Martin Weiss, Transportation Specialist, FHWA
HEPI-20, 1200 New Jersey Avenue, S.E., Washington D.C. 20590
202-366-5010 (v) 202-366-3713 (f)
The evening after beginning the narrative, I was listening to the radio. They were playing the song "Ironic" from the 1995 album Jagged Little Pill by Alanis Morrisette. I thought that the two contradictory trends noted in the introduction represented a form of irony. Then I noticed that the other portions of the paper reminded me of other songs on the album or of the structure of the album itself. When I gave the presentation to at the TRB summer meeting in 1999, people in the audience seemed to see the same irony, jaggedness, etc. so the final TRB published (in a compact disk) version left these comments in the section heading. Since then, some people who weren't at the meeting told me they liked the heading because it was a type of mnemonic device. Of course, other people didn't get it, but even there it still served as a mnemonic device anyway. On balance, the headings do more good than bad and they are retained in this internet version of the paper.
In antiquity, commerce thrived along the royal roads of the great Empires. In our own day, technology growth corridors are identified by the highway that provides them access. Strange then, that there would be controversy on the proposition that transportation improvement has potential for economic growth. Yet controversy exists. Indeed, the controversy is so extensive that the 'transportation improvement brings economic growth' proposition, especially where 'transportation improvement' means 'highway improvement' has been contested on many different, sometimes mutually incompatible, grounds. These challenges are sometimes stated as fact, sometimes as conjecture, sometimes as stories and to some extent have been the basis of statute (e.g., physical capacity increases not being eligible for CMAQ funding). A sample of such challenges, which have been taken from 'anti build' themes of newsletters, seminars and conferences of various groups, follows:
Along with this, another trend is underway. Many economic development policy interests are finding that other prospective economic development tools, such as preferential taxation mechanisms and regional education initiatives are becoming more problematical. These interests are turning to transportation, usually highway, improvements as a tool to bring commerce to depressed or low job growth regions. One manifestation of this latter trend is the number of highway corridor advocacy groups, some of whom use the interstate shield in their group name as a symbol of their vision for a corridor. To some extent, these visions have also been given a basis in statute (e.g. future Interstate status to certain high priority corridors).
A little too ironic, don't you think.
Since the Interstate era, considerable research has been done to try to understand the relationship between transportation improvements and economic growth. In the 60s, an HRB publication addressed this issue as, "Will Model Building and the Computer Solve Our Economic Forecasting Problems"[i]. Through the 70s, it was partly model development and partly empirical analysis of a regional nature. From the 80s on, the use of econometrics and other economic modeling, especially on a national scale has been increasing, sometimes supplemented with interviews to establish cost and utility functions within the modeling. Regional empirical studies became less common.
During this later period, individual research efforts, with various degrees of clarity and success, tackled the problems of data availability and accuracy, defining 'economic growth', defining 'transportation improvement', identifying hidden or deeply buried analysis assumptions and data interpretation conventions, separating causes from coincidences and explaining results in an understandable way. Along the way, some of these efforts attracted considerable attention by seeming to predict: counter intuitive results; inexplicable outcomes; or, outcomes, which while using seemingly reasonable assumptions, were inconsistent with other outcomes. Substantial numbers of conferences, seminars and so forth resulted. Researchers and policy practitioners sometimes supported, sometimes criticized each other over data, analysis techniques, interpretation and so on. Of course, one crucial group of issues is inherently difficult in such modeling, namely job growth, job quality and related matters.
At this point we don't know enough to make quick and confident judgements regarding economic development outcomes for use in developing TIPs, STIPs, Transportation Plans or environmental documents. Completing these documents requires decision brokering trade offs between improvements in different jurisdictions, between different types of transportation improvements, between maintenance type projects and increased capacity type projects and between projects that support different, sometimes contradictory regional goals.
In an effort to, at the very least, ground ourselves in information that is broadly understandable and closer to the 'taste, touch and smell', FHWA has made some efforts to renew the regional empirical element. To begin with, it will be instructive to review some of the work of the 70s.
In 1970, FHWA submitted a report to Congress[ii] in which a wide variety of impacts were discussed. Two are worth reviewing at this time.
One significant effect of the Interstate System was on employment. During a five year period (1958-1963), job growth in high density urban areas with topographic barriers of some kind (rivers, hill) served by freeways was compared to job growth in similar urban areas without freeway service.
|Benefits of the Interstate Program|
|Job Growth Per Thousand Population|
|With Freeway Access||Without Freeway Access|
|Fast Growing Cities in SE, Midwest, NW||43||23|
Deeper into this early report, it seems apparent that improved access had increased land values and the ability of local business to market products. A follow-up report[iii] noted that population had migrated to those high job growth areas, and, in some cases, the population migration was faster than the job growth.
By 1980, a report on the impact of the Interstate System on non metropolitan areas[iv] had been released which included analysis of 1960-1975 data. The effects were similar to those noted earlier for densely populated cities. Of course, the positive relationship was not universal. For example, some counties, that contained an Interstate highway, had no interchanges and thus were not served by the Interstate. It would be unsurprising if in some of these counties, employment dropped when the Interstate was constructed.
Of course, even by the 70s, much of the highway program funds were being expended to improve non Interstate highways and, in the 80s, the highway program funds began to be used for improving transit, and in the 90s for such uses as transportation enhancements.
Thus, by the 90s, relatively little of the highway program was being expended on new Interstate highway construction. Furthermore, the domestic economy has evolved steadily and is now less oriented to manufacturing and mining and more oriented to information technology. Hence, the degree to which 70s research still applied is open to some question.
Four studies done in the 90s seem to shed some light on the relationship of transportation improvement and economic development. Two studies concentrated on the Appalachian region, which includes portions of 13 States and two concentrated on the Lower Mississippi Delta region, which includes portions of 7 States.
The first of the Appalachian region studies[v] looked at income. This was considered appropriate because the goal of the Appalachian Development program was to promote economic development and decrease poverty in the region. The study examined which low income counties (defined as having per capita income less than two thirds the national average) were successful in climbing out of that situation. Some examples of success and non success were:
|Poverty Alleviation in Appalachian Region|
|State||# of low income counties in 1965||# of low income counties in 1990|
The study notes a number of reasons for the relative success of Pennsylvania, North Carolina and Georgia and the relative lack of success of West Virginia and Kentucky. One reason is that the 80s was the decade in which the mining industry restructured and retrenched. Indeed, West Virginia had only 16 counties in the low income status in 1980 and Kentucky had 35. However, two other reasons were noted. The first was that the low income counties in those two States were more distant from metropolitan areas. The second was that, in these two States, substantial portions of the highway corridors to be improved as part of the Appalachian Regional program were not yet improved.
The second of the Appalachian region studies[vi] compared 391 of the counties in the Appalachian region to 391 nearby counties and parishes outside the region which were demographically similar (no matches were sought for the two counties added to the Appalachian region in the early 90s and no match was found for 6 counties in the region). The comparison included changes in demographics and earnings by major economic sector.
|Relative Income Growth Sectors in Appalachian Region|
|Percentage by which the Appalachian county income increased more than its counterpart.|
|Finance, Insurance and Real Estate||135%|
The effect of the Appalachian highway program may be partially understood by examining the differences between the growth rate of subsets of these counties. One comparison shows the relative success of counties which had at least 3 miles of Appalachian highways and those which had an interstate highway or who had growth center funding (the latter was a program that provided substantial funding in the 60s and early 70s). It is useful to note that the counties with interstates include those with no interchanges or with only a diamond interchange and the growth center counties include those containing centers which didn't grow.
|Benefits of the Appalachian Highway Program|
|Percentage by which the Appalachian county increased more than its counterpart in Per Capita Income.||Number of County Pairs|
|Appalachian Highway Counties||32%||110|
|Appalachian Growth Center Counties||14%||90|
|Appalachian Interstate Hwy Counties||15%||152|
The first Lower Mississippi Delta region study[vii] concentrated on employment because of a 1990 CNN video report. In this report, the virtually unanimous opinion in on-camera interviews was the first priority of the region should be more jobs. This, despite the well documented problems in education, health, etc. The study itself was undertaken as a follow up to a Commission established in the late 80s which had resulted in 55 transportation improvement recommendations (the vast majority of which were highway related). Two participants in the Commission are worth noting. The chairman of the Commission was then Arkansas Governor Clinton, who at the time of the update was President Clinton. A prominent advisor to the Commission was then Arkansas Highway Commissioner Slater, who at the time of the update was FHWA Administrator Slater and shortly thereafter was DOT Secretary Slater.
