Introduction

In the United States, the importance of transportation facilities to the nation's economic strength and efficiency is generally accepted. A fundamental requirement of manufacturers is to distribute their products to appropriate markets quickly and inexpensively; people must be able to get to work and to conduct business. And as illustrated by recent disasters -- flooding along the Mississippi and Missouri rivers and the Northridge earthquake in California -- any significant disruption in the movement of goods or people economically impacts a great number of businesses and huge population groups. This recognized link between transportation and economic development continues to justify significant public expenditures in transportation systems at the local, state, and federal levels. Nevertheless, many of the intuitive relationships are not analytically established.

Although evaluations of relatively localized user costs and benefits of investments in transportation projects are possible, it is difficult to quantitatively link such investments to national or regional growth, economic development, industry or national productivity, growth in economic welfare, or the nation's competitiveness in the international market.

The Clinton administration recognizes the importance of transportation infrastructure investment to revitalize the economy and to enhance U.S. global competitiveness. That's why one of the U.S. Department of Transportation's (DOT) strategic goals is to get the economy going and create new jobs by making strategic transportation investments.

The Intermodal Surface Transportation Efficiency Act (ISTEA) of 1991 has given us the tools to improve transportation -- to channel innovative thinking, to harness technology, to strengthen old partnerships and build new ones. These new partnerships blend business, labor, government, and the environmental and educational communities into a united force for economic growth and job creation.

In support of these efforts, the Federal Highway Administration (FHWA) is building on a priority research area, established in 1989, which examines the linkage between investment in highways and bridges and the nation's capacity to sustain economic performance and growth. In this research, the FHWA is acting as an intelligent consumer -- studying all relevant data including previous, related research.

The purpose of this article is to explain the FHWA research agenda -- "Interrelationships Between Transportation Infrastructure Investment and Productivity" -- and to offer a summary of research from this program. (1)

This research will clarify, as quantitatively as possible, both the short- and long-term influences of highway infrastructure investment and service on the economy to make possible more informed decision making and to respond to the productivity slowdowns. The research program seeks to answer the question: "How do changes in highway investment or service translate to private productivity?"

Research Approaches

Three approaches are used in this research. The first approach, macroeconometric analysis, investigates national and state-level linkages between infrastructure capital stock or investment and economic growth and productivity, using econometric methodologies such as the production function. The second dimension, microeconomic industry analysis, explores the connection between individual industries or firms and transportation infrastructure through analysis of logistics relationships. A third perspective, highway system assessment, examines service, network, and system characteristics to determine whether a causal relationship exists.

Together the three dimensions will incorporate leading analytical perspectives to assess the contribution to economic performance and growth of systems, corridors, and projects as reflected in investment programs at local, state, regional, and national levels. Several methodologies and analysis levels are explored to locate the best way to understand the relationship between highway infrastructure and economic vitality. These approaches are only a starting point for a thorough analysis.

 

 

FHWA began investigating this issue in fiscal year 1990. Table 1 shows several projects are ongoing and others are scheduled for fiscal year 1994.

Summaries of specific research projects are presented according to the approach undertaken. Collateral studies with the National Cooperative Highway Research Program (NCHRP), the Transportation Research Board (TRB), and the American Association of State Highway and Transportation Officials (AASHTO) are also included. The goal is to promote an objective and policy-useful understanding of the significance of highway investment and service at the global, national, state, local, and industry levels. While results of individual studies show mixed results, they are presented to emphasize the difficulty of answering the question of how highway infrastructure is related to economic performance.

Background

The study of productivity and economic growth has always been an important issue. However, interest in the sources of such activity rises in times of recession such as we experienced in the early 1990s. U.S. productivity growth has been lower in this decade than in the past. At the same time, the United States has experienced lower productivity growth than other industrialized nations. While several measures reflect the health of the economy, it is difficult to ascertain which measure is the best, since slightly dissimilar objectives are captured in different measures. While one measure may imply a healthy economy, another may show tremendous cause for concern.

Growth implies a healthy, expanding, flourishing economy -- an economy where output is continually expanding. Productivity is the ratio of outputs to inputs. Increases in productivity typically means more output is produced with the same amount of input. Productivity for a specific input can be measured separately. For example, labor productivity measures how much output increases for a particular amount of labor input. Individual productivity can be measured by the amount of product or service produced per hour.

