A Comparison of High Occupancy Toll Lanes and  High Occupancy Vehicle Lanes in the Sacramento Region  Caroline J. Rodier and Robert A. Johnston  University of California One Shields Avenue Davis, CA 95616 word count: 5914 Rodier, Caroline. (530) 752 - 6303 (phone); (530) 752-3350 (fax); cjrodier@ucdavis.edu Johnston, Robert. (530) 582-0700 (phone); (530) 782-0707 (fax); rajohnston@ucdavis.edu January, 2001 TABLE OF CONTENTS     (Headings are links) ABSTRACT INTRODUCTION LITERATURE REVIEW METHODS      TRAVEL DEMAND MODELING        ECONOMIC WELFARE MODELS 2015 SCENARIOS IN THE SACRAMENTO REGION RESULTS CONCLUSIONS ACKNOWLEDGEMENTS REFERENCES ABSTRACT             As the evidence mounts that HOV lanes may not deliver expected reductions in congestion and emissions, alternatives are being sought.  High occupancy toll lanes are one attractive alternative.  A region-wide system of new HOV lanes in the Sacramento region is compared to a system of HOT lanes for the year 2015.  These policies are also combined with land use intensification measures and auto pricing policies to further enhance their ability to promote carpooling and transit use and reduce congestion and emissions.  The travel effects are simulated with the Sacramento regional travel demand model (SACMET96). The economic benefits for both personal travel and commercial vehicle travel are obtained from economic welfare models developed by us for use with the SACMET96 model.  The scenarios are evaluated against travel, total consumer welfare, and equity criteria.  Emissions effects are inferred from reductions in auto travel.  Key words: high occupancy toll lanes, high occupancy vehicle lanes, travel modeling, economic benefits |TOP| INTRODUCTION To date nearly 1,200 miles of high occupancy vehicle (HOV) lanes have been build nationwide.  Federal and state policies currently promote HOV lane projects. In air quality non-attainment regions, HOV lanes are virtually the only roadway projects approved.  The rationale behind these policies is that HOV lanes foster carpooling and transit use and thus will reduce congestion and emissions.  However, increasingly the evidence has suggested that HOV lanes may not produce expected reductions in congestion and emissions (1,2).  Therefore, alternatives are being sought.  High occupancy toll (HOT) lanes are one attractive alternative.  HOT lanes have been successfully implemented on State Route 91 in Orange County, CA, I-15 in San Diego, CA, and I-10 (Katy) in Houston, TX, and many other regions are actively considering HOT facilities.             A region-wide system of new HOV lanes in the Sacramento region is compared to a system of HOT lanes for the year 2015.  These policies are also combined with land use intensification measures and auto pricing policies to further enhance their ability to promote carpooling and transit use and to reduce congestion and emissions.  The travel effects are simulated with the Sacramento regional travel demand model (SACMET96). The economic benefits for both personal travel and commercial vehicle travel are obtained from economic welfare models developed by us for use with the SACMET96 model.  The scenarios are evaluated against travel, total consumer welfare, and equity criteria.  Emissions effects are inferred from reductions in auto travel. |TOP| LITERATURE REVIEW Recent evidence has challenged the rationale behind the adoption of HOV lane projects, namely that HOV lanes foster carpooling and transit use and thus will reduce congestion and emissions.  Rodier and Johnston (1) simulate an extensive system of HOV lanes in the Sacramento region for the year 2015 and find only a modest reduction in congestion, an increase in emissions, and a loss in consumer welfare when the unobserved private cost of additional auto travel is considered.  Joy Dalgren (2) develops a model to estimate person-delay and emissions for a number of HOV and general purpose lanes alternatives.  She finds that HOV lanes will only be more effective in reducing congestion and emissions than general purpose lanes when there is a high level of congestion and high proportion of HOV in the general purpose lanes.              As the evidence mounts that HOV lanes may not deliver expected reduction in congestion and emissions, HOT lanes are increasingly becoming an appealing alternative.  