Methodology for Determining the Economic Development Impacts of Transit Projects (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.

8 2. SYNTHESIS: THE STATE OF THE ART AND THE STATE OF THE PRACTICE In this section of the report we describe existing academic theory and research methods to estimate the economic impacts of transit investments, focusing on agglomeration impacts. This is followed by a description of how public agencies estimate and understand the potential economic impacts of project proposals and evaluations, based on our interviews with staff as well as our review of reports and other documents. Our focus was on three main questions: • How are transit infrastructure investments thought to affect the economy? • How have those possible impacts been investigated and estimated? • What are the practical challenges in carrying out such estimates? We begin with an overview of existing federal planning and evaluation processes for fixed-guideway transit systems. We then summarize our extensive review of academic literature on transportation and agglomeration (the full review can be found in Appendix C). We next turn to a description of our interviews with practitioners familiar with New Starts and Small Starts funding applications to the Federal Transit Administration (the full description of interviews is Appendix D). Finally, we describe documentary sources and guidance that explain how economic impacts are estimated in practice (the full review of these practice documents is in Appendix E). Existing Federal planning and evaluation processes The Federal Transit Administration’s (FTA) New Starts and Small Starts programs are the primary federal funding resource for capital investments in fixed-guideway transit systems. SAFETEA-LU identifies specific criteria that the FTA must consider in order to advance a New Starts project through the project development process and to enter into a funding agreement. SAFETEA-LU requires that FTA consider the economic development effects of New Starts projects, however this criterion was not required for the FY 2008 and FY 2009 evaluation cycles because FTA "desires through the rulemaking process to work with the industry on the development of appropriate factors for measuring the economic development effects of candidate projects." Economic development impact is currently included as an optional measure under the “Other Factors” project justification criterion. This criterion is documented by project sponsors in a "Making the Case" report that is submitted to FTA. Specific reporting requirements are not provided. FTA published the Proposed New Starts Economic Development Criterion - October 2008, which lays out a method and reporting requirements for a new, stand-alone economic development criterion first applicable to FY 2011 projects. This criterion is based on the developability of land near stations, the presence of transit-supportive plans and policies, and the economic climate. Land development impacts of proposed fixed-guideway projects are documented through population and employment forecasts, tax assessment data, a build-out analysis of the total additional development that could be accommodated under existing or proposed zoning, and a subjective market assessment by a local analyst.

9 Transit-supportive plans and policies are defined by FTA as those that support pedestrian mobility and accessibility, and include pedestrian network connectivity, building setbacks, parking design, requirements, and regulations, the land use mix, and residential and commercial densities. These are documented through an inventory of relevant plans, policies, and ordinances, as well as a narrative description of potential barriers such as environmental contamination. Economic climate is documented through long-term metropolitan growth forecasts, recent growth in station area and project corridor property values, commercial and residential rents, and commercial vacancy rates. Academic literature review A rich theoretical literature can be applied to the question of how a transit project might increase economic activity. The empirical literature is less detailed but also useful. Our research focuses on “additional economic benefits”—benefits of transit investments beyond travel time and/or cost savings, the capitalization in land prices of those savings, or the higher development density that may ensue. Travel time and cost savings, and their immediate effects, can be more readily calculated; such estimates are already required in FTA guidelines for funding applications. Transit projects are hypothesized to have several types of additional economic impacts, closely related to each other. These impacts are called “wider economic impacts” by the UK Department for Transport. They are related to the “induced” impacts of transportation investments—the economic and population growth that have been shown to follow some transportation investments—but they are entirely distinct from those changes. In fact, in the UK, only the interactions between firms enabled by lower travel times are included when calculating agglomeration economies. It is those interactions that matter in increasing productivity. What is agglomeration? Agglomeration is perhaps the most widely observed feature of the spatial organization of economic activity, evident in the existence and growth of cities, in the formation of industrial regions and districts, and in the clustering of like activities within a neighborhood. Agglomeration is a term also used to refer to interactions between firms and households that may be made easier by transportation improvements without any such physical changes in the built environment. There are multiple kinds of agglomeration mechanisms that may result in economic benefits. These mechanisms include greater innovation due to more frequent contacts among a specialized labor force, reduced costs of producing goods when production equipment and knowledge are shared, and better matching of workers to firms. Which of these mechanisms might arise from transit projects is an understudied question, but an important one. Most additional economic impacts of transit investments can be characterized as higher industrial productivity due to greater agglomeration economies enabled both by faster travel, and by densification of development near transit stops and across the metropolitan area. While “agglomeration” perhaps most commonly refers to the size or density of a city, or of an industrial cluster, it also refers to industrial concentration within an urban economy, or the percentage of total employment in a particular industrial sector, or even increases in “effective density,” which is essentially equivalent to decreases in travel time. There are various causes of such economic returns to firm concentration, including access to large and specialized labor markets, better sharing of a variety of inputs to production, and the

