WEBVTT - Ercot’s Extremes: Weather, Solar and Bitcoin 0:00:00.080 --> 0:00:02.480 This is Dana Perkins and you're listening to Switch on 0:00:02.600 --> 0:00:05.360 the B and EF podcast. So, as the saying goes, 0:00:05.600 --> 0:00:08.560 everything's bigger in Texas, and for the last few years 0:00:08.640 --> 0:00:12.720 that has also included swings in extreme weather conditions. When 0:00:12.760 --> 0:00:15.680 storm Jury hit the state in February of twenty twenty one, 0:00:15.960 --> 0:00:18.360 Texans rushed to heat their homes and the demand for 0:00:18.400 --> 0:00:23.000 electricity outstripped supply, sending prices soaring to nine thousand dollars 0:00:23.000 --> 0:00:27.400 a megawatt hour and causing blackouts At a critical moment. Urkott, 0:00:27.520 --> 0:00:30.319 the state's grid operator, brought in reforms to avoid the 0:00:30.320 --> 0:00:32.800 issue persisting, but from late June of this year, a 0:00:32.880 --> 0:00:37.600 heat wave once again caused electricity prices to a times skyrocket. 0:00:38.200 --> 0:00:41.680 So what makes Urkut and these price fluctuations so unique? 0:00:41.720 --> 0:00:44.239 And as we prepare for more extreme weather, how might 0:00:44.280 --> 0:00:47.879 the state, as an independent energy island of sorts adjust? 0:00:48.560 --> 0:00:50.800 To explain what we might be able to learn from 0:00:50.800 --> 0:00:53.320 this unique market. In today's show, I get to speak 0:00:53.320 --> 0:00:56.600 with benf's head of research for North America, Thomas Rowland's 0:00:56.640 --> 0:01:01.320 Rees and BNF's US power specialist Natalie Mandy Bratta. Together 0:01:01.400 --> 0:01:04.360 we discuss Urcot and how it operates the Texas grid 0:01:04.480 --> 0:01:07.679 and what this means for electricity prices. We also discuss 0:01:07.840 --> 0:01:11.520 Texas's growing renewable energy rollout and where solar fits into 0:01:11.520 --> 0:01:15.080 the equation, and finally, Bitcoin in Texas and how it 0:01:15.120 --> 0:01:19.080 fits into grid flexibility. As always, if you like this podcast, 0:01:19.200 --> 0:01:22.520 if you subscribe, you'll receive an update on future episodes, 0:01:22.720 --> 0:01:24.480 and if you give us a review, it'll make us 0:01:24.520 --> 0:01:27.440 more discoverable by others. But right now, let's jump into 0:01:27.520 --> 0:01:40.520 the conversation with Tom and Natalie about Urkott. Natalie, thank 0:01:40.560 --> 0:01:41.480 you for joining us today. 0:01:41.560 --> 0:01:43.280 Thanks so much. Shanna, it's great to be here. 0:01:43.400 --> 0:01:45.160 And Tom, thank you for coming on. 0:01:45.000 --> 0:01:47.000 Switched on, Thank you so much for having me. 0:01:47.240 --> 0:01:49.280 So, Tom, you've been with the business as long as 0:01:49.320 --> 0:01:52.560 I have, which is, you know, since before we were 0:01:52.600 --> 0:01:55.080 a part of Bloomberg. And for those who are listening now, 0:01:55.080 --> 0:01:57.160 I thought we would have this moment where I was 0:01:57.200 --> 0:02:00.240 thinking about saying, Tom, welcome back to the show today, 0:02:00.560 --> 0:02:03.800 But actually you haven't been on with me, and you 0:02:03.880 --> 0:02:06.440 were on one of our initial pilot episodes, which I 0:02:06.440 --> 0:02:08.760 believe may not have seen the light of day. So 0:02:08.880 --> 0:02:11.800 what is it that you're hoping to realize in today's show. 0:02:12.160 --> 0:02:16.040 Well, firstly, I hope that we can give an entertaining podcast. 0:02:16.040 --> 0:02:18.320 Then I think that so long as I'm not too boring, 0:02:18.360 --> 0:02:20.160 we should be able to do that because we're going 0:02:20.200 --> 0:02:23.280 to be talking about URKOT, the Texas power market, which 0:02:23.440 --> 0:02:27.239 is like the Wild West of power markets. Everything is 0:02:27.360 --> 0:02:29.120 just maxed out in URQUT. It's one of the most 0:02:29.160 --> 0:02:32.560 interesting things that I've had the opportunity to study in 0:02:32.639 --> 0:02:34.760 my varied career at BNFS. 0:02:34.800 --> 0:02:35.720 So I'm really. 