HOT lanes allow non-carpoolers and some carpoolers to use HOV lanes by paying a toll. HOT lanes have been implemented on State Route 91 in Orange County, CA, I-15 in San Diego, CA, and I-10 (Katy) in Houston, TX.  Since the State Route 91 Express Lanes opened in December 1995, there has been a dramatic reduction in peak period congestion on adjacent non-toll lanes (3). This reduction has been gradually eroding, but congestion has not even come close to pre-express lanes conditions (3).  When the express lanes were initially opened, significant gains in carpooling were found, but since that time gains in carpooling have been found to be insignificant (3).  In San Diego, the HOT lanes on I-15 have been open since December 1996.  There has been considerable demand for use of these HOT lanes (4) as well as an 11% increase in carpooling (5).  An express bus service in the corridor has been launched with the HOT revenues (5).  The I-10 HOT lanes in Houston have only recently been opened, but demand is expected to be significant (4).  Areas throughout the U.S. are considering HOT lanes including Dallas, TX, Sonoma County, CA, Contra Costa County, CA, Alameda County, CA, Maryland, Milwaukee, WI, Portland, OR, Phoenix, AR, Denver, CO, Hampton Roads, VA, Los Angeles, CA, and Minneapolis, MI.             HOV and HOT lanes policies could potentially be made more effective by combining them with land use and travel pricing measures.  These measures have been found to be very effective at reducing congestion and emissions.  A number of studies have found that higher density cities reduce vehicle miles traveled (VMT) per capita (6, 7, 8, 9).  Studies of higher densities near transit or transit oriented developments (TODs) show reductions in auto travel on the order of 4% over 30 years in the Seattle region (8), 8% over 20 years in Portland, Oregon (10), 4% to 7% over 20 years in the Sacramento region (1,11), and 20% over 20 years based on a review of several simulation studies in the U.S. (12, 13).  Other studies show that land use measures effectively reduce auto travel or are made significantly more effective when combined with travel pricing policies and/or improved transit and walk and bike facilities (7,10,14,15).  Reduced emissions generally result from reduced auto travel.             Many studies show travel pricing measures to be effective at reducing auto travel and emissions. Cameron's simulation study of auto pricing in Southern California (16) found that VMT could be reduced by about 12% and pollutants by about 20% with a peak period road congestion charge of $0.15 per 1 mile, employee parking charges of $3 per day, retail and office parking charges of $0.60 per hour, emissions fees averaging $110 per year per vehicle, and deregulated transit services.  Wilson and Shoup's empirical studies of large employer sites (17) show from 20% to 30% reductions in commute trips to the sites when employees pay fully for their parking.              Road pricing has been advocated by economists for decades.  Morrison's review of the literature (18) shows large potential welfare benefit from road charges.  Studies have shown that tolls can benefit all income groups (19,20).  Small's paper (21) develops a spending program for anticipated revenues from the Southern California pricing policies (16).  He demonstrates financial benefits to all consumers when pricing policies are combined with tax rebates and transit improvements. |TOP| METHODS Travel Demand Modeling This study uses the SACMET96 travel demand model (22).  The model was developed with a 1991 travel behavior survey conducted in the Sacramento region.  Some of the key features of this model include (1) model feedback of assigned travel impedances to the trip distribution step; (2) auto ownership and trip generation steps with accessibility variables; (3) a joint destination and mode choice model for work trips; (4) a mode choice model with separate walk and bike modes, walk and drive access modes, and two carpool modes (two and three or more occupants); (5) land use, travel time and monetary costs, and household attribute variables included in the mode choice models; (6) all mode choice equations in logit form; and (7)  a trip assignment step that assigns separate A.M., P.M., and off-peak periods |TOP| Economic Welfare Models Kenneth Small and Harvey Rosen (23) show how a consumer welfare measure known as compensating variation (CV) can be obtained from discrete choice models:                                                                where l is the individual's marginal utility of income, Vm is the individual's indirect utility of all m choices, p0 indicates the initial point (i.e., before the policy change), and pf indicates the final point (i.e, after the policy change).  