This is a free preview of a paid episode. To hear more, visit www.douglewin.comWhat steps do we need to take in Texas to foster growth in this important emerging market? To explore this question, there's no one better than John Berger, founder and CEO of Sunnova Energy. Headquartered in Houston, Sunnova is a leader in residential solar, battery storage and energy management technologies across the US John has been in the industry for about 30 years now, and he understands the value proposition and the barriers to renewable energy and distributed generation as well as anybody. I really enjoyed this episode and I hope you do too. This episode is for paid subscribers only and only the free preview will be listed publicly on podcast apps. To become a premium subscriber, please visit www.douglewin.com/subscribeTimestamps4:03 - Current trends in the power market and how that’s impacting the value of distributed generation and resilience7:49 - Bull vs bear market; changes in the power market in the past 30 years and demand growth12:25 - What policy/regulatory changes needed to grow distributed solar17:11 - Potential for greater competition in distribution and publicly available distribution resource plans, challenges of central planning23:30 - Why is Texas lagging in distributed generation?28:17 - A performance mechanism for microgrids31:06 - Sunnova’s Adaptive Retail Plan with David Energy33:51 - Virtual Power Plants36:05 - Benefits of distributed generation and VPPs to consumers; what’s holding back the growth of VPPs38:52 - How to make solar available and affordable to all 42:03 - Sunnova’s Project Hestia with the Department of Energy44:41 - Impact of Inflation Reduction Act investments on the market 46:57 - Are solar supply chains shifting to the US / North America50:07 - EVs and EV chargingShow NotesSunnovaSunnova’s Adaptive Retail Plan with David EnergyCritics Say CenterPoint CEO’s “Relationship” Influenced $818M Deal For Generators - Texas MonthlyCenterPoint spent $800M on mobile generators. Where are they post-Hurricane Beryl? - Houston Chronicle

I spoke with Jeff Goodell, the bestselling author of The Heat Will Kill You First: Life and Death on a Scorched Planet, on the dangers of heat waves, especially when in combination with other natural disasters.Timestamps4:12 - Physiological impacts of heat and how to stay safe9:13  - Scale of heat-related deaths and illness12:28 - What is a Hurricane Katrina of heat and how is heat a threat multiplier17:58 - Resilience solutions for heat and extreme weather 23:52 - Maintaining hope in the face of the realities of climate change32:43 - What will happen to frontline climate communities like Houston as climate change continues to exacerbate extreme heat and other weather conditions36:51 - Economic costs of natural disasters41:49 - Need for corporate and political leadership; reimaging how we structure cities and communities; strategies and technologies for mitigation and adaptation50:08 - Climate denialism and climate nihilism 56:51 - Recommended reading and Jeff’s future researchShow NotesThe Heat Will Kill You First: Life and Death on a Scorched Planet by Jeff GoodellBig Coal: The Dirty Secret Behind America's Energy Future by Jeff GoodellThe Water Will Come: Rising Seas, Sinking Cities, and the Remaking of the Civilized World by Jeff GoodellThe Heat Wave Scenario That Keeps Climate Scientists Up at Night - New York Times Op-Ed by Jeff GoodellObama Takes on Climate Change: The Rolling Stone Interview with Jeff GoodellBeryl was the weakest a hurricane could be. Why does it feel like Houston isn't the same? Article from Sarah Smith, Houston ChronicleHeat-related mortality in Europe during the summer of 2022 - Nature Magazine This is a public episode. If you’d like to discuss this with other subscribers or get access to bonus episodes, visit www.douglewin.com/subscribe

In the wake of Hurricane Beryl, after millions of Houstonians lost power and at least 36 people died, many of them because they didn’t have AC in the heat following the storm, Texans were once again left with many of the same questions we’ve been asking for the past decade. How do we make sure that the power stays on when a storm hits or disaster strikes? How can our state, rich in all its energy resources, have its electric grid so badly pummeled by a Category 1 storm? When we are faced with the next storm, hurricane, or cold snap, how will we make sure things are different? How can we keep critical facilities powered, vulnerable residents safe, and generally ensure a reliable and resilient grid? For CenterPoint, the utility serving most of the Houston area, the response so far has been a plan to purchase additional generators. This approach has been met with considerable skepticism, given that CenterPoint had already spent $800 million on generators that proved unusable during Beryl — a cost that continues to be borne by their customers.However, there are alternative solutions that can enhance community resilience without breaking the bank. One of the most promising is microgrids. Microgrids consist of interconnected resources like solar panels, gas generators, and batteries that connect directly to homes, facilities, and other sites, allowing them to operate independently from the main grid. Microgrids have been deployed across the state, including in Houston, and customers with microgrids during Beryl had drastically different experiences than others in the city and the region.To learn more about microgrids, and understand how they can help save lives and maintain stability in future storms, I spoke to Allan Schurr, the COO of Enchanted Rock, which has been deploying microgrids for almost 20 years. 100% of Enchanted Rock’s microgrids that were tasked with providing power during the Beryl aftermath were able to do so, including their microgrids at H-E-B stores and distribution centers, which helped ensure many Texans had access to air conditioning, food, and supplies. Allan and I dug in how Enchanted Rock’s microgrids work and what types of customers are currently served by them. We talked about how microgrids can and should be deployed at nursing homes, hospitals and other critical facilities; why the company chose to switch to natural gas generators instead of diesel fuel; the impact of state legislation on microgrid deployment; and how utilities can make it easier for all types of customers to get microgrids. This one is part of a larger series I am doing on the aftermath of Hurricane Beryl focused on microgrids. In the next few weeks, we’ll be releasing short episodes (like this one) interviewing folks working on solar microgrids and storage, understanding the impact of natural disasters on marginalized communities, the importance of DERs, and more.Thank you for being a listener and don’t forget to like, share, and leave a five star review wherever you listen to podcasts.Timestamps3:48 - How Enchanted Rock got started6:14 - What is a microgrid? How does it work?9:12 - Enchanted Rock’s partnership with H-E-B and benefits to the community12:14 - Nursing homes13:07 - Other types of clients that benefit from microgrids14:44 -Generators vs microgrid and Texas Energy Fund legislation for microgrids22:04 - Benefits of microgrids to the overall grid; microgrids as part of Virtual Power Plants26:03 - Winter Storm Uri and Enchanted Rock28:45 - Accessing gas for microgrids during severe winter weather or storms31:15 - Can you have resilience without microgrids?33:53 - How can utilities make it easier for customers to get microgrids35:42 - Enchanted Rock’s Bridge-to-Grid program and microgrids at data centersShow NotesEnchanted RockEnchanted Rock’s Bridge-to-Grid programWhat’s in a name? Preparing for long-duration outages from “anonymous” storms by Allan SchurrHouston’s post-Beryl outages highlight benefits of distributed energy from Canary MediaAfter Hurricane Beryl, Microgrid-Equipped Grocery Stores and Homes Weathered Outages for More Than a Week in Houston from Microgrid KnowledgeCenterPoint spent $800M on mobile generators. Where are they post-Hurricane Beryl? - from the Houston Chronicle This is a public episode. If you’d like to discuss this with other subscribers or get access to bonus episodes, visit www.douglewin.com/subscribe

During Winter Storm Heather this past year, battery storage reduced power costs in Texas by $750 million. This finding was only one of many in a report released by Aurora Energy Research this May, “The Role of Battery Energy Storage Systems in the ERCOT Market.”  The report is one of the most in-depth analyses to date on the impact of battery storage, which has grown exponentially in Texas over the past decade, even surpassing growth in markets with major battery subsidies. In today’s episode, I’m joined by Olivier Beaufils, one of the report’s primary authors and the lead of the ERCOT Market team at Aurora Energy Research, a company focused on providing power market forecasting and analytics for energy markets around the world.Olivier was gracious enough to go through the key takeaways of the report and break down its findings with me. We discussed how the study calculated system’s savings during Heather, whether these types of savings can be expected in the future, how batteries are pairing with natural gas and renewables in the ERCOT market, the potential of long-duration energy storage, and more. Olivier also detailed what happened in Texas on September 6th – wherein ERCOT called an Energy Emergency Alert level 2, the first time it did that since Winter Storm Uri – and how batteries kept the system running and Texans’ lights on.We recorded this episode prior to Hurricane Beryl's impact on Texas. While we don’t discuss its specific impacts and aftermath here, we are preparing a series of episodes in the coming weeks that will focus on Hurricane Beryl, its effect on Houston, and strategies for improving Texas’s resilience against future hurricanes and extreme weather events. Despite not directly addressing Beryl, this episode remains pertinent to these conversations, particularly as home battery storage systems are crucial tools for community safety and resilience.I learned a lot from this report and from the conversation. After listening, I encourage you to take a look at the report as well; it is free and publicly available. Thank you for being a listener and don’t forget to like, share, and leave a five star review wherever you listen to podcasts.Timestamps3:17 - About Aurora Energy Research4:25 - Summary and key takeaways from Aurora’s Battery Energy Storage Systems in the ERCOT Market report9:02 - Pace in battery storage growth: what spurred it12:18 - How Aurora calculated battery savings during Winter Storm Heather and if this level of savings is typical17:57 - Batteries as ancillary services 21:55 - Shift in scarcity hours in TX, the duck curve, and how batteries are helping solve peak demand challenges28:09 - Findings on peaker plants, natural gas, and the need for flexible resources33:05 - Demand growth37:42 - What happened on September 6th and how batteries kept the system running41:23 - Long duration energy storageShow Notes:Aurora Energy ResearchRole of Battery Energy Storage Systems (BESS) in the ERCOT Market - report from Aurora Energy ResearchEnergy Unplugged Podcast from Aurora Energy Research This is a public episode. If you’d like to discuss this with other subscribers or get access to bonus episodes, visit www.douglewin.com/subscribe

I speak to energy experts from all facets of the energy industry and the one thing that keeps coming through in these conversations these days is the need for better planning to deal with rising load growth, extreme weather, higher costs and emissions. As I think about these problems, I try to think of people who I can learn from and that will help my audience understand these issues as well.My guest on the podcast this week, Hala Ballouz, has as much experience with these challenges as anyone I know. She's the president and CEO of Electric Power Engineers (EPE), and she's built a rapidly growing team of over 200 power system engineers and energy professionals with a singular focus, holistically forming the electric grid to enable a resilient, affordable, and carbon-free energy future. Hala envisions a future grid that is decentralized, incorporating microgrids, distributed generation, demand response, grid edge technologies, all enabling consumers to participate in energy markets and contribute to system reliability and resiliency. It's a very compelling and exciting vision and few describe it better than Hala. In our conversation, we talked about the need for better planning to integrate large loads and new supply and demand resources on the grid including AI data centers, but also distributed energy resources and electric vehicles, and other grid edge technologies on the demand side. We explore the challenges and opportunities in building a clean and resilient energy grid, highlighting necessary technology and regulatory innovation needed to address congestion and curtailment issues. Hala also underscores something that isn't often talked about in these kinds of conversations. It's not all technology and technocratic fixes. There's also an increasing importance and significance for stakeholder engagement and the need to redefine reliability and resiliency requirements in ways that are meaningful and understandable to the general public. I really enjoyed this conversation and I can’t wait to hear your thoughts about it. Don’t forget to check the show notes where we have links to Hala’s writing!Timestamps3:48 - Hala’s background, about Electric Power Engineers (EPE), and women in engineering6:20 - Changes to the ERCOT grid in the past 20 years9:40 - Expected loads and opportunities for load shifting 16:11 - What is the grid edge and decentralization of the grid20:04 - Reimagining the grid, bi-directional power flow, and distributed energy sources (DERs)23:35 - What are the barriers for tapping into DERs and increasing reliability and the importance of better grid planning29:27 - Importance of and trends in distribution resource planning, including “8760” analyses34:46 - Optimizing current transmission and planning for the future40:25 - Utilities turning down new loads outside the ERCOT market, how to avoid this in Texas, and challenges to central planning46:47 - Grid enhancing technologies (GETs)48:23 - Storage as a transmission and distribution asset 54:09 - Solving transmission congestion59:01 - Hydrogen1:00:13 - Existing government policies and regulatory structures that need to change for a reliable, clean grid and the importance of stakeholder engagement1:03:26 - Hala’s vision for the grid in 5 to 10 years1:08:13 - The need for strategy to reduce costs and engage ratepayers Show NotesHala’s Engineering Transition NewsletterA Thousand Points of Light by Hala BallouzElectric Power Engineers (EPE) This is a public episode. If you’d like to discuss this with other subscribers or get access to bonus episodes, visit www.douglewin.com/subscribe

This is a free preview of a paid episode. To hear more, visit www.douglewin.comHi All: We initially distributed this episode on Wednesday. Unfortunately, due to technical issues with Substack, some of our subscribers did not receive it. If you've already downloaded the episode, there's no need to do so again. For those who have yet to receive it, we apologize for the delay and hope you enjoy the episode.To understand what's happening, you have to know how you got there. You need a sense of the history. That's one of the reasons I've enjoyed interviewing past Commissioners and Chairs of the Texas Public Utility Commission (PUC). At the Texas Power Podcast, I interviewed Pat Wood and Barry Smitherman, both former PUC Chairs. Here at Energy Capital, the very first episode I hosted Commissioner Will McAdams, who had just stepped down from the PUC a few weeks prior. These episodes remain incredibly valuable for anyone interested in Texas energy. Today, I have the pleasure of speaking with Becky Klein, who served as a Commissioner on the Public Utility Commission from 2001 to 2002 and as Chair from 2002 to 2004. Her tenure was pivotal as the restructuring of the Texas electricity market, initiated by the legislature in 1999, was being implemented. Becky played a critical role in this transformative period.After leaving the PUC, she served as both Vice Chair and Chair of the Lower Colorado River Authority, which owns several gigawatts of power plants. She is also an active board member for various water utility and energy companies And if all that isn't enough, Becky founded and serves on the board of the Texas Energy Poverty Research Institute (TEPRI), an organization dedicated to reducing Texans’ energy burden and ensuring energy is accessible and affordable for all Texans.There are few people better situated to help understand what's happening today, especially as we grapple with rising demand, a rapidly changing resource mix, evolving market design, technology breakthroughs, an affordability crisis, and much more. During the episode, Becky shared her past work on Texas energy and regulatory matters and her vision for the future of the Texas grid. We discussed her work with TEPRI and considered the challenges and opportunities in the energy transition to alleviate energy poverty. We also explored strategies for ensuring reliable electricity access, enhancing affordability, and the evolving role of oil and gas within the broader energy landscape.This podcast is for paid subscribers only and only the free preview will be listed publicly on podcast apps. For details on how to listen to the full episode on your favorite podcast app, please refer to this information from Substack. If you like the episode, please don’t forget to recommend, like, and share on Substack, Apple Podcasts, Spotify, or wherever you listen. I look forward to hearing your thoughts; don’t hesitate to share them with me and fellow listeners in the comments. Thank you for listening and for being a subscriber! Transcript, show notes, and timestamps are below.Timestamps3:02 - Becky’s background7:26 - How Texas’ market restructuring came to be 8:49 - The central problems facing the Texas grid and the importance of first principles 11:58 - What Becky believes are the principles guiding the ERCOT market today 14:39 - Becky’s views on what caused Winter Storm Uri and how that shapes our energy market and grid goals17:33 - What would are the most important things we can do to improve reliability and affordability23:49 - The need for more clean dispatchable or clean firm resources, energy efficiency, and operational flexibility26:48 - How can the energy transition benefit everyone not just some and the work of the Texas Energy and Poverty Institute (TEPRI)37:17 - What are the metrics we should use to assess the impact of energy efficiency programs for low-income households; Energy Poverty Index41:10 - Should Texas reinstate the System Benefit Fund?45:48 - TEPRI’s new program with the Bransfield Community Development Corporation48:00 - Current and expected trends in oil and gas during the energy transitionShow NotesThe Texas Energy and Poverty Research Institute (TEPRI)2024 Community Voices in Energy Survey – Statewide Report from TEPRIFebruary 2023 Senate Business and Commerce Hearing featuring testimony from former PUC Chair Becky Klein. Her testimony begins at 7:51.Burden of the global energy price crisis on households. Study published in Nature Energy.EIA Energy Insecurity Data