While some economists suggest that growth is not always a panacea, many in the United States are concerned that without continued productivity growth, our standard of living will not continue to rise. Indeed, without productivity growth, the only way for our incomes as a whole to increase would be through a redistribution of the current level of wealth.

In light of this, one might ask, "What causes increases in productivity?" or "What causes declines in productivity?" To answer the first question, increases in productivity result from increases in technology or innovation, increases in labor skills or education, and growth in the quality or quantity of capital stock. However, there is much disagreement over the second question. Table 2 shows several possibilities.

Table 2 - Potential Causes of productivity Slowdowns

• Changes in the composition of the labor force.
• Changes in the growth rate of private capital stock and its flow of services.
• An increase in energy prices.
• Declining investment in private research and development.
• Diversion of capital resources to pollution abatement and government worker safety regulations.
• Mismeasurement of output.
• Shift to a service-oriented economy.
• Declining investment in public infrastructure.

One hypothesis postulates that declines in the quantity and quality of public capital infrastructure investment led to the current slowdown in productivity. Indeed, investment in infrastructure as a percentage of gross domestic product has fallen from 3.4 percent in 1970 to 2.5 percent in 1992. Figure 1 shows the decrease in investment in highways as a portion of all infrastructure investment from 1970 to 1992. While analysts first addressed this question in terms of percentages -- investigating how investment in infrastructure has changed over the years -- this sort of investigation cannot prove a causal relationship. Indeed, such efforts are highly criticized because they neglect several other issues and offer no statistical proof of a linkage. To understand the causal relationships between infrastructure investment and productivity, one must perform a thorough regression analysis -- a statistical process of estimating a causal relationship between an assumed known constant (e.g., highway capital stock) and a random variable (e.g., output growth).

 

 

Several objectives exist and infrastructure's effectiveness in promoting change in each of these areas varies tremendously. In practice, the U.S. transportation objective has focused on mobility and an intuitive sense of the benefits to connectivity rather than specific distributional or economic growth objectives. Economist David Lewis differentiates areas of economic activity where "the influence of infrastructure investment is verifiable" into distributional and growth categories. Infrastructure investment may cause distributional effects on the structure of employment, personal income, regional output and income, and sectorial output and income. Likewise, infrastructure investment can yield growth in economic output, productivity, employment, and economic welfare. (2)

The Macroeconometric Approach

Studies often use national, state, or metropolitan area data in a production function to estimate the magnitude of the causal relationship between the inputs -- labor, private capital, and public capital -- and the output.

Output = function of (labor + private capital + public capital) + error term

This perspective follows the structure and logic of early efforts by economists David Aschauer and Alicia Munnell. They found an enormous role for public capital infrastructure, specifically highways, in output growth. A wide range of elasticities are found in the literature. These efforts address impacts of transportation investment or capital stock on economy-wide output growth. While studies to explore this magnitude have been commonplace in the history of economics, previously, public capital was not included as a production factor. (3,4,5)

Several journal articles have been published that explore the possibility of a causal relationship from public capital infrastructure or specific portions of public capital infrastructure -- such as highways, water and sewer systems -- or education to economic productivity. Other studies investigate the question of complementarity or substitutability between public and private capital.

Sensitivity testing of national-level macro studies

The FHWA study "Highways and Macroeconomic Productivity, Phase 1" was completed in January 1992. This project critically evaluated the hypotheses and conclusions of frequently cited macroeconometric analyses concerning the relative magnitude of effects of public and private capital investment on economic growth. (6)

Michael Nienhaus at the Volpe National Transportation Systems Center pursued replication and sensitivity testing of Aschauer's and Munnell's production function results using national aggregate time series data. Aschauer and Munnell's results are frequently criticized because of the unbelievably large size of the coefficients. Many economists suggest that the results do not indicate causation between public capital and output growth; rather they suggest that the results may arise from simple correlation or that output growth may be causing a rise in public capital investment. (3,4)

Although Nienhaus duplicated key results of the Aschauer and Munnell studies, slight changes to the data and mixing of the data sets led to sensitive results. While Aschauer and Munnell's analyses served as ground-breaking introductions to an important issue, their results are not strong enough to justify the conclusions; other levels of analysis are needed.