10 rapid dissemination of specialized production knowledge. The relative importance of each of these agglomeration mechanisms is not well understood empirically, though there is a well- developed set of theories. Theory has it that concentration of economic activity, such as a city or a business park, takes place because of economic returns to doing so. The existence of returns to scale of concentration is demonstrated by the fact that settlement is organized in cities, rather than scattered across the landscape. Agglomeration economies are to a large extent external to firms; that is, firms do not capture all of the benefits of their decision to locate near other firms. Those benefits instead accrue to all members of the agglomeration. When choosing a location, a firm will take account of the benefits of concentration on its own activities and profitability, but not the impact of its own decisions on the activities of other firms. Because of this, firm clusters, CBDs, and cities themselves may be less concentrated than is optimal. There are two main ways that transportation improvements may affect economies of agglomeration: • If a transportation improvement causes the relocation of activity (i.e. densification), there may be positive agglomeration impacts where the activity relocated to (and negative impacts where it relocates from) • The lower cost of movement will increase accessibility to economic activity, especially by increasing the pool of accessible labor; this may lead to further positive agglomeration impacts. These various agglomeration effects can be evaluated as they might relate to transit (see Table 1, below). Each implies a different possible set of empirical measures and estimation methods to determine whether these economic impacts are in fact large enough to help distinguish proposed projects, and if so, how best to estimate the impacts.

11 TABLE 1 Transit projects and hypothesized agglomeration mechanisms Agglomeration mechanism Likely facilitated by transit projects? Input sharing (enabling horizontal vertical disaggregation, that is, small, nimble firms instead of large ones; and supplier specialization) No, unless transit projects reduce freight congestion Knowledge spillovers (skilled labor learning from each other; quick dissemination of innovative practices) Indirectly, (a) by facilitating local and walk- accessible firm concentrations, and possibly (b) by increasing speed of business travel (but only where transit is used for business travel and/or where transit reduces road congestion) Labor market pooling (enabling better matching of workers to jobs; less turnover) Yes, by increasing the size of the labor pool within commuting distance Reduced cost of negotiations (enabling vertical disaggregation and supplier specialization) Indirectly, by facilitating local and walk-accessible firm concentrations Infrastructure sharing (closely related to economies of scale in transportation provision) Yes—there is shared access to transit infrastructure, but this particular mechanism of agglomeration is already included in FTA guidelines by counting travel times along with anticipated densification near stops. Amenity sharing (specialized public and private goods) Yes—but this is a consumer-side benefit, and is hard to measure except very indirectly, via land prices Empirical studies of the effects of transportation investments on agglomeration are diverse and incomplete. Depending on the causal theory being tested, the studies use different dependent variables: changes in productivity, firm revenues, wages, and land values are all examples. Studies also use different independent measures: accessibility changes, and changes in density, are the most common examples. The unit of analysis also ranges widely, from the metropolitan region to small areas. There is little or no direct study of how transit investments might have economic impacts beyond travel time savings. The uncertainty on the empirical side is illustrated by Table 2 below, which gives an example of how in the case of different levels of pent-up demand the theory fails to shed much light on the specific effects of any particular transportation investment. The response of travelers and development to projects will vary; in some cases costs will exceed benefits and in other cases the reverse will be true. Additional economic benefits may be a substantial fraction of project benefits in some cases, such as a project that helps relieve a major bottleneck or that occurs in a region with industries that benefit from agglomeration. Local context and conditions can make a difference and any single study will not be able to disentangle all the possible associations.