0:02:35.520 --> 0:02:39.000 Hoping that we can convey why it's so interesting, why 0:02:39.000 --> 0:02:41.200 it's so extreme, and also maybe to make a little 0:02:41.240 --> 0:02:43.960 bit of a defense of ERCOT, because you know, one 0:02:43.960 --> 0:02:45.640 of the things we'll talk about is it's kind of 0:02:45.639 --> 0:02:48.880 fashionable to dunk on ERKOT because it's so extreme and 0:02:48.919 --> 0:02:52.239 it's Texas. Everyone loves to say, oh, those crazy Texans 0:02:52.240 --> 0:02:54.960 with their crazy power market, but it's actually been functioning 0:02:55.520 --> 0:02:58.840 reasonably well lately in very challenging conditions, and I think 0:02:58.840 --> 0:03:00.840 that's a story that is worth telling well. 0:03:00.840 --> 0:03:02.520 And I bring up the fact that you haven't been 0:03:02.560 --> 0:03:04.520 on the show yet because I almost couldn't believe it 0:03:04.560 --> 0:03:06.799 when you said it, because you're one of my favorite 0:03:06.800 --> 0:03:09.040 people to talk to about some of the more interesting 0:03:09.080 --> 0:03:11.520 things that we are covering. So one of the reasons 0:03:11.560 --> 0:03:14.240 we decided now was the right time to delve in 0:03:14.320 --> 0:03:17.480 again into this very fascinating market had to do with 0:03:17.600 --> 0:03:22.240 this recent wave of extremely warm weather across parts of 0:03:22.280 --> 0:03:25.200 the US. But we're going to talk specifically about Texas. 0:03:25.280 --> 0:03:27.720 Can you explain what happened and kind of the knock 0:03:27.760 --> 0:03:29.160 on effects to the grid. 0:03:29.600 --> 0:03:33.760 Sure, So, in early June, the weather forecasts started really 0:03:33.800 --> 0:03:37.520 shooting up to I'm now in the US and I've 0:03:37.680 --> 0:03:41.880 completely converted to speaking in terms of fahrenheit. I'm a convert, so. 0:03:41.840 --> 0:03:44.160 I'm a convert to celsius. So now I'm gonna have 0:03:44.200 --> 0:03:46.400 to calculate it on my phone while you're talking. 0:03:46.720 --> 0:03:50.280 So basically, what I've learned is the reason fahrenheit works 0:03:50.360 --> 0:03:52.800 is that anything above a hundred is like super hot. 0:03:53.040 --> 0:03:57.040 And for about two months now, Texas has had temperatures 0:03:57.080 --> 0:03:59.360 that are above a hundred during the day, and so 0:03:59.480 --> 0:04:02.200 that is really when you're in the times when heat 0:04:02.240 --> 0:04:06.040 starts to affect infrastructure, it's it's not safe. 0:04:06.360 --> 0:04:07.800 So it's it's been very extreme. 0:04:07.880 --> 0:04:11.280 And to put this in context, in the US and 0:04:11.320 --> 0:04:13.960 a lot of other power markets around the world summertime 0:04:14.040 --> 0:04:16.040 and the sort of the hottest day of the year 0:04:16.279 --> 0:04:18.520 is what they all plan for because that's when air 0:04:18.520 --> 0:04:21.640 conditioning demand is at its highest, and so it's when 0:04:21.760 --> 0:04:25.880 electricity demand is at its highest. So the combination of 0:04:25.880 --> 0:04:29.360 a sustained period of heat and that has then led 0:04:29.400 --> 0:04:32.600 to record power demand means that URKOT has really been 0:04:32.640 --> 0:04:35.799 getting put to the test over the course of this summer. 0:04:36.279 --> 0:04:39.279 And you'd reference that our cuts are particularly interesting. But also, 0:04:39.360 --> 0:04:42.159 let's say unique power market, So can you explain what 0:04:42.360 --> 0:04:44.560 is it that makes it different from let's say, other 0:04:44.600 --> 0:04:45.880 states in the United States. 0:04:46.200 --> 0:04:49.159 There's a couple of things that make it different. One 0:04:49.200 --> 0:04:52.960 is just generally across the United States power markets, the 0:04:53.080 --> 0:04:56.280 regulatory structure is a lot more varied, and you can 0:04:56.320 --> 0:05:01.680 see a lot more of hybrids between regulated and liberalized markets. 