This is a free preview of a paid episode. To hear more, visit www.douglewin.comTo understand what's happening, you have to know how you got there. You need a sense of the history. That's one of the reasons I've enjoyed interviewing past Commissioners and Chairs of the Texas Public Utility Commission (PUC). At the Texas Power Podcast, I interviewed Pat Wood and Barry Smitherman, both former PUC Chairs. Here at Energy Capital, the very first episode I hosted Commissioner Will McAdams, who had just stepped down from the PUC a few weeks prior. These episodes remain incredibly valuable for anyone interested in Texas energy.Today, I have the pleasure of speaking with Becky Klein, who served as a Commissioner on the Public Utility Commission from 2001 to 2002 and as Chair from 2002 to 2004. Her tenure was pivotal as the restructuring of the Texas electricity market, initiated by the legislature in 1999, was being implemented. Becky played a critical role in this transformative period.After leaving the PUC, she served as both Vice Chair and Chair of the Lower Colorado River Authority, which owns several gigawatts of power plants. She is also an active board member for various water utility and energy companies And if all that isn't enough, Becky founded and serves on the board of the Texas Energy Poverty Research Institute (TEPRI), an organization dedicated to reducing Texans’ energy burden and ensuring energy is accessible and affordable for all Texans.There are few people better situated to help understand what's happening today, especially as we grapple with rising demand, a rapidly changing resource mix, evolving market design, technology breakthroughs, an affordability crisis, and much more. During the episode, Becky shared her past work on Texas energy and regulatory matters and her vision for the future of the Texas grid. We discussed her work with TEPRI and considered the challenges and opportunities in the energy transition to alleviate energy poverty. We also explored strategies for ensuring reliable electricity access, enhancing affordability, and the evolving role of oil and gas within the broader energy landscape.This podcast is for paid subscribers only and only the free preview will be listed publicly on podcast apps. For details on how to listen to the full episode on your favorite podcast app, please refer to this information from Substack. If you like the episode, please don’t forget to recommend, like, and share on Substack, Apple Podcasts, Spotify, or wherever you listen. I look forward to hearing your thoughts; don’t hesitate to share them with me and fellow listeners in the comments. Thank you for listening and for being a subscriber! Transcript, show notes, and timestamps are below.Timestamps3:02 - Becky’s background7:26 - How Texas’ market restructuring came to be 8:49 - The central problems facing the Texas grid and the importance of first principles 11:58 - What Becky believes are the principles guiding the ERCOT market today 14:39 - Becky’s views on what caused Winter Storm Uri and how that shapes our energy market and grid goals17:33 - What would are the most important things we can do to improve reliability and affordability23:49 - The need for more clean dispatchable or clean firm resources, energy efficiency, and operational flexibility26:48 - How can the energy transition benefit everyone not just some and the work of the Texas Energy and Poverty Institute (TEPRI)37:17 - What are the metrics we should use to assess the impact of energy efficiency programs for low-income households; Energy Poverty Index41:10 - Should Texas reinstate the System Benefit Fund?45:48 - TEPRI’s new program with the Bransfield Community Development Corporation48:00 - Current and expected trends in oil and gas during the energy transitionShow NotesThe Texas Energy and Poverty Research Institute (TEPRI)2024 Community Voices in Energy Survey – Statewide Report from TEPRIFebruary 2023 Senate Business and Commerce Hearing featuring testimony from former PUC Chair Becky Klein. Her testimony begins at 7:51.Burden of the global energy price crisis on households. Study published in Nature Energy.EIA Energy Insecurity Data

In April, ERCOT's Regional Planning Group published projections that ERCOT's electricity demand may double by 2030 compared to 2021. Initially anticipating a peak demand of about 90 gigawatts by the end of the decade, ERCOT now expects a staggering 152 gigawatts, up from approximately 75 gigawatts at the beginning of the decade. This would mark the largest growth rate in the decade since the post-World War II era. There are many factors driving the surge, but one of the most significant is the rise of Bitcoin mining and AI data centers. The size and speed and scale of AI data center growth that has emerged over the last year has been surprising to say the least, even to close industry observers.However, one company foresaw this trend and built its business model around addressing data center energy needs. That company is Crusoe and on today’s episode I had the pleasure of speaking with Cully Cavness, Crusoe’s co-founder, President, and Chief Operating Officer. He has a robust background in energy and, along with Chase Lochmiller, founded Crusoe in 2018.What sets Crusoe apart is their innovative approach to siting and powering data centers. They locate them near oil and gas drilling operations that have a lot of venting and flaring of methane gas. They also site data centers near constrained renewable developments. This means that they are using energy or power that would otherwise be merely wasted and instead converting it into useful electricity to power data centers. With the example of flaring and venting, this can reduce emissions by as much as 70%.I talked about all of this with Cully and more, including how data centers could be flexible loads, the efficiency of AI chips, reducing emissions through methane flaring. It was a fascinating conversation and I hope you enjoy it.I look forward to hearing your thoughts on this episode and on geothermal energy. Thank you for listening and for being a subscriber!Timestamps2:47 - About Cully and Crusoe8:47 - Crusoe digital flare mitigation (DFM) strategy and how they are reducing emissions14:54 - Crusoe’s digital renewables optimization (DRO), negative pricing, and how Crusoe is managing stranded wind power20:46  - Expected and current loads and load growth from data centers27:45 - Flexibility of data center demand and AI; training vs. inference functions in AI37:36 - Opportunities for greater efficiencies in AI chips40:34 - Trends in carbon accounting; location matching42:52 - Tally’s Law and the Energy Transition49:22 - Cully’s thoughts on needed policy/regulation changes for the energy transitionShow NotesCrusoe Energy Crusoe Careers PageTally’s Law and the Energy Transition by Cully Cavness The Extraction State by Charles BlanchardAI, Data Centers and Energy, Interview with Michael Terrell - Redefining Energy PodcastAI is poised to drive 160% increase in data center power demand - Report from Goldman SachsNuclear? Perhaps! - Interview with Jigar Shah on the Volts PodcastTexas Advanced Nuclear Reactor Working Group at the Texas Public Utility CommissionThe Energy Capital Podcast with Former PUC Commissioner Will McAdams"The Name of the Game is Flexibility," a Conversation with ERCOT's Pablo VegasTranscriptDoug Lewin Cully Cavness, welcome to the Energy Capital Podcast.Cully CavnessThank you so much for having me.Doug LewinReally looking forward to this conversation. Crusoe is really a fascinating company. You guys are doing some really innovative, interesting, and different things. So why don't we start with you, Cully? Tell us a little bit about your background and about Crusoe. Explain to the audience a little bit who you guys are as a company, if you would.Cully CavnessGreat. I'm excited to be here and share a little bit about what we're doing at Crusoe, where we came from, where we're going. In terms of my personal background, I grew up in Denver, Colorado. I went to Middlebury College in Vermont to study and I studied geology and economics thinking I was gonna go into oil and gas. But at Middlebury, anybody who's familiar with the school will know that the climate conversation was a huge theme and a huge focus in that student body. And it made a big impact on me.And so I actually, right after I graduated from college, I was awarded a Thomas Watson Fellowship, which is a program where you're sort of banished from your home country for a year and you get to go study whatever subject you really want to study for that year. And I wanted to think about this sort of morality of energy and the balance between energy and the economy and the environment. And so I was really fortunate to be able to go to Iceland where I worked with a lot of geothermal power and hydro producers. I went to China where I was much closer to coal. And then I went to Spain. I worked with wind and solar developers for the CFO of a large renewables group there. And then I went to Argentina and I worked with a hydroelectric engineer. And I got to really see a pretty broad survey of the global energy system, everything from finance to project development and management to engineering and operations. I saw power plants that had broken and were in stages of repair and learned a lot from that experience. And from that, I ended up going into the geothermal energy industry. I had a mentor who was the CEO of a company called Global Geothermal, and he took me under his wing. And for the first few years of my career, I was developing geothermal power plants, mostly internationally. And then sort of long story short, I ended up doing an MBA over at Oxford in England and came back to an oil and gas focused investment bank here in Denver. It was sort of the one energy focused investment banking role in Denver, primarily oil and gas clients. And that brought me back into the oil industry. I ended up being a Vice President of Finance for a private equity backed oil and gas company after that. And we were drilling some exploratory oil and gas wells in Eastern Colorado. That was sort of a step out from the core shale play, the Niobrara. We were miles away from the core of the activity. We drilled some wells that ended up being good oil wells, but there was no natural gas pipeline infrastructure in that area. And so the default then is, at least at the time was, all right, if you can't get the gas into a pipe, you put the oil into a truck and you send the truck to the refinery. That's how you sell the oil. And you can't do that with the gas, so you just light it on fire and you burn it. It's called a flare. And I thought that was pretty insane. And I was frankly, I was embarrassed about it. You know, just considering the path that I'd gone through and that I had really wrestled with that intersection of climate and environment on one side, but then the economic and human benefits of energy access on the other.Wasting the energy the uncombusted methane emissions. I had a big problem with that and I've been you know, I've been playing around with mining Bitcoin as a hobby in my basement and my wife was observing that you know, the there's like hot wind coming out of the basement and our power bill had dribbled and that's also a commercial problem related to energy and an environmental problem related to energy. And the insight was basically maybe one of these problems can solve the other. What if we could package a modular data center that could go to the oil field, actually sit on pad next to a flaring well site, capture that gas that was being flared, turn it into electricity, use the electricity to power the modular data center and basically new way to, we called it the digital pipeline. So a new way to take that gas molecule and convert it into value. The view was that it would be simpler and more cost effective to transport a bit than a molecule or an electron in these stranded gas locations. And that thesis really bore out.Doug LewinThis is really interesting. So, you know, you, what you're, what you're describing in Eastern Colorado was obviously a problem all over the place in the oil and gas industry. And obviously this podcast is very focused on Texas. It's a huge problem in the Permian and, a little bit less in the Eagleford though. It's still a problem down there too. There's just a lot more of a gas focus down there, but particularly in the Permian where it's mostly oil and what you're describing, as I understand it, is what is often called associated gas, right? You're really drilling for the oil. That's the valuable product.  Gas, depending on what's going on in the market, when today it's very low, right? 2022 is obviously, it was very high, but over the last six, seven, eight years, it's generally been very low. One of my favorite energy books is called Extraction State by Charles Blanchard, a History of Natural Gas in the United States. And he describes for those listeners that are Seinfeld fans, the famous episode of the muffin tops where they start this business. Like everybody only wants the top of the muffin. Nobody wants the bottom of the muffin. So they create this business and just sell muffin tops. And they end up in this huge problem because they can't dispose of all of the bottoms of the muffin that nobody wants. They can't get anybody to haul it away. And Blanchard in his book, Extraction State, likens the associated gas to the bottom of the muffins, the oil to the muffin top. So what you're describing is a very interesting solution to instead of, and you mentioned flaring as a matter of fact, like in Texas, an even bigger problem is the venting. Like people think when they see those flares, that's a problem. And it is, it's actually a bigger problem when you can't see it and it's being vented directly into the atmosphere. So what you're doing is trying to use that. Can you talk just a little bit and help me and the audience understand? Because I think for a lot of folks who approach this from a climate perspective, they're going to say, well, wait a minute, but you're still burning gas to power the data centers or the Bitcoin mines or whatever. That's still contributing to emissions. But what is the difference? And you don't have to get into great gory detail, but rough order of magnitude, the difference in the emissions, how much are you reducing emissions by actually using that gas for power rather than flaring or venting it?Cully CavnessYeah, some great questions. On the emissions reduction piece, so if we just compare the status quo, there's a big ball of fire, it's a flare situation, and we come in and we deploy what we call digital flare mitigation. That's what we call our technology. And we reduce that flaring volume. The important thing to know is that flares don't fully burn methane, and there have been a lot of studies on this over the years that have shown that probably 8 or 9% of the methane is escaping to the atmosphere, uncombusted from that flare. And it can vary. I mean, it can be as much as half in a really bad, poorly run flare, and it can be much more efficient, but that's sort of like a reasonable average to think about. And methane is 82 times as potent as CO2 over a 20 year timeframe. So when we talk about mid-century 2050 kind of climate goals, there's an 82 multiplier to take into account there. And our generators get 99.9% combustion of the methane. So dropping that basically to zero, it's absolutely true. We're still emitting CO2 from converting the gas through the combustion process. But by eliminating the methane, it's reducing about two thirds to 70% of the CO2 equivalent emissions compared to status quo flaring. The other way to think about that is that, you know, like of all the gas that does get burned into CO2 by the flare, the remaining residual methane is three quarters of the total CO2 equivalent impact of the flare. And so by dealing with that, you're dealing with like the majority of the CO2 equivalent problem, especially for those newer term climate goals, which I really do think about a lot because I think the crux of these climate goals isn't really actually about where do we get to from a parts per million perspective ultimately, like what's the right level in the atmosphere. Often there's a focus on it. We can't go beyond 450 ppm. There was an organization called 350.org that said we shouldn't go beyond 350. The bigger issue, I believe from a biodiversity standpoint, is actually the speed at which you get there. Because species just can't adapt fast enough to accommodate some of these faster pathways. And methane is kind of the key culprit in that acceleration and that speed side. So getting it from methane to CO2, it sort of buys time, I call it extending the climate runway. And it gives it literally a couple more decades for species to adapt. And you're moving from a semi-arid environment habitat to an arid habitat. Well, if that happens over the span of 10 years, you might go extinct. If it happens over the span of 100 years, there's a potential for a little bit more adaptation or migration for the species to be stressed but not extinct. And, you know, so speed really does matter in that context, which is why methane is kind of like the low-hanging fruit that makes a lot of sense to go for first. It's also something that we can do that's economic. So it turns out there's a virtuous cycle of this does make sense economically. We can provide basically a free solution for flaring that mitigates the methane. It provides us with access to a low-cost energy source that was previously