Government-owned property is exempt from local taxes almost everywhere in the United States, but this situation is based less on logic than on now-outdated historical considerations. Remarkably, there are no comprehensive estimates of the value of these exemptions. For comparison, the value of property tax exemptions for nonprofit institutions (excluding houses of worship) was about $900 billion in 1997, and charitable properties (including hospitals and universities) accounted for about $500 billion of this figure (Cordes, Gantz and Pollak 2002, 89). Even without comprehensive data, it is clear that the amount of government-owned land is vastly greater than nonprofit holdings. However, the exempt status of government land barely provokes complaint (except in the western states where federal landholdings are enormous) whereas exemptions for nonprofit organizations are frequently challenged.

Historical Background and Federalism Today

Government-owned property traditionally has been exempt from taxation in order to avoid an empty ritual whereby the sovereign taxed itself. The implicit assumption of a single sovereign was quite reasonable in Elizabethan England, where the property tax first took root, but not so in the U.S. today. The myriad school districts and special districts that now compete with counties and municipalities for property tax revenues were virtually nonexistent in the nineteenth century. Today there is no economic reason to exclude all government property from the tax base.

Exemptions for private, nonprofit entities grew out of the government exemption. In the seventeenth century, private parties did not always wait for the Crown to repair their bridges, causeways, seawalls or highways. They assumed this responsibility whenever self-interest required and the purse permitted. The capital-intensive nature of such activities that relieved government of a burden made a property tax exemption a logical tool for encouraging private initiative. Thus the first charitable exemptions were a type of quasi-government exemption, subsidizing private parties who discharged public responsibilities.

Charitable exemptions for the alleviation of poverty began as a separate category, because reducing poverty was not originally considered a government responsibility. The change in this attitude over time had the effect of diminishing the distinction between alleviating poverty and relieving government of a burden, but these remain two separate bases for the charitable exemption. Before the New Deal of the 1930s, U.S. counties had the primary governmental responsibility for poor relief, through maintaining almshouses and work farms. The principal public expenditure required for them was for land and construction, since the residents did the day-to-day work of running these facilities. In this situation, a property tax exemption made sense. If a charitable organization did not build such a facility, the responsibility would fall to county government and would be funded through property taxes. It was easy to see a clear and convincing connection between the alleviation of poverty, relief of a government burden and a property tax exemption.

Modern U.S. federalism has undermined these connections. There is no single sovereign now, but rather 87,000 units of government, including 19,000 municipalities, 16,600 townships and towns, 3,000 counties, 13,700 school districts and 34,700 special districts, which often overlap in complex ways. The property tax is virtually the sole source of internally generated revenues for school districts and special districts. A government exemption can be administered so that no unit of government need pay taxes to itself, while taxpayers outside the taxing jurisdiction who benefit from the property would pay the tax.

Valuation of unique government property and infrastructure is a problem, but it is not insurmountable. A new addition to generally accepted accounting principles requires local governments to carry on their balance sheets the depreciated value of their physical assets, including infrastructure, which can be a starting point for valuing such property. Already local government property is taxable in 11 states, provided it lies outside the owner’s boundary. For example, a reservoir owned by a water district can be taxed by the town or county where the reservoir is located, and the tax can be collected through increased water rates charged to the utility’s customers.

The strong consensus in favor of exempting government property is due to inertia, power and precaution. The federal government has vast landholdings, collects no property taxes, and therefore would oppose any tax on government property. Besides, the Constitution shields it. State governments also have extensive holdings and do not benefit from property taxes to any significant degree, so they too would oppose taxing government property. Local governments, special districts and school districts would be the net beneficiaries if government property were taxed, since their own property holdings are small in comparison to federal and state governments, yet the property tax provides almost 40 percent of their revenue (U.S. Census Bureau 1998).

Charitable Exemptions as Sovereign Exemptions

As long as government property is exempt, the case for charities is strengthened. Evelyn Brody (1998; 2002) argues that the states, by conferring benefits of sovereignty on nonprofit institutions, are acknowledging the underlying independent, self-governing nature of those institutions. “Tax exemption carries with it a sense of leaving the nonprofit sector inviolate, and the very concept of sovereignty embodies the independent power to govern” (Brody 1998, 588). Under federal tax law, neither charitable institutions nor local governments are taxed on net income, contributions or interest income from bonds, but both are taxed for payments made for services rendered. Considering charitable nonprofit institutions as quasi-sovereign allows us to make sense of “the rules in the tax scheme that operate to curtail rather than enhance the economic strength of the charitable sector. After all, rival sovereigns rarely feel comfortable letting the other grow too powerful” (Brody 1998, 586).

The U.S. Supreme Court, in Walz v Tax Commissioners, 397 U.S. 664 (1969), supports the position taken by Brody: “[Exemption] restricts the fiscal relationship between church and state, and tends to complement and reinforce the desired separation insulating each from the other (emphasis added).” Churches, and by extension other nonprofit institutions, are sovereigns in their own domain, which is circumscribed by a higher sovereign—state government.

Conversely, arguments used to attack certain charitable exemptions can also be applied to the governmental exemption. Critics of nonprofit tax exemption focus on large, property-rich and financially strong organizations, calling them commercial enterprises (Balk 1971; Hyman 1990; Gaul and Borowski 1993). This category includes colleges, universities, hospitals and nursing homes. No state prohibits charities from engaging in commercial activities, but 8 states out of 43 responding to the survey described below prohibit charities from earning a profit, even for institutional purposes. All states prohibit the charitable owner of exempt property from distributing profit to private parties. “It is a well-established principle of law that a charitable institution does not lose its charitable character and its consequent exemption from taxation merely because recipients of its benefits who are able to pay are required to do so, as long as funds derived in this manner are devoted to the charitable purposes of the institution” (American Jurisprudence 1944).

Commercial enterprises of local government are generally tax exempt, including air and marine ports, electric power generating facilities, water treatment and distribution plants, golf courses, package liquor stores and parking garages, to name a few. If commercial activity is to be the test for taxation, this should be applied evenhandedly and extend to government property as well.

A Survey of State Charitable Exemptions

Every state exempts charitable property, but the meaning of “charitable” varies quite a lot because its legal antecedents are traceable to the English Statute of Charitable Uses of 1601. Policy makers have shown considerable ingenuity in adapting an ancient law to modern needs, and ingenuity breeds variety. A Lincoln Institute-sponsored survey explored the laws in each of the 50 states to clarify the definition and application of “charitable” property tax exemptions.

As befitting a sovereign, private nonprofit institutions enjoy a constitutionally protected tax exemption in almost as many states as do local governments. The constitutions of 38 states make reference to exemption of local government or private institutions, or both. States have probably been reluctant to define charity statutorily because the judicial branch is the final arbiter of constitutional matters. Four states authorize legislatures to grant exemptions without giving specific direction; only 9 (including all 6 New England states) are silent. Specific exemptions are mandated in 27 states, and are discretionary in 16. Arizona, Missouri, Nebraska, North Carolina and Virginia are in both categories because they mandate some exemptions (usually governmental) but give their legislatures discretion with respect to other classes of institutional property.

Only 10 states have statutory definitions, and they show very little similarity (see Figure 1). Four of them define charity in terms of a public benefit, two in terms of relieving government of a burden, and one (Florida) could be placed in either category. Other individual states define charity in terms of relief of poverty or deriving income in the form of donations, or simply by listing exemption-eligible activities, with a slight overlap with relief of poverty. Five state definitions (Florida, Nebraska, New Hampshire, North Carolina and Pennsylvania) are extremely broad, which essentially punts the issue to the judicial branch.

