This appendix describes the evaluations that are summarized in the section in Chapter 3 entitled Highway System Performance. The descriptions note uncertainties that arise from simplifying assumptions and data limitations in the studies.
Neither the federal government nor the states conduct systematic retrospective evaluations of the costs and benefits of projects. The U.S. Department of Transportation (USDOT) does, however, prepare biennial reports to Congress (the Conditions and Performance studies) that estimate the benefits of alternative future rates of capital spending for highways.
The estimates are derived from a model (the Highway Economic Requirements System) that uses data on traffic, geometry, and state of repair of each of a sample of road segments reported to the Federal Highway Administration by the states. A set of cost factors allows the model to project infrastructure and user costs for each segment for specified assumptions about future road improvements and traffic growth. Given a forecast of traffic and a budget, the model selects the most cost-effective highway improvements.
The USDOT studies present historical trends for physical measures of highway condition (pavement smoothness, bridge structural condition, and numbers of bridges with obsolete designs) and performance (congestion, average speed, and accident risks). The 2002 study concluded that physical conditions of highways were unchanged or slightly improved during the 1990s. For example, the fraction of all vehicle miles on main roads that were on pavements meeting an engineering standard for minimum acceptable ride quality was nearly constant (at 90 percent) over the decade (USDOT n.d., ES-4). Performance was found to have deterio-
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 202
The Fuel Tax and Alternatives for Transportation Funding: Special Report 285 APPENDIX A Highway Benefits Estimates This appendix describes the evaluations that are summarized in the section in Chapter 3 entitled Highway System Performance. The descriptions note uncertainties that arise from simplifying assumptions and data limitations in the studies. U.S. DEPARTMENT OF TRANSPORTATION HIGHWAY CONDITIONSAND PERFORMANCE STUDIES Neither the federal government nor the states conduct systematic retrospective evaluations of the costs and benefits of projects. The U.S. Department of Transportation (USDOT) does, however, prepare biennial reports to Congress (the Conditions and Performance studies) that estimate the benefits of alternative future rates of capital spending for highways. The estimates are derived from a model (the Highway Economic Requirements System) that uses data on traffic, geometry, and state of repair of each of a sample of road segments reported to the Federal Highway Administration by the states. A set of cost factors allows the model to project infrastructure and user costs for each segment for specified assumptions about future road improvements and traffic growth. Given a forecast of traffic and a budget, the model selects the most cost-effective highway improvements. The USDOT studies present historical trends for physical measures of highway condition (pavement smoothness, bridge structural condition, and numbers of bridges with obsolete designs) and performance (congestion, average speed, and accident risks). The 2002 study concluded that physical conditions of highways were unchanged or slightly improved during the 1990s. For example, the fraction of all vehicle miles on main roads that were on pavements meeting an engineering standard for minimum acceptable ride quality was nearly constant (at 90 percent) over the decade (USDOT n.d., ES-4). Performance was found to have deterio-
OCR for page 202
The Fuel Tax and Alternatives for Transportation Funding: Special Report 285 rated: the fraction of all travel on freeways and principal arterial streets that is in congested conditions increased from 34 to 40 percent between 1987 and 2000 in urban areas with populations exceeding 3 million and from 18 to 22 percent in urban areas with populations under 500,000. The 2002 study projected the effects of alternative rates of highway capital expenditure on highway user costs (travel time, vehicle operating costs, and accident costs) over the period 2001–2020. In the projections, the future level of highway travel depends on the cost of travel. The projections indicated that, if all highway capital projects nationwide with a benefit–cost ratio greater than 1 were carried out, annual capital spending would average $107 billion (in 2000 dollars). To maintain overall conditions and performance at 2000 levels, annual capital spending of $76 billion would be required, 17 percent above actual capital spending of $65 billion in 2000. The latter estimate suggests that the present spending level plus normal growth in spending may be nearly sufficient to maintain performance to 2020. The estimated discrepancy between actual and maximum justified spending in the 2002 report was somewhat less than in the previous report (USDOT 2000). The reduction presumably reflects