0:05:01.920 --> 0:05:06.760 Actually URKO is the least regulated, which of course everyone says, huh, 0:05:06.839 --> 0:05:09.880 typical Texas, But actually being the least regulated means that 0:05:09.920 --> 0:05:13.520 it most resembles a European power market in that generation 0:05:13.760 --> 0:05:17.400 is competitive and retail is competitive, and actually having competitive 0:05:17.400 --> 0:05:20.240 retail is somewhat rare in the US. But also is 0:05:20.279 --> 0:05:23.039 that it is an energy only market, which means that 0:05:23.320 --> 0:05:26.440 all of the incentives for say, plants that just are 0:05:26.480 --> 0:05:28.600 there for that hottest day of the year are not 0:05:28.720 --> 0:05:31.320 through things like capacity markets like you see in other 0:05:31.560 --> 0:05:35.599 markets like PJM or in GB in the UK, but 0:05:35.720 --> 0:05:38.320 it's all through a mechanism where it just says, oh, 0:05:38.360 --> 0:05:41.159 on those days is designed to have prices go to 0:05:41.240 --> 0:05:43.960 really extreme levels so that those plants can make their 0:05:44.000 --> 0:05:46.120 money in the space of a couple of hours a year. 0:05:46.240 --> 0:05:48.919 That makes it viable for them to stay online. But 0:05:48.960 --> 0:05:51.400 that means that it's it's quite volatile, and it's quite 0:05:51.440 --> 0:05:53.839 People sometimes describe it as a casino because if that 0:05:54.000 --> 0:05:57.080 day where prices explode doesn't come, then those plants don't 0:05:57.120 --> 0:05:59.160 make enough money. And if you get more of those 0:05:59.240 --> 0:06:01.640 days than you need, then those plants make way more 0:06:01.640 --> 0:06:04.680 money than they need. And ultimately the cost is passed 0:06:04.720 --> 0:06:07.320 on to consumers via the competitive retailers. 0:06:07.360 --> 0:06:07.760 And then I. 0:06:07.720 --> 0:06:10.560 Suppose the competitive retailers have this challenge of how do 0:06:10.600 --> 0:06:13.280 we offer the best deal whilst managing this risk. 0:06:13.520 --> 0:06:15.880 So that's kind of why Texas is so interesting. 0:06:16.080 --> 0:06:18.839 I'll just add on the last point you had about 0:06:18.880 --> 0:06:22.800 the energy only market. The mechanism that Urcott has are 0:06:22.839 --> 0:06:26.080 these scarcity adders, and those are happening in real time. 0:06:26.279 --> 0:06:29.960 So when reserves drop a certain threshold, Urcott adds to 0:06:30.080 --> 0:06:34.159 the power price to incentivize more power plants to come online. 0:06:34.240 --> 0:06:36.839 I think a coworker compared it to Uber surge pricing, 0:06:37.000 --> 0:06:41.000 and I say, the other uniqueness of URCOT that people 0:06:41.120 --> 0:06:43.400 remember is that it's an island grid, so it's not 0:06:43.440 --> 0:06:47.160 connected to the eastern or western connection, so when supplies 0:06:47.520 --> 0:06:52.200 are low, they can't rely on imports from neighboring balancing authorities. 0:06:52.080 --> 0:06:55.760 No ability to buy or sell any access capacity or 0:06:55.800 --> 0:06:56.840 required capacity. 0:06:57.000 --> 0:06:58.360 Correct And just. 0:06:58.320 --> 0:07:01.600 To put it in context, Natalie, what does say the 0:07:01.640 --> 0:07:04.560 power price on a typical day and how high can 0:07:04.600 --> 0:07:06.440 it go when we have these extreme prices. 0:07:06.800 --> 0:07:11.440 I'd say probably between sixty e eighty dollars per megawat 0:07:11.840 --> 0:07:15.040 and a typical day. But on these extreme days, for example, 0:07:15.080 --> 0:07:18.200 on June twenty, we saw prices go to a close 0:07:18.200 --> 0:07:20.760 of five thousand dollars per megawatt, and just on July 0:07:20.920 --> 0:07:23.680 thirty one, we saw prices to twenty three hundred and 0:07:23.680 --> 0:07:24.440 fifty dollars. 