAKES A DECISION TO INVEST IN A $500 MILLION ASSET BASED ON ONE YEAR OF, OF, OF REVENUES, RIGHT? AND SO FUNDAMENTALLY, THAT EXPECTATIONS OF FUTURE REVENUE IS WHAT CREATES THE INVESTMENT CASE, RIGHT? AND THIS IS ALSO WHAT WE'RE MODELING.
UH, JUST TWO, UH, FEW VERY QUICK POINTS.
UM, YOU'VE GOT SCENARIOS HERE BASED ON THE LONG-TERM LOAD FORECAST THAT IS THE ERCOT TRANSMISSION PLANNING LONG-TERM LOAD FORECAST, CORRECT? UH, WELL, IT'S ACTUALLY THE, THE, WELL, I MEAN IT, IT'S THE ONE THAT IS MANDATED BY HB 50 66, RIGHT? UH, AND, AND THAT IS THE OFFICIAL 2025 LONG-TERM LOAD FORECAST THAT ERCOT PUBLISHES AS THE LONG-TERM LOAD FORECAST.
SO IT, IT DOESN'T, I MEAN, I THINK IT'S IN LINE WITH THE TRANSMISSION PLANNING, BUT THAT, I THINK THAT'S ALSO THE ONE YOU HAVE TO USE FOR
[02:35:01]
AS A LOAD FORECAST FOR, FOR PLANNING, FOR BUDGETARY REASONS, NO, NOT THE BUDGETARY REASONS, BUT FOR THE TRANSMISSION PLAN TRANSMISSION PLANNING.RIGHT, WHICH IS A NON COINCIDENT 90TH PERCENTILE LOAD FORECAST.
BUT I DON'T THINK THERE'S ANOTHER LONG-TERM LOAD FORECAST, CDRI MEAN, THIS IS IT, RIGHT? THAT'S A, THE CDR HAS A 50 50 LOAD FORECAST THAT IS, THERE IS A, THERE IS A FORECAST IN THE CDR R AS WELL, BUT AGAIN, I THINK WE FELT THAT THE, IN TERMS OF OKAY, FAIR.
IN TERMS OF THE BOUNDS OF WHAT THE STUDY WAS DOING, WE THOUGHT THE LONG-TERM LOAD FORECAST WAS A, A REASONABLE HIGH END ASSUMPTION.
UM, UH, SO, UH, I VERY MUCH APPRECIATE YOUR DISCUSSION ABOUT BENCHMARKING AND THE FACT THAT YOUR MODEL, YOUR BENCHMARKING BACK TO, UM, UH, UH, PREVIOUS OUTCOMES.
I, I, IF THIS HAD BEEN SOMETHING THAT HAD COME ACROSS MY DESK FOR REVIEW, I WOULD'VE REALLY POKED AT THAT AND I WOULD'VE POKED AT THE REVENUE THAT WAS BEING GIVEN TO UNITS IN THE DIFFERENT SCENARIOS.
BASED ON, YOU KNOW, THE ORDC, UH, DRRS, I'M WONDERING IF YOU WOULD BE WILLING TO PUBLISH THAT DATA, UM, SO THAT WE CAN SEE, UH, REVENUE OUTCOMES.
I DIDN'T NECESSARILY SEE THAT IN THE DECK.
REVENUE OUTCOMES, MAYBE BY TECHNOLOGY TYPE IN THE DIFFERENT, UM, SCENARIOS.
UH, I THINK THAT'S A GOOD QUESTION.
I THINK IT'S PROBABLY SOMETHING WE CHECK WITH
AND MY LAST POINT HERE, I MEAN, MY, MY CONCERN IF I HAD BEEN REVIEWING THIS DOCUMENT WOULD BE THAT THE OUTCOMES THAT WE KIND OF TALKED ABOUT WITH THE RDC WITHOUT THE, UM, DRS PLUS MAY NOT BE ACCURATELY FULLY ACCOUNTING FOR THE REVENUE THAT WOULD BE PROVIDED TO GENERATORS.