The lack of a discernable pattern in judicial opinions arouses suspicion that courts must work backwards from a desired result to develop standards and tests. The situation today parallels the first half of the twentieth century, when bureaucrats and judges were gatekeepers to the nonprofit sector, approving or denying a petition for a nonprofit corporate charter, and they “used their control to promote the causes they believed in” (Silber 2001, 6). Awarding a nonprofit charter is now a ministerial act, but property tax exemption for charitable purposes remains subject to a variety of state laws with idiosyncratic judicial interpretations in every state. Confusion in the public debate over the charitable property tax exemption is the sure result. In devising tests, courts sometimes conflate public benefit with relief of poverty, and the result is unenforceable. Either one or the other must take precedence. Unless statutes are clear, courts are free to choose and to switch back and forth.

The case of hospitals is illustrative. Although one will find exempt hospitals in every state, the law is ambivalent. Hospitals have constitutional protection in only 3 states, while in 17 they are exempt only because the court regards them as “institutions of purely public charity.” The famous 1985 decisions in the supreme courts of Utah and Pennsylvania that undermined hospital tax exemption were health care cases. The courts concluded that the hospital (Utah) and the consortium of hospitals (Pennsylvania) were not in fact charities. Without putting too fine a point on it, the judicial remedies were based on the principle of relieving poverty.

Much angst and legal conflict could be averted if relief of poverty could be treated as separate and distinct from public benefit and relieving government of a burden, and fortunately it can be quantified. If a legislature wants a particular type of institution (e.g., hospitals) to relieve poverty, then the state should tax the hospitals, but award each property owner in the group a tax credit equal to the amount of service they give away up to their tax liability. This proposal raises the thorny question of how to measure the value of services priced below market, but the problems are surmountable (see Bowman [1999] for a method for hospital services). Solutions to these complexities are not likely to introduce the element of arbitrariness that pervades judicial decisions today.

H. Woods Bowman is associate professor in the Public Services Program at DePaul University in Chicago, Illinois. He was a visiting fellow at the Lincoln Institute in 2001 and he contributed to the Urban Institute book, Property Tax Exemption for Charities, edited by Evelyn Brody (2002).

References

American Jurisprudence. 1944. Taxation 51 § 602.

Balk, Alfred. 1971. The Free List: Property Without Taxes. New York: The Russell Sage Foundation.

Bowman, Woods. 1999. Buying charity care with property tax exemptions. Journal of Policy Analysis and Management vol. 18, no. 1 (winter): 120–125.

Brody, Evelyn. 1998. Of sovereignty and subsidy: Conceptualizing the charity tax exemption. Journal of Corporation Law vol. 23, no. 4 (summer): 585–629.

_____. 2002. Legal theories of tax exemption, quasi and real. In The Property Tax Exemption: Mapping the Battlefield, Evelyn Brody, ed., 145–172. Washington, DC: Urban Institute.

Cordes, Joseph J., Marie Gantz, and Thomas Pollak. 2002. What is the property-tax exemption worth? In The Property Tax Exemption: Mapping the Battlefield, Evelyn Brody, ed., 81–112. Washington, DC: Urban Institute.

Gaul, Gilbert and Neill A. Borowski. 1993. Free Ride: The Tax-Exempt Economy. Kansas City: Andrews McMeel.

Hyman, David A. 1990. The conundrum of charitability: Reassessing tax exemption for hospitals. American Journal of Law and Medicine vol. 6, no. 3: 327–380.

Silber, Norman I. 2001. A corporate form of freedom: The emergence of the nonprofit sector. Boulder, CO: Westview Press.

U.S. Census Bureau, U.S. Department of Commerce. 1998. Statistical Abstract of the United States 1998, table 500 (reporting 1995 data).

Figure 1: Statutory Criteria for Charitable Organizations

Arizona requires “qualifying charitable organizations” to spend at least 50 percent of their budgets on services to state residents who receive “temporary assistance to needy families benefits or low income residents…and their households” [A.R.S. § 43-1088 G(2)].

In Florida, “Charitable purpose means a function or service which, if discontinued, could legally result in the allocation of public funds for the continuance of the function or service. It is not necessary that public funds [actually] be allocated, but only that such allocation is legal” [F.S. §196.012]. Houses of worship are exempt under a separate statute.

Hawaii defines charitable purposes as “community, character building, social service, or educational nature, including museums, libraries, art academies, and senior citizens housing facilities qualifying for a loan under the laws of the United States” [H.C.A. § 246-32(c)(2)].

In Montana charities must accomplish their activities “through absolute gratuity or grants” [M.C.A. § 15-6-201(2)(a)(i)].

In Nebraska charities must operate “exclusively for the purposes of the mental, social, or physical benefit of the public or an indefinite number of persons” [R.S.N.A. § 77-202(1)(d)].

A New Hampshire charity is one that performs “some service of public good or welfare advancing the spiritual, physical, intellectual, social or economic well-being of the general public or a substantial and indefinite segment of the general public that includes residents of the state of New Hampshire…” [R.A. § 72:23-1].

In North Carolina, “A charitable purpose is one that has humane and philanthropic objectives; it is an activity that benefits humanity or a significant rather than a limited segment of the community without the expectation of pecuniary profit or reward. The humane treatment of animals is also a charitable purpose” [N.C. Gen. Stat. § 105-278.3(d)(2)].

Pennsylvania requires: (1) relief of poverty; (2) advancement and provision of education, including secondary education; (3) advancement of religion; (4) prevention of treatment of disease or injury, including mental retardation and mental disorders; (5) government or municipal purposes; or (6) accomplishment of a purpose that is recognized as important and beneficial to the public and that advances social, moral, or physical objectives” [10 Penn. Stats. § 372].

A South Dakota public charity “must receive a majority of its revenue from donations, public funds, membership fees, or program fees generated solely to cover operating expenses; it must lessen a government burden by providing its services to people who would otherwise use government services; it must offer its services to people regardless of their ability to pay for such services…” [S.D.C.L. § 10-4-9.1].

Texas defines charity by reference to the type of activity such an organization undertakes. T.T.C. § 11(d) lists 19 activities, including: (d)(1) “providing medical care without regard to the beneficiaries’ ability to pay…”

La información sobre los valores de las casas es fundamental para los investigadores y formadores de políticas interesados en el establecimiento de políticas públicas bien informadas en temas tales como tributación e inversiones de infraestructura. Los métodos de ventas repetidas, como el índice de S&P/Case-Shiller, que se usan comúnmente en los Estados Unidos para estudiar la dinámica de los precios de las casas, no son prácticos en los países en vías de desarrollo, donde los precios declarados de las transacciones de las casas frecuentemente son menores a su valor real, por razones diversas.

Los propietarios de casas también necesitan una referencia confiable para medir el valor de una casa para ayudar en la toma de decisiones personales tales como la jubilación (Lusardi y Mitchell, 2007), el consumo (Campbell y Cocco, 2007), los ahorros (Juster et al. 2005) y la composición de la deuda de la unidad familiar (Disney, Bridges y Gathergood 2006). como para dictar políticas públicas. Se ha demostrado que el valor de una casa es una variable clave en las decisiones de un propietario, por ejemplo en materia de jubilación

Hemos evaluado la fiabilidad de los datos de encuestas de hogares sobre las estimaciones del valor de las casas en México efectuadas por sus propietarios. Argumentamos que, en promedio, las estimaciones de los propietarios de corto plazo tienden a ser razonablemente objetivas y precisas. Las estimaciones de los propietarios sobre el precio de mercado de una propiedad, obtenidas por medio de estas encuestas, pueden ser la manera más conveniente y confiable de efectuar el seguimiento de los precios de las casas bajo ciertas circunstancias.

La valuación de las casas en los Estados Unidos

Todas las encuestas de hogares principales en los Estados Unidos – el censo decenal, el Estudio de Panel sobre Dinámicas de Ingresos (Panel Study of Income Dynamics) la Encuesta de Hogares de los Estados Unidos (American Housing Survey) y la Encuesta de Finanzas del Consumidor (Survey of Consumer Finances)hacen preguntas tales como: “¿Cuánto vale esta propiedad? Es decir, ¿por cuánto cree usted que se vendería esta propiedad si se pusiera a la venta?” El argumento principal a favor de usar este tipo de pregunta sobre la valuación de una casa es que es fácil de recolectar los datos. Pero también es crucial evaluar la fiabilidad de estas valuaciones autoestimadas de las casas.