the increased rate of highway spending in the late 1990s. The 2002 report did not present estimates of returns on investment. However, the previous report estimated that if all projects with benefit–cost ratio greater than 1 were carried out over the 20-year period 1998–2017, the average benefit–cost ratio would be 3.7 (USDOT 2000). The projections showed that, at all spending levels analyzed up to the maximum economically justified level, congestion will be little improved. The fraction of urban travel that is under congested conditions increases by 2020, although annual hours lost to congestion per driver fall slightly at the higher spending levels (USDOT n.d., 9-8). The Conditions and Performance studies also compare the mix of kinds of projects that USDOT estimates would be most beneficial with the mix that highway agencies have been carrying out in recent years. The 1999 study concluded that benefits would be increased if agencies shifted spending from capacity expansion to system preservation. The 2002 report, attributing the change in part to large investments in preservation starting in the mid-1990s, concluded the opposite—that the mix should now be shifted to expansion (USDOT n.d., iii). The USDOT model used to produce these projections has been critiqued by a Transportation Research Board committee (TRB 2003, 56–58, 127) and by the General Accounting Office (GAO 2000), which concluded that the studies have value for the purposes intended. The model also was relied upon by the Congressional Budget Office to analyze highway spending effectiveness (CBO 1988, 4–20). CBO prioritized categories of highway investment in terms of national average rates of return (in the 1980s) as follows: projects to maintain current conditions, 30 to 40 percent annual rate of return; new construction in urban areas, 10 to 20 percent; projects to fix roads not meeting minimum engineering
OCR for page 202
The Fuel Tax and Alternatives for Transportation Funding: Special Report 285 design standards, 3 to 7 percent; new construction in rural areas, “low” except for bridge replacements to carry large trucks. The USDOT model has important limitations. It does not support comparisons of highway expansions with demand management alternatives. It does not incorporate a network model; consequently, the estimate of benefits from expansion of a highway link does not change if a decision is made to expand a substitute or complementary link simultaneously. It cannot evaluate trade-offs between capital expenditure and maintenance. Environmental costs are not taken into account. The model takes as given certain design standards and regulatory requirements that strongly affect project costs. Data quality is a concern, and projections are sensitive to values of elasticity parameters in the model, which are not known with a high degree of certainty. Finally, the model’s projections of justified spending levels assume that projects are performed in order of their rates of return, with the highest-payoff projects given highest priority until the budget is exhausted or the minimum acceptable rate of return is reached. In practice, state and local highway agencies do not rank projects exclusively according to economic returns, and there may be regional disparities in rates of return that result from the state allocation formulas in the federal-aid program. However, priorities in state and local capital programs certainly are influenced by factors (including volume of traffic, severity of congestion, degree of deterioration of pavement and structures, and project costs) that are related to rate of return. PRODUCTIVITY BENEFITS OF HIGHWAY INVESTMENTS Several economic studies in recent years have produced estimates of the return on highway infrastructure investment by using statistical methods to examine how infrastructure affects production costs or contributes to output, for the national economy or for industry groups at the national or regional level (reviewed by Shirley and Winston 2004, 399; ICF et al. 2001; Aaron 1990). This section describes the results of the four studies that are highlighted in Chapter 4. Keeler and Ying (1988) This study estimated a cost function for the intercity trucking industry (that is, the relationship of total annual production costs in the trucking industry to industry output, the prices of inputs, and external factors that influence productivity, including highway infrastructure). The study was based on aggregate industry data from the Interstate Commerce Commission for each of nine regions for 1950 to 1973. Highway capital stock in each region was represented by an index similar to that described in Chapter 2 and shown in Figure 2-12. Only state-maintained roads (or federal-aid roads for some calculations) were included in the capital stock
OCR for page 202