UM, UH, WHICH WOULD BE, IF THAT WERE THE CASE, AND I'M NOT SAYING IT IS, BUT IF THAT WERE THE CASE, THAT WOULD KIND OF, UM, UH, SET APART THOSE SCENARIOS FROM THE DRS PLUS SCENARIO WHERE THERE'S A FIXED KIND OF, YOU'RE GETTING THIS INCOME.
UM, AND THEN I WOULD ALSO POINT OUT THAT IT'S EXTREMELY IMPORTANT FOR PEOPLE TO UNDERSTAND THAT THE MODELING YOU'RE DOING, ESPECIALLY THE DRRS AND THE ORDC EXTENSION MODELING IS ALL BASED ON $5,000 PER MEGAWATT HOUR SYSTEM-WIDE OFFER CAP, CORRECT? CORRECT.
AND THEN WHEN YOU COME BACK AFTERWARDS AND YOU SAY, LET'S COMPARE THE SCENARIOS BASED ON CONSUMER IMPACT, YOU THEN JACK THAT UP SEVEN TIMES AND YOU SAY, BUT WE'RE GOING TO ASSUME THAT THAT VOLATILITY IS NOW WORTH $35,000 PER MEGAWATT HOUR.
SO IT'S, AND, AND THE $400 MILLION OFFSET THAT'S BASED ON $35,000 PER MEGAWATT HOUR, NOT BASED ON THE SYSTEM-WIDE OFFER CAP, THE SCARCITY THAT WE KIND OF, THE VALUE OF SCARCITY THAT WE ASSUME IN THE ACTUAL REALTIME MARKET.
I MEAN, TO TOTALLY FAIR COMMENT.
UM, I THINK ULTIMATELY WE'RE SORT OF JUST FOLLOWING THE DECISIONS THAT THE PUBLIC UTILITY COMMISSION MADE, RIGHT? WHERE THEY SAID THE CAP IS GONNA BE 5,000 AND THEN THE VOL IS GONNA BE 35,000, AND SO WE'RE JUST MAKING THE SAME CALCULATIONS.
COMPLETELY UNDERSTAND, BUT IT'S NOT NECESSARILY FULLY TRANSPARENT IN THE, I MEAN, IT IS IN THERE.
WELL, I MEAN THE, I THINK THOSE, THOSE NUMBERS FROM THE PUC ARE PRETTY SET, RIGHT? YES.
WE'RE, WE'RE JUST FOLLOWING THOSE.
AGAIN, THANK YOU ALL, AND THANK YOU FOR BEING HERE SO LATE IN DECEMBER.
VERY MUCH APPRECIATE YOUR PARTICIPATION.
HEY THERE, BRIAN SAMES WITH CALPINE.
I, I WOULD LIKE TO, UM, AGREE WITH WARREN.
I'D LIKE TO SEE A BREAKDOWN OF, OF WHERE THE MONEY GOES BY RESOURCE TO, UM, JUST CURIOUS ABOUT THAT.
UM, ARE YOU GUYS FAMILIAR WITH THE, THE BATES AND WHITE REPORT? UH, SORRY, SAY THAT AGAIN? ARE YOU FAMILIAR WITH THE BATES AND WHITE REPORT FROM 2023? UH, YES.
IT, I'VE, IT'S BEEN A LONG TIME, BUT I'VE, I'VE SEEN IT.
I, I, I KIND OF CONSIDER THAT AS THE ORIGINAL FAMOUS DRRS.
YOU KNOW, WE'VE GOT LOTS OF DIFFERENT VERSIONS OF DRS NOW BETWEEN DRS MINUS DRS PLUS, BUT, UM, THAT ONE HAD SOME NUMBERS IN THERE IN TERMS OF, OF, UM, THE NET COST.
