WEBVTT - Japan Powers Up: Market Liberalization Sparks Interest 0:00:00.160 --> 0:00:02.560 This is Tom Rowland's Reese and you're listening to Switched 0:00:02.600 --> 0:00:06.400 on the BNF podcast. Today we're discussing the Japanese power market, 0:00:06.600 --> 0:00:08.960 a market which, as one of our guests boldly claims, 0:00:09.039 --> 0:00:11.640 could be the most important power market in the world. 0:00:11.960 --> 0:00:15.200 Since the Fukushima disaster and its entire nuclear power fleet 0:00:15.240 --> 0:00:18.079 being shut down, the Japanese power market has continued to 0:00:18.120 --> 0:00:21.159 face our host of challenges, not only natural disasters, but 0:00:21.200 --> 0:00:24.160 the global energy crisis and more extreme periods of weather. 0:00:24.360 --> 0:00:27.159 To many, this may seem unattractive or too challenging, and 0:00:27.240 --> 0:00:29.800 yet the world's third largest liberalized power market has been 0:00:29.840 --> 0:00:33.400 attracting more attention from foreign investors than ever. While Japanese 0:00:33.400 --> 0:00:35.640 power demand has been in decline in recent years, we 0:00:35.680 --> 0:00:38.520 are now seeing this trend reverse with new data centers 0:00:38.560 --> 0:00:42.160 and semiconductor manufacturing plants connecting to the grid. With these 0:00:42.240 --> 0:00:45.840 comes the need for further power sources. Utility scale renewables 0:00:45.880 --> 0:00:47.840 such as wind and solar would be a natural fit 0:00:47.880 --> 0:00:49.800 for most in this situation, but as we are to 0:00:49.840 --> 0:00:52.960 discuss today, land constraints and local factors opposing their own 0:00:53.000 --> 0:00:55.920 set of challenges For these two to discuss, the Japanese 0:00:55.960 --> 0:00:58.320 power market further. Today, I'm joined by two of BNF's 0:00:58.360 --> 0:01:01.800 Global Power Market's Teamly it's Mariko O'Neil and Yumi Kim, 0:01:01.960 --> 0:01:04.920 to review findings from the report Japan Power Market Outlook 0:01:04.920 --> 0:01:08.320 too H twenty twenty four Tailwinds Ahead. BNF clientsill be 0:01:08.360 --> 0:01:11.360 able to access this and other related research at BNF 0:01:11.360 --> 0:01:14.160 go on the Bloomberg terminal and at BNF dot com. Now, 0:01:14.280 --> 0:01:16.840 let's talk to Marico and Num about the Japanese power market. 0:01:27.000 --> 0:01:29.960 So I'm joined here today by Mariko. Mariko, thank you 0:01:30.040 --> 0:01:31.000 for joining. 0:01:30.840 --> 0:01:33.000 Thanks for having me, and Yumi. 0:01:33.160 --> 0:01:34.600 Thanks a lot for inviting me too. 0:01:34.920 --> 0:01:38.480 So we're talking about the Japanese power market. First off, 0:01:38.720 --> 0:01:41.240 there's a lot of interest in it right now. It 0:01:41.280 --> 0:01:43.600 seems to be a hot topic, not just in Japan 0:01:43.800 --> 0:01:47.120 but around the power world. So why is it that 0:01:47.160 --> 0:01:48.760 everyone's talking about Japan right now? 0:01:49.040 --> 0:01:52.800 Yeah, good question. So this might be controversial, but I 0:01:52.800 --> 0:01:55.240 think there's an argument to be made that the Japanese 0:01:55.280 --> 0:01:58.640 power market might be the biggest power market in the 0:01:58.680 --> 0:02:02.760 world never And we can get into sort of like 0:02:02.800 --> 0:02:06.520 all the parameters of whether it is or it isn't later, 0:02:06.680 --> 0:02:09.960 But in terms of why now, Japan has been slowly 0:02:10.040 --> 0:02:13.360 but surely liberalizing its power market over the past couple 0:02:13.400 --> 0:02:17.000 of decades, and its efforts really accelerated after twenty eleven 0:02:17.120 --> 0:02:19.160 and when they had to remove nuclear capacity from the 0:02:19.200 --> 0:02:21.560 grid after Fukushima. So you could say that it was 0:02:21.600 --> 0:02:25.440 properly liberalized in twenty sixteen, and since then trading volumes 0:02:25.480 --> 0:02:28.160 have been climbing. And given that it's an island with 0:02:28.280 --> 0:02:32.880 no almost no natural resources and almost entirely dependent on 0:02:33.000 --> 0:02:35.919 fuel imports, there's been a lot of interest sort of 0:02:36.000 --> 0:02:38.679 across the commodities space on what's going on in the 0:02:38.760 --> 0:02:39.920 Japanese power system. 