Aug Concentration: Power Top-1 Still 33.2% / Top-3 56.5% — Coal Rebound Tip ~48% China
Late-Aug energy-systems concentration after IEA Mid-Year Update: Ember electricity Top-1/Top-3 carried flat at 33.2%/56.5%; path meters add China ~42% of TWh growth, ~48% of the +1.4% coal rebound, and ~50% of ~610 TWh solar add — while EU/Japan wholesale >+30% prices the unchanged US LNG tip.
Loading interactive charts…
Our Q3 concentration lens re-scored Top-1 / Top-3 / HHI after Ember’s Global Electricity Review 2026 and IEA World Energy Investment 2026: electricity Top-1 China 33.2%, Top-3 56.5%, US LNG Top-1 24%, clean investment Top-1 China ~34% of $2.2T, solar modules still 80% China. The August Mid-Year update then answered the level and mix question — demand 3.6% / 3.8%, renewables widening toward 37%, coal generation rebounding +1.4%, EU/Japan wholesale >+30%. This late-Aug concentration cut asks the distribution follow-up Mid-Year forces: did the security shock thicken the top of the share ladder, or only re-price a tip that was already thick?
The dashboard above is that 202608 concentration lens. Toggle Scoreboard, Stock vs growth, Aug path meters, Trade & shock, and Capex & vintage. Metric controls flip Top-1 / Top-3 / HHI / Δ vs Q3; lens filters isolate demand, path, export, manufacturing, and investment. The punchline is deliberately two-sided: Ember stock tips are carried flat (no newer period census), while Mid-Year path meters show who owns the next TWh add, who owns the coal rebound, and who owns the solar generation increase — and the wholesale scatter shows which import-dependent systems pay for the unchanged LNG export tip.
The Aug scoreboard: carried stock, new path tips
| Lens | Q3 concentration | Aug 202608 | Δ |
|---|---|---|---|
| Electricity generation Top-1 | 33.2% (China) | 33.2% | 0 pp (carried) |
| Electricity Top-3 | 56.5% (CN+US+IN) | 56.5% | 0 pp |
| TPES demand Top-1 | 27.6% | 27.6% | 0 pp |
| Demand-growth add Top-1 | ~42% (China) | ~42% | 0 pp (restated path) |
| Coal generation add Top-1 | — | ~48% (China) | NEW path tip |
| Solar generation add Top-1 | — | ~50% (China) | NEW path tip |
| LNG exports Top-1 | 24% (US) | 24% | 0 pp (priced, not rewritten) |
| Hard coal exports Top-3 | ~73% | ~73% | 0 pp |
| Clean investment Top-1 | ~34% | ~34% | 0 pp (WEI carried) |
| Solar PV modules Top-1 | 80% | 80% | 0 pp |
| EU/Japan wholesale YoY | >+30% | >+30% | Shock meter (disclosed) |
Read the table as a stability-plus-path scoreboard. Mid-Year did not publish a new Ember country census or a new Energy Institute primary-share tape. What it rewrote is the trajectory: demand accelerates, coal generation rebounds once, renewables widen their mix lead, and wholesale prices split by LNG exposure. Desks that treat “renewables to 37%” as “concentration eased” are reading the mix column and ignoring the geography columns.
Scoreboard: what moved is the path tip, not the Ember tip
Open Scoreboard. The vintage-delta bars are mostly zeros on electricity Top-1/Top-3, LNG Top-1, solar manufacturing, and growth-add Top-1 — those are carried from the Q3 concentration companion. The bars that jump are coal-add Top-1 and solar-add Top-1: Mid-Year’s new objects, estimated as shares of the global coal TWh add and of the ~610 TWh solar generation increase.
Toggle Δ vs Q3 on the ladder and the same geometry appears in market space. Path lenses print large positive deltas because they did not exist as concentration meters on the Q3 board; stock lenses print flat. That is the methodological hinge for this vintage: a Mid-Year Update can thicken the distribution of flows without rewriting the distribution of stocks.
Stock vs growth: same China tip, different denominators
Switch to Stock vs growth. Leave the curve on Stock (Ember TWh). The Lorenz panel still clears 33.2% at rank 1 and 56.5% by rank 3 — identical to Q3 because the Ember GER 2026 census has not been superseded. Flip to Growth add and the tip steepens: China alone is about 42% of the world TWh add on IEA’s 3.6% (2026) / 3.8% (2027) path, with China + India + US near 68%.
India still prints the fastest major-system growth rate (~7%) on a smaller base; China still wins volume concentration of the add. That distinction is why the August update showed regional rates and this post converts them into share of the add. Averaging a 7% India print with a 1.8% US print and calling the average “global demand” is how desks understate Asia’s tip.
Aug path meters: coal rebound and solar add geography
Open Aug path meters. On Demand path, bars and the TWh line climb from 28,600 TWh (2025) toward 30,700 TWh (2027e). The growth-add pie repeats the 42% / 68% tip. Switch to Coal rebound — the uncomfortable Mid-Year companion. Global coal-fired generation is expected to rise 1.4% in 2026 after the Q3 Ember year that celebrated fossil generation −0.2%. Estimated add shares put China near 48%, India near 22%, and the EU’s slowed decline as a residual — Top-3 of the add near 78%.
That is not a share reclaim of the global generation mix. Renewables still take the #1 slot and widen toward 37% by 2027 with VRE at 21%. It is a concentration statement about who owns the one-year dispatch rebound the Hormuz / LNG shock reopened. Analysts who treated Ember’s 2025 fossil halt as a permanent trajectory without a geographic add lens will mis-rank 2026.
