The electric grid is the largest machine ever built — and it is really three systems bolted together. Every discussion of power markets, reliability, and energy investment starts with this anatomy: generation makes electricity, transmission moves it in bulk, and distribution delivers it to homes and businesses. Understanding who owns and operates each layer — and why the layers are regulated so differently — is what makes everything else in energy legible.
Generation: where the power starts
Generation is the asset-by-asset layer. Power plants — gas turbines, nuclear reactors, wind farms, solar arrays, hydro dams, coal units — convert energy into electricity and sell it. The United States has roughly 1,100 gigawatts of generating capacity spread across tens of thousands of individual units, from 1,300-megawatt nuclear reactors down to single-megawatt commercial solar rooftops.
The single most important distinction in grid economics is between capacity (what could be produced) and energy (what actually is produced). A 1-gigawatt nuclear plant produces power at full output roughly 90 percent of the hours in a year — its capacity factor. A 1-gigawatt solar farm in a sunny region might manage a 25 percent capacity factor; a wind farm, 35 to 45 percent. Two plants with identical nameplates can deliver wildly different amounts of energy, and reliability analysis lives entirely in that gap.
Who owns generators depends on where you are. In restructured states — Texas, much of the Northeast, the Mid-Atlantic, the Midwest, California — generators are independent companies competing in wholesale markets run by regional grid operators. In vertically integrated states — much of the Southeast, the Northwest, and parts of the Mountain West — regulated utilities own their power plants outright and earn a regulated return on the invested capital, with fuel costs passed through to customers.
Transmission: the bulk highway
Transmission moves electricity from generators to load centers over high-voltage lines: 115 kilovolts and above, with major backbones running at 230, 345, 500, or 765 kV. Higher voltage means lower resistive losses, which is why bulk power travels hundreds of miles efficiently — for illustration, losses on a 500 kV line run on the order of a few percent per hundred miles, versus far higher losses at distribution voltages.
The transmission network is not one grid but effectively three: the Eastern Interconnection, the Western Interconnection, and ERCOT in Texas, plus a handful of smaller systems. Within each interconnection, every large generator spins in synchrony at 60 hertz. Power flows between interconnections only through a small number of direct-current (DC) ties, which convert AC to DC and back again — a reminder that the “national grid” is really a set of continental-scale machines with limited doors between them.
Who runs transmission involves a split worth memorizing. Transmission owners — utilities and standalone transmission companies — build and maintain the wires. System operators dispatch generation and direct flows: regional transmission organizations (RTOs) and independent system operators (ISOs) cover roughly two-thirds of U.S. electricity demand, while elsewhere utility-run balancing authorities do the job. Owners are regulated monopolies earning FERC-authorized returns on invested capital; operators are neutral traffic controllers that do not own the assets they direct.
Distribution: the last mile
Distribution is the local network — substations, feeders, poles, and wires typically operating at 4 to 35 kV, stepped down to 120/240 volts at homes and 480 volts and up at commercial buildings — owned almost universally by regulated utilities franchised by states and cities. It is the most expensive layer per unit of energy delivered, because it is the most extensive: millions of miles of line serving every structure in the country, versus roughly 200,000 miles of high-voltage transmission.
It is also where most outages actually occur. The vast majority of customer outage-minutes in the U.S. come from distribution-level faults — a tree limb across a feeder, a car striking a pole, a substation equipment failure — not from bulk-system events. When people talk about grid reliability as experienced by customers, they are mostly talking about distribution performance. Bulk-system failures are rarer but vastly more consequential: one cascading event can darken tens of millions of customers at once.
Why the layers are regulated differently
The regulatory split follows the economics of each layer. Generation is potentially competitive — anyone can build a power plant — so restructured states let markets set wholesale prices. Transmission and distribution are natural monopolies — it makes no sense to build two competing sets of poles down the same street — so they are regulated: utilities earn an authorized rate of return on prudently invested capital (the rate base), with rates set to cover operating costs plus that return.
This is why utility investment behavior looks the way it does. A regulated utility’s earnings grow when its rate base grows — building a substation, reconductoring a line, hardening feeders against storms. It does not earn more by operating cheaply. Understanding this incentive is the key to understanding utility capital plans, rate cases, and merger logic.
Who answers to whom
Oversight follows the three-part split. The Federal Energy Regulatory Commission (FERC) sets wholesale market rules and transmission policy, including the authorized returns transmission owners earn. The North American Electric Reliability Corporation (NERC) writes and enforces mandatory reliability standards for the bulk power system, with fines for violations. State public utility commissions regulate distribution utilities, approve their capital plans, and set retail rates.