0:02:40.280 --> 0:02:42.960 So, actually, I mean on this topic, you both spoke 0:02:43.000 --> 0:02:45.840 at a seminar that we held in our Tokyo office 0:02:45.960 --> 0:02:49.320 quite recently after releasing the Japan Power Market Outlook, and 0:02:49.320 --> 0:02:51.519 it was part of Japan Power Week, which I presume 0:02:51.639 --> 0:02:53.600 is like a bigger thing where everyone is coming in. 0:02:53.880 --> 0:02:56.640 So how much of the people in the room there 0:02:56.919 --> 0:03:01.280 would you say were people who do their business in 0:03:01.360 --> 0:03:03.840 Japan versus global companies. 0:03:04.160 --> 0:03:07.200 Yeah, at the moment, it's quite a nice, almost straight 0:03:07.240 --> 0:03:09.440 down the middle, half and half split, i'd say, of 0:03:09.480 --> 0:03:13.720 market participants, with half of them being domestic Japanese players, 0:03:13.840 --> 0:03:16.919 and then the other half being from everywhere from Sydney 0:03:16.919 --> 0:03:21.560 to Singapore to Hamburg to Chicago. So that's been quite international. 0:03:21.919 --> 0:03:24.919 One of the things about Japan having liberalized its power 0:03:24.960 --> 0:03:28.560 market more recently is that those who've been involved in 0:03:28.720 --> 0:03:32.120 the US or European power markets are looking at it 0:03:32.160 --> 0:03:34.960 and thinking, I can play the same game or a 0:03:35.000 --> 0:03:37.240 similar game to what I did in the US or 0:03:37.240 --> 0:03:38.400 in Europe in Japan. 0:03:39.080 --> 0:03:42.520 So just before we started recording, I was talking about 0:03:42.520 --> 0:03:45.640 my cats. I adopted two kittens recently, and one of 0:03:45.680 --> 0:03:48.480 the things that my wife said, and she'd never had cats, 0:03:48.720 --> 0:03:51.320 was a surprise to her, is that if you didn't 0:03:51.400 --> 0:03:53.280 know cats, you'd think they're all the same, and when 0:03:53.280 --> 0:03:56.600 you have them, you realize they have distinct personalities. So 0:03:56.680 --> 0:04:01.000 liberalized power markets, you know, and you mentioned that people 0:04:01.000 --> 0:04:03.760 who are familiar with liberalised power markets see japan'ze here 0:04:03.800 --> 0:04:06.520 is a power market that is becoming more like what 0:04:06.560 --> 0:04:09.160 I'm used to. But just like a cat, the Japanese 0:04:09.160 --> 0:04:12.400 power market must have its own personality. So what are 0:04:12.440 --> 0:04:16.360 the characteristics that define the Japanese power market as distinct 0:04:16.440 --> 0:04:18.120 from the others that we're used to. 0:04:19.000 --> 0:04:22.200 One of the things might be like geography. So, as 0:04:22.240 --> 0:04:26.680 I mentioned, Japan is an island with no electrical interconnections 0:04:26.720 --> 0:04:29.440 to any other country, so in that sense, it's quite 0:04:29.520 --> 0:04:32.400 self contained. Those people who've been involved in Europe or 0:04:32.440 --> 0:04:34.160 the US, this might be a bit different to them, 0:04:34.200 --> 0:04:37.680 where there's a bit of interconnection between different markets. Another 0:04:37.720 --> 0:04:40.679 thing is that it's sort of like a very long 0:04:40.760 --> 0:04:43.880 and thin island, you could say, so in terms of 0:04:44.000 --> 0:04:47.360 climate and weather, you have subarctic regions like hock Guido 0:04:47.440 --> 0:04:50.000 in the north, all the way down to subtropical regions 0:04:50.040 --> 0:04:52.960 like Qushu in the south, with really different climates throughout 0:04:52.960 --> 0:04:56.880 the year, which obviously affects electricity consumption or what I 0:04:56.880 --> 0:04:59.479 guess we call power demand. And another thing actually that 0:04:59.600 --> 0:05:02.000 is a very particular quirk is that the grid is 0:05:02.080 --> 0:05:06.520 essentially split into two. So East Japan, which is composed 0:05:06.560 --> 0:05:09.279 of the region's hock Guide or Tohoko and Tokyo, is 0:05:09.320 --> 0:05:12.640 on fifty hertz, and then West Japan, composed of the 0:05:12.680 --> 0:05:16.040 six other power regions of Japan, is on sixty hurts, 0:05:16.080 --> 0:05:20.159 and there's very little transformer capacity between the two, which 0:05:20.240 --> 0:05:22.279 makes it sort of a tale of two cities. 