being wasted and valued almost at zero. And there's an economic incentive to do more of that. So today we've deployed more than 200 megawatts of that across seven states, including your state in Texas. I was just in Midland and Hobbs, New Mexico earlier this week. And we've got kind of our fastest growing area of operations is down there in the Permian where there is a big flaring challenge, but hopefully we've got a solution that's going to make a dent in that really quickly. You know, I'd also just point out that we're also moving into a lot of renewables based projects as well. So we can talk about that later. And there are other forms of stranded energy that are on the other end of the energy transition where there's an inefficiency that needs to be dealt with and computing in a very interesting way can be the solution there as well.Doug LewinSo let's talk about that. But before we do, just really quickly on the, I think this is really important on methane, because I agree with you, the speed at which the climate impacts hit us, our species and other species, and affect biodiversity and all that matters. I think the other piece of that, Cully, too, is that, you know, if you're, if you can reduce the most potent greenhouse gasses, which, which methane is, is certainly in that group, CFCs and things like that go in there too. You're, you're also then, buying time for technological improvements to come along. Right. I mean, and I’m not am you know, I'm anticipating the criticism I'll get and it's, and it's legitimate criticism. I'm not representing some kind of techno optimist like view of this that like oh technology is going to save us. We've got a lot of things we got to do right. But one wedge of the pie is technology. And we have seen tremendous technological advancements in the energy and climate space. Just where we are right now with solar and storage, it would have been very hard to even 10 years ago, imagine the kinds of price declines that we've seen. Some people imagined it, but not many. So I think that that's part of it too, is you've really got to address that most potent piece. And so anything we can do on methane just kind of has an outsized impact. So let's do talk about renewables because there's sort of an interesting kind of corollary here, almost like the associated gas part of renewables or the bottom of the muffin, if you will, in that you have times and they're becoming quite common now, particularly in Texas, I'm not as familiar as with other markets, how much we're seeing this, but in Texas, the amount of curtailment and congestion that we see on the system is very, very large and rising very fast as we're not able to keep up with the transmission needs as generation is coming into the market. And so you end up with a lot of times where the renewable energy that is produced is not able to reach any place where it's actually used. So they're literally just curtailing, just shutting down wind or solar power. And it's not just wind or solar, there's actually other kinds of power that can get congested, caught behind a congested node and just not able to be used. So if I understand what you guys are doing, similarly what you're doing with the flares, where you're putting a data center next to it and capturing that flare and making a power plant, you can kind of do the same thing near congested nodes where you have a lot of renewable production and actually reduce energy waste and actually use that at the data center. Is that correct? And either correct me if I got something wrong and then maybe give us some more context and details around how that works.Cully CavnessYeah, that's right. So I mentioned we called that first business model DFM, digital flare mitigation. And the second model we call DRO, digital renewables optimization. Meaning you can bring a data center into one of these congested nodes, especially if it's really saturated with wind power where you have this intermittency effect. You can have 20% to 30% of the hours of a year in some nodes, and Texas is a good example, that are negatively priced. So that has a number of interesting knock-on results. One is that older wind farms that don't have the production tax credit… Let's back up and say how is something negatively priced? How is there actually a negative price on the grid? What's happening there is there’s so much wind power that the grid can't accept it all. Either there's a transmission constraint or there just simply isn't demand at the end of the lines and so people are effectively having to pay transmission and distribution fees and receive zero revenue because there's no bid for that power. And so you're getting like an all in price that's actually a negative price. And you can do that if you get a rebate from the federal government in the form of the production tax credit. So you could sell for negative one penny if you get two and a half pennies from the federal government, your net price is actually positive one and a half. Those production tax credits expire after about 10 years. If you're an older wind farm, you don't get that anymore. So when the price on the node goes negative, you actually just have to stop producing wind power, which is like the worst outcome from an energy transition perspective. You literally could be producing more renewable power, but for an economic constraint reason, it's just actually being shut off. Even for newer wind farms that are still receiving the production tax credit, this is obviously like a very frustrating problem. And it's like a breakdown in the supply and demand connection in the market. So our view again is that bringing energy to the data center isn't always the right answer. Sometimes you should bring the data center to the energy. And that's really our origin story is we've been bringing data centers out to stranded energy locations. This is just another form, frankly, of stranded energy. They're also hard places to operate. They're not your traditional data center markets like in Virginia and around Dallas, and the Pacific Northwest. This is, you know, to again use the Texas example, this is rough West Texas kind of desert territory without a lot of the existing infrastructure. However, if you do it the hard way, there's access to a lot of low-priced clean energy. I believe more in the location-based approach to emissions accounting rather than the market-based approach. The market-based approach would say, I built my data center anywhere, I bought renewable energy credits for all the megawatt hours, and therefore I have zero emissions. The location-based is more like, depending on where you located that load, what actually happened in the physical real world. And if you located in some areas, the answer is you spun up a coal power plant. And if you located in other areas, it's like you absorbed a lot of otherwise curtailed or negatively priced wind power and maybe drew some power from the grid as well but the average emissions of that location-based view leads you to some very different outcomes compared to the market-based approach to carbon accounting.And so we've really embraced that location-based energy first approach to how and where we locate our data centers. We've got projects going in Iceland, for example, the ultimate stranded renewable energy resource, a place that I spent several months of my life. And they've got gigawatts of potential of geothermal and hydropower and 300,000 people. And what are they going to do with it? When I used to live there, they were talking about running a power line to Scotland. So we're bringing AI training workloads to Iceland, and we're deploying what will be the largest computing cluster in the country of Iceland. And it will be serving AI customers with 100% geothermal and hydropower as another example of how this can work from a location-based standpoint.Doug LewinI do have questions about the location based and sort of some of this move towards 24/7, but before we go there, I just, I want to cover a couple of things that I think are really important to AI. Can you help again, me and the audience kind of understand the magnitude of the AI loads? This is like, obviously it's almost becoming cliche cause every conference you're at, it's all AI all the time. But I think to a certain extent, and this is one of the reasons I was so excited to have you on is, you know, you hear all these things floating around, but there's so many different sources of information. You guys are working on this very directly. Can you talk a little bit about just sort of like what size of data centers you're seeing? And we'll put a link in the show notes. There's a pretty good report a couple of weeks ago by Goldman Sachs, and they talked about 160% increase in data center power demand. I think it's very important when you're thinking about that too, to your point about location, that's not going to be 160% increase everywhere, right? It's going to be concentrated in areas where, you know, like Virginia, that's this well-known kind of data center hub, Iceland, you were just talking about, but Texas, I think will be one of those places too, because of the energy abundance and the general ability to get low cost power, which isn't super easy, but it looks like you guys have a big part of the solution to sort of make that happen. But anyway, talk a little bit about the size of these data center loads and what that means for an energy system. You could talk about it in different places, but in Texas, our peak so far is somewhere around 85,000 megawatts, a little higher. Anyway, yeah, talk a little bit about the size of the load, if you would.Cully CavnessYeah, maybe again zooming back like where's the load coming from? It's interesting to just think about physically what's happening in these AI data centers compared to a traditional data center. So traditionally a data center, you go inside there are racks, racks and racks of servers. And those racks traditionally were like seven kilowatt racks. That was a very standard, maybe 14 kilowatt racks. Those were kind of standard power densities. And the current kind of leading edge GPUs coming out of Nvidia, for example, are really optimal around a 50 to even 100 kilowatt rack. And there are prototypes that are not far off that they're being demonstrated at conferences that are hundreds approaching a megawatt in like a rack. And it's this incredible density of electrical and thermal cooling capabilities. So literally liquid cooling to every single chip on the server. There's an in and an out of cold water and hot water going to every single chip on the server. It's allowing them to just compress more and more computing power into smaller footprint. There's that going on. And then there's this insight, which is that the more GPUs can be networked together on the same cluster, the more performant that cluster will be at training, for example, a large language model. And there's a physical constraint element to that. So you actually have fiber distances that have to be considered, how far away can the farthest GPU on the cluster be away from the center of the cluster? And that leads you to wanting to have these really dense configurations of lots of computers close together so they can all be networked together on the high-performance networking. And when you play that out, what that's meaning is that what used to be a big data center was like a five megawatt data center, and then a 10 and a 20 megawatt data center. These were kind of beyond belief huge, even just a decade or two ago. Now it's sort of looking like we're going towards 100 megawatt data centers that have 100,000 GPUs all networked together in a single cluster. And perhaps that's even going to be small in the not too distant future. We might go to hundreds of megawatts and campuses that are going to be gigawatts. And there's this arms race of who can train the best models and who can operate the best inference off of that model. And it seems right now that size is going to matter and it's leading to a real land grab around access to megawatts, access to digital infrastructure. And our view is that it's really important how and where we kind of locate those megawatts on the grid and potentially even off grid. And so that's really where we take that energy first approach to development and construction and operation of data centers. We've got a piece of our business which is building the data centers, the 100 megawatt scale. We have our own design, which we think is really an optimized design around heating, cooling, those physical constraints and distances within the data center. And those can be essentially offered and leased to larger technology companies for them to host their own GPUs in there and do their own workloads. We also have our own GPU cloud product called Crusoe Cloud where we put our own computers in there. We rent them out by the hour, by the three-year contract. We have different models, but we actually are the cloud provider within the infrastructure as well in some cases. And we have two different ways to then basically take those megawatts to market. But it all starts with energy first locations that we can develop into that data center infrastructure.Doug Lewin Thank you for that. And so just to put that in perspective for some folks that may not be as in the weeds on power as others, when you talk about the old racks used to be 7 or 14 KW. A home on an average day might be using two or three KW, an average sized home, couple thousand square feet or something like that. On a really hot day, it might be five, six, seven. So, one rack would have been equal to like one home on a very hot day. I'm speaking in generalities, obviously it matters how much insulation and what kind of HVAC and what you're doing inside your house. If you're running multiple hairdryers or something at once, it might be different. But these Nvidia racks you're talking about, now you're getting up to 500 to, excuse me, 50 to a 100 KW. So right now you're talking about something like 10 to 20 homes. And when you talk about a data center that's a hundred megawatts and even clustered, and actually one of the previous podcasts was with ERCOT CEO Pablo Vegas. And he talked about how we're seeing 500 megawatt, I believe the numbers he said were 500 megawatts, 700 megawatts popping up on the grid, like all at once. So that's probably what you're describing as one of these campuses or clusters. Now you're talking about like small cities. It's like a small city popping up on the grid potentially in the space of 6 to 12 months, give or take. About right?Cully CavnessI think your order of magnitude is right. I used to think about a house as one KW. Doug Lewin Yeah, not in Texas. Everything's bigger down here. Cully CavnessBut yes, I mean, roughly that's right. I mean, these are huge sources of power demand when they come online at this kind of scale.Doug LewinAnd so can you talk a little bit about the flexibility of that demand? So this is becoming a bigger and bigger issue. In Texas, ERCOT has established a large flexible load task force, which spends a lot of time, has spent a lot of time thinking about Bitcoin, but is going to more and more need to be thinking about these AI data centers. And I've heard a lot of different things. There's no flexibility at all. There's a lot of flexibility. I've heard everything in between. Can you just kind of share your perspective on how much these can be flexible and if flexibility can be built into the design. You mentioned you guys are designing some for heating and cooling. Are you thinking about building flexibility in? And let me just, before I turn it over to you to answer that, I would imagine with what you guys are doing, that flexibility would be important because when you're siding next to a congested node, like you wanna be running as much as you can during those periods. But then when it's not congested and power prices are getting really high on the grid and there's actually not enough power, maybe you could move that wind or solar rather than using it, the data center, you could move it onto the lines and let it flow to Dallas or Houston or whatever. And the prices are higher than you would want to use less there. So maybe you do that through changing around how much cooling is going on or you site batteries there or if it's a large language model, maybe you're batching functions and not running them 8,760 hours, which is the number of hours there are at a year, but maybe 8,500 and even not needing power for even a couple hundred hours would make a huge difference. So can you talk a little bit about what flexibility there is or is not with AI data centers?Cully Cavness Yeah, I think it comes back to the customers that are using the servers. And right now, I think a lot of the market is still stuck in an old mental model of data centers need to be tier four, meaning 99.999% uptime reliability. And it seems to me that many of the incumbents still have that perspective as it relates to this new wave of AI data centers, it's not clear to me that all the use cases of AI computing require that kind of uptime. And I think that you actually had an insight, which I've been thinking about a lot, which is this batching idea. So you can checkpoint these models, meaning if the power goes out in a worst case scenario, you don't have to restart training the whole model. It kind of falls back to your last point that you saved the model. It's like if you're playing Mario and Bowser kills you, you get to start back at the beginning of the level. You don't have to go all the way back to the beginning of the game. And that's important, right? Because that means that, okay, maybe you would tolerate a certain amount of outage and it becomes an economic decision. If I can offer you a much lower price point per hour of training on average throughout the course of the year, provided that I can interrupt your workload 1% of the time or 10% of the time or something. Is that a trade that you're willing to make? I believe that for some percentage of the customers that will be a trade that they're ultimately able to make, the market hasn't really moved there yet. But I think some of the kind of frontier folks are starting to realize that. And we would really advocate for that because that would allow for more of the demand response feature that, for example, in Texas is such a big deal in you're deregulated market, you've got this really kind of beautiful, again, market-based incentive driven system to provide flexibility back to the