En Kish y Lansing (1954), se les pidió a propietarios en ciudades estadounidenses que estimaran el valor de mercado de sus casas, y las mismas estimaciones fueron efectuadas más tarde por valuadores profesionales. La conclusión principal fue que el sesgo promedio en las estimaciones de la gente oscilaba alrededor de cero. Es decir, si bien algunas estimaciones individuales podían ser bastante distintas del valor asignado por el valuador, los errores parecían cancelarse al hacer el promedio. Esta fue una conclusión importante para justificar el uso de esta pregunta en grandes encuestas. Cuando los investigadores se concentraron en distintos subgrupos, descubrieron que las estimaciones más precisas del valor de la casa eran las efectuadas por propietarios recientes. No aumentaba la exactitud si el encuestado era jefe de hogar o tenía más educación, ni si el valuador había podido entrar a la propiedad para hacer la valuación.

Usando los mismos métodos y datos similares, Kain y Quigley (1972) confirmaron que los errores en gran medida se compensaban, pero que estaban correlacionados con el nivel socioeconómico del encuestado. Una mayor educación estaba asociada a un menor sesgo positivo en la estimación del propietario. Esta investigación también trató de determinar por qué algunos encuestados no respondían a esta pregunta, y descubrió que aquellos con mayor nivel de ingresos y educación pero menor tiempo de tenencia tenían una mayor probabilidad de proporcionar una estimación del valor de su casa.

Dado que la estimación ideal del precio de mercado de una casa es el precio de venta más reciente, algunos estudios han comparado los datos de venta de transacciones recientes con las estimaciones de los propietarios. Goodman e Ittner (1992), por ejemplo, compararon las estimaciones de los propietarios con los precios de venta subsiguientes de la misma propiedad usando la Encuesta de Hogares de los Estados Unidos de 1985 y 1987. Encontraron que el propietario promedio en los EE.UU. sobreestima el valor de su casa en un 6 por ciento por encima del precio de venta real, y que el error absoluto promedio es de alrededor del 14 por ciento. El error no está relacionado en su mayor parte con las características del propietario, de la casa o del mercado local.

Otro enfoque consiste en comparar las valuaciones tributarias con las autoestimaciones de los propietarios (David, 1968). El problema obvio de usar las valuaciones tributarias es que éstas pueden no estar siempre actualizadas para reflejar las condiciones actuales del mercado, causando una impresión fallida del valor de la vivienda. En general, estos estudios concluyeron que, en promedio, los propietarios en los E.U.A. tienden a sobreestimar el valor de sus casas en alrededor del 5 por ciento. Esta sobrevaluación no está relacionada con las características del propietario o de la casa, sino con la duración de la tenencia de la casa. Llegamos entonces a la conclusión de que se pueden usar las encuestas en forma fiable para obtener, a muy bajo costo, estimaciones razonables de la valuación de las casas en los mercados de vivienda de los EE.UU.

Mercados de vivienda en países en vías de desarrollo

En los países desarrollados, el acceso al suelo ocurre principalmente por medio de compras formales, mientras que en los países en vías de desarrollo es bastante común que una porción sustancial del crecimiento urbano ocurra en terrenos ocupados informalmente, especialmente por parte de los grupos de bajos ingresos. Los habitantes se pueden organizar para invadir tierras estatales, áreas protegidas y hasta propiedades privadas. Cuando los tribunales hayan establecido la ilegalidad de tales actos, muchas veces ya habrá algún político a quien le conviene brindar protección y servicios a los invasores a cambio de votos y respaldo político, en vez de desalojarlos de los suelos ocupados ilegalmente.

En algunos casos, los gobiernos locales expropian la tierra para después transferirla a sus electores. Aquellos que adquieren su propiedad en estas condiciones tienen más dificultad para determinar el valor monetario de la propiedad, ya que no pagaron inicialmente por ella. Y en ocasiones, la propiedad no se puede vender fácilmente, ya que no tiene un título válido.

Otra característica distintiva de los mercados de vivienda en los países en vías de desarrollo es la gran proporción de casas construidas por sus propietarios. La preponderancia de viviendas autoconstruidas, en vez de ser erigidas por compañías de construcción especializadas, ocurre porque el sistema financiero está subdesarrollado y los préstamos hipotecarios no existen o son muy caros. Las familias adquieren una vivienda construyéndola con sus propias manos durante un período prolongado de tiempo.

La falta de desarrollo de los mercados hipotecarios puede obligar a las familias a usar métodos de construcción ineficientes, porque la casa se construye por etapas, usando los materiales de construcción que ellos puedan solventar y tengan a su disposición en ese momento. Estas restricciones pueden dar lugar a costos de construcción innecesariamente altos, así como también a prácticas de construcción no reguladas e ineficientes. Si la gente estima el valor de su casa como la suma de los gastos incurridos para construirla, aquellos propietarios que se encuentran en zonas donde no haya acceso a los préstamos hipotecarios probablemente avalúen sus casas autoconstruidas a un valor más alto.

La autoconstrucción también puede influir el acceso a información sobre el valor actual de mercado de dicha propiedad. Los proyectos inmobiliarios desarrollados por compañías de construcción generalmente tienen un mayor grado de homogeneidad, de manera que las ventas en el barrio generan información sobre el valor actual de las propiedades circundantes. Cuando una casa es autoconstruida, este canal de información desaparece, porque las casas construidas de esta manera no representan adecuadamente el valor de otras casas en el barrio.

Jiménez (1982) ha proporcionado uno de los pocos estudios sobre los valores de las casas en un país en vías de desarrollo. Usando datos de un barrio pobre en las Filipinas, encontró que los valores medios de las estimaciones respectivas del propietario y del valuador no se pueden distinguir estadísticamente. Su muestra de las Filipinas se compara bien con los resultados de Kain y Quigley (1972) para St. Louis, Missouri, en cuanto a las diferencias de las valuaciones promedio. No obstante, para estimaciones individuales, Jiménez encontró que el valor absoluto promedio de las diferencias entre la valuación del propietario y la del valuador es aproximadamente un 55 por ciento de la valuación media, mientras que la cifra comparable para Kain y Quigley es de aproximadamente el 20 por ciento.

Encuestas de hogares y datos de valuación

Nuestro análisis usa datos de una encuesta de hogares y valuaciones de las mismas casas efectuadas por un valuador que también es corredor inmobiliario. La encuesta, realizada entre mediados de febrero y mediados de marzo de 2006, incluyó aproximadamente 1.200 viviendas en las afueras de la ciudad de Acayucán, en el estado de Veracruz, en el centro de México. Las casas de la muestra se encontraban en los distritos más pobres de la ciudad, donde las calles no estaban pavimentadas y muchas casas no contaban con servicios vitales de agua y alcantarillado. La casa promedio de la muestra tenía 2,5 cuartos. Sólo el 63 por ciento de las viviendas tenían un baño interno, y el 60 por ciento tenían un techo de placas metálicas, amianto u hojas de palma. Además, el 12 por ciento de los propietarios reportaron que no tenían un título de propiedad.

Las familias entrevistadas fueron propietarios-ocupantes, no inquilinos, y la mayoría vivía en viviendas pequeñas de un solo piso, en lotes bien demarcados. Una de las preguntas de la encuesta fue: “¿ Por aproximadamente cuánto dinero cree usted que se vendería esta casa actualmente?” El propietario promedio estimó que el valor de su casa era $19.948, mientras que la valuación promedio fue de sólo $12.123 (en dólares estadounidenses de 2006). La diferencia de las medianas, si bien mucho menor ($1.545) y la diferencia del logaritmo de las medias también son significativas en ambos casos. (Para una descripción detallada de la encuesta, véase González-Navarro y Quintana-Domeque [2007, 2008].)

