High Occupancy Vehicle (HOV) Lanes first principles assessment

In theory, HOV priority lanes can be an efficient way of using road capacity as they discourage inefficient use of the road space by single occupant cars by giving priority to public transport, two-wheel vehicles and car sharing. If every single occupancy driver can be persuaded to car-share, then the people-moving capacity of a given road will be doubled, and additional benefits will accrue from any that switch to van-pools or existing public transport services.

They can also represent a more-equitable use of road space, since they allocate more of the available roadspace to HOVs and buses which impose less congestion per-person-trip on other road users than Single Occupancy Vehicles (SOVs). (VTPI, 2004).
However some argue that they can be less efficient than additional road capacity for all users (Orski, 2001; Dahlgren, 1998; Johnston, 1996). Wellander and Leotta (2001) report that effectiveness and benefits of HOV lanes depend on the criteria and assumptions used to evaluate them. This would also change according to type of HOV facility (e.g. additional or converted lanes). Some oppose additional lanes on the basis that they increase total road capacity and encourage longer-distance commuting and in some cases people believe that HOV lanes can have negative impacts (Leman et al 1994; Dalhgren, 1998). Moreover, HOV lanes can often take years to reach their full potential, since they affect long-term decisions such as where consumers live or choose to work.

There are various studies related to the evaluation of HOV lanes which are mainly based on US experience of existing HOV lanes, and on modeling studies. Status of HOV lane developments in Europe is discussed in ICARO (1999).

Conclusions and results vary from one study to other. There are numerous reports and papers available at the TRB HOV committee website at http://www.hovworld.com. Arguments on HOV lanes can be divided into three groups. However it should be noted that the arguments also vary whether the HOV lanes are “converted” or “added” lanes.

Those that find that HOV lanes are effective
Level of reduction in vehicle trips on HOV corridors is reported to vary between 4 to 30% (e.g. Comsis, 1993; Pratt, 2000; Ewing, 1993). Apogee (1994) estimated that HOV lanes can reduce up to 1.4% of vehicle miles travelled (VMT) and up to 0.6% of vehicle trips in a region. For example, in Texas, HOV lanes carry as much as 40 percent of the people moved on the freeway. Surveys have shown that willingness to car share and bus use increases after the opening of an HOV lane (Stockton et al, 1999). Similar results are also found in the UK first HOV lane in Leeds (LCC, 2002) (see Evidence on Performance section).

While some theoretical research has suggested that maximum delays of 30 minutes are necessary to justify an HOV lane over a general purpose lane (Dalgren, 2002) other monitoring results show that as little at 10 minutes delay can result in very successful HOV lane operation (Stockton, 1999). For example, currently the Shirley Highway (Virginia)’s 28-mile reversible HOV lanes carry an average of 10,400 person trips and 2,800 vehicles in the morning peak hour. These lanes provide an average travel time savings of 31 and 36 minutes for the morning and evening peak travel periods.
Those that find HOV lanes to be less effective than general purpose lanes
Orski (2001) argued that HOV lanes are less effective than additional general purpose lanes. By using a simulation model based on the Sacramento Area Transportation Study, Johnston et al (1996) also argue that benefits of HOV are generally temporary and that the higher speeds soon induce longer non-work trips, time shifting to peak periods, mode shifts from bus to HOV and abstraction of non-car-available travellers from PT to join SOV drivers. The study results also show higher VMT with the new HOV lanes.
Similarly, based on the model results for an hypothetical corridor study, McDonald et al (2000) concluded that “construction of new mixed flow lanes or conversion of existing HOV lanes can lead to increases in VMT that are likely to have negative environmental impacts. This occurs due to modal shifts and rescheduling effects without consideration of possible inducement of new trips”. This study also argues that “HOT lanes offer the possibility of larger reductions in vehicle and person hours of travel time, because they preserve incentives for higher vehicle occupancy and allow more efficient use of lane capacity”.
Dahlgren (1998) argues that reduced delays for both HOV and LOV can induce additional traffic growth, shifts from other routes and times. He summarises the possible effects of constructing a HOV lane as illustrated in the figure below

There are those who suggest that benefits of HOV lanes can be increased by turning them into HOT (High Occupancy Toll) lanes. The argument is summarised in the following paragraph.
Despite the widespread implementation of HOV lanes in other major US cities and inter-urban corridors, ridesharing amongst commuters has been declining – from 20% in the 1970s to 13% in the 90s – and under-utilisation has led to criticism of HOV lanes. This has led to increasing interest in High Occupancy Toll (HOT) lanes and studies studies have increasingly demonstrated that HOT lanes with toll differentiation provide a cost-effective way to reduce traffic congestion (e.g. Turnbull et al., 1991; O'Sullivan, 1993; Yang et al, 1999). The HOT lanes scheme in San Diego in the USA costs sole drivers up to $4, depending on distance, and HOVs with 2+ occupants can travel free. In Houston, Texas, the HOT scheme allows HOVs with 2+ occupants to use the HOV 3+ lane for a fixed fee.

The table below illustrates ways providing an extra HOV lane might be expected to affect the number of car kilometres travelled.