The jurisdictional seams matter. A state commission can order its utilities to harden distribution feeders; it cannot order a regional capacity market to value that hardening. FERC can redesign wholesale markets; it cannot site a transmission line over a state’s objection (with narrow exceptions). When something goes wrong on the grid, the first diagnostic question is always: which layer failed, and which regulator has jurisdiction over it?
AC vs. DC: why the grid is alternating current
The grid runs on alternating current because of a 19th-century engineering tradeoff that still holds. AC voltage can be stepped up and down efficiently with transformers — cheap, reliable devices with no moving parts. That made long-distance transmission practical: generate at moderate voltage, step up to hundreds of kilovolts for the journey, step back down for delivery. DC could not do this easily until modern power electronics.
High-voltage DC (HVDC) has carved out important niches where its advantages outweigh converter-station costs: very long point-to-point transfers (where DC’s lower line losses win over distance), undersea cables (where AC’s capacitive losses are prohibitive), and asynchronous interconnections (the DC ties between the Eastern, Western, and Texas grids). For illustration, an HVDC line becomes economically attractive roughly beyond 400–500 miles overhead, closer for cable. But the meshed network — the web of interconnected AC lines that provides redundancy — remains AC, because AC circuit breakers and protection are simpler and the entire system was built around it.
Losses: why voltage wins
Every mile of wire converts some electricity to heat. These resistive losses scale with the square of current — double the current, quadruple the loss. Since power equals voltage times current, transmitting the same power at higher voltage means lower current and dramatically lower losses. This square-law relationship is the entire economic logic of high-voltage transmission: a 765 kV line can move roughly four times the power of a 345 kV line with similar or lower losses per megawatt.
Losses also explain the shape of the system. Bulk transmission at high voltage keeps losses to a few percent across hundreds of miles. Distribution at low voltage loses a larger share per mile — acceptable because distances are short, but one reason distributed generation (rooftop solar, batteries) sited near load has an inherent efficiency advantage: it avoids the losses of the journey entirely.
The money flow: how each layer gets paid
Follow a dollar from your electric bill backward. The distribution utility collects it and keeps the regulated distribution charge — typically the largest single component of the bill. It passes the wholesale energy cost to generators (in restructured states, through market purchases; in integrated states, through its own plants’ costs) and pays transmission charges set under FERC-approved tariffs to transmission owners.
Transmission owners earn a FERC-authorized return on equity — historically in the neighborhood of 9 to 11 percent — applied to their invested capital. This is why transmission has attracted independent developers and private capital: it is a regulated, inflation-linked return on very long-lived assets. Generation, by contrast, earns whatever the market pays — feast or famine depending on the price environment, which is why merchant generators hedge heavily and integrated utilities prefer the certainty of rate-based ownership.
This money flow explains recurring industry behavior. Utilities propose capital-intensive solutions (they earn on capital). Developers propose generation where market prices are highest. Transmission owners propose lines. Everyone’s preferred answer is, not coincidentally, the product they sell. Reading any grid proposal starts with asking who earns the return.
Why this anatomy matters for what’s next
Every major grid debate is, underneath, an argument about which layer should change and who should pay. Data centers want to connect quickly — that is a distribution and transmission interconnection question. Wind and solar developers want to build where the resource is — that is a transmission expansion question. Policymakers want cleaner generation — that is a generation question, but one whose feasibility depends entirely on the other two layers.
The most common analytical error in energy is treating one layer as the whole system: celebrating generation additions without asking whether transmission can deliver them, or demanding reliability without distinguishing distribution failures from bulk-system failures. The three-layer anatomy is a diagnostic tool. When a blackout happens, ask which layer failed. When a policy is proposed, ask which layer it acts on and which layers must cooperate. When an investment is pitched, ask which layer’s rate base or market it earns from.
The grid’s next decades will stress this anatomy. Electrification grows the distribution layer’s job. Remote renewables grow transmission’s job. Retirements reshape generation’s job. But the structure — make it, move it, deliver it, balance it continuously — will not change, because it follows from physics, not policy.
One more number to carry into the rest of this series: the U.S. power system delivers roughly 4,000 terawatt-hours of electricity per year to more than 150 million customers through infrastructure valued in the trillions of dollars. It is the enabling platform for the entire economy, and its reliability is assumed rather than noticed — until it fails. That invisibility is the industry’s greatest achievement and its greatest political vulnerability: nobody funds the maintenance of a miracle they take for granted.
The constraint everything serves
One number governs the entire machine: electricity must be consumed the instant it is produced. There is no warehouse, no inventory buffer — supply and demand must match continuously, across a continent, in real time. Generation provides the supply, transmission provides the delivery path, distribution provides the last mile, and operators hold the whole thing in balance second by second.
Every market design, every reserve margin, every reliability standard you will read about in this series exists to manage that single constraint across three very different systems operating as one. Get the anatomy right and the rest follows.