The change in indirect utility is converted to dollars by the factor, 1/l, or the inverse of the individual's marginal utility of income.  Small and Rosen show how marginal utility of income can be obtained from the coefficient of the cost variable in discrete choice models.             The compensating variation formula (1) from above was adapted to suit the specifications of the SACMET96 mode choice models.  In the home-based work, shop, and other mode choice models, households are segmented into income/worker categories, and person trips are generated for those categories. To obtain compensating variation for each income/worker category h, the following formula was applied for all modes m and for all trips Q between all origins i and all destinations j:       where l is provided by the coefficient of the cost variable in the mode choice equations.  Total compensating variation was obtained by summing the compensating variation obtained from each income/worker group.  Compensating variation was also obtained from the non-home-based mode choice models; however, these models are not stratified by household/income classes. Based on a review of the literature (e.g., 24), it is assumed that total operating costs are $0.40 per mile.             Economic benefits to commercial vehicles travel resulting from transportation policies were obtained from the trip distribution model in SACMET96, which distributes commercial vehicle trips as a function of zone-to-zone travel times.  The following formula was applied: where B is equal to the net benefits to commercial vehicle travel.  These include travel time costs, operation and maintenance (O&M) costs, and revenue benefits.  Travel time is obtained from the model and converted to dollars with the average wage rate of truck drivers in the region ($12 per hour).  VMT for commercial vehicles is also obtained from the model for each scenario.  Total O&M (excluding wages) costs and revenue benefits for the scenarios are obtained by multiplying the average per mile costs for the region ($0.90 for O&M and $0.95 for revenues) by VMT.  Truck wages, operation and maintenance costs, and revenues were developed based on national data and in consultation with the California Trucking Association.  Values used correspond to low estimates for the state of California, which are reasonable for the Sacramento region. |TOP| 2015 SCENARIOS IN THE SACRAMENTO REGION             Base Case.  The base case scenario represents a financially conservative expansion of the Sacramento region’s transportation system and will serve as a point of comparison for the other scenarios examined in this study.  This scenario includes a relatively modest number of road-widening projects, new major roads, one freeway HOV lane segment, and a limited extension of light rail (east to Mather Field Road).             High Occupancy Vehicle Lanes (HOVs).  The HOV lane scenario represents an extensive expansion of the Sacramento region’s HOV lane system to encourage the use of carpools and reduce traffic congestion and emissions. The HOV lane system is expanded east on SR-50 past Folsom near the El Dorado County line, northeast on I-80 to Douglas in Roseville, northwest on I-5 to the Sacramento International Airport, and west on I-80 to Davis.  In this scenario, HOV lanes are increased from 26 lane miles in the base case scenario to 179 lane miles.  Mixed-flow freeway lanes are increased by 6% over the base case scenarios. Express bus service that takes advantage of the HOV lanes is also added to the transit network. HOV lanes are separately coded in the highway network used in SACMET96.             High Occupancy Tolled Lane (HOTs).   In this study, the HOV lanes in the HOV lane scenarios are converted to HOT lanes.  To simulate the use of HOT lanes in SACMET96, the mode choice model for the home-based work trip purpose was expanded to include the HOT mode (using variable coefficients specific to the drive alone mode) and HOT trips were assigned to the separately coded HOT lanes.  A $0.05 per mile toll was charged for using the HOT lanes.  Tolls were charged to single occupant vehicles, and two occupant vehicles but no tolls were charged to vehicles with three or more people.  