0:07:24.880 --> 0:07:27.400 What does this mean for retail customers like you and me? 0:07:27.720 --> 0:07:30.960 That's a really interesting question because it is the job 0:07:31.000 --> 0:07:34.800 of the retailers to manage this risk and offer customers 0:07:35.280 --> 0:07:38.240 a deal that is palatable. So from a customer's point 0:07:38.280 --> 0:07:41.240 of view, they should just see that their energy provider 0:07:41.520 --> 0:07:43.400 offers them a deal for certain period of time and 0:07:43.440 --> 0:07:46.400 a price by magwa hour, and those energy providers have 0:07:46.520 --> 0:07:49.480 to go into the market and procure their energy and 0:07:49.680 --> 0:07:51.880 manage this risk, and you know, the deal they offer 0:07:51.920 --> 0:07:54.200 customers is kind of a reflection of can they make 0:07:54.240 --> 0:07:56.760 a slight profit of while managing this risk. The reason 0:07:56.760 --> 0:08:00.000 I say it's interesting is because everything was so unregulated 0:08:00.200 --> 0:08:03.720 in Texas and they had competitive retail, which is somewhat 0:08:03.800 --> 0:08:07.040 unique in the US, although it's kind of commonplace in 0:08:07.080 --> 0:08:09.200 Europe and men. There was a lot of innovation and 0:08:09.320 --> 0:08:12.920 one of the very innovative ideas was a retailer called 0:08:12.960 --> 0:08:15.760 Griddy who said, you know what, rather than just say 0:08:15.800 --> 0:08:19.400 offering customers are fixed tariff, why don't we say let's 0:08:19.520 --> 0:08:23.360 just give them exposure to wholesale power prices. I either 0:08:23.400 --> 0:08:26.280 hourly market will meet that using a smart meter and 0:08:26.360 --> 0:08:28.440 build them for it, and they'll just pay us a 0:08:28.440 --> 0:08:30.280 monthly fee. And you know, you run the numbers and 0:08:30.320 --> 0:08:32.440 this would say, like in twenty sixteen, they were saying this, 0:08:32.480 --> 0:08:34.280 you run the numbers on look, look at how wholesale 0:08:34.280 --> 0:08:36.680 prices have been. That'll end up working out cheaper for 0:08:36.800 --> 0:08:38.920 you if you're okay with taking on that risk. I 0:08:38.960 --> 0:08:41.200 guess we're going to get on to talking about winter 0:08:41.240 --> 0:08:43.920 Storm Ury in February twenty twenty one, when the lights 0:08:43.960 --> 0:08:47.160 went out but prices went to these high levels. That 0:08:47.240 --> 0:08:51.199 deal worked out extremely badly in that set of circumstances, 0:08:51.240 --> 0:08:54.480 and it was because no one anticipated that something like 0:08:54.640 --> 0:08:58.280 winter Storm Ury would happen, and so these customers half 0:08:58.320 --> 0:09:00.120 the time they didn't have any power, and then when 0:09:00.160 --> 0:09:02.720 the power was on, it was costing ten thousand dollars 0:09:02.760 --> 0:09:04.920 per megawat hour or something like that, and they were 0:09:05.040 --> 0:09:07.920 ending up with bills that were tens of thousands of 0:09:07.960 --> 0:09:10.480 dollars that they couldn't possibly pay. And there are a 0:09:10.520 --> 0:09:13.840 lot of defaulting and financial pain happening off the back 0:09:13.880 --> 0:09:16.440 of this. So, in answer to your question, retailers are 0:09:16.440 --> 0:09:18.959 supposed to be innovative, supposed to manage the risk, there 0:09:19.000 --> 0:09:21.200 have been some missteps along the way. 0:09:21.440 --> 0:09:24.120 We did a podcast back in twenty twenty one after 0:09:24.280 --> 0:09:27.640 Storm Mury which was this extreme cold front in Texas, 0:09:27.679 --> 0:09:31.680 which also did cause extreme prices and actually had some 0:09:31.840 --> 0:09:35.600 really terrible consequences. What do you think between then and now, 0:09:35.679 --> 0:09:38.400 between Storm Mury so extreme cold and now we're dealing 0:09:38.480 --> 0:09:42.120 with extreme heat, how has the grid innovated, how has 0:09:42.160 --> 0:09:44.440 it changed, how is it adapted? And how is the 0:09:44.440 --> 0:09:46.319 story different this time around? 0:09:46.920 --> 0:09:49.040 I think there are too answer to that. The first 0:09:49.040 --> 0:09:50.760 one is the retail one, if you want to tackle 0:09:50.800 --> 0:09:53.240 that time, and I can talk about Urkard broadly. 0:09:53.240 --> 0:09:56.080 Sure, I mean, I suppose on the retail side, I 0:09:56.120 --> 0:09:58.600 don't think. I mean, Nanati can correct me if I'm wrong. 