Dedicated Revenue Mechanisms for Freight Transportation Investment (2012)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

77 Leading Revenue-Generation Options Development and evaluation of a revenue mechanism to fund freight infrastructure entails a multilevel series of choices and tradeoffs. At the highest level is the basis of fees or taxation itself—the activities, assets, or transactions that will be taxed, tolled, or surcharged. The choices and tradeoffs progress through identification of target vehicles, activities, and payers to details of the revenue collection mechanism and handling of privacy issues. The research team evaluated a wide range of proposed fed- eral freight infrastructure revenue generation mechanisms and concluded that the leading feasible options are a fuel tax sur- charge, distance/vehicle VMT fees, and vehicle registration fees. Other options, such as taxes on freight tonnage, freight value, ton-miles, value added, or transportation cost, were found to be impractical. Some options, such as international trade fees, were found to be ill-suited as national domestic funding mechanisms, although they may be effective when applied to specific port or trade needs. This section discusses the research team’s conclusions on the leading feasible revenue-generating options, a poten- tial complementary role for PPPs and ITCs, the potential for increased effectiveness by using the leading feasible revenue-generating options in combination, and, finally, the relationship of each of the leading feasible revenue- generation options to current or proposed overall funding systems. Fuel Tax Surcharge Increasing the existing federal fuel taxes on diesel fuel and gasoline would initially appear to be the most straight- forward way to increase overall revenue for transportation infrastructure. An overall fuel tax increase is supported by key industry stakeholders and appears to have relatively good public acceptance. However, an overall fuel tax increase would not yield a dedicated revenue stream for freight infra- structure. A separate, dedicated fuel tax surcharge for freight vehi- cles entails multiple complications. A fuel tax surcharge could be targeted to medium- and heavy-duty trucks either by giving tax refunds or credits to other vehicles or by iden- tifying taxable vehicles at the point of fuel purchase. The tax refund/credit option is low-tech and low-cost, but places the administrative burden on those who do not pay the tax. A tax surcharge on diesel fuel for Class 4–8 trucks with freight body types would cover only around 58% of the medium- and heavy-duty truck fleet, leaving approximately 3 million service trucks and gasoline-powered freight trucks out of the new revenue system. Covering all diesel Class 4–8 trucks would still leave out about 3.1% of the fleet, or 1.6 million vehicles. If the dedicated fuel tax surcharge covered both diesel and gasoline, it would involve virtually all 250 million U.S. vehicles. The existing fuel tax system is highly efficient because it does not require vehicle identification, or vehicle-by-vehicle transactions and accounting. Making a fuel tax surcharge vehicle-specific negates much of the fuel tax’s efficiency advantage. Technological means to identify vehicles at the point of fuel purchase would dramatically increase the imple- mentation and collection costs of that option. Inland waterway operators pay a fuel tax of $0.20 per gal- lon, with the revenue dedicated to infrastructure projects via the Inland Waterways Trust Fund. There is an industry- backed proposal to raise this tax by at least $0.06 per gallon. It is technically feasible to levy a fuel tax on railroads. Some states do so now, although there is a legal challenge in progress. It would be difficult, however, to justify charging a mode that provides its own infrastructure a fuel tax to fund infrastructure. The LUST Trust Fund is a mode-neutral fund created in 1986 to fund federal oversight or cleanup of leaking under- ground fuel storage tanks. The LUST fund is capitalized C h a p t e r 7 Findings and Conclusions

78 through a $0.001 per gallon excise tax on gasoline and other motor fuels, which produced $188 million in 2009. Revenue is collected through the Highway Trust Fund (highway and rail), Inland Waterway Trust Fund, and the Airport and Air- way Trust Fund. The LUST excise tax is a concrete example of a practical, dedicated revenue mechanism to support specific transportation infrastructure investment. It provides a prec- edent for operating a mode-neutral freight fund. Distance/Vehicle VMT Fees Distance/vehicle VMT fees would vary by vehicle class, and operators would be charged according to the miles traveled by each vehicle. These VMT fees would have the advantages of varying directly with mileage and fulfilling the policy desire for “user fees” rather than taxes. The cost of implement- ing distance/vehicle VMT fees, however, would be relatively high. VMT fees could be implemented through a costly, high- tech OBU approach or through a low-cost/low-tech mix of OBU, commercial, and self-reporting systems. In either case, there would be high federal implementation, collection, and enforcement costs and a longer implementation period than other options. Part of the attraction of VMT fees to planners and policy makers is the potential for incorporating TDM or congestion-pricing features into the revenue system. These options, however, would not be feasible with distance/vehicle VMT fees. VMT fees have relatively poor support or opposi- tion from the public and the transportation industry. Time/location VMT fees would permit incorporation of TDM, tolling, and congestion pricing. National time/ location VMT fees for freight and service trucks, however, do not appear to be feasible or publicly acceptable for the foreseeable future. The high costs, lack of national benefits, and persistent privacy issues make it difficult to imagine cir- cumstances under which such a system would be palatable to the industry or the public. Moreover, the TDM benefits of time/location VMT fees depend on local implementation of TDM features. VMT fee systems, particularly time/location VMT fees, raise serious public and political issues of privacy. These issues may be able to be resolved legally and technologi- cally, but they will nonetheless remain powerful emotional and political barriers to VMT fee systems. In particular, there appears to be strong opposition to systems that use GPS or equivalent technologies to determine and record detailed vehicle location records. The technological issues can be resolved through the use of so-called “thick” OBU clients that calculate and pay VMT fees rather than trans- mitting location data. Commercial systems that aggregate and pay VMT fees for entire fleets likewise address the privacy issues. Legality is unlikely to be an issue for these approaches, although there are lingering legal concerns over the ongoing collection of detailed location data. Avail- able surveys and other data, however, strongly suggest that adverse public reaction to the perception of privacy invasion would make time/location or GPS-based VMT fees publicly unacceptable. Vehicle Registration Fees The simplest and most cost-effective means of generating revenue for a dedicated freight infrastructure fund would be a federal registration fee on Class 4–8 trucks of all types. A regis- tration fee would not be a direct user fee, but it would be a proxy for potential truck infrastructure impact and requirements. Such registration fees could be implemented by expanding the HVUT system, which presently covers Class 7–8 trucks. Alter- natively and perhaps more efficiently, a federal registration fee could be “piggybacked” on state registration fee collection. The system could incorporate the IRP, which already collects and allocates registration fees across multiple states and Canada. A registration fee would be the quickest and most efficient option to implement because it would require no new technology and could be collected and enforced via minor expansions of exist- ing systems. Public-Private Partnerships and Investment Tax Credits The research team analyzed the different types of PPPs used in transportation infrastructure projects and evaluated the potential opportunities to leverage public funding. The research team determined that the major value of PPPs is not in providing capital that would otherwise be inacces- sible, but in facilitating more rapid capital investment at a comparable or even lower financing cost. The sources of PPP funding can, for the most part, be accessed through revenue bonds or other instruments. The efficiency attributes of private- sector development and operation are, theoretically, accessi- ble through outsourcing and design-build contracts without private financing. PPPs, however, may prove to be a quicker and more flexible means of tapping those funding sources and efficiencies. In that respect, the true function of PPPs may be more institutional than economic. These findings suggest that PPPs are not a substitute for revenue-generation mechanisms, but a complement that can make other mecha- nisms more effective by giving public agencies more choices over the timing of investments. For railroads and potentially for other modes, ITCs for capacity enhancements are widely considered a viable means of inducing additional beneficial private investment. An ITC can be effective in encouraging additional private investment of a particular type or in a general development direction. As such, an ITC can supplement the public sector’s own infrastructure

79 investment efforts. There is very limited evidence, however, on the amount of new private investment that is induced by ITCs. ITCs appear to be a potentially valuable part of a com- prehensive freight transportation financing package, but can do only part of the job by themselves. Their usefulness is par- ticularly limited until decisions are made on the kinds and amount of freight infrastructure to build. Combined Strategy Options The complexity of the freight transportation industry and the infrastructure funding challenge suggests that the candi- date funding mechanisms may be more effective in combi- nation than separately. In particular, registration fees might supplement fuel taxes or VMT fees to do the following: • Cover electric or hybrid vehicles that would not pay fuel taxes commensurate with their infrastructure needs or impacts. • Capture the impact of vehicle size or weight characteristics that are not reflected in a VMT fee system. • Create incentives to turn the vehicle fleet over faster, to use fuel-efficient tires, or to take other actions consistent with national policy goals. A combined federal freight infrastructure revenue strat- egy could also embrace PPPs and ITCs. As the research team found, these options can facilitate access to multiple capital sources and accelerate the project funding and development process. The availability of multiple revenue and funding tools would thus increase the ability of federal planners and decision-makers to create an equitable, effective revenue strategy with appropriate incentives. Relationship to Overall Infrastructure Funding While this study focused on separate revenue mechanisms for freight transportation and infrastructure, the research team noted that the options differ in their relationship to current or proposed overall funding systems. Targeting fuel tax increases on freight movements or medium- and heavy-truck movements introduces compli- cations that the current fuel tax system avoids. The current fuel tax system is efficient in part because there is no need to differentiate between vehicles or vehicle uses. Targeting the tax requires a means of either differentiating vehicles at the point of purchase via some technological means (e.g., a vehi- cle ID system) or adjusting the post-purchase tax burden (via refunds). The increased cost, complexity, and evasion possibilities inherent in targeting the tax negate some of the fuel tax advantages. Those advantages could be preserved by increasing all fuel taxes and allocating a portion of the rev- enue to freight infrastructure based on a formula, guidelines, or legislative outcomes. That approach, however, is essen- tially the current system operating with higher revenues and does not guarantee freight infrastructure funding or link the level of funding to the level of freight activity. The infrastructure required for a VMT fee system would likely exhibit very strong scale economies. Whether the data are communicated via roadside readers, wireless connec- tions at gas stations, Internet connections to OBUs, or cel- lular connections to smartphones, a truck-only or freight- only system would require the same national coverage as an all-vehicle system. Application of VMT fees or higher VMT fees to the trucking sector would appear much more efficient if a passenger-vehicle VMT system were already in place. Revenue This section discusses the research team’s conclusions regarding revenue and costs, tax or fee rates, federal collec- tion and enforcement costs, maximum revenue potential, long-term revenue outlook, and indexing as they relate to the revenue-generating mechanisms studied. Revenue and Costs For revenue-generating mechanisms, costs and cost/benefit comparisons are critical. Existing fuel taxes are an efficient, cost-effective revenue mechanism due to the simplified col- lection mechanism and other features that have become institutionalized over the 70+ years that fuel taxes have been collected. Restricting additional fuel taxes to freight trucks or even to all Class 4–8 trucks, however, significantly compli- cates fuel tax collection and increases its cost. By comparison, VMT fees would be costly to collect and enforce; implement- ing the technology and collecting the fees for 9 million trucks would involve high capital and compliance costs. In contrast to VMT fees, collection mechanisms are already in place for vehicle registration fees. Although registration fees are cur- rently more costly to administer, collect, and enforce than fuel taxes in percentage terms, this is not the case in terms of absolute dollars. Tax or Fee Rates The tax or fee rates associated with each of the three lead- ing revenue-generating options depend on the revenue goal, the bases for taxes or fees, and the annual quantities for those bases. Table 51 shows the relationship for an annual gross revenue target of $5 billion.