The Fuel Tax and Alternatives for Transportation Funding: Special Report 285 measure, as an approximation of the network of main freight-carrying intercity roads. The authors used data only through 1973 because their data showed highway capital stock growing so slowly in the following decade that the later observations added nothing to their ability to estimate the effect of changes in the highway stock on trucking costs. The results showed savings reaching $6 billion to $9 billion annually by 1973 (in 1973 dollars) for the total of U.S. intercity truck traffic that would have occurred without the expansion of the highway system. The savings depends on the value assumed for the elasticity of truck traffic volume with respect to trucking costs. The estimated annualized capital cost of the intercity highway network during the period was $18 billion per year based on a 12 percent interest rate and a 25-year life of new stock and $11 billion per year based on a 6 percent interest rate. Therefore, these truck savings justified one-third to one-half of the total capital cost of the intercity highway system at the higher interest rate and 55 to 80 percent of capital costs at the lower interest rate. Highway maintenance and operation expenditures were $9.6 billion in 1973 (FHWA 1997, Table HF-210), so the estimated trucking cost savings equaled between one-quarter and one-half of total capital costs plus operating expenses. These estimates cover only part of the total benefit of highways. They exclude the benefit derived from the additional truck traffic stimulated by highway system expansion during the study period as well as benefits derived from business traffic other than freight trucks and from personal travel. Freight truck traffic was 5 percent of all vehicle miles of travel in 1973. Year-by-year estimates showed that by the early 1970s, the marginal benefits of increases in the highway capital stock were becoming small. This is not an implausible result since by the early 1970s a basic network of Interstate highways had been completed. The authors noted that this is the same period during which expansion of the system also slowed and speculated that the system may have been responding to considerations of economic efficiency. Shirley and Winston (2004) A model of the inventory holding costs in U.S. industry was developed. Transportation system improvements are expected to reduce inventory holding costs by increasing the speed and reliability with which firms can replenish inventory. Firms are willing to bear the capital and operating costs of holding inventories of production inputs in order to avoid the costs of lost sales and production disruption if they run out. Also, if transportation is expensive, firms will tend to order replacements in large quantities to gain bulk shipment economies. If transportation becomes cheaper, quicker, and more reliable, the cost of running out of stock is reduced because inventory can be replenished quickly, replacement orders will be smaller and more frequent, and inventory size and cost will decline.
OCR for page 202
The Fuel Tax and Alternatives for Transportation Funding: Special Report 285 The study estimated the reduction in inventory costs caused by expansion of the highway system in the period 1973–1996. The authors estimated the parameters of an econometric model relating a plant’s inventory to variables representing the firm’s expected demand for its product, highway infrastructure, and the interest rate and other variables representing inventory holding and stockout costs. Annual establishment-level inventory data were from a Census Bureau survey. Highway infrastructure was represented by capital stock measures for the nation and for the state in which each establishment was located. The results showed that at all times during the study period, additional investment in the highway capital stock caused plants to reduce their inventories. Finally, inventory savings were scaled to an estimate of total economywide logistics cost savings by multiplying by a constant factor. The authors estimated that the annual rate of return on net investment in the highway capital stock from these savings was 18 percent during the 1970s (i.e., an additional $1 of net highway capital stock reduced costs by $0.18), 5 percent during the 1980s, and 1 percent during the 1990s. The authors speculated that the decline in the rate of return over time was the result of finance- and management-related factors, including the growth in project earmarking in the federal-aid program, misdirection of spending for expansions, and lack of congestion pricing. However, they had little evidence that these shortcomings were more severe in recent years than in the 1970s. Both the trucking cost study and the inventory cost study found steep declines in rates of return over the periods studied, but the former found a low rate of return in the 1970s and the latter found a relatively high return during that period. Fernald (1999) This study took a more aggregate approach than those described above. It examined the link between highway capital expansion and industry productivity growth. The study used data on production and inputs for 29 industry groups (covering the entire U.S. private economy except agriculture and mining) for the period 1953–1989. It estimated the relationship of growth in total