12 TABLE 2 Possible variation in additional economic benefits Cost Time savings “Other” economic benefits Notes Example 1 $1.5bn $0.80bn $0.10bn Low demand for project Example 2 $0.50bn $0.75bn $0.50bn Relieve transportation bottleneck Example 3 $0.50bn $0.10bn $0.20bn Post-project travel/congestion Example 4 $2.50bn $0.90bn $1.80bn Ability to attract agglomeration of firms There may also be other additional economic benefits of transit investments that are related to but distinct from agglomeration effects, including: • Transit network effects • Economic multiplier effects • Reduced income-tax-related deadweight loss • Greater price competition The first of these is the increasing returns to scale in waiting, walking, and transfer times that come with higher-density transit networks. Some transit investments may reduce waiting, walking, and transfer times, but these depend on specific details of the transit network and also of the corresponding pedestrian network. The benefits associated with network effects will likely be enhanced by having a large existing transit network that is well integrated with a pedestrian network. The second is economic multiplier effects. These are typically estimated by regional economic models, such as the REMI (Regional Economic Models, Inc.) model, which addresses the connection between transport costs and productivity via the ability of firms to access labor markets, and the potential variety and concentration of those labor markets. Calculating such effects is beyond the scope of this study, but economic multipliers may be intimately related to agglomeration economies. We describe such models below and in Appendix E. The third of these is the additional value of labor market participation that goes beyond increased wages. Agglomeration can lead to higher employment rates or more work hours in several ways, such as improved transit access to jobs encouraging more unemployed residents to join the labor market. If a proportion of the time saved traveling is spent working more hours; or if local labor supply increases at highly productive (and agglomerated) locations, deadweight losses associated with taxation of income may decrease, which would be another economic benefit. The fourth of these additional economic benefits is related to but distinct from agglomeration effects. Lack of accessibility and high transportation costs can be a barrier to competition between firms. These effects are likely to be minimal given that transportation networks (particularly the road network) are mature in the US, so transit investments would likely have a trivial impact on increasing competition between firms. Lower prices and more competitive markets effects may occur only if the transit project reduces major congestion of

13 roads that are used by freight, leading to the possibility of effectively larger market areas for some industries. Interviews We conducted eighteen practitioner interviews with transit agencies, metropolitan planning organizations (MPOs), and their technical consultants in the US, UK, and Australia from January 2009 to April 2009. The purpose was to obtain their assessment of how the economic impacts of proposed transit projects are assessed in practice, including what works well and what does not, how methods might be improved and made consistent among different locales, and what types of products from this effort would be useful. We sought to understand current techniques, expertise levels, and needs of various agencies. We asked questions about the methods currently used to forecast economic and non-economic benefits of proposed projects; data quality and availability; and transferability across different regions and transit modes. The US interviews also addressed how the assessment of economic impacts relates to the New Starts/Small Starts federal funding process. An interview guide was developed to ensure all interviews followed the same structure and collected similar information. A copy of the interview guide is included at the end of Appendix D. The guide covered the following topics: • Benefits measures used in the transit project development process; • Methods for both forecasting and ex-post monitoring of economic impacts; • Data availability and ease/cost of acquisition, including any systematic delays or difficulties; • Any local, customized modeling tools for forecasting impacts and their relative success; • Examples and evidence of good practice and problems encountered previously relevant to the research objectives; • Transferability of methods between regions and project types; and • Any other views on what a practical evaluation methodology should consist of. Interview subjects in the US included transit agency staff, their consultants, staff of MPOs, and a representative of APTA. Outreach to identify interview subjects was conducted at the January 2009 Annual Meeting of the Transportation Research Board in Washington, DC, through presentations at select committee meetings. Subjects were also selected from recommendations made by the TRB Review Panel, through direct outreach to transit agencies, and through personal contacts of the researchers. Interview subjects in the UK and Australia included transit and other government agency staff and their consultants. Members of the research team interviewed a diverse set of transit professionals with experience in rail, light rail, and streetcar projects in mature, developing, and planned systems in a variety of metropolitan areas. Subjects included technical staff and consultants with broad experience in project planning and economics, as well as agency leadership with knowledge of the political complexities of transit development. In total, 18 interviews were conducted, primarily via telephone (thirteen in the US, four in the UK and one in Australia). More detailed accounts of the interviews are presented in Appendix D. Here we summarize the most salient results. The interviews suggest there is no standard practice in the US transit industry for conducting economic benefits analysis. Respondents cited a variety of methods, including cost–

