The Q1 20 South Korean supply-demand balance—currently forecast to be very loose— will tighten considerably if the energy ministry enacts a proposed plan to cut coal-fired capacity by up to 11 GW in December 2019-March 2020. We think the ministry will likely enact some restrictions—supporting gas demand—but the cuts are unlikely to be as restrictive as in the initial proposal. At most, we think coal plant restrictions of 7 GW in peak winter would be more realistic, which would increase gas demand by 0.7 Mt for every month there are restrictions. The high risk of at least some coal-fired capacity cuts this winter is supportive for the JKM-TTF spread.
South Korea posted another month of strong imports in August (2.67 Mt, +0.7 Mt higher y/y), pushing LNG storage levels to 4.2 Mt by month-end, a record y/y increase of 2 Mt, with the country certainly undertaking some restocking. LNG storage should reach maximum capacity by the end of November, leaving the country well supplied ahead of winter, especially given much higher scheduled nuclear availability (+4.7 GW y/y). But there is uncertainty around the likely volume of LNG imports this winter, given that South Korea’s National Council on Climate and Air Quality (NCCA) has recommended that the government restrict some 30% (11 GW) of the country’s coal-fired capacity (see E-mail alert: South Korean proposal to restrict coal-fired capacity could boost LNG imports by 3.1 Mt this winter; bullish for JKM-TTF spread, 13 September 2019).
Some officials in the Korean energy ministry are said to be concerned that that the proposed loss of some 11 GW of coal-fired capacity over peak winter would leave the power system unable to meet demand in some circumstances. We estimate that winter peak power demand in Korea is likely to be around 80 GW. The country has a 106 GW of firm generation capacity (excluding renewables), to give a robust demand peak reserve margin of 33%. The current outage schedule shows that an average of 3 GW of nuclear capacity will be offline for maintenance during the peak winter months. Another 11 GW of lost coal-fired capacity would reduce the peak reserve margin to around 15%. As power system operators tend to prefer that peak margins be no lower than 20%, the energy ministry’s resistance to the proposed halt of 11 GW of capacity is unsurprising. We expect that KEPCO and the power regulators will push for a level of reserve margin of no lower than 20%, and that suggests that the 7 GW is the highest level of coal plant outage that is likely to be accepted for peak winter. For any month that 7 GW of coal-fired plant is offline, that would add around 0.7 Mt of demand to South Korean LNG imports. Coal capacity restrictions of 7 GW in December 2019-March 2020 would therefore add 2.8 Mt of gas demand (provided the nuclear maintenance schedule does not change), bringing total winter LNG imports to 25.5 Mt, 0.8 Mt higher y/y, compared with 2.0 Mt lower y/y currently forecast.
|Fig 1: South Korean LNG stock movement, Mt||Fig 2: South Korean LNG imports, Mt|
|Source: JODI, Energy Aspects||Source: JODI, Energy Aspects|
Energy ministry officials reportedly agree that measures need to be taken to reduce fine dust from coal-fired plants, but say that the suggested number of plant closures in the NCCA plan is unnecessarily restrictive, given the government was able to successfully lower particulate emissions in spring 2019 with a much shorter duration of coal plant shutdowns. Coal-fired capacity was cut by 12.9 GW in May—2.2 GW more y/y and 1.9 GW more than NCCA’s planned March shutdown—but the May restrictions were only for 15 days. Some ministry officials are said to be concerned that the closures would result in heavy financial losses for KEPCO, which would have already hedged its winter coal requirements and would be required to unwind them in the market.
The NCCA is scheduled to meet and discuss its plan with the energy ministry this week. The ministry is expected to present an alternative plan to tackle fine dust emissions, which could provide more clarity on winter gas demand.