Where congestion was not eliminated on the HOT lanes with the $0.05 toll, a $0.50 per mile toll was imposed in order to achieve non-congested conditions on those roadway segments.  This toll level was selected after numerous other tolls were tried to obtain the lowest toll level that would eliminate congestion on the HOT links.             Road Oriented Development (RODs).  RODs were developed to examine a road oriented land use intensification scenario that might be similar to a transit oriented land use intensification (or TOD) scenario with respect to potential travel and emission benefits.  RODs, like TODs, are mixed-use centers with a relatively high intensity mix of shopping, jobs, and housing.  RODs are different from TODs in that they would be located at major intersections along HOV or HOT lane corridors as opposed to light rail stations.  The purpose of RODs would be to facilitate carpooling and express bus service.  Less emphasis would be placed on improving the walk and bike environment of the RODs than in the TODs. In this study, RODs have an average density of 15 households per acre, 10 retail employees per acre, and 20 non-retail employees per acre.  These density levels were developed based on a review of current land use densities in Sacramento areas that are considered to be TOD prototypes.  To achieve the ROD densities, growth in households, retail employment, and non-retail employment from 1995 to 2015 in the outer zones were moved to the zones in the RODs.  The ratio of the household classification was held constant in all zones in the input files, and thus only the total number of households were changed in zones. Sixty-six percent of new regional household development and 52% of new employment development were moved from outer zones to RODs for the scenario.  RODs have a one fourth-mile radius.  School enrollment was also adjusted to correspond to the changes in households. Ramps and arterials were expanded as needed to keep congestion levels without the RODs approximately the same as with the RODs to avoid impeding access to HOV and HOT lanes.             Pricing Policies.  In this study, parking cash-out, VMT tax, and peak period pricing policies were examined in combination with the other scenarios described above.   The parking-cash out policy was modeled after the California law (California Health and Safety Code Section 43845) that requires that employers (who rent parking spaces from a third party and have 50 or more employees) offer commuters the option to choose cash in lieu of any parking subsidy offered.  This law applied to regions that do not meet California’s state clean air standards (such as Sacramento).  To simulate the parking cash-out scenario in SACMET96, home-based work trips were charged five dollars for work place parking in 59 zones identified by Sacramento Area Council of Governments staff as likely to be eligible under the policy in the future. The income benefits of the parking cash-out program were simulated in the consumer welfare estimates by returning paid parking charges to the travelers. Many have advocated the imposition of a VMT tax that captures the monetary and non-monetary external costs imposed on society by auto travel.  A $0.05 VMT tax for the Sacramento region was obtained from average national estimates of the external costs of auto use provided by Mark Delucchi’s The Annualized Social Cost of Motor-Vehicle Use in the U.S., 1990-1991 (25).  External costs captured by the SACMET96 models were excluded.  The low estimates of total national external costs for autos were summed and divided by the total VMT for autos to obtain the $0.05 per mile VMT tax.  The per mile cost of auto travel (currently $0.05) included in the mode choice models of SACMET96 was increased by $0.05 to represent the VMT tax.  The peak period pricing policy is intended to reduce congestion during peak travel hours.  To simulate this policy, the per mile cost of auto travel for the home-based work trip purpose was increased by $0.10. |TOP| RESULTS  The effects of the policy scenarios in the Sacramento region are more strongly evident in the daily mode share projections for home-based work trips (see Table 1) because these are the trips largely targeted by the HOT and HOV policies.  