14 benefit analysis, proprietary models such as REMI and TREDIS, input-output models, and real estate investment studies (these are discussed in the next section). A large share, or in some cases all, of the calculated economic benefits in these studies amount to monetized travel time benefits—a form of double-counting, since travel time reductions are already reported to the FTA as part of the application process. These methods lack an explicit attention to productivity-increasing responses to transit investments. Respondents were generally unfamiliar with agglomeration or labor search, and how they might lead to economic benefits from transit investments. Interviewees also commonly wanted to discuss the New Starts/Small Starts application process. The process was generally reported to be cumbersome and there was reluctance to increase the complexity of evaluation, which is what is implied by an additional “economic benefits” criterion. Respondents generally expressed the interests of their agencies. They preferred finding a process that would reflect positively on their agency’s local circumstances. Many felt the existing process was biased against their agencies. A common theme among respondents was that the unique economic and land development contexts of different regions, as well as the relative maturity of systems, should all be taken into consideration. Respondents also expressed a strong interest in having explicit environmental benefits considered as part of the calculation of economic benefits. In the United Kingdom, the Department for Transport (DfT) publishes standard guidance for overall assessment of transportation investments, which was frequently cited by UK respondents as a key resource. The guidance is standardized across different types of transportation projects and only contains mode-specific advice where there are technical modeling issues, and in parameters such as the value of time. It covers a range of areas including environmental impacts, safety impacts, economic impacts, and accessibility; it utilizes both quantitative and qualitative measures, which are compared to national standards. The guidance includes a methodology for assessing additional economic impacts such as agglomeration economies. In contrast with the UK, most practitioners in the United States are unaware of the ways in which transit is thought to have the potential to increase economic productivity. To the extent that they are, there is little consensus about how impacts should be measured. Practice reports and guidance To supplement our review of the academic literature and our practitioner interviews, we conducted a review of current practice in the US and the UK for assessing the economic impacts of transit projects, focusing on reports and practice guidelines. In the US, these consist largely of the results of input-output models and integrated land use and transportation models. Such models do not explicitly attempt to estimate transit-caused increases in economic productivity. The economic multiplier effects due to travel time reductions may vary from region to region. As we note below, however, it is difficult to conceive of a clean way to make estimates distinguishing regional economic impacts on this basis. Models used to calculate these impacts are region-specific and can be tweaked to yield desired answers. Table 3 summarizes the methods used by the sample of studies that we were able to gain access to and found most relevant. Appendix E provides details on each of these studies.

15 TABLE 3 Summary of methods used by sample studies Study Cost–benefit Analysis Input-Output/CGE Real Estate Ad Hoc Chicago Metropolis X TREDIS Access to the Region's Core (ARC) X REMI X Metrolinx X Proprietary X MAROps X REMI X New York Cross Harbor X REMI Chicago Region Freight (CREATE) X REMI X Howland Hook Marine Terminal X NJ Transit Consumer Spending X X Portland Streetcar X DART Fiscal Impact X X Phoenix Metro Light Rail X DART TOD Guidelines X X Based on the information from the interviews, we conducted a detailed examination of REMI and the UK method, and discuss the latter in some detail in Appendix E. Many of the other methods used by agencies are relatively ad hoc, or are studies aimed at gaining local support for specific projects. Based on our review, none of the US studies tackled the productivity-increasing effects that are the focus of this study. Most studies used input-output analysis to track the local employment impacts of construction spending. Others relied on traditional travel demand modeling, mainly to assess direct travel time benefits. The REMI model is one of the more widely used models for evaluating economic benefits, not only of transportation projects, but of other regional investment strategies. We studied publicly available information on this model to determine whether it evaluates agglomeration benefits, as this was unclear in the studies we reviewed. We concluded that it was not possible to tell from available documentation. More details on this issue are provided in Appendix E. The UK guidance has the benefit of being clearly written, non-proprietary, and more explicitly linked to economic theory. While this guidance is still in draft form, it is being applied by consultants and local authorities to projects within the UK. The first and most obvious disadvantage of this guidance for analyzing transit projects in the United States is that the agglomeration elasticity estimates are based on UK data. There are other issues, as we note in the framework below. Agglomeration benefits in the UK approach are calculated based on estimates relating employment accessibility (rather than development density) to economic output measures. The accessibility measure accounts for the concentration of employment within a zone and interactions with every other zone, discounting by an exponential function of travel time, or more accurately, by the generalized cost of travel. This generalized cost estimate is taken from a travel demand model, with and without the specified project. The change in accessibility is converted to productivity gains using empirical evidence on the relationship between effective density and productivity, or agglomeration elasticities. The UK method also estimates labor market impacts separately, although these effects are possibly more clearly understood as another particular kind of agglomeration economy (as

16 defined in the academic literature; see Appendix C). The two most relevant for transit projects are increased employment due to increased access and increased productivity due to a larger number of jobs in more productive firms and industries. The increase in labor supply (that is, more people choosing to work) is spurred by the effective reduction in the costs of working by reducing transport costs. This is an important point, especially when considering the distributional impacts of investment and how transport costs disproportionately affect lower income employees. In order to estimate this properly, one must derive an elasticity of employment with respect to wages. Estimates are likely available in the literature and could be used, although these would not be locally specific. In the UK approach, the increase in accessibility is also considered to have the potential to increase local labor supply to dense and highly productive locations. In the presence of income taxation, any resulting productivity increase will carry a component (i.e. the tax element) that is not counted elsewhere. This element may simply consist of a transfer between workers and the government, so it is not clear whether it should be included as an economic benefit.