Solar panels on the Stirling Arms Hotel, Guildford, Western Australia. Source: Wikimedia Commons
In past articles for the Dispatch Energy newsletter, I’ve written about the American electricity industry, especially its technology, its regulation, and the pressures now coming from data centers. But my focus has been domestic, and it would be understandable if you concluded from all this that American arrangements are, shall we say, idiosyncratic. This industry is physically and economically complicated, built on enormous infrastructure. Is it really this weird everywhere, or just here?
Generation Down Under
I start, as always, with the physical reality that electricity does not care about state lines. A generator in one state can serve customers in another. Transmission networks let regions share reserves, tap different resources, smooth out weather and demand, and increase competition, lowering the cost of keeping the lights on.
Some of the most important changes, though, are happening at the opposite scale. Rooftop solar panels, batteries, electric vehicles, and flexible loads sit behind millions of individual meters, and their owners know things no grid operator can know for them, like what they’re willing to pay, when they can shift consumption, or how much inconvenience they’ll tolerate.
Modern electricity, therefore, poses an institutional puzzle: While coordination across enormous areas benefits some decisions, decentralized knowledge and individual choice drive others. International comparison helps clarify these competing demands by showing how different institutional arrangements divide authority, coordinate across scales, and adapt to similar technological and economic pressures. Australia is similar enough to the U.S. to make a comparison worthwhile. Both countries have a large land mass, a federal system dividing authority between a national government and the states, and long distances separating both population and varied energy resources.
Australia has built what I’ll call nested institutions: Different organizations make decisions at different scales, but their roles connect deliberately. The United States, meanwhile, more often has layered institutions: Federal, regional, state, utility, and local authorities overlap, sometimes productively and sometimes at seams that impose real transaction costs. The difference shows up at both ends of the grid, with wholesale markets and transmission at the top and rooftop solar at the bottom.
Where the two grids began
Australia’s electricity industry started, like America’s, as a decentralized mix of private, municipal, and small public utilities in the late 19th century. But by the 1930s, Australian states had consolidated their systems into large public utilities and used that ownership to electrify rural areas. The United States took a different route, mostly keeping investor-owned utilities in place and reaching rural areas instead through the Rural Electrification Administration, which financed member-owned cooperatives with low-interest federal loans rather than nationalizing the industry.
By the eve of the 1990s reforms, Australia’s system looked like the traditional American utility model, but with public rather than investor ownership, with each state running its own vertically integrated utility and connecting to its neighbors only at the margins. That legacy of public ownership created a different starting point for reform than America’s landscape of primarily investor-owned utilities and cooperatives.
Two reform paths
In the 1990s, the Australian government undertook a broad program of reforms similar to those in the United Kingdom under Prime Minister Margaret Thatcher—namely, privatization and market mechanisms for big infrastructure industries like power and telecommunications. That program meant unbundling the vertically integrated state authorities into generation, transmission, distribution, and retail, with implementation varying by state while the wires stayed regulated monopolies everywhere. The states then worked together to build a single National Electricity Market, launched in 1998 across the interconnected eastern and southern states.
The United States restructured on a similar timeline but a more decentralized path. The Energy Policy Act of 1992 reduced legal barriers to competitive wholesale markets, and utilities were encouraged, not required, to join the resulting regional markets. Several formed instead of one—PJM, New York ISO, ISO-New England, California ISO, and Texas’ ERCOT—while much of the Southeast and West never joined one at all.
Both countries confronted the same economic insight in the 1990s: Generation gets more competitive when transmission opens up and more generators compete and are dispatched across larger areas. And they built that insight into very different forms of federalism—Australia through a single national bargain, America through a landscape of separate regional agreements.
A national market without national command
The National Electricity Market did not eliminate the role of the states, and Australia did not create a national regulator to plan and operate the industry. Different institutions instead acquired different functions: The Australian Energy Market Operator (AEMO) runs the wholesale market and power system, the Australian Energy Regulator polices monopoly networks, the Australian Energy Market Commission writes the rules, states retain authority over energy policy, and firms make their own investment decisions.
The market design reflects the same instinct toward specialization. Rather than layering on a separate administrative capacity market of the sort that some U.S. markets use to induce generation investment, Australia made the National Electricity Market an energy-only market: Generators earn their revenue through electricity and ancillary-service prices alone, including potentially very high prices during genuine scarcity, while financial derivatives let participants manage that price risk on their own. The goal, in the Australian Energy Market Commission’s own account, is economic efficiency—dispatching the cheapest available resources first and letting scarcity pricing do the work a capacity market would otherwise do by administrative fiat. In the U.S., only Texas’ ERCOT market operates under similar rules.
This division of institutional labor is easiest to see in transmission. Australia tracks price spikes and investigates market power, and a 2005 review found transmission congestion behind a large share of high-price periods, triggering institutional change. AEMO gained a system-wide planning role, and new projects had to prove their economic benefits, while regulated companies kept building them. This structure in which national information sits above decentralized implementation is nested institutional coordination.
America’s version looks similar on paper, but its seams run deeper, worsened by a basic incentive problem: A utility earns its regulated return by expanding its own network inside its own territory, so a line that mostly benefits customers elsewhere is somebody else’s problem to fund. The problem isn’t the number of organizations; a utility, a state regulator, and FERC can each do their job fine on their own. The real question is whether these layers fit together well enough for a decision at one level to account for costs and benefits at another.