The results show that the HOT scenario reduces the drive alone mode share slightly but more than the HOV scenario, the HOV scenario increases shared ride more than the HOT scenario; and the HOT scenario increases transit, walk, and bike mode shares more than the HOV scenarios.  The addition of the RODs to the HOT and HOV policies intensifies this trend.  The addition of the pricing policies to the ROD scenarios continues the intensification of this trend with the exception of the drive alone mode share, where the reduction in the drive alone share is slightly less for the HOT scenario than for the HOV scenario. The mode share trends for total regional trips (see Table 2) are similar to those described above but differences among scenarios are smaller because they include all trip purposes.  Table 1.  Daily Percentage Mode Share for Home Base Work Trips 2015 Policy Scenarios for the Sacramento Region                       Drive Alone   Shared Ride   Transit   Walk & Bike                   Base Case   77.7%   15.1%   2.3%   4.9%                   HOV   77.3%   15.5%   2.5%   4.8%     (-0.5%)a   (2.1%)   (5.0%)   (-2.1%)                   HOV/ROD   76.1%   16.4%   2.3%   5.2%     (-2.1%)   (8.2%)   (-0.2%)   (7.4%)                   Pricing HOV/ROD   73.7%   17.7%   2.7%   5.9%     (-5.1%)   (16.8%)   (13.7%)   (22.3%)                   HOT   77.5%   15.3%   2.5%   4.7%     (-0.7%)   (0.5%)   (6.3%)   (-3.0%)                   HOT/ROD   76.2%   16.2%   2.4%   5.2%     (-2.2%)   (6.7%)   (0.4%)   (7.0%)                   Pricing HOT/ROD   74.2%   17.4%   2.6%   5.8%     (-4.8%)   (14.8%)   (10.4%)   (18.8%)     a Figures in parentheses are percentage change from the base case scenario.      Table 2.  Daily Mode Share Projections   2015 Policy Scenarios for the Sacramento Region                           Drive Alone   Shared Ride   Transit   Walk & Bike                       Base Case   50.0%   41.8%   0.7%   7.5%                       HOV   50.0%   41.9%   0.8%   7.4%       (-0.0%)a   (0.2%)   (3.4%)   (-1.5%)                       HOV/ROD   49.8%   41.9%   0.7%   7.5%       (-0.4%)   (0.4%)   (2.2%)   (0.4%)                       Pricing HOV/ROD   49.4%   42.1%   0.8%   7.7%       (-1.3%)   (0.8%)   (12.2%)   (3.1%)                       HOT   50.0%   41.8%   0.8%   7.4%       (-0.0%)   (0.2%)   (4.2%)   (-1.5%)                       HOT/ROD   49.8%   41.9%   0.7%   7.5%       (-0.4%)   (0.3%)   (2.5%)   (0.4%)                       Pricing HOT/ROD   49.4%   42.0%   0.8%   7.7%       (-1.2%)   (0.7%)   (10.5%)   (2.7%)                       a Figures in parentheses are percentage change from the base case scenario.              Daily vehicle travel projections are presented in Table 3.   The HOV and HOT scenarios both result in small increases in vehicle trips (0.2% and 0.3%, respectively) and VMT (1.9% and 2.2%, respectively) and reductions in VHD (5.2% and 7.7%, respectively).  Compared to the HOV scenario, the HOT scenario provides larger reductions in VHD and greater increases in vehicle trips and VMT.  The addition of RODs to the HOV and HOT scenarios produces an increase in vehicle trips (1.6% and 1.7%, respectively) because of the increased auto accessibility in the RODs but reduced VMT (1.7% and 1.8%, respectively).  Thus, it appears that the ROD scenarios result in more but shorter vehicle trips.  The addition of the RODs to the HOV and HOT scenarios significantly improves the reduction in VHD (11.7% and 15.7%, respectively).  The HOT/ROD scenario provides greater increases in vehicle trips, smaller reductions in VMT, and greater reductions of VHD than the HOV/ROD scenario.  The addition of the pricing policies to the HOV/ROD and HOT/ROD scenarios tends to dampen the increase in vehicle trips (1.0% and 1.1%, respectively), increase the reduction in VMT (4.4% and 4.1%, respectively), and increased the reduction of VHD (21.9% and 23.2%, respectively).   Table 3.  Daily Vehicle Travel Projections 2015 Policy Scenarios for the Sacramento Region                       Vehicle Miles   Hours of Travel     Trips (millions)   Traveled (millions)   Delay (thousands)a               Base Case   6.8   62.2   243.3               HOV   6.8   63.3   230.7     (0.2%)   (1.9%)   (-5.2%)               HOV/ROD   6.9   60.9   214.8     (1.6%)   (-1.9%)   (-11.7%)               Pricing HOV/ROD   6.9   59.4   189.9     (1.0%)   (-4.4%)   (-21.9%)               HOT   6.8   63.5   224.5     (0.3%)   (2.2%)   (-7.7%)               HOT/ROD   6.9   61.0   205.2     (1.7%)   (-1.8%)   (-15.7%)               Pricing HOT/ROD   6.9   59.6   186.9     (1.1%)   (-4.1%)   (-23.2%)               a  Vehicle hours of delay are vehicle hours traveled under congested speeds minus vehicle hours of travel under free flow speeds on the same  facility. b Figures in parentheses are percentage change from the base case scenario.                