Public transport services

A first principles assessment

Why introduce changes to service levels?

There are a number of broad, and sometimes conflicting, aims which may influence considerations about public transport service levels and patterns.  These include:

1. To provide a service responding to travel needs of people who do not currently use public transport. There can be several underlying reasons for this, for instance:

1. to increase passenger base to secure viability of the public transport services;
2. to reduce congestion by persuading people to switch from private vehicle to public transport;
3. to support employment by improving connectivity between residential and employment areas (see Marsden et al. 2013);

2. To improve the quality of service for existing passengers - to help maintain the existing passenger base and, in some cases, generate additional trips.
3. To be cost effective - many private bus operators attempt to tailor the service levels they offer to meet demand, e.g. lower in the peak period.
4. To contribute to social inclusion, for instance by providing services meeting needs of people without private vehicles and for whom walking and cycling are not viable.

Demand impacts

When service levels change they influence the level of demand for public transport. In general, all other things being equal, an increase in service levels will increase patronage, whilst a decrease in service levels will reduce patronage. The size and direction of the change in demand following a change in service levels can be expressed in terms of a service elasticity and is defined as,

For example, if the service elasticity of bus demand with respect to service frequency is 0.4, and all service frequencies were to increase by 10% we would expect patronage to increase by 4%. The service elasticity is therefore a measure of the sensitivity of bus passengers to service levels.

The absolute size of the service elasticity conveys information on the sensitivity of demand to changes in the factor affecting demand and its sign conveys information on the direction of the change. Service elasticities are defined as inelastic if they are less than 1.0 and elastic if they are greater than 1.0.

A wide range of factors influence the size of service elasticity and these are listed below:

• Service levels – the lower the current level of service the more sensitive passengers will be to service level changes.
• Size of service level change – the larger the change in the service level the more sensitive passengers will be to the service level change.
• Income levels – those on low incomes are less likely to be sensitive to changes in bus service levels and more sensitive to changes in fares.
• Competition from other modes – strong competition from other public transport operators will make passengers more sensitive to service level changes.
• Demographic factors – The elderly and school children are more sensitive to changes in service levels.
• Journey purpose – travellers commuting to work tend to be less sensitive to service level changes, whilst leisure travellers are more sensitive.
• Urban vs Rural – passengers in rural areas tend to be more sensitive to service changes than passengers in urban areas.
• Area - passengers tend to be less sensitive to service level changes in large cities compared with smaller conurbations.

Whilst these factors can be discussed in isolation it is likely that more than one of them will exert an influence at the same time. In general terms TRL (2004) reports a short run service frequency elasticity with respect to bus of around 0.4 and a long run elasticity of 0.7, with slightly higher figures for rail. These figures relate to mainly UK based evidence whereas the TRB report (2003) is based mainly on US evidence. TRB (2003) reports general service elasticities of around 0.5 in the short run, with most recent studies tending to group their observations around 0.3 and 1.0. The former figure tends to represent studies that are based on central city urban locations and the latter on suburban systems that have undergone well thought out expansion in a growing economy where public transport is well regarded. The expected responses and situations to changes in service levels are outlined below.

The extent of modal switch between bus and car will be dependent upon the service cross elasticity between modes. Cross elasticities measure the change in demand for one mode as a result of the change in one of the factors associated with another transport mode (mainly fare/cost or service frequency). The size of the cross elasticity will therefore depend upon how demand for car will alter (and therefore how demand for bus will alter) due to a change in bus service frequencies.

As time spent travelling within the public transport vehicle is only a proportion of the total journey experience faced by the passenger, it is necessary to individually analyse the various component elements of the entire journey made by public transport. Kelly (1996) attempts to do so as follows:

• An enquiry: Finding out the route of the service and whether it actually serves the desired destination of the user
• A walk: Walking to the stop/station
• A wait: Waiting for the vehicle
• A ride: The actual travel inside the vehicle
• A walk: Walking to the destination

Various instruments can be employed to improve the quality of public transport. It is not intended to list all instruments here but to give a flