The Grid That Wouldn't Die

Samuel Insull built the modern American electric grid in the 1920s, and in some fundamental sense we are still living inside the system he designed. Insull, a former personal secretary to Thomas Edison who had transformed Chicago’s patchwork of competing power companies into a unified utility by 1907, understood that electricity was a natural monopoly. You don’t want two sets of wires running down every street. You want one company, regulated by the state, serving everyone at a price set by a public commission. This model spread across the United States over roughly three decades, and by the time Insull’s own empire spectacularly collapsed in 1932 — one of the largest corporate failures in American history to that point — the institutional framework he created had become the permanent template.

Nearly a century later, the grid looks startlingly similar. The United States still has approximately 3,000 separate electric utilities. The transmission infrastructure is divided into over 60 balancing authorities. The regulatory framework is split between federal oversight (the Federal Energy Regulatory Commission, created in its current form in 1977) and state public utility commissions, each of which has its own rules, rate structures, and political constituencies. Electricity that flows across state lines enters a maze of jurisdictional disputes that would have baffled an engineer and delighted a lawyer.

The obvious question is why. The grid is widely acknowledged to be inadequate for the energy transition: it needs to carry far more power, over longer distances, to balance variable renewable generation. The American Society of Civil Engineers gives it a grade of D+ in infrastructure report cards. The International Energy Agency estimates the United States needs to add or replace 80 million kilometers of power lines by 2040. That is, roughly speaking, twice the total existing grid. The scale of required investment is staggering, the institutional obstacles are even more so.

The first obstacle is the regulated utility model itself. A utility earns its return not on the power it sells but on the capital it deploys — a system called rate-of-return regulation. If a utility spends money on transmission infrastructure, that capital goes into the rate base, and the utility earns a regulated return (typically 9 to 11 percent) on it. This sounds like it should encourage investment. In practice it encourages the specific kind of investment that regulators will approve, which means local distribution infrastructure rather than long-distance transmission, which requires coordinating with neighboring utilities, crossing state lines, and navigating federal approval processes that routinely take a decade. The path of least resistance is always local.

The second obstacle is incumbent generation. When a new high-voltage transmission line is built connecting cheap wind power in Wyoming to load centers in California, someone wins and someone loses. The winners are Wyoming wind developers and California consumers. The losers are the existing California generators whose output was previously valuable because there was no competition from elsewhere. Those generators have been fighting transmission for decades, and they are very good at it. The arguments shift — grid reliability, environmental impact, visual blight, property rights — but the underlying dynamic does not. Incumbent generators systematically resist the infrastructure that would dilute their market position.

The third obstacle is what economists call the problem of cost allocation. If three states benefit from a new transmission line but the line physically crosses only one of them, who pays? Federal rules require that costs be allocated in proportion to benefits, but quantifying benefits is extraordinarily contentious. Every utility’s engineers will produce models showing that their customers derive minimal benefit and should pay minimal cost. The resulting negotiations can last longer than the construction would take. The Midcontinent Independent System Operator, which manages the grid across 15 states, spent roughly eight years developing a cost allocation methodology for a major transmission expansion that was approved in 2011. The projects themselves took another decade to build.

What is remarkable is that this is not uniquely American dysfunction. Germany’s Energiewende, the ambitious energy transition launched in 2000, ran directly into the same problem: wind resources concentrated in the north, industrial demand concentrated in the south, and transmission infrastructure that could not be built fast enough to connect them. By 2019, Germany was spending roughly 1.4 billion euros per year on “redispatch” — paying generators to turn up or down to compensate for congested transmission lines that couldn’t carry power where it needed to go. The problem is not cultural. It is structural.

The structural problem, stripped to its essence, is that transmission is a public good built through private decisions. The economic benefits of a well-connected grid are diffuse and largely captured by consumers over decades. The costs are immediate, local, and fall on specific property owners and specific utilities. The political economy of immediate concentrated costs versus diffuse long-term benefits is brutal for any infrastructure that requires coordination across jurisdictions.

What has actually worked, historically, is federal power. The interstate highway system was built by the federal government, which could impose uniform standards, acquire right-of-way through eminent domain, and distribute costs across the entire country. The Rural Electrification Administration, created in 1935, extended power to farms across America not through market mechanisms but through federal financing and technical assistance to cooperatives that had no profit incentive to serve unprofitable rural customers. The Tennessee Valley Authority, also 1935, built hydroelectric infrastructure across seven states because a federal agency could coordinate what no private entity or state government could.

The Infrastructure Investment and Jobs Act of 2021 allocated $65 billion for grid improvements, the largest single federal investment in transmission infrastructure since the New Deal. The Inflation Reduction Act of 2022 added further incentives. And yet by 2026, the interconnection queue — the list of power plants waiting for permission to connect to the grid — contained over 2,600 gigawatts of proposed projects, mostly renewables, waiting for studies that could take five years and approval that might never come. The money existed. The institutions remained incapable of processing the applications.

The situation illuminates something important about how infrastructure systems calcify. When Samuel Insull’s model was being established in the 1920s, it was genuinely innovative: it produced cheap, reliable power and funded universal service through cross-subsidies. The regulatory framework that grew around it was designed to solve real problems. But institutions designed to solve one era’s problems become the following era’s obstacles. The rate-of-return model works well when you need local distribution but badly when you need long-distance transmission. State-by-state regulation works well for local monopolies but badly for interstate infrastructure.

The economists’ solution is restructuring markets so that transmission is planned and compensated at the regional or national level rather than by individual utilities. The Federal Energy Regulatory Commission’s Order 1000, issued in 2011, tried to do this by requiring regional transmission planning and competitive solicitation of projects. By 2025, the results were mixed at best. The order created processes but not outcomes. Processes designed by committees of utilities, state regulators, and other stakeholders tended to produce compromises that satisfied no one’s infrastructure needs while satisfying everyone’s bureaucratic requirements.

What the American grid story really illustrates is that the difficulty of changing infrastructure is not primarily technical. The United States has the engineering knowledge, the financial capital, and the physical space to build a modern grid. What it lacks is a political mechanism for overriding the distributed veto power of incumbents who benefit from the existing system. Regulation creates beneficiaries. Beneficiaries resist change. The more successful the original system, the more powerful the incumbent interests it created, and the harder subsequent change becomes.

There is a grim historical irony here. Samuel Insull himself was eventually destroyed by the system he created. His holding company empire, built on pyramiding utility assets across states, collapsed when the stock market crash of 1929 drained liquidity and exposed the extent to which the structure depended on perpetual growth. The subsequent investigation by the Federal Trade Commission laid the groundwork for the Public Utility Holding Company Act of 1935, which broke up the interstate holding companies and entrenched state regulation even more firmly. The system that replaced Insull’s empire proved even more resistant to change than his original creation, precisely because it was designed to prevent the kind of centralized control that had made his empire possible — and that would also have made national grid modernization possible.

The lesson is not that monopoly is good or that deregulation is bad. It is that every institutional solution encodes the political compromises and power relationships of its moment of creation, and those compromises become harder to revisit as the institutions mature and the people who benefit from them organize. The American electric grid is not primarily a technical system. It is a political system that happens to carry electrons.

Whether the grid will modernize fast enough for the energy transition is genuinely uncertain. The federal government has shown it can move when it commits fully — the New Deal programs moved very fast once the political will was clear. But political will is not clear, and the distributed opposition is well organized and well funded. The most likely scenario is incremental improvement: enough to prevent catastrophic failure, not enough to achieve the transformation that climate targets require, and a growing gap between what the grid does and what we need it to do. Insull would recognize the pattern.