There will be few surprises for summer 2019 Japanese gas demand unless there are substantial changes to this summer’s nuclear maintenance schedule. Our balance indicates that end-April stocks were broadly seasonally normal. Restocking should push LNG imports up by a moderate 0.9 Mt y/y this summer. The looming uncertainty for Japan is LNG demand in 2020, given the Nuclear Regulation Authority’s (NRA) refusal in late April 2019 to grant safety compliance extensions to 3.6 GW of nuclear plant at risk of closure next year. The market will need to closely watch for any updates to the NRA decision, but for now we have a base case scenario that assumes a shutdown of three months for each plant. That would boost thermal power demand, resulting in a 2.4 Mt y/y rise in LNG demand across 2020, a substantial revision from the 1.2 Mt y/y rise we had previously forecast.
Japanese LNG imports in April were a touch higher y/y at 5.6 Mt (+0.02 Mt), although our balance suggests there was still a small stockdraw, as colder weather y/y gave res-com demand a boost. Our balance also indicates that end-April stocks were broadly seasonally normal, less than 0.1 Mt higher y/y at around 4.08 Mt.
The variables that will affect Japanese gas demand for the rest of 2019 are fairly clear. Largely speaking, the large drops in power sector gas demand will moderate in June-October, as the impact of 2018’s March-June ramp-up in nuclear capacity eases. The two main factors that could lead to small adjustments to Japanese gas demand are moderate changes to the nuclear maintenance schedule and the weather during peak summer, both factors that the market is already accustomed to closely following. This month, our forecast for gas demand through 2019 is little changed from our previous forecast.
However, our forecast for 2020 has shifted substantially—we now expect LNG demand to rise by 2.4 Mt y/y, an upward revision from the 1.3 Mt we had forecast last month. This is based on our assumption of substantial risk that the NRA will shut down reactors next year that have not complied with necessary counter-terrorism measures (see E-mail alert: Japanese nuclear reactors could be shut again on compliance deadline, providing upside to 2020 LNG demand, 24 April 2019).
How big is the 2020 bull?
It is important to emphasise that there is still a significant lack of clarity regarding these potential 2020 nuclear shutdowns. What we do know is that in late April 2019, the NRA refused to extend a deadline for utilities to build anti-terrorism measures. Four plants totalling 3.6 GW are at risk of closure in the financial year starting March 2020. Kyushu Electric Power’s 0.89 GW Sendai unit 1 is the plant with the first deadline, of 17 March 2020, followed by the 0.89 GW Sendai unit 2 with a deadline of 21 May 2020. Kansai’s Takahama units 3 and 4 (0.87 GW each) will need to comply by August 2020 and October 2020, respectively.
The retrofits require plants to have separate buildings that allow units to be remotely operated in the event of a terrorist attack. Surprisingly, there has been no new information or updates since the NRA decision last month to deny utilities an extension to the compliance deadline. Utility Shikoku called the decision ‘severe’ and said that the company would attempt to shorten the necessary construction time to meet the deadline. But it is unclear how quick construction will be, although it should be noted that both Kyushu and Kansai sought to postpone the five-year deadline by one to three years, citing the need to carry out major construction works.
Given there is no present indication the NRA will provide utilities with an extension, we have settled on a base case scenario assuming that the four plants will be shut down for three months each. Aggregate capacity losses would peak in August and October at 1.5 GW and 1.7 GW respectively. If all of the lost nuclear generation is replaced by gas-fired output, power sector gas demand would rise in a range of 0.01-0.16 Mt y/y each month in March–October, with demand peaking in August. We would still expect power sector gas demand to fall y/y in November–December, owing to the planned return of the 0.79 GW Shimane unit 2 in October 2020 and the 0.78 GW Takahama unit 1 in November 2020. We would expect to see 2020 LNG imports rise by a total of 2.3 Mt y/y. If all shutdowns are averted, our current balances show that Japanese LNG demand will total about 83.2 Mt, 1.7 Mt more y/y.
A six-month shutdown of each plant would take even more capacity offline during the late summer peak cooling month—2.6 GW in August and 2.3 GW in September—and would have a much deeper impact in Q4 20, as the start-up of the 0.79 GW Shimane 2 and 0.78 GW Takahama 1 would be offset by the shutdown of Takahama units 3 and 4 (each 0.87 GW). Our balances show a 2.8 Mt y/y rise in 2020 LNG demand, with a 0.3 Mt y/y/ rise in Q3 20 alone, compared to a 0.06 Mt y/y rise in Q3 20 under a three-month shutdown scenario.
A 12-month shutdown of each plant would push offline capacity in August and September to a peak of nearly 2.7 GW, and Q4 20 impacted capacity would rise to 1.95 GW. If all lost nuclear generation was replaced with gas, it would push power sector gas demand up by a range of 0.4-0.28 Mt each month across March–December, boosting incremental LNG demand in 2020 to 3.1 Mt.
|Fig 1: Japanese reactors at risk of shutdown||Fig 2: 2020 LNG imports forecast, Mt, y/y|
|Source: Energy Aspects||Source: METI, Energy Aspects|