0:05:22.320 --> 0:05:25.799 You could say, Yeah, I'd also like to add that 0:05:26.080 --> 0:05:29.440 the fuels market in Japan is quite unique because, as 0:05:29.520 --> 0:05:33.120 Marco said, Japan is quite important dependent. It doesn't have 0:05:33.200 --> 0:05:35.920 a lot of natural resources within that country. So when 0:05:35.920 --> 0:05:38.800 we look at the normal power mix in Japan, there's 0:05:38.880 --> 0:05:41.680 almost a fifty to fifty split between call and gas, 0:05:41.800 --> 0:05:46.200 and within those fuel technologies, there's quite complicated contract structure 0:05:46.279 --> 0:05:49.159 compared to markets like Europe and US. So that's another 0:05:49.200 --> 0:05:51.360 feature in Japan that's quite unique to the market. 0:05:51.600 --> 0:05:53.680 We're going to talk about that, that sort of the 0:05:53.680 --> 0:05:57.160 complexities of the spot versus contract markets, because I think 0:05:57.200 --> 0:05:59.479 that is something to understand the Japanese paner market you 0:05:59.520 --> 0:06:02.240 have to really understand as much as the power dynamics. 0:06:02.360 --> 0:06:06.040 But just generally on this point around fuels all being imported, 0:06:06.279 --> 0:06:11.800 does that translate to relatively expensive power in Japan, then, I'd. 0:06:11.600 --> 0:06:15.240 Say so to a certain extent. In Japan because as 0:06:15.240 --> 0:06:18.440 we highlighted, it has to import almost all of its fuel. 0:06:18.640 --> 0:06:22.000 That's particularly true for gas because gas prices tend to 0:06:22.040 --> 0:06:25.440 be more expensive than cool So unlike US, where it 0:06:25.480 --> 0:06:29.520 has a huge capacity of domestic gas procurement, like Japan 0:06:29.600 --> 0:06:32.200 has to really quite a lot on the spot procurement 0:06:32.240 --> 0:06:34.839 and also the long term contracts from other countries like 0:06:34.920 --> 0:06:37.760 Middle East and also US, so all of that costs 0:06:37.800 --> 0:06:40.800 is translated into higher power prices in Japan. 0:06:41.360 --> 0:06:44.880 So we have an island kind of divided into sort 0:06:44.880 --> 0:06:47.200 of two islands. We could say a fifty hertz and 0:06:47.240 --> 0:06:50.719 a sixty hertz island. Could we just explain quickly for listeners, 0:06:51.160 --> 0:06:55.560 why are there two different grids with different frequencies? 0:06:56.120 --> 0:06:59.040 They have a different frequency when it comes to grid 0:06:59.160 --> 0:07:03.320 because Japan's powermark used to be dominated by nine incombine utilities, 0:07:03.520 --> 0:07:07.640 so they used to have control over generation, transmission and 0:07:07.680 --> 0:07:11.200 operation and also the retail side. So because it was 0:07:11.400 --> 0:07:15.440 basically a monopoly of those nine incombineutilities, they had very 0:07:15.440 --> 0:07:18.480 little incentive to cooperate with each other because they kept 0:07:18.520 --> 0:07:22.800 control over specific power regions. But because Japan now faced 0:07:22.920 --> 0:07:25.840 the regulation of the power markets, now obviously they have 0:07:25.880 --> 0:07:29.440 to cooperate more. But unfortunately, because of this historical legacy 0:07:29.720 --> 0:07:33.640 of incombine utilities, the greed system is not as holistic 0:07:33.680 --> 0:07:34.480 as we want to be. 0:07:35.120 --> 0:07:38.280 So we have this thin island with very diverse weather, 0:07:38.720 --> 0:07:42.480 not a lot of its own natural resources, and relatively 0:07:42.520 --> 0:07:46.840 expensive imported fuel, and then we know that if those 0:07:46.880 --> 0:07:50.600 are the equations you're dealing with, maybe a logical thing 0:07:50.640 --> 0:07:54.320 would be nuclear. And obviously nuclear has had a rollercoaster 0:07:54.440 --> 0:07:57.560 of a story in Japan. Could you tell us where 0:07:57.920 --> 0:07:59.760 we are right now? Give it maybe a little bit 0:07:59.800 --> 0:08:01.560 of a background into the history of it, but could 0:08:01.560 --> 0:08:04.520 you tell us what the current dynamic sell when it 0:08:04.560 --> 0:08:05.480 relates to nuclear. 