grid. And when it gets really hot and everybody's AC is on, you can get paid to turn off if you're able to. And we need that economic signal to flow through to the customers that are doing the training workloads. I would say there's going to be a difference between training and inference. So broadly in this AI computing world, there's these two categories. Training is the very kind of it's a longer term, it can be days, weeks, months of running a model over large amounts of data to find the insights and create the weights and balances, weights and biases in the model that sort of builds the model, let's say. And then inference is once that's been built, it's using the model to do tasks. So it might be for a self-driving car to detect, is that a stop sign or a yield sign? That's like using the model to make one inference or for ChatGPT, it would be answering the question that you just posed in the chat box. That's an inference. And, you know, clearly that is a more, you need to be up and available to provide that service. Can these be federated across multiple data centers? So one can be offline at any given time if as long as another one is online. I think these are really interesting questions and it's just a totally new approach to the digital infrastructure and the kind of interaction of digital infrastructure and the energy infrastructure. It just, we need to get a lot more sophisticated on that. That's what I'm really excited about at Crusoe is we have a team that's a hybrid of energy professionals from grids and utilities, from upstream oil and gas producers, from renewables developers. And we have a team of really seasoned executives who have built and operated data centers and cloud products. And you know, we are sort of merging those two disciplines in a pretty special way that we're going to try to find and use these kind of insights of how digital infrastructure can and should interact with energy infrastructure.Doug LewinYeah, I think that's really, really interesting and a really sort of potent mix, one that is really needed at this point. We'll put in the show notes a link to, there was a podcast, I believe it was Redefining Energy did one with Michael Terrell from Google. And he was describing how they're not there yet, I think, if I remember right, he was talking about sort of where this might be headed in that with what you're talking about with inference, where the task needs to be done, you need to know, is that a stop sign or a yield sign? You can't wait till later, but they could move those functions. And you sort of suggested this just a minute ago, but just to dive a little deeper into it, you could move those functions to different data centers based on how much energy is available. And I was sort of getting this image in my head of like, there's the saying, the wind's always blowing somewhere, the sun's always shining somewhere. And so if you picture some of these data functions actually moving around the world with the sun, so that solar power is providing a lot of these tasks, but at different data centers. And you don't really care as a customer, as an end user, I don't care what's happening in Iceland or in Texas, right? It doesn't matter as long as when I need the information, it's there, right? And so you could kind of see some really interesting, and it's going to take this close integration of, as you're describing, this sort of energy expertise and the data center expertise for where is that possible, where is it not.Cully CavnessYeah, I think it's, you're probably more right than wrong. And if you think about what that means from an infrastructure build perspective, it's like a huge undertaking. It's like a global rebuilding of the digital infrastructure, the fiber infrastructure, and all the energy that has to go into it. I mean, we're talking about many trillions of dollars of capital investment. I think on that specific idea, there's this latency constraint, which is some things need to be really fast. Like that if it's a stop sign, you really need to know it's a stop sign and you need to know it immediately. There's other forms of inference that you probably could wait a couple hundred milliseconds to get the ping back from Asia. If there was, if it was sunny in Asia, it's nighttime here, you know, you could you could actually imagine that. Right now, most of the latency, if you type in a prompt into ChatGPT, is still the model itself doing the computing. But as they're getting more efficient, that portion is getting smaller and it's going to be the percentage that's borne by the networking latency is going to be a larger percentage as that happens. And so there will eventually be this interesting question of how close does the workload need to be to the problem? And for some, it'll have to be very close. And for others, it could maybe be anywhere on Earth. Doug LewinYou were also talking earlier about how much of these loads are actually cooling loads, right? And so there's probably also some demand flexibility there, I would think. I don't know. And so let me phrase it in the form of a question. Could you do, like for instance, as on the residential side, if you have a home that is well insulated, you can pre-cool your home. So Texas is a great example. It's going to be a hundred degrees basically every day this summer. We've already hit a hundred a couple times and we're not even as we're recording out of May yet. But you know, you have all these days that are a hundred degrees, tons of solar power, pretty low prices at two o'clock and three o'clock in the afternoon. It's really not an issue. The issue on the grid, the tightness is going to be seven, eight, nine o'clock as the sun is going down. So you pre-cool the home you use less in those evening hours. Is there the ability to do some of that at data centers too, where you're actually making it a little colder, or does it have to be like, it's got to be at this temperature and it can't vary?Cully CavnessIt's more of a run rate kind of problem. I mean, you've got just 100 megawatts of heat being produced 24-7, and that has to be evacuated on a very continuous basis. There's a little bit of thermal inertia if you were to pre-cool, but…Doug LewinNot as much as with a home.Cully Cavness Never say never. I mean, maybe you get a big reservoir of cool water and you kind of like pre-chill some big thermal mass or something. But in your normal data center, it's less of an opportunity. There's definitely a lot of efficiency opportunities. Having the most efficient chillers and being thoughtful on the designs, the engineering, there's a lot to do there. Doug LewinAnd there's efficiency in the chips as well, right? Can you talk about that? You were talking about the Nvidia chips and they're obviously, they use a lot of power right now, but I think Nvidia, at one of their last events did talk about how their chips are more efficient. Are you seeing that on the data center side?Cully CavnessThe chips are more efficient, but there's more chips and bigger chips. And so the power consumption of the server isn't necessarily declining. In fact, it might even be rising. I mean, again, if you back up to just kind of first principles of what's going on here, you've got the sophistication of making these chips has gotten to such an insane point that there has to be some kind of like tailing off of the efficiency gains. So just as an example of how these things are made. They're mostly made in Taiwan and there's a factory there owned by TSMC. A lot of people might have heard of this company. It's the largest semiconductor fab in the world. And they have this machine which is made by a company called ASML. And this machine costs something like $800 million per machine. And it's transported in like four jumbo aircraft to Taiwan.And they have 80 plus of these machines in TSMC, as far as I understand. Each one of these things is like a major installation engineering project to put this thing together. And then once it's there, the way it works is they're liquefying a droplet of metal that's falling into like a convergence of a bunch of lasers. These lasers are blowing up this droplet of metal into a very specific wavelength of ultraviolet light that bounces off a bunch of mirrors. And then it goes on to a piece of silicon that's like 10 atoms thick or less. Doug LewinWow.Cully CavnessAnd it's etching a pattern of circuitry into this and then another deposition of a substrate and another blast of this extreme ultraviolet light. And it's doing this like tens of thousands of times per second across this whole factory. When you're measuring things and just like atoms thick, you're running up against real boundaries of physics can do and I don't know how much you can just depend on the chips getting more efficient.Are there different versions of chips and different designs of chips coming that are definitely different approaches? Yes, absolutely. There's some new startups that have released some new chips that look like they could be way more efficient on the inference side, for example. The training piece, it's a little bit less clear that you've got huge gains that can still be made there. But again, never say never when it comes to technology. I'm just pointing out that they've been at this for like decades now and it's gotten really optimized. And there are certain pieces of it that might be kind of asymptoting.Doug LewinYeah. And it's a little bit of a law of diminishing returns, right? At a certain point, like you just, you reach that kind of limit where, I mean, we're seeing this like on, we're not there yet, there's still more cost declines in solar and storage, but when you see those curves and you can only get so close to zero before there, there wouldn't be any money left to be made in it. So at some point you hit that terminal point.Cully CavnessMaybe. I mean, I just worry when people say it's going to be solved by more efficient chips. There might be limits there.Doug Lewin No, that makes a lot of sense. That makes a lot of sense. All right, so let's come back to something you mentioned earlier. You were talking about carbon accounting and the location-based. There's, again, a lot of emphasis among some of the major tech companies, but I'm seeing this get talked about a lot more, what is often called kind of the 24-7. So you were describing this earlier. A lot of companies want to have 100% clean power, 100% renewable power. Sometimes they describe it different ways. But what they're trying to do is do like hourly matching. You were talking about location matching. Are you seeing more and more companies actually asking about that sort of thing, actually getting much more granular about carbon accounting? Obviously the majors, the Googles and the Microsofts and that are doing that sort of thing. Is it getting beyond those companies or is it still pretty concentrated in a few companies that are doing that kind of work?Cully CavnessI think it's still kind of early days on a bunch of this and they keep modifying the rules and editing them over time as I think people just realize there are better approaches to carbon accounting and the standards that go into it. You know, location matching is important, but it's also like really important like what's happening at that location. I guess that was the point I was trying to make is that when you add a load to a specific node, it causes a specific set of energy generation resources to get called up. And in some of those situations produce a lot more emissions than others would. So just kind of first being thoughtful about the location and the realities on the grid of what happens when you add load there. Then to me, the second thing is, okay, you're buying RECs and you're providing that economic incentive for the renewables to get developed, which is valuable. It's a great thing that we have that mechanism in the market. But to me, the bigger question is kind of that physical reality of where was the load located?Doug LewinFor sure. And you're starting to see, I think they're called TEAKS. There's like time energy attribute or something like that. There's a time-based attribute kind of accounting that is already going on. It'd be interesting to see if that could be extended to location as well. I think it's a really interesting idea. I actually hadn't, I'm embarrassed to say I hadn't really thought about it before, but it's a little bit embedded in the time thing because where you are would matter, but it's not explicit and it probably could be teased out more. It's really interesting.So I want to ask you just a couple more things before we wind down here. A couple of years ago you wrote an article on Medium. We'll put a link to it in the show notes. Tally's Law and the Energy Transition. Can you talk a little bit about, give people a little preview if they go and read that. What are they going to see? What's sort of your thesis in that piece?Cully CavnessYeah, I mean, this goes back to some of that early, my early experiences in the energy industry and wrestling with just the identity of wanting to work in the energy industry. I mean, I think it's the most fascinating industry. It impacts people. It impacts human lives in a lot of really positive ways. Just having access to any form of energy is huge and transformational for so many people. And back in the early 2000s when I went from Iceland to China, China's pretty wealthy, but in the mid 2000s, it was still pretty rough for the majority of the people there. And most of them could not care less if the power was coal or solar or wind or the climate impacts of any of it. And they just wanted to have the economic means to have another meal that day, you know, like to add protein to their diet, to upgrade from a bicycle to a motor scooter or from a motor scooter to a small car or something like that.And, you know, starting in Iceland, it's the exact opposite. It's like this abundance of riches of unlimited renewable power. And it's very easy to say, yes, we'll just have 100% renewable power. We only have 300,000 people and we have more hydro and geothermal than we know what to do with. So just two very different realities for those two countries at that time. And it led me to that framework of the triangle of ease, energy, the economy, the environment. And I sort of formed a formula of how I thought about the world along these lines, which was RT equals PQV. And people can look it up if they want to, but basically natural resources times technology on one side of the equation and population, quality of life, and environmental health on the other side of the equation. It's a mental model for like, if you change one of these five variables, what happens? So if you increase population or increase the quality of life of a population and you don't have any change on the energy technology side or the natural resource side, then the variable that has to fall is V, environmental health. And conversely, like if you can innovate with technology and increase the left side of the equation, you could increase population, you could increase quality of life without impacting environmental health. And, you know, there's more to it than that, but I think that's like kind of a simple way to summarize the viewpoint. And I guess I do take a little bit of the techno-optimist’s viewpoint. And well, in the sense that I do think technology will largely be the solution or it won't be, but we're definitely not shutting off the energy. And buying time, more time for technology to end up being the solution is really the most important thing we can do right now. So the low-hanging fruit being things like waste, things like methane emissions. Let's solve those now so we have more time for the technology to grow and increase the T variable so that we can continue to have better quality of life and more people without having a big climate impact. And I'm hopeful on that.  