Interactions Between Transportation Capacity, Economic Systems, and Land Use (2012)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

45 C h a p t e r 8 The T-PICS system has been designed to allow practitioners to add new case studies to the database. This chapter provides guidance on how to conduct such studies. Data Collection Review of Similar Cases The T-PICS database includes 100 case studies of projects throughout the United States and abroad. Reviewing cases in the system before conducting new case studies can help the researcher identify potential sources of background information and the types of people or organizations that should be interviewed and provide insights into the types of questions that can be asked to elicit the most useful data and information. The database can be sorted to help the researcher identify projects that most closely correspond with his or her project. Collection of Background Documents and Literature As a starting point for each case study, it is useful to gain an understanding of the context in which the project has been introduced and matured. An Internet search should be under- taken to gain general knowledge of the project and the region in which it was built. Good places to start include aaroads.com and state DOT websites, as well as local economic develop- ment agency websites. A web search of the project itself can also turn up environ- mental impact reports and other project-related documents, as well as newspaper articles about the project. It is also useful to search the name of the community and any development projects related to the investment of which you are aware. The literature search will provide the researcher with a general understanding of the project and can be used to tailor inter- view questions to collect the best information for under- standing the project and its impacts and for relating the story of the project in the project narratives. Any useful documents or websites should be recorded for entry into the system. Quantitative (Empirical) Data Collection T-PICS includes empirical data for each case study. Those adding new cases to T-PICS will need to collect background demographic and economic data on a local, regional, and statewide basis to populate the database. Such data usually can be collected from published sources. The researcher may not be able to fill in all fields; that is all right, although researchers should try to fill in as possible. The categories of data are (a) Project Data—general information about the project; (b) Setting—information about the project’s local area; and (c) Impact Measures—economic activity levels rep- resenting 1 year before construction and 5 or more years after project completion. The specific empirical data to be assembled were defined in Chapter 3, and categories of project types were defined in Chapter 2. Data sources are listed in the Data Dictionary and in a supplemental spreadsheet table, which are both available on the T-PICS website (http://transportationforcommunities .com/t-pics; click About T-PICS). Additional notes and pointers are provided here. • Impact area. The impact area typically is the counties in which the project passes or is located. However, for some large projects there may be additional counties of impact identified through the interview process. • Impact measures. Measures may include (a) employment, (b) income, (c) business sales, (d) property values, (e) tax revenues generated, (f) square feet of building construc- tion, and (g) value of investment in terms of construction cost. Items (a)–(e) are measured in terms of annual levels for two points in time representing conditions before and after the project. Items (f) and (g) describe activity occur- ring between the two points in time. All may be measured at the local, county, and state levels. Conducting Future Case Studies

46 • Time periods. “Preproject” data are collected for the year before construction begins and “postproject” data for at least 3 years after project completion. The postyear selec- tion may depend on the project type. The full economic effect of an access road may take only 2 years to be observ- able, while the full effects of an interstate may take 5 to 10 years. • Employment. Employment is measured by place of work (i.e., it represents the number of people working at loca- tions within the study area, regardless of where they live). This information should not be confused with data on employment by place of residence, which represents a measure of local labor force. Average worker income is similarly measured by place of work. Interview Data Collection Although some of the empirical impact data (such as employ- ment trends) can be collected via public sources, other types of impacts require local information (such as property values and building construction information). In addition, the case studies should include information about causal factors affect- ing project impacts (including supporting infrastructure, land use policies, and business programs). To obtain this local infor- mation, the researcher must conduct interviews with key public officials (e.g., local or regional planning agencies) and private- sector representatives (e.g., Chamber of Commerce or develop- ers), as well as review available local documents. The purpose of the interviews should be to develop a coherent narrative describ- ing the planning, implementation, and results of the project. A list of basic interview topics was presented in Chapter 3. Questions do not need to be followed verbatim; they are sim- ply guidelines for the types of information to be collected. Interviews generally are more effective if they are conversa- tional, as opposed to asking a numbered series of questions. Thus, interviews should start with an explanation regarding the purpose and use of the case study database and why there is interest in this specific project case. Questions may also be amended or added, based on issues identified from the back- ground