factor productivity in each sector to growth in the national highway system and to the stock of motor vehicles that each industry owns, as a share of its total capital stock. The premise of this model was that if expansion of the highway system contributes to productivity growth, industries that make more intensive use of motor vehicles will benefit more from highway expansion. An estimate of the rate of return earned by highway investment was derived from the observed “excess” productivity growth in the motor vehicle–intensive industries. The results indicated that for the period as a whole, road expansion contributed strongly to productivity growth and that the return on additional road investment greatly exceeded the normal private-sector rate of return. The estimates
OCR for page 202
The Fuel Tax and Alternatives for Transportation Funding: Special Report 285 implied rates of return exceeding 100 percent, at least in the first part of the study period—that is, the addition of $1 worth of roads reduced business costs by more than $1 per year. Separate estimates for the pre- and post-1973 periods indicated that after 1973 the rate of return was somewhat lower and that the difference between it and the normal rate of return was not statistically significant. The author emphasized a qualitative interpretation of the results: “roads appear strongly productive before 1973. After 1973, the productivity of roads is statistically significantly smaller, and we cannot reject that roads have a normal (or even zero) return” (Fernald 1999, 632). The uncertainty in the estimate of the post-1973 rate of return is very large, because there is little variability in road growth after that date (in fact, the stock of roads nearly stopped growing in the period 1973–1989). These findings—large overall returns but an apparent sharp decline in the early 1970s—closely parallel those of Keeler and Ying. Nadiri and Mamuneas (1998) The final study, sponsored by the Federal Highway Administration, estimated the contribution of highway capital to productivity in 35 industries and in the entire U.S. economy for the period 1950–1991. Demand and cost functions were estimated for each industry. Highway capital stock and the stock of other publicly provided infrastructure were explanatory variables in the cost functions. The marginal benefits of highway capital for each industry were calculated from the parameters of the demand and cost equations. The study found annual rates of return on highway capital of 54 percent for 1960–1969, 27 percent for 1970–1979, and 16 percent for 1980–1991 (Nadiri and Mamuneas 1998, Table 12). That is, in the 1980–1991 period, an additional $1 of highway capital stock produced annual cost savings in private business equal to 16 percent of the total social cost of providing the additional capital. (The social cost of an added $1 in capital spending may be a little more than $1 because of economic distortions in the private sector caused by taxation.) The authors observed that by the end of the period, rates of return on highway and private-sector capital appear to have converged. REFERENCES Abbreviations CBO Congressional Budget Office FHWA Federal Highway Administration GAO General Accounting Office TRB Transportation Research Board USDOT U.S. Department of Transportation
OCR for page 202
The Fuel Tax and Alternatives for Transportation Funding: Special Report 285 Aaron, H.J. 1990. Discussion. In Is There a Shortfall in Public Capital Investment: Proceedings of aConference Held in June 1990(Alicia H. Munnell, ed.), Federal Reserve Bank of Boston, Boston, Mass., pp. 50–63. CBO. 1988. New Directions for the Nation’s Public Works. Sept. Fernald, J.G. 1999. Roads to Prosperity? Assessing the Link Between Public Capital and Productivity. American Economic Review, Vol. 89, No. 3, June, pp. 619–638. FHWA. 1997. Highway Statistics Summary to 1995. GAO. 2000. Highway Infrastructure: FHWA’s Model for Estimating Highway Needs Is GenerallyReasonable, Despite Limitations. June. ICF Consulting, HLB Decision Economics, and Louis Berger Group. 2001. Freight Benefit/Cost Study:Compilation of the Literature (Final Report). Federal Highway Administration, Feb. 9. Keeler, T.E., and J.S. Ying. 1988. Measuring the Benefits of a Large Public Investment: The Case of the U.S. Federal-Aid Highway System. Journal of Public Economics, Vol. 36, pp. 69–85. Nadiri, M.I., and T.P. Mamuneas. 1998. Contribution of Highway Capital to Output and ProductivityGrowth in the U.S. Economy and Industries. www.fhwa.dot.gov/policy/gro98cvr.htm. Federal Highway Administration, Aug. Shirley, C., and C. Winston. 2004. Firm Inventory Behavior and the Returns from Highway Infrastructure Investments. Journal of Urban Economics, Vol. 55, March, pp. 398–415. TRB. 2003. Special Report 271: Freight Capacity for the 21st Century. National Academies, Washington, D.C. USDOT. 2000. 1999 Status of the Nation’s Highways, Bridges, and Transit: Conditions and Performance. USDOT. n.d. 2002 Status of the Nation’s Highways, Bridges, and Transit: Conditions and Performance:Report to Congress.