I mean, I think there are a lot of things that are on the horizon that look great. It appears carbon capture and sequestration is getting there. A lot of interesting things happening on batteries. Radia just came out with this new wind turbine that looks incredible. I don't know if you've looked into this company at all. You should have Mark on your podcast if you want to talk to a really interesting entrepreneur in the wind space right now.You can go on and on, all the nuclear stuff that's happening. There's a bunch to be excited about, but it's clearly going to take a while for this to mature and to really scale up. So again, back to extending the climate runway, we got to stop emitting all this methane just needlessly. We got to stop wasting energy. Let's stop curtailing renewables. Let's tap into the latent stranded renewables where we can, if we can locate a load in Iceland versus in the Eastern Seaboard where it's going to be much more fossil oriented. Maybe that's a good idea. So again, we're going to energy first, trying to be thoughtful about these things and extend the climate runway while still accommodating this wave of AI demand, which is clearly happening. So it's just sort of like, how can you influence it to be least impactful?Doug Lewin Yeah, I mean, it's happening and I think we're also gonna see increasing demand for all sorts of different things, including quality of life, right? There's still, like you're talking about China 15, 20 years ago, there's still a billion people in the world that don't have access to electricity and they need to get it. That's just wrong. We've got to figure out ways. So that formula you just described I think is really, really interesting. I'll tell you, I mean, I think we're all creatures of our experiences. And, you know, we have, obviously as any two people have had different experiences. I do think technology will play a big role in it, but I think there's this kind of interplay happening all the time between technology and markets and policy, right? And so if you're in a place where like you could have the greatest technology, but if the policy makers set up a system where they're favoring other technologies or just not allowing market entry, right? And there's this interplay between those things because you can only hold back, you know, the water, if you will, hold back the deluge so long and a technology will kind of like water flowing, right, will find its way to get in even where policy is trying to keep things out. But policy can really slow down or to put it more positively, policy can really enable markets to bring technologies in really quickly. Anyway, I just, while technology will have a big role, I do think there's always going to be this interplay and depending, I don't know if you have any thoughts on that, any place in particular that you're operating where you see like policy stands in the way or makes things easier. I kind of like this Texas market because what you were describing earlier with like, it's going to depend on the customer when I asked you about flexibility, right? We have a market that kind of accommodates that. Like if you're willing to pay thousands of dollars a megawatt hour, you can have the power whenever you want it, but everybody's going to have their price as to when they're willing to curtail. So there's an example of a market sort of enabling technologies and flexibility. That's mostly on the large customer side. We need more of it on the small customer side, but that's a topic for a different podcast.Cully CavnessThis isn't as close to our business, but I would really love to see a streamlining of nuclear permitting. I just think there's so much potential there and the fear of it has led to a permitting regime that makes it almost impossible to get anything done. And it's sort of disproportionate to the risks at this point. I think if you really talk to experts, they would say that, which I'm not a nuclear expert, but I've been to how many hundreds of energy conferences, I've heard enough to know that it's really, really safe. When it goes wrong, it goes really wrong, but other things go a little bit wrong all the time. And the actual impact on human health and the environment is probably worse from a lot of other energy sources than nuclear running very safely for 99.999% of the time and having a big accident even once in a while, which it does appear they're even getting better at really minimizing that outside risk with the new generations of nuclear reactors. And updating our framework to allow more nuclear to get permitted and built would be huge. Obviously, the grid transmission permitting is a big deal. At some point we might have to get a little bit more aggressive with like eminent domain or something. I don't know, you just can't let every potential objection, get in the way of building out the really critical infrastructure that we need for the whole country. And there's a balance. Personal property rights are essential. They're like the cornerstone of the country. But there has to be a practical solution to get things done. And these things really need to get done. And so that's a hard problem. But we've got to figure that one out.And I think, yeah, the Texas grid is a good example. I mean, you guys get a bad rap with some of the outages from some of the winter storms recently, sometimes. But I would say by and large, it is viewed as the epicenter of entrepreneurship and dynamic business models in the power industry because of the way it's been deregulated. And I think ERCOT’s being pretty thoughtful. It's like they're still regulating and controlling the really bad edge cases, right? There are like caps to power pricing. There are kind of like curtailment mechanisms to avoid the blackouts that things are changing so fast. They're clearly not perfect yet, but they seem like they understand the problem and they're moving in the right direction really fast. But in the middle range, like the normal course of business stuff, just let the free market operate and it'll tell you what kind of power resources and what kind of, you know, do you need battery storage? Do you need peaker plants? do you want more renewables? Let the market figure all that out and not picking too many winners there. That seems great and also kind of reducing the gatekeepers of the traditional grid players, the traditional utility players, letting all that kind of be a little bit more flexible so more participants can come in and find their niche, it just seems to be like it's on a really healthy track in Texas in general, despite some of the headlines, which, you know, look, I wasn't there when the power went out, and I know a lot of people were, got hammered by that in really bad ways. And maybe, though, in the ultimate grand view, it's like a price that was paid for making the grid a lot better in the long run and making it this kind of free market approach where a lot of really cool innovations can come in and make the grid much better 20 and 50 and 100 years from now.Doug LewinAnd I think for anybody listening that wants to dive more into Winter Storm Uri, we'll put a link in the show notes to the very first podcast of the Energy Capital Podcast we recorded, which was with former Commissioner Will McAdams. He was just a week or two out of the Public Utility Commission. We recorded it. And we talked there about what happened during Uri and how it really wasn't a quote unquote market failure. It wasn't like the... That was really if we weren't yet regulating, and I would argue we're still not yet regulating natural gas supply to be winterized. Power plants weren't well enough winterized. And we hadn't weatherized homes and buildings, so demand was kind of off the charts. And that's something TUC continues to look at and may do more on. On the nuclear piece, and we're going to have, there's going to be lengthy show notes for this one. This is good. We've covered a lot of ground. We will put a link. There is an advanced nuclear working group at the Public Utility Commission. Governor Abbott put Commissioner Glotfelty in charge of that. I'm really intrigued by nuclear. I think, you know, just today, the day we're recording, Secretary Granholm, obviously in the Biden Administration, was at a ribbon cutting in Georgia for the new nuclear plants. And she said, we need 200 more. We got two here. We need 198 more. It's one of the few areas where I think Republicans and Democrats seem to really want to kind of both pursue a similar solution. And to your point about the dangers of nuclear versus other energy sources, we'll put a link to a great podcast, Dave Roberts, the Volts Podcast did with Jigar Shah and comparing, for instance, the radiation of coal ash sitting around on coal sites compared to what's at a nuclear plant, far worse. What is just sitting around like basically uncovered in these coal retention ponds. Cully CavnessThat's what I was sort of referencing. I mean, there's this, people are afraid of crashing in an airplane much more than crashing in a car, right? Because the airplane crash, the mental model is so gruesome and horrible and the car crash seems like something you do every day. So it's not something to be as afraid of. But the reality is the car is gonna kill you a hundred times more often than the airplane is gonna kill you.Doug Lewin Yeah, it's like 40,000 deaths a year in automobiles or something like that and a very low number…Cully CavnessI think it's a little bit like that with nuclear. I mean, it's scary because you think of Chernobyl, but if you actually do the kind of risk math, it's not the scary one to be worried about.Doug Lewin Yeah. Cully, I really appreciate you doing this. The model Crusoe has with this, you know, really citing, big energy loads next to areas where there is abundant and so abundant that it's often wasted energy is really fascinating, really thrilled to hear you're in Texas and doing things here. We'll definitely want to follow your company very closely and encourage our audience to do the same. Is there anything else you'd like to say? Anything I should have asked you that I didn't that you'd like to talk about before we end?Cully CavnessYou know we're hiring and if anybody wants to join the team we'd love to have you. We've got a really special team of talented, entrepreneurial, hard-working, creative folks here and we need more of them. So check out our website, Crusoe.Ai  and look at the careers page. I'm recruiting for a Chief of Staff which I'm really excited about. So if anybody wants to check that roll out in particular I'm just starting to look through resumes right now. Yeah, I think other than that, we're just happy to introduce ourselves to your audience and we'd love to stay in touch and do it again sometime.Doug LewinThanks so much, Cully. And we'll put a link to the, if you send us one, we'll put a link to where that job posting is. And yeah, I can't think of, you know, I mean, it's gotta be on the short list of most interesting places if somebody really wants kind of a front row seat into the energy transition and AI data centers, rising load growth, all these things that are going on. What an interesting position you have open. So we'll be sure to, to put a link to that as well. I learned a lot from this discussion. I really appreciate you taking the time. Thanks so much, CullyCully CavnessMe too. Thanks for having me. This is a public episode. If you’d like to discuss this with other subscribers or get access to bonus episodes, visit www.douglewin.com/subscribe

Cindy Taff discusses the potential of geothermal energy in Texas and Sage Geosystems' innovative approach to tapping into this energy source. Topics include enhanced geothermal systems, geothermal storage innovations, seismic concerns, technology integration, power generation cycles, investment potential for oil companies, cost competitiveness, and job growth in the geothermal energy sector.

One of the hardest things about energy efficiency is measuring it. It's difficult to do, but with increasingly sophisticated software platforms, it's getting easier and more precise. As homes and buildings begin to use more electricity for things like heat pumps and electric vehicles, and as solar and wind power continue to grow, there will be many times a day and most times of the year when it will actually be energy efficient to use more electricity. But at certain times and places, we'll need to use less or even send power from solar and storage, and even the storage in electric vehicles, at homes and businesses back to the grid. My guest this week was Carmen Best, Chief Policy Officer at Recurve, a company working to measure the time and locational benefits of flexible demand in the most rigorous way possible by actually measuring the end use changes and power consumed at the meter. Grid operators and policymakers are increasingly focused on dispatchable generation in Texas, sources they can turn on with a push of a button. But what they should be focused on is not only dispatchable generation, but dispatchable resources of all kinds, including demand side resources.To have confidence in them though, grid operators and utilities have to be confident they're deployed at the right time and in the right place. In this podcast, you'll learn the policies needed to truly scale demand side resources, including energy efficiency, demand response, and solar and storage, and how a leading company in this space is approaching the challenges. Carmen has spent over 20 years working on energy efficiency with much of that time focused on how to evaluate, measure, and verify energy savings.She's one of the leading experts on demand flexibility, energy efficiency, and the policies needed to make sure all demand side resources can contribute to reliability and affordability. I learn something every time I talk to Carmen, and this podcast was no exception. This episode is particularly well timed as today was the deadline for comments in response to the PUC’s energy efficiency questions. I answered some of the questions in this article. You can see the responses of over 20 commenters, including Recurve, here. I look forward to hearing your thoughts about this episode and about measuring and valuing energy efficiency. As always, thank you for listening and for being a subscriber! Transcript, show notes, and timestamps are below.Timestamps4:00 - About Recurve5:00 - The challenge of measuring demand reductions and Recurve’s approach to solving that problem9:32 - How Recurve’s data can be used by system operators to manage distributed resources, improve system reliability, and empower consumers13:41 - The link between energy efficiency and demand response18:57 - How can we be sure energy efficiency delivers value?23:33 - System benefit targets vs. savings targets; how this impacts consumers, costs, and competition.27:42 - Shifting energy efficiency from a utility program to a competitive market35:06 - Using energy efficiency measurement pilots to better determine costs and benefits45:17 - Texas’ ADER (Aggregated Distributed Energy Resource) Task Force; the role and potential of demand-side resources in ancillary services markets52:35 - Utility shutoffs. Challenges and opportunities in programs targeted low-income customers 58:34 - Policies to prioritize to increase reliability and affordabilityShow NotesRecurve FLEXMarketMission:dataSmart Thermostat Pilot filing with the Texas PUCEnergy Efficiency Questions filing with the Texas PUCProject 56517: The energy efficiency docket (here you can see all the responses to the PUC’s questions)The PUC’s Efficiency Questions: A Potential Turning Point for Grid ReliabilityCommunity Voices in Energy Survey: Texas Statewide Report from the Texas Energy and Poverty Research Institute (TEPRI)TranscriptDoug LewinCarmen Best, welcome to the Energy Capital Podcast.Carmen BestThanks for having me, Doug.Doug LewinThanks so much for doing this. I am a big fan of yours and the work you do and of Recurve, have been following Recurve for a long time, but I suspect some of our audience will not yet be familiar with Recurve and what it does. Can you just start us off by talking a little bit about the company and what it does?Carmen BestYeah, sure. So, Recurve is a company that is focused on analytics and we're a software company. And we're contributing to the clean energy future by looking at energy consumption patterns in effect. So, changes in energy consumption are really important to distributed energy resources and demand flexibility. And we kind of got our start in the energy efficiency space. And one of the problems we were trying to solve was having standardized measurement and verification. And that our open source products have kind of evolved to be a broader swath of solutions that include being able to forecast where demand resources are going to be best suited for the grid, be that distribution or decarbonization initiatives, look at how they performed. So, utilizing that measurement verification interface of how well they did.And then the last part is really about settlement. If you have the information about how they performed, how does that track to the value that you want out of these resources and how do you pay folks for what they've delivered as opposed to just having fixed incentives for single technologies or something? We're really trying to build out a way for utilities and other entities to be able to buy demand flexibility direct from the market.Doug LewinSo much there. This is so exciting and so like music to my ears and hopefully to a lot of our listeners as well. So we, on a lot of the podcasts, we've had discussions in the Jason Ryan podcast and Will McAdams, Pablo Vegas, so on and so forth. I've been talking a lot about the distributed resources that are coming. And I use that word resources very explicitly because some of the resources are supply like solar, storage and the like. There's others as well: fuel cells, Bloom boxes, there's all kinds of different things that can be supply resources on the demand side. But then there's also the actual reductions of demand. And this is where it gets really tricky because you're kind of trying to measure a reduction as a, it's much easier to measure when you're adding something, it's kind of more difficult to measure the subtraction. Can you talk a little bit about how y'all do that and how you've evolved a system to do that?Carmen BestSure. I think at the crux of it, it's really about a counterfactual analysis. And this is something that's being done in lots of different industries to understand or academic spaces to understand what would have happened if we wouldn't have done what we did. Right? And it's not really an esoteric question. It is a quantifiable, answerable thing, but you need to be able to agree on a couple of important things like… well, what was happening and what was likely to continue to happen. We answer that by looking at past energy consumption and can make some assumptions about what energy usage would be in the future in a business as usual situation. And then the other component is having information of what changed. So how are you using energy after an intervention? And that intervention might be installing a heat pump, it might be upgrading your windows, etc. It might be an event-based intervention. So you did something in response to a call for load reduction or something like that. And you can be looking at that during a certain timeframe, maybe four hours or two hours or whatever the fact may be. But you're building that counterfactual analysis so you can see what that change in energy consumption was and then tie that back to the value that it was delivering during that point in time.Doug LewinAnd when you guys are doing that, Carmen, you're actually doing that measurement at the meter. So this is not a bunch of spreadsheets and slide rules and whatever else. This is not some grand estimate. You're actually measuring whether it's output of something like a solar panel or a battery or the reduction from something like a smart thermostat cycling, or like you said, a heat pump replacing some other system, right?Carmen BestYeah, we're typically looking at a whole building analysis, so be that a home or a commercial building and we're looking at their energy consumption at the utility meter, usually an hourly estimate. And then we have a timestamp of when an intervention, whatever we're tracking happened. And then we're looking at the energy consumption pattern after that intervention. It's also weather normalized. So you're factoring in how people typically respond to heat or cool or cold because that is one of the main drivers of our energy usage. And then more recently, we've been factoring in other big use cases like is there solar on the site? Is there an electric vehicle on the site, etc.? And those are other ways that you kind of have to factor in that site level energy consumption pattern either by backing them out or doing other analytics around how that change in consumption is happening.Doug LewinYeah. So just to take just one step back, we're going to go, we're going to go back into the, into the, you know, details