0:08:05.760 --> 0:08:08.960 Yeah, so funny that you asked this bit. So in 0:08:09.040 --> 0:08:12.520 twenty eleven, after the Mount Dan up for Kushuma, Japan 0:08:12.600 --> 0:08:15.560 turned off all of its nuclear reactors for power generation, 0:08:15.760 --> 0:08:18.680 so it went from having significant chunk of power generation 0:08:18.840 --> 0:08:22.440 being nucleared to none and thankfully, actually there were limited 0:08:22.480 --> 0:08:26.000 blackouts because they had pretty strong contingency plans for other 0:08:26.040 --> 0:08:28.560 sources of power. But it has meant that obviously they've 0:08:28.600 --> 0:08:32.480 substituted a lot of nuclear generation with gas and coal generation. 0:08:32.800 --> 0:08:35.240 And since twenty eleven there's been a big effort to 0:08:35.320 --> 0:08:38.319 restart some of the reactors. So they want to basically 0:08:38.480 --> 0:08:40.680 check all of the nuclear reactors to make sure they're 0:08:40.720 --> 0:08:43.280 safe to operate and in the event of anything like 0:08:43.320 --> 0:08:46.959 an earthquake or another tsunami, there is a sufficient backup 0:08:47.000 --> 0:08:50.760 plan in place to ensure everyone's safety, and they've restarted 0:08:50.760 --> 0:08:55.079 a handful, particularly in West Japan, so I think nuclear 0:08:55.120 --> 0:08:57.960 generation last year was just shy of ten percent of 0:08:58.000 --> 0:09:01.880 power generation in Japan. However, they have set quite ambitious 0:09:01.880 --> 0:09:06.280 targets for restarting the nuclear fleet, So because of its 0:09:06.400 --> 0:09:10.400 role as sort of cheap, marginal cost and low carbon 0:09:10.679 --> 0:09:13.720 source of power generation, Japan wants to have twenty to 0:09:13.760 --> 0:09:17.000 twenty two percent of its power generation coming from nuclear 0:09:17.000 --> 0:09:20.880 by twenty thirty. However, the process of restarting the reactors 0:09:20.920 --> 0:09:25.000 has been really slow and riddled with delays, and actually 0:09:25.120 --> 0:09:29.040 the first nuclear reactor in East Japan has restarted since 0:09:29.120 --> 0:09:31.160 twenty eleven, so if you think about that, that's taken 0:09:31.480 --> 0:09:34.960 over thirteen years to restart that reactor. And based on 0:09:35.000 --> 0:09:38.559 our recent analysis, actually even if Japan were to restart 0:09:38.679 --> 0:09:41.800 all of the reactors that have applied to its Nuclear 0:09:41.840 --> 0:09:46.040 Regulation Authority for approval to restart, it still wouldn't hit 0:09:46.160 --> 0:09:48.800 this twenty thirty target of having over a fifth of 0:09:48.800 --> 0:09:52.040 its power generation coming from nuclear. So it is kind 0:09:52.040 --> 0:09:55.200 of stuck in a tricky position because obviously, even if 0:09:55.280 --> 0:09:58.480 the government would like to emphasize nuclear as a source 0:09:58.520 --> 0:10:02.679 of clean and low cost power generation. It's extremely politically contentious, 0:10:02.800 --> 0:10:06.520 and there's very little control of an independent regulator deciding 0:10:06.520 --> 0:10:09.280 whether or not a reactor is safe to restart. So 0:10:09.559 --> 0:10:12.959 that's sort of the whole pickle with nuclear power. 0:10:13.120 --> 0:10:16.520 That's really interesting, And I would have just thought that 0:10:16.600 --> 0:10:20.800 in this scenario that the natural next question would be 0:10:20.960 --> 0:10:24.680 why aren't we seeing a huge increase in the buildout 0:10:24.720 --> 0:10:28.320 of renewables, whether it's wind or solar. Seems like a 0:10:28.440 --> 0:10:31.240 market primed for that, But they're not. I mean, I 0:10:31.240 --> 0:10:32.880 don't know what the numbers are. Am I right in 0:10:32.880 --> 0:10:35.120 saying they're not actually a huge part of the film mix? 0:10:35.800 --> 0:10:39.160 Yeah, you're right that for solar and wind products, it's 0:10:39.240 --> 