AND I'M CURIOUS IF YOU'VE COMPARED THE NET COST THAT WAS IN THAT PROPOSAL VERSUS WHAT'S ON SLIDE 15 HERE? UH, THAT'S A GREAT QUESTION.
UH, WE HAVE NOT DONE ANY COMPARISON TO THAT STUDY.
UM, THIS IS SOMETHING WE CAN LOOK AT.
YEAH, I, I'D ENCOURAGE YOU TO LOOK AT, I, UM, I THINK YOU'LL BE PLEASANTLY SURPRISED ABOUT THE COST OF THIS COMPARED TO WHAT WAS IN, IN THAT STUDY.
YEAH, I MEAN, WE, WE WOULD HAVE TO LOOK AT THE ASSUMPTIONS, RIGHT? I
[02:40:01]
MEAN, OBVIOUSLY THE TRICK WITH THE MODELING IS, UM, I'M NOT SURE EXACTLY THE, I CAN'T REMEMBER THE ASSUMPTIONS IN THAT STUDY BACK IN 2023, BUT TO BE ENTIRELY FAIR, UH, I THINK THE SYSTEM AND THE WORLD WE LIVE IN RIGHT NOW, AND THE DISCUSSION WE'RE HAVING IN 20, ALMOST 2026, ARE NOW SLIGHTLY DIFFERENT FROM THE DISCUSSION WE WERE HAVING IN 2023 WHEN IT COMES TO, UH, DATA CENTER LOAD GROWTH, UH, OVERALL LOAD GROWTH IN NARCO SORT OF RESOURCE POLES.I MEAN, EVEN BATTERY LIKE IN 2023, TO BE FAIR, THE, I, I DON'T KNOW HOW MANY BATTERIES THEY HAD IN THE SYSTEM, BUT IT WAS, YOU KNOW, WE HAD WHAT, 500 MEGAWATT ON THE AIRCRAFT SYSTEM VERSUS 15 GIGAWATT NOW.
SO, UH, I, WE'LL DEFINITELY TAKE A LOOK, BUT, UH, I'M, I'M NOT ENTIRELY SURE HOW COMPARABLE THOSE TWO ARE.
YEAH, I, I GUESS MY POINT IS THAT, UM, MY RECOLLECTION FROM 2023 WAS THAT, UH, THE, UH, THAT VERSION OF DRS WAS, WAS, UH, AS PART OF, OF A, PART OF A PACKAGE AS AN ALTERNATIVE TO PCM.
AND SO IF THIS IS SOMETHING THAT'S, UH, A LESSER COST COMPARED TO THAT NET COST, IT, IT SHOULD BE ATTRACTIVE.
YEAH, WE, WE SHOULD DEFINITELY TAKE A LOOK.
UM, SO A ACTUALLY, UH, BRIAN, I'M, I'M GLAD YOU BROUGHT UP THE BATES WHITE REPORT BECAUSE THAT, THAT HAS BEEN ONE OF THE CURIOSITIES I'VE HAD IS HOW THE DRS ASS, NOT THE AS PLUS, BUT HOW THE D-R-S-A-S COMPARES TO WHAT WAS IN THAT, SINCE THAT WAS THE, UH, THE, THE, YOU KNOW, THE FONT, THE FOUNTAIN HEAD FROM WHICH THE DRS CONCEPT SEEMED TO, UH, YOU KNOW, FLOW INTO OUR LIVES.
UM, SO YOU MAY NOT, THAT MAY BE SOMETHING FOR FOLLOW UP, UM, OR SOMETHING WE CAN TALK ABOUT IN THE, IN THE NEXT WORKSHOP ABOUT 1309 AND HOW THAT COMPARES.