0:09:58.679 --> 0:10:01.200 I don't think that there has been any the major 0:10:01.320 --> 0:10:05.240 reform apart from there's just a lot more caution. So 0:10:05.559 --> 0:10:07.959 I don't think anyone in their right mind is either 0:10:08.040 --> 0:10:12.040 signing up to or selling contracts that expose customers to 0:10:12.320 --> 0:10:16.199 hourly wholesale prices, because everyone saw how badly that went 0:10:16.240 --> 0:10:18.680 in twenty twenty one, so it's it's maybe just a 0:10:18.679 --> 0:10:21.520 little bit more conservative now, and perhaps that's not a 0:10:21.559 --> 0:10:24.360 bad thing. I know we talk a lot about retail innovation, 0:10:24.600 --> 0:10:28.079 and that's part of the rationale for having competitive retail 0:10:28.120 --> 0:10:31.480 in both Europe and Texas, but maybe sometimes too much 0:10:31.520 --> 0:10:34.840 innovation is too risky, and so my understanding is that 0:10:35.080 --> 0:10:37.120 it's maybe a little bit more conventional in the way 0:10:37.120 --> 0:10:40.360 that they are are hedging the risks and then billing customers. 0:10:40.720 --> 0:10:44.880 Urkat has introduced a few changes since Winter Storm Mury. 0:10:45.120 --> 0:10:48.480 The first is like short term market changes, the major 0:10:48.559 --> 0:10:51.160 one being during Winter Storm Muri, your price is went 0:10:51.200 --> 0:10:53.959 to nine thousand dollars per megawatt. That was the system 0:10:54.000 --> 0:10:56.960 cap for their scarcity adder. They reduced that to five 0:10:57.000 --> 0:11:00.000 thousand dollars per megawat, which is why we saw price 0:11:00.200 --> 0:11:02.880 is around that level this summer. The scarcity adder also 0:11:03.120 --> 0:11:07.880 kicks in at a higher reserve level, so it's more conservative. Basically, 0:11:07.920 --> 0:11:11.640 prices will increase at a higher level than before to 0:11:11.720 --> 0:11:14.160 try and get power plants to come online. They've also 0:11:14.200 --> 0:11:19.040 added more ancillary services, which is another reliability service that 0:11:19.200 --> 0:11:23.400 Urcott has. But they've done a whole study to improve 0:11:23.520 --> 0:11:27.080 long term market design. The major one that passed earlier 0:11:27.120 --> 0:11:30.920 this year is called a performance credit mechanism. It's a 0:11:30.960 --> 0:11:34.120 capacity like payment. So as we talked about before, other 0:11:34.160 --> 0:11:37.520 markets have capacity markets that procure enough supply over a 0:11:37.520 --> 0:11:40.400 certain number of years. Orcott has famously been an energy 0:11:40.440 --> 0:11:43.280 only market, but is trying to introduce this new payment 0:11:43.360 --> 0:11:47.000 that is more predictable and can be another revenue to 0:11:47.080 --> 0:11:50.080 power plants other than these extreme price fikes. 0:11:50.360 --> 0:11:52.760 See the other thing I would say, though, that has changed, 0:11:52.880 --> 0:11:55.480 and maybe you don't say Urcott has changed this. It's 0:11:55.559 --> 0:11:58.600 just something that has changed generally is that there's a 0:11:58.679 --> 0:12:02.239 huge amount of solo beein to the grid. So typically 0:12:02.480 --> 0:12:05.840 these price spikes, apart from February twenty twenty one, which 0:12:05.880 --> 0:12:10.360 was an exceptional set of circumstances, typically you're expecting these 0:12:10.440 --> 0:12:13.480 price extremes, this lottery as we describe it, to happen 0:12:13.559 --> 0:12:16.559 during the summer and when the temperature is high, i e. 0:12:16.760 --> 0:12:17.960 During the day all day. 0:12:18.320 --> 0:12:21.080 Now, there's so much solar being added and still being 0:12:21.120 --> 0:12:24.200 added in Texas, it's enough now that we're at a 0:12:24.240 --> 0:12:27.400 stage where for most of the day you can rely 0:12:27.679 --> 0:12:31.600 on solar generating enough to meet that high demand, and 0:12:31.800 --> 0:12:34.640 so these price spikes aren't going to happen and won't 0:12:34.640 --> 0:12:37.600 be necessary. And it's only now during a more limited 0:12:37.640 --> 0:12:41.120 time window that during the late afternoon when the solar 0:12:41.120 --> 0:12:43.560 generation is ramping down but it's still hot that you 0:12:43.720 --> 0:12:46.559 still have the possibility of these price extremes. And so 0:12:46.679 --> 0:12:51.439 while while solar doesn't completely eliminate this dynamic, it certainly 0:12:52.040 --> 0:12:55.439 reduced the probability of it and made it maybe a 0:12:55.480 --> 0:12:58.520 little bit more healthy and restricted, I suppose. 