This essay is part II of a series authored by Tanya Abrahamian looking at the expansion of takings following the Supreme Court’s decision in Kelo v. New London. The essays support a reversion to a narrow conception of public use by describing the economic underpinnings of the authority and by countering what it deems to be the primary arguments in favor of its expansion: namely, that eminent domain use for private development is justified because it produces positive externalities and that such use is a necessary solution to the holdout problem. This segment of the series will look at the historical expansion of eminent domain authority that was caused by the erosion of the term “public goods” to mean “public benefit.” The next segment of this ongoing series will analyze the flaws of this expansion of eminent domain authority when the “just compensation” associated with a taking fails to take into account the externalities associated with the government’s use of the land.
II. The History of Eminent Domain: From “Public Goods” to “Public Benefit”
A. The “Public Good”
Eminent domain was initially employed to foster localized development and overcome the under-provision of public goods. In this sense, eminent domain historically arose as a remedy to the public good problem. Public goods are goods that are neither rivalrous nor excludable in consumption. This renders their provision unprofitable and, consequently, such goods will be under-provided barring government intervention in the market. For example, though certain infrastructure may be valued by the community as a whole, there is little incentive for any individual to voluntarily provide private land for its construction since the marginal benefit derived will almost certainly be less than the cost of ceding the required land. Because the good is non-excludable and non-rivalrous, it is not readily valuable, there can be no profits extracted from its provision, and hence no private market will provide it. Consequently, public goods are generally provided by the government because they would otherwise be under-provided. The first uses of eminent domain were to ensure that the land was put to its highest value in cases where the social value of its use for infrastructure development exceeded the private value to the land’s current occupant. Such was the conventional theory justifying the authority, barring the use of development incentives or tax subsidies.
The underlying, but oftentimes overlooked, problem that causes this under-provision of goods is the difficulty of attaching a societal value to the land. As stated above, the under-provision of public goods occurs because there is a discrepancy between the social value to the public of a good–in this case, the land–and its private value. In a negotiation, the government will cap its offer on the land at this societal value, but the problem lies in its inability to determine this value. Because the cap is unknown, the current landowner has an incentive to holdout for some potentially better, but yet unknown, offer. Consequently, eminent domain is a means of overcoming this barrier by subverting this valuation criteria altogether. Its solution to the holdout problem is to render the acceptance of the government’s valuation—equivalent to its “just compensation” determination—mandatory. The failure to find a more apt alternative to eminent domain is perhaps partially due to a failure to understand why it is necessary: inabilities in valuing a social good coupled with the holdout problem.
Whether intentionally or not, the Court’s historical narrow definition of “public use” coincided with this economic justification. For example, in early jurisprudence the Supreme Court recognized that “a law that takes property from A. and gives it to B . . . is against all reason and justice.” The most common use of eminent domain at the time of the nation’s founding was takings for government welfare functions, as opposed to private takings. Up through the nineteenth century, it was applied to facilitate the construction of roads and other communal infrastructure such as mills and dams. Accordingly, under this interpretation, public use required literal use by the public. The definition of the term, therefore, meant either that the property was under government use or that the entire public had a right to use it. This is analogous to stating that the good provided must be non-excludable. Because land use for infrastructure is generally non-rivalrous, given that its use by one individual does not reduce its utility to another, the earliest uses of eminent domain were for the provision of public goods. Early public uses of eminent domain therefore coincided with the definition of an economic public good—one that is both non-excludable and non-rivalrous.
B. Cognitive Dissonance: The History of Eminent Domain
To facilitate the rapid economic development of the Industrial Revolution in the late nineteenth century, the courts began to move towards a broader interpretation of public use. Yet, though the Supreme Court adopted the economic justification of eminent domain as necessary for the provision of public goods, its rulings were incongruous with its justifications. The Court sanctioned a broader definition of public use to mean “public advantage, convenience, or benefit.” The Courts in Clark v. Nash and Strickley v. Highland Boy Mining Co. found that an eminent domain taking was only permissible in “exceptional” cases when “the very foundations of public welfare could not be laid without requiring concessions from individuals to each other upon due compensation which under other circumstances would be left wholly to voluntary consent.” This language appears to loosely coincide with the public goods justification for eminent domain.
The language that eminent domain can only be used when “the very foundations of public welfare could not be laid” corresponds to the idea that eminent domain is necessary for the provision of public goods, but not for the provision of private goods. It suggests that eminent domain is justified when the market would not otherwise function to provide the good or when transactions conducted with “voluntary consent” fail. These transactions fail when there is a market failure justifying some government intervention: either under-provision of a public good or a holdout problem.
Yet, oddly, despite these aforementioned statements, the Strickley Court found constitutional a Utah statute permitting the exercise of eminent domain for “roads, railroads, tramways, tunnels, ditches, flumes, pipes, and dumping places to facilitate the milling, smelting, or other reduction of ores, or the working of mines.” The Court’s holding is incongruous with its rule statement which loosely corresponds to the definition of a public good.