La otra fuente de datos son las valuaciones de las casas efectuadas por un valuador profesional y agente de bienes raíces. Al hacer que una sola persona realice todas las valuaciones, se reducen los riesgos de decisiones subjetivas y el uso de distintas prácticas de valuación. El valuador visitó una de cada dos casas donde sus residentes habían sido entrevistados, y las valuaciones se realizaron a más tardar dos meses después de la encuesta de hogares, reduciendo así la preocupación sobre la posibilidad de inflación o volatilidad en los precios de las casas.

En las encuestas completadas, la tasa de respuesta para la pregunta sobre la estimación del propietario del valor de la casa fue de aproximadamente el 74 por ciento. Una de las ventajas importantes del procedimiento en dos etapas del estudio es la posibilidad de investigar las valuaciones de las casas del 26 por ciento de los encuestados que no respondieron a la pregunta sobre el valor de su casa. Por lo tanto, a diferencia de cualquier estudio anterior, podemos investigar si la falta de respuesta a la pregunta por parte del propietario está relacionada con el valor de la casa medido por el valuador.

Encontramos evidencia de que la valuación promedio de una casa es igual tanto para los encuestados que proporcionaron una estimación del valor de su casa como para los que no lo hicieron. Además, la edad, sexo, ser o no jefe de familia, y poseer o no un título y tenencia de la propiedad no estuvieron correlacionados con la probabilidad de respuesta. Esto sugiere que los propietarios que no proporcionaron una estimación del valor de su casa son un subconjunto aleatorio de la muestra. Esta es una conclusión importante si se van a utilizar las valuaciones de los propietarios en otros estudios para estimar los precios promedio de las casas en países en vías de desarrollo.

Suponemos que el valor determinado por el valuador es muy cercano al valor de mercado de la casa. También creemos que es razonable interpretar que la discrepancia entre los valores de las casas obtenidos de los propietarios y los del valuador se debe a percepciones equivocadas sobre el valor de mercado por parte del propietario. Hay varias razones que justifican dicha interpretación. Primero, el valuador hace una estimación más precisa que los propietarios sobre el tamaño del lote (uno de los determinantes más importantes del valor de una casa). Segundo, el valuador parece inferir otras características de la vivienda con precisión. Tercero, es más probable que el valuador tenga conocimiento de los factores de mercado que influyen en la valuación de las casas en la ciudad.

La relación entre las estimaciones del propietario y el valuador

La tabla 1 (ver anexo) muestra el grado de error promedio y la falta de precisión o exactitud en las estimaciones de los propietarios para varias submuestras. Como en estudios previos, los resultados se muestran para distintas medidas de sesgo (la diferencia entre el valor de la casa estimado por el propietario y el valuador, y la diferencia porcentual con respecto a la estimación del valuador) e inexactitud (la diferencia absoluta y la diferencia porcentual absoluta).

Considerando a todos los propietarios de la muestra, la diferencia promedio entre la estimación del propietario y la valuación es de alrededor de $7.800, indicando que los propietarios tienden a sobreestimar el valor de sus casas. La diferencia porcentual de las medias es el 124% de la valuación. En cuanto a la falta de exactitud o precisión, la diferencia absoluta de las medias es aproximadamente de $13,500, reflejando cuán distintas son las estimaciones del valuador de las del propietario. En promedio, los propietarios de la muestra tienen una expectativa demasiado alta del valor de su casa.

Estos resultados se contraponen a la evidencia disponible para los Estados Unidos y las Filipinas. Tanto Kain y Quigley (1972) en St.Louis y Jiménez (1982) en las Filipinas reportan una diferencia porcentual de las medias de menos del 0, 5 por ciento. En términos de precisión, encontramos también que los resultados son muy distintos. En nuestra muestra, la diferencia porcentual absoluta se estima en más del 150 por ciento, mientras que en las Filipinas es de aproximadamente el 55 por ciento y en St. Louis el 20 por ciento. Por otro lado, los resultados de error e inexactitud para propietarios con tenencia reciente (menos de dos años) son estadísticamente cercanos a cero. El mismo resultado se obtiene para el error porcentual de la media, y el error porcentual absoluto se reduce en más del 50 por ciento.

Como se mencionó anteriormente, una de las diferencias entre los mercados de viviendas desarrollados y subdesarrollados es la falta de construcción de viviendas por firmas especializadas y la falta de información acerca de la distribución de los precios de las viviendas en estos últimos. La última columna de la tabla 1 identifica el conjunto de casas construidas, en vez de autoconstruidas. Para este subgrupo, el error medio y el error porcentual medio no son estadísticamente distintos de cero, mientras que el error porcentual absoluto (33 por ciento) es el menor de los cuatro grupos. Esto puede sugerir que los propietarios de casas autoconstruidas proporcionan estimaciones con un sesgo positivo porque la autoconstrucción gradual a lo largo del tiempo es más cara que comprar una casa ya construida.

Determinantes del sesgo individual y la inexactitud

Los resultados sugieren que el sesgo e inexactitud en las estimaciones de los propietarios se pueden atribuir a la tenencia a largo plazo. En González-Navarro y Quintana-Domeque (2009), demostramos que ni la discrepancia en la estimación del tamaño del lote ni las características socioeconómicas del encuestado parece estar correlacionadas con el error o la falta de precisión en las estimaciones de los propietarios. Sólo la tenencia está correlacionada significativamente con el sesgo y la falta de precisión.

Dados estos resultados, estimamos a continuación los valores promedio de las casas a nivel de distritos censales para estudiar el desempeño del valor autoestimado en función de la longitud de la tenencia. Como argumentamos anteriormente, el valor autoestimado por los propietarios de tenencia breve proporciona una estimación más exacta del valor promedio de las casas. Una cuestión importante al incluir solamente a los propietarios recientes es que el tamaño de la muestra es chico. La tabla 2 (ver anexo) muestra que las dos medidas de sesgo (la diferencia de las medias y la diferencia porcentual de las medias) para propietarios de tenencia breve proporcionan una estimación menos distorsionada del valor medio de las casas, como también las medidas de diferencia absoluta de las medias y la diferencia porcentual absoluta de las medias.

Si bien hemos hecho todo el esfuerzo posible por obtener precios de mercado para confirmar las valuaciones profesionales, esta búsqueda resultó esquiva. Este hecho resalta la importancia de evaluar la fiabilidad de los valores autoestimados de las casas en los países en vías de desarrollo, donde los valores de mercado estimados por los propietarios parecen ser la medida más accesible del valor de las casas.

Conclusión

En nuestra muestra, el sesgo de valuación asociado con una tenencia más prolongada es positivo, confirmando los resultados de varios estudios en los EE.UU. efectuados con casas recientemente vendidas. Nuestra conclusión principal es que el sesgo debido al período de tenencia es potencialmente mucho mayor en el contexto de un país en vías de desarrollo. Los propietarios de tenencia prolongada sobreestiman, en su mayoría, el valor de sus casas, con un error porcentual absoluto de la media del orden del 150 por ciento. Por otro lado, las familias con una tenencia de dos años o menos realizan estimaciones razonablemente exactas y objetivas del valor de sus casas. Un grupo de casas similares construidas por una firma de construcción especializada muestra un sesgo nulo y estimaciones mucho más precisas que las de otros subgrupos.

Encontramos que la falta de respuesta a la pregunta sobre el valor de la casa no está correlacionada con su valuación y otras características demográficas del propietario. Esto sugiere que se pueden obtener estimaciones objetivas del valor promedio de grupos de casas por medio de encuestas de hogares. Además, el sesgo y la inexactitud no tienen una correlación pronunciada con las características socioeconómicas, como los ingresos familiares o el nivel de educación del encuestado.

