Lessons Learned, and a Framework in the Sky - Economic Development

The first two and last two lessons were addressed in a TRB paper presented in January 2003 [4] and will not be substantively addressed here. The typology that was addressed in that paper will be briefly reviewed, however, and the remainder of this paper will address the quantifying methodologies and present a framework within which to evaluate economic development highway projects.

The Typology

As determined in the earlier paper (January 2003), there are six types of projects. Each type is associated with a set of primary objectives and a set of ex-ante analysis considerations that are to some extent objectively addressable. The typology matrix in the earlier paper is reproduced below.

Table 1 - Typology of Projects, Objectives, and Ex-ante Analysis Considerations

Project Type	Primary Objectives	Ex-ante Analysis Considerations
1 Better access to employment or production or distribution centers within a generally rural area	Serve specific existing employers or centers and encourage new compatible employers or centers	Jobs maintained/generated, plant investment, tax base effects
2. Connectivity and/or circulation improvements to service roads, access roads, and relief roads that supplement a major highway through a generally rural area	Promote and accommodate commercial development along major highway	Revitalization potential, jobs potentially generated, tax base effects, new businesses created
3. Connectivity improvements between cities or between a city and a production area through a generally rural area	Promote and accommodate commercial development within a major corridor and at both ends of the corridor	Potential developable sites, interest in existing employers to expand capacity, investment leveraging considerations, tax base effects
4. Better access between a workforce and a production center within a generally urban area	Serve existing employers or centers, and encourage new, compatible employers in existing centers	Workforce availability to employers, jobs maintained and/or generated, business investment likely, tax base effects, diversification of employers
5. Better connectivity between one production/distribution center and another within a generally urban area	Serve existing employers or centers, encourage new compatible employers in existing centers, and create potential for more high-prestige businesses	Workforce availability to employers, Jobs maintained and/or generated, business investment likely, tax base effects, possible new industries (including high wage or "high potential growth: employers), diversification of employers
6. Access/circulation changes to support changing land use in either a rural or urban area	Protect or enhance quality of life; facilitate changing character of employers	Business investment likely, tax base effects, higher wage employment, diversification of employers

Initial Analysis Context

The next step in analysis of the economic development potential of a project requires assessment of the statutory context. A number of states have economic development programs; however, few of these states have programs with a scope containing more than one or two types of projects. Typically, a state program will simply require selection between projects of the same type.

After classifying projects into the six types noted above, the next step is disaggregating them to differentiate between those that are in areas that are economically distressed from projects not in such areas. There are a number of ways to do this. One way is to use any pertinent state specific statutory language. Another way is the use of specific criteria. In the TRB 2003 paper, two criteria in particular (population loss and high unemployment) were developed for illustrative use. The FHWA is not, at this time, recommending that states use these specific criteria, but the FHWA does consider it desirable that some criteria be used. An elected member of the Alabama state legislature pointed out how important it is to consider the issue of economic distress. [5]

The second distinction is disaggregating projects where improvement advocates have both a realistic cost estimate for the improvements and specific ideas about what kind of employers might be established (or expanded) if such improvements were made. One lesson learned in the Initiative was that both a realistic cost estimate and some thoughts on potential employers were required to produce a realistic economic development study. If this information is not available in applications or in the initial project information, it must be acquired as part of the economic development study. Obtaining a cost estimate will typically require discussions with a state or local infrastructure agency. Obtaining the potential employer information is sometimes the byproduct of obtaining a baseline of the area's economic structure.

After the initial analysis, there are, in effect, four subtypes of projects within each major project type: The table below shows this subtypology within the type 2 projects.

Table 2 - Subtypology of Projects, Economic Distress Classification, and Information of Project Cost and Potential New Employers

Project Type (illustrative)	In an Economic Distress Area	Initial Analysis Information Availability	Presumptive Rating from Initial Analysis Information
2.	Yes	Yes	Best
2.	Yes	No	OK, improvable
2.	No	Yes	OK
2.	No	No	Weak

It can certainly be argued that (a) projects in economically distressed areas, and (b) projects that already have initial analysis information available should be given priority over projects without these characteristics. The former consideration is based on the principle that economic development in areas that are not economically distressed is inherently less of a public good than economic development in areas that are economically distressed. The latter consideration is based on specific experience with the Initiative, which determined that the cost of creating the initial analysis information ranges from about $5,000 (for a small-scale project) to about $50,000 (for something like a 100- mile-long corridor).

A possible additional distinction is between projects that represent a "new" or "newly recognized" need and those that represent a recognized need but where highway improvement advocates wish to accelerate the implementation of the improvement. For now, the FHWA is not making a judgment about which is better.

Methodologies of Detailed Analysis

The jobs generated due to construction of a highway improvement may or may not be a consideration in choosing between two or more highway economic development investments. If the jobs would be created in an area of economic distress there would be more justification than otherwise, but clearly the number of highway jobs is generally proportional to the cost of the highway. If an estimate of the number of jobs is to be generated, there are a number of methods of doing it. In the initiative, the RIMS II model (an input-output program created by the U.S. Department of Commerce, Bureau of Economic Analysis) was used a number of times, but there are other models. A member of the U.S. Congress, while attending an initiative function, emphasized how important even temporary construction jobs are in economically distressed areas [6] .

The more difficult analysis task is in developing an estimate of how many permanent jobs are likely to be created from a given highway improvement. In all cases in the Initiative, this required some knowledge of non-highway economic development efforts. Within the context of the Initiative itself, this analysis was done using a variety of methods; those methods are enumerated below.

