Cover Image

Not for Sale



View/Hide Left Panel
Click for next page ( 18


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



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 17
17 Kentucky, the primary sources of revenue include fuel taxes, expenditures to build and maintain roads (Selle 2006). registration and license fees, usage taxes, road tools, other Oregon has avoided this issue entirely by virtue of its con- motor carrier taxes, federal taxes, and miscellaneous taxes stitutional requirement that all highway-user revenue be and fees. This table demonstrates how highway-user fees are used for the construction, rehabilitation, and maintenance attributed to each vehicle class. For example, the Kentucky of roads in the state. weight distance tax is allocated to trucks weighing in excess of 59,999 lb based on VMT. The regular Kentucky motor The 1994 Texas HCAS excluded the impacts of state fuels tax is allocated to all highway-user classes based on a highway-user taxes that had been used to fund education revenue analysis that considers VMT, rates of fuel consump- programs in the state and federal funds used on mass transit tion, and tax rates. projects (Euriat 1994). In 2000, however, a study sponsored by the Texas DOT recommended exploring scenarios in the TREATMENT OF DIVERSIONS OF HIGHWAY-USER next Texas HCAS that would allocate all revenues regardless REVENUES TO OTHER USES of their use (Luskin et al. 2001). States have varied in their treatment of diversions of highway- EXPERIENCE IN SELECTED STATES user revenues to other uses. Some HCASs have identified the diversion of highway-user revenues to other uses but have Brief highlights of HCAS work by 11 selected states are eliminated them from consideration, whereas others have at- summarized in this section. These states have been selected tempted to judge each diversion based on its relative merit. For either because (1) they have been leaders in the field, (2) the example, most but not all states consider the highway patrol as amount of work they have done, or (3) they have performed an essential element of the highway program and allocate these work that may be of significant interest to other states that costs accordingly (Idaho Department of Transportation 2002). may want to borrow from what they have learned or use the results of their work directly or indirectly. The 1997 Federal HCAS allocated expenditures of high- way-user revenues on mass transit. The principle underlying this approach was that expenditures on mass transit facilities Arizona represent a form of congestion management and, therefore, should be allocated to highway users. Thus, mass transit Arizona is one of four states that had conducted five or more expenditures were allocated to automobiles, pickups, and vans HCASs and responded to the survey. Each HCAS was com- in proportion to VMT on higher-order urban highways. It is pleted within the last 14 years, with the last being completed worth noting that practitioners interviewed for this study indi- in 2005. It completed a comprehensive study in 1992 to 1993 cated that because heavy-vehicle classes also benefit from that included all of the currently recommended analyses and congestion-relieving transit investment, it would be more ap- methods, with the exception of some refinements based on propriate to allocate these costs according to congested PCE. the 1997 Federal HCAS. It is also worth noting that not all forms of transit investment ease congestion. For example, although the Central Phoenix/ In 1999, Arizona developed and applied a simplified East Valley Light Rail Project is expected to reduce VMT by method based on the 1993 study and then applied it again four-hundredths of one percent in the region, vehicle speeds twice over the next two years. The amount of effort involved are expected to drop by one-tenth of a mile per hour in the re- was reduced by an order-of-magnitude. The output of both gion and one-fifth of a mile per hour in the corridor served by the simplified and traditional Arizona HCAS model is pre- light rail (Final Environmental Impact Statement . . . 2002). sented in Table 6. As shown, the results are consistent This counterintuitive result is expected to occur because the between the two models for the auto and bus classes. How- light rail will be constructed in an existing roadway (thus re- ever, the simplified model generated results that varied sig- ducing roadway capacity), trains will be given signal pre- nificantly from the Arizona HCAS model for pickups and emption rights that will disrupt signal synchronization with SUVs (20.7% difference), single-unit truck class (56.7% dif- vehicular traffic, and train tracks are expected to block direct ference), and combination trucks (14.8% difference). When access to many driveways and side streets along the route. studying broader vehicle classes (e.g., basic vehicles versus heavy trucks), the results are fairly consistent. See the last Other HCASs exclude these diverted revenues. For section of chapter five dealing with simplified models for fur- example, the 1999 Nevada HCAS identifies the total value ther discussion of the implications of these observations and of diverted higher-user funds at the federal, state, and local how improved models of this type might be developed. levels in Nevada but does not allocate them noting: "Since the purpose of highway cost allocation is to identify whether different vehicle classes are contributing in propor- California tion to their cost responsibility, only the taxes that come from highway users and are used for highways are included California has conducted only two HCASs (1984 to 1987 and in our study" (1999 Highway Cost Allocation Study 1999). 1995 to 2000), but both made important contributions--the The 2005 Vermont HCAS also focuses exclusively on state first primarily in terms of principles and methodology and the