En resumen, los resultados de este estudio advierten contra el uso de las estimaciones de los propietarios para hacer un análisis de conducta individual, pero sugieren que estas estimaciones se pueden usar como una aproximación razonable de los valores medios de grupos de casas. Si el objetivo es estimar el valor medio de la casa, las respuestas de los propietarios con un periodo corto de residencia se podrían usar exitosamente en encuestas futuras realizadas en países en vías de desarrollo.

Sobre los autores

Marco González-Navarro es Robert Wood Johnson Scholar de Investigación sobre Políticas de Salud en la Universidad de California, Berkeley. Recibió su doctorado en economía de la Universidad de Princeton.

Climent Quintana-Domeque es profesor asistente del Departamento de Economía de la Universidad de Alicante, España. Recibió su doctorado en economía de la Universidad de Princeton.

Referencias

Campbell, J.Y. and Cocco, J.F. 2007. How do house prices affect consumption? Evidence from micro data. Journal of Monetary Economics 54(3): 591–621.

David, E. L. 1968. The use of assessed data to approximate sales values of recreational Property. Land Economics 44(1): 127–129.

Disney, R., Bridges, S. and Gathergood, J. 2006. Housing wealth and household indebtness: Is there a household ‘Financial Accelerator’? Manuscrito no publicado. Nottingham, UK: University of Nottingham.

Gonzalez-Navarro, M. and Quintana-Domeque, C. 2007. Description of the Acayucan Standards of Living Survey. Unpublished manuscript. Princeton, NJ: Princeton University.

———. 2009. The reliability of self–reported home values in a developing country context. Journal of Housing Economics 18(4): 311–324.

Goodman, J.L. and Ittner, J.B. 1992. The accuracy of homeowners’ estimates of house value. Journal of Housing Economics 2(4): 339–357.

Jimenez, E. (1982) The value of squatter dwellings in developing countries. Economic Development and Cultural Change 30(4): 739–752.

Juster, F.T., Lupton, J.P., Smith, J.P., and Stafford F. 2005. The decline in household saving and the wealth effect. Review of Economics and Statistics 87(4): 20–27.

Kain, J.F. and Quigley, J.M. (1972) Note on owner’s estimate of housing value. Journal of the American Statistical Association 67(340): 803–806.

Kish, L. and Lansing, J.B. 1954. Response errors in estimating the value of homes. Journal of the American Statistical Association 49(267): 520–538.

Lusardi, A. and Mitchell, O.S. 2007. Baby boomers retirement security: The role of planning, financial literacy and housing wealth. Journal of Monetary Economics 54(1): 205–224.

A large share of national energy consumption takes place in cities—in the United States about three-quarters of energy use is in or related to urban areas. Accordingly, cities offer significant opportunities for energy savings from increased efficiency, but important issues remain: Will market forces produce efficiency gains when appropriate, or will market failures such as imperfect information, unavailable financing, or misunderstood risks impede market solutions? How much do people value energy savings, and how sensitive are they to changes in energy prices? The Lincoln Institute hosted a conference on energy efficiency and cities in October 2012 to address these and related issues, and a few highlights follow.

Valuing Energy Efficiency

Consumers should be willing to pay more for built space that uses less energy. Evidence indicates that users of commercial space value energy efficiency and are willing to pay more for it, and many studies indicate that LEED-certified office and commercial space sells or rents at a premium over traditional space. There is much less evidence of such preferences for residences, in part because it is difficult for most homebuyers to determine the energy efficiency of a dwelling, especially a new one with no operating record.

Some residential developments are now being classified using procedures similar to LEED certification or to the Energy Star ratings such as those used for major appliances. Dwellings in California that have the highest energy efficiency ratings sell at a premium of about 9 percent above units with average energy efficiency. Similar price premiums have been observed in the Netherlands for houses certified at the highest efficiency level using a European certification procedure. Some of these premiums may reflect the improved comfort levels that these buildings provide in addition to energy savings. It also seems likely that the energy efficiency premium observed in California is up to three times greater than the incremental cost of the higher efficiency of these dwellings.

Determining Cost

The cost of integrating energy efficiency into new buildings is less than the cost of improving the efficiency of older buildings. A home built since 2000 uses about 25 percent less energy per square foot than one built in the 1960s or earlier. The technical potential for improved energy efficiency in older homes seems evident, but homeowners face two challenges: to determine which improvements have the highest payoff per dollar spent, and to obtain a contractor and financing for the work.

While many diagnostic tools are available to assess existing dwellings, their accuracy varies widely and depends critically on detailed inputs about both the dwelling’s attributes and the household’s living style. Obtaining a contractor and financing can involve high transaction costs for households in effort, time, and money. Many utility companies are offering both technical and financial support for energy retrofitting, but progress has been slow.

Changing Energy Consumption

It may be easier to change residential living styles than to retrofit old buildings, and many utilities are experimenting with schemes to modify household behavior. The most common program involves “nudging” households toward more efficient habits by providing periodic home energy reports that compare their recent energy use with that of their neighbors. Analysis indicates that these reports have both a short-term impact on household energy consumption and a longer-term cumulative impact that continues after the reports end. The energy savings from these programs are small, ranging from a half to one kilowatt hour per day for a household, but the program’s low cost makes the results as cost-effective as many other policies.

Recognizing John Quigley

This conference was designed with John Quigley, economics professor at the University of California at Berkeley, who passed away before the conference took place. In addition to the original papers on energy and cities, papers on urban economics were presented by some of his former students, colleagues, and coauthors. All of the papers will be submitted for a forthcoming special edition of Regional Science and Urban Economics, which will recognize his contributions over a long and distinguished career.

The property tax is the most widely unpopular tax in America. States have responded to this public opposition by enacting a range of tax relief policies, especially for homeowners (Cabral and Hoxby 2012). Among the most commonly adopted programs are homestead exemptions and property tax credits; all but three states have at least one of these programs. But despite their broad use and their potentially large impact on the distribution of property tax burdens, there has been remarkably little data available on the tax savings generated by property tax exemptions and credits.

Two new resources, available through the Lincoln Institute’s Significant Features of the Property Tax subcenter, begin to fill this need. These tables provide information for each state on the share of homeowners eligible for these programs and the level of tax savings they receive, as well as an analysis of how eligibility and benefits vary across the income distribution (see box 1, p. 26). This article draws on these resources to provide the first national study of property tax exemptions and credits with estimates of tax savings from these programs. With this information, policy makers have a critical tool to evaluate and improve the effectiveness of their property tax relief programs.

————————

Box 1: State-by-State Details on Property Tax Exemptions and Credits

The Significant Features of the Property Tax sub-center provides three key resources with information on property tax exemptions and credits in all 50 states; it is accessible at www.lincolninst.edu/subcenters/significant-features-property-tax.

Tax Savings from Property Tax Exemptions and Credits

This online Excel file includes estimates of tax savings from programs in individual states (see abbreviated example below), plus overview tables that make it easy to compare across states. For each program, the file provides estimates of the number of eligible homeowners and the median benefit, as well as a distributional analysis by income quintile. This is the first time that detailed data are available for most of these programs.

Summary Table on Exemptions and Credits

This online Excel file includes a set of tables for 167 programs displaying the value of exemptions expressed in terms of market value; criteria related to age, disability, income, and veteran status; the type of taxes affected (i.e., school or county taxes); whether the tax loss is borne by state or local governments; local options; and more. The summary table makes it easy to conduct quantitative analysis of these programs or make quick state-by-state comparisons. The information in these tables was used to generate the tax savings estimates.

Residential Property Tax Relief

This section of the Significant Features website includes detailed descriptions of property tax exemptions and credits, which were used to create the online Summary Table on Exemptions and Credits. It also describes other types of property tax relief, such as circuit breakers and tax deferral programs.

