South Korean LNG imports in July rose by 0.28 Mt (10%) y/y to 3.03 Mt, boosting already high LNG storage. While recent changes to the nuclear maintenance schedule will provide an opening for thermal generation in the power sector in August–September, we expect coal to fill the void rather than gas. We expect the weakness seen in power sector gas demand through H1 19 will intensify as we move into winter 2019-20, putting LNG storage on course to reach a record high by November. We have revised up our Q3 19 LNG import forecast to 8.9 Mt (0.13 Mt lower y/y) from 8.6 Mt last month, owing to strong July and August imports, but we expect that LNG imports in Q4 19 will drop by 0.9 Mt y/y (7%) as storage approaches its limit and demand in the power sector remains weak.
A combination of strong July LNG imports and weak power sector gas demand boosted end-July South Korean LNG stocks to 3.37 Mt (+0.05 Mt y/y) according to our balance. With August LNG imports on course to post a 0.2 Mt y/y rise, and power sector gas demand set to remain weak, it looks as though South Korea will finish this season without posting a summer stockdraw, putting storage on course to be just shy of capacity by the end of October.
As South Korea is now well into the peak summer period, seasonal maintenance and air-quality restrictions on coal-fired plants have likely eased, raising available coal-fired capacity. Indeed, Kepco said this month that “preventative maintenance” for the spring season ended in June. This means that while recent upward revisions to the nuclear maintenance schedule (there will be 0.5 GW less capacity available in y/y terms across August and September, while 1.2 GW more capacity y/y had previously been expected) should curb late summer nuclear generation, coal is more likely than gas to fill the generation gap. We have previously detailed how new air-quality control rules would limit Q2 19 gas-fired generation losses (see South Korea Monthly, 15 May 2019) relative to coal, with the dynamic reversing in Q3 19. Historically, average ultra-fine dust (PM 2.5) levels peak in March and early summer but ease sharply by June, which would have allowed generators to bring air-restricted coal capacity back online by late Q2 19.
|Fig 1: Korea power sector gas demand, TWh||Fig 2: Korean LNG stock movement, Mt|
|Source: KEPCO, Energy Aspects||Source: JODI, Energy Aspects|
Indeed, compared to heavy coal-fired losses over April–May, Kepco power generation data show coal-fired generation losses to be minimal in June, and this should continue into winter 2019-20, especially now that the new 1.4 GW Shin Kori unit 4 has finally started up this month. Strong Q2 19 nuclear generation (+10.4 TWh y/y) because of record-high y/y nuclear capacity growth coincided with a heavy drop in thermal generation. But where coal-fired generation bore the brunt of early Q2 19 thermal generation losses, dropping by a total of 5.3 TWh y/y in April-May, the trend reversed in June, with gas-fired output falling by nearly 3 TWh (24% y/y) and coal generation easing by just 1% (0.2 TWh) y/y.
While power generation data for July are not yet available, gas-fired output now taking the brunt of thermal losses instead of coal would explain why July power sector gas demand dropped by 0.15 Mt (-13%) y/y, despite the heaviest y/y nuclear capacity losses (1.7 GW) in over a year. And given that renewable generation growth has been broadly steady at around 0.5 TWh y/y over the past six months, we expect total power sector gas demand to come in around 0.5 Mt y/y lower across Q3 19. These fundamentals put LNG storage on course to reach 4.8 Mt by end-October, just shy of capacity, painting a very bearish picture for Q4 19 LNG import demand—we see a drop of 0.9 Mt y/y (7%) over the quarter.