information. analysis Net Economic Impact Net economic impact is calculated as the change in employ- ment or other impact metrics between a preproject year and a postproject year, which may reflect the net result from a mix of positive changes (such as new jobs created at one part of the study area) and negative changes (such as job loss else- where in the study area). Information for statewide trends over the same period are also collected to enable additional comparisons of how local changes differ from the effects of underlying trends and business cycles that also affect broader state and multistate regions. Attribution of Causal Credit The attribution of causality for observed economic impacts is another important consideration. In other words, the impact of a highway project is not necessarily the difference between economic measures before and after construction. For instance, if there are 5,000 local jobs before a highway’s construction and 6,000 after its construction, this does not mean that the highway is responsible for creating 1,000 jobs. There are other factors that may have come into play during the highway’s construction period that may have had nothing to do with the project. Direct versus Total Impacts Impacts on business activity (including employment, income and output changes) may be calculated in either of two ways. • The first way is to observe direct effects, defined as changes in adjacent or nearby areas, and then apply a localized input- output multiplier to calculate a total impact figure for the surrounding area that also accounts for “indirect effects” (growth of other area businesses that supply products and services to the directly growing business) and “induced effects” (growth of other area retail and service businesses due to spending of income by the additional workers). • The second way is to observe changes in the broader econ- omy of the county or multicounty study areas, in which case total impacts are already being captured. The likely source for direct impact information is employ- ment data, which can be obtained for multiple points in time by census tract or zip code. (Employment by place of work can be acquired from the census tract files of the Longitudinal Employer-Household Dynamics database [http://lehd.did .census.gov/led] or from the zip code files of County Business Patterns [www.census.gov/econ/cbp].) Local data often can also be obtained on property sales, construction activity, and tax receipts from inquiries made during local interviews. Direct impacts on jobs can also be estimated if the researcher obtains information on the square feet of new development built as a result of the transportation improvement. These estimates are available from sources such as the Urban Land Institute (www.uli.org), which reports on typical ratios of workers per 1,000 square feet of occupied building space. (The estimates vary but are typically in the range of 1.0 for warehouses, 2.1 for industrial space, 2.2 for retail space, 4.2 for office space, and 0.7 for hotels.)

47 Input-output economic multipliers reflect the ratio of total direct effects. There are separate multipliers for employment, income, and output changes, and they also vary by county (or aggregation of counties). They are also affected by industry mix. The T-PICS case studies have all used IMPLAN model multipliers that have been customized for the applicable study areas of each of the 100 projects. Construction of a Narrative A full understanding of the impacts of a transportation investment requires not only data analysis but also a distilla- tion of findings from interviews, local data collection, and a review of previous local studies. The narrative should be a relatively brief (3- to 5-page) story of how the project came about and its impacts on the local area. The structure should be in the following order: • Synopsis. A one-paragraph summary of the project history and its outcomes. The summary should include a descrip- tion of the project, its location, dates of construction, proj- ect cost, and impacts in terms of jobs or types of businesses attracted. • Background. Describe the local project context. The back- grounder should include a brief economic history of the region, population and employment trends, description of major transportation routes and facilities that serve the area, travel time to the nearest commercial airport, and other transportation features. • Project description and motives. Describe the project (type, cost, and so forth) and why it was built. • Transportation impacts. Discuss the implications of the project on local transportation, such as changes in average annual daily trips, travel time savings, or other factors. • Demographic, economic, and land use impacts. Discuss pre- construction and postconstruction data and impacts attributed to the project, such as new firms attracted and retained and changes in employment, land use, and land development. • Nontransportation factors. Discuss other factors that influenced project outcomes (e.g., supportive policies and incentives). If several factors combined with the transportation investment to create a climate for eco- nomic growth, then transportation investments can only be attributed a portion of that growth. The allocation of causality for each project should be discussed with interviewees. • Resources and citations. Compile a list of studies and links to websites used in the case study. • Interviews conducted. Compile a list of organizations par- ticipating in the interview process. Challenges Although much of the requested data for case studies can be relatively straightforward to collect, the availability of some data elements varies from project to project. The level of effort needed to collect each data element also varies by proj- ect type and scale, although certain elements are particularly elusive. This includes information regarding the following: • Complementary actions; • Interventions; • Land use patterns and policies; • Future