Past experience with emissions modeling using the California Department of Transportation emissions model and the California Air Resources Board’s EMFAC7F emission factors with the SACMET model indicate that scenarios rank with VMT for emissions reduction even with small increases in vehicle trips. Thus, both the HOV and HOT scenarios should result in increased emissions over the base case scenario, and the HOT scenario should produce greater increases than the HOV scenario.              The addition of the RODs to the HOV and HOT policies should reverse the increase in emissions.   However, the reduction in VMT and emissions would be relatively small considering the large percentage of new growth that would need to be shifted to achieve these results--not to mention the political will needed to implement such a controversial policy.  In another study, we evaluate TODs with similar regionwide shifts in new development and light rail and feeder bus service along similar corridors (26).  The TOD policy resulted in a 7% reduction in VMT and a 5% reduction in total organic gases at the regional level (26).  These results suggest that land use intensification policies may be more effective in reducing VMT and emissions when they are combined with transit rather than HOV or HOT lane policies.                The addition of pricing policies to the HOV/ROD and HOT/ROD scenarios should increase the reduction of emissions.  Pricing policies are also considered to be quite controversial.  The TOD policy with similar pricing policies resulted in a 9% reduction in VMT and a 7% reduction in total organic gases at the regional level (26).  Thus, the combination of pricing and land use intensification measures may be more effective at reducing VMT and emissions when combined with transit rather than HOV and HOT facilities.               The economic benefits (1995 present value) are presented in Table 4.  With respect to economic benefits, the HOT scenario is clearly superior to the HOV scenario.  The HOV scenario results in a small economic loss ($0.003 per trip) when the full unobserved cost of additional travel is included in the analysis (which is assumed to be $0.40 per mile).  The HOT scenario results in an economic benefit of $0.026 per trip because of travel time savings to travelers with high values of time, which more than offset the unobserved cost of additional travel.  The approach to refunding revenues in the HOT scenarios was conservative.  It is assumed that a private company runs the HOT lanes and keeps the profit after paying for the capital and O&M costs of the scenario.  Benefits would be higher if it was assumed that the HOT facilities were build by the public and revenues were used to fund additional transportation improvements in the corridor (e.g., new express transit service).  This approach would be similar to the I-15 HOT lanes in San Diego.   Table 4.  1995 Present Value of Economic Benefits 2015 Policy Scenarios for the Sacramento Region                   Total per Daya   Total per Tripa       (1)   (2)             HOV     -$250,370   -$0.003             HOV/ROD     $3,805,380   $0.041             Pricing HOV/ROD     $8,172,840   $0.087             HOT     $2,465,840   $0.026             HOT/ROD     $7,282,660   $0.078             Pricing HOT/ROD     $13,281,400   $0.142             a Includes Capital and O&M Costs.                The addition of the RODs and then the pricing policies to the HOV and HOT scenarios both increase the economic benefits of the scenarios.  This is because of the travel time savings that result from the ROD and the pricing policies.  It is also assumed that revenues from the pricing policies are returned to the public through, for example, lower sales taxes.  Overall, the scenarios that combine RODs and pricing policies with HOT lanes have higher benefits (approximately double) than those combined with HOV lanes.  The economic benefits for the HOT scenarios are also higher than those obtained from the TOD scenarios, but the benefits for the HOV scenarios are lower than for the TOD scenarios (26).              