Responses and situations

Response	Reduction in road traffic	Expected in situations
		Because of reduced delays in peak hours, some trips might be shifted to those hours
	/	The extra lane for HOVs may allow these vehicles to switch to a more-direct HOV route from a more- indirect route chosen previously to avoid the congestion Conversely, the extra capacity may encourage HOV drivers to choose a longer (but faster) HOV-lane route in preference to a more-direct but more-congested alternative
	/	Increased accessibility of locations (e.g. work, education) might be positive or negative effects in relation to planning policies.
		The impact of an HOV on accessibility is unlikely to be large enough to significantly change overall trip frequency
		The bus-lane component of the HOV will help encourage mode-shift from single occupant cars to buses.
	/	Reduced car ownership through car sharing Induced car ownership due to reduction in delays from the extra road capacity
	/	HOV lane corridors can become attractive to commuters due to time savings. This is likely to increase car kilometres due to the impact of the additional road space.

	= Weakest possible response,		= strongest possible positive response
	= Weakest possible negative response,		= strongest possible negative response

= No response

The table below illustrates ways in which excluding Single Occupancy Vehicles (SOVs) from a lane on a moderately-congested link might be expected to affect the number of car kilometres>travelled.

For the average individual, the general impact of providing an extra HOV lane might be as the following:

Financing requirements
Costs of HOV lanes include project construction, management and enforcement. These costs depending on the type of HOV lanes, e.g. converted or added. While converting an existing general purpose lane into a HOV lane is relatively cheap, constructing a new lane can be expensive, especially if physical barriers are included in the HOV lane scheme.
Accompanying measures, such as marketing, consultation etc. can also be costly. Enforcing the measure can also involve substantial costs. Usually, there are no sources of income associated with the HOV lanes. A possible source of income could be the introduction of a combined HOV lane / pay lane (HOT lane), where solo drivers are also accepted upon payment of a toll, which can be varied according to the time of day and the level of congestion.
HOV can be a relatively low-cost measure; the Trondheim scheme, for example, cost 750,000NOK (£60,000) for signing, markings, information and marketing. In comparison, many of the American states have spent close to $1m (£550,000) in information campaigns alone.

Expected impact on key policy objectives
Expected benefits of HOV lanes include increased travel speeds and reliability for HOV passengers (including those using public transport), increases in car occupancy and person throughput. It makes the alternatives to driving alone more attractive. They are also perceived to provide significant benefits to local authorities including road and parking facility cost savings, public transport operating cost savings, congestion and pollution reductions, and consumer benefits. (VTPI, 2004.)

However some argue that HOV lanes (as extra road capacity) encourage urban sprawl and contribute to poor air quality (Leman, Schiller and Pauly, 1994) and they are an inefficient use of road capacity (Orski, 2001; Johnston, 1996).

The impact of HOV lanes tends to be measured in terms of improvements in car occupancy and person throughput, and there are few direct measurements of congestion. In many cases, quoted journey times and hence delays to carsharers have improved, but in some case this is at the expense of the mixed use lane.
The attractiveness of the lanes largely depends on journey time differentials, but this is constrained in an urban setting by buses stopping in the HOV lane, junctions and access by frontagers, etc. Another key variable is people’s propensity to rideshare. Research in the US has shown that this is not linked to socio-economic characteristics, but that trip lengths and the ease of finding a partner are important. The latter is likely to become more difficult as more employees work flexibly.

The contribution of the provision of an extra HOV lane to policy objectives are summarised below.

Contribution of an Extra HOV Lane to Policy objectives

The policy contribution of excluding SOVs from an existing moderately-congested link are summarised below.

Expected Impact on Problems
The impacts of HOV lanes will be largely determined by the relative proportion of SOVs to HOVs in the traffic, the speed-differential between general and HOV lanes, the nature of the traffic using the route (in particular whether origins and destinations are sufficiently limited to make car-sharing a realistic alternative for a significant proportion of the SOV drivers). As usual, the pattern differs between providing an extra HOV lane and restricting the use of existing lanes. The table below relates to the provision of an extra HOV lane.

Expected winners and losers
Some argue that HOV priority is unfair as it favours one group (HOV users) over other road users (Orski, 2001). Others consider HOV priority a fairer allocation of road space by giving travellers who use less space, and therefore contribute less to traffic congestion, priority over those who use more space. Some critics argue that HOV lanes do not meet the needs of people who cannot use public transport (VTPI, 2004).
HOV lanes will benefit car sharers and public transport users, which includes a high proportion of low income and disadvantaged people (Pratt, 2000).

Barriers to implementation
HOV lanes can be controversial and provoke substantial resistance, so not all politicians will be keen on this instrument especially in the case of a ‘converted’ lanes (ICARO, 1999). HOV facilities in the US have closed after few months of operation because of public opposition (Fuhs and Obenberger , 2001). An example of this phenomenon recently occurred in New Jersey where the State removed HOV lanes on I-287 and I-80. (McDonald, 2000). Arguments include that “they are unfair and ineffective, in preference to general-purpose lanes” or that “they increase total road capacity, leading to increased total vehicle traffic and urban sprawl” (VTPI, 2004).
Therefore public and political support is an essential factor for successful application of HOV facilities. Marketing is essential to create awareness and acceptance in reducing traffic congestion and pollution and to promote car sharing (ICORO,1999).

Appropriate contexts
It was suggested that HOV lanes are most effective at reducing car use in major urban areas with large employment centres, heavy congestion and complementary policies where public transport provides time savings of at least 5 to 10 minutes per trip (Turnbull, 2001; Pratt,1999). Similar conclusions were also supported in ICARO (2000).

Enforcement
Enforcement is crucial for gaining public support for HOV, but providing manual enforcement is expensive and takes the police away from their core work, while automated systems are still unproven. In Leeds, the cost of continuous police enforcement was a factor in the decision to restrict the lane to peak period operation.