0:12:58.800 --> 0:13:02.160 So between this cold snap which happened in twenty twenty 0:13:02.200 --> 0:13:06.480 one which really impacted the gas pipelines and wind power, 0:13:06.679 --> 0:13:09.640 we now have this additional solar that's been coming online 0:13:09.640 --> 0:13:12.959 over the past several years that's really creating or evening 0:13:13.000 --> 0:13:16.680 out the access in summertime. But as you point out rightfully, 0:13:16.840 --> 0:13:19.720 these spikes do tend to happen in terms of demand 0:13:19.920 --> 0:13:22.480 in the afternoon till late afternoon, often as the sun 0:13:22.520 --> 0:13:25.199 is setting and people are heading home returning home after work. 0:13:25.320 --> 0:13:27.199 And my question really has to do with with all 0:13:27.240 --> 0:13:30.000 of this renewables roll out in Texas, are we also 0:13:30.120 --> 0:13:34.160 seeing storage or are we seeing grid flexibility being built 0:13:34.200 --> 0:13:36.440 in What are the mechanisms in place in order to 0:13:36.480 --> 0:13:39.800 meet that late afternoon peak energy demand. 0:13:40.120 --> 0:13:44.240 Solar actually generates during that late afternoon peak of six pm, 0:13:44.360 --> 0:13:47.360 but it's we are now seeing that peak net demand, 0:13:47.400 --> 0:13:50.840 which is demand minus renewables, happen when solar comes offline. 0:13:50.960 --> 0:13:55.240 Eight to nine pm, so that more predictable renewables waiting off, 0:13:55.400 --> 0:13:58.440 more solar waning off could cause price spikes that happened 0:13:58.520 --> 0:14:01.680 later in the day. That what we've typically seen, which 0:14:01.679 --> 0:14:04.240 has been driven by demand in people coming home. 0:14:04.640 --> 0:14:07.640 And to your question around storage data, that is creating 0:14:07.640 --> 0:14:10.400 an opportunity for storage. It's a very well defined time 0:14:10.440 --> 0:14:13.040 window where high prices are expected. You know, we've seen 0:14:13.080 --> 0:14:16.720 it in California that solar gets built and then storage follows. 0:14:17.040 --> 0:14:19.760 We're seeing the same happening in Texas. It's maybe at 0:14:19.800 --> 0:14:22.680 an earlier stage, but yeah, we are expecting in the 0:14:22.720 --> 0:14:25.640 next few years now a substantial amount of storage to 0:14:25.640 --> 0:14:28.480 get built in Texas to complement the solar. 0:14:30.560 --> 0:14:33.320 Another show that we previously did, which we probably should 0:14:33.360 --> 0:14:35.680 have also featured you in tom was one that was 0:14:35.720 --> 0:14:41.080 around servers and as a potential source of flexible capacity. 0:14:41.240 --> 0:14:44.800 And this brings me to another point around well, certainly 0:14:44.840 --> 0:14:47.360 something that puts a lot of demand on servers, which 0:14:47.400 --> 0:14:50.840 is bitcoin mining. There's an interesting connection here to Urka, 0:14:50.960 --> 0:14:52.840 and we'll come to that, but before we go there, 0:14:53.160 --> 0:14:56.200 talk a little bit about well, bitcoin mining and why 0:14:56.240 --> 0:14:59.640 is it so energy intensive and require so many servers? 0:15:00.560 --> 0:15:02.760 So that's a good question because in a way, I 0:15:02.880 --> 0:15:06.840 understand very well the impact that bitcoin mining has on 0:15:06.880 --> 0:15:10.240 the energy system, but I'm certainly not a bitcoin expert. 