80 Federal Collection and Enforcement Costs Estimates of federal collection and enforcement costs are summarized in Table 52. The lowest estimates are for the diesel fuel taxes with non-freight tax refunds, because that option uses the existing fuel tax system. There would be a small additional cost to process refunds or credits for sev- eral million non-freight diesel fuel purchasers. The costs for a diesel/gas option are much higher because refunds or credits would have to be processed for over 240 million vehicles. The freight VMT fee options do not exhibit economies of scale as the number of affected vehicles rises because much of the cost is driven by vehicle-by-vehicle transactions. There would be economies of scale as the amount of the fee rose. The registration fees do have economies of scale in collection because there is an existing system and the marginal cost of adding vehicles to that system would be small. Maximum Revenue Potential The potential maximum, long-term, net revenue from the various options is influenced by annual federal costs of collec- tion and enforcement; diesel tax revenue lost from conversion to gasoline-powered trucks; and fuel, VMT, and registration revenue lost from truck-to-rail modal shifts. Table 51. Tax or fee rates for $5 billion gross revenue. Revenue Option Fuel Tax Surcharge Options 2008 Vehicles 2008 Gallons Purchased Diesel fuel tax with non-freight tax refunds Class 7&8 freight 3,769,296 15,360,180,292 Class 4-8 freight 4,892,531 19,937,451,447 Diesel/gas* tax with non-freight refunds Class 7&8 freight 4,147,443 16,901,160,573 Class 4-8 freight 7,075,838 28,834,600,558 Diesel fuel tax with vehicle ID Class 7&8 freight 3,769,296 15,360,180,292 Class 7&8 all types 3,932,630 16,025,777,776 Class 4-8 freight 4,892,531 19,937,451,447 Class 4-8 all types 6,227,636 25,378,110,126 Diesel/gas* tax with vehicle ID Class 7&8 freight 4,147,443 16,901,160,573 Class 7&8 all types 4,327,163 17,633,532,832 Class 4-8 freight 7,075,838 28,834,600,558 Class 4-8 all types 9,006,738 36,703,169,928 VMT Fee Options 2008 Vehicles 2008 Vehicle MilesTravelled VMT Distance/ Vehicle Fee - OBU/Options Class 7&8 freight 4,147,443 104,740,386,072 Class 7&8 all types 4,327,163 109,279,065,702 Class 4-8 freight 7,075,838 178,694,663,105 Class 4-8 all types 9,006,738 227,458,000,401 VMT Distance/ Vehicle Fee - OBU Only Class 7&8 freight 4,147,443 104,740,386,072 Class 7&8 all types 4,327,163 109,279,065,702 Class 4-8 freight 7,075,838 178,694,663,105 Class 4-8 all types 9,006,738 227,458,000,401 Excise Tax Options 2008 Vehicles 2008 Vehicles Operated Annual Registration Fee Class 7&8 freight 4,147,443 4,147,443 Class 7&8 all types 4,327,163 4,327,163 Class 4-8 freight 7,075,838 7,075,838 Class 4-8 all types 9,006,738 9,006,738 Source: Tioga Group Analysis of 2002 VIUS (U.S. Census Bureau, 2004) and Table VM-1 of Highway Statistics 2009 (FHWA, 2009). Notes: * "Gas" in this case includes gasoline, natural gas, propane, alcohol fuels, and blends. Basis $/Gal Required for $5 Billion Gross Revenue 0.33$ 0.25$ 0.30$ 0.17$ 0.33$ 0.31$ 0.25$ 0.20$ 0.30$ 0.28$ 0.17$ 0.14$ $/VMT Required for $5 Billion Gross Revenue 0.048$ 0.046$ 0.028$ 0.022$ 0.048$ 0.046$ 0.028$ 0.022$ $/Vehicle Required for $5 Billion Gross Revenue 1,206$ 1,155$ 707$ 555$

81 Table 53 provides estimates of these adjustments and net federal revenue for 2012, assuming pass-through to custom- ers is complete by then. The results are shown graphically in Figure 11. The research team calibrated the revenue model at $5 billion annually. Therefore, the net revenue results are most trustworthy at that revenue target. As the revenue target and tax burden escalate from $5 billion to $20 billion annu- ally, the efficiency of most options declines slightly as trans- portation activity shifts from truck to rail intermodal. The exception is the diesel-only fuel tax surcharge, which loses some of its value due to increased conversion of truck fleets to gasoline. The team modeled two behavioral responses—diesel-to- gas conversion and truck-to-rail mode switch. There are Table 52. Estimated annual collection and enforcement costs. Revenue Option Fuel Tax Surcharge Options 2008 Vehicles Annual Federal Collection and Enforcement Cost Diesel fuel tax with non-freight tax refunds Class 7&8 freight 3,769,296 6,230,704$ Class 4-8 freight 4,892,531 5,107,469$ Diesel/gas* tax with non-freight refunds Class 7&8 freight 4,147,443 245,852,557$ Class 4-8 freight 7,075,838 242,924,162$ Diesel fuel tax with vehicle ID Class 7&8 freight 3,769,296 37,692,961$ Class 7&8 all types 3,932,630 39,326,297$ Class 4-8 freight 4,892,531 48,925,309$ Class 4-8 all types 6,227,636 62,276,359$ Diesel/gas* tax with vehicle ID Class 7&8 freight 4,147,443 41,474,434$ Class 7&8 all types 4,327,163 43,271,631$ Class 4-8 freight 7,075,838 70,758,380$ Class 4-8 all types 9,006,738 90,067,377$ VMT Fee Options 2008 Vehicles Annual Federal Collection and Enforcement Cost VMT Distance/ Vehicle Fee - OBU/Options Class 7&8 freight 4,147,443 195,160,519$ Class 7&8 all types 4,327,163 201,450,710$ Class 4-8 freight 7,075,838 297,654,328$ Class 4-8 all types 9,006,738 365,235,818$ VMT Distance/ Vehicle Fee - OBU Only Class 7&8 freight 4,147,443 195,160,519$ Class 7&8 all types 4,327,163 201,450,710$ Class 4-8 freight 7,075,838 297,654,328$ Class 4-8 all types 9,006,738 365,235,818$ Excise Tax Options 2008 Vehicles Annual Federal Collection and Enforcement Cost Annual Registration Fee Class 7&8 freight 4,147,443 46,150,189$ Class 7&8 all types 4,327,163 46,517,004$ Class 4-8 freight 7,075,838 50,000,000$ Class 4-8 all types 9,006,738 50,000,000$ Source: Tioga Group Analysis of 2002 VIUS (U.S. Census Bureau, 2004) and Table VM-1 of Highway Statistics 2009 (FHWA, 2009). Notes: * "Gas" in this case includes gasoline, natural gas, propane, alcohol fuels, and blends. Basis Collection C Average Collection Cost per Vehicle 1.65$ 1.04$ 59.28$ 34.33$ 10.00$ 10.00$ 10.00$ 10.00$ 10.00$ 10.00$ 10.00$ 10.00$ Average Collection Cost per Vehicle 47.06$ 46.55$ 42.07$ 40.55$ 47.06$ 46.55$ 42.07$ 40.55$ Average Collection Cost per Vehicle 11.13$ 10.75$ 7.07$ 5.55$ osts other responses with less drastic impacts on net revenue, such as higher load factors, greater truck utilization, and fewer backhauls. Moreover, as tax rates rise, so will the industry response. Finally, one open question is how enforcement costs are likely to change as taxes increase. Neither of the behavioral responses modeled are directly affected by a registration fee. As mentioned, a registration fee would create an incentive to do the same amount of shipping with fewer trucks (higher load factors, better utilization, and so forth). For the registration fee diversion effect, the research team converted lost truck miles to reduced truck purchases based on the average miles per truck over the entire fleet (roughly 38,000 miles per truck). Lost revenues were calculated by applying the average registration fee per truck to “lost trucks.”

Table 53. Revenue targets and net revenue. Option Revenue Adjustments 5,000,000,000$ Federal Collection and Enforcement 4,752,587$ Diesel to Gas Conversion Loss 355,769,487$ Truck to Rail Mode Switch Loss 112,042,354$ Net Revenue 4,519,895,702$ Federal Collection and Enforcement 70,758,380$ Diesel to Gas Conversion Loss -$ Truck to Rail Mode Switch Loss 86,593,064$ Net Revenue 4,836,250,883$ Federal Collection and Enforcement 297,654,328$ Diesel to Gas Conversion Loss -$ Truck to Rail Mode Switch Loss 95,525,793$ Net Revenue 4,599,554,493$ Federal Collection and Enforcement 50,000,000$ Diesel to Gas Conversion Loss -$ Truck to Rail Mode Switch Loss 43,054,872$ Net Revenue 4,901,076,525$ Diesel Tax Diesel & Gas Tax VMT Fee Registration Fee 10,000,000,000$ 15,000,000,000$ 20,000,000,000$ 4,752,587$ 4,752,587$ 4,752,587$ 1,615,614,452$ 3,973,834,308$ 7,244,679,437$ 443,418,999$ 994,129,934$ 1,764,175,159$ 7,906,374,161$ 9,960,383,379$ 10,867,672,971$ 70,758,380$ 70,758,380$ 70,758,380$ -$ -$ -$ 331,921,358$ 735,984,882$ 1,298,783,636$ 9,572,797,229$ 14,138,880,660$ 18,534,501,175$ 347,654,328$ 397,654,328$ 447,654,328$ -$ -$ -$ 349,536,807$ 762,033,041$ 1,333,014,495$ 9,276,224,217$ 13,782,354,844$ 18,117,946,373$ 50,000,000$ 50,000,000$ 50,000,000$ -$ -$ -$ 169,277,094$ 378,666,668$ 671,223,591$ 9,757,649,555$ 14,519,719,092$ 19,187,285,135$ 2012 Revenue Target

83 The loss of revenue is limited by the underlying inelastic- ity of demand for freight transportation. As the economic impacts and analysis suggest, the adverse impact on com- modity production and consumption would be relatively small. This observation is confirmed by the long-term resil- ience of freight demand in the face of steady increases in underlying fuel prices that exceed the tax or fee levels envi- sioned in this analysis. Long-Term Revenue Outlook The long-term revenue available from each option depends on the growth expected in the tax or fee base (fuel use, VMT, or truck registrations) and the offsetting costs and diversions discussed above. Table 54 summarizes available forecasts and extrapolations of forecasts for transportation activity in VMT, fuel use, and ton-miles. The growth rates vary by transportation segment, forecast source, and measurement basis. These factors were used to determine the long-term net revenue potential for each of the modeled options. Table 55 shows an increasing tax burden for the private sector for 2012 to 2035 and the net revenue available under each option. The patterns and relationships are shown graph- ically in Figure 12. For the diesel-to-gas conversion, the team used a 10-year linear phase-in. This basically assumes that diesel trucks operate about 10 years, that the average age of the current fleet is random (i.e., that there are as many 7-year-old trucks in operation as 2-year-old trucks), and that the diesel-to-gas conversion is made at the end of a truck’s useful life. Truck-to-rail mode switch is assumed to phase-in based on contract expirations (the only thing preventing imme- diate diversion where the economics work). The phase-in value for tax-burden pass-through to customers was used as a proxy for contract expirations. For the two fuel tax scenarios, the first, second, and third year values are 67%, 100%, and 100% respectively. For the VMT and excise tax scenarios, the equivalent values are 33%, 67%, and 100%. As the graph in Figure 12 shows, there would be a multi- year implementation lag for revenue from the two technology- dependent sources, the diesel and gas tax with vehicle ID and the VMT fee. Net revenue from the vehicle registration fee and the diesel and gas tax surcharge tracks the gross tax bur- den closely. Net revenue from the VMT fee and the diesel- only tax surcharge grow at similar rates but at a lower level due to the high collection cost of VMT fees and the conver- sion to gasoline trucks for a diesel tax surcharge. These growth patterns differ from those predicted for pas- senger vehicles or fuel use and VMT as a whole. Table 56 displays the most recent U.S. Department of Energy forecasts of long-term energy use by mode. As is apparent, the growth rate for heavy-duty vehicles (and rail) is much higher than for light-duty vehicles (cars and light trucks). Although the fuel economy of trucks has improved and will continue to improve, the improvements for heavy-duty trucks are much smaller than the improvements for passenger cars, and increasing truck sizes and loads tend to offset technical fuel economy gains. Truck fuel use, VMT, and registrations therefore tend to grow at similar rates, and the long-term rev- enue growth outlook for the various options is likewise similar. Figure 11. Revenue targets and net revenue. $5,000,000,000 $7,000,000,000 $9,000,000,000 $11,000,000,000 $13,000,000,000 $15,000,000,000 $17,000,000,000 $19,000,000,000 $5,000,000,000 $7,000,000,000 $9,000,000,000 $11,000,000,000 $13,000,000,000 $15,000,000,000 $17,000,000,000 $19,000,000,000 2012 Net Federal Revenue 20 12 R ev en ue T ar ge t Diesel Tax Diesel & Gas Tax VMT Fee Registration Fee

84 Table 54. Forecasts of U.S. transportation activity. Factor Information Source or Methodology 2010 Estimate Passenger VMT 3,313,673 Truck VMT* 249,710 Freight Truck VMT TL VMT 106,912 LTL VMT 3,099 Private Fleet VMT 102,243 Non-Freight Truck VMT Private Service Truck VMT 12,486 Government Truck VMT 19,977 Military Truck VMT 4,994 Truck VMT by Class Class 1 3,063,963 Class 2 25,697 Class 3 23,361 Class 4 9,089 Class 5 7,939 Class 6 18,099 Class 7 8,992 Class 8 156,533 Truck VMT by Road Type Not on NHS 61,067 Interstate 124,557 Non-Interstate STRAHNET 13,709 STRAHNET connector 1,181 Other NHS 48,159 Approved Inter-Modal Connector 1,038 Highway Gasoline, diesel and other fuels (million gallons) 185,867 Truck, total 37,561 Single-unit 2-axle 6-tire or more truck 9,320 Combination truck 28,241 Truck (percent of total) 20 Rail, Class I (in freight service) Distillate / diesel fuel (million gallons) 3,916 Water Residual fuel oil (million gallons) 5,248 Distillate / diesel fuel oil (million gallons) 2,078 Gasoline (million gallons) 1,310 Pipeline Total VMT: millions of miles Extrapolated from FHWA Reported Statistics 1980-2000 Assumes 85% (non-government, non- private fleet share) -- Vehicle type shares assumed to grow from 2008 distribution at rate consistent with ATA forecast (but maintaining linear control totals) Federal Highway Statistics 2008 (10% of heavy trucks government owned, assumes comparable share of VMT) Actual VMT from Federal Highway Statistics (extrapolated as above); classification shares derived from VIUS: http://www.census.gov/svsd/www/97vehinv. html as of July 1, 2005 Extrapolated truck VMT from above allocated to highway systems based on FAF/NHPN from FHWA website. Fuel Use: millions of gallons Estimates extrapolated from federal highway statistics -- if a trend appeared non- linear, an average is noted. On-highway vs. off-highway diesel is not readily available 265,571 -3.30% 5,287,600 0.95% 23,921 1.73% 2,229,420 1.86% 2,786,096 1.92% 97,723 -7.64% Declining NA 40,963 -1.04% 246,269 -0.53% 2,242 1.53% 576,097 -0.10% Natural gas (million cubic feet) 519,677 TOTAL U.S. ton-miles of freight (millions) 4,380,007 Air carrier, domestic, all services 16,989 Intercity truck 1,541,024 Class I rail 1,904,373 Domestic water transportation 478,993 Coastwise 152,771 Lakewise 50,478 Internal 274,089 Intraport 1,656 Oil pipeline 587,249 Freight Transportation: millions of ton-miles These estimates are extrapolated from available federal transportation statistics on U.S. ton-miles 2030 Estimate 2010-2030CAGR 4,516,142 1.56% 348,836 1.69% 152,762 1.80% 4,922 2.34% 138,826 1.54% 17,442 1.69% 27,907 1.69% 6,977 1.69% 4,167,306 1.55% 35,898 1.69% 32,634 1.69% 12,697 1.69% 11,091 1.69% 25,284 1.69% 12,561 1.69% 218,671 1.69% 82,053 1.49% 178,333 1.81% 19,920 1.89% 1,543 1.35% 65,596 1.56% 1,390 1.47% 232,003 1.11% 43,588 0.75% 8,709 -0.34% 34,879 1.06% 18 -0.59% 3,916 0.00% 5,248 0.00% 2,078 0.00% 1,865 1.78%