OCR for page 17
18 TABLE 6 A credible HCAS cannot be done for any state's local EQUITY RATIOS AND COMPARISON OF SIMPLIFIED governments without use of a specially designed sur- MODEL TO ARIZONA HCAS RESULTS vey of local governments. Such surveys should be Equity Ratios built around existing data recently reported by local Simplified Model Arizona HCAS Model governments. Vehicle Class Autos 1.33 1.30 Other technical contributions of the 1984 to 1987 Califor- nia HCAS included: Pickups/SUVs 1.45 1.74 Complete adaptation of the 1982 Federal HCAS methods Buses 0.93 0.90 for application at the state level, reflecting the differences in available databases and differences in a variety of Single-Unit Trucks 1.41 0.90 technical methods used in pavement and bridge design. Development of new operational procedures for cross- Combination Trucks 0.81 0.93 walks between (1) expenditure databases and new data developed for expenditures for cost allocation categories Total 1.20 1.20 such as new pavements, pavement rehabilitation, pave- ment maintenance, similar breakdowns for bridge work, Source: 1999 Update of the Arizona Highway Cost Allocation Study (1999). grading and drainage, etc. (see Table 11 in the guidelines section of chapter five), and (2) registered weights and second in terms of demonstrating the value of organizing an operating weights. A significant aspect of these efforts in- HCAS so that it can be open to rapidly changing circumstances. volved analyses of detailed data from project files, weigh- in-motion (WIM) data, and data from weigh stations. The 1984 to 1987 study was one of the first to be initiated Development of well-defined criteria for defining vehi- after publication of the 1982 Federal HCAS report and the cle classes. subsequent national dialog about the impact of that bench- Development of procedures for distinguishing full-fee mark study. In performing that study the California DOT paying vehicles and exempt, partially exempt, or spe- made a major commitment to doing the study in a manner cial-fee-paying vehicles for each vehicle class. that was fully responsive to the 1982 federal study in terms Development of procedures for revenue attribution that of both economic concepts and the logic behind each of the provide reliable estimates of total taxes and fees paid technical advancements of the 1982 study. for each vehicle class for both full-fee-paying vehicles and others. Some of the contributions of the 1984 to 1987 study went beyond the federal study in defining economic principles in In developing these procedures, California was able to operational terms and implementing them: avoid having to deal with various complex issues that are not Highway-user payments should be defined to include all important in terms of basic cost allocation principles or types of payments that are unique to the use of highways, equity concerns. These include issues that some states have regardless of where or how those revenues are spent. struggled with, such as the amount of evasion there is for var- Highway expenditures and future costs should be de- ious taxes and fees and how to deal with tax subsidies that all fined to include all expenditures regardless of the legislatures have created for various categories of vehicles. source of funds for the expenditures and regardless of which agency is responsible, and all costs that are pub- The improvements made in the 1984 to 1987 California lic responsibility regardless of whether they have yet HCAS have been used and improved on in subsequent HCASs been formally adopted or budgeted. in California, Vermont, Arizona, Idaho, Minnesota, and Expenditures for other modes of transportation (pri- Oregon, and in the State HCAS Model developed for FHWA. marily transit) should be included if they are publicly recognized as being of substantial benefit to highway The 1995 to 2000 California HCAS built on and refined the users, either in terms of congestion reduction or conser- work of the earlier study, and was noteworthy not for the types vation of highway capacity for future growth. of improvement advances described earlier, but for two unique Each level of government should be analyzed sepa- aspects that differentiated it from all other state HCASs. rately so that results can be presented in the most flexi- ble manner, either separately or in any combination. The first is how the study dealt with a complex set of im- Careful attention should be given to defining what expen- pacts that were central to the purpose of that study. Because ditures are properly considered part of the analyses for of recent changes in federal law, California had to make basic each level of government. For example, state aid to local changes in how weight fees were being collected on both governments is part of state programs and federal aid to power units and trailers or risk a loss of more than $100 mil- state or local governments is part of the federal program. lion per year. The challenge California faced was how to make

