# How We Built Institutions for the Wrong Grid ### Paper I: The history, in service of the argument *Companion papers: II. The Great Inversion (the diagnosis) and III. Completing the Market (the reforms).* --- The grid no longer suffers from scarce electricity. It suffers from scarce permission. Paper II documents that inversion; this paper explains where the permission structure came from, because many of the rules now constraining the queue were sensible answers to the engineering and institutional problems of their time. Three settlements built the modern grid. Understanding what each one solved, and what each one assumed, is what makes the third settlement proposed in Paper III legible as evolution. ## Settlement one: the regulatory compact (1907) Electricity began as a competitive business, and an ugly one. After Edison energized Pearl Street in 1882, dozens of firms strung duplicate wires through American cities, and city councils sold franchises in ways that made "utility" and "corruption" synonyms. Samuel Insull, Edison's former secretary and the builder of Chicago's Commonwealth Edison, proposed the fix that stuck: one large generator and one set of wires served a city more cheaply than five competing systems, so let the state grant a monopoly and discipline it through a commission instead of through rivals. Wisconsin and New York created the first public utility commissions in 1907. The deal, an exclusive franchise plus an obligation to serve, in exchange for rates regulated around prudently incurred costs and an authorized return on capital, is the regulatory compact, and it still governs the wires outside a few restructured states [1][2]. Two features of the compact matter for everything that followed. Insull campaigned for it himself, because a state commission was more predictable, and more manageable, than municipal politics; the observation that regulation can serve the regulated as well as the public is as old as the institution, and Stigler formalized it in 1971 [3]. And the compact's payment formula, a return on invested capital, hard-wired a preference that Averch and Johnson would name in 1962: a firm paid for capital will choose capital, whatever the cheaper alternative [4]. Neither feature was a design flaw at the time. Capital was exactly what a country electrifying from nothing needed, and the commission was a real improvement on the city council. The compact scaled until it broke financially. Holding companies pyramided utilities across state lines through the 1920s; the 1929 crash took down the leverage, with Insull's empire among the era's largest business failures, and the New Deal responded with the Public Utility Holding Company Act and the Federal Power Act of 1935. The 1935 settlement drew the jurisdictional seam that still defines the industry: states govern retail rates and distribution, the federal government governs wholesale sales and interstate transmission. Most of the paralysis Paper II documents lives on that seam, because the resources that now matter most for flexibility (batteries, aggregators, flexible loads, distributed generation) operate across or near it. ## The assumption fails (1965 to 1978) For four postwar decades the compact delivered falling real prices, and nobody audits a machine that prints discounts. The discounts stopped in the 1970s: scale economies in thermal generation ran out, oil shocks and inflation drove costs up, nuclear programs hemorrhaged capital, and commissions had to learn to say no. The intellectual foundation cracked with the cost curves. Joskow and Schmalensee's *Markets for Power* (1983) argued that generation, unlike wires, could support competition [5], and PURPA (1978) proved it by accident: required to buy from independent cogenerators at avoided cost, utilities discovered that outsiders could build and run plants perfectly well. The monopoly on generation had been a policy choice wearing an engineering costume. ## Settlement two: the half-finished unbundling (1992 to 2000) The second settlement acted on that discovery at higher voltages. The Energy Policy Act of 1992 opened wholesale generation to competition. FERC's Order 888 (1996) forced utilities to open their transmission lines to all comers on equal terms, encoding the principle that a wire owner who also owns generators cannot referee access to the wire. Order 2000 produced the independent regional operators that now dispatch two-thirds of American load. The pricing architecture came from Schweppe's *Spot Pricing of Electricity* and Hogan's locational marginal pricing: let the price of power vary by time and place to reflect the physics [6][7]. Turning Kirchhoff's laws into a price system remains one of the great applied results in market design, and it works: at the transmission level, thousands of independent generators compete to serve load at prices that track real scarcity. The important distinction is that wholesale unbundling did not eliminate monopoly. It separated competitive functions from transmission's natural-monopoly platform. This series extends that logic into distribution; it does not assume that distribution itself has ceased to be a natural monopoly. The settlement