85 Table 55. Long-term revenue estimates. Diesel Tax Diesel & Gas Tax VMT Fee Registration Fee 2012 5,000$ 4,885$ -$ -$ 4,921$ 2013 5,192$ 4,997$ -$ -$ 5,095$ 2014 5,384$ 5,143$ -$ -$ 5,284$ 2015 5,576$ 5,287$ -$ -$ 5,472$ 2016 5,662$ 5,328$ 5,516$ 5,289$ 5,556$ 2017 5,747$ 5,368$ 5,566$ 5,332$ 5,640$ 2018 5,833$ 5,406$ 5,649$ 5,374$ 5,724$ 2019 5,918$ 5,443$ 5,732$ 5,453$ 5,808$ 2020 6,004$ 5,479$ 5,815$ 5,532$ 5,892$ 2021 6,083$ 5,508$ 5,892$ 5,605$ 5,970$ 2022 6,163$ 5,580$ 5,969$ 5,678$ 6,048$ 2023 6,242$ 5,652$ 6,045$ 5,751$ 6,126$ 2024 6,321$ 5,724$ 6,122$ 5,824$ 6,204$ 2025 6,400$ 5,796$ 6,199$ 5,897$ 6,281$ 2026 6,494$ 5,880$ 6,290$ 5,983$ 6,373$ 2027 6,588$ 5,965$ 6,380$ 6,070$ 6,465$ 2028 6,681$ 6,050$ 6,471$ 6,156$ 6,557$ 2029 6,775$ 6,135$ 6,562$ 6,242$ 6,649$ 2030 6,869$ 6,219$ 6,652$ 6,328$ 6,741$ 2031 6,976$ 6,317$ 6,757$ 6,428$ 6,847$ 2032 7,084$ 6,415$ 6,861$ 6,527$ 6,952$ 2033 7,192$ 6,512$ 6,965$ 6,626$ 7,058$ 2034 7,300$ 6,610$ 7,070$ 6,726$ 7,164$ 2035 7,407$ 6,707$ 7,174$ 6,825$ 7,269$ Year Tax Burden Net Revenue (millions) Figure 12. Long-term revenue estimates. $4,000 $4,500 $5,000 $5,500 $6,000 $6,500 $7,000 $7,500 $8,000 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 $ M illi on s Tax Burden Net Diesel Tax Revenue Net Diesel & Gas Tax Revenue Net VMT Fee Revenue Net Registration Fee Revenue

86 Figure 13. The impacts of inflation and fuel economy on federal motor fuel taxes. Indexing Any proposed revenue mechanism would need to incor- porate indexing or some other means of adjustment to fulfill long-term freight infrastructure funding needs. There are at least two purposes to indexing taxes or fees: • To keep the taxes or fees in proportion with broader cost and price trends so that fuels, VMT fees, or registration fees do not become markedly more or less expensive than other goods and services and introduce market distortions. • To preserve (or improve) revenue buying power for build- ing and maintaining transportation infrastructure. Federal fuel taxes have not been increased since 1993, and there is no mechanism for adjusting them to reflect inflation. Revenues have not risen with inflation as consumer goods prices typically do, and the buying power of fuel tax revenue has eroded as the cost of infrastructure and maintenance has risen. Figure 13 illustrates the erosion of fuel tax buying power by comparing existing Highway Trust Fund Highway Account fuel tax rates with indexed rates based on the Consumer Price Index (CPI, with 1956=100) and on fuel economy (with 1970= 100). The graph therefore shows what the 1956 fuel tax rates would have become if they had increased at the same rate as the CPI or fuel economy. These trend lines understate the magnitude of the full problem as they do not account for the impact of increased truck weights. The following conclusions are evident: • Since 1956, the tax on diesel fuel has largely kept pace with inflation while the much more financially important tax on gasoline (approximately 70% of fuel tax revenue) has lagged inflation since the early 1970s. • Neither tax has kept pace with the combined impact on buy- ing power from both inflation and increased fuel economy. These observations make a clear case for indexing future taxes or fees used to support infrastructure. The pattern shown in Figure 13 is not intrinsic to the fuel tax. Any federal or state tax or fee that is not appropriately indexed will face the same issues. VMT fees offer the advantage of increasing with vehicle travel and road use, regardless of vehicle technol- ogy or power source. This feature will not, however, obviate the need to either index VMT fees or adjust them periodi- cally to account for inflation. Registration fee revenues would grow as the subject vehicle fleet grows over time. While there is thus some parallel between registration fee revenue and infrastructure needs, the increasing size and capacity of vehi- cles, and the rising unit costs of infrastructure would require a registration fee to be indexed as well. Table 56. Long-term energy consumption by mode (quadrillion Btu). Segment 2009 2035 CAGR Light-duty vehicles 16.71 18.39 0.4% Heavy-duty vehicles 4.53 6.68 1.5% Air 2.66 3.07 0.6% Marine 1.24 1.35 0.3% Rail 0.56 0.76 1.2% Pipeline 0.65 0.67 0.1% Total 26.35 30.92 0.6% Source: U.S. Energy Information Administration, April 2011.

87 Figure 14. Comparison of highway construction cost indexes. There are two cost indexes specifically designed to reflect highway construction costs—FHWA’s NHCCI and an over- all PPI for the Highway and Street Construction Industry compiled by the Bureau of Labor Statistics (BLS) (“BHWY”). NHCCI can be used both to track price changes associated with highway construction costs and to convert current dol- lar expenditures on highway construction to real or constant dollar expenditures. NHCCI tracks price changes through a database of bids received on state transportation projects in 45 states. NHCCI is intended to replace FHWA’s dated Bid-Price Index (BPI). “BHWY” covers inputs, principally materials and services, but does not include labor. Figure 14 compares these two indexes with the BLS CPI on a common basis of 2003=100. There is a marked diver- gence between the BLS PPI/BHWAY and the FHWA NHCCI starting in 2007. This may be because the prices of material inputs, especially those derived from petroleum, continued to rise during the recession (as reflected in the BLS PPI/ BHWAY). Bids may have declined due to scarcity of con- tracts and intense competition (FHWA NHCCI). There are also indexes published by trade associations and industry journals. The American Road and Transportation Builders Association in the monthly Construction Material Cost Report tracks changes in material prices. The Engineer- ing News-Record, a leading engineering industry journal, publishes a Construction Cost Index (CCI), a Building Cost Index (BCI), and a Materials Cost Index (MCI) that are widely used in the construction industry. All three indexes have a materials and labor component. The CCI can be used where labor costs are a high proportion of total costs. The BCI is more applicable for structures. The MCI covers concrete, steel, and lumber cost changes. The final choice of a basis for indexing freight taxes or fees will be affected by analytical concerns and policy goals. Nonetheless, there is a need to find or develop a workable cost index methodology. Implementation and Costs This section discusses the research team’s conclusions regard- ing implementation and compliance costs and implementation and phase-in of the leading revenue-generating options. Implementation and Compliance Costs Table 57 provides estimates of federal government and freight industry implementation costs. The fuel tax is well established, and its operation is embedded in industry prac- tice. Any new option will inevitably seem disruptive and com- plex by comparison, but, with sufficient industry knowledge, the disruption can be assessed and the risks reduced. Revenue mechanisms that build on existing tax, fee, or regulatory sys- tems without introducing new technology or infrastructure have the lowest implementation, collection, and compliance costs, and the closest implementation horizon. For the fuel tax surcharges with tax system refunds or credits, the compliance costs would be born primarily by non-freight purchasers who would need to file for refunds or credits. The estimates for industry compliance costs shown in Table 58 are lower for low-tech options that do not require technology and higher for high-tech options that do. Implementation and Phase-In The varying implementation costs, implementation time- lines, and annual net federal revenues of the leading options would yield significantly different levels of funding during the phase-in period. Table 59 displays the estimated net fed- eral revenue for the first 10 years after approval and the

Table 57. Implementation cost estimates. Fuel Tax Surcharge Options Federal Implementation Cost Avg. Federal Implementation Cost per Vehicle Diesel fuel tax with non-freight tax refunds Class 7&8 freight -2ndPgs -$ -$ Class 4-8 freight -$ -$ Diesel/gas* tax with non-freight refunds Class 7&8 freight -$ -$ Class 4-8 freight -$ -$ Diesel fuel tax with vehicle ID Class 7&8 freight 1,500,000,000$ 398$ Class 7&8 all types 1,500,000,000$ 381$ Class 4-8 freight 1,500,000,000$ 307$ Class 4-8 all types 1,500,000,000$ 241$ Diesel/gas* tax with vehicle ID Class 7&8 freight 3,000,000,000$ 723$ Class 7&8 all types 3,000,000,000$ 693$ Class 4-8 freight 3,000,000,000$ 424$ Class 4-8 all types 3,000,000,000$ 333$ VMT Fee Options Federal Implementation Cost Avg. Federal Implementation Cost per Vehicle VMT Distance/ Vehicle Fee - OBU/Options Class 7&8 freight 3,000,000,000$ 723$ Class 7&8 all types 3,000,000,000$ 693$ Class 4-8 freight 3,000,000,000$ 424$ Class 4-8 all types 3,000,000,000$ 333$ VMT Distance/ Vehicle Fee - OBU Only Class 7&8 freight 3,000,000,000$ 723$ Class 7&8 all types 3,000,000,000$ 693$ Class 4-8 freight 3,000,000,000$ 424$ Class 4-8 all types 3,000,000,000$ 333$ Excise Tax Options Federal Implementation Cost Avg. Federal Implementation Cost per Vehicle Annual Registration Fee Class 7&8 freight -$ -$ Class 7&8 all types -$ -$ Class 4-8 freight -$ -$ Class 4-8 all types -$ -$ Source: Tioga Group Analysis of 2002 VIUS (U.S. Census Bureau, 2004) and Table VM-1 of Highway Statistics 2009 (FHWA, 2009). Notes: * "Gas" in this case includes gasoline, natural gas, propane, alcohol fuels, and blends. Annual Equivalent Assuming 10-Year Life Freight Industry Implementation Cost Avg. Industry Implementation Cost per Vehicle Annual Equivalent Assuming 5-Year Life -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ 150,000,000$ 376,929,607$ 100.00$ 75,385,921$ 150,000,000$ 393,262,970$ 100.00$ 78,652,594$ 150,000,000$ 489,253,095$ 100.00$ 97,850,619$ 150,000,000$ 622,763,594$ 100.00$ 124,552,719$ 300,000,000$ 414,744,339$ 100.00$ 82,948,868$ 300,000,000$ 432,716,314$ 100.00$ 86,543,263$ 300,000,000$ 707,583,795$ 100.00$ 141,516,759$ 300,000,000$ 900,673,766$ 100.00$ 180,134,753$ Annual Equivalent Assuming 10-Year Life Freight Industry Implementation Cost Avg. Industry Implementation Cost per Vehicle Annual Equivalent Assuming 5-Year Life 300,000,000$ 1,036,860,847$ 250.00$ 207,372,169$ 300,000,000$ 1,081,790,786$ 250.00$ 216,358,157$ 300,000,000$ 1,768,959,488$ 250.00$ 353,791,898$ 300,000,000$ 2,251,684,414$ 250.00$ 450,336,883$ 300,000,000$ 2,073,721,693$ 500.00$ 414,744,339$ 300,000,000$ 2,163,581,572$ 500.00$ 432,716,314$ 300,000,000$ 3,537,918,975$ 500.00$ 707,583,795$ 300,000,000$ 4,503,368,829$ 500.00$ 900,673,766$ Annual Equivalent Assuming 10-Year Life Freight Industry Implementation Cost Avg. Industry Implementation Cost per Vehicle Annual Equivalent Assuming 5-Year Life -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ -$ -$

