Time-bound emergency coal use without emissions lock-in: Japan and South Korea as precedents, Türkiye as application, March–September 2026

ABSTRACT

On 10 September 2026 the IEA projected that global coal demand would rise by 1.2% in 2026 to a record 8.94 billion tonnes, reversing a pre-war forecast of slight decline, as Europe, Japan, Korea and China switched from gas to coal. This brief, addressed to Türkiye’s Ministry of Energy and Natural Resources, the Ministry of Environment, Urbanisation and Climate Change and EMRA, argues that the long-term emissions cost of emergency coal depends on the form of the measure more than on the tonnage burned. Comparing Japanese, Korean, Chinese and German measures through Winzer’s narrow supply-continuity definition and Cherp and Jewell’s vulnerability approach, it recommends four rules: sunset clauses tied to LNG triggers, sizing to the buffer gap, pairing with low-emission capacity decisions, and no long-life fossil approvals under emergency procedures.

1. The problem: a rational short-term choice with a variable long-term price

According to the IEA’s mid-year coal update of 10 September 2026, global coal demand will rise by 1.2% in 2026 to 8.94 billion tonnes, with gas-to-coal switching in Europe, Japan, Korea and China as the main source of the increase; before the war the agency expected a slight decline (International Energy Agency [IEA], 2026a). The reversal reflects the restriction of LNG supply through Hormuz. The IEA expects combined LNG exports from Qatar and the UAE to fall by about 45% in 2026, while the Asian Platts JKM averaged $17.5/MBtu in the second quarter, 45% higher year on year (IEA, 2026b). Japan, China and South Korea cut LNG imports by a combined 10 million tonnes (Rystad Energy, 2026).

Current policy among importers treats emergency coal as a single decision: burn more or burn less. The evidence from March–September 2026 supports a different framing. Time-bound regulatory exemptions can be withdrawn when the crisis ends without any new investment decision; coal use that settles in through market substitution, or crisis decisions that commit capital to fossil capacity, tend to persist. The recommendations below apply this distinction to Türkiye, for which the sources reviewed record no emergency coal measure, so the form of any future measure remains open.

2. Two definitions, two accounts

Winzer (2012) confines energy security to risks to supply continuity and keeps economic efficiency and environmental sustainability outside the concept. The narrow definition makes the trade-off visible: emergency coal raises supply security in the narrow sense, and the emissions cost is booked to a separate policy objective. Cherp and Jewell (2014) define energy security as the low vulnerability of vital energy systems and argue that the “four As” classification fails to identify where vulnerability originates. Their approach requires naming the vulnerability a coal measure reduces. In 2026 that vulnerability was dependence on imported LNG passing through Hormuz; storable coal from multiple origins replaces it with a more dispersed supply structure. Read together, the two frameworks sort measures by which vulnerability they reduce and whether they can be reversed.

3. What the precedents show

Japan’s Ministry of Economy, Trade and Industry temporarily lifted capacity penalties on old coal plants with efficiency below 42%, targeting about 500,000 tonnes of LNG savings a year, against domestic LNG stocks of about 4 million tonnes (Daiss, 2026). The saving equals one eighth of the stock (0.5/4 = 0.125): a limited, temporary exemption designed to stretch a buffer. On 26 August Japan also announced plans to rebuild up to 5 reactors by the 2040s and up to 14 by the 2050s, and requested ¥1.9838 trillion for energy and GX spending in fiscal 2027 (Seo, 2026).

South Korea lifted coal generation caps in March 2026, raised nuclear utilisation to up to 80% and introduced its first gas price cap in 30 years; Qatar supplied 15% of Korea’s LNG imports in 2025 (Ramage, 2026). These data date from March, and the sources contain no verified new measures for August–September. The source does not define the Korean cap removal as time-bound; it forms part of an emergency package.

China shows the market channel. Thermal, mainly coal-fired, generation rose 3.4% year on year to 2.53 trillion kWh in the first five months of 2026, and coal’s share of generation, which had fallen to 51.4% in 2025, is rising again (Reuters, 2026). The IEA adds that high oil prices created extra coal demand in China’s chemical sector (IEA, 2026a). Reversal in China depends on price signals, with no regulatory decision to withdraw.