UH, ONE THING THAT IS, THIS IS MORE OF A REACTION JUST TO SOME OF THE COMMENTS I'VE HEARD FROM FOLKS AND, YOU KNOW, I RECOGNIZE THAT, YOU KNOW, EVERYONE IS, AND MYSELF INCLUDED, ARE TRYING TO UNDERSTAND THE MECHANICS AND HOW THIS WILL, WILL WORK WITH THE CURRENT, YOU KNOW, ENERGY MARKET CONSTRUCT AND, UM, AND FRANKLY HOW SUSTAINABLE, UH, YOU KNOW, THAT IS.
'CAUSE WARREN, I I APPRECIATE THE POINT YOU MADE THAT LONG-TERM EQUILIBRIUM IS NOT A FIXED DESTINATION.
IT IS, YOU KNOW, ONE OF MY, ONE OF MY FAVORITE LESSONS FROM, UH, ADVANCED MACROECONOMICS IS IT'S REALLY ACTUALLY, IT FOLLOWS ALMOST LIKE FISH POPULATIONS.
AND YOU LOOK AT DYNAMICS THERE, YOU'RE CONSTANTLY CIRCLING AROUND A POINT, BUT YOU NEVER QUITE GET THERE.
AND THEN AN EXOGENOUS SHOCK KIND OF MOVES THAT OFF AND YOU START CIRCLING AROUND THAT.
AND ANYWAY, UM, IT'S A MORE FISH.
NO, AND ACTUALLY NO, IT'S WHALE POPULATION.
WE HAD MANAGED TO STAY AWAY FROM THAT
UM, BUT, YOU KNOW, SO I, I APPRECIATE THAT.
I APPRECIATE THE, THE COMMENTARY THAT, YOU KNOW, WHAT WE'RE LOOKING FOR ISN'T NECESSARILY SOMETHING THAT'S JUST HOW DO WE, YOU KNOW, GET TO 2030.
IT'S, WHAT WE SHOULD BE LOOKING AT IS WHAT IS A MARKET DESIGN THAT WILL PROVIDE LONG-TERM DYNAMIC, UH, SIGNALS THAT EVOLVE WITH, YOU KNOW, CHANGES IN, IN THE, THE MAKEUP OF THE FLEET.
AND I THINK CRITICALLY, THIS IS THE OTHER PIECE OF IT ACHIEVES THE RELIABILITY OUTCOMES THAT, YOU KNOW, THE MARKET HAS BEEN, YOU KNOW, REQUIRED TO ACHIEVE.
THAT'S, YOU KNOW, THE LEGISLATURE'S GIVEN SOME OBJECTIVES.
THERE'S A CLEAR DESIRE FOR THOSE TO BE, THOSE OBJECTIVES TO BE ACHIEVED THROUGH DISPATCHABLE GENERATION, DISPATCHABLE, THERMAL GENERATION, UH, IN PARTICULAR.
AND SO WE'RE, WE'RE TRYING TO CO OPTIMIZE MULTIPLE CONSTRAINTS AND NOT JUST, UH, NOT JUST THAT ONE, YOU KNOW, NOT JUST ONE OUTCOME OR, YOU KNOW, THE STATUS QUO OUTCOME.
THE QUESTION I HAD WAS ON THE DEMAND CURVE, UM, AND I APOLOGIZE, DIDN'T ASK THIS EARLIER, BUT YOU KNOW, YOU'VE GOT A SHAPE TO IT, AND WE'VE JUST GONE THROUGH AN EXERCISE AND, YOU KNOW, TALKING ABOUT SHAPING OF DEMAND CURVES WITH THE A STC GO LIVE, WE'RE ALL FAMILIAR WITH THE ODC SHAPING AND HOW THAT, UH, WAS PUT TOGETHER.
SO YEAH, I WAS CURIOUS IF YOU COULD EXPLAIN HOW YOU GOT TO THAT, UM, THAT S-CURVE SHAPE IN THE DEMAND CURVE AND, AND WHAT IT'S REPRESENTING.
YEAH, AND I MEAN, TIM, TIM CAN COMMENT A LITTLE BIT MORE ON THE SORT OF THE, THE DRAWING OF THE DEMAND CURVE.