0:15:10.440 --> 0:15:14.080 But the way bitcoin works is that the system is 0:15:14.120 --> 0:15:18.520 built around that you create new bitcoins by solving complex 0:15:18.560 --> 0:15:22.520 computational problems, and you get rewarded with bitcoins for doing that, 0:15:22.640 --> 0:15:26.680 and you also, in doing so facilitate processing of transactions 0:15:26.800 --> 0:15:30.320 in the bitcoin market, and over time, the amount of 0:15:30.480 --> 0:15:34.480 energy required to create every bitcoin increases, the idea being 0:15:34.480 --> 0:15:37.720 that eventually all the bitcoins that are possible to be 0:15:37.760 --> 0:15:40.000 created will be created, but at that point in time, 0:15:40.040 --> 0:15:42.960 it will be really energy intensive to create a bitcoin. 0:15:43.240 --> 0:15:45.080 Now I know that when I say, oh, it's very 0:15:45.160 --> 0:15:47.920 energy intensive, everyone has a sort of a knee jerk 0:15:48.000 --> 0:15:51.960 reaction and says, that's so bad, that it's so energy intensive, 0:15:52.120 --> 0:15:54.960 And I actually just want to maybe temper that a 0:15:54.960 --> 0:15:58.160 little bit. If you have, say, servers that consume a 0:15:58.200 --> 0:16:01.880 megawatt of power, and it costs barely any energy to 0:16:02.000 --> 0:16:05.440 produce some bitcoins, then you're just going to produce more bitcoins. 0:16:05.720 --> 0:16:07.960 The fact that it's more energy intensive just means that 0:16:08.000 --> 0:16:10.240 you're still going to run your server the same amount, 0:16:10.360 --> 0:16:13.400 you'll just get less bitcoins from doing it, And actually 0:16:13.440 --> 0:16:15.960 that means you're more likely to switch off when power 0:16:16.000 --> 0:16:19.520 prices are too high, which is normally when fossil generation 0:16:19.640 --> 0:16:22.160 is happening. So Bitcoin gets a little bit of a 0:16:22.160 --> 0:16:24.640 bad rap for being so energy intensive, and I sometimes 0:16:24.680 --> 0:16:27.080 think that maybe misses the point a little bit. 0:16:27.400 --> 0:16:30.680 But there is this flexibility element just built in inherently 0:16:30.720 --> 0:16:32.160 to the way the market functions. 0:16:32.680 --> 0:16:35.880 Yeah, and you only get that flexibility element because it's 0:16:35.960 --> 0:16:36.800 energy intensive. 0:16:37.080 --> 0:16:40.000 Well, let's back up still one more, which is is 0:16:40.040 --> 0:16:43.640 bitcoin very common in Texas? Like what's the connection between 0:16:43.920 --> 0:16:45.080 er cut and bitcoin. 0:16:45.360 --> 0:16:47.680 I'm going to let Natalie give you the numbers, although 0:16:47.720 --> 0:16:50.320 I know what they are, but I will just anecdotally 0:16:50.360 --> 0:16:50.720 tell you. 0:16:51.920 --> 0:16:53.200 I'll just anecdote. 0:16:52.760 --> 0:16:55.240 Because I feel like she found out these numbers, but 0:16:55.240 --> 0:16:57.640 I'll anecdotally tell you when she was working on the 0:16:57.760 --> 0:17:01.000 Erko market outlook and she told me, oh, Tom, oh, 0:17:01.040 --> 0:17:04.320 bitcoin is a big factor in Texas, and I was thinking, oh, 0:17:04.320 --> 0:17:06.920 that's nice, a nice little side story, and she told 0:17:06.920 --> 0:17:10.800 me the numbers, and I was My first reaction was like, Natalie, 0:17:10.880 --> 0:17:13.080 I think you might have got your units mixed up 0:17:13.200 --> 0:17:16.560 because the numbers were so unbelievable. But Natalie tell us 0:17:16.560 --> 0:17:18.480 what those numbers are, and I will just say she 0:17:18.600 --> 0:17:20.240 hasn't got her units mixed up. 0:17:20.760 --> 0:17:24.480 Yeah, Urkott has about two gigawatts of bitcoin mining capacity 0:17:24.920 --> 0:17:27.560 online that has sprung up in the last year or two. 0:17:28.000 --> 0:17:31.359 Baseload demand in Urcott is forty five to fifty gigawatts, 0:17:31.400 --> 0:17:34.800 so that's not sensitive to temperature. So assuming bitcoin is 0:17:34.880 --> 0:17:38.520 running around the clock except during the stressful grid hours, 0:17:38.800 --> 0:17:42.040 that's about two to three percent of Urcott's basedload demand. 0:17:42.359 --> 0:17:44.800 Now, what does it look like going forward? And is 0:17:44.880 --> 0:17:47.960 Texas a growing market for bitcoin miners. 