stopped at the substation, for two reasons worth distinguishing. The defensible reason was trauma: California's 2000-01 crisis, a failure of design and manipulation rather than of competition itself [8], froze retail restructuring in half the states and made "market" a dirty word in utility regulation for a decade. The less defensible reason was that nobody powerful needed the distribution system opened. In 2000, nothing behind the meter could generate, store, or flex, so a distribution monopoly that owned the wires, planned the network, ran the interconnection process, held the data, and sold the default product seemed harmless. The local utility kept every one of those roles. It holds them today. ## The wrong grid Which is how America arrived at institutions for the wrong grid. The compact assumes the utility must build everything; a 2,061 GW queue of independent developers is waiting to build instead [9]. Cost-of-service ratemaking assumes capital is the scarce input; capital is lined up and connection is scarce. The 1935 seam assumes a clean boundary at the meter; batteries and flexible data centers live on the boundary. Order 888 assumes the gatekeeping problem ends at transmission voltage; the same conflict of interest now operates on every distribution feeder. Each assumption was true when written. The technology evolved beyond them, and the institutions, unlike the technology, have no cost curve pushing them forward. The pattern across both prior settlements is the useful lesson. Each came only after the old arrangement failed visibly and expensively, each was fought by the incumbents of its day, and each was later defended by many of the same interests that fought it. The third settlement, unbundling the distribution platform and making connection a matter of rules at every voltage level, is specified in Paper III. Its timing follows the same pattern: the visible, expensive failure is already here. ## The precedent that actually fits Policy arguments lean on analogies, and most of the popular ones fit this case badly. Airline deregulation removed price and entry controls from a service with no physical network bottleneck, so it says little about a wires monopoly. Telecom local-loop unbundling is closer and mostly a cautionary tale: regulators tried to share one copper network among competitors by administratively pricing its parts, and the regime collapsed into litigation. Neither is the model here. One precedent matches the structure point for point: natural gas restructuring under FERC Orders 436 and 636 (1985-1992) [10]. Gas pipelines had been merchants that bought the commodity, transported it, and sold the bundle, exactly as utilities bundle electrons with delivery. FERC separated the roles: pipelines became open-access carriers paid regulated rates for transportation, the commodity became a competitive market, and storage, the physical feature that makes a commodity warehousable and hedgeable, developed into a traded service. The parallel is structural, because batteries are beginning to give electricity some of the temporal flexibility that storage gave gas, converting part of an instantaneous delivery service into something with a forward curve. The gas transition also previews the hard parts, since it required explicit treatment of stranded contracts and a decade of litigation, and it ended with a liquid national market that nobody proposes to re-bundle. Where the analogy fails is instructive too: electricity moves near the speed of light on a network that must balance every second, which is why this series pairs open access with operating envelopes and obligations instead of copying gas rules verbatim. By first principles, the gas precedent shows the destination is reachable. --- ## Sources [1] Thomas P. Hughes, *Networks of Power: Electrification in Western Society, 1880-1930* (Johns Hopkins, 1983). [2] Richard F. Hirsh, *Power Loss: The Origins of Deregulation and Restructuring in the American Electric Utility System* (MIT Press, 1999). [3] George J. Stigler, "The Theory of Economic Regulation," *Bell Journal of Economics and Management Science* 2:1 (1971). [4] Harvey Averch and Leland L. Johnson, "Behavior of the Firm Under Regulatory Constraint," *American Economic Review* 52:5 (1962). [5] Paul L. Joskow and Richard Schmalensee, *Markets for Power* (MIT Press, 1983). [6] Fred C. Schweppe et al., *Spot Pricing of Electricity* (Kluwer, 1988). [7] William W. Hogan, "Contract Networks for Electric Power Transmission," *Journal of Regulatory Economics* 4 (1992). [8] Severin Borenstein and James Bushnell, "The U.S. Electricity Industry After 20 Years of Restructuring," *Annual Review of Economics* 7 (2015); Frank A. Wolak's CAISO Market Surveillance Committee analyses. [9] Joseph Rand et al., *Queued Up*, Lawrence Berkeley National Laboratory (annual editions). [10] FERC Orders No. 436 (1985) and No. 636 (1992), restructuring interstate natural gas pipelines as open-access carriers; Paul MacAvoy, *The Natural Gas Market* (Yale, 2000). *General-audience companions: Gretchen Bakke, "The Grid" (2016); Forrest McDonald, "Insull" (1962); Julie A. Cohn, "The Grid: Biography of an American Technology" (2017).*