While roads are the conventional example of a public good, the use of the authority for pipelines and railroads (both of which are excludable, hence their provision by private enterprises) does not appear to correspond to their statement that eminent domain is justified in cases where “the very foundation of public welfare c[an]not be laid” because the good cannot be otherwise provided through voluntary transactions. The Court recognizes this carve-out for common carriers when it points to the “inadequacy of use by the general public as a universal test,” a departure from prior eminent domain use. The Court’s reasoning neglects to address the fact that, barring a holdout problem, most railroads and pipelines are private entities whose provision can occur through “voluntary consent.” Consequently, eminent domain is not a necessary prerequisite for their existence as the Court appears to suggest. Even assuming that the Court is only restricting eminent domain to cases where a holdout problem is common, its failure to establish a criteria for identifying a holdout problem established the basis for future expansions of the “public use” definition. Though the Court recognized that eminent domain should only be used when necessary, or when the land is required for the provision of a public good, it has not adhered to this rule in deeming constitutional state statutes conferring such authority.
This same dissonance exists in more modern discourse that attempts to justify a broader interpretation of public use. In articulating a broad position, the Nevada Supreme Court stated that
[I]f public occupation and enjoyment of the object for which land is to be condemned furnishes the only and true test for the right of eminent domain, then the legislature would certainly have the constitutional authority to condemn the lands of any private citizen for the purpose of building hotels and theaters. Why not? . . . The public have the same right, upon payment of fixed compensation, to seek rest and refreshment at a public inn as they have to travel upon a railroad . . . . One purpose is, so far as the legal rights of the citizen are concerned, as public as the other.
While the Nevada Supreme Court correctly analogized between railroads and other private enterprises, such as hotels and theaters, it draws an incorrect conclusion from this comparison. The court incorrectly identified this class as falling under the category of public goods. Common carriers such as railroads and pipelines are not public goods in the economic sense, even though they are a “necessary and natural accompaniment to the emergence of civil society.” The reality is that railroads are a semi-public good with certain characteristics of a public good, namely that they are non-rivalrous. Yet railroads, like hotels and theaters, are excludable and are not an economic public good. In this sense, the court’s evolving definition of public use took a logical path: from an economic public good, to a semi-public good, to ultimately any tangential “public benefit.”
C. The “Public Benefit”
This expansion of eminent domain authority occurred with the conflation of the economic “public good” with public benefit. Under this broad interpretation of public use, a private entity’s attribution of an incidental amorphous public benefit is sufficient to render a use “public.” The U.S. Supreme Court’s ruling in Kelo v. City of New London largely substantiated this transition by permitting the use of eminent domain any time a condemned property has the potential of contributing to the local economy. This decision merely affirmed earlier rulings that established that the authority need only be “rationally related to a conceivable public purpose” and that “beneficiaries of an eminent domain action need not constitute ‘any considerable portion’ of the community.” In this sense, as Gideon Kanner states, the decision sanctioned an “Orwellian” interpretation of the term, in which “clearly private uses of land by redevelopers and their customers qualif[ied] as ‘public’ under the Constitution.” The decision re-interpreted public use from its origins as an economic public good to any private activity that has some positive externality.
The Supreme Court has therefore sanctioned a broad use of eminent domain authority by defining public use to include a use with any tangential public benefit. It justified this expansion by assuming that the same justifications for eminent domain use by the government applied to private entities.
Part III to follow.
 William Treanor, The Origins and Original Significance of the Just Compensation Clause of the Fifth Amendment, 94 Yale L.J. 694-716 (1985) (finding that the first cases of eminent domain use were for the provision of aqueducts and roads); Carrie B. Kerekes & Dean Stansel, Takings and Tax Revenue: Fiscal Impacts of Eminent Domain, Mercatus Working Paper (2014) (finding that the first uses of eminent domain authority were for the government provision of public facilities and infrastructure).
 Paul Samuelson, The Pure Theory of Public Expenditure, 36 The Review of Economics and Statistics 378-9 (1954); Richard Cornes & Todd Sandler, The Theory of Externalities, Public Goods, and Club Goods (1996) (describing a non-rivalrous good as one for which, for any level of production, the cost of providing it to an additional individual is 0); Joseph Stiglitz, Knowledge as a Global Public Good, The World Bank (1998) (describing a non-excludable good as one in which no one can be excluded from its use).
 E.g., Ronald Coase, The Problem of Social Cost, 3 J.L. Econ. 1-44 (1960); Arthur C. Pigou, The Economics of Welfare (1920); Thomas Helbling, Externalities: Prices Do Not Capture Costs, International Monetary Fund (March 28, 2012), http://www.imf.org/external/pubs/ft/fandd/basics/external.htm.
