Historic designation and the rebuilding of neighborhoods: new evidence of the value of an old policy tool

Abstract

Rebuilding central cities has focused on big-ticket items such as entertainment complexes and neighborhood-based initiatives. Historic preservation as one neighborhood-based strategy seeks to capitalize on intact collections of historic architecture and the pedestrian-friendly character of these pre-automobile neighborhoods to elevate property values and enhance the tax bases of central cities. While a majority of past studies indicate historic designation does enhance property values, some found no or negative impacts. By quantitatively assessing the impact of local historic district designations on the enhancement of residential property values, this paper contributes to the literature on the relationship between historic preservation and urban revitalization. The multi-city analysis focuses on preservation’s impacts on residential property values and the resulting positive and negative outcomes from district designation.

Keywords:

• historic preservation
• economic development
• revitalizing core cities

Notes

1The coefficients represent the change in dollar of sales price attributed to one unit of change in the associated variable.

1. The research focuses solely on locally designated historic districts, not National Register of Historic Places historic districts. The former typically bring additional rules and restrictions about alterations and demolition to historic properties. The latter is a largely honorary distinction, with few to no mandates for property maintenance, upkeep or alterations.

2. In the years studied, each of the fast-growth areas were indeed robust, with rising employment levels and home values. While the Dallas area has done better than many others with regard to employment and housing since the recession’s onset, Phoenix has not. The current situation in Phoenix does not interfere with the study’s objectives, given the years for which data were collected.

3. Most cities’ preservation ordinances are structured slightly differently. The most common approach is for a historic district or landmarks commission to review applications for alterations. In some cases, though, this responsibility falls to the planning commission or individual review bodies established for each district. Any research on local historic districts must take into account these local variations.

4. For the remainder of this article, the phrase “historic district” refers to local historic districts.

5. Cities have the authority to regulate land use as specified by each state’s enabling legislation. Acceptance of a city’s right to establish zoning and land use codes is stipulated and accepted by landowners through the use and purchase of property within a city’s borders. Additionally, the state and federal courts have long upheld cities’ rights to control private property through zoning and historic designation. See Berman v. Parker 348 U.S. 26 (1954), Penn Central Transportation Co. v. New York City, 438 U.S. 104 (1978), Duerksen (1983).

6. For insight into the debates about preservation regulation, the need for incentives, and issues of property rights see Throsby (1997), Schuster (1997), and Costonis (1997).

7. For an overview of the multiplicity of values driving historic preservation decisions, see Mason (2006) and Mason and Avrami (2002).

8. For example, see Seattle, Washington’s transfer of development rights (TDR) program (Seattle.gov n.d.). While, in theory, this initiative can balance out the supply issues, it is a complicated and infrequently used option. In a broader context, TDR programs are troubling in that, at some point, the amount of sellable “development rights” must be calculated. As has been reinforced by the US Supreme Court, owners are not guaranteed unlimited use of their land. It is thus an imprecise calculation to figure out how much development rights have been restricted via the creation of a historic district.

9. To determine regional population change, we used the Metropolitan Statistical Area (MSA) geography as defined by the US Census Bureau.

10. The suburban communities in five of the six Metropolitan Statistical Areas gained population, despite the primary city’s fate. The only exception was Pittsburgh MSA, where suburban communities’ population declined 1.2% from 2000–2006. In the fast-growth regions, suburban areas had at least 15% population growth during the same period.

11. Limitations in available resources constrained the study to six regions.

12. A full review of the history and policies for each historic district is beyond the scope of this paper. For more information on Whittier Mill, see Atlanta Urban Design Commission n.d.; for Junius Heights, see Dallas Landmark Commission n.d.; for Alvarado, see Phoenix Historic Preservation Commission n.d.; for Betts-Longworth, see Cincinnati Historic Conservation Board n.d.; for Ohio City, see Cleveland City Planning Commission n.d.; and for Allegheny West, see Pittsburgh Historic Review Commission n.d.

13. For each city, the historic and non-historic areas are located in the same jurisdiction and receive the same urban services (trash collection, sewers, utilities, fire, police). Furthermore, they fall within the catchment of the same school district. The six school districts are: Dallas Independent School District, Atlanta Public Schools, Cleveland Metropolitan School District, Cincinnati Public Schools, Pittsburgh Public Schools, and Phoenix Union High School District.

14. Simons et al. (1998) provide evidence that separation of at least one mile is sufficient to minimize the possibility of spillover benefits from an historic district.

15. This research focuses on the observable, comparative characteristics of historically designated and non-designated neighborhoods. There are certainly other factors that might influence why a neighborhood has historic designation or not (the activism of local residents, a desire to keep out unwanted uses, concern over property rights infringement in an area, etc.). It is beyond the scope of this research to investigate those factors. Future research should explore methods of incorporating more qualitatively assessed data into quantitative property values research.

16. Existing literature makes it clear that there is no simple and universal way to determine either the complete set of relevant characteristics for price determination, or the true functional form for whatever set of characteristics is used in any given application. In this case, due to the variation in the structure of the data provided by the various cities, we faced two options. The first was to omit from any of the models any variable not available from any one of the cities. This would have the advantage of analytically treating all of the cities in exactly the same way, thus enhancing comparability between cities. However, such comparability would have been achieved only at the cost of potential bias attributable to omitting relevant variables from the analysis. The second option was to use all of the variables available for each city, but to use a slightly different set of characteristics for each city. This option minimized the problems associated with omitted variable bias, but limited comparability between cities. We chose the latter option, because we judged that the benefits of minimizing omitted variable bias outweighed the benefits of enhanced comparability. Accordingly, it should be clear that while the models reported herein were calibrated with the idea in mind of making them as comparable as possible given the data provided to us by the various city governments, they are not strictly comparable. The notation “n/a” (not available) indicates when one or more data elements were not available for the properties in one of the cities.

17. Even the best-fitting and best-behaved of the possible regression models fit to these data exhibited a couple of unsettling features. In a couple instances, these features could be explained. For example, a negative coefficient on the number of bedrooms initially seemed to be counterintuitive; however it became completely intuitive as soon as we realized that when total square footage is held constant, a negative sign for bedrooms reflects smaller individual rooms. Fewer and larger bedrooms command higher prices, as units with larger individual rooms are seen as more desirable and more valuable. In other instances, statistical and analytic tests set any associated concerns to rest. For example, variance inflation factors indicated no reason for concern with multi-collinearity in any of the models, and other residual analysis indicated no reason for concern with model pathologies. Thus we can say confidently that the variation in price significantly and consistently accounted for by the historical designation variable is not collinear with any of the control or other independent variables in these models.

18. What counts as historic is a somewhat debated idea in scholarly circles, but fairly well defined in practical terms. For an overview of the scholarly debates, see Tomlan (1998). Most cities establish their own criteria for historic designation, but largely follow those set forth by the National Park Service for listing on the National Register of Historic Places. For an overview of the latter, see National Park Service (1997).