State-level econometric analyses

As part of the macroeconometric approach in FHWA's research program, state-level production functions have also been investigated to address the question of a proper scale -- national, statewide, regional, or industry -- for analysis. Indeed, many researchers postulate that the magnitude of the linkage varies according to the scale of analysis. While a relationship is intuitive, the question remains: "Is the relationship so strong that it surpasses the role of private capital?" or "Do the public good benefits, in combination with output effects, outweigh the effects of public capital?" Research on state-level analysis has investigated whether a linkage exists between transportation infrastructure and productivity. The research has also tried to reconcile the various methodologies, choose an appropriate functional form for analysis, and deal with problems of measurement error.

In the FHWA study "Highways and Macroeconomic Productivity: Phase II," Therese McGuire investigated state-level relationships using pooled cross-section, time series data on gross state product as output and the following inputs: private capital stock for structures and equipment; public capital stock, including educational buildings and highway capital as separate categories; and labor, either nonagricultural or total labor. (7)

Gross State Product = function of (state labor + state private capital + state public capital) + error term

Results indicated a positive and statistically significant role for public capital in determining output.

When McGuire tested highways separately from public capital, highways had the strongest effect of the public capital components included. This means that for a 1-percent increase in highway capital stock--the output measure -- gross state product will increase by 0.121 to 0.127 percent. Overall, McGuire concluded that public capital has a small but positively significant impact on productivity.

Capital stock and investment data

Consistent with the macro dimension of the FHWA research agenda, an NCHRP/AASHTO project "Macroeconomic Analyses of the Linkages Between Transportation Investments and Economic Performance" addressed the problem of limited data. This project generated two industry-specific, state-level private capital stock, time-series data bases using the U.S. Bureau of Economic Analysis (BEA) national figures on employment and output as allocators. Two state-level highway stock data bases were generated using national-level FHWA and U.S. Bureau of the Census (Census) data for highway investment and distributed to the state level using a variety of allocators such as employment. These two highway capital data bases are in addition to the national-level BEA data base used by other researchers. Additional state-level data used Census investment data for six types of infrastructure, including four transportation categories: highways, mass transit, airports, and waterways. These improvements to the data will help answer the question, "Do results of the above studies differ due to differences in the data or insufficient historical data?" (8)

Estimates indicated that for a 1-percent increase in highway capital stock, gross state product -- the output measure -- will increase by 0.121 to 0.127 percent. Overall, McGuire concluded that public capital has a small but significant, positive impact on productivity.

The Industry Studies Approach

In 1990, the FHWA and jointly FHWA and AASHTO through the National Cooperative Highway Research Program (NCHRP) began a set of six industry case studies to identify firm-specific relationships between transportation -- specifically highways -- and productivity. Each effort used a different methodology; however, the main emphasis was on a logistics-style analysis. By collecting new data from firms in particular industries, these studies have tried to identify the mechanisms through which highway investment is converted into increased private sector productivity at the microeconomic or industry level. This dimension of the FHWA research agenda asks questions such as: How does the effect of public capital vary among firms and modes of transportation? What is the overall impact of transportation service on specific industries such as bulk agricultural products, heavy materials, or light-weight and high-value products? How can changing logistical needs and truck size and weight policies effect transportation's economic performance?

Two studies, FHWA's "Industries Studies of the Relationship Between Highway Transportation and Productivity" and the joint FHWA/AASHTO study "Measuring the Relationship Between Freight Transportation Services and Industry Productivity," have tried to measure "direct" costs such as those associated with travel time, vehicle maintenance, fuel use, and safety, as well as "indirect" benefits such as innovation in design and operation of production and distribution systems, which influence productivity. (9,10) For example, transportation improvements allow firms to design more productive processes based on lower-cost, reliable freight transportation. Improvements to the system also may increase the range of territory from which firms can gather inputs or seek markets. As final consumers, the public benefits through a wider array of goods and services which can be offered at lower costs.

By identifying industry benefits and costs which are not usually quantified -- such as transit time reduction, reliability of arrival time, inventory reductions, decreased product damage rates, and decreased freight costs -- we will better understand the linkages between highway infrastructure investment and economic performance.

Efforts to measure the indirect economic benefits of the highway network to industry indicate, as expected, that cost and logistics savings would not be possible without a reliable highway system. Indeed, the FHWA study's preliminary estimates indicate that in the first year, a 6.6 percent rate of return accrues to the manufacturing sector as a whole, due to highway investment. Results for the economy as a whole may be even higher, when other sectors of the economy are incorporated into the analysis. Results for specific sectors of the economy will be additive, therefore the rate of return for the total economy may be considerably larger. These results are preliminary, and additional industrial studies will offer further detail on this issue.