Germany shows lock-in through investment. Chancellor Friedrich Merz called the nuclear exit “a serious strategic mistake”, while the government treats it as irreversible and plans 12 GW of mainly gas-fired flexible capacity not expected before 2031 (Bielamowicz, 2026). The persistence risk there sits in gas, not coal. In Europe, summer heat also cut nuclear and hydro output and drought lowered river levels (Sommer, 2026), limiting low-emission flexibility in 2026.

4. Recommendation 1: Write a sunset into every emergency coal measure

Addressee: Ministry of Energy and Natural Resources and EMRA. Rationale: the Japanese exemption is reversible because it is a temporary regulatory act. A Turkish measure should lapse automatically on a stated date or on an observable LNG trigger, whichever comes first. The IEA’s own conditional statement supplies the trigger logic: coal demand may fall in 2027 if LNG flows normalise and stay high if disruption continues (IEA, 2026a). Sequencing: draft the sunset together with the measure, before any winter activation. Trade-off: an automatic lapse may end relief while prices remain high. Indicator: the measure lapses on schedule and coal dispatch returns to its pre-measure level.

5. Recommendation 2: Size the exemption to the buffer gap

Addressee: Ministry of Energy and Natural Resources. Rationale: Japan’s measure targeted a defined LNG saving relative to a known stock. Sizing to the gap between stocks and expected shortfall keeps the coal increment proportionate. Sequencing: publish the targeted LNG saving in bcm or tonnes at activation. Trade-off: a narrow measure may prove insufficient in a harsh winter and require a second decision. Indicator: LNG saved per month against the target.

6. Recommendation 3: Pair emergency coal with a low-emission capacity decision

Addressee: Ministry of Energy and Natural Resources. Rationale: Japan matched coal relief with a reactor-rebuilding plan and Korea with higher utilisation of existing nuclear plants; in Japan’s case the crisis leaves a low-emission legacy alongside the temporary coal increment. For Türkiye the matching decision already exists: first electricity at Akkuyu is targeted by the end of 2026 (NTV, 2026), a target not yet verified as met. Sequencing: link the end of any coal exemption to Akkuyu’s grid connection. Trade-off: tying the sunset to a project date imports project delay risk into the coal measure. Indicator: Akkuyu’s first-electricity date.

7. Recommendation 4: Approve no long-life fossil capacity under emergency procedures

Addressee: Ministry of Energy and Natural Resources, with the Ministry of Environment, Urbanisation and Climate Change. Rationale: Germany shows that the durable cost of the crisis can arise from new capacity with decades of life, regardless of fuel. Emergency procedures shorten scrutiny and should therefore be restricted to operating rules for existing plants. Sequencing: state this restriction before winter. Trade-off: some investment that would also suit normal conditions will be delayed until standard review. Indicator: the number of new fossil capacity approvals issued under emergency procedures, with a target of zero.

8. Implementation risks

Three risks follow from the evidence. Exposure is misallocated: according to Ember’s June 2026 analysis, the Hormuz crisis could add about $14 billion to Türkiye’s 2026 energy bill, of which $7.7 billion oil, $6.4 billion gas and $95 million coal (Gümüş, 2026), so coal measures address only the gas-to-power part of the cost. Measures without explicit sunsets have no built-in end point; the Korean cap removal is not defined as time-bound in the source. And the evidence base is incomplete: the sources contain no 2026 country-level emissions data, so the climate cost is assessed through coal demand and measure form.

9. Outlook: Türkiye, Saudi Arabia and the World

For Türkiye, November brings COP31 in Antalya (Tekin et al., 2026), which places the trade-off between crisis coal measures and emissions targets on an agenda Ankara will chair. A sunset-based design would give the host a defensible position if winter forces gas-to-coal switching.

For Saudi Arabia, the relevant channel is Asian oil demand. The IEA finding that high oil prices created coal demand in China’s chemical sector (IEA, 2026a) implies that, as long as high prices persist, part of oil-based feedstock in Asia can be replaced by coal, making the Asian demand base for Saudi crude more price-sensitive.

Globally, the IEA’s World Energy Outlook 2026 is due in November with several scenarios (IEA, 2026c). If the Hormuz restriction persists through winter, coal demand will stay high in 2027 on the IEA’s reasoning; if LNG flows normalise, the withdrawal of Japan’s temporary exemption and Korea’s coal-cap relief will be the first test of whether the short-term switch is reversed.