The results of the equity analysis are presented in Table 5.  Generally, for the HOV scenarios there is no significant difference among the benefits and losses for the different income classes.  The exception to this is the higher benefits obtained for the highest income class for the pricing HOV/ROD scenario.  This is because of the time savings to travelers with a higher value of time.  For the HOT scenarios, benefits tend to increase with level of income.  Like the pricing HOV/ROD scenario, this is because higher income classes value travel time savings more than lower income classes.  The HOT only scenario results in a loss to the lowest income class.  Losses to the lowest income class are eliminated when HOT, HOV, and pricing policies are combined with RODs because RODs provide greater access to carpooling and express transit service to the lower income class.    Table 5.  1995 Present Value of Economic Benefitsa by Income Class 2015 Policy Scenarios for the Sacramento Region                   Income Class One   Income Class Two   Income Class Three     ($0 to $10,000)   ($10,001 to $35,000)   ($35,001 and above)               HOV   -$0.001   -$0.005   $0.000               HOV/ROD   -$0.004   -$0.005   $0.066               Pricing HOV/ROD   $0.050   $0.043   $0.100               HOT   -$0.002   $0.002   $0.071               HOT/ROD   $0.004   $0.011   $0.149               Pricing HOT/ROD   $0.129   $0.108   $0.200               a Includes Capital and O&M Costs.   CONCLUSIONS HOV lanes and, to a lesser extent, HOT lanes are considered politically feasible policies to address the problems of congestion and emissions associated with regional transportation systems.  The results of this study indicate that HOT lane policies may be significantly better than HOV lane policies at reducing congestion.  However, both the HOV and HOT scenarios may increase VMT and emissions compared to a no-build scenario, and the increase in emissions may be greater for the HOT scenario compared to the HOV scenario.              The potential added benefits of combining land use intensification policies (RODs) and pricing policies to the HOT and HOV scenarios with respect to congestion and emissions reductions were also examined.  The RODs and pricing policies were both found to reduce congestion but more so in the HOT scenarios than in the HOV scenarios.  The RODs and pricing policies were also found to decrease VMT and emissions; however, the auto travel reduction and emissions benefits for the HOT/ROD and HOV/ROD scenarios were small, especially when compared to TOD scenarios.  These results suggest that land use intensification policies may be more effective at reducing VMT and emissions when they are combined with transit rather than HOV and HOT lanes.             With respect to the economic benefits of the scenarios, the HOT lane policies are clearly superior to the HOV lane policies.  The HOV lane scenario resulted in small losses when the analysis included the full unobserved cost of additional auto travel.  However, the HOT lane scenario resulted in economic benefits over and above the full unobserved cost of additional auto travel because of reduced congestion for and greater travel time savings to travelers with higher values of time.  Losses for the HOV lanes are offset and gains for the HOT lanes are increased when the policies were combined with RODs and pricing policies.  These increased economic benefits resulted from greater accessibility to carpooling and express transit service in the RODs and reduced congestion resulting from the pricing policies.  The HOT/ROD scenario also provided greater benefits than a comparable TOD scenario.                  The equity analysis generally indicated that there were no significant differences in losses and benefits among income classes for the HOV scenarios; however, the opposite was true for the HOT scenarios where benefits increased with income level.  Losses to the lowest income class are eliminated when HOT, HOV, and pricing policies are combined with RODs because RODs provide greater access to carpooling and express transit service to the this income class.  This suggests that if the revenues from HOT lanes are used to compensate lower income classes monetarily or through transportation improvements (e.g., more transit service), then losses to the lowest income classes may be avoided. ACKNOWLEDGEMENTS We would like to thank the University of California Transportation Center and the Environmental Protection Agency for funding this research.  