UH, SO TO, TO BE COMPLETELY FAIR, RIGHT? LIKE THIS DEMAND CURVE WAS SET BY US IN THE MODELING, RIGHT? SO, SO THIS ISN'T SOMETHING THAT'S SET IN THE NPR OR YOU KNOW, SOMETHING THAT'S TAKING, SO, SO TO AN EXTENT THIS IS A MODELING ASSUMPTION AND YOU CHANGE IT, YOU'LL CHANGE A LITTLE BIT THE RESULTS.
UM, THE OTHER PART OF IT IS, THE IDEA IS THAT FOR LOW LEVEL OF PROCUREMENT, IT WOULD ONLY GIVE YOU A VERY SMALL FRACTION OF THE BUDGET.
SO IT'S REALLY THAT THIS, THE BOTTOM PART OF THE S IS REALLY LIKE, WE DON'T WANT TO DISTRIBUTE VERY MUCH MONEY AT NOON IN THE MIDDLE OF, YOU KNOW, SPRING WHERE THERE'S TONS OF SOLAR, WIND, ET
[02:45:01]
CETERA.UM, AND THEN THE STEEP PART IS THAT WE REALLY WANTED TO PICK UP RIGHT? WHEN WE NEED A LOT OF PROCUREMENT FOR THE SORT OF 8:00 PM SUNSET IN AUGUST OR THE 7:00 AM MORNING RAMP IN FEBRUARY.
AND SO IT WAS REALLY JUST SKEWING THE DISTRIBUTION MASSIVELY TOWARDS HOURS THAT WE KNOW ON A MONTHLY BASIS ARE HOURS OF HIGHER RELIABILITY RISK VERSUS HOURS THEY'RE WORTHLESS, RIGHT? YOU DON'T WANT TO BE PAYING GENERATORS TO BE THERE WHEN YOU DON'T REALLY NEED THEM TO BE THERE BECAUSE YOU ALREADY HAVE PLENTY OF CAPACITY.
SO THAT'S REALLY THE SKEWING IDEA WHERE YOU DON'T WANNA BE SPENDING MONEY.
IT, THIS IS MEANT FOR THE MECHANISM NOT TO BE JUST A, A RENT TO EXISTING GENERATION TO JUST BE THERE.
IT'S MEANT TO PAY FOR WHEN WE NEED YOU.
AND YOU DON'T WANT TO CREATE, UH, YOU KNOW, A HUGE OPPORTUNITY COST IN SAY THE, THE SHOULDER PERIODS WHERE THERE'S PLENTY OF PROBABLY SOLAR AND WIND AND BATTERY CAPACITY THAT, YOU KNOW, THERMAL GENERATION CAN TAKE OUTAGES.
UM, THAT, YOU KNOW, YOU STILL WANT TO HAVE THOSE TIME PERIODS WHERE, YOU KNOW, EVERYONE CAN TAKE CARE OF THEIR MAINTENANCE NEEDS AND NOT, YOU KNOW, BE KIND OF, OF BITING THEIR FINGERNAILS ON WHAT THE OPPORTUNITY COST IS.
UM, SO I, I CAN APPRECIATE THAT.
I GUESS THE QUESTION IS, WAS THERE A, UM, WAS THERE A PARTICULAR FORMULATION THAT WENT INTO THAT CURVE OR IS IT, UH, I RECOGNIZE IT'S A SKEW YOU'RE TRYING TO SKEW TOWARDS THE HIGHER, HIGHER VALUE PERIODS? YEAH, WE, I MEAN WE, WE CAN BE TRANSPARENT ON HOW WE BUILT IT IN THE MODEL.
BASICALLY WE USE, YOU KNOW, AS YOU SAID, SCUR FUNCTION, RIGHT? AND YOU KNOW, THE, AS OLIVIER SAID, THERE'S A NUMBER OF DIFFERENT WAYS THAT YOU COULD DRAW IT THAT ARE GONNA GET YOU DIRECTIONALLY SIMILAR RESULTS.