89 net present value (NPV). In both cases, the estimates are for application to Class 4–8 freight trucks; wider application would yield higher total numbers but also greater disparities. The large differences in NPV are traceable to the higher implementation costs and longer lead times before the first revenue from technology-dependent options. Revenue Efficiency This section discusses the research team’s conclusions regarding net revenue and net cost and scale economies as these relate to the leading revenue-generating mechanisms. Net Revenue and Net Cost Table 60 provides estimates of net annual federal revenue ($5 billion less collection and annualized implementation costs) and annual industry cost ($5 billion plus annualized implementation and compliance costs), and these costs are broken down into net annual revenue per vehicle and annual cost per vehicle. Table 60 also provides the ratio of net federal revenue to annual industry cost. This ratio can be used to compare the overall revenue efficiency of the various options. The numerical values shown in Table 60 would be regarded as rough approximations suitable for relative comparisons. Table 60 shows the following: • The highest efficiency ratios are for the diesel fuel tax with tax refunds and the vehicle registration fees, due to their absence of significant implementation costs. Both of these options build on existing collection systems. • Options that rely heavily on technology, namely, the fuel taxes with vehicle ID and the VMT fees, have high Fuel Tax Surcharge Options Annual Private-Sector Compliance Cost Diesel fuel tax with non-freight tax refunds Class 7&8 freight 155,767,598$ Class 4-8 freight 127,686,726$ Diesel/gas* tax with non-freight refunds Class 7&8 freight 6,146,313,915$ Class 4-8 freight 6,073,104,051$ Diesel fuel tax with vehicle ID Class 7&8 freight 188,464,804$ Class 7&8 all types 196,631,485$ Class 4-8 freight 244,626,547$ Class 4-8 all types 311,381,797$ Diesel/gas* tax with vehicle ID Class 7&8 freight 207,372,169$ Class 7&8 all types 216,358,157$ Class 4-8 freight 353,791,898$ Class 4-8 all types 450,336,883$ VMT Fee Options Annual Private-Sector Compliance Cost VMT Distance/ Vehicle Fee - OBU/Options Class 7&8 freight 447,923,886$ Class 7&8 all types 467,333,620$ Class 4-8 freight 764,190,499$ Class 4-8 all types 972,727,667$ VMT Distance/ Vehicle Fee - OBU Only Class 7&8 freight 447,923,886$ Class 7&8 all types 467,333,620$ Class 4-8 freight 764,190,499$ Class 4-8 all types 972,727,667$ Excise Tax Options Annual Private-Sector Compliance Cost Annual Registration Fee Class 7&8 freight 103,686,085$ Class 7&8 all types 108,179,079$ Class 4-8 freight 176,895,949$ Class 4-8 all types 225,168,441$ Source: Tioga Group Analysis of 2002 VIUS (U.S. Census Bureau, 2004) and Table VM-1 of Highway Statistics 2009 (FHWA, 2009). Notes: * "Gas" in this case includes gasoline, natural gas, propane, alcohol fuels, and blends. Table 58. Compliance cost estimates.

1 2 3 4 5 6 7 8 9 10 NPV Diesel fuel tax with non-freight refunds -$ 4,995$ 4,995$ 4,995$ 4,995$ 4,995$ 4,995$ 4,995$ 4,995$ 4,995$ 33,812$ Diesel/gas tax with non-freight refunds -$ 4,757$ 4,757$ 4,757$ 4,757$ 4,757$ 4,757$ 4,757$ 4,757$ 4,757$ 32,202$ Diesel fuel tax with vehicle ID (300)$ (300)a$ (300)$ (300)$ 4,801$ 4,801$ 4,801$ 4,801$ 4,801$ 4,801$ 18,984$ Diesel/gas tax with vehicle ID (600)$ (600)$ (600)$ (600)$ 4,629$ 4,629$ 4,629$ 4,629$ 4,629$ 4,629$ 17,203$ VMT Distance/ Vehicle Tax - OBU/Options (600)$ (600)$ (600)$ (600)$ 4,402$ 4,402$ 4,402$ 4,402$ 4,402$ 4,402$ 16,256$ VMT Distance/ Vehicle Tax - OBU Only (600)$ (600)$ (600)$ (600)$ 4,402$ 4,402$ 4,402$ 4,402$ 4,402$ 4,402$ 16,256$ Annual Registration Fee -$ 4,950$ 4,950$ 4,950$ 4,950$ 4,950$ 4,950$ 4,950$ 4,950$ 4,950$ 33,508$ Revenue Option Federal Net Revenue by Year for Class 4-8 Freight Trucks (millions) a Parentheses mean cost rather than revenue. Table 59. Net federal revenue during phase-in for Class 4–8 freight trucks.

91 implementation and collection costs that dilute their revenue-generating effectiveness. • The lowest ratio is for the diesel/gas fuel tax with tax refunds since it would require the great majority of vehicle owners—about 240 million—to file for refunds. The other key statistic shown is the annual average cost per vehicle. The diesel fuel tax with non-freight refunds achieves part of its apparently high efficiency by placing a relatively heavy burden on relatively few trucks. The lowest average cost per truck is for the widest possible application of regis- tration fees, that is, to all Class 4–8 trucks. Scale Economies Technology-intensive revenue-generation options dem- onstrate significant economies of scale because their imple- mentation, collection, and compliance costs are essentially independent of the tax or fee rates and the resulting gross revenue. Table 61 displays tax or fee rates, net federal rev- enue, industry cost, and the ratio of revenue to cost for gross revenue targets of $5 billion, $10 billion, and $20 billion. The ratio of revenue to cost rises for all options as the revenue goal increases, but the improvements are greatest for the high-tech options—fuel taxes with vehicle ID and the VMT fees. Even at the $20 billion scale, however, the ratios for the OBU-only VMT options remain noticeably lower. Technical Feasibility Technical feasibility within the freight sector is not an issue for any of the options, although time/location VMT fees do present some technical challenges. The technical differences show up in implementation time, cost, and evasion potential. Fuel Tax Surcharge Options Net Annual Federal Revenue Net Annual Avg. Rev. per Vehicle Annual Industry Cost Annual Avg. Cost per Vehicle Ratio Diesel fuel tax with non-freight tax refunds Class 7&8 freight 4,993,769,296$ 1,325$ 5,155,767,598$ 1,368$ 0.97 Class 4-8 freight 4,994,892,531$ 1,021$ 5,127,686,726$ 1,048$ 0.97 Diesel/gas* tax with non-freight refunds Class 7&8 freight 4,754,147,443$ 1,146$ 11,146,313,915$ 2,688$ 0.43 Class 4-8 freight 4,757,075,838$ 672$ 11,073,104,051$ 1,565$ 0.43 Diesel fuel tax with vehicle ID Class 7&8 freight 4,812,307,039$ 1,277$ 5,263,850,725$ 1,397$ 0.91 Class 7&8 all types 4,810,673,703$ 1,223$ 5,275,284,079$ 1,341$ 0.91 Class 4-8 freight 4,801,074,691$ 981$ 5,342,477,166$ 1,092$ 0.90 Class 4-8 all types 4,787,723,641$ 769$ 5,435,934,516$ 873$ 0.88 Diesel/gas* tax with vehicle ID Class 7&8 freight 4,658,525,566$ 1,123$ 5,290,321,037$ 1,276$ 0.88 Class 7&8 all types 4,656,728,369$ 1,076$ 5,302,901,420$ 1,225$ 0.88 Class 4-8 freight 4,629,241,620$ 654$ 5,495,308,657$ 777$ 0.84 Class 4-8 all types 4,609,932,623$ 512$ 5,630,471,636$ 625$ 0.82 VMT Fee Options Net Annual Federal Revenue Net Annual Avg. Rev. per Vehicle Annual Industry Cost Annual Avg. Cost per Vehicle Ratio VMT Distance/ Vehicle Fee - OBU/Options Class 7&8 freight 4,504,839,481$ 1,086$ 5,655,296,055$ 1,364$ 0.80 Class 7&8 all types 4,498,549,290$ 1,040$ 5,683,691,777$ 1,313$ 0.79 Class 4-8 freight 4,402,345,672$ 622$ 6,117,982,396$ 865$ 0.72 Class 4-8 all types 4,334,764,182$ 481$ 6,423,064,550$ 713$ 0.67 VMT Distance/ Vehicle Fee - OBU Only Class 7&8 freight 4,504,839,481$ 1,086$ 5,862,668,224$ 1,414$ 0.77 Class 7&8 all types 4,498,549,290$ 1,040$ 5,900,049,934$ 1,363$ 0.76 Class 4-8 freight 4,402,345,672$ 622$ 6,471,774,294$ 915$ 0.68 Class 4-8 all types 4,334,764,182$ 481$ 6,873,401,433$ 763$ 0.63 Excise Tax Options Net Annual Federal Revenue Net Annual Avg. Rev. per Vehicle Annual Industry Cost Annual Avg. Cost per Vehicle Ratio Annual Registration Fee Class 7&8 freight 4,953,849,811$ 1,194$ 5,103,686,085$ 1,231$ 0.97 Class 7&8 all types 4,953,482,996$ 1,145$ 5,108,179,079$ 1,180$ 0.97 Class 4-8 freight 4,950,000,000$ 700$ 5,176,895,949$ 732$ 0.96 Class 4-8 all types 4,950,000,000$ 550$ 5,225,168,441$ 580$ 0.95 Source: Tioga Group Analysis of 2002 VIUS (U.S. Census Bureau, 2004) and Table VM-1 of Highway Statistics 2009 (FHWA, 2009). Notes: * "Gas" in this case includes gasoline, natural gas, propane, alcohol fuels, and blends. Table 60. Revenue efficiency comparisons.

Annual Industry Cost (billion) Ratio $/Gal Net Annual Federal Revenue (billion) Annual Industry Cost (billion) Ratio 5.2$ 0.97 0.65$ 10.0$ 10.1$ 0.99 5.1$ 0.97 0.50$ 10.0$ 10.1$ 0.99 11.1$ 0.43 0.59$ 9.9$ 13.8$ 0.72 11.1$ 0.43 0.35$ 9.9$ 13.8$ 0.72 5.3$ 0.91 0.65$ 9.7$ 10.3$ 0.95 5.3$ 0.91 0.62$ 9.7$ 10.3$ 0.95 5.3$ 0.90 0.50$ 9.7$ 10.3$ 0.94 5.4$ 0.88 0.39$ 9.6$ 10.4$ 0.92 5.3$ 0.88 0.59$ 9.6$ 10.3$ 0.93 5.3$ 0.88 0.57$ 9.5$ 10.3$ 0.93 5.5$ 0.84 0.35$ 9.5$ 10.5$ 0.90 5.6$ 0.82 0.27$ 9.4$ 10.6$ 0.88 Annual Industry Cost (billion) Ratio $/VMT Net Annual Federal Revenue (billion) Annual Industry Cost (billion) Ratio 5.7$ 0.80 0.095$ 9.6$ 10.7$ 0.90 5.7$ 0.79 0.092$ 9.5$ 10.7$ 0.89 6.1$ 0.72 0.056$ 9.5$ 11.1$ 0.85 6.4$ 0.67 0.044$ 9.4$ 11.4$ 0.82 5.9$ 0.77 0.095$ 9.6$ 10.9$ 0.88 5.9$ 0.76 0.092$ 9.5$ 10.9$ 0.88 6.5$ 0.68 0.056$ 9.5$ 11.5$ 0.82 6.9$ 0.63 0.044$ 9.4$ 11.9$ 0.79 Annual Industry Cost (billion) Ratio $/Vehicle Net Annual Federal Revenue (billion) Annual Industry Cost (billion) Ratio 5.1$ 0.97 2,411$ 10.0$ 10.0$ 0.99 5.1$ 0.97 2,311$ 10.0$ 10.0$ 0.99 5.2$ 0.96 1,413$ 9.9$ 10.1$ 0.99 5.2$ 0.95 1,110$ 9.9$ 10.1$ 0.98 f Highway Statistics 2009 (FHWA, 2009). nue Goal $10 Billion Revenue Goal $/Gal Net Annual Federal Revenue (billion) Annual Industry Cost (billion) Ratio 1.30$ 20.0$ 20.2$ 0.99 1.00$ 20.0$ 20.1$ 0.99 1.18$ 19.8$ 26.1$ 0.76 0.69$ 19.8$ 26.1$ 0.76 1.30$ 19.8$ 20.3$ 0.98 1.25$ 19.8$ 20.3$ 0.98 1.00$ 19.8$ 20.3$ 0.97 0.79$ 19.8$ 20.4$ 0.97 1.18$ 19.7$ 20.3$ 0.97 1.13$ 19.7$ 20.3$ 0.97 0.69$ 19.6$ 20.5$ 0.96 0.54$ 19.6$ 20.6$ 0.95 $/VMT Net Annual Federal Revenue (billion) Annual Industry Cost (billion) Ratio 0.191$ 19.4$ 20.7$ 0.94 0.183$ 19.3$ 20.7$ 0.94 0.112$ 19.3$ 21.1$ 0.91 0.088$ 19.2$ 21.4$ 0.90 0.191$ 19.4$ 20.9$ 0.93 0.183$ 19.3$ 20.9$ 0.93 0.112$ 19.3$ 21.5$ 0.90 0.088$ 19.2$ 21.9$ 0.88 $/Vehicle Net Annual Federal Revenue (billion) Annual Industry Cost (billion) Ratio 4,822$ 20.0$ 20.1$ 0.99 4,622$ 20.0$ 20.1$ 0.99 2,827$ 20.0$ 20.2$ 0.99 2,221$ 20.0$ 20.2$ 0.99 $20 Billion Revenue GoalRevenue Option Fuel Tax Surcharge Options $/Gal Net Annual Federal Revenue (billion) Diesel fuel tax with non-freight tax refunds Class 7&8 freight 0.33$ 5.0$ Class 4-8 freight 0.25$ 5.0$ Diesel/gas* tax with non-freight refunds Class 7&8 freight 0.30$ 4.8$ Class 4-8 freight 0.17$ 4.8$ Diesel fuel tax with vehicle ID Class 7&8 freight 0.33$ 4.8$ Class 7&8 all types 0.31$ 4.8$ Class 4-8 freight 0.25$ 4.8$ Class 4-8 all types 0.20$ 4.8$ Diesel/gas* tax with vehicle ID Class 7&8 freight 0.30$ 4.7$ Class 7&8 all types 0.28$ 4.7$ Class 4-8 freight 0.17$ 4.6$ Class 4-8 all types 0.14$ 4.6$ VMT Fee Options $/VMT Net Annual Federal Revenue (billion) VMT Distance/ Vehicle Fee - OBU/Options Class 7&8 freight 0.048$ 4.5$ Class 7&8 all types 0.046$ 4.5$ Class 4-8 freight 0.028$ 4.4$ Class 4-8 all types 0.022$ 4.3$ VMT Distance/ Vehicle Fee - OBU Only Class 7&8 freight 0.048$ 4.5$ Class 7&8 all types 0.046$ 4.5$ Class 4-8 freight 0.028$ 4.4$ Class 4-8 all types 0.022$ 4.3$ Excise Tax Options $/Vehicle Net Annual Federal Revenue (billion) Annual Registration Fee Class 7&8 freight 1,206$ 5.0$ Class 7&8 all types 1,155$ 5.0$ Class 4-8 freight 707$ 5.0$ Class 4-8 all types 555$ 5.0$ Source: Tioga Group Analysis of 2002 VIUS (U.S. Census Bureau, 2004) and Table VM-1 o * "Gas" in this case includes gasoline, natural gas, propane, alcohol fuels, and blends. $5 Billion Reve Table 61. Scale economies.

