Asia-Pacific
Australia’s energy-only NEM with five-minute settlement, Japan’s JEPX, nodal Singapore, India’s exchanges, and China reforming toward a unified national market by 2030.
Words the industry uses are marked like this at the sentence that defines them, so you can tell a term you will hear on a desk from a phrase this book happens to be using.
The Asia-Pacific has no shared electricity design, no regional regulator and no coupling algorithm binding it together. Its systems were built at different times, by governments with very different views about what a market is for, and several of them are being rebuilt right now.
Which makes the region the best available test of whether Part Three was the right way to organise this book. If the three axes are the questions that actually determine how a system behaves, then five systems that share almost nothing else should still be legible once their three answers are known.
Australia: energy-only, and the most solar-saturated grid on earth
The National Electricity Market (NEM) runs down the eastern seaboard, dispatches centrally, settles every five minutes, and pays nothing for capacity. On the third axis it made the same choice as Texas.
It made a different choice on the second, using five regional prices rather than nodes, with marginal loss factors recalculated annually to handle losses, as Chapter 15 described. And it chose a very different number for the ceiling that an energy-only market depends on. The Australian market price cap for 2026-27 is 23,200 Australian dollars per megawatt hour. Against ERCOT's 5,000 US dollars, and even allowing for the exchange rate, Australia permits scarcity prices several times higher than the largest energy-only market in America.
What makes Australia the system everybody else studies is the solar. By the middle of 2026 the NEM carried around 26.4 gigawatts of rooftop capacity across roughly 3.9 million installations, in a market whose demand rarely exceeds the mid-thirty gigawatt range. Renewables supplied 42.1% of NEM generation in the second quarter of 2026, a record for that quarter, while coal output fell around 5% year on year.
The consequences are visible in the price to a degree seen nowhere else. In the final quarter of 2025 negative prices occurred in about 31% of all NEM intervals. In South Australia the figure reached 48.4%, and in Victoria 43.1%. For nearly half of all five-minute intervals in one state, the wholesale price of electricity was below zero.
Rooftop solar is invisible to the market that it dominates. A household array is behind the meter, so it never offers into the auction and never appears in the merit order. It shows up as demand that failed to arrive. An operator facing a system with more rooftop capacity than some states' entire peak load is dispatching against a supply source it cannot see, cannot instruct and, until recently, could not curtail.
Australia is where the questions of the 2030s are being answered a decade early, and it got there through a subsidy programme and a sunny climate rather than through market design.
Singapore: nodal, in a country you can drive across in an hour
Singapore runs a nodal market, with prices computed at hundreds of points across a system serving a single island city-state.
The example matters because it disposes of the most common objection to nodal pricing. Chapter 15 set out the argument that nodal designs are for very large systems, since the complexity is only worth carrying when the network is big enough to develop serious internal constraints. Singapore is the size of a mid-sized American county and computes locational prices anyway, because the case for doing so rests on whether constraints exist rather than on how far apart the ends of the system are.
Japan: a market built after the fleet
Japan liberalised late and in stages. Retail competition opened fully in 2016, and the legal separation of transmission from the incumbent regional utilities followed in 2020. The Japan Electric Power Exchange, which had existed since the mid-2000s as a thin voluntary venue, grew into the clearing point for a substantial share of the country's power, with daily volumes reaching the order of a terawatt hour and covering roughly a third of national demand.
Two features follow from arriving late. Japan added a capacity market rather than attempting an energy-only design, because it was building the market around a fleet that already existed and whose owners needed revenue certainty to keep it running. And the country retains ten regional transmission areas with limited interconnection between them, plus a split between 50 hertz in the east and 60 hertz in the west, an artefact of nineteenth-century equipment purchases from Germany and the United States that still constrains how much power can move between halves of the country.
The 2020 to 2021 winter is the episode to know. Prices on the exchange rose by a factor of tens over several weeks during a cold snap that coincided with LNG supply tightness, and retailers who had sold fixed-price contracts while buying on the spot market were destroyed. It was the same structural failure that would appear in Texas six weeks later.
India: coupling used against a monopoly
India built exchanges rather than a single market operator, and the largest of them, the Indian Energy Exchange, ended up with something close to a monopoly, holding a share of traded volume in the region of 85% and above.
The regulator's response is the reason this belongs in the book. India is implementing market coupling, the same mechanism Chapter 19 described in Europe, under which bids from every exchange are pooled and matched centrally to produce one clearing price, with Grid India acting as the coupling operator. The Central Electricity Regulatory Commission told the Supreme Court in 2026 that the rules would take effect within weeks, and the court declined the incumbent exchange's request for a stay while its challenge proceeds.
Europe adopted coupling to join separate national markets across borders. India is adopting it to break a concentration of liquidity inside one country, because an exchange that holds most of the volume attracts the rest of it automatically, and no competitor can enter on price. The mechanism is identical and the purpose is competition policy rather than integration, which is a genuinely novel use of it.
China: from no market at all
China is building the largest electricity market in the world from a starting point that was not a market, and doing it on a published schedule.
Document 136, issued by the national planning and energy authorities in early 2025, required provincial governments to move all wind and solar generation to selling through the market by the end of that year, replacing fixed feed-in arrangements. It paired that with a contract-for-difference scheme, with strike prices and volumes set province by province. A separate notice called for spot market coverage nationwide by the end of 2025. In February 2026 the State Council set out the roadmap for the next five to ten years, under which the unified national market should be largely complete by 2030, with essentially all generation and consumption participating directly, and fully established by 2035.
Whether those dates hold is a separate question, and provincial protectionism has been the recurring obstacle, since provinces have long preferred to run their own plants rather than import cheaper power from neighbours. The direction is what matters for this book. A system organised entirely around administrative allocation is being converted into one organised around prices, at continental scale, inside a decade.
The region on three axes
Table 20-1: Five systems, three questions
| Who dispatches | How location is priced | How capacity is paid | |
|---|---|---|---|
| Australia | Single central operator | Five regions plus annual loss factors | Energy-only, very high cap |
| Singapore | Central operator | Nodal | Energy plus reserve products |
| Japan | Regional operators, exchange separate | Regional areas, weakly linked | Capacity market |
| India | Grid India, with competing exchanges being coupled | Regional | Mostly long-term contracts with distribution companies |
| China | State grid companies, provincial spot markets | Provincial, moving toward national | Administrative, with capacity payments being introduced |
The axes hold. Nothing in that table required a new question, and knowing the three answers for any of these systems tells a reader most of what determines how its prices behave.
What Part Four adds up to
Europe's 2024 market reform made the two-way contract for difference the standard instrument for financing new wind, solar and nuclear. China's Document 136 introduced a contract-for-difference scheme for renewables in the same period. A union of liberal democracies coordinating through published regulation, and a state-directed system converting administrative allocation into markets, reached for the same instrument within about a year of each other.
They did so for the reason Chapter 8 identified. A fleet with near-zero marginal cost cannot finance itself from a price set by marginal cost, and every system that builds enough of it eventually discovers the same gap and reaches for a contract to bridge it. The politics differ completely. The arithmetic does not.