10. Conclusion

Emergency coal in 2026 was a rational way to reduce LNG vulnerability tied to Hormuz, and it conflicts with emissions targets. The size of the long-term cost depends on design. Japan bounded its coal relief with a temporary regulatory exemption, and Japan and Korea matched coal measures with nuclear measures, a rebuilding plan in Japan and higher utilisation in Korea; China’s increase rests on market substitution; Germany’s persistence risk sits in a new gas capacity decision. For Türkiye, the four rules proposed here keep any emergency coal measure reversible: a sunset tied to an LNG trigger, sizing to the buffer gap, a link to Akkuyu, and no long-life fossil approvals under emergency procedures.

SOURCES

Bielamowicz, J. (2026, 6 April). Iran war shock: Germany regrets nuclear exit, eyes coal. Defence24. https://defence24.com/other/iran-war-shock-germany-regrets-nuclear-exit-eyes-coal

Cherp, A., & Jewell, J. (2014). The concept of energy security: Beyond the four As. Energy Policy, 75, 415–421. https://doi.org/10.1016/j.enpol.2014.09.005

Daiss, T. (2026, 22 July). Japan’s emergency energy moves threaten its decarbonisation goals. Gas Outlook. https://gasoutlook.com/analysis/japans-emergency-energy-moves-threaten-its-decarbonisation-goals/

Gümüş, B. S. (2026, 12 June). Hürmüz krizi Türkiye’ye 14 milyar dolara mal olacak [The Hormuz crisis will cost Türkiye $14 billion]. Ember. https://ember-energy.org/app/uploads/2026/06/TR_Hurmuz-krizi-Turkiyeye-14-milyar-dolara-mal-olacak-1.pdf

International Energy Agency. (2026a, 10 September). Global coal demand set to increase this year amid Middle East conflict. https://www.iea.org/news/global-coal-demand-set-to-increase-this-year-amid-middle-east-conflict

International Energy Agency. (2026b). Gas market report, Q3-2026 – Executive summary. https://www.iea.org/reports/gas-market-report-q3-2026/executive-summary

International Energy Agency. (2026c). World Energy Outlook 2026 [Event page]. https://www.iea.org/events/world-energy-outlook-2026

NTV. (2026, 9 June). Türkiye için dönüm noktası: Akkuyu’da ilk üretim gerçekleşecek [A turning point for Türkiye: First generation at Akkuyu]. https://www.ntv.com.tr/ntvpara/turkiye-icin-donum-noktasi-akkuyuda-ilk-uretim-gerceklesecek-1727834

Ramage, T. (2026, 25 March). The Iran war is stress-testing South Korea’s energy model. Korea Economic Institute of America. https://keia.org/the-peninsula/the-iran-war-is-stress-testing-south-koreas-energy-model/

Reuters. (2026, 24 June). China’s coal power on the rise again in 2026, reversing first-in-a-decade decline. Engineering News. https://www.engineeringnews.co.za/article/chinas-coal-power-on-the-rise-again-in-2026-reversing-first-in-a-decade-decline-2026-06-24

Rystad Energy. (2026, 28 August). Short-term outlook: Europe draws in additional LNG volumes. https://www.rystadenergy.com/insights/short-term-outlook-europe-draws-in-additional-lng-volumes

Seo, H. (2026, 26 August). Japan to rebuild up to 14 nuclear reactors, cutting reliance on the Strait of Hormuz. The Financial News. https://en.fnnews.com/news/202608261428078107

Sommer, A. (2026, 8 September). Summer heat and conflict stoke winter risks. Axpo. https://www.axpo.com/es/en/newsroom/magazine/energy-market/summer-heat-and-conflict-stoke-winter-risks.html

Tekin, E., Solmaz, F. Z., Demir, İ., & Taşdemir, K. (2026, 20 September). Turkish President Erdogan arrives in US to attend UN General Assembly. Anadolu Agency. https://www.aa.com.tr/en/turkiye/turkish-president-erdogan-arrives-in-us-to-attend-un-general-assembly/4062456

Winzer, C. (2012). Conceptualizing energy security. Energy Policy, 46, 36–48. https://doi.org/10.1016/j.enpol.2012.02.067

Furkan Mahmut Tameroglu