OCR for page 17
19 these changes and eliminate most trailer fees for commercial A fairly small central office staff used to working to- vehicles so that (1) fees would bear the closest possible match gether cooperatively without the need to create special with cost responsibility of vehicles owned and operated by the task forces with formally delegated powers; and many different classes of industry in the states, and (2) how to Recent experience in conducting two HCASs. best select from a wide range of options available for achiev- ing this in a way that did not unduly burden any segment of the trucking industry or the general public. Indiana Indiana is one of three states known to have experience with To meet this challenge, a very user-friendly spreadsheet HCASs that did not respond to the survey. Our only source was developed for California (similar in concept to the one of information about this experience is from the website of described in the weight fee subsection of the guidelines in the Joint Transportation Research Program of Purdue chapter five) that permitted the project team, California DOT University and the Indiana DOT and publications by the staff, and any interested participant from industry to easily Director of the Joint Transportation Research Program, develop and evaluate proposed alternatives. This facilitated Professor Kumares Sinha and others. The 1988 HCAS used more than a year's worth of technical work and dialog among the incremental approach, employed extensive data sets, and all interested parties before the HCAS report and recommen- used numerous allocators, including ESALs, to apportion dations were prepared in 1999. costs among attributable and non-attributable classes of costs (Sinha et al. 1989). The other unique aspect of the 1995 to 2000 California HCAS was a completely unanticipated legislative action that fundamentally changed California's highway-user tax struc- Kentucky ture. The legislature decided to phase out and totally elimi- nate the state's $1.5 billion per year vehicle license fees (an Kentucky is the second of the four states responding to the ad valorem annual fee on all vehicles including trailers). survey that had conducted five or more HCASs, and like Arizona had conducted all of them in recent years (every two California decided to have its HCAS consultant work with years from 1992 through 2000). The studies were done as an the department of motor vehicles (DMV) to develop and eval- initiative of the Kentucky Transportation Cabinet (KTC), uate alternatives for increasing other fees on a schedule that and had been effective in calling attention to the inequities of would create revenue neutrality at each step in the phase-out pe- the tax structure. riod for the vehicle license fees. The project team had to adopt the findings, database, and methods developed for that HCAS According to the KTC's survey responses, recent HCASs to develop and evaluate alternative fee schedules for these were criticized for their treatment of a small number of method- phases based on HCAS principles, the revenue neutrality re- ological issues. The principal issue revolved around the low tax quirements, and a requirement that this revenue neutrality be rate for the weight-distance tax relative to heavy-truck cost re- maintained for each agency that received a formula share of all sponsibility. The explanation for the decision by KTC to stop highway-user revenues. The required set of fee increases was conducting HCASs on a regular basis is not that the tax burden developed in consultation with the concerned agencies and was would be too much with an increase in weight-distance tax submitted to the legislature in 2000 (Sydec Inc., et al. 1999). rates, but that "the evasion was too high." Evidence to support this conclusion came from an analysis of evasion that compared Idaho actual tax collections with an estimate of tax liability based on VMT estimates for vehicles weighing in excess of 59,999 lb. Idaho has conducted two relatively recent HCASs--in 1994 Based on the outcome of this analysis, weight-distance tax lia- and 2002--and several years ago made an attempt to apply bility was estimated at $86.6 million as compared with actual FHWA's State HCAS Model but found that it required greater tax receipts of $70.2 million (Osborne et al. 2000). The argu- information technology (IT) expertise than was available at ment behind this decision is that although an increase in the time. Based on that experience, the Idaho Transportation weight-mile tax rates would bring about equitable payments Department recently undertook a third HCAS using FHWA's from heavy-truck classes as a whole, tax payments would ex- State HCAS Model with the assistance of two Washington ceed the cost responsibility calculated for the majority of the