UH, THE MAIN THING THAT WE WANTED TO ENSURE IS THAT WE STRUCK THAT RIGHT BALANCE, AND WE DISCUSSED WITH THE ERCOT TEAM BETWEEN, YOU KNOW, DRAWING A CURVE THAT'S ONLY REALLY, UM, YOU KNOW, PROVIDING VALUE AT THE HIGHEST PROCUREMENT LEVELS AND PROVIDING A HEALTHY AMOUNT OF VALUE ON A CONSISTENT BASIS.
SO IT'S, YOU KNOW, SOMETHING THAT'S MORE PREDICTABLE, UH, MORE BANKABLE.
YOU COULD GO IN AND YOU COULD ADJUST THAT S-CURVE, YOU KNOW, AND I'M SURE ERCOT WILL DO SOMETHING LIKE THAT DURING ACTUAL IMPLEMENTATION, UH, TO DESIGN IT SUCH THAT IT STRIKES THAT RIGHT BALANCE, BUT WE ARRIVED AT A VALUE THAT WE, YOU KNOW, SINCE CHECKED WITH THEM AND THE TEAM THAT FELT LIKE IT WAS, YOU KNOW, ADEQUATELY PROVIDING THAT PREDICTABILITY WHILE STILL CONCENTRATING THE PAYMENTS DURING THE TIMES WHEN IT'S ACTUALLY, YOU ACTUALLY NEED THE STUFF TO SHOW UP.
UH, JUST WANNA MAKE SURE I'M READING THE CHARTS ON 15 AND 16 CORRECTLY, WHICH, WHICH DESIGN INCENTIVIZES THE MOST DISPATCHABLE
SO HERE, IF YOU WOULD LOOK AT THE NETS, NAME PLATES, THE BALANCE RIGHT, WOULD JUST BE SLIGHTLY POSITIVE IN THE DRS PLUS.
SO BECAUSE YOU ARE LOSING, AND I'M TALKING ABOUT THE DRS AS PLUS SCENARIO, RIGHT? OBVIOUSLY THE OTHER TWO ARE POSITIVE, SO NO, NO, NO QUESTIONS OR, OR VIRTUALLY POSITIVE.
BUT THE, THE ONE WHERE YOU ACTUALLY HAVE A, A DIFFERENCE, UH, YOU WOULD END UP WITH NET PROBABLY CLOSER TO, TO ONE GIGAWATT OF ADDITIONAL CAPACITY BECAUSE YOU ARE LOSING EFFECTIVELY FOUR GIGAWATTS OF SHORT DURATION BATTERIES IN SOLAR.
UH, AND SO THAT, THAT IS OBVIOUSLY AN CONSEQUENCE OF THAT DESIGN MECHANISM.
IT'S TAKING SOME MONEY FROM THE ENERGY MARKET, PUTTING IT INTO THE ANCILLARY MARKET TO, TO INCENTIVIZE THIS, THESE RESOURCES.
JUST WANNA MAKE SURE I WAS READING THOSE CORRECTLY.
UH, STEVE RE INVIEW CONSULTING IN MY FIRST, I'D LIKE TO START OFF WITH A DEEP, DEEP APOLOGY FOR ASKING A QUESTION QUESTION 20 MINUTES AFTER THE END OF THE MEETING.
UM, THE, UH, BUT I HOPE IT'S, UH, I HOPE IT'S, UH, ONE THAT FOLKS FIND WORTHWHILE.
CAN WE GO BACK TO THE DEMAND CURVE SLIDE PLEASE? UH, AND WHILE YOU'RE GETTING THERE, WE'RE ALL USED TO, UH, WE'VE BEEN DOING A LOT OF ANCILLARY SERVICE DEMAND CURVES AROUND HERE.