93 For all three candidate mechanisms, low-tech implemen- tation approaches result in lower implementation, collec- tion, and compliance costs; shorter implementation time- lines; and greater revenue-generation efficiency. In all three cases, low-tech collection methods could build on existing systems. • Diesel or diesel/gas tax increases with refunds or credits for non-freight or light-vehicle operators could be imple- mented through relatively small changes in the existing fuel tax and income tax systems. • VMT fees for trucks could be implemented through a combination of IFTA participation, commercial fuel card systems, and self-reporting enforced through cross-checks with state registration systems. • A federal vehicle registration fee (or an extension of the HVUT) could be implemented through the existing HVUT system or by piggybacking on state vehicle regis- tration systems. High-tech collection solutions may offer greater preci- sion, potential linkages to future traffic-management sys- tems, or other non-revenue advantages, but would entail higher implementation, collection, and compliance costs. Implementation costs would include technology develop- ment, system and network deployment, and vehicle ID tag or OBU installation. Implementation costs would recur due to technology turnover, equipment lifespan, and new vehicle purchases. Federal collection and enforcement costs would include data bandwidth, systems operation and maintenance, auditing, and enforcement actions. Compliance costs would include OBU maintenance and replacement, record-keeping, and tax/fee transaction costs. Multimodal Application One major issue to be addressed across all the alternatives is the extent to which they could be applied to multiple freight modes or to trucks only. The nation’s freight infrastructure encompasses highways, roads, railroads, ports, inland and coastal waterways, airports, and pipelines. From a national planning and policy perspective, it would be desirable to have an infrastructure funding mechanism that could cover all these modes as required to achieve revenue and policy objectives. In most contexts, “freight infrastructure” refers to the surface modes—highway, rail, and water—and excludes pipeline and air. This study follows that convention. Airport infrastructure and operations, including air cargo, have their own revenue sources and funding infrastructure. There is little or no public involvement in pipeline infrastructure. This section discusses the three leading revenue-generating options—fuel tax surcharge, VMT fee, and registration fee— in terms of their potential application to multiple freight trans- portation modes. Multimodal Fuel Tax Surcharge A fuel tax surcharge could cover all modes that use fuel. Off-road uses, which include railroads, waterways, and air freight operations, are currently exempt from the major fed- eral fuel taxes. Railroads and waterways do, however, contrib- ute to the LUST fund, so there is a precedent for fuel taxation across the surface modes. Table 62 illustrates multimodal fuel tax surcharge options. Currently, diesel trucks pay $0.244 per gallon in federal fuel tax and inland waterway vessels pay $0.20. Of the total, only the inland waterway portion, $73.9 million, is dedicated to Class 4-8 Diesel Trucks Class I Railroads Inland Waterways 19,937,451,447 3,911,178,000 369,500,000 Rates 0.244$ -$ 0.200$ Revenue 4,864,738,153$ -$ 73,900,000$ Rates 0.364$ -$ 0.260$ Revenue 7,257,232,327$ -$ 96,070,000$ Rates 0.450$ 0.206$ 0.206$ Revenue 8,980,962,631$ 807,489,697$ 76,285,826$ Rates 0.504$ 0.260$ 0.260$ Revenue 10,048,475,529$ 1,016,906,280$ 96,070,000$ * Rate from Freight FOCUS Act and ITMS Plan. Fuel Tax Option 2008 Gallons Used Current Tax Rates 2010 Proposed Infrastructure Tax Rates* Uniform Infrastructure Rate IMTS Rate 4,938,638,153$ 73,900,000$ 7,353,302,327$ 2,488,564,174$ 9,864,738,153$ 5,000,000,000$ 11,161,451,809$ 6,296,713,656$ Revenue Total Dedicated to Freight Infrastructure Table 62. Multimodal fuel tax options.

94 Table 63. 2008 estimated vehicle counts. freight infrastructure. Applying the two 2010 infrastructure proposals, the Freight FOCUS Act and the Inland Marine Transportation Systems (ITMS) plan, would add $0.12 to the truck tax and $0.06 to the waterways tax, yielding about $2.5 billion for freight infrastructure. Of the $2.5 billion, $96.1 million would be dedicated to waterways. A uniform infrastructure fuel tax of $0.206 applied across all modes would yield the $5 billion gross revenue target for freight infrastructure used in other study comparisons, of which $76.3 million would be dedicated to inland waterways. Finally, an infrastructure fuel tax equal to the ITMS pro- posal for a minimum increase in the waterways tax ($0.26 per gallon) would yield $6.3 billion for freight infrastructure, of which $96.1 million would be dedicated to inland waterways. These examples, however, raise the issue of dedicated multimodal funds versus dedicated modal funds. One of the problems with the Highway Trust Fund is that it is not dedicated solely to highway uses. Trucking industry support for fuel tax increases is contingent on dedication to high- way uses. The Inland Waterways Trust Fund is already a dedicated modal fund. The Freight FOCUS Act, as drafted, stipulated that 92.5% of the revenue from any one mode be dedicated to projects for that mode and that half of any multimodal revenue would be prorated by the ratio of single- mode contributions. These observations suggest that a sin- gle multimodal fuel tax surcharge for freight infrastructure is technically feasible, but an attempt to create and allo- cate such a fund is likely to encounter serious institutional barriers. Multimodal VMT Fee The research team analyzed the potential for application of VMT concepts to railroads and waterways; however, “vehicle miles traveled” is basically a highway transportation metric. For both railroads and waterways, defining and applying VMT metrics would be highly problematic. For railroads, the definition of “vehicle” can be ambiguous, since trains are made up of changing mixes of freight cars and locomotives. The actual miles traveled are not recorded in railroad opera- tions. Moreover, a VMT fee for trains to travel over private railroad lines would be hard to justify. For waterways, “miles traveled” is difficult to define and is unrelated to infrastruc- ture requirements. Attempting to apply VMT fees to rail or water modes would thus entail costly and potentially oner- ous record-keeping, administrative, and enforcement costs, without a clear connection between such fees and freight infrastructure requirements. Multimodal Registration Fee Vehicle registration fees can conceivably be applied to the full range of freight transportation modes and the equipment they employ (absent any specific statutory prohibitions). Federal excise taxes are now applied to heavy-duty truck reg- istrations, the sale of new heavy-duty trucks, and heavy-duty truck tires. In general, vehicles that operate on public roads or waterways are registered at some level of government (or perhaps more than one level) and pay fees. Vehicles that operate on private land or off road (notably railroad equip- ment) are not registered and do not pay fees. Excise taxes have previously been assessed on a wide range of commodi- ties, and there is no conceptual reason why federal registra- tion fees could not be assessed on other trucks, railroad cars or locomotives, or waterborne vessels. As Table 63 shows, there were roughly 9.5 million medium- duty trucks, heavy-duty trucks, locomotives, freight cars, tow and tug boats, and barges in U.S. service in 2008. Of those, the trucks account for 95%. A federal registration fee of $555 (see Table 51) would yield an additional $262 million annu- ally if applied to rail and inland waterway vehicles. Linkage between Use and Payment Fuel taxes and VMT fees are user fees, with the linkage between use and payment dependent on program details. The ability of distance/vehicle VMT fees to link use, impact, and payment depends on the payment scale applied to trucks of different classes and weights. Registration fees are not closely linked to infrastructure use. The policy attraction of VMT systems is based in part on a preference for user fees over taxes. Many published sources consider VMT to be a better user fee than fuel taxes or excise taxes, which are regarded as an indirect measure of use. The “use” of a highway, road, or other transportation infrastructure is a function of four interrelated variables— miles traveled, GVW, gross axle weight, and congestion impact. The degree to which a revenue mechanism can take these factors into account determines how closely the fee can correspond to the use. Miles Traveled Miles traveled is a relatively clear concept and relatively easy to define for trucking. Miles traveled is the major fac-

95 tor in the need for infrastructure capacity and in the wear and tear on roads. Most highway vehicles would rarely travel significant distances off public roads, but there are obvious exceptions for vehicles used in agriculture, resource extrac- tion (e.g., mining and lumber), construction, and the mili- tary. The present system allows those vehicles to use untaxed fuel. To continue the exemptions, a VMT fee would either have to exempt the vehicles or exempt the off-road miles (possible with GPS or self-reporting systems). A registration fee system would exempt the vehicle based on type, owner- ship, or place of use. Gross Vehicle Weight The weight of the vehicle is another major factor in the need for infrastructure and infrastructure maintenance. Actual GVW includes the tare weight of the vehicle, the weight of the fuel and equipment on board, the weight of the load, and the weight of the driver and any passengers. As a practical matter, it is impossible to determine actual gross in- use vehicle weight without weighing every single vehicle on every single trip leg. Vehicles are classified by their gross vehi- cle weight restriction (GVW or GVWR), which is one basis for reliably attributing weight characteristics. A fuel tax indi- rectly captures some of the weight and size impact because larger, heavier vehicles typically have higher fuel consump- tion and thus pay higher effective tax rates per mile. If the impact is a function of GVW, then the GVW classification or an equivalent classification system can be used to graduate appropriate user or registration fees. Registration fees need only classify the truck once, when first registered. A gradua- tion method could depend on the relationship between GVW and some average operating weight and between that average operating weight and highway impact. The study team could locate no usable data on average gross operating weight. The concepts of average operating weight and the distribution of operating weight by GVW class appear to have received little attention to date. Gross Axle Weight Gross axle weight, in its simplest form, is the GVW divided by the number of axles. More complex issues of weight distri- bution entail application of complex “bridge formulas,” but those issues are more often addressed by vehicle configura- tion and inspection than by fees. The tire “footprint” also makes a small difference. Fuel taxes do not vary with axle weight. Axle weight distinctions have not been addressed by VMT fee discussions to date, but the number of axles is a factor in some toll calculations and state registration fees. Vehicles with more axles usually pay higher tolls and fees, presumably under the assumption that more axles mean more weight. VMT fees could be varied by the number of axles, but, because for any given weight, road impacts decline with the number of axles, the VMT rates should go down for additional axles in the same vehicle class. Congestion Impact The congestion impacts of trucks are sometimes measured in PCEs. The minimum PCE for trucks would ordinarily be the ratio of their length to passenger-car length, which var- ies between 1.0 and 3.0. The effective PCE for heavy trucks increases in congested traffic, in stop-and-go traffic, and on grades because the slower acceleration of heavy trucks increases the impact on other vehicle speeds. On steep grades with heavy congestion, truck PCEs may exceed 4.0. As with axle weights, there have been no VMT proposals that attempt to incorporate PCEs into the fee schedule. Fuel taxes capture PCEs only to the extent that PCEs correspond to weight and fuel use. Registration fees can also take length into account for straight trucks, truck tractors, and trailers. These considerations illustrate the challenges faced in making any revenue source correspond as closely as possible to vehicle infrastructure impacts. The impact of gross vehicle weight and axle weight on pavement has been a controver- sial issue, and the congestion impact of trucks in mixed flow is a new area of study. Attempting to build a more precise reflection of use impacts into a VMT or a registration fee schedule will likely raise serious political and equity issues as well as technical difficulties. One option may be to charge all vehicles the same distance-based VMT fee and use other mechanisms, such as registration fees, to reflect the difference between vehicles and their impact. This option would have the advantage of simplifying VMT fee compliance, collection, and enforcement. Incentives The revenue mechanisms analyzed would not have strong productivity or behavioral incentives, largely because their impacts would be small compared to other freight and service industry cost factors. The incentives that the revenue mecha- nisms may provide are described in the following: • A fuel tax surcharge or VMT fee that focused on trucks with freight body types would encourage the use of other truck types to move freight, including the use of freight trailers behind service vehicles such as medium-duty pickups. • Mileage-based VMT fees would encourage truck driv- ers and dispatchers to minimize mileage, but since they already have strong incentives to do so, the marginal impact would be small. • The use of OBUs for a VMT fee system would likely accelerate the adoption of telematics by the trucking industry.