The Highway Service and Performance Assessment Approach

 

Projects using FHWA's highway system assessment approach focus more directly on the efficiency and productivity of the services provided through the use of highway facilities, rather than either their asset value or investment expenditures for highways over time. This approach studies the highway system and its connectivity as seen in its unique network aspects, public utility aspects, pervasiveness, extensiveness, and maturity. This approach takes advantage of FHWA's existing data bases describing highway capital expenditures, highway facilities, and highway system performance. Changing industrial transportation demand impacts and the effects on the highway system are also considered. A national approach captures spillover benefits of the network. This perspective addresses the following questions: How do impacts of transportation spending vary by the type of expenditure? What are the effects of networks and of connectivity? How do congestion, capacity, and highway performance measures affect highway system or economy-wide productivity?

The FHWA study "Performance-Based Measures of the Transportation Productivity Linkage" focuses on the importance of highway density and network connectivity -- both interstate and intrastate -- to economic growth. Using data on highway performance and conditions, the study assessed measures for operating characteristics, physical conditions, and safety in relation to economic vitality. (11)

This study uses correlation analysis to assess whether a coincident relationship exists between random variables -- performance and condition measures -- and economic output. This is opposed to regression analysis, which estimates a causal relationship between an assumed known constant -- highway capital stock -- and a random variable -- output growth. Early results of cross-section time-series correlations for the 48 contiguous states in the United States over two time periods 1985-88 and 1980-88 complement the macroeconometric findings of production and cost functions. They show positive correlations between highway infrastructure measures and the level and growth rate of gross state product (i.e., income) per labor force member.

The study identified several similar positive relationships and similar trends. The study indicated that pavement quality (PSR) has a positive relationship with income growth. The trends for urban interstate congestion and gross domestic product (GDP) are similar; however, causality cannot be implied from simple correlation analysis. We are not yet certain what network analysis measures to pursue or how to incorporate this element into the economic analysis. However, this study indicates that performance measures of the highway system may provide additional insight into the economic linkages between networks and productivity.

 

 

Future Research

FHWA's fiscal year 1994 research agenda "Interrelationships Between Highway Infrastructure and Productivity" plans several approaches.

The FHWA research from the macroeconometric perspective has moved away from production and cost function analyses towards other methodologies at the national level, such as input-output analysis. A new effort explores input-output models to explain the production and consumption impacts of technological improvements of highway transportation. Other projects have begun to extend in-house research and model-building efforts to assess the employment effects of highway spending.

The FHWA microeconomic industry analysis approach continues to examine and develop a variety of methodologies at the industrial level. Continued logistics-style analysis will expand the current efforts by the FHWA and the NCHRP. Other efforts will explore the availability of additional firm-level data.

To better understand the linkage of highways to economic performance, network and service characteristics of the highway system will be explored further. Specifically, spatial economics and location theory will be used to investigate the highway system network attributes. Other efforts will consider how these research products can be incorporated at both program and project levels. Cost-benefit analysis, one example of such a methodology, can contribute to project evaluation and selection, but applications at a national level remain inadequate.

Issues of Structural Change

Economists Bruce McDowell and Bruce Bell identify several structural transformations underway throughout the world: changes in production processes and in the structure of the industrial sector, shifts in the location of various economic activities, the increasing importance of the service sector in the economy. (12) Economist Randall Eberts also identifies increased international competition as a potential cause of restructuring in America. (13) These changes may alter our demands for infrastructure in the future. For example, will the development of advanced telecommunications capabilities serve as a complement or substitute to transportation infrastructure?

The future transportation outlook includes: a shift toward intermodalism, increased responsiveness to customer needs, more attention to performance measures, accountability, efficiency, and cost effectiveness.

Conclusion

This article has presented a discussion of FHWA research from the "Interrelationships Between Transportation Infrastructure Investment and Productivity" priority initiative. The research program pursues three unique dimensions to address several questions about the linkage between highways and the economy. While the projects within these approaches are unique, they offer several levels of analysis and divergent methodologies to achieve a full complement of answers to questions such as: How do highways promote employment, income, output or economic growth? Are public and private capital substitutes or complements? Are industries affected differently by transportation infrastructure? How does the quality or performance of highway infrastructure effect economic vitality? How can we enhance highway system efficiency and productivity?