here in a little bit. But if we take a step back, I actually want to go back to, and folks, if you haven't listened to this already after this podcast, you go back and listen to this other one with Pablo Vegas, where the CEO of ERCOT, and I asked him about his kind of ten year vision for where their ERCOT grid was going. And I, wasn't really sure what he was gonna say in that answer. Thought he might talk about, it could have been any number of different things, but actually the first thing he said was, and I'm quoting, “I see a future where there are distributed generation resources scattered throughout every community we live in. Those resources could be electric vehicles. They could be HVAC systems in your house that are connected to a smart internet connected grid thermostat.” He went on, he kind of gave this list of different things and said, I think this is gonna be one of the defining features of the grid in 10 years is all these distributed resources.So again, it's been a theme running throughout these podcasts and a lot of my writing on the Texas Energy and Power Newsletter and a lot of the discussions that are just happening at ERCOT, at the PUC, at the legislature and around the country and around the world is how do we orchestrate all of that stuff, right? We're going from a world where you had, you know, a few hundred really large power plants sending power one way. Now you're going to this world where you have thousands, even millions of tiny little points.And if I understand, and I'm not sure I do, so correct me if I've got this wrong, but I understand what Recurve is doing, you guys have built this kind of platform where all of these different resources can be on that platform and whether it's a utility or grid operator, essentially whoever's playing the function of the distribution system operator doesn't have to be somebody actually named capital D capital S capital O, but whoever is playing that function can then see at each of these nodes or whatever, what is happening and actually orchestrate all that tiny stuff up into something that is at a system level going to work to increase reliability, keep costs low, and deliver other values as well. So did I describe that right? Correct me where I got it wrong. If it is right, I'd love for you to talk a little bit more about that and some of the values that you can actually realize, because it's not just there's an emergency push a button. It is that, but there's a lot of other value streams in this, right?Carmen BestRight. I think a couple things that I encounter when I'm talking to potential customers or current customers. What you described of kind of that distributed world of devices is definitely one aspect of the future energy grid that we're looking at. And I think for folks to be able to understand how all those different devices are operating there's going to be different degrees of control that's going to be of interest to customers. Right? Some folks will want to allow their smart thermostat to be controlled, other folks might not. And what we're really offering to the market is being able to see where different types of interventions could have the biggest bang for the buck for that distribution system value. So you can use current patterns in energy consumption to understand where you might want to do a smart thermostat installations or heat pump installations, etc. And on the flip side, when those interventions have happened, be them an orchestration of a DER, maybe through a DERMS [Distributed Energy Resource Management System] platform, which we are not, we run alongside a DERMS and give that extra layer of insight of how did this actually deliver? How did it perform? And how did it perform not just for those devices, but in relation to the rest of the grid? We can do comparison analyses that are really insightful to be able to have that kind of bigger picture of how things are delivered. And then what that really enables is different types of aggregations that can be really valuable to the grid. So I think a lot of times when people are thinking about virtual power plants, or they're thinking about dispatchable devices, et cetera. They get really granular and they think about all the little widgets and how we're going to have levers, who can I envision, little levers all over the place that we're going to be able to control these things. But there's a lot of potential out there for just base load virtual power plants, right? Energy efficiency. And what we haven't really had is an ability to tap into that in a meaningful way that's really tied to the complementary grid value that you could get from installing technologies that are going to deliver a quote unquote shape that's going to be really grid beneficial for the wholesale grid, but then also for distribution systems. And also be able to be targeted to places where there may have been historical inequities, et cetera. So you can combine these different win-win propositions to values as they were to be able to deliver aggregations of systems and really optimize this ever increasingly distributed energy model that we have. So it can be bringing benefits to lots of different folks in the system, customers who are engaging with those devices, but then also all the rest of the people who are just trying to live their lives and have affordable energy in the world.Doug LewinYeah. And I love that you brought that back to energy efficiency, because it's funny. I like anybody who's listening to this podcast probably knows I talk about energy efficiency all the time. Some people get tired of me talking about energy efficiency, but I'm going to keep talking about it because it's incredibly important. But even like the way I asked the question, I was kind of leaving out energy efficiency, right? I'm asking about like, there's all these little resources that are able to respond at the push of a button. It's very easy for our minds to go there because that dispatch ability is important, but it's also important alongside, as you put it, these kinds of more long-term load shaping, right? That with energy efficiency, talk a little bit about, and you must see this all the time in the data, it's sort of how energy efficiency and demand response are actually not only complimentary, but almost kind of, maybe I'm overstating it, but kind of almost need each other.Carmen BestYeah, I think it's been one of the interesting evolutions of my career too, because I've spent most of my time in energy efficiency, not so much in demand response. Over my career, there are different people who have been working in demand response and energy efficiency, et cetera. One of the analogies that really has stuck with me is that demand response is really about building the fire station, as it were, making sure you're ready so you can drop load and be reactive very quickly. It can be a high cost for a really short event or for a really short intervention. Energy efficiency can be thought of as kind of pruning trees or making sure that you're mitigating the risk of needing to drop load at an immediate time. And how I see those really coming together in this modern context is that there are more and more devices that are controllable. So as we're investing in energy efficiency, like a heat pump, you're also getting the control value of being able to control that device in a DERM system to drop load, move load, whatever, in addition to the long-term kind of base load value of reducing a higher efficiency unit, right? Compliment that with insulation, for example, and having a shell that's also going to allow that heat pump to operate more efficiently and do pre-cooling and all that other stuff. Those two things can be coming together as well. Now where it gets interesting is how do we have programs and policy structures where we can be paying for both of those at the same time? Because right now I've heard folks say, well, I don't want us to do energy efficiency because then that's going to mean that I don't have as much load to drop when I need to. Well, sure, but you're also going to have less flexibility within the system to be able to manage the loads. And wouldn't it be nice if you don't have to drop loads in an emergency, but rather you're managing loads on a more regular basis. And I think for both historic demand response and historic energy efficiency to have a path for the future, which is really demand flexibility, they need to be able to come together and the value streams of that long-term avoided cost value from efficiency should be paid for and the short-term load reductions should also be paid for. Demand donation is kind of, I don't know if you came up with that term, Doug, but it's one that floats around our place a lot. If we're asking people to reduce load, they should be compensated for it. And when you can aggregate that all together with the different solutions that are out there, those two things can come together quite nicely and be able to create a system where you can really be bringing customers along on their own energy journey to optimize their own, improve their own resilience to be able to respond in emergency events, you know, extreme weather on either end of the spectrum of too hot, too cold, and also be able to address affordability at the same time.Doug Lewin Yeah, let's, so first of all, I didn't come up with demand donation. I like that. I often say uncompensated demand response, and I will give credit where it's due. I first heard Michael Murray of Mission:data. So folks should look up, we can put the link to that great organization in our, uh, in the show notes, but uncompensated demand response is usually the, the term I use, but I like demand donation as well.Let's stay though, for a little bit longer on energy efficiency. So in Texas, just a couple of days ago, we're recording at the end of April, just a couple of days ago, the PUC staff has created a new docket and put some questions out there. Looks like they are beginning to make some moves to implement Senate Bill 1699 by Senator Johnson, who was on this podcast a few weeks ago. Energy efficiency, I think it would be hard for anybody to argue the counter at this point. There's a lot more potential for energy efficiency in Texas right now. But we're kind of struggling to get over the hump of really sort of putting something bigger into place. And I think there's a few reasons for that, but I think one of those kind of comes back to, there's this kind of skepticism of what the savings are and are they really being realized? And I know we've already covered this a little bit, but I just wanna make sure we do cover it quite thoroughly because…Carmen BestIt’s my favorite subject.Doug LewinRight, because it's your favorite subject. This is one of the reasons why I asked you to be on this podcast. Cause this is really what I want to talk about is how do we really know the savings are materializing? And, and you know, the, the system that we use right now obviously has a lot of deemed savings that you basically, right, there's a calculation. If I do this, I replace this thing with this thing. I get X amount of savings. But what you all are doing is actually again, measuring that at the meter. But that means you're also just not getting this kind of like peanut butter, here's what it's saving over the course of a year. But here's what it's saving at, oh, let me just randomly pick a time, 7:30 or 8:00 on a summer evening or 8am on a winter morning, right? You can really see where the problems on your grid, where the savings are materializing or where they're not, where you might be spending dollars on energy efficiency and looks like this thing we said we were doing rather than having years and years of studies and reviews and all this, you have the meter data instantaneously. I'm going to let you respond. I'm so sorry. I asked really, really long questions, But there's one other aspect of this too, right, which is location, right? You can see where those are happening. So it's not just this kind of peanut butter throughout the entire Houston area, the entire DFW. No, there's a transmission constraint somewhere and we need, you know, not only demand response, but there's all this discussion right now of a transmission constraint that caused problems on September 6th last year, and they want to go after demand response. They could conceivably go after energy efficiency too, as long as, but the grid operator has to know that those savings are really there, right? Carmen BestRight. Yeah, I think, you know, the reason I say it's kind of my favorite subject is because I'm, as a former PUC staffer myself, I'm all about the accountability. This has been the trajectory of my career the whole time of like, how do we know this stuff actually showed up? And it's not always so popular to be the evaluator on the back end, but I think it's becoming increasingly important. And it's become a lot easier to be able to do that in effect because of our computing power and kind of enabling software and smart meter infrastructure, etc. that's available to us today. So there's really no reason not to use it in my humble opinion. I think we're even seeing states kind of move into a paradigm where they're like, measurement is our default. We're going to start with that and we'll make exceptions where we can't measure it. In those cases we'll use deemed savings estimates, which is just fixed estimates for a single technology. And I'm really encouraged by that, because I think it does just what you're saying. It allows you to align that measured savings impact with your end in mind. So if you know that you are going to have hotspots on the grid that correspond with solar turning off, four to nine, for example, that's kind of where things happen a lot, then you can make sure that the efficiency interventions that are being installed are going to be kind of aligning with that value and you couple that with controls that are also going to deliver a value at that time. What we've seen is when you set that up of here's what we need, when we need it, and where we need it and put an appropriate value next to it, aggregators and folks out in the market can definitely respond to that. That is no new news to Texas, I'm sure, because you've had markets out there for a long time, but I think that are constructed with that intention in mind of being able to drive down price, etc. But I think that being able to have that avoided cost curve that is aligned with grid value is a really important piece of the puzzle to have out there.The Texas PUC recently adopted kind of a new avoided cost, but I think it's not quite yet aligned with more of the distribution system value that might be out there to get a little more targeted for locational value that could be captured. The emissions value would be another thing that could be captured in there as well. And then we've also experimented and had some success with using adders for equity. So making sure that populations that have historically been kind of left out of the equation can be part of the solution going forward. And I kind of went off track there for a second. Let me come back to why measurement as a default is really important. It's important for us to know, like for the Public Utilities Commission to know and for ERCOT to have visibility, et cetera. But it's also what can drive business and appropriate incentives and compensation for the folks that are doing the work. When you have just an average for every technology that's installed, it really doesn't differentiate between a quality project and a non-quality project. You're kind of just counting widgets installed. And at the end of the day, you have this broad assumption of what they might have delivered, but you don't have that tied back to what they actually delivered.Sometimes when I first say this, folks say, well, how are they going to know or isn't that too risky, et cetera? And I think what we've found is as we've run what we call FLEXmarkets and put that price out there and ask people to respond, it maybe takes one cycle, maybe two for them to figure out kind of what the right projects are going to be to be able to deliver on that and start making a return on that. And it's so much nicer because they have so much more flexibility of how they can meet their customer and deliver products and services to the customer that are going to resonate. They don't have to deal with all the fixed incentives and that almost becomes like baggage versus being able to deliver a value stream that really fits into their business models and how they're delivering services to customers already.Doug LewinAnd when you're saying ‘their’, you're talking about the folks you're referring to would be like it could be HVAC companies, it could be energy efficiency implementation companies, it could be retail electric providers, who else? Carmen BestYeah. Well, those are the main ones. I mean, we kind of have this block of folks that we call aggregators, and they're really all the entities that have creative business ideas to go out and get these impacts, right? I hesitate to say savings because sometimes it's going up, sometimes it's going down, and really what we want is the value…Doug Lewin And sometimes we want it going up too, right?Carmen BestExactly.Doug LewinI mean, we're starting to have a system where we want to get more resources to come on the grid, but if those resources really don't have any value for many, many hours during the day, we actually may want to start driving energy use and demand up at certain points. And that's also energy efficiency.Carmen BestYep, exactly. And that's where some states are starting to consider moving away from savings targets and move into system benefit targets. Even in Texas, the performance incentives are really anchored with the system benefits that are delivered. No extra commentary on if it's the right amount at this point, but the concept of being able to set a goal based on system benefits that's inclusive of all the different types of benefits, not just the savings, I think is one that needs to kind of continue in the conversation because it actually normalizes across a lot of other things that could be out there and allows you to translate that back into a market price to go out and set up systems, simplified systems to go get energy efficiency and demand response potentially that doesn't have a lot of the centralized planning baggage that we've historically had with program design.Doug LewinSo yeah, not only is it not centrally planned, it's