development capacity; • Financial incentives/business climate; • Congestion; • Property values; • Property tax revenue; • Private investment; and • Commercial space. Difficulty collecting information on these data elements can be attributed to one or more of the following challenges. Time Series Not Available Although planning and land use context information often is available in database form, it generally is not available as time series data. A researcher interested in a particular project can obtain current land use information from the planning department covering the project’s jurisdiction, but if the project crosses city or county lines, the researcher has to visit several planning departments. It is also unlikely that the plan- ning department can provide land use data covering previous periods, making before and after changes to land use difficult to determine other than anecdotally. No Centralized, Consistent Source Economic development intervention and support policies are a perfect example of information that is difficult to collect because it is not housed in a centralized source. In the United States—and even in individual states—there is no single agency charged with economic intervention or provision of financial/business attraction incentives. In fact, such efforts often come from multiple levels of government with varying degrees of coordination. Furthermore, economic develop- ment intervention and support policies are heterogeneous, ranging from streamlined permitting processes, to shovel- ready sites, to tax credits and direct cash transfers. A retail center at a major highway visible site created by a transporta- tion investment could receive various incentives from any number of sources. Sometimes such support is tracked either

48 formally or informally by an economic development agency, but because support can come in many forms and from many different entities, it can be difficult for a researcher to identify all of the agencies with relevant information. The interview process can help with this task, but if the information is scat- tered across numerous agencies, the level of effort needed to obtain complete information can become substantial. Data covering property values and property taxes can be obtained from a centralized source (the local property tax assessor’s office) but neither assessed value nor tax col lections data are defined consistently across jurisdictions. Obtaining property value from the tax assessor is problematic because each jurisdiction assesses property value differently. In some jurisdictions assessed value is meant to represent the full market value of a property, and when updated regularly, gen- erally reflects market values. However, if properties are not routinely reassessed, then over time values in the assessor’s database will deviate from market values. Some jurisdictions use a percentage of market value as assessed value, while other jurisdictions, such as those in California, are statutorily lim- ited in how much value may increase from year to year, which tends to artificially hold assessed values far below market values. Therefore, it is not enough for a researcher to simply col- lect property value data from a local assessor’s office. The researcher also needs to understand the local system concern- ing how property values are assessed (full, partial, statutorily) and how often assessed values are updated. Analysis of prop- erty tax data can also be problematic, for although most assessors’ databases can capture time series data, property tax rates are subject to change from year to year. Thus, in addi- tion to property tax associated with a particular property or total property tax for a jurisdiction, the researcher needs to know the prevailing tax rate for each time period for which data are collected to ensure that fluctuations are the result of actual changes in underlying property value and not simply changes in tax rates. Data Availability/Accessibility Limitations Some data elements exist but cannot be readily accessed the way researchers interested in studying the impacts of trans- portation impacts need them. For instance, it may be rela- tively simple to obtain jurisdiction-wide totals for assessed values or taxes paid, but subjurisdictional or parcel-level data may not be available. Although some jurisdictions have sophisticated GIS-based database systems and are willing to do specialized data runs, other jurisdictions have basic sys- tems for which subarea data runs would be an overly time- consuming imposition on the assessor’s staff. In the case of the commercial space data discussed above, market and submarket definitions used by the data source may not match those relevant to the project of interest, and the pri- vate firms that collect the data may not be willing or able to do specialized data runs or may charge a fee for the service. Collection of some data elements is stymied by a combina- tion of the above. Data tracking total commercial space before and after a project typically lacks a centralized source and consistency. Commercial real estate broker firms often collect data for the larger real estate markets reflecting total space, rents, and vacancy levels by product type. However, they do not typically maintain time series data, nor do they cover smaller, nonmetropolitan markets. Broker interviews can be used to get a general sense of current property values, but few brokers track property values over long periods of time. Scale of the Data Collection Effort All of the preceding variables must be considered in the con- text of the larger data collection effort. The researcher collect- ing each of the above may be collecting dozens of other pieces of data from a broad range of sources, sometimes from mul- tiple jurisdictions, sometimes at the subjurisdictional level, for many projects across the country, all under time and bud- get limitations. If this effort is multiplied by a number of separate case study projects, the challenge becomes clear.