Notes: Total tax savings from the Senior and Disabled Property Tax Homestead Exemption ($392M) is less than the combined total of the programs for Seniors ($378M) and the Disabled ($22M), because homeowners who are 65+ and disabled cannot claim the exemption twice. The online Summary Table shows that the Senior and Disabled Exemption is a $25,000 exemption for homeowners who are 65+ or disabled; the two Rollback programs are percentage exemptions of 2.5% and 10% for all owner-occupied residences. Source: Lincoln Institute of Land Policy (2015).

————————

How Property Tax Exemptions and Credits Work

Property tax relief programs come in a variety of forms. Homestead exemptions reduce the amount of property value subject to taxation, either by a fixed dollar amount or by a percentage of home value. Property tax credits, in contrast, directly reduce the homeowner’s tax bill by a fixed dollar amount or certain percentage.

As table 1 illustrates, programs designed to provide identical benefits to owners of $200,000 homes have widely different impacts on homeowners with higher- and lower-valued properties. Given a 1% tax rate, a $20,000 flat dollar exemption reduces property taxes for each homeowner by $200 ($20,000 x 1%). This program has a progressive impact on the property tax distribution because lower-income households tend to have less valuable homes, and the exemption represents a larger share of their home values. In this case, the $20,000 exemption reduces property taxes by 20% on the $100,000 home, 10% on the $200,000 home, and 5% on the $400,000 home.

A percentage exemption, in contrast, provides the same percentage reduction in taxes for all three homeowners—in this example, 10%. In dollar terms, however, percentage exemptions favor owners with higher-valued homes: a 10% across-the-board reduction lowers property taxes by only $100 on the $100,000 home but $400 on the $400,000 home.

In the case of flat dollar credits, homeowners with lower-valued homes usually receive the largest tax cuts in percentage terms. In contrast, the percentage tax credit again provides the owner of the $400,000 home the largest tax cut in dollar terms.

An important feature of property tax exemptions and percentage credits is that the dollar reduction (but not the percentage reduction) in taxes increases with tax rates. For instance, if the homes in table 1 were subject to a 2% tax rate, the dollar savings to their owners would double under the $20,000 exemption, 10% exemption, and 10% credit. While the dollar savings from flat dollar credits do not vary with tax rates, the percentage savings to homeowners decrease as tax rates rise.

Critical Features of Exemptions and Credits

The design of homestead exemption and property tax credit programs varies significantly across the 50 states. Figure 1 (p. 28) summarizes the number and share of state programs with the following key characteristics.

Benefit Calculation

Perhaps the most important feature of property tax relief programs is how benefits are calculated. In 2012, 59% of state programs provided flat dollar exemptions, 19% provided percentage exemptions, and the final fifth used property tax credits or other more complicated formulas to determine the amount of tax relief for each homeowner.

While the programs work in similar ways, their effects differ dramatically. As the examples in table 1 show, flat dollar exemptions and credits make the property tax distribution more progressive, while percentage exemptions and credits do not. As a result, to provide a certain level of tax relief for the median homeowner, percentage exemptions are more expensive than other programs because they result in larger property tax cuts for owners of higher-valued homes. Instead of changing the distribution of property taxes among homeowners, percentage exemptions are primarily a way to shift the tax burden away from homeowners as a group to businesses, renters, and owners of second homes.

State vs. Local Funding

The ultimate impact of exemptions and credits on property tax bills depends on how the programs are funded. Figure 1 shows that in 2012 only 28% of these programs included full state reimbursement to cover local revenue losses, while 57% had local governments bear revenue losses on their own. For 15% of programs, state and local governments shared the revenue loss in some way. (Broad-based programs for all homeowners or all seniors are more likely to receive state funding than programs for smaller groups such as veterans or the disabled. In 2012, 43% of tax relief programs for all homeowners or seniors were state-funded, 48% were locally-funded, and the rest split the revenue loss [Lincoln Institute of Land Policy 2014].)

The primary argument in favor of state funding of property tax exemptions and credits is that it can help mitigate disparities in property wealth across localities. Poorer communities and those without a significant business tax base typically have higher property tax rates, and these communities receive more funds per homeowner under state-funded programs. Without this assistance, communities with higher tax rates will experience larger revenue losses from tax relief programs unless they increase tax rates even further.

Seniors vs. All Age Groups

A number of states provide property tax relief for seniors. In 2012, more than a third favored seniors in some way: seven had statewide programs solely for this group, while 11 also covered younger homeowners but provided higher benefits for older homeowners. Other states provided either the same level of benefits for homeowners of all ages (15 states) or did not have broad-based programs (18 states).

Common arguments for targeting senior homeowners is that property taxes account for a larger share of their incomes, and local governments spend less on seniors than on younger homeowners with school-aged children. While it is true that property taxes account for a larger share of income for seniors than for working-age homeowners, the two groups devote nearly identical shares of their incomes to total housing costs because seniors are far less likely to have mortgages (Bowman et al. 2009, 11). In addition, property taxes are payments for public services, not user fees (Kenyon 2007, 36). Younger households without children in public schools do not benefit from property tax relief under these programs. The preferential tax treatment of seniors may simply reflect the fact that older households are a politically powerful group that votes in high numbers.

Estimating the Benefits of Exemptions and Credits

To estimate tax savings from homestead exemptions and property tax credits, the first step was to create the online Summary Table on Exemptions and Credits, which describes the key features of each program (see box 1 for description). These data draw almost entirely from the Residential Property Tax Relief Programs section of the Lincoln Institute’s Significant Features of the Property Tax database.

The second step was to combine this information with household-level data from the 2008–2012 American Community Survey (ACS). This nationally representative survey has data on more than 6.5 million U.S. households, including the household characteristics that determine program eligibility (age, income, disability, veteran status, etc.) and level of benefits received (home values and property tax bills). For a full explanation of the methodology used to estimate tax savings from exemptions and credits, see Langley (2015).

It is important to note that the estimates reported here are gross property tax savings. Tax relief programs often lead to higher property tax rates, especially under locally-funded programs where jurisdictions raise tax rates to offset the drop in the tax base from the exemptions. Estimates of net property tax savings would be lower in those communities, because the higher tax rates offset some of the direct tax relief provided from exemptions and credits.

Figure 2 shows that total property tax relief from homestead exemptions and property tax credits varies widely across states, but is generally small relative to total property tax revenues. In 14 of the 45 states with these programs, total savings are less than 0.5% of property tax revenues; in 27 states, the savings are less than 2.5%. At the same time, though, tax savings in nine states equal or exceed 10% of total property tax revenues. Indiana’s program is particularly generous, offering all homeowners a $45,000 exemption, then an additional 35% exemption for the first $600,000 in assessed value and a 25% exemption for value above $600,000.

Tax Savings for Different Types of Programs

Most states have more than one property tax exemption or credit program, with different programs targeting different groups of taxpayers—typically all homeowners, seniors, veterans, or the disabled. Figure 3 presents estimates on the share of homeowners eligible for these programs, along with the level of tax savings they receive.

Homeowners

Programs in 26 states are for nearly all homeowners, but usually limited to owner-occupied primary residences. In the typical state with these programs, the median homeowner receives a 12.5% cut in property taxes. On the high end, however, the median property tax cut was at least 25% in more than a quarter of states with these programs.

Seniors

Property tax relief programs in 18 states target older homeowners (typically at least age 65). These programs are much more generous than those covering all homeowners, with a median tax reduction of nearly 30% in the typical state. More than half of these programs provide a median tax cut of at least 25%, while only a sixth of them provide a median tax savings of less than 10%.

In the median state, 19.6% of homeowners are eligible for the programs, but eligibility rates vary greatly across states depending on whether there is an income ceiling. In the seven states that provide property tax relief to seniors regardless of income, 25–30% of homeowners are typically eligible. But in seven states with low income cutoffs ($10,000 to $30,000), only 5–10% of homeowners qualify. The other four states with property tax relief programs for seniors do not fit neatly into these two categories because they have higher income ceilings, strict wealth limits, or other eligibility criteria.