0:10:42.480 taking quite a bit of time in Japan to be developed, 0:10:42.600 --> 0:10:46.000 and that's because of relatively high cost of development compared 0:10:46.040 --> 0:10:48.920 to other countries. And the main reason is that Japan 0:10:49.040 --> 0:10:51.680 there's quite a lot of mountains, so it's really difficult 0:10:51.840 --> 0:10:55.439 to build large scale solar products, for example, within its 0:10:55.520 --> 0:10:58.760 land and that's why it's been pretty expensive to generate 0:10:58.800 --> 0:11:02.880 from renewable sources. And that provides a reason why Japan's 0:11:03.040 --> 0:11:06.760 government wants source focus on nuclear products by restarting them, 0:11:06.840 --> 0:11:10.760 because it's less expensive compared to building new large scale 0:11:10.760 --> 0:11:13.640 solar and wind products in Japan. But even with those 0:11:14.160 --> 0:11:18.000 slower build out of solar and wind products in Japan, 0:11:18.080 --> 0:11:20.640 we're starting to see quite a lot of latility when 0:11:20.679 --> 0:11:23.600 it comes to power price dynamics because there's still quite 0:11:23.640 --> 0:11:26.240 a bit of solar products coming in in certain regions 0:11:26.320 --> 0:11:29.600 like qu in the down South, and that's causing some 0:11:29.840 --> 0:11:33.240 interesting power price dynamics, like we're seeing much lower prices 0:11:33.320 --> 0:11:36.680 during the midday when more solar products tends to generate 0:11:36.720 --> 0:11:37.760 electricity together. 0:11:38.160 --> 0:11:41.800 And I mean your point around the sort of the 0:11:42.120 --> 0:11:46.439 geography is taken, you know, the mountainous aspects, are we 0:11:46.559 --> 0:11:50.320 seeing significant uptake of rooftop solar. 0:11:50.600 --> 0:11:53.120 There's quite a lot of share of roof puff solar 0:11:53.200 --> 0:11:56.280 because of the same issue that we highlighted earlier, because 0:11:56.320 --> 0:11:59.640 of a lot of mountains hindering large scale utility solar. 0:11:59.800 --> 0:12:02.880 So we're seeing a lot of top solar products being 0:12:02.960 --> 0:12:06.040 built on top of residential properties, and that tends to 0:12:06.080 --> 0:12:09.280 take up a lot of share in Japan's solar product capacity. 0:12:09.720 --> 0:12:10.040 Got it. 0:12:10.080 --> 0:12:12.520 So if we're thinking solar in Japan, probably our starting 0:12:12.559 --> 0:12:16.600 point is rooftop solar, right interesting, and then we alluded 0:12:16.640 --> 0:12:19.560 to it earlier, but the sort of the situation when 0:12:19.559 --> 0:12:23.000 it comes to fuel is a little bit complicated. And 0:12:23.200 --> 0:12:26.199 this is a market dynamic that is maybe I'd be 0:12:26.200 --> 0:12:28.640 wrong set it's unique to Japan, but it's not something 0:12:28.679 --> 0:12:30.360 you see in every market, and you do see it 0:12:30.400 --> 0:12:32.439 in Japan. I know, you me, You've spent a lot 0:12:32.440 --> 0:12:35.400 of time trying to figure this out. But when we're 0:12:35.400 --> 0:12:38.320 thinking about power market dynamics, obviously a thing we always 0:12:38.320 --> 0:12:41.000 think about is what is the fuel price? And in 0:12:41.040 --> 0:12:43.280 a lot of markets as easy, you know, because whether 0:12:43.320 --> 0:12:47.320 it's gas or coal, there's pretty liquid market with prices 0:12:47.320 --> 0:12:49.520 that are constantly reported, and so it's very easy to 0:12:49.640 --> 0:12:51.880 sort of relate what's going on power markets to that. 0:12:52.080 --> 0:12:54.720 It's not quite so simple in Japan. So can you 0:12:54.760 --> 0:12:56.880 just explain a little bit why it's not quite so simple? 