96 • If graduated by size or weight, an annual registration fee could encourage users to choose the smallest (and presum- ably most fuel-efficient and lowest-emissions) truck that will perform the task at hand. • A higher or broader federal registration fee would discour- age operators from keeping older medium- and heavy- duty trucks for occasional or seasonal use. Such uses are often dominated by old, inefficient, high-polluting trucks with questionable safety status, so such an incentive might be seen as a public benefit. • A diesel fuel tax surcharge could have more behavioral impact. A tax surcharge that increased the cost of diesel fuel relative to gasoline could lead to significant substitu- tion of gasoline-powered medium- and heavy-duty trucks for the present diesel fleet. The environmental impacts of the three leading revenue options would likely be equally modest. VMT fees, especially time/location VMT fees that incorporate congestion pric- ing, are expected to have significantly favorable impacts on passenger-car driving habits, based on the results of pilot pro- grams. Passenger-car drivers, however, don’t have the strong existing cost incentives to minimize mileage that truck drivers do, nor do passenger-car drivers have the supervision of cost- conscious managers and dispatchers. Impacts The relatively modest incentives created by the various options suggest that transportation system impacts of the taxes or fees will be minimal. With the exception of the pos- sible substitution of gasoline trucks for diesel trucks with the diesel tax surcharge, the taxes or fees would not tend to change the way freight haulers operate. The analysis of modal shift in Chapter 6 (see Table 46) esti- mated annual lost truck miles at between 3.2 and 3.7 billion, roughly 1.5% of the annual total for Class 4–8 trucks. The relatively small impact is due to the predominance of truck movements under 500 miles where rail is not competitive and to the relatively small increase in overall trucking costs. In a multimodal application, an identical diesel fuel tax surcharge could have very different operating cost impacts on the highway, rail, and inland water modes. Table 64 shows esti- mates of the percentage operating cost impact of a $0.25 per gallon diesel fuel tax surcharge. These estimates are imprecise due to the difficulty of obtaining comparable data on oper- ating cost structures. The estimates do, however, illustrate the range of impacts. The percentage increase in the example is actually lowest for trucks, because trucks already pay the highest fuel costs. The impact is greatest for the inland water- ways carriers because fuel is such a large part of their cost structure. Chapter 6 analyzed economic impacts in terms of (1) costs imposed on industry and (2) output lost by industry. The resulting economic impact estimates are shown in Table 65. The greater impacts of the diesel/gas tax and the VMT fee are due to their higher implementation and industry compliance costs. The estimates are in the vicinity of $500 million, about 10% of the $5 billion gross tax revenue target used for the analysis. As Chapter 6 noted, the primary economic conse- quence of the tax (from industry’s perspective) is slower GDP growth and lower income rather than long-term employ- ment loss. Equity For most revenue-generating options, equity is an issue of design and implementation. There are equity issues with the fuel and excise taxes as currently implemented, but those issues could be resolved. The equity effects on freight, ser- vice, and passenger users will depend on fee system design. Proposals to tax freight trucks exclusively would raise equity issues between freight and service truck operators. Targeting Freight Vehicles To fund infrastructure through a tax or fee on “freight” vehicles or their operations, it is necessary to define the sub- ject vehicles with precision. While most discussions of freight infrastructure and funding conjure up images of heavy-duty Table 64. Impact of $0.25/gallon diesel fuel surcharge. Sources: ATA, Coosa-Alabama River Improvement Association, U.S. DOE, Tioga Group estimates. Operating Cost 2008 Fuel $/Gal. Increase at $0.25/Gal. 3.

97 trucks; in reality, the truck population includes a wide range of vehicles used for many purposes. This issue is treated at greater length in Appendix B. Figure 15 illustrates the problem graphically. Class 4–8 medium- and heavy-duty trucks are a subset of all Class 1–8 trucks. Diesel trucks are also a subset of the Class 1–8 trucks. Freight trucks (defined either as freight body types or trucks used in goods movement applications) are another subset of Class 1–8 trucks, and most are a subset of Class 4–8 medium- and heavy-duty trucks. The intersection of these three sub- sets is medium- and heavy-duty diesel freight trucks—about 3.1 million vehicles, or 3.8% of the 2002 VIUS total for Class 1–8 trucks. A diesel fuel tax surcharge would affect the 7.4% of the fleet that uses diesel, about 6.1 million trucks, but only about half of them haul freight. For the fuel tax surcharge and VMT fee alternatives, a substantial part of the implementation, compliance, and col- lection costs would be incurred in identification of freight vehicles. As shown in Table 66, however, a large portion of the medium- and heavy-duty trucks in use do not haul either freight or passengers, but are used in the provision of services. The majority of the U.S. fleet, about 60%, is in mixed-use sectors, notably construction. In many instances, the use of diesel fuel is implicitly equated with freight transportation. As Table 67 shows, however, even in the heaviest classes of trucks, other fuels account for about 9% of the vehicles. In medium and light- heavy vehicle classes, which together account for about the same number of vehicles as the heavy-heavy class, only about half of the vehicles use diesel fuel. A diesel fuel tax, therefore, would cover only part of the freight trucking industry. A tax or fee focused on a subset of the whole Class 4–8 medium- and heavy-duty truck fleet would therefore raise serious equity concerns. The most pragmatic approach, which would also yield the highest revenue, would be to apply the tax or fee to all Class 4–8 medium- and heavy-duty trucks, with the understanding that freight and service vehicles of similar sizes have similar infrastructure requirements and would benefit similarly from infrastructure improvements. Appendix B discusses freight and truck fleets in more detail. Electric Vehicles Fully electric or plug-in, hybrid, heavy-duty freight or ser- vice vehicles are likely to be uncommon in the near future, but could eventually become a factor. Electric freight or Table 65. Economic impacts of tax scenarios as lost 2021 GDP ($m2012). *Does not include net tax revenue. Figure 15. Truck population segments. CLASS 4-8 TRUCKS 6.5% DIESEL TRUCKS 7.4% FREIGHT TRUCKS 5.1% ALL TRUCKS CLASS 1-8 DIESEL FREIGHT 3.8%

98 service vehicles have been used in the past for urban delivery and will likely become a practical possibility with continued advances in battery technology. Applying fuel taxes to electricity is difficult. Power con- sumption would need to be tracked at either the point at which vehicles were charged or on the vehicle itself. Electric vehicles can be charged from almost any conventional elec- tric power outlet with appropriate converters or on-board equipment, so metering at the point of charging would likely be impractical. On-board metering or tracking, like VMT fee systems, would require some means of reporting use and cal- culating fees. Most electric vehicles have regenerative brak- ing, so while going downhill they “make” fuel. Electric power use, therefore, is not an accurate proxy for electric vehicle impact on infrastructure. VMT fees are a logical way of assessing electric or partially electric vehicles for infrastructure impacts. On-board OBUs and self-reporting options would both be applicable. An annual registration fee system would apply equally well to electric or partially electric vehicles. Since the registration fee could vary by vehicle, it would be possible to create incen- tives for electric vehicle ownership and use or higher fees to offset loss of fuel tax revenue. International Trucking There is a substantial volume of cross-border trucking activity that may pose problems in some fuel tax, VMT fee, or registration fee systems. At present, U.S. and Canadian trucks move freely across the border. The situation at the Mexican border is much different and highly controversial. The predominant pattern is relatively short shuttle move- ments across the border between factories, distribution cen- ters, or drop lots. Longer trips to and from U.S. inland points are typically handled by U.S. truckers operating to and from border drop lots rather than crossing the border. Longer Table 66. Uses of medium- and heavy-duty trucks. Sector Total Share Freight 4,330 34% For-hire Transportation & Warehousing 1,280 18% Retail Trade 1,531 7% Wholesale Trade 736 5% Manufacturing 783 4% Mixed Freight/Service 12,526 60% Construction 4,541 19% Agriculture, Forestry, etc. 2,240 14% Not Reported/Not Applicable 3,286 11% Vehicle Leasing or Rental 859 6% Waste Hauling, Landscape, Admin/support 743 5% Utilities 679 3% Mining 178 2% Service & Personal Transport 68,318 6% Other Services 2,127 3% Accommodation & Food Services 284 1% Information Services 377 1% Personal Transportation 65,343 1% Arts, Entertainment, Recreation 187 0% Total 85,174 100% Source: 2002 VIUS (U.S. Census Bureau, 2004). Table 67. Truck classes and fuel used. Medium (GVW Class 4-5) Light-Heavy (GVW Class 6) Heavy-Heavy (GVW Class 7&8) Total Fuel Share of Truck Type Gasoline 54.6% 40.1% 8.5% 30.0% Diesel 44.8% 57.5% 90.9% 69.1% Alternative 0.6% 2.4% 0.6% 0.9% Source: 2002 VIUS (U.S. Census Bureau, 2004, Table 5).

99 trips within Mexico are a mirror image, handled by Mexican truckers to and from drop lots south of the border. Under the North American Free Trade Agreement (NAFTA), Mexican truckers were to have been permitted to operate in the United States beyond the border zone. That permission has been withheld, however, and operation of Mexican trucks in the United States remains an industry and political hot button. The infrastructure needs along both borders are already substantial and will increase as trade with both partners grows. If VMT became the primary basis for infrastructure finance, border states could be expected to demand that Canadian and Mexican trucks pay their fair share. When and if Mexican trucks handle more trips north of the border, there will be a corresponding loss of fuel tax revenue unless Mexico is brought into the IFTA system. Most Canadian provinces are partners to IFTA. Fuel taxes paid by U.S. trucks operating in Canada and Canadian trucks operating in the United States are therefore already allocated between states and provinces based on VMT. Mexican states are not yet parties to IFTA, and there is little reason for them to join while Mexican trucks are prohibited from operating in the United States beyond the border area. An extension of IFTA to Mexican states and truckers would be a logical step when and if the NAFTA logjam is broken. It would be difficult to justify a U.S. mandate to equip foreign cross-border trucks with VMT-compliant OBUs or EOBRs. Some firms specialize in cross-border trips while others rarely or never make such trips. Moreover, such a requirement could violate the terms of NAFTA or other cross-border agreements. Direct fuel tax revenue from Canadian and Mexican trucks is likely to be minimal. The heavy-duty Class 7 and 8 vehicles that are likely to be used in cross-border shipping typically have large fuel tanks holding up to 250 gallons, or enough fuel to travel 1,000–1,200 miles. U.S. and Canadian trucks moving between border and near-border states would rarely need to refuel in the other country (but might do so if diesel is cheaper across the border). Diesel fuel is much cheaper in Mexico, so Mexican truckers avoid buying fuel in the United States. Mexican diesel can have higher sulfur content and damage U.S. fuel systems, so U.S. truckers rarely buy fuel in Mexico. U.S. state and federal governments do, however, receive a prorated share of Canadian fuel tax revenue accord- ing to the miles traveled south of the Canadian border. There is no comparable revenue stream from Mexican trucks. Some alternatives for addressing the problems that may arise with some fuel tax, VMT fee, or registration fee systems in international trucking would include the following: • Requiring cross-border trucks to install VMT-compliant OBUs and report U.S. mileage separately. • Requiring cross-border trucks to be enrolled in IFTA. • Requiring a manual log and VMT fee payment for travel in the U.S. VMT-compliant OBUs capable of tracking and reporting U.S. mileage separately would have to be GPS-enabled, or they would have to be units of the kind required for time/ location VMT as opposed to mileage-only units sufficient for distance/vehicle VMT. The need for more elaborate OBUs would greatly increase the compliance cost for cross-border truckers, particularly if U.S. truckers did not need them. If this were the case, it would tend to perpetuate the cross- border shuttle system prevalent in Mexican trade and perhaps encourage such a system for Canadian trade as well. There would be an unknown but probably significant increase in border area congestion, VMT, and emissions, and a corre- sponding loss of efficiency. Canadian trucks regularly engaged in cross-border moves are normally enrolled in IFTA. These vehicles could convert to distance/vehicle VMT fees at little or no cost. Extension of IFTA to Mexico would involve up-front political and institu- tional investments and possibly protracted negotiations, but cost would be minimal and compliance cost very low (as little as $25 per month for IFTA filing). Non-IFTA trucks could have an option of a one-time fee at the border or could inter- change the shipment with an IFTA-enrolled trucker. Manual filing of VMT fees for border crossings would be cumbersome, costly, and unreliable. The audit and enforce- ment capability would be very limited. One workable alterna- tive might be a one-time payment in lieu of OBU installation or IFTA enrollment. Federal registration fees would not apply directly to Cana- dian or Mexican trucks, but an apportionment plan along the lines of the IFTA or UCRP system would appear to be applicable. Exemptions New funding mechanisms offer an opportunity to elimi- nate or phase out current fuel tax exemptions and promote equity among classes of vehicles and users. Certain users are exempted from paying federal excise taxes on motor fuels for on-highway use, including vehicles that fall into the follow- ing categories: • Vehicles exclusively used by federal, state, county, or local government. • School buses. • Qualified intercity and local buses. • Vehicles used by non-profit educational organizations. Individual states also have their own categories of exempt highway users. If these exemptions are not eliminated, they will have to be accommodated within the new system.