Toggle Solar add. IEA’s ~610 TWh solar generation increase (+23%) is a near-record repeat of Ember’s +636 TWh 2025 print. China still supplies about half of the increase; India and other Asian economies rise toward ~17% of that add. Greening the mix and concentrating the build are compatible — and the manufacturing tip at 80% China for modules remains the supply-chain twin of that generation tip.
Trade & shock: LNG tip unchanged, wholesale prices the tip
Switch to Trade & shock. Fuel-export stacked bars keep coal (Indonesia Top-1 ~36%, Top-3 ~73%), LNG (US 24%, Top-3 ~63%), pipe gas, and crude on the carried tape. Mid-Year does not rewrite those seaborne shares in a single print. What it does is price the LNG tip: the temporary loss of nearly 20% of global LNG supply pushed Asian and European gas prices to their highest levels since 2022–23, and Q2 2026 wholesale electricity averages rose >+30% YoY in the EU and Japan, stayed roughly flat in the US, rose less than 10% in India, and fell about 45% in Australia.
The import-dependence × wholesale scatter makes the exposure geometry concrete. Japan (~88% import dependent) and the EU (~58%) sit with the shock prints. The US (net energy exporter on this meter) is flat. Australia, deep export surplus with renewables and batteries cutting peak gas reliance, prints −45%. Concentration of export supply and concentration of import exposure remain two sides of one trade map — Mid-Year simply lights up the price channel the research ledger already mapped in stock terms.
Capex & vintage: capital tip carried, slope confirms the freeze
Open Capex & vintage. Clean investment geography is still the IEA WEI 2026e tape: China ~34% of $2.2T, Top-3 (China + US + EU) ~58%, fossils $1.2T. No newer World Energy Investment period print arrives in the Mid-Year electricity update, so the capital tip is carried — another reminder that greening the numerator of investment does not automatically diversify the geography of who builds modules and cells.
The vintage slope (Research → Conc 2026 → Q3 → Aug) shows electricity Top-1 climbing 31.5% → 32% → 33.2% → 33.2%, LNG Top-1 21% → 22% → 24% → 24%, and solar manufacturing frozen at 80% across the entire theme arc. Aug’s novelty is the coal-add and solar-add path tips sitting beside that frozen stock slope — not a restatement of Ember ranks.
Who is exposed — and what would rewrite the tips
Exposed on the stock clock: desks that treat “renewables → 37%” as proof of geographic diversification; Western OEMs and utilities whose growth models still assume OECD TWh shares hold mid-2010s norms; coal-exposed systems that ignore China’s dual role as renewables builder and coal-add Top-1 (~48%).
Exposed on the path clock: LNG importers in Europe and Northeast Asia priced into >+30% wholesale YoY; buyers without destination-flexible contracts when the US–Australia–Qatar tip still clears ~63%; analysts who read US LNG growth as pure security without noticing Top-1 concentration remains elevated even when the share print is unchanged.
What would rewrite the next concentration cut: a new Ember country census that moves electricity Top-1 off 33.2%; an EI Statistical Review that moves TPES China off ~27.6%; Hormuz shipping fully normalised with gas prices collapsing coal’s relative merit order early (shrinking the coal-add tip); or a WEI refresh that diversifies clean-capital geography away from China’s ~34%.
Caveats and methodology
- Carried ≠ confirmed unchanged forever. Flat Δ vs Q3 means no newer census, not a guarantee the next Ember print will hold the tip.
- Primary ≠ electricity. TPES and TWh share tips live on different denominatorsdo not average them into one “energy concentration” slogan.
- Coal +1.4% is global generation growth, not a mix reclaim. Renewables still lead and widen toward 37%.
- Coal-add and solar-add shares are estimated from Mid-Year regional narratives (China largest share of coal increase; China ~half of solar increase)treat as directional path meters, not disclosed national add tables.
- Growth-add Top-1 ~42% converts regional growth rates into share of world TWh add; it is sensitive to the path assumption (3.6% / 3.8%).
- Wholesale YoY figures are Q2 2026 spot averages as summarised by IEAnot full-year realised retail bills.
- Clean investment $2.2T and manufacturing 80% / 75% tips are carried from WEI / IEA supply-chain secondary prints.
- HHI values are constructed from rounded country shares for board comparisonnot antitrust filings.
- Import-dependence meters mix Eurostat-style energy dependence with research-ledger framing; Australia’s deep negative dependence is an export-surplus visual, not a literal “−140% import” identity.
- This post is a concentration companion. For Mid-Year levels and mixes use the August update; for Ember/WEI Q3 levels use the Q3 update; for the prior Top-k tape use the Q3 concentration print and the 2026 concentration print; for country mix ledgers use the research post.
Bottom line
Late-Aug does not dissolve energy-system concentration — it separates stock tips from path tips. Electricity Top-1 China remains 33.2%; Top-3 56.5%. US LNG Top-1 stays ~24% with Top-3 ~63%, now priced into EU/Japan wholesale >+30%. Clean investment Top-1 stays China at ~34% of $2.2T. Solar modules remain 80% China. What Mid-Year adds to the distribution question is the flow tip: China ~42% of the demand add, ~48% of the coal rebound add, and ~50% of the solar generation increase — while renewables widen the mix lead toward 37%. Composition can green while geography stays top-heavy.
Related reading: Q3 concentration lens, August Mid-Year update, Q3 Ember/WEI update, 2026 concentration print, Energy systems research, and Chokepoint commodities Q3 concentration.