0:12:57.240 --> 0:12:57.640 Sure? 0:12:58.040 --> 0:13:00.880 I do think that the fuel's market and Japan is 0:13:01.360 --> 0:13:04.280 very unique, and sometimes it can be causing a lot 0:13:04.320 --> 0:13:07.520 of headache to the traders. It is quite complicated, and 0:13:07.559 --> 0:13:10.680 that's because it has made two ways of procurement. And 0:13:10.840 --> 0:13:12.920 first of all, the fuels can come from a spot 0:13:13.000 --> 0:13:17.680 market or the currently traded market where utilities can buy 0:13:17.880 --> 0:13:21.000 fuels on the spot or relatively for a short shorter 0:13:21.080 --> 0:13:24.680 period of time, and then there's another procurement method for 0:13:24.880 --> 0:13:29.360 long term contracts and for Japan's gas market, oftentimes those 0:13:29.400 --> 0:13:32.760 long term contracts are indexed to oil prices like Brent, 0:13:33.080 --> 0:13:36.120 and those deals can last up to almost fifteen to 0:13:36.200 --> 0:13:40.080 twenty years. When I started covering this sector, it didn't 0:13:40.080 --> 0:13:43.120 seem very obvious why long term contract guess would be 0:13:43.120 --> 0:13:45.920 indexed to oil, and that seemed very random to me. 0:13:46.120 --> 0:13:48.760 And actually there's quite a bit of long history behind 0:13:48.800 --> 0:13:52.200 this because the utility started indexing long term contract to 0:13:52.240 --> 0:13:55.760 oil back in eighties and nineties and that's when Japanese 0:13:55.840 --> 0:13:59.400 utility started inputting gas to using the power sector. And 0:13:59.640 --> 0:14:03.120 back then is used oil indexes to price that because 0:14:03.160 --> 0:14:05.840 they are more familiar with the oil pricing and that 0:14:06.000 --> 0:14:08.119 could give more stability in terms. 0:14:07.880 --> 0:14:09.000 Of price outlook. 0:14:09.280 --> 0:14:12.240 So fast forward, even today, a lot of the procurement 0:14:12.280 --> 0:14:15.360 are still based on this kind of long term contract 0:14:15.440 --> 0:14:18.120 gas prices index to oil and it can take up 0:14:18.200 --> 0:14:22.280 around almost ninety percent of Japan's gas procurement currently. Even 0:14:22.320 --> 0:14:25.400 for coal, it does have a quite complicated structure. It 0:14:25.800 --> 0:14:28.680 both has spot and contract market to it, But for 0:14:28.800 --> 0:14:32.960 coal it's less complicated because contract coal prices are index 0:14:33.040 --> 0:14:36.200 to spot core prices, so it's easier compared to gas 0:14:36.240 --> 0:14:37.040 market in Japan. 0:14:37.360 --> 0:14:39.600 I remember you were telling me the other day that 0:14:39.800 --> 0:14:44.160 the rules have changed recently about when say, if you're 0:14:44.200 --> 0:14:48.080 a gas generator you're placing your bids in the power market, 0:14:48.240 --> 0:14:52.440 what you should treat as your underlying fuel price. So 0:14:52.720 --> 0:14:55.200 what are those rule changes and what led to those 0:14:55.240 --> 0:14:56.360 rule changes happening? 0:14:56.600 --> 0:14:59.880 Sure, so when it comes to power prices, you can 0:15:00.040 --> 0:15:04.480 imagine that that's determined why utilities bidding into the wholesale 0:15:04.600 --> 0:15:07.840 power market. And that's the biggest driver for that is 0:15:07.880 --> 0:15:11.080 the marginal cost of generation or the fuel procurement costs 0:15:11.080 --> 0:15:14.440 for these utilities. And as I just explained, because there's 0:15:14.520 --> 0:15:17.760 a huge component of long term contract gas prices in 0:15:17.760 --> 0:15:21.720 gas procurement, you two ties used to calculate those beasts 0:15:21.720 --> 0:15:25.160 almost based on their long term contract gas prices. And 0:15:25.240 --> 0:15:27.480 then like in the recent few years, there has been 0:15:27.520 --> 0:15:30.280 more volatility in the spot gas market, and then the 0:15:30.360 --> 0:15:34.000 utilities realized that they could resell the cargoes that they 0:15:34.040 --> 0:15:36.400 procure from the long term contract market to the spot 0:15:36.440 --> 0:15:39.160 gas market when there's such an opportunity, and then that 0:15:39.280 --> 0:15:42.240 caused a huge spike in the power market because utwoties 0:15:42.280 --> 0:15:44.960 were more incentivized to sell gas rather than to use 0:15:44.960 --> 0:15:47.640 it to burn gas at their gas generation fleet. So 0:15:47.720 --> 0:15:51.440 then Japanese government decided to revise the guideline in terms 0:15:51.440 --> 0:15:55.600 of how the utilities calculate those bidding costs. And nowadays, 0:15:55.720 --> 0:15:59.120 after twenty twenty and twenty one, now the youtuities tend 0:15:59.160 --> 0:16:02.920 to reflect more spot gas prices in their beating costs, 0:16:02.960 --> 0:16:05.720 and to a certain extent that has raised the power 0:16:05.760 --> 0:16:09.120 prices in Japan. But also for you tities, now they 0:16:09.120 --> 0:16:12.480 don't have to resolve gas in the spot market when 0:16:12.480 --> 0:16:14.320 there's such a spike in the gas prices. 