The views expressed and any errors are those of the authors.       REFERENCES 1.         Rodier, Caroline J. and Robert A. Johnston. Travel, Emissions, and Welfare Effects of Travel Demand Management Measures. Transp. Res. Record 1598, pp. 18-24.1997. 2.         Dalgren, J. In what situations do high occupancy vehicle lanes perform better than general purpose lanes? Transportation Research Board. Preprint. 75th Annual Meeting, Washington, D.C. 1996. 3.                  ARDFA. Interim Results for SR-91.  Cal Poly, San Luis Obispo, CA June 25, 1997. 4.                  ITE Task Force.  High-Occupancy/Toll (HOT) Lanes and Value Pricing: A Preliminary Assessment.  ITE Journal.  June 1998. 5.                  Hamberg, L. HOT Idea for Easing Bay Area Traffic.  San Francisco Chronicle. June 8, 1998. 6.         Newman, P.W.G., and J.R. Kenworthy.  Cities and automobile dependence: A Sourcebook. Gower Technical, Brookfield, MA, 1989. 7.         Webster, F.V., P.H. Bly, and N.J. Paulley.  Urban Land-Use and Transport Interaction:  Policies and Models. Avebury, Brookfield, MA, 1988. 8.         Watterson, W.  Linked Simulation of Land Use and Transportation Systems:  Developments and Experience in the Puget Sound Region.  Paper for the Conference on Transportation and Global Climate Change:  Long-Run Options, Transportation Research Board, Asilomar, CA, August 1991. 9.         Natural Resources Defense Council and the Sierra Club.  Explaining Urban Density and Transit Impacts on Auto Use.  Docket No. 89-CR-90, California Energy Commission, Sacramento, CA, January 1991. 10.              Cambridge Systematics, Inc. (1996).  Analysis of Alternatives. Making the Land Use Transportation Air Quality Connections.  Portland, Oregon. May 1996. 11.       Johnston, Robert A. and Raju Ceerla. 1995. Land Use and Transportation Alternatives, in Transportation and Energy: Strategies for a Sustainable Transportation System, ed. by D. Sperling and S. Shaheen. ACEEE, Wash., D.C. 12.       Keyes, D.L.  Energy and Land Use:  An Instrument of US Conservation Policy? Energy Policy. Vol. 4, No.3, September 1976, pp. 225-236.    13.       Sewell, W.R.D. and H.D. Foster.  Analysis of the United States Experience in Modifying Land Use to Conserve Energy.  Working Paper No. 2, Lands Directorate, Canada, 1980. 14.              Replogle, M.  Computer Transportation Models for Land Use Regulations and Master Planning in Montgomery County, Maryland.  In Transportation Research Record 1262, TRB, National Research Council, Washington, D.C., 1990, pp. 91-100. 15.       Urban Travel and Sustainable Development. OECD, Paris. 1995.   16.       Cameron, M.  Transportation efficiency: Tackling Southern California's Air Pollution and Congestion. Environmental Defense Fund, Los Angeles, March 1991. 17.       Wilson, R.W., and Shoup, D.C.  Employer-Paid Parking:  The Problem and Proposed Solutions.  Paper for the Association for Commuter Transportation Conference, Seattle, WA, December 1990. 18.       Morrison, Steven A.   A Survey of Road Pricing.  Transportation Research:  A.  Vol. 20A, No. 2, 1986, pp. 87-97. 19.       Small, Kenneth A.   The Incidence of Congestion Tolls on Urban Highways. Journal of Urban Economics, Vol. 13, January 1983, pp. 90-111. 20.       Small, K.A., C. Winston, and C. Evans.  Road work : A new highway pricing and investment policy. The Brookings Institution, Washington, D.C., 1989. 21.       Small, Kenneth A.   Using the Revenues from Congestion Pricing:  A Southern California Case Study.  Paper for the Reason Foundation.  Dept. of Economics, University of California, Irvine, August 17, 1992. 22.              DKS & Associates. Model Development and User Reference Report. Sacramento, CA., October 1994. 23.       Small, K.A., and H.S. Rosen.  Applied Welfare Economics with Discrete Choice Models.  Econometrica, Vol. 49, No. 3, January 1981, pp. 105-130. 24.       Small, K.A.  Urban Transportation Economics. Harwood Academic Press, 1992. 25.       Mark Delucchi’s The Annualized Social Cost of Motor-Vehicle Use in the U.S., 1990-1991, 1996. 26.       Johnston, R.A., C. J. Rodier, and M. G. Choy. 1997. Transportation, Land Use, and Air Quality Modeling.  Proceedings from the 1997 ASCE Meeting on the Transportation, Land Use, and Air Quality Connection. Portland, OR.