In the course of doing the study, FHWA found that counties and parishes with job growth at better than the national average, or with average job growth in spite of the loss of a military base, had made substantial highway improvements. Although FHWA was unable to obtain detailed and comprehensive information about improvements in non transportation areas, some such information was obtained. Some examples of these success stories showing both transportation and non transportation influences were:
|Caldwell, Lyon and Trigg||KY||+22%||major highway improvements
|Rapides||LA||+8%||major highway improvements
airport access improvement
loss of England AFB
|Howell||MO||+16%||some highway improvements
more improvements programmed
employee/local govt programs
|Madison||TN||+25%||many highway improvements
improved airport access
improved education facilities
expansion of gaming industry
|Mississippi||AR||+7%||many highway improvements
local self help efforts
loss of Eaker AFB
|Tunica||MS||+80%||major highway improvements
expansion of gaming industry
The second Lower Mississippi Delta region study[viii] was a follow up to the update. The objective was to estimate how much of the increase in employment in seven non metropolitan counties and parishes (Avoyelles Parish, LA, Crockett County, TN, Drew County, AR, Howell County, MO, St. Francis County, AR, Trigg County, KY and Tunica County, MS) with rapid job growth could be attributed to transportation improvements. Since, among other problems, there is no way to obtain this information directly, the methodology was to use focused interviews, which is an indirect, rather than direct empirical method and, which is also subject to various possible errors (for example interviewees may subconsciously or consciously tell interviewers what they think the interviewer wants to be told). However, focused interviews do have the advantage of aggregating opinions of people who do have substantial direct evidence, and to do so at a reasonable cost. The principle investigators, subsequent to review of interview information plus other information publicly available, made the following estimate:
|Component of Job Growth Attributable
to Transportation (mostly highway) Improvements
|Counties/Parishes without gaming||40% to 65%|
|Counties/Parishes with gaming||25% to 40%|
In the course of the interviews, the officials being interviewed provided numerous direct links between improved highways and job growth. In the counties without gaming, these, typically, were observations such as, "... and the XYZ company moved in here to take advantage of 4 lane highway ... and...". In the other Counties, a typical observation was, "... we knew the casino would need a 4 lane highway but would generate tax revenue in return, so we....".
These responses to interviews also show two general causal sides to the transportation and economic development relationship, that is, the 'who caused what' issue.
What can we make of all this? Well, for one thing, it seems irrational to hold that the 'anti build' themes cited at the beginning of this paper are, as a general rule, true. Furthermore, one can qualitatively argue that: transportation and trade jobs do tend to be reasonably high in salary; productivity improvements do result from certain transportation improvements; transportation, especially Interstate highway, access features prominently in the advertising that economic development professionals do to attract high paying industry; and, industry centers are frequently named for the highway that serves them, etc. However, it is fairly easy to document or, at least imagine, situations where: freeway construction (or heavy rail line) was followed by decay of certain neighborhoods; or, a highway improvement was quickly followed by the movement of near by low wage manufacturing to Mexico; or, a highway improvement was followed by the establishment of numerous fast food businesses: or when a highway (or rail or bus or other transportation) improvement didn't bring an increase in manufacuring jobs. Thus, there are times when TIP, STIP and plan developers will be less than confident of their projections. It does not seem that we will ever be sure under what conditions highway/transportation improvements lead to a growth in 'good jobs'. It is not even certain whether or not research can answer this question as fast as the economy changes the meaning of the phrase 'good jobs'. However, FHWA does intend to promote an expansion of the scope of economic development before/after type studies to other parts of the country and to improvements other than new access or capacity expansion. These efforts would at the very least deepen our understanding of the transportation improvement and economic development relationship. In the most optimistic case, they would result in some sort of 'rules-of-thumb' to support economic development considerations in projects, programs and plans.
[i] Highway Research Record 149; Forecasting Models and Economic Impact of Highways; National Research Council; 1967
[ii]2. Benefits of Interstate Highways; FHWA, USDOT; 1970
[iii] Social and Economic Effects of Highways; USDOT; 1974
[iv] The Impact of Interstate Highway System on Non-Metropolitan Growth; University of Texas for RSPA/USDOT; 1980
[v] Core-Periphery Models, Regional Planning Theory, and Appalachian Development; Tyrel G. Moore of the University of North Carolina at Charlotte; Professional Geographer: Vol 46, no 3, August 1994.
[vi] The Economic Effects of the Appalachian Regional Commission: An Empirical Assessment of 26 Years of Regional Development Planning; Andrew Isserman and Terance Rephann of the University of West Virginia and the Regional Research Institute; Journal of the American Planning Association, Vol 61, No 3, Summer 1995.
[vii] Linking the Delta Region with the Nation and the World, An Update by the Federal Highway Administration on Progress Achieved in Transportation and Employment in the Lower Mississippi Delta Region; FHWA with assistance from numerous Federal, State, local and quasi governmental agencies and other individuals and agencies; December 1995.
[viii] Lower Mississippi Delta Development Center, Inc. (LMDDC) Final Contract Report; LMDDC and the University of Memphis for FHWA; Sept. 1996.