0:16:14.640 --> 0:16:17.240 This is interesting if we take a step back, what 0:16:17.280 --> 0:16:21.640 you've just described to me in totality seems like an 0:16:21.680 --> 0:16:26.200 example of a market going from this very regulated world 0:16:26.520 --> 0:16:30.360 of you know, regulated monopolies, bilateral contracts doing what is 0:16:30.440 --> 0:16:32.640 need to sort of keep the lights on, and then 0:16:32.760 --> 0:16:36.080 trying to transition towards a very liquid, liberalized market. And 0:16:36.320 --> 0:16:38.520 it seems like this sort of thing are kind of 0:16:38.560 --> 0:16:40.560 some of the I don't know whether you would want 0:16:40.600 --> 0:16:43.200 to call it growing pains that you hit along the 0:16:43.200 --> 0:16:45.520 way where suddenly there are things happening that seem very 0:16:45.560 --> 0:16:49.360 contradictory and complexity because you're going from one system that's 0:16:49.400 --> 0:16:51.760 very simple to another system that's very simple. But in 0:16:51.760 --> 0:16:54.280 the middle of that transition you have complexity. And that 0:16:54.320 --> 0:16:56.400 seems to me part of what's going on here in 0:16:56.440 --> 0:16:59.160 Japan right now. What makes it an interesting market is 0:16:59.160 --> 0:17:01.840 that it's not fully there yet in terms of being 0:17:01.840 --> 0:17:03.240 a liberalized power market. 0:17:03.440 --> 0:17:05.320 I think I think you're right, it's sort of a 0:17:05.359 --> 0:17:08.959 long road. And I would say, given sort of some 0:17:09.000 --> 0:17:13.040 of the geographic constraints that we've discussed, energy cost and 0:17:13.320 --> 0:17:17.360 energy security is obviously a big concern for regulators, and 0:17:17.640 --> 0:17:20.440 I think it makes them quite risk averse to move 0:17:20.480 --> 0:17:23.560 too quickly. Obviously, like the electricity grid is such a 0:17:23.600 --> 0:17:27.280 complicated thing, and you know, finding yourself at any moment 0:17:27.359 --> 0:17:30.480 a little bit short of power can really lead to disaster. 0:17:30.600 --> 0:17:32.880 And so, you know, I mentioned that it has been 0:17:32.880 --> 0:17:35.600 liberalizing for several decades, and we sort of assume that 0:17:35.640 --> 0:17:37.840 the reforms will, you know, continue. 0:17:38.160 --> 0:17:42.480 So, speaking of being worried about keeping the lights on, 0:17:42.760 --> 0:17:45.560 everywhere in the world, when we talk about power, there 0:17:45.600 --> 0:17:48.960 is always a question about Ai data centers, And I 0:17:49.000 --> 0:17:51.119 know this is something that you looked at in the 0:17:51.480 --> 0:17:54.119 recent Japan power market outlook. So we've talked about this 0:17:54.160 --> 0:17:58.119 really tight power market, you know, particularly with the historical 0:17:58.240 --> 0:18:00.359 retirements of nuclear and then it being very load to 0:18:00.359 --> 0:18:02.680 bring them back, and then on top of that, users 0:18:02.720 --> 0:18:04.600 to say, oh, there's a new source of demand in 0:18:04.640 --> 0:18:07.480 the form of data centers. Is that a going to 0:18:07.520 --> 0:18:10.320 be a major part of this story in Japan? 0:18:10.720 --> 0:18:14.119 Yeah, So I guess that there are a couple of 0:18:14.200 --> 0:18:18.840 components to this query about the recent boom in AI technologies. 