Coal has been the main fuel of industrialization over the past century and a half. To this day, it remains a key feedstock for power generation (thermal coal) and in making steel (coking coal). Yet demand for coal appears to be peaking, with seaborne coal trade volumes plateauing at just over 1.2 billion tonnes.
While coking coal remains indispensable for
steel, thermal coal is seeing competition from cleaner fuels. Carbon Brief
published a study this month showing
that worldwide electricity generated
from coal dropped by 3 percent, or 300 terawatt hours, this year. While shipping
data still showed a slight increase in trade, it seems clear growth has topped
out.
Peak
demand in America & Europe
Global coal use may be
peaking, but there are big regional differences. In the United States, thermal coal has
been steadily pushed out by cheap shale gas and renewables. As a result, thermal
coal consumption has been declining for half a decade.
In Europe, coal also looks like it is in terminal decline. Shipped imports into former coal powerhouses Britain and Germany have collapsed over the past five years amid rising renewable and natural gas capacity while electricity consumption peaked in the early 2000s in both Britain and Germany, and has steadily declined since. Germany still burns a fair bit of domestic lignite to generate electricity, but Britain today uses barely any coal for power generation.
Asian coal use still strong
Asia is different. Shipped thermal coal
imports by the top four consumers—China, Japan, India and South Korea—have remained
strong at record levels above 400 million metric tons. While consumption is
falling in traditional import powerhouses Japan and South Korea for the same
reasons as it is in Europe, consumption is still rising in key emerging
economies like China and India. While both have pledged to combat pollution and
climate change, they still rely heavily on coal-fired power. In both countries,
coal also receives political support as authorities fear a move away from coal
would trigger rising unemployment from mining closures.
Many emerging economies across Asia—including
Indonesia, Vietnam, and India—also so far prefer
domestic coal to imported liquefied natural gas (LNG), as governments resist
spending big to develop LNG’s costly infrastructure, which will only raise the
country’s import bill.
Still, with consumption declining in Japan
and South Korea but growing in China and India, Asia’s thermal coal demand is creeping
up. Considering North America and Europe’s decline, it is likely the world has
reached peak thermal coal demand.
So what (for oil)?
Peak coal demand has not gone unnoticed. Also
fearing a peak
in demand and potentially stranded assets, oil producers are putting off spending
on future output. Meanwhile, investors are cutting exposure to petroleum assets
to comply with sustainability targets and shareholder pressure.
Yet peak oil demand has yet to happen. Consumption
this year rose by 1 percent and will for the first time average above 100
million barrels per day. While growth is slowing, consumption will likely
increase for years to come.
When it eventually peaks, a look at coal gives
a glimpse of what could happen in other sectors. Peak demand does not mean consumption will
fall off a cliff—coal demand has so far plateaued at or near records, with
pockets of growth still around.
Serving those pockets will remain a
profitable business for miners that produce the sought-after coal. The most
modern coal-fired power stations—euphemistically named “ultra-supercritical” and
mostly under construction in Asia—use different coal than the ageing coal-slingers.
This will favor different coal exporters. For instance, high quality thermal
coal from Australia is better suited to meet demand from new power stations than
coal from Indonesia, which tends to be of lower quality.
And while many European and American banks
now shun coal, credit remains available as investors, including from Japan,
where most “ultra-supercritical” turbines are made, still lend to projects they
perceive as relatively clean.
Similar trends may be emerging in oil. Many
investors prefer small and short-cycle oil assets like US shale or projects in
fully developed markets like Europe’s North Sea to risky, costly and long-cycle
new production like Brazil’s deepwater fields.
Like coal, the oil industry will see a
shift in demand for certain crude grades even as overall consumption growth
stalls, plateaus, or peaks. In shipping, this is already happening as a sulfur cap from January 2020 in marine
fuels has
pushed up demand for niche crude grades that are medium and sweet in quality. Tightening
environmental regulation will soon hit gasoline and diesel consumption from cars,
albeit at high levels. However, demand for oil from other sectors like petrochemicals
(e.g., household chemicals, textiles, and consumer goods) will still grow for
years. This will trigger another change in demand away from heavy/sour grades
commonly used for transportation to lighter/sweeter types used more by the
chemical industry. For specialized producers and investors, such an outlook
offers opportunity even amid a peak in overall consumption.
Henning Gloystein is director, global energy & natural resources at the Eurasia Group. You can follow him on Twitter @hgloystein
EnergySource provides analysis and insight on key energy issues, making sense of key energy trends and their implications for geopolitics, geoeconomics, policy, and markets.
When it eventually peaks, a look at coal gives a glimpse of what could happen in other sectors. Peak demand does not mean consumption will fall off a cliff—coal demand has so far plateaued at or near records, with pockets of growth still around. What could this mean for oil?