State Economists. Although this study was effectively com- motor carriers who comply with current tax systems. Further, pleted in 2007, the results have not been published. the argument was advanced that if the KTC reduced evasion associates with weight-mile taxation, heavy-truck payments Idaho is an appropriate state in which to conduct a second would equal cost responsibility under the current system and low-cost test of FHWA's State HCAS Model, because it has no adjustments to tax rates would be necessary. the following advantages: This decision is one that might have benefited from the An interest in periodic performance of HCASs; findings of the Oregon Weight-Mile Tax Study. The same tax Good databases for most of the inputs required; evasion issues had been raised, and Oregon DOT's response

OCR for page 17
20 was to investigate the reality of the claim. The study found that 2. Analyses of special classes of vehicles that have, or evasion of the weight-mile tax was not excessive in compari- potentially might have, reduced tax rates based on rev- son with evasion rates for other taxes and fees. The study also enue contributions beyond their cost responsibility examined the trade-off between the weight-mile tax evasion (Result of the Minnesota . . . 1990). rate and the level of enforcement and concluded that, although the level of enforcement was somewhat below the optimal level (i.e., the enforcement level of effort that would result in Nevada the lowest sum of enforcement costs and revenue lost as a re- sult of evasion), the level of evasion was within an acceptable Nevada has completed six HCASs and is the third of the four range at 3% to 7% (Cambridge Systematics et al. 1996). The states that completed five or more HCASs and responded to authors of the report recommended a detailed program of the survey. Nevada HCASs were conducted in 1984, 1988, increased enforcement that would both reduce evasion and in- 1990, 1992, 1994, and 1999. crease the cost-effectiveness of the enforcement program. An outside audit was conducted in 1995 in response to questions and comments by stakeholders and the legislature. Maine The audit included a thorough review and assessment of the procedures and analyses used by the Nevada DOT in the first Maine DOT has completed at least four HCASs--in 1956, four Nevada studies, resulting in recommendations for 1961, 1982, and 1989 (Maine Highway Cost Allocation refinements that were incorporated in the 1999 study proce- Study . . . 1989). Maine's work in at least the last of these dures (Sydec 1994). studies is of interest because the authors of this work devel- oped original approaches, while generally following the best thinking regarding basic principles of cost allocation. The Oregon 1989 Maine HCAS was required by legislation designed to improve equity in the state. Oregon is the fourth of the four states that completed five or more HCASs. It conducted the first HCAS (called "cost re- The 1989 study examined a small number of expenditure sponsibility" studies in that state until recently) in 1937 and categories: highway construction, maintenance, bridge con- has conducted studies fairly routinely ever since, having struction, local assistance, and other outlays. The 1989 Maine completed its fifteenth in 2007--more than twice as many as HCAS used standard approaches for overall study design any other state. (e.g., cost-occasioned approach and allocating actual ex- penditures as opposed to costs). It did, however, use some Oregon has been the developer of most of the basic prin- unusual allocators and approaches for attributing costs, in- ciples that have come to be widely accepted in this field. It cluding VMT, ESALs, PCEs, truck miles traveled, standard has been among the first to adopt the results of research per- vehicle equivalent, the Delphi method, overhead, and other formed by the FHWA and others and adopt new national allocators. The Delphi method, as applied in the Maine HCAS methods for use at the state level. HCAS, allocated some expenditures based on the judgment of maintenance experts. The study also used miscellaneous The 1999 Oregon HCAS was the first state study to adopt allocators such as fuel consumption and the number of regis- FHWA's new NAPCOM for allocation of pavement costs trations by vehicle class. and was the first to make use of the results of other research and methodology from the 1997 Federal HCAS. The soft- ware developed for the 1999 Oregon study was the first to use Minnesota Excel's new Visual Basic programming language, and it The Minnesota DOT is believed to have conducted only formed the basis for FHWA's subsequent development of the one HCAS (1989 to 1990) using an improved version of the generalized HCAS model developed for use by other states, interlinked set of spreadsheets developed and applied for as described previously. Vermont. When compared with the earlier Vermont study, the Minnesota study incorporated more detailed analysis of Recent Oregon studies have also included a rather com- revenues, expenditures, pavement designs and types; more prehensive set of issue papers covering most of the common detailed program categories and classes of highways; and choices facing complex state HCASs. The 1999 Oregon more detailed analysis of expenditures by local governments. HCAS issue papers included: The study finished with two sets of analysis requested by Pavement issues legislative staff: Alternative methods for allocating pavement cost responsibility, 1. An evaluation of specific alternative changes in tax Load and non-load-related damage shares for pave- structure. ment costs,