100 In 2007, ATRI estimated an overall fuel tax revenue loss of $900 million annually due to these exemptions, as shown in Table 68. The federal portion was $570 million. The exemption issue affects all three of the leading candi- date mechanisms identified in this study. Increasing or sur- charging current fuel taxes would not affect exempt vehicles or users. VMT fees could potentially be assessed on vehicles and users that do not now pay fuel taxes. The barriers would be political and institutional. Registration fees could likewise be extended to the full range of vehicles if politically and publicly acceptable. Many off-road and non-vehicle operations, including rail- roads, barges, machinery, stationary engines, electric genera- tors, and so forth would be easier to segregate in a VMT system because VMT for that equipment would not be tracked. It is likely that there would be no VMT fees for farm equipment, for example, as any on-road travel would be ancillary to routine operation. Off-road highway vehicle travel poses a challenge to poten- tial VMT systems. Currently, vehicles that operate off road can use untaxed fuel. Untaxed fuel is either bought in bulk or delivered to jobsites using mobile equipment. Closing this exemption would reduce the current evasion problem, but would create other difficulties. Many off-road vehicles (e.g., construction equipment) do not have odometers and lack the electronic systems that would be used to connect OBUs. It is also a conceptual leap to apply infrastructure taxes or fees to vehicles that spend their time operating off the infrastructure that the fees are designed to support. Exemption of government use vehicles currently reduces net federal fuel tax revenue by an estimated $363 million annually. Slightly more than 2% of the trucks in 2008 were owned by public agencies (see Table 69). Most of the medium-duty, public-sector trucks are owned by county and local agencies (see Figure 16). For those public-agency fleets that use fuel cards to obtain gas or diesel at commercial gas stations or card- lock facilities, the withdrawal of the exemption could be accomplished via procedural and accounting changes. If third-party VMT tracking via commercial OBUs becomes common practice, there would be multiple options for public-sector vehicles: • They might not be tracked. • They could be tracked but their VMT and TDM fees reported at zero (no payment). This option would have the distinct advantage of generating VMT and TDM data for private-sector vehicles. • They could be tracked and their VMT fees reported and paid at a different (presumably lower) rate. Public agencies that purchase fuel in bulk could either report fuel use or VMT and calculate fees manually or use OBUs for that purpose. There are also commercial auto- mated and manual systems that can track vehicle mileages in conjunction with bulk fuel use. If both VMT and fuel usage were reported, discrepancies could again trigger audits. Trucks and other vehicles owned by public agencies are registered, but public agencies do not pay the same registra- tion fees as owners of private vehicles. Assessing federal excise taxes or registration fees on these vehicles would be relatively simple and would not have extensive technology or reporting requirements. Application of fuel taxes, VMT fees, or registration fees to government vehicles, however, would be a transfer between government entities and would by itself bring no more funds into the public sector. To pay fees while retaining their same net budget level, government agencies would have to increase revenue from other sources by the same amount. For federal Table 68. Total annual federal and state fuel tax exemptions. Sector Federal Exemptions State Exemptions Total Government Use Vehicles 363,000,000$ 155,000,000$ 518,000,000$ School Bus Use 146,000,000$ 126,000,000$ 272,000,000$ Transit Use 61,000,000$ Unknown 61,000,000$ Federal Use N/A 29,000,000$ 29,000,000$ U.S. Postal Service N/A 27,000,000$ 27,000,000$ Charitable Organizations Unknown Unknown Unknown Total 570,000,000$ 337,000,000$ 907,000,000$ Table 69. Truck fleet composition—2008.

101 agencies, funding fees would presumably entail offsetting increases across the whole mix of federal revenue sources. Unless federal agencies accepted a net budget loss, the impact of federal agency freight fees would be spread across the general population. Taxes or fees paid by state and local government agen- cies would presumably be funded by increases in their cur- rent mix of revenue sources. The increased expenditures on transportation infrastructure would likely pass through state governments, principally the state highway departments. The amounts may not balance, however, which would result in friction between state and federal levels (just as there is now friction over “donor” and “donee” states under the Highway Trust Fund). Within a state, the effect would be to transfer funds from other state agencies and their revenue sources to the state Federal-Aid highway and infrastructure program. On the local and regional level, taxes or fees would be paid from the existing spectrum of revenue sources, which can include state and federal funds. In effect, subjecting public vehicle fleets to fees would thus result in passing the fees through to the general public. To the extent that there is a net increase in federal revenue from extending a freight fee to public agencies that are exempt from the fuel tax, the burden will thus ultimately be transferred to the existing mix of non-fuel-tax revenue sources. The economic impact is likely to be roughly the same as transferring General Fund revenue to the Highway Trust Fund and increasing the General Fund tax burden accordingly. Public and Industry Acceptance Overall Indications There is limited information on the overall acceptance of freight-focused revenue mechanisms. Most of the studies and surveys have focused on public acceptance of taxes or fees on passenger vehicles. Explicit linkage to environmental issues tends to increase support for any type of fee or tax. It also appears that public acceptance may vary greatly with the details of implementation. The ATA, and other industry organizations, and most likely their members, are concerned that VMT fees or other pro- posals do not address one of their greatest concerns over the existing fuel tax: diversion to non-highway uses (American Transportation Research Institute, May 2007). The industry is very unlikely to support any revenue mechanism that does not incorporate funding dedication to highway infrastructure. The Associated General Contractors of America, on the other hand, has voiced their concern over the rising costs of infra- structure coupled with the decline in effective funding and endorsed the whole range of potential solutions (Statement of James D. Waltze, National Surface Transportation Policy and Revenue Study Commission Field Hearing, Los Angeles, CA, February 22, 2007). There is little support in the literature for applying new revenue mechanisms to non-highway modes. The ATA sup- ports inclusion of other modes in Highway Trust Fund rev- enue collection if those modes are eligible for Highway Trust Fund spending. Yet, there appears to be little support for using Highway Trust Fund revenue for other freight modes. The ATA also supports development of a multimodal funding mechanism for multimodal freight projects. Fuel Taxes and Surcharges The National Surface Transportation Infrastructure Financing Commission noted strong public opposition to motor fuel tax increases as one of the problems associated with the fuel tax. As of mid-2010, the current administration had eliminated fuel tax increases from consideration. There are, however, specific proposals and indications of support for fuel tax increases. The Freight FOCUS Act of 2010, described earlier, would raise the diesel tax by $0.12 per gallon and index the tax for inflation to fund a Goods Movement Trust Fund. The proposal did not pass in 2010, but can be regarded as an initial placeholder for more serious efforts in 2011 and beyond. The measure is supported by the Figure 16. Public-sector Class 1–5 trucks in service—2009. Federal Agencies 13% State Agencies 17% County & Local (less police) 70% Source: Automotive Fleet 2009

102 trucking industry, whose members would bear the brunt of the increase. ATA has also proposed an increase in the die- sel tax and tied it to proposals to reduce diversion, reduce evasion, and narrow exemptions. The AASHTO proposal to convert the fuel tax to a sales tax would effectively increase the tax rate because fuel prices are expected to rise. VMT Fees Studies and surveys to date suggest that the public is very skeptical regarding VMT systems. There is a range of con- cerns, with privacy being the most prominent and consis- tent. The potential use of GPS in connection with VMT fees is apparently a cause for great concern. Acceptance of VMT fees within the freight and service trucking industry will depend on the amount of the fees and the means of implementation. The principle concerns of truck owners and operators will be the following: • The base per-mile rate as applied to trucks. • The graduation of fees with increasing truck weight or GVW class. • The higher capital, collection, compliance, and enforce- ment costs of VMT fees compared to existing fuel taxes. • The potential need for VMT-compatible technology investments that may duplicate or make obsolete com- mercially justified systems. • The higher potential for evasion. The ATA remains strongly opposed to weight-distance taxes of any kind, and a distance/vehicle VMT fee would be a form of weight-distance tax. The analysis of VMT fee structures opens the possibility of large annual fee increases for the larger truck classes. Experience with weight-distance taxes in Oregon, New York, New Mexico, and Kentucky leads to legitimate concerns over record-keeping and compliance costs, as well as over the expense of the tax itself. Acceptability within the industry would also depend on the revenue collection mechanism. About 25% of the 9 mil- lion medium- and heavy-duty trucks in the United States are enrolled in IFTA, for which the owners are already recording and reporting VMT. Another significant but unknown per- centage of trucks are equipped with commercial or propri- etary telematics or EOBR systems that are probably capable of recording and reporting mileage. Truck owners would quite reasonably object to VMT implementation plans that duplicated existing capabilities at additional cost. Acceptance of VMT fees by other modes is very unlikely. The application of VMT fees to railroads or coastal and inland waterways vessels would be awkward at best and is hard to justify. Registration Fees Registration fees and excise taxes are not popular in the truck- ing industry. When business conditions are difficult, mar- ginal operators sometimes cannot make the annual HVUT payment. Tire taxes and sales taxes on new trucks are, with some justification, criticized for discouraging new invest- ments that could advance safety and environmental objec- tives. At present, the three federal excise taxes are assessed on the heaviest trucks, but the revenue is not dedicated to trucking infrastructure. The literature reviewed for this study had relatively little coverage of non-fuel excise taxes or registration fees. The excep- tions are testimony on behalf of the Associated General Contractors of America, which mentions a vehicle sales tax as an alternative to motor fuel taxes, and the work of the Mineta Institute, discussed below (Waltze, February 22, 2007). The acceptance of a federal registration fee might be more widespread if the fee were paid quarterly or at some other interval rather than an annual lump sum. Shifting from an annual basis, however, would likely eliminate the potential for piggybacking on state registration fee collection. Public acceptance of a broader federal truck registration fee might also be greater if the truck tire and truck sales taxes were allowed to expire. Such an option, however, would result in a major loss of Highway Trust Fund revenue. Survey Results The Mineta Transportation Institute at San Jose State Uni- versity has conducted a series of surveys to explore public opinion regarding a range of revenue-generating options. A 2005 white paper summarizes Mineta Institute survey findings in which three alternatives stand out: tolled high- ways (treated in this study as a project-specific option); higher fuel taxes; and registration fees tied to emissions and fuel use (Weinstein and Dill, February 22, 2007). While the survey mainly concerned passenger vehicles and their owners, the results (see Table 70) indicate stronger support for fuel taxes than many observers had expected and weak support for mileage-based (VMT) fees. A subsequent June 2010 report by the Mineta Institute reviewed surveys to date and conducted a new survey of attitudes toward transportation tax increases (Agrawal and Nixon, June 2010). Table 71 shows the study’s findings regarding previous surveys. Most surveys found the percent- age of respondents favoring a mileage tax to be quite low. The most favorable result was the most recent (HNTB in 2010), but that survey apparently phrased the question to refer to a “mileage use tax” without specifying a mile-by-mile VMT fee. The 2009 California survey found that responses were more favorable when the proposal was linked to vehicle