Veterans

State programs for veterans are more common than for any other group of homeowners, although eligibility is often limited to those who are disabled. Indeed, only 10 states provide property tax exemptions or credits for all veterans, even those without disabilities. In the median state with these programs, the typical beneficiary receives a property tax cut of just 3.2%.

There are 31 states that provide property tax exemptions or credits to veterans with service-connected disabilities. Because of the disability requirement, most veterans are ineligible for the programs. Indeed, only 15% of veterans qualify in the typical state. Overall, just 0.6% of homeowners are eligible for these programs in the median state.

Moreover, most of the 31 programs base eligibility and benefit levels on disability ratings from the Department of Veterans Affairs. Just seven states have programs for all partially disabled veterans, and veterans with lower disability ratings typically receive modest tax savings. On the other hand, 18 states restrict eligibility to veterans who are permanently and totally disabled. These programs benefit a very small share of veterans, but they usually provide a full 100% exemption.

Disabled

Programs in 23 states cover disabled homeowners, but really target two distinct groups: disabled homeowners and blind homeowners. In 2012, 12 states had programs for disabled homeowners, seven states had programs for the blind, and five states covered both groups. Programs for the disabled typically require beneficiaries to be permanently and totally disabled, but exact criteria vary. In the median state, 2.3% of homeowners are eligible for these programs and they receive a median property tax cut of 21%.

Conclusion

Homestead exemptions and property tax credits are an important part of the property tax system. These programs are used in nearly all states and can make the distribution of property taxes significantly more progressive. It is therefore critical that policymakers have good data on the property tax relief that these programs actually provide.

New research makes this information available for the first time. Using the Lincoln Institute’s Significant Features of the Property Tax subcenter, policymakers can easily compare key features of property tax exemption and credit programs across states, and see estimates of eligibility and tax savings. These data make it possible to evaluate the impacts of property tax exemptions and credits in their particular states as well as find ideas for program improvements.

Adam H. Langley is Senior Research Analyst at the Lincoln Institute of Land Policy. Special thanks go to Andrew Reschovsky, who provided extensive comments on this article and other related papers.

References

Bowman, John H., Daphne A. Kenyon, Adam Langley, and Bethany P. Paquin. 2009. Property Tax Circuit Breakers: Fair and Cost-Effective Relief for Taxpayers. Cambridge, MA: Lincoln Institute of Land Policy.

Cabral, Marika, and Caroline Hoxby. 2012. “The Hated Property Tax: Salience, Tax Rates, and Tax Revolts.” Cambridge, MA: National Bureau of Economic Research. Working paper 18514. November.

Kenyon, Daphne A. 2007. The Property Tax-School Funding Dilemma. Cambridge, MA: Lincoln Institute of Land Policy.

Langley, Adam H. 2015. “Estimating Tax Savings from Homestead Exemptions and Property Tax Credits.” Working paper. Cambridge, MA: Lincoln Institute of Land Policy.

Lincoln Institute of Land Policy. 2014. Significant Features of the Property Tax. Residential Property Tax Relief Programs: Summary Table on Exemptions and Credits in 2012. www.lincolninst.edu/subcenters/significant-features-property-tax/Report_Residential_Property_Tax_Relief_Programs.aspx

Lincoln Institute of Land Policy. 2015. Significant Features of the Property Tax. Tax Savings from Property Tax Exemptions and Credits in 2012. www.lincolninst.edu/subcenters/significant-features-property-tax/Report_Residential_Property_Tax_Relief_Programs.aspx

Scholars and public officials concerned with social justice consider redistribution of land values to be an especially important objective of urban policy in Latin American countries, where great differences in access to scarce urban infrastructure and services result in an unfair distribution of land values. However, value capture policies and instruments used in principle to “redistribute the valorization gain” or “promote redistribution of land value increments” are rejected by some progressive sectors because they believe that, in spite of the redistributive connotation, those instruments are not really aimed at redistribution in practice.1 This article explores a number of questions that must be addressed to achieve a better understanding of the value capture concept and its potential to play a truly redistributive role in Latin America.

The Distributive Principle and the Redistributive Goal

The basic principle of value capture 2 is to return to the community the land value increments resulting from community action. The most usual way to define those increments is to focus on particular increases in land value that result from specific and dated public actions. The corresponding value capture instruments could, therefore, be thought of as devices to recover for the public the increase in land value associated with public actions that otherwise would be captured by private entities. The aim of this distributive policy is to restore a previous state of distribution that, in essence, is taken as a proper or given one.

An alternative interpretation is based on the principle stated by Henry George that all land value, irrespective of its origin, is the product of community effort. In this view, only when all of the land value is taken into consideration and the goal of altering the current state of land value distribution is introduced can the value capture idea acquire a truly redistributive perspective.

Redistributing land values is but one of the possible goals of urban land policy. Other goals are raising public revenues to finance urban services, regulating and managing urban land uses, and controlling undesirable outcomes of the functioning of urban land markets. That is, redistribution may be a guide to more progressive distributive policies, but it is not necessarily the basic principle of value capture.

Thus, we can distinguish between the distributive principle of value capture policies-to restore a certain state of distribution-and a redistributive goal of urban land policies-to alter a certain state of distribution. This distinction allows us to address the confusion about distribution and redistribution applied to land values and to the value capture idea.

The Practice of Value Capture in Latin America

In its generic sense, the value capture idea applies to any levy or planning tool intended to distribute land value increments. Almost all Latin American countries have experience with the property tax, and many have other planning tools such as the compulsory donation of land for public purposes in land parceling or subdivision projects. Historically, the development of the value capture idea has been associated with a specific instrument known as Contribución de Valorización/Mejoras. This special assessment or valorization charge, incorporated into the legislation of most Latin American countries, aims at capturing a portion of special benefits (land valorization) that arise from public investments in infrastructure and services, to finance such investments.

Even with this narrow definition, the implementation of value capture has been limited and controversial. Both the political influence of landowners and the technical (but also often legal) shortcomings of adequately assessing land values have been identified as restraints to its use in many countries. Colombia is perhaps the only country with an established tradition of using the instrument, but even there its implementation is subject to serious limitations. Some observers acknowledge its incapacity for redistribution and others claim it frequently loses the link with the distributive principle and becomes simply a practical way to pay the community for the costs of a public action that generates benefits for only some individuals.

A closer look at concrete Latin America experiences with the implementation of value capture instruments leads to a disturbing conclusion. Rather than evolving from the ethical principle of fairness, whereby the increment of land value resulting from community action returns to the community, the value capture idea seems to have been adopted in Latin America as a pragmatic cost-recovery mechanism to overcome the chronic shortage of public revenues to finance urban infrastructure. The major goal of such value capture instruments has been ultimately to raise public revenues, whether based on a distributive principle or not.

Linking Value Capture and Redistribution

Even when the distributive principle is secured, the goal of raising public revenues can differ from or even contradict other goals of urban land policy, including the important redistributive goal. For instance, when a public investment in urban infrastructure generates land value increments in a highly valued area, and then associated income from the use of a value capture instrument is reinvested in the same area, the result is not redistributive and can even be regressive.

To understand the contradictions that arise between the traditional use of value capture instruments to raise general revenues and the necessity of incorporating the redistribution goal into those policies, we need to consider value capture as a more comprehensive concept. Even when limited to its usual definition centered on specific land value increments, at least three non-autonomous public actions or decisions must be associated with the distributive principle of value capture:

1. an original public action (regulation, investment, etc.) that results in land value increments;

2. a second action to capture (some of) this value; and

3. a third action related to the destination or use of collected resources.

While the second action implies the use of a general or specific value capture instrument, the first and third actions, though related to specific decisions, cannot be separated from two basic questions concerning public decisions as a whole: How are public works allocated in space, and how is the general revenue distributed?