really, yeah, it can be almost exactly the opposite. It becomes a very open competitive free market with the thing that the regulator or policymaker or whoever's making that decision does need to decide is what are we after, right? Like what is the outcome we're trying to reach here? Is it reductions at a particular time of day? Is it trying to bend the cost curve for the cost of the distribution system, which has been rising very, very rapidly over the last few years and putting a lot of upward pressure on prices actually? I get asked this question all the time, right? Like why if renewables are so cheap aren't energy bills going down? And one reason is that while the cost of generation, at least this is true in Texas, the cost of generation has gone down, the cost of distribution has gone up a lot and has eliminated a lot of those savings so they don't show up. There's other reasons. It's not just that simple, but that is one of the reasons. So as long as a, whoever the policymaker is, in this case, it'd probably be the Public Utility Commission would say, here's the goal and the outcome we're after. And now we have, you would have a system where you're measuring the outcomes and then saying to the market, I don't really care if you do windows, thermostat, HVAC, solar, storage, EV charging. I want the outcome. You guys figure out how to get there and then these companies are competing in that market. Is that right?Carmen BestYep, exactly. So it creates more customer choice, which is a good thing for everyone. It should be able to drive the cost down. I think it's also appropriate for the state agency, at least in the early days, to lay out kind of a willingness-to-pay price of” this is how much we're valuing this avoided cost value”. And then the TDUs or Transmission Distribution Utilities could be including, along with that, the distribution system value that might complement the overall. Then you can stack as many other values on top of that as you might want to. But yeah, at the end of the day, you have this open platform that can also continue to evolve. So as you solve different parts of the grid challenges that exist today, you can shift that price to other parts or other components. It can continue to be factored into services and offerings and new technologies that we haven't even thought of yet that might be available to customers to manage their energy consumption or optimize their bills, meaning bring them down as low. In most cases, maybe there's folks who want to optimize their bills the other way. But all of those things combined can be much more streamlined way to get at that otherwise stranded resource that's out there. Folks don't do efficiency just on their own a lot of times because there are other barriers to adoption and timelines and things like that. It's totally appropriate to kind of build that into the grid value that is going to be coming from those interventions. And it should be something that utilities can really buy from the open market without having to have, you know, high cost programs or rate payer funded programs that don't really have that accountability. Let me touch on that point for a second too, because I think another component of accountability that I think really resonates with this sort of model. You can call it FlexMarket Model or the Market Access Model that we've been working on. Since it's a pay for performance model, that if the savings or the impacts are delivered based on what everyone's agreed to as the value and they match that measurement regime, i.e. they show up, then ratepayers are only on the hook for what's actually delivered as opposed to having to make a good faith investment in a program or a big construct of, we don't know necessarily if the program is going to deliver or not. But when you're deploying it like this with this open market model, it's scalable. If the price is wrong and folks don't show up, well, you won't get anything. If the price is not aligned with what the market is willing to deliver, then ratepayers are actually protected from not getting what they paid for, right? I don't know if that came out quite right…Doug LewinYeah, no, that makes sense. That makes perfect sense. I mean, you're, again, because it's actually measured and you actually know what happened, you're, I mean, it's pay for performance. It's paying for what actually happens. Carmen BestYou're paying for accountability. I think I wanted to add a couple of anecdotes too because it's been interesting for me. One of the great things that I like about working at Recurve is I get to kind of explore the whole country. Texas has been one place that I've been learning a lot about in the last year and a half. There's two other states that are kind of interesting that recently took this measured by default approach. It was really on the basis of this ratepayer accountability construct that they kind of, was in essence the tipping point. So Kansas in adopting their energy efficiency paradigm said, we want to make sure that it's measured on the basis of accountability. And Louisiana was the other state that did that recently. So they adopted a new statewide energy efficiency program paradigm. And they too said, we're going to measure it first. And that's one way that we can build that confidence that it was actually delivered.Doug LewinVery interesting. But they're doing that within the context of what you would still call a utility program, right? Or no? Carmen Best Exactly. Yeah.Yeah. Okay. So there is this kind of like a little bit of a threshold question. I think it's a really interesting one for Texas as to whether or not, and I think this is very much an open question in Texas right now. Maybe it's not an open question. It's certainly out there. Are the utilities the right place to have the energy efficiency programs, is there sort of a need to go maybe from quote unquote programs to more of a market based structure? And that's a little bit, little bit scary because nobody's really sure exactly what that means. Does it live at ERCOT? Like where does that market even live? So there's, there's a little bit of like a crawl, walk, run there, right? Of like, the first step is to within the utility programs, move to more of this pay-for-performance see how it works. I think a lot of people want to kind of jump to the absolute, but if you kind of take that crawl, walk, run, even if you don't move past the crawl, you've made some good progress because now you've made sure that there's accountability and more accountability. Not that there's not any, there's definitely, don't want to sound like, the programs right now are great for what they do. The EM and V that's there shows like $4 a benefit for every dollar spent. So this let’s be very, very clear there that the programs as they are today are good, but as with anything, there's always room for improvement. So if you look at this as kind of an evolutionary step of now we'll measure it, make sure that the things that are showing the biggest, again, I almost said savings, but the biggest impact that we're looking for are the things we continue to do. And as you're doing that, then that gives policymakers in Texas, that at least historically, maybe a little less so today, but hopefully still favor competitive markets, the opportunity to gain some lessons learned and look at how this might actually evolve from a program to a market.Carmen BestYeah. And I think there's two really interesting opportunities that are out there that kind of demonstrate that experimental stage as it were. One is, you know, Texas is obviously on the cusp of some reliability challenges. And to me it occurred... Doug LewinCusp? I think we're there. I think we’re past the cusp.Carmen BestOkay, we're there. The fact that Texas is facing some reliability challenges, I think there's an opportunity to take a no-regret step to carve out a space that could test this open market model and really position it to be incremental to the existing portfolio. This has been done in other states and all the tools and capabilities are readily available for the utilities or in partnership with REPs even to be able to go out and do this, set up an open market model, you can mirror it after the FLEXmarket model or the market access program design. Set that up and then give a little bit of space from the existing regulatory models of the performance incentives and kind of all the energy efficiency legislative mandate stuff and say, we're going to run this alongside in the interest of reliability. No regrets. Ratepayers are only going to pay for what’s delivered. Try it out and see how it goes. And then based on the data and the information that comes back, see if we want to scale it up, scale it down, get rid of it, whatever you need to do. I think that really aligns with the objectives of SB 1699. It could also align with the opportunities for federal funding that are present in the Inflation Reduction Act and the HOMES program, which allows for a measured pathway. So that would be the kind of space where you could have a residential pathway that is using this measured pay for performance sort of model, which could be synergistic with whatever the state energy office ends up doing with those funds, but it'd be effectively compliant with that IRA funding mechanism. And I think by having kind of a sandbox or a space in which you could be testing that out, you’d be able to, like you're saying, do the crawl, walk, run and kind of get a sense of, okay, what's the right value for this? That would definitely be something that would need to be figured out. And then, who's showing up to deliver it? How did they do? And do we want to do it again? And I think that could happen in the EEIP, in the efficiency proceeding. It might also be possible in the ADER task force where they're calling for more blocky resources. You know, first round of ADER injection resources, etc. I was fortunate to be on the task force and kind of contribute to that, but mostly witness kind of how it played out. And I think the next iteration I'm really hopeful that we can get to some of these longer term resources too that might not have that granularity in the data specifications that have to be SCED following, etc. But they're offering other resources like that long-term load shaping that can also be really valuable to the grid. So there's lots of places where it could be tested, and it's really a matter of which ones are going to resonate at this moment within Texas.Doug LewinYeah, let's, I want to come back to the ADER task force in just a minute. I do, I do want to talk about that, but before we go there, I just want to go a little further with what you're talking about there. So what you're suggesting is the, if I'm hearing this right, is you basically… So the PUC has this requirement right now it's in Senate Bill 1699 that says you got to reduce average residential load.  They have just put out some questions, hired a Director of Energy Efficiency. The energy efficiency programs, the way the timeline usually works is the utilities put in their plans on April 1 for the next year. So they've already put in those plans. Two PUC meetings have come and gone. The PUC hasn't, none of the commissioners, at least publicly in the meetings, have said anything to the utilities about their plans. Those are probably gonna be the plans in 2025. What they could do though, and there's two things that are really kind of top of mind for me here as far as something like you're suggesting, there's like an add-on. Like you do this market on top of your existing utility efficiency programs. There is this, um, looks like there's going to be an RFP for demand response. Again, what they're, what they're trying to do here is you have a, you have a major line from south to north. They are worried that line is going to be overloaded so they're curtailing generation on the south part. They need to reduce demand on the north part to keep everything balanced. So basically anywhere north of like somewhere south of San Antonio. So you're talking about San Antonio, Austin, pretty big areas. The value of that DR goes down when you get up towards Dallas, Fort Worth, but pretty valuable there in central Texas. They could put something like this in place for those areas and not limit it to demand response because to the point with this kind of measurement, you can actually see at the time the constraint is binding or whatever. When that is happening, they could measure, wow, that energy efficiency intervention we did delivered more than the demand response or delivered less. I don't know how that's gonna come out. We can all guess, we can all sit around and guess, but the point is to actually have the data and we can have that, right? So that was one example. I'm gonna give one more and then I'll stop talking. The other example would be, we're recording in the end of April, but like winter's not that far off. And if you wanna actually get after resistance heat, that could be, they could say, okay, rather than utilities change all your programs and focus on resistance heat, we're gonna do X number of dollars, take a $30 million day that we're gonna do this and like really go after resistance heat and we're gonna measure it so that we can see that all these people out there, the SPEER, the South-Central Partnership for Energy Efficiency Resource, American Council for Energy Efficient Economy, people like me that are always shouting about resistance heat, like let's get the data and see if these people are right. Maybe they are, maybe they aren't. There's one way to find out is to actually get a market going and measure and see. So those are two examples. Would those make sense to you? Can you think of others? Carmen BestYeah. Those to me, the first one in particular sounds a lot like a non-wires alternative sort of solution where you're saying, here's where the constraints are, go get it. It also occurs to me that the timelines for being able to deploy that are a lot shorter than being able to build other resources to fill that gap, right? So that was one of the reasons why the FLEXmarket was launched really quickly in response to summer reliability issues and within about six months, it was starting to deliver impacts. Now, did it meet all those obligations? No, but it was really a no regrets sort of investment to reduce load at certain times of the day. I guess I'd boil it down to, and then the second one on heat pumps and resistance heating. That's really looking at like there's clearly an opportunity here for reducing constraints. Those could be geolocated as well. In addition to the general no regrets technology targeted solutions as well, you could also target it to build impacts, I would say too, because there's probably a lot of folks who have resistance heating that are suffering from really high bills that don't need to. So that'd be another dimension that you could do that targeting. I think the main thing right now of looking at a potential pilot or an opportunity for standing up that type of FLEXmarket or open market model could be deployable in all of those different situations in my view. And it kind of boils down to three key things. One is setting measurement as a default. So if we're going to run these things, we're going to measure them. We've got the data to do it. Let's do it. The second one is let's make sure that we're not leaving resources out, demand flexibility resources. So if we're going to run demand response ops, let's put energy efficiency in there too, at least don't exclude it. Then the third thing being, let's use an open market model so that we can bring as many brains as possible to the creative technologies and solutions that could solve that problem, then just limiting it to one technology potentially. I think another one of the solutions that's out there right now is the smart thermostat pilots that were just proposed. I think they're a great solution for getting to summer reliability this year. I think that you could do more by saying, let's measure these impacts. And then let's also make sure that they're tied to a longer term solution. There's urgency now to get this done, but I also think that it could be tied to a continued partnership between TDUs and REPs, which is kind of illustrated in the proposals that are in front of the commission right now to really be going after these resources together and making sure that they're addressing reliability for the state and reliability, affordability, and sustainability too, because it also delivers a lot of emissions reductions.Doug LewinYeah, absolutely. And so, yeah, you're referring to the filing by the transmission distribution utilities. We'll put a link in the show notes so people can see that. Give a lot of credit there that they are, I'm critical a lot of times. I like to give credit when good things are happening too. They have proposed a smart thermostat pilot program. I believe Oncor is one of the four Investor-Owned Utilities within ERCOT that wants to implement that this summer, CenterPoint 2025, AEP-TNMP 2026. And the Retail Electric Providers put in comments in support of that too. Doesn't happen a whole lot where you see sort of TDUs and REPs, the Retail Electric Providers and the Transmission Distribution Utilities in Texas agreeing on something like that. So that is really exciting to see. And I think that's a great point that one could be a really big deal for this summer and could scale a lot faster if we really had the ability to measure where the results are happening. That makes a lot of sense. I do want to talk about the ADER Task Force a little bit. You mentioned you are on the Task Force. You have seen how that has evolved. In the beginning, it is really pretty focused on, I think as you referred to it earlier, an injection. That really is like storage injecting power into the grid. My understanding of this is that it is being set up such that aggregated load resources, aggregations of for instance, smart thermostats or sensors on electric hot water heaters or any number of other things could also participate there. And that is a value stream we really haven't talked about yet. We've talked about a bunch of different ones, but we really haven't talked about ancillary services, which have gotten a lot of attention over the last year for being very expensive. There's arguments about how expensive they actually are, but there's no doubt it's in the billions of dollars. And this is a potential to have more competition in those markets