The market has long anticipated a y/y drop in summer 2019 Korean gas demand and LNG imports owing to higher nuclear capacity, but April surprised with a 0.19 Mt y/y import rise. The uptick likely holds some clues for what to expect during the rest of this summer. While Q2 19 is still scheduled to see the largest of 2019’s y/y nuclear capacity additions, it may be that coal-fired power takes the biggest hit from this in Q2 19, given mounting public pressure to improve air quality, and as such we have softened our expectations for gas demand losses this quarter and raised our forecast for LNG imports. Conversely, gas could be the bigger loser over coal in Q3 19, but the current nuclear schedule indicates that incremental nuclear generation will be light that quarter, meaning less thermal generation will be displaced. Across the summer quarters, we now expect LNG imports to drop by a lighter 0.37 Mt y/y.
Following a mild winter, April HDDs were up by 11% y/y and 14% above the 10-year average, boosting res com demand by 0.16 Mt (11%) y/y. The cold spell was brief, however, as May weather is on track to be the warmest in the past 10 years, eliminating the last of any early summer heating demand. For the rest of the summer, focus turns to the power sector. The Korean power market has been expecting that South Korean power sector gas demand will be down y/y in summer 2019, owing to higher nuclear capacity. But repeated revisions to the nuclear maintenance schedule, combined with new air quality restrictions and brisk renewable generation, mean that much uncertainty remains as to how severe the loss will be.
Certainly, y/y gas-into-power losses will be the largest in Q2 19, although we have revised them down from last month’s forecast owing to maintenance schedule revisions. Nuclear capacity rose by 6.2 GW y/y in April, delivering a 0.16 Mt (12%) reduction for gas into power. May and June nuclear capacity is now expected to rise by 4.9 GW and 3.3 GW, respectively, revised down from the 5.8 GW and 3.8 GW listed last month, but we expect that to result in a 2.4–3.6 TWh y/y rise in nuclear generation across those months. And if we can assume that the fairly steady 0.5 TWh y/y or so rise in renewable generation we saw across Q1 19 will stretch into summer, some 3–4 TWh of thermal generation should be displaced each month in May-June unless power demand rises significantly.
Which fuel takes the bigger hit?
The big question is: how is that thermal loss split between gas and coal? As noted above, gas will see losses in Q2 19, but we think that the impact of air quality regulation will mean that the brunt of the thermal power losses in the quarter will fall on coal.
As it did in 2018, South Korea suspended some of its older coal-fired plants this spring in order to improve air quality, with a total of four plants comprising 2 GW of capacity shut from March-June. This year, the government seems to be under even more pressure to raise air quality. In February 2019, the Special Act on Particulate Matter Reduction and Management, which allows local governments to impose restrictions on coal power plants, came into effect. Local municipalities are now obliged to take emergency dust reduction measures if the daily average level of ultra-fine dust (PM 2.5) exceeds 50 micrograms per cubic metre and is forecast to do so again the following day.
It is unclear how the Special Act differs from rules announced in May 2018 (see E-mail alert: Korea promises air quality restrictions on coal, bolstering LNG demand, 29 May 2018). Included in the May 2018 measure is that 42 of 68 coal/oil-fired power plants are to be limited to generating at 80% below their current levels when fine dust warnings are issued. Coal saw some minor losses (an average of 50 GWh) across summer 2018, but record high cooling demand and heavy nuclear outages meant that coal-fired output was a touch higher y/y in some months despite the new regulation.
But recent public pressure amid particularly high pollution levels in Q1 19 suggests to us that coal will take the primary hit in Q2 19 thermal output. Historically, particulate emissions have risen through winter, peaking in late Q1 and early Q2, the result of a combination of high fossil fuel demand paired with a temperature inversion layer that occurs on cold nights and traps air pollution closer to the ground. Although aggregate power demand often peaks in early Q3, PM levels are generally at their lowest in the second half of summer. This is because the Asian Dust or Yellow Sand—a weather phenomenon caused by high-speed winds blowing fine desert sands from northern China over Northeast Asia—eases through summer. Asian Dust blows all year round, but northern Asian is most affected during the spring time, particularly in March.
Last month, Seoul PM 2.5 measurements fell to the lowest level for the time of year since at least 2015. With roughly 5 TWh of incremental nuclear and renewable generation in April, both coal and gas will have taken a hit, but the sharp drop in PM 2.5 measurements, following especially high levels in Q1 19, could indicate that the Special Act restrictions are hurting coal more than gas. By June, when we expect around just 3 TWh of incremental nuclear and renewable generation, we could even see gas demand stay flat, or make small y/y gains, given that June 2018 weather was cooler than average. We have thus revised our forecast for Q2 19 aggregate gas demand to 8.8 Mt, 0.2 Mt less y/y, compared to a 0.6 Mt y/y loss previously. More aggressive restocking should push Q2 19 LNG imports to 9.6 Mt, 0.11 Mt less y/y.
We have revised down our Q3 19 LNG demand forecast as well, on the expectation that restrictions on coal-fired generation will ease in the second half of the summer. The only change to the Q3 19 maintenance schedule is in July, where nuclear capacity is now scheduled to be 1.8 GW lower y/y, compared to previous expectations of 2.2 GW less y/y. And given that PM levels generally trough in July, this provides an opportunity for coal-fired plants to pick up lost nuclear share instead of gas. We now expect Q3 19 aggregate gas demand to drop by 0.17 Mt y/y, and have revised our Q3 19 LNG imports down to just under 8.8 Mt, 0.25 Mt less y/y.
|Fig 1: Seoul PM 2.5 levels, pcm||Fig 2: Korean LNG stock movement, Mt|
|Source: Berkeley Earth, Energy Aspects||Source: JODI, Energy Aspects|