Allocation of public works

When raising revenues and promoting redistribution are concurrent goals, the second does not necessarily follow the first. In Latin America these goals are often contradictory. Under conditions of highly uneven distribution of wealth and scarce funds to finance public works, it is usually easier to guarantee the raising of revenues through the allocation of public works (original action) in areas where more absolute revenues can be collected. Even with the use of a value capture instrument, when the subsequent decision (destination of resources) maintains the same state of wealth distribution, the whole public action becomes regressive.

On the other hand, rejection of value capture instruments does not prevent the misallocation of public works. In fact, it just contributes to the status quo. For example, the facelift of Copacabana in Rio de Janeiro, which replaced old trees and modernized sidewalks, was financed by the general revenue, not by a specific value capture device. However, many of the poor areas of the city have neither sidewalks nor a single tree on their streets. Recognizing this irony reinforces the need for a new framework for value capture policies that can allocate public works more equitably.

Distribution of general revenue

Latin America presents extreme relative and absolute differences in public infrastructure provision, calling for equity criteria to evaluate distributive policies. Yet, equity criteria are subjective and there are distinct visions on what is fair. Given the disparities in wealth and in access to serviced land, it is important to consider not only relative differences but also the absolute differences between highest and lowest levels.

To illustrate this point we can apply the classic redistributive argument to the distribution of land values in a society with 10 units of wealth (i.e., land value) distributed between two groups: the higher group has 8 units or 80 percent of the wealth and the lower group has 2 units (see Table 1). This example can represent the typical differences between serviced areas occupied by the rich and unserviced areas occupied by the poor in Latin American cities. An increase of 50 percent in this wealth (5 total units), if distributed in the same ratio, does not change relative differences, but the absolute difference between the two groups is increased by 50 percent, from 6 to 9 units.

TABLE 1: Distributive Value Capture Policies Total Wealth Lower Group Higher Group Relative Differences Absolute Difference original: 10 units 2 units 8 units 1:4 6 units increased: 15 units 3 units (2+1) 12 units (8+4) 1:4 9 units

Another important consideration is the level of the group in the lowest position. Value capture instruments are justified as distributive tools to return to the community special benefits resulting from a public action that only some individuals receive. But that justification in turn raises the need to clearly separate special benefits from basic needs. If we consider access to urban infrastructure as a basic need, the society must decide on the minimum level of access for the lower group. Priority should be given to actions that achieve those minimum levels before other benefits accrue to the higher group. If this society decides that the minimum level of wealth should be 6 units for the lower group, then an increment of 5 units of land value would be distributed in such a way as to decrease both relative differences and absolute differences (see Table 2).

TABLE 2: Linking Value Capture and Redistribution Total Wealth Lower Group Higher Group Relative Differences Absolute Differences original: 10 units 2 units 8 units 1:4 6 units increased: 15 units 6 units (2+4) 9 units (8+1) 2:3 3 units

Value Capture and Socio-spatial Equity

Urban planning decisions, such as the norms and regulations on land use and development rights, also affect the distribution of urban land values and must be integrated into value capture policies. In Latin America, where the differences in access to public infrastructure and urban services are marked by severe social segregation and exclusion, this integration implies the inclusion of a socio-spatial dimension that can deal with the disparities between serviced rich center cities (for the few) and unserviced poor peripheries (for the majority). Therefore, land value redistribution policies acquire a particular political context in which the generation of land value increments and the destination of corresponding funds are fixed in distinct socio-economic areas of the city.

However, even when this socio-spatial dimension is incorporated, most redistributive value capture instruments provide necessary but not sufficient conditions for a better distribution of land values. While redistribution from rich areas to all areas involves altering the distribution of general revenue to achieve its equity objective, redistribution from all areas to poor areas involves altering the allocation of public works and/or development rights on land to arrive at a better distribution of land values.

Since these approaches involve greater institutional changes, a third option seeks to stimulate the generation of land value increments in rich areas in order to raise revenues that can be redistributed to poor areas. These so-called “Robin Hood” policies are being considered to deal with urgent needs in poor areas, combined with specific opportunities and demands in rich areas. One example is the “linkage operation” recently popularized in many large Brazilian cities, where the negotiation of legal exceptions for development generates payments earmarked for social housing. However, a careful examination of this transfer tool shows that stimulation of land value increments in rich areas actually increases intra-urban differentiation and as a result may exacerbate the gap between rich and poor areas.

This and other largely unanticipated perverse outcomes show that the development of value capture policies and instruments for Latin American countries cannot be considered independently from an urban land policy oriented to the reduction of socio-spatial inequalities. The latter can be attained only by direct actions geared to altering the current distribution pattern of land values. This means that redistribution, although not necessarily implied in the value capture idea, must be incorporated deliberately into the development of distributive value capture policies.

Guidelines for Implementing Value Capture Policies

This discussion reinforces the argument that value capture policies in Latin America must be preceded by changes in the process of distributing land values in the broadest sense, especially where redistribution is pursued as a major goal of urban policy. This perspective would help to consider in an integrated manner, in each public decision concerning a specific way of distributing urban land values, several other ways in which the public sector contributes to this distribution, including:

• the way taxes on land are designed and collected;
• the way public revenue is allocated for public works;
• the way specific value capture instruments are applied (or not);
• the way the collected resources are apportioned; and
• the way land uses and development rights are defined.

The potential and limits of specific value capture instruments are conditioned by those distributional public actions and decisions. When specific value capture instruments are used independently from this consideration, the whole process may be undermined. Collection of land taxes is usually neglected; public investments tend to be allocated unjustly; political impediments to the use of value capture instruments abound; revenues are not distributed in a socially equitable manner; development rights are incorporated in ownership rights, etc. As a consequence, redistribution cannot be attained and the distributive principle is imperiled.

The challenge in Latin America, then, is to work out the preconditions for improved use of the value capture idea, rather than simply to focus on overcoming procedural difficulties in applying existing instruments or to reject those instruments in favor of replacement tools usually subject to similar shortcomings. To have a chance of being truly redistributive, these distributional decisions should account for all components of land value, including accumulated, potential and specific increments, not only land value increments in the strictest sense. Efforts in this direction may contribute to a redistributive perspective on value capture policies.

How much value capture is “enough” will vary among countries, but the balance of policies should include these basic guidelines:

• improvement and strengthening of the property tax, especially its land component based on all of the land value as opposed to specific land value increments;
• universalization of the provision of public infrastructure and urban services (i.e., basic needs as opposed to special benefits); and
• socially responsible approach to the definition and regulation of ownership rights and development rights on land.

These guidelines are strongly associated with urban land value increments in the broadest sense, and they can be used to reduce absolute and relative socio-spatial differences. If they continue to be neglected, and value capture policies are confined to specific land value increments, attempts at redistribution in Latin American countries are bound to fail. Furthermore, the implementation of value capture instruments will continue to serve as an anti-social mechanism that only exacerbates the already great differences between rich and poor.

Fernanda Furtado is a fellow of the Lincoln Institute. She received a dissertation fellowship from the Institute to help complete her Ph.D. thesis on “Urban Land Value Recapture in Latin America” at the Faculty of Architecture and Urbanism at the University of São Paulo, Brazil.

See the Latin American Program and Land Lines sections of this website for additional articles and reports on this topic in both English and Spanish.

Notes

1 See Donald Shoup, “Is under-investment in public infrastructure an anomly?” in Gareth A. Jones and Peter Ward, eds. 1994. Methodology for Land and Housing Market Analysis. Cambridge, MA: Lincoln Institute of Land Policy. Shoup’s piece includes the debate held during the 1991 Fitzwilliam Workshop on Land Values and Land Valorization in Developing Countries at the University of Cambridge on whether value capture instruments are intended to redistribute the valorization gain or are just a device to strengthen government finance.

2 It would be more precise to speak of value recapture, because besides better representing public interventions in order to return to the community the unearned land value captured by private entities, the term alludes to redistribution as a specific way of developing such policies. However, the more generic term value capture is used in this article.