A: Using estimates from other studies. In several of the corridors studied in the initiative, previously completed economic development studies had estimated the employment created for specific activities (e.g., wind energy farms, biotech production plant). The sizing of employment and investment in plant and equipment is drawn from firms in the industry. Analysis of corporate governance reports of employers [7] provides information on the typical size of facilities, the investment cost, and the number of full-time employees. In this case, all that needs to be done is to determine what percentage of the jobs can be considered directly dependent on the transportation improvement being implemented [8] .

B: Using extrapolations based on local interviews. In several of the corridors, local officials were queried. They often had reasonable estimates on what type of development would likely be created if a specific highway improvement were to be completed. This approach requires an optimistic outlook; to some extent, it implies a belief that highway improvements are the "last" or nearly the "last" obstacle to capturing economic development opportunities. In the Initiative, the FHWA found that, once the highway improvement was defined realistically (not always a trivial task), the local officials gave reasonable estimates on development. Similarly, a number of state DOTs have found that when local officials take responsibility for an estimate of project jobs, they provide realistic estimates. In this case it was necessary to determine employment at existing businesses in the general region similar to those that would be created upon completion of the highway improvement [9] .

C: Using the assumption of "induced regression to the mean." In several corridors the initial analysis showed that the area was deficient in retail or other specific types of business activities. Based on interviews with local officials it was determined that the provision of the highway improvement would provide access and in some cases make land available for development. The highway improvement was then deemed to be the critical element to induce additional development so that the area would reach approximately the same business density observed in peer areas. In this case it was necessary to determine employment levels of the types of businesses that would need to be added to reach the peer business density level [10] .

D. Using the assumption that targeted businesses would develop to serve a defined customer base. In one case a defined level of tourism was predicted based on a non-transportation model. In that case it was necessary to determine the employment needed to serve that level of tourism. It was also necessary to state that both highway improvements and other improvements were required to meet the requirements for this level of tourism [11] . The linkage of highway investment to economic development was made by specifying what access was needed to expand the tourism market, then what internal circulation was needed to fully serve tourist venues.

E. Using the assumption that business density would increase in specific industry groups. This assumption was based on previous studies, but with some modifications made to reflect the implementation timetable for the highway improvement and other developments. In this case it was necessary to essentially reexamine a previously completed study. This earlier study made overly optimistic assumptions about the completion of a highway improvement. Also, changes in the character of the work force and business mix had occurred since that time, but the basic vector of the early study was still valid [12] . This type of analysis is typically called "cluster analysis."

F. Using the assumption that business density in a specific industry group (for a new industry-specific corridor) would resemble business density for the same industry in a peer corridor. In application, the industry in question should both (a) be growing, and (b) have an existing presence in reasonable proximity to the development speculated for the highway investment. In this case it was necessary to disaggregate highway-specific determinants in the peer case to be applied to the new corridor [13] . The industry growth is then reallocated, with some portion attributed to the corridor where the highway investment is presumed to occur.

G. Hybrids. Various combinations of the above six methods were used in other corridor studies.

H. Other methodologies. These methodologies apply to the project types noted in the table below. If more urban projects had been part of the Initiative, it is likely that other methodologies might well have been used.

At this time, the FHWA is unable to provide guidance on when a particular methodology should be used or when it shouldn't (this would be an appropriate subject for a future research effort). However, there is reason to believe that projects analyzed using more than one methodology may be more reliable simply because of the internal check.

Table 3 - Typology of Projects, Methodology to Estimate the Number of Permanent Jobs

Project Type	Methodologies
1.	A, B
2.	A, B, C, G
3.	A, B, C, D, E, F, G
4.	No studies of this type were done in the initiative.
5.	No studies of this type were done in the initiative.
6.	No studies of this type were done in the initiative.

Framework for Evaluation

A reasonable way to assess projects against each other is to put them all in an evaluation table. It should be readily apparent that comparing different types of projects requires additional considerations that are not easily quantifiable. It should also be readily inferable that comparing projects that were analyzed using different methodologies requires similar considerations. Notwithstanding these caveats, one could align projects in a table as shown below. The columns marked "Other Economic Factors" and "Other Non-Economic Factors" could be used to give such information as the work scope of the project, the implementation time required, the tax base consequences of the project, etc. Since there is no purely quantifiable way to rank projects against each other, use of an evaluation panel (where the person who puts together the matrix is not one of the panel) seems a reasonable way to proceed. If the project types under consideration are limited to one or two, it is reasonable to expect this framework will lead to defensible selections.

Table 4 - Illustrative Evaluation Matrix for Projects (all values illustrative)

Project Type	Rating (from table 2 factor)	Project Cost (in thousands)	Analysis Method	Estimated Permanent Jobs Created	Other Economic Factors (e.g., tax base)	Other Non-economic Factors (e.g., complexity of project)
2.	Best	$10,000	A, B	1,000	.....	....
2.	OK	$ 5,000	A	300	.....	....
3.	OK	$35,000	A, C, D	2,000	.....	....
3.	Weak	$55,000	A, C, D, G	2,500	.....	....

A Word about Multipliers

Corridor improvement advocates have on a number of occasions expressed a considerable interest in the concept of multipliers. Conceptually, by attracting businesses with high multipliers, a project could maximize the number of jobs created [14] . However, in practice, the types of businesses that are attracted to an improved corridor are somewhat limited; they generally are businesses that are similar to or logistically related to existing businesses, or to embryonic industries [15] .

Final Thoughts in Song*

This song summarizes FHWA's economic development efforts preceding and including the Initiative. Penned by FHWA's Martin Weiss, it is sung to the tune of Ghostriders in the Sky (with apologies to Stan Jones who wrote that song in 1949 and the many people, e.g. Johnny Cash, Lorne Green, Burl Ives, etc. who've sung it on records).

Econdev Initiative - born one dark and windy day.
Corridors established to be built upon the way.
Meetings to advise on just what jobs they saw.
And quantifications of jobs upon which to draw.