0:17:48.640 --> 0:17:53.000 Yes, Urcott has its own like large load Interconnection Q, 0:17:53.480 --> 0:17:57.280 similar to on the renewable side of Interconnection Q, except 0:17:57.320 --> 0:18:00.199 for these large loads, and it has about thirty seven 0:18:00.280 --> 0:18:03.760 gigawatts of these they're most likely duplicate projects and not 0:18:03.840 --> 0:18:06.880 all of them will come online, but we did here 0:18:07.240 --> 0:18:10.640 enough create a project database based on company announcements, and 0:18:10.880 --> 0:18:13.000 there could be five to six gigawatts in the next 0:18:13.080 --> 0:18:13.639 few years. 0:18:13.880 --> 0:18:16.720 Now, when you first looked at those numbers, Tom certainly 0:18:16.760 --> 0:18:18.880 told us what his initial reaction was when you brought 0:18:18.880 --> 0:18:21.520 them to him. But when you stumbled upon this you 0:18:21.520 --> 0:18:25.000 know or didn't stumble upon when you researched this particular 0:18:25.160 --> 0:18:28.720 heart of Texas's energy grid. What was your initial thought. 0:18:28.960 --> 0:18:31.439 Yeah, it was an interesting thing to track since there 0:18:31.480 --> 0:18:34.000 was a lot of news articles of bitcoin mining flocking 0:18:34.040 --> 0:18:37.480 to Texas and a lot of politicians welcoming them into 0:18:37.560 --> 0:18:40.720 their towns and cities. So we wanted a way to 0:18:40.800 --> 0:18:44.440 quantify how big of an impact this was, which is 0:18:44.480 --> 0:18:46.480 why we had to track in on a project a 0:18:46.560 --> 0:18:50.440 project basis and found those massive numbers. And the other 0:18:50.480 --> 0:18:53.080 thing that was hard to ignore was the amount of 0:18:53.080 --> 0:18:56.879 demand growth that happened last summer. So the previous demand 0:18:56.880 --> 0:19:01.720 record are caught was seventy five gigawatts set in twenty nineteen, 0:19:01.880 --> 0:19:04.880 and that that held until last summer, and it broke 0:19:04.920 --> 0:19:07.119 the demand record close to a dozen times and reached 0:19:07.160 --> 0:19:10.600 eighty gigawatts, So that's six gigawatts of peak demand growth. 0:19:10.640 --> 0:19:12.919 Is quite rare in other parts of the US to 0:19:13.000 --> 0:19:16.560 have so much load growth, so we wanted to try 0:19:16.600 --> 0:19:19.359 and measure how much of it was the bitcoin that 0:19:19.480 --> 0:19:21.640 was adding to this tremendous load growth. 0:19:21.920 --> 0:19:25.040 So we do know that servers generally do pose an 0:19:25.080 --> 0:19:28.800 opportunity for flexibility within an energy system, and I believe 0:19:28.880 --> 0:19:32.880 that you also came across some really interesting examples of 0:19:32.920 --> 0:19:36.400 where that has happened, specifically with bitcoin. Can you elaborate 0:19:36.440 --> 0:19:37.080 on some of those. 0:19:37.480 --> 0:19:40.320 Last July, there was another heat wave in Texas and 0:19:40.400 --> 0:19:43.960 prices also exploded, and based on quotes and what we 0:19:44.240 --> 0:19:47.560 modeled demand, we saw that around one hundred and fifty 0:19:47.760 --> 0:19:50.800 dollars per megawatt, based on what the bitcoin price was 0:19:50.800 --> 0:19:53.560 at at that time, was about the break even price 0:19:53.600 --> 0:19:57.040 of bitcoin. So if anything above it, it made economic 0:19:57.200 --> 0:20:01.040 sense for these minds to shut off. And that's broadly speaking, 0:20:01.119 --> 0:20:03.040 a lot of these minds have different efficiencies. 0:20:03.320 --> 0:20:04.760 Did many of them take advantage of it? 0:20:05.000 --> 0:20:05.200 Yeah? 0:20:05.240 --> 0:20:08.280 I mean I think they pretty universally took advantage of 0:20:08.320 --> 0:20:11.960 it because either they have demand response agreements in place, 0:20:11.960 --> 0:20:14.480 but if they're exposed to wholesale power prices, it makes 0:20:14.520 --> 0:20:17.399 no sense for them to mine bitcoin at a loss. 0:20:17.600 --> 0:20:20.240 And I suppose this kind of speaks to just why 0:20:20.280 --> 0:20:23.240 this is interesting to me, even if I'm not interested 0:20:23.280 --> 0:20:25.679 in bitcoin. Is one thing that makes bitcoin unique as 0:20:25.680 --> 0:20:28.600 a load, and this is different from other servers, is 0:20:28.640 --> 0:20:32.640