 See Thomas J. Miceli, The Economic Theory of Eminent Domain (2011).
 See Ed Nosal, The Taking of Land: Market Value Compensation Should be Paid, Journal of Public Economics, 431, 432 (2001) (assuming that “government’s expropriation decision is motivated by the maximization of social welfare”); Compare Rick Rybeck, Using Value Capture to Finance Infrastructure and Encourage Compact Development, 8 Public Works Management & Policy, 249, 253 (2004) (stating that land value is determined by the value of public goods and services that are available to particular sites) with Philip E. Graves, Valuing Public Goods, 46 Challenge: The Magazine of Economic Affairs (2003) (demonstrating the difficulties inherent in valuing public goods).
 E.g., Ronald Coase, The Problem of Social Cost, 3 J.L. Econ. 1-44 (1960); Arthur C. Pigou, The Economics of Welfare (1920); Thomas Helbing, Externalities: Prices Do Not Capture Costs, International Monetary Fund (March 28, 2012), http://www.imf.org/external/pubs/ft/fandd/basics/external.htm.
 See Ed Nosal, The Taking of Land: Market Value Compensation Should be Paid, Journal of Public Economics, 431, 432 (2001) (assuming that “government’s expropriation decision is motivated by the maximization of social welfare”).
 Calder v. Bull, 3 U.S. 386, 388 (1798).
 Errol E. Meidinger, The ‘Public Uses’ of Eminent Domain: History and Policy, 11 Envtl. Law 1 (1980); Kohl v. United States, 91 U.S. 367 (1875) (condemning property for construction for public buildings); United States v. Great Falls Manufacturing Co., 112 U.S. 645 (1884) (approving the condemnation of property for the construction of aqueducts to provide cities with drinking water); Carrie B. Kerekes & Dean Stansel, Takings and Tax Revenue: Fiscal Impacts of Eminent Domain, Mercatus Working Paper (2014) (finding that the first uses of eminent domain authority were for the government provision of public facilities and infrastructure).
 Errol E. Meidinger, The ‘Public Uses’ of Eminent Domain: History and Policy, 11 Envtl. Law 1 (1980).
 Errol E. Meidinger, The ‘Public Uses’ of Eminent Domain: History and Policy, 11 Envtl. Law 1 (1980) (citing Bloodgood v. Mowhawk & Hudson R.R., 18 Wend. 9, 52-62 (N.Y. 1837)); Sadler v. Langhorn, 34 Ala. 311, 333 (1859); Charles E. Cohen, Eminent Domain After Kelo v. City of New London: An Argument for Banning Economic Development Takings, 29 Harv. J.L. & Pub. Pol’y 491, 505-6 (2006).
 Stephen J. Jones, Trumping Eminent Domain Law: An Argument for Strict Scrutiny Analysis Under the Public Use Requirement of the Fifth Amendment, 50 Syracuse L. Rev. 285, 292 (2000) (referencing Lawrence Berger, The Public Use Requirement in Eminent Domain, 57 Or. L. Rev. 203, 208 (1978)).
 Strickley v. Highland Boy Mining Co., 200 U.S. 527, 531 (1906).
 Utah, Rev. Stat. 1898, § 3588
 Strickley, 200 U.S. at 531.
 Id. (referencing the court’s decision in Clark v. Nash, 198 U.S. 361 (1905)).
 Dayton Gold & Silver Mining Co. v. Seawell, 11 Nev. 394, 410 (1876).
 E.g., Adam Gopnik, The Plot Against Trains, The New Yorker (May 15, 2015), http://www.newyorker.com/news/daily-comment/the-plot-against-trains.
 Kelo v. New London, 545 U.S. 469 (2005).
 Hawaii Housing Authority v. Midkiff, 467 U.S. 229, 241 (1984); Nadia E. Nedzel et al., Eminent Domain: A Legal and Economic Critique, 7 U.Md. L.J. Race, Relig., Gender & Class 140 (2007) (citing Hawaii Housing Authority v. Midkiff, 467 U.S. 229, 244 (1984)).
 Gideon Kanner, Kelo v. New London- Brave New World or Return of the Robber Barons? (2005).
 See Thomas J. Miceli, The Economic Theory of Eminent Domain (2011).
 Doug Campbell, The Economics of Eminent Doman, Region Focus (Fall 2005), https://www.richmondfed.org/~/media/richmondfedorg/publications/research/region_focus/2005/fall/pdf/feature3.pdf.
 E.g., City of Buffalo v. J.W. Clement Co., Inc., 28 N.Y.2d 241, 258 (1971); Rose v. State of New York, 24 N.Y.2d 80, 87 (1969); Marraro v. State of New York, 12 N.Y.2d 285, 292-293 (1963).
 Thomas J. Miceli, Free Riders, Holdouts, and Public Use: A Tale of Two Externalities, Economics Working Papers (2009) (referencing the Supreme Court’s decision in Kelo).