103 Table 70. 2005 Mineta Institute survey findings. Revenue Option Description of Option from Questionnaire % of Respondents Supporting the Option Truck-only toll (TOT) lanes There were proposals in some congested regions to build new toll lanes for trucks right next to existing freeways. Trucks would be required to use these toll lanes instead of the regular freeway. (Survey 2) 64 High-Occupancy/Toll (HOT) lanes Open underused carpool lanes to solo drivers who were willing to pay a toll. (Survey 1) 55 Toll roads One option for building new highway projects without increasing taxes is to borrow money to build the road, charge tolls for driving on the new highway, and use the money collected to pay back the loans and maintain the highway. (Survey 2) 47 Environmental vehicle registration fees Increase the vehicle registration fee to an AVERAGE of $62 per year for all vehicle owners, but vary the fee according to how much pollution the vehicle emits and how much gas mileage it gets. Vehicles that emit more pollution or get lower gas mileage would pay HIGHER fees and those that emit less pollution or get better gas mileage would pay LOWER fees. (Survey 1) 44 Express toll lanes Building new freeway lanes alongside existing highways and charging a toll to drivers who use those NEW lanes. (Survey 2) 44 Gas tax Increase the 18-cents-a-gallon state gas tax by one cent per year for ten years. (Survey 1) 40 Sales tax Adopt a half-cent increase in the statewide sales tax. (Survey 1) 40 Vehicle license fee Raise the vehicle LICENSE fee to 1%. The vehicle license fee is currently 0.65% (point six-five percent) of your vehicle's value, so the new fee would be 1%, with the additional revenue dedicated to transportation purposes. (Survey 1) 40 Tolls on new highway lanes One way to pay for new highway lanes is to charge tolls for using them. (Survey 1) 40 Registration fees Increase the vehicle REGISTRATION fee to $62 per year per vehicle, from its current level of $31. 32 General obligation bonds One proposal is for the state to pay for new freeways and transit programs with general obligation bonds. These don't require a tax increase. But paying off the bonds from the state's general fund over 30 years would use money that otherwise might be spent for other programs and services. 30 Indexed gas tax Index the gas tax to inflation. Under this proposal, the gas tax could increase slightly each year based upon inflation. For example, in 2004, inflation in California was about 3%, so the tax would have gone up by about a half cent per gallon. (Survey 1) 27 Mileage fee Eliminate the 18-cents-a-gallon gas tax altogether and replace it with a so-called "mileage fee" based on the number of miles a vehicle is driven. Each driver would pay a fee of one cent per mile for every mile driven within the state. For example, every 100 miles driven would incur a mileage fee of $1. Each vehicle would be equipped with an electronic means to keep track of miles driven and the fee would be paid at the pump when drivers buy gas. (Survey 1) 22 N: Survey 1: 2705; Survey 2: 815

104 Table 71. Mineta Institute study summary of mileage tax survey results. Sponsor (and Author, if Different) Survey Data Sampling Frame Findings HNTB Corporation (Kelton Research) 2010 U.S. residents 39% of respondents agreed with the statement "the U.S. should try to reduce transportation greenhouse gas emissions by reducing the number of miles that vehicles travel through a mileage use tax." Mineta Transportation Institute (Agrawal et al.) 2009 California residents 28% of respondents "supported" replacing the state gas tax with "a fee of 1-cent per mile for every mile driven within the state." Respondents were informed that "vehicles would be equipped with an electronic means to keep track of miles driven, and the fee would be paid when drivers buy gas." Support for the proposal increased to 50% for a variation in which "vehicles that pollute the least would pay less, and vehicles that pollute the most would pay more per mile." Mineta Transportation Institute (Weinstein et al.) 2006 California likely voters 23% of respondents "would vote for" replacing the state gas tax with a mileage fee where "each driver would pay a fee of 1 cent per mile for every mile driven within the state." Respondents were informed that "vehicles would be equipped with an electronic means to keep track of miles driven, and the fee would be paid when drivers buy gas." Rasmussen Reports 2009 U.S. residents 18% of respondents "favored" some form of mileage tax "to help fund the building and repair of roads and bridges." Civitas Institute 2009 North Carolina registered voters 12% of respondents "would view favorably" a switch to "a plan that would charge all drivers based on the number of miles they drive in North Carolina." (The question did not specify what the "current system" was.) pollution. Ironically, the fuel tax is linked much more closely to vehicle pollution than VMT fees. Figure 17 shows the results of the Mineta Institute survey for 2010 and 2011 (Agrawal and Nixon, June 2011). A mileage tax of $0.01 per mile was the least popular option, with 21–22% respondent support. Gas tax variations were generally more popular than VMT fees. As the report notes, support increased with linkage to specific environmental issues and benefits. Figure 18 shows the actual question asked regarding a “mileage tax” and the responses in the 2010 Mineta Institute survey (Agrawal and Nixon, June 2010). The responses were in fact highly negative, with over 60% strongly opposing a mileage tax. A public outreach effort in connection with the Nevada DOT VMT fee study (Rawlins, April 20, 2010) found that following a public meeting, there was this response from attendees: • 68% indicated that the public meeting helped address their concerns. • Areas that were designated as needing more attention were privacy (37%) and policy (28%). • 515 indicated “Yes” or “Maybe” on a question asking about their willingness to participate in a pilot study. • 58% indicated that they would not be willing to have VMT technology in their vehicle. Table 72 summarizes results from a Minnesota DOT sur- vey of public options regarding VMT fee or MBUF systems. Respondents were presented with a high-tech approach involving GPS-based OBUs and a low-tech approach that did not involve GPS. The results show a much stronger concern for privacy with the high-tech GPS approach. The author of the presentation described GPS as a “deal breaker” (Buckeye, April 20, 2010). Respondent opinions of VMT fees are indeed generally more favorable after they participate in pilot programs. Fig- ure 19 shows opinions shifting from negative towards posi- tive from screening (start of pilot program) through baseline (during pilot) to exiting (after pilot). Their concerns regard- ing privacy tend to diminish, and they are more willing to accept less detailed statements in return for greater privacy as their trust in the system increases.

105 Figure 17. Results from Mineta Institute survey. 21 23 33 33 39 43 43 30 22 24 36 36 39 45 45 48 50 56 62 0 10 20 30 40 50 60 70 Mileage tax: flat rate of 1¢ per mile Gas tax: 10¢ increase Gas tax: 10¢ increase with information about average driver's annual costs Milegae tax: rate varies by vehicle's pollution level (average 1¢ per mile) Gas tax: 2¢ increase per year, for 5 years 0.5¢ Sales tax Gas tax: 10¢ increase with revenue dedicated to transportation projects to reduce global warming Gas tax: 10¢ increase with revenue spent to reduce local air pollution Gas tax: 10¢ increase with revenue spent on projects to add more modern, technologically advanced systems Gas tax: 10¢ increase with revenue spent on projects to reduce accidents and improve safety Gas tax: 10¢ increase with revenue spent on projects to maintain streets, roads, and highways Percent Support levels for the tax options surveyed in 2010 and 2011 2011 2010 "Support" is the sum of those who said they strongly or somewhat strongly supported the tax option. Figure 18. Mileage tax (VMT) question from Mineta Institute survey. Unweighted Weighted Strongly support 6% 9% Somewhat support 14 12 Somewhat oppose 15 15 Strongly oppose 64 61 Don't know (volunteered) 2 3 8a. One idea (a DIFFERENT idea) is to adopt a new tax based on the number of miles a person drives. Each driver would pay a tax of 1 cent for every mile driven. For example, someone driving 100 miles would pay a tax of 1 dollar. Vehicles would have an electronic meter to keep track of the miles driven, and the tax would be paid each time drivers buy gas. Would you strongly support, somewhat support, somewhat oppose, or strongly oppose this new mileage tax?

106 Table 72. MBUF public opinion study results. Source: Buckeye, April 20, 2010. % % Figure 19. VMT pilot project survey results. Very Positive Positive Neutral Somewhat Negative Very Negative Screening 0.15 0.23 0.33 0.11 0.04 Baseline 0.17 0.39 0.27 0.13 0.04 Exiting 0.29 0.41 0.12 0.09 0.08 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 How do you feel about the idea of replacing the gas tax with a mileage-based road user fee? Source: Hanley and Kuhl, April 20, 2010. All of the pilot tests to date, however, have been tax-neutral to the participants in that none have paid more in VMT fees than they would have in fuel taxes. The attitude of the public toward VMT fees in addition to fuel taxes has not been tested in pilots, nor have surveys to date addressed the issue. Figure 20 provides results from the 2009 Mineta Transpor- tation Institute survey indicating greater support for registra- tion fees than for mileage-based fees (VMT fees) (Agrawal, Dill, and Nixon, June 2009). A survey on “green” taxes and fees in California found that 41% supported an increase in vehicle registration fees. That support rose to 63% for “green” registration fees linked to vehicle emissions. Table 73 provides additional survey results, which tend to support inclusion of fuel consumption and emissions factors in registration fees. Implications The limited amount of information available on accep- tance by the public and by industry suggests that VMT

107 Table 73. Support for vehicle registration fees. Note: Percentages may not total to 100% due to rounding. Source: Agrawal, Dill, and Nixon, June 2009. fees face a number of obstacles, notably the privacy issue. The public is generally more comfortable with fuel taxes or registration fees, particularly when these fees are linked to “green” initiatives. Industry acceptance appears to parallel public acceptance, although for slightly different reasons. The details of implementation matter a great deal to both groups. Summary Comparisons Table 74 summarizes and compares the leading revenue options based on key evaluation criteria. The ratings are rela- tive, rather than absolute. Moreover, many of the rankings depend on the details of revenue mechanism design rather than fundamental characteristics. Technical feasibility per se is not an issue for the lead- ing options. The significant difference is in the degree to which each alternative relies on technology, with the more technology-intensive options being costlier and taking lon- ger to implement. The potential for evasion varies; tax sur- charges or VMT fees that vary between vehicles or depend on technology are more vulnerable to evasion than fuel taxes or registration fees that use existing systems. The three leading options all have the potential to yield substantial revenue to fund freight infrastructure. There are no inherent limits, so the revenue potential is primarily an issue of political, industry, and public acceptance of appro- priate taxes or fees. The sustainability of fuel tax revenue over time in the freight sector is higher than in the passenger sec- tor, but still subject to diminishing growth. VMT fee revenue will grow as VMT continues to grow. Truck registration fees vary with the size of the fleet, which tends to diminish or not grow as quickly during economic slowdowns. Figure 20. Support for VMT fees versus registration fees. 0.2 0.41 0.5 0.63 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 Mileage Fee Vehicle Registration Fee "Flat-Rate" "Green " Source: Agrawal, Dill, and Nixon, June 2009.

108 There are marked differences in implementation time and cost, with the high-tech variations on VMT fees and fuel tax surcharges taking longer and costing more. A federal regis- tration fee would be the quickest and least expensive option. The differences in cost translate into differences in revenue efficiency. The low-tech variation on fuel tax surcharges and the federal registration fee yield the greatest net federal rev- enue at the lowest industry cost and are therefore the most efficient. VMT fees would likely show the greatest scale econ- omies, with efficiency rising as the fee rises. The potential of fuel taxes and registration fees to cover multiple modes depends on system design and policy choices. Trucks and inland waterway vessels use public infrastructure and pay both fuel taxes and registration fees. Railroads use private infrastructure and pay neither (except in a few states that assess fuel taxes on railroads), but there are no technical barriers to federal taxes on railroad fuel or fees on equipment. VMT fees, however, would not be practical for rail or water- borne transportation. As with other factors, the degree to which each option is a “user fee” depends on system design. Fuel taxes reflect both vehicle mile and vehicle weight because fuel consumption varies with both. VMT fees reflect vehicle mileage by defini- tion, but would need to vary by vehicle class or weight. Reg- istration fees do not reflect mileage, but can vary by weight, class, axle loading, or other impact variables. The tax and fee rate levels contemplated in this analy- sis would not create strong incentives for change in freight industry operations. The changes in overall cost structure would be minor—far less than the fuel cost variability that Table 74. Revenue mechanism comparison matrix.

109 the industry has recently experienced. Freight and ser- vice truck operators already have stronger cost incentives to minimize mileage and fuel use, so any new incentives would be slight. The transportation system impacts, envi- ronmental impacts, and economic impacts would all be modest. The exception would be a diesel tax surcharge that widened the cost gap between diesel fuel and gasoline, thus encouraging the substitution of gasoline-powered trucks for diesel trucks. Equity among users, vehicle types, and industry segments is likewise a matter of system design more than an intrinsic feature of each option. One serious issue is the definition of “freight vehicles” and the choice of fuels to tax. There is no straightforward, unambiguous way to classify vehicle types or industry segments as “freight,” and medium- and heavy-duty trucks use gasoline as well as diesel. Each option, if imple- mented, would also have to cope with exemptions for off-road and public-sector uses, electric and hybrid vehicles, and inter- national trucking. Combining revenue sources (for example, using registration fees to cover electric vehicles that would not pay fuel taxes) may be a more effective strategy than attempt- ing to adapt one option for all applications. Public acceptance is problematic. The current admin- istration has indicated its opposition to higher fuel taxes, but multiple industry associations have supported fuel tax increases. Planners and policy makers tend to favor VMT fees, while the industry and the public tend to oppose them. Of the three options, VMT fees may face the greatest barriers due to the perception of a privacy problem, even if the privacy issue can be resolved legally and technologically.