From interstate wires to rooftop panels
Another instructive comparison comes from how the two countries have managed distributed energy resources, mainly rooftop solar, sitting behind millions of individual meters. Australia has become one of the world’s great rooftop-solar experiments, with more than 4 million small systems installed, roughly one for every three homes.
Why? Subsidies, for one. The federal government cut the upfront price of installations, and several states offered extraordinarily generous feed-in tariffs, making solar privately profitable for households where it otherwise wouldn’t have been.
This investment was not necessarily socially efficient—a subsidy drives a wedge between private and social cost, and invoking emissions damages does not show rooftop solar is the least-cost way to abate them. High solar adoption is not evidence of good institutions: Pay people enough to buy something, and they generally will.
But subsidies are only part of the story. Australia also made installing rooftop solar routine. A subsidy changes the return on an investment; removing unnecessary permitting and paperwork reduces the resources an investment consumes in the first place. The distinction contrasts with American residential solar’s unusually expensive installations, even though installers in both countries buy the same globally traded panels and inverters. A routine American installation can run into utility rules, permits, inspections, building codes, and jurisdiction-specific fees—thousands of small resource-consuming requirements. Australia regulates rooftop solar too, but it has made the ordinary installation standardized and administratively boring.
California’s experience drives the point: Few places have pushed rooftop solar harder with layers of subsidies and net metering, and yet Australia has achieved far higher penetration at lower cost. California illustrates American institutional layering—state regulation, utility rules, municipal permitting, building codes, and local inspection, seams that repeated streamlining has never quite closed. The comparison is not regulation versus deregulation—both California and Australia regulate heavily—but whether institutions are nested into channels for coordination, or merely layered into transaction costs.
When success creates a new scarcity
Australia’s rooftop solar boom eventually produced a different problem. Distribution networks were built to move electricity in one direction, from generators to customers. Put enough panels on enough roofs, and customers start sending it back the other way.
Solar adoption in a legacy one-way distribution network becomes an economic problem at scale: Local feeders have finite capacity, and midday solar output can grow abundant enough that another kilowatt-hour has little value, even as electricity a few hours later stays valuable (which is why batteries are such economically transformative technologies). A network built for passive consumers must now allocate scarce capacity among millions of small producers.
The first responses were blunt: fixed export limits that capped how much a household could send back regardless of conditions. Australia has since moved toward flexible exports. In South Australia, for example, smart inverters let households export more when local capacity allows it, varying access with conditions instead of designing every connection around the worst possible moment (which is what we tend to do in the U.S.). The same logic applies to batteries, EV charging, and flexible demand: Electricity at noon is not worth what it is at 7 p.m., and a battery behind a constrained feeder can be worth more than an identical battery elsewhere.
Here the role of prices and distributed knowledge applies, with a caveat: The mechanism is administered, not organic. Using a dynamic operating envelope, the network operator calculates a number from real-time voltage and loading data on the local feeder, feeding it back to smart inverters as a synthetic limit rather than a price discovered through voluntary trades. It still provides real coordination, because no grid operator has to know what any household wants, and no household has to know the condition of every generator and feeder on the system. The distribution business only measures its own wires and computes a number, and the household only responds to it.
Nested institutions, layered institutions
Australia should not be lionized: Its electricity system is intensely political, heavily regulated, and burdened with its own distortions. Its solar subsidies offer a textbook case of the price distortions market-oriented economists rightly scrutinize. But American diversity should not be demonized: Multiple regional markets test different designs, and states and utilities adapt to local circumstances, creating experimentation that a uniform system might suppress.
Having many layered centers of authority instead of one isn’t a problem in itself. The question is whether they’re connected well. A good system has authority operating at the right scales, joined up so information and learning can travel between them, while a badly layered one produces veto points, incompatible rules, and boundaries where nobody accounts for the gains from working together.
Australia’s institutions look nested, with a common wholesale market sitting above regional networks, national planning incorporating local information, distribution companies experimenting with flexible access within common rules, and households staying free to invest as they choose. The United States contains more institutional layers and more institutional seams. Sometimes those seams protect valuable autonomy; sometimes they are just transaction costs. Distinguishing between the two situations is the hard economic question.
Finding the right scale
Technological change makes that question more urgent because power systems now need coordination across vastly different scales. Wholesale trade wants large geographic markets, transmission needs planning beyond utility territories, distribution needs granular local information, and solar, batteries, EVs, and flexible loads depend on millions of private decisions no central planner should make.
Australia’s reforms are interesting because they move in both directions at once, integrating markets upward while letting decisions and experimentation move downward. Rather than being a contradiction, this may be the central institutional problem of the modern grid.
The problem isn’t centralization versus decentralization; it’s matching decisions to the scale where the relevant knowledge, costs, and benefits actually reside, and then connecting those levels effectively. The choice is between institutions that are nested and institutions that are merely stacked. Electricity needs both large markets and local knowledge, and designing institutions that can use both is the difficult work ahead.
Originally published in the Dispatch Energy newsletter at https://thedispatch.com/newsletter/dispatch-energy/australia-electrical-system-grid-solar/
Lynne, this comparison helped me see the American system more accurately rather than simply as institutional disorder. Australia’s nested structure may coordinate wholesale markets and transmission more effectively, while America’s messier layers preserve multiple regional experiments and more opportunities for local discovery. As you show, the real challenge is not choosing between centralization and decentralization. It is integrating enormous electricity systems while capturing the rapidly expanding knowledge emerging behind individual meters. I would be interested in whether Australia is better at carrying that local knowledge upward—or whether America’s greater openness ultimately produces more useful experimentation despite its considerable transaction costs.