We will use the information from this research to enhance the productivity and efficiency of the highway system and to respond to structural changes in the economy.

Further study of the relationship between infrastructure investment and productivity growth is needed to develop strategic infrastructure investment policies that will provide policy guidance at the program level and to determine tradeoffs to maximize economic vitality.

In the past, the U.S. transportation objective focused mainly on mobility and connectivity. While early research on the linkage served as a ground breaking introduction, it has taken time to learn the right questions to ask. We are just now beginning to get some results, especially in the industry studies, and we anticipate developments in the network analysis and the macro input-output modeling efforts.

The benefits from our highway system cross all levels of society and are exhibited in several ways. The industry studies indicate that evolving management forms and logistic cost savings would not be possible without our expansive highway system. For example, a lower-cost, efficient, reliable highway network allows transportation consumers to redesign production processes and access more markets, thereby providing a wider array of goods and services at lower costs. Reductions in logistic costs will continue to offer consumers and producers extraordinary benefits. It is our hope that this research agenda will provide an overall estimate of the level and process by which these economic benefits become a reality. By improving our understanding of this linkage process, we can begin to offer more direction on how to support continued economic benefits to the economy as a whole from the transportation network. Without this increase in economic productivity, we cannot expect our standard of living to rise or even be maintained.

References

(1) Assessing the Relationship Between Transportation Infrastructure and Productivity, Searching for Solutions: A Policy Discussion Series, Federal Highway Administration, Publication No. FHWA-PL-92-022 Washington, D.C., August 1992.

(2) David Lewis. "Objectives and Decision Criteria for Infrastructure Investment: A Summary of Research Findings of Project 2-17 of the National Cooperative Highway Research Program," American Association of State Highway and Transportation Officials, Washington, D.C., 1993.

(3) David A. Aschauer. "Is Public Expenditure Productive?" Journal of Monetary Economics, Volume 23, pp. 177-200, 1989.

(4) Alicia Munnell. "Why Has Productivity Growth Declined? Productivity and Public Investment," New England Economic Review, January/February 1990, pp. 3-22.

(5) Alicia Munnell. "How Does Public Infrastructure Affect Regional Economic Performance?" Is There a Shortfall in Public Capital Investment? Federal Reserve Bank, pp. 69-194, 1990.

(6) "Highways and Macroeconomic Productivity: Phase I," Federal Highway Administration, Washington, D.C., 1991. (Unpublished)

(7)"Highways and Macroeconomic Productivity: Phase II" Federal Highway Administration, Washington, D.C., 1992. (Unpublished)

(8) Michael Bell "Macroeconomic Analyses of the Linkages Between Transportation Investments and Economic Performance" National Cooperative Highway Research Program Project 2-17(3), American Association of State Highway Transportation Officials, Washington, D.C., 1994.

(9) "Industries Studies of the Relationship Between Highway Transportation and Productivity," Federal Highway Administration, Washington, D.C., 1993. (Unpublished)

(10) David Lewis and Daniel Brod. "Measuring the Relationship Between Freight Transportation Services and Industry Productivity," National Cooperative Highway Research Program Project 2-17(4) Federal Highway Administration and the American Association of State Highway Transportation Officials, Washington, D.C., 1993.

(11) "Performance-Based Measures of the Transportation-Productivity Linkage: Draft Technical Memorandum on Phase I," Federal Highway Administration, Washington, D.C., 1993. (Unpublished)

(12) Bruce McDowell and Michael Bell. "The Value of Infrastructure to America," U.S. Army Corps of Engineers, Washington, D.C., 1991. (Unpublished)

(13) Randall W. Eberts. "An Assessment of the Linkage Between Public Infrastructure and Economic Development,"Final Report, National Council on Public Works Improvement, Washington, D.C., 1986.

Theresa M. Smith is an economist in the Industry and Economic Analysis Branch of the Transportation Studies Division of the FHWA. She is a member of the American Economic Association and the Western Economic Association. For more than three years, she has worked principally on analyzing the linkage between transportation infrastructure investment and economic productivity. She is currently the contract manager on "Performance-Based Measures of the Transportation Productivity Linkage" and the "Trucking Industry Size Study." She holds a master's degree in economics from the University of North Carolina at Chapel Hill.