0:18:18.960 --> 0:18:22.440 The main one is around data centers and basically the 0:18:22.520 --> 0:18:25.560 build out of these data centers that are really energy intensive, 0:18:25.720 --> 0:18:28.240 and to what extent is that going to strain the grid. 0:18:28.840 --> 0:18:32.280 Having looked at basically what's in the pipeline in Japan, 0:18:32.440 --> 0:18:35.000 it looks like most of the data centers that will 0:18:35.040 --> 0:18:37.280 be coming online in the next couple of years will 0:18:37.320 --> 0:18:40.800 be in areas that already have really high power demand, 0:18:40.920 --> 0:18:44.560 so very populous areas like Tokyo and Orsaka, which is 0:18:44.560 --> 0:18:47.440 in the Kansai area. And there's a question, I guess, 0:18:47.480 --> 0:18:49.440 when you're looking a little bit further out as to 0:18:49.480 --> 0:18:52.760 whether that will continue to be the trend. So having 0:18:52.960 --> 0:18:57.160 demand concentrated in certain areas is not ideal for Japan because, 0:18:57.359 --> 0:19:00.200 for example, the less populous areas of how I Do 0:19:00.320 --> 0:19:02.960 and Qsu have a lot of renewable generation. There's a 0:19:02.960 --> 0:19:05.280 lot of power supply there, but not a lot of demand. 0:19:05.320 --> 0:19:07.840 And then you know, when you get into Orsaka, Tokyo, 0:19:07.880 --> 0:19:09.719 you have a lot of power demand but not so 0:19:09.800 --> 0:19:13.560 much power supply, let alone clean power supply and new 0:19:13.600 --> 0:19:17.560 AI technologies like training large language models, doesn't have as 0:19:17.600 --> 0:19:20.360 much of a need to be close to the end 0:19:20.480 --> 0:19:22.919 user so close to big cities. So that's sort of 0:19:22.920 --> 0:19:24.760 an open question, I guess as we look a little 0:19:24.800 --> 0:19:26.879 bit longer term. But the other facet of it in 0:19:26.960 --> 0:19:29.840 Japan is that the government has big ambitions to make 0:19:30.000 --> 0:19:32.040 Japan a bit of a powerhouse in terms of the 0:19:32.080 --> 0:19:36.600 supply chain and in particular for semiconductor manufacturing, so it 0:19:36.640 --> 0:19:40.360 has quite generous subsidies for companies looking to open semiconductor 0:19:40.440 --> 0:19:43.640 fabs or fabrication plants we call them. And the tricky 0:19:43.720 --> 0:19:45.720 thing with those, I guess, is that a lot of 0:19:45.760 --> 0:19:48.639 these technology companies, be it for the data centers or 0:19:48.680 --> 0:19:52.879 the semiconductor manufacturing, have quite ambitious climate targets as well, 0:19:53.119 --> 0:19:56.679 even though their processes really require around the clock power. 0:19:57.000 --> 0:19:59.240 And so this is kind of something that has also 0:19:59.320 --> 0:20:03.639 been reopened the nuclear question in Japan because obviously wind 0:20:03.680 --> 0:20:06.800 and solar is intermittent and only generates power when the 0:20:06.840 --> 0:20:09.200 sun is shining or the wind is blowing. But there 0:20:09.240 --> 0:20:13.280 are lots of market entrants, particularly given recent geopolitical risks, 0:20:13.520 --> 0:20:16.520 that are looking to open facilities in Japan that need 0:20:16.560 --> 0:20:18.639 to consume a lot of power out of twenty four 0:20:18.680 --> 0:20:22.639 to seven around the clock basis, while also meeting company 0:20:22.640 --> 0:20:25.679 wide targets to have that be carbon free. And so 0:20:25.920 --> 0:20:29.000 there's been a lot of attention on nuclear as that 0:20:29.119 --> 0:20:32.080 low carbon, around the clock source of power. But also 0:20:32.440 --> 0:20:35.480 the government I suppose, has been trying to promote some 0:20:35.520 --> 0:20:38.480 of these facilities to be in those areas like cock 0:20:38.480 --> 0:20:40.919 Guide or Qushu that have a lot of renewable power. 0:20:41.200 --> 0:20:44.280 Water is also an important question. So yeah, those are 0:20:44.280 --> 0:20:46.399 some of I guess the variables at play when it 0:20:46.400 --> 0:20:48.760 comes to AI. But in the sort of like next 0:20:48.760 --> 0:20:51.320 couple of years, we looked at the data centers that 0:20:51.440 --> 0:20:54.879 will probably come online, and I think in terms of prices, 0:20:55.080 --> 0:20:57.760 it won't really do that much. There's not a huge 0:20:57.800 --> 0:20:59.960 amount I think in the pipeline, and some of us 0:21:00.080 --> 0:21:03.560 scenarios are seeing like a plus minus one percent I 0:21:03.600 --> 0:21:06.560 think change in power prices over the coming couple of years. 0:21:06.800 --> 0:21:09.360