SO WE'RE USED TO ANCILLARY SERVICE DEMAND CURVES WHERE, UH, YOU KNOW, IT, IT'S, IT'S A SCARCITY SINGLE SIGNAL, AND AS YOU GET MORE SCARCE, UM, YOU, YOU GET A HIGHER PRICE.
THIS IS, THIS IS BACKWARDS, UH, FROM THAT, AND I NOTICED THAT IN THE TEXT BELOW YOU, YOU REFER TO A, A BUDGET ALLOCATION.
UM, IS IT, IS IT POSSIBLE THAT THIS IS MORE A BUDGET ALLOCATION CURVE THAN AN ANCILLARY SERVICE DEMAND CURVE THAT WE WOULD NOT, WE WOULD NOT SEE
[02:50:01]
AS MORE STUFF SHOWED UP A HIGHER PRICE PER MEGAWATT BEING PAID TO THAT STUFF.BUT THIS IS JUST REFLECTING THAT YOU ARE, UH, PLANNING ON DISTRIBUTING MORE OF YOUR REVENUE TOWARDS, UH, A HIGHER TO, TO HIGHER NEED HOURS.
I, I THINK THAT WE, WE BUILT IT AS PART OF THE MODELING HERE.
UM, YOU KNOW, TO, TO, TO REFLECT SORT OF HOW WE WERE MODELING THINGS FOR CLARITY.
I THINK, I THINK THERE'S A, I THINK IT WOULD BE FAIR TO SAY YOU COULD CONSTRUCT THE EXACT SAME MECHANISM AND JUST BUILD IT MORE LIKE A TRADITIONAL ANCILLARY DEMAND CURVES WITH, YOU KNOW, SCARCITY PRICING IN MIND AND JUST STILL HAVE THE BUDGET ALLOCATED ACROSS THE YEAR AND LOOK AT HOURS OF HIGHEST SCARCITY FOR, FOR THAT DEMAND CURVE.
SO I THINK THE WAY YOU BUILD THE ANCILLARY SORT OF SERVICE DEMAND CURVE, IT COULD GO EITHER WAY, BUT DO THE SAME THING OR SOMETHING FAIRLY SIMILAR.
SO I WOULDN'T, I WOULDN'T CALL THIS A BUDGET DISTRIBUTION CURVE, IF THIS IS, THIS IS WHERE YOU'RE GETTING TO, IT'S STILL A DEMAND CURVE.
WE JUST BUILT IT MAYBE IN A WAY THAT'S A BIT REVERSED TO, TO THE WAY YOU'RE BUILDING IT, TYPICALLY FOR ANCILLARIES.
BUT, BUT YOU COULD EQUALLY BUILD IT THE OTHER WAY, RIGHT? O OKAY.
SO IF YOU BUILT IT THE OTHER WAY, YOU'D BE BUILDING DIFFERENT ANCILLARY SERVICE DEMAND CURVES FOR DIFFERENT HOURS AND YOU'D HAVE HIGHER ANCILLARY SERVICE DEMAND CURVES FOR WHERE YOU WANTED THE CORRECT, THE CAPACITY TO SHOW UP.
AT THIS POINT, WE DON'T HAVE ANY MORE QUESTIONS IN THE QUEUE.
ALRIGHT, WELL, DO YOU WANNA WRAP IT UP? WELL, I, I DO.
THIS WAS AN ENGAGING DISCUSSION TO SAY THE LEAST.
UH, I REALLY APPRECIATE THE, THE PARTICIPATION TODAY.
THIS HAS BEEN, UM, EXCITING IN A WAY, UH, I, I GLAD TO SEE SO MUCH INTEREST AND EXCITEMENT IN D-R-S-D-R-S ANSWER SERVICE PLUS, AND WE LOOK FORWARD TO, UH, CONTINUED DISCUSSION THROUGHOUT THE STAKEHOLDER PROCESS.
SO THANK YOU TO AURORA FOR BEING HERE TODAY, AND THANK YOU ALL.