too, and for customers to actually earn some back. Are you bullish on the potential for demand response to actually participate in ancillary services or, because our listeners are probably going, well, of course she is, that's obvious. But I think the question though is, is that really the best place for demand to realize value? Or is that, because I think there's at least a potential, I'm not sure about this, I don't think this way, I'm just sort of putting it out, there's a possibility, is like, maybe there's other places where there's a lot more value that could be tapped into and sort of chasing after ancillary service markets, maybe that's better left to others and demand should really play somewhere else. I'm curious how you see that. Does aggregated demand response have a big role to play in ancillary service markets? And tell us anything else you wanna say about ADER Task Force and the pilot and where you think it might be headed.Carmen BestYeah, I mean, first of all, I'll just say like I'm a huge fan of task forces. I think just the timeline and the impact of that group in such a short timeframe, I've been involved in many working groups over my career, but this task force was like in terms of delivering on a promise quickly was really impressive. So I'll start with that. That's not self-congratulatory. It was more like to say that the team was really focused and got the job done. And having Commissioner McAdams and Commissioner Glotfelty at the helm really made a big difference on that front too. Now, where it goes, I don't know either Doug if that's going to be the best place for demand response to play. But I am generally a fan of open definitions on a lot of stuff, like trying to be able to consider all the different resources that can come to bear for different services that we might need and then be able to value them properly. So ancillary services, it's not something I've spent a lot of time in my career looking at. I think there is probably some value there, but I also think that isolated by itself, we probably won't tap into the full value that could be out there. So having kind of scaled options for being able to get, you could do an ABC of, okay, in some situations we're just going to have energy value, long-term value, and then in certain instances you'll also be able to be paid for the ancillary value. So in effect, if you can deliver more services, you'd be paid for, excuse me, paid a premium for delivering on that maximum value. Same with SCED following. If you can follow that …Doug LewinAnd go ahead and explain, because we do have some folks that are not as in the industry. So do explain SCED following if you would. Carmen BestOh gosh, now I'm going to have to explain that…Doug LewinWell, it basically is following a market price. It's following, right?Carmen BestYeah. In my simplified version of it, you have to follow basically the market price on a minute-to-minute, almost minute-to-minute basis, which is a pretty complex technical task that was achieved, might I add, and impressively so. Kudos for that.  But, that is not the threshold that every demand response or energy efficiency service needs to be held to because there's a lot of value to be had and to be seen on an hourly basis, for example. So all of that to say that if it can be delivered that way, I think it's appropriate to pay a premium. And then for the other resources you may pay less per unit, but the volume and the scale is where you're going to be able to do more comprehensive projects that are really going to extend the value of this distributed system being more effective all the way out to the endpoints, endpoints being homes, businesses, et cetera, and not just the injection of specific technologies, injection resources from specific technologies, et cetera. So that's kind of where I land on that one.Doug LewinYeah, so, if I were to summarize that, and I think this is where I fall on this too, is sure, like DERs, whether it's demand response or what have you, should be able to participate in the ancillary service markets, whether or not the aggregators that are serving customers are able to make that work or not, that's a market function. In other words, you need to enable it, make it possible. And then you see if the value stream is big enough to justify it or if, you know, maybe large battery operators are just able to do it more efficiently, from an economic perspective more efficiently, then they win that competition. And that's fine. And you look for places that demand response and energy efficiency can realize other values. But you don't foreclose that. You don't just take it off the table. You see what happens at the competition.Carmen BestWell, and I think that also brings up in my mind, Doug, some of the debates around like double counting or where you can participate in multiple markets. I think in most situations, we want to be able to enable folks to capture value streams from multiple services and solutions, within reason, obviously. We don't want people getting paid twice for the same thing. But in most cases, they're delivering different things and it's possible to segment the value streams for each of those things and therefore we can pay for them separately such that we can get more resources and more flexibility onto the system, which will reduce costs for everyone in the long term.Doug LewinMakes good sense. Speaking of reducing costs, you said something earlier that caught me and I wanted to follow up on. You were talking about the DR RFP that looks like it's gonna be coming out from a PUC or ERCOT, I guess it'd be from ERCOT sometime in the next month or two and about heat pumps. And you mentioned you could also have a focus on customers that are having trouble paying their bills. So we know in Texas, I believe it was a TEPRI study, the Texas Energy Poverty Research Institute, that 40% of Texans they surveyed were struggling to pay their power bills. That was down a little bit from 45% the US census had in 2022, but that's two different data points that tell us just shy of one out of every two Texas families are struggling to pay their energy bills.So with this kind of a model, you can also then, as you were pointing out, and I'd love for you to talk about this more, I don't know if it's happened other places, but in the state of Texas, and Texas is in a minority of states on this, the state of Texas does not even track utility shutoffs, or maybe it's tracked, but it's not reported. There's no place you can go on the PUC website or anywhere else and see that last month, X number of customers were shut off. But obviously the utilities know where those customers are because they're doing the remote disconnect. And I don't know exactly what the, in great, great detail, what the process is before you shut somebody off, but it seems like somewhere in that process, there should be a step where somebody looks at, oh, this person, I can see on a tax record, this person lives in a 1200 square foot house and their energy bill was $450 last month. Something's wrong, we probably should target some energy efficiency savings here, so they can actually afford their bill and not be shut off. What is the potential? I mean, there's potential for that in sort of usual, and maybe the answer is just, yeah, if you have a FLEXmarket and you have this sort of measured rather than deemed, sure it's possible, but it's no different than the other programs. Is there some additional sort of benefit of evolving utility programs to actually get the measured performance that extends to low-income customers at risk or who are actually about to be shut off?Carmen BestYeah, we don't have a lot of the intel on the about to be shut-off question because we just haven't deployed our platform in that respect. But effectively what we're doing is we're looking at those consumption patterns and then can match that to rates that align with low-income customer streams. If those exist, in Texas, you could be layering those two pieces of information together and then be targeting interventions for those populations or those individuals. The other component is obviously geographic targeting because since, for better or worse, poverty is usually concentrated geographically, that's also a pretty straightforward way that you could be looking at different pockets of the population and then be going out to offer services and solutions for them.Now, I don't have a long experience of low-income programs either, and I know they introduce a lot of unique challenges of delivering services from building infrastructure. It's hard to retrofit a home that's on the brink of falling over, for example. So I know other states though have used their public funds to be able to address kind of those basic issues and not have to jump over serving those customers, but be able to do the basic upgrades to their homes such that they can do the retrofits and really tighten up their homes, etc. So having those structures that are supporting customers to get to a place where they can reduce their energy waste is one of the solutions that I've heard. And I think it's in Missouri has a good example of that where they use those funds to do that.I think for the grid value though, those homes are wasting a lot of energy for the grid and for the individuals that are living in them. So there's no reason to segment who's benefiting from those or not benefiting from those homes being upgraded. All customers in Texas will benefit from those homes being better insulated, a more solid shell, et cetera, and reducing the cost for those individual customers will also reduce the cost for the whole system because there will be less energy that's wasted likely in the system. So I guess back to my point of like there's multiple value streams that should be recognized and captured. And there's been, just to give one example of where we've set up a FLEXmarket wherein we have the avoided cost value. That's kind of the base price for delivering the energy services, so you may do a retrofit. And then we have adders that are built into that market. If you're delivering to a customer that has a low income, you'll get a 2X incentive. If you're doing the work with a local contractor that actually serves that community, so there's local economic development benefits, you'll get another adder. So since you just have this market price, you can kind of stack these other components on top of it to draw people into places where you want them to be delivering services and getting value both for the customers and for the grid. And it doesn't have to be like a whole separate program, etc. You can really just be calibrating it to address the different needs that are out there in the market.Doug LewinGot it, that makes a lot of sense, that's great. I want to just ask you, you know, sometimes I started this podcast with the intention of asking every guest like the same three questions and it just hasn't really panned out that way because the conversations end up being really interesting to me and I run out of time. But I do want to ask you, because we talked about it a little bit beforehand, the two to three energy policies you think would have the biggest impact to increase reliability, lower systems costs, reduce pollution. I thought your answer is really good. I think it'll probably also work as a pretty good summary of this conversation. So what are the two to three policies you think would have the biggest impact?Carmen BestSo it kind of comes back to the core things that Recurve has been doing over the last years and that really is measure everything because we can. And it's not measurement just for the academic exercise, but really because it drives the ability to target and get to where the resources are going to have the biggest impact. Make sure that they delivered on those resources, and then also have fair compensation for quality projects. And that means fair compensation both for the folks doing the work, but also fair compensation for the customers that might be participating in demand response, etc. So that's kind of my number one, and that's been a thread throughout my career that I'm sticking with. And that's another thing that Recurve is really operationalizing in a way that's streamlined and accessible. It's not really that we invented measurement, not by a long shot, but it's where we're putting it into the process to make it accessible and actionable is really where we're trying to innovate on that front. So measure everything and make it accessible so people can act on it. The second one is really being like we talked about, making sure that demand flexibility is all-inclusive of energy efficiency and demand response when and where we can. So you have, don't leave one out of the other and definitely don't leave them competing with one another. Come up with systems wherein they can be operationalized together and the value streams that are delivered both from that long-term intervention and the short-term can be calibrated and managed with a common measurement framework. That's what we're doing in the Recurve platform. We can measure both of them using the same basic models. It's just different time horizons, which allows you to do both and not, quote unquote, eat each other's lunch, right? Doug LewinYeah, and before you go to the last one, I'm gonna mention this RFP that ERCOT’s gonna put out again, because I think it just shows where there is, in the grid operator mentality, and I can understand this, it's like I have to be able to push a button to get it, but you really have to take a step back and say, is that actually true? If I have reduced the demand, right? And the key is you have to know that it's there. And I think that's where, this is why I wanted to have this conversation. I hope this is becoming clear to the audience or was clear long ago, if not as clearing up now, because this is really what Recurve potentially brings is the ability to then measure that energy efficiency. So as a grid operator, a regulator, do I really care if it's demand response or energy efficiency? What I need is the demand reduction. And I totally get not being able to run a grid on a deemed savings, but on a measured savings, that's different, right? Carmen BestYeah. And it's really giving visibility to that end in mind. Like, what do we want? We want grid impacts. When do we want it? Now? That sort of theme, right? Yeah. And then I would say the last thing is really allowing for open market models to kind of thrive and answer these questions. I think you'd brought it up on a podcast a couple of weeks ago. When we have open markets, it allows us to bring all the brains together to be able to solve these problems. And I think as the grid is evolving and as technologies and customer journeys and energy are evolving from electric vehicles to all kinds of different technological solutions, our ability to adapt to that is going to be very limited if we're trying to pick each technology and design a program around it. If we have an open structure that's focused on the end in mind, we're going to be able to have a lot more fluidity in evolving the solutions in addition to having more accountability for driving towards those end goals. So that's really what the market access model and the FLEXmarket design was really all about. And having the tools, and solutions, and software to enable that transactive visibility and that end in mind is really what Recurve is all about.Doug LewinYeah. And it actually makes your head hurt if you do this sort of, you know, empathetic, you know, work of putting yourself in the position of a regulator who is like among everything else they've got going on. And they've got, you know, I think they said today at the PUC open meeting, they've got like 25 rule makings out of the last session alone, 25 out of the one before that. And now you need to like design, you need to actually work with utilities and stakeholders to design a program for EV charging for electric hot water heater sensors for HVAC cycling for, I could go on and on and on and on. You don't need to do all that. You need to set up one market where you measure the results and pay based on the results. And now, not only do you not have to design each of those individual ones, you've created a structure where when the next thing comes along, you don't have to design it. It can just come right into the market.Carmen BestRight. Yep. And you imagine all the vendors that are out there as well, like knocking down utility doors too trying to sell them the next greatest thing too. It's another way that you can kind of synergize across all those solutions and use it as a test bed as well. Like what does it deliver in this context? And then make decisions and resource allocations based on that as well. But also give more choice to folks so they can take their own journey into this next, this huge energy transition that we're all experiencing.Doug Lewin And they can really accrue to the benefit of customers. It gives customers a chance to be part of that energy transition, put some dollars back in their pockets. This doesn't have to all be a one way street from the customers to the energy companies for more and more, turn that flow around and let customers get some back too. Carmen BestYep. Doug LewinCarmen, is there anything that I should have asked you that I didn't?Carmen BestOh boy. I don't think so. I think this was a great conversation. It's really been exciting for me personally to get to know Texas and learn more about how things work. I have to confess, I was a little intimidated in the early years, but what I've found is people are really solutions oriented and really excited about trying to find solutions. And that's really inspiring and invigorating. So I've just been really honored to be part of the conversation at all and hopefully have contributed to the next solutions that might be viable for addressing reliability, affordability, and sustainability.Doug LewinYeah, you already have and I'm looking forward to see what's next. And I think you're right, like the ADER Task Force is a great example of where Texas can move really fast. And maybe what we need is, now like a demand flexibility spinoff of the ADER Task Force or something. Carmen BestAll right. Now you’re talking.Doug LewinStay tuned. All right, cool. Carmen, thanks so much for being with us.Carmen BestThanks, Doug. This is a public episode. If you’d like to discuss this with other subscribers or get access to bonus episodes, visit www.douglewin.com/subscribe