OCR for page 17
21 Allocation of load-related portion of pavement and 3. Bridge costs, and shoulder costs, 4. Common costs. Use of AMT as an allocator for selected costs, Reliability of data supporting pavement damage The 2002 Texas HCAS allocates bridge costs based on the relationships, and modified incremental approach. It allocates common costs Appropriate environmental factors for allocation of proportionally based on VMT. To allocate flexible and rigid pavement costs. pavement construction, rehabilitation, and maintenance Bridge issues costs, the study uses five allocation methods: Definition of increments, Cost allocation methods, 1. Generalized method--Allocates costs based on a New bridges and bridge replacement, hypothetical facility specially designed for groups of Bridge rehabilitation costs, and vehicle classes, with costs for the base facility Bridge maintenance. allocated based on VMT and the load-related costs Width-related cost issues allocated to vehicle classes based on the optimal de- Other attributable cost issues sign for each combination of vehicle classes, highway Allocation of costs of capacity improvement projects, type, and climatic region. Allocation of right-of-way costs, 2. Proportional based on ESALs. Allocation of climbing lane costs, and 3. Modified incremental analysis--Allocates costs incre- Allocation of rest area costs. mentally, with some cost elements allocated to specific Common and residual cost issues vehicle classes and others allocated to multiple vehicle Other cost elements and time frame issues classes. 4. Variable lanes approach that allocates costs based on the Exclude or include congestion and other external costs, lanes required for different classes of vehicles--that is, Use of expenditure versus cost-based approach, automobiles require more lanes than heavy trucks. Use of historical versus forecast data, and 5. FHWA State HCAS Model. Treatment of federal and local revenues and expen- ditures. Each allocation method was used to determine total cost responsibility and assign equity ratios to each vehicle class Since the 1999 Oregon study, there has been a shift in em- examined in the study. phasis in cost allocation procedures from use of engineering and axle weight allocators toward vehicle use allocators. Vermont Achieving an appropriate balance between the engineer's and the economist's perspective has, in effect, become a new Vermont has conducted four HCASs--in 1982, 1990, 1993, HCAS policy issue as a result of Oregon's recent experience. and 2005. The first was reportedly a comprehensive one. Then in 1989 and 1990 Vermont performed an extensive Texas HCAS using an interlinked set of spreadsheets before the de- velopment of Excel's Visual Basic software that was devel- The Texas HCASs have employed innovative techniques to oped and used in later HCASs. The 1990 Vermont study was conduct highway cost allocation. For example, the 2002 unusual in that it was conducted by Vermont's Legislative Texas HCAS examined the climatic factors that affect the Council, with substantial assistance from VTrans. In 1993, durability of highways. The study used numerous climatic an in-house update of the 1990 study was performed. factors to differentiate local climates, and based on a statisti- cal analysis, found the following factors could be used to In 2005, VTrans, using the expertise of an IT professional establish the relationship between climate and pavement de- with substantial private sector experience, was the first known terioration: Thornthwaite Index (index of moisture), average transportation department to complete an in-house HCAS winter temperature, total freeze-thaw cycles in one year, and using FHWA's State HCAS software after it was developed total precipitation or rainfall. This analysis was used to es- for Oregon and later generalized for all states. The 2005 tablish five relatively homogeneous climatic regions within HCAS was requested by the legislature to evaluate proposed the state. Establishing these regions affected how costs asso- changes to DMV fees. VTrans examined actual expenditures ciated with the deterioration of pavement were allocated and revenue for the previous fiscal year. VTrans then ran dif- among highway-user classes. ferent DMV revenue scenarios to calculate new equity ratios. Although the truck/auto equity ratios moved one percentage Four major cost components were considered for alloca- point closer to equity, that was by coincidence. Comparison tion in the 2002 Texas HCAS: with neighboring states was a more important consideration. VTrans' only outside assistance in this effort was a review by 1. Pavement construction costs (including reconstruction), the manager of the consultant team that developed the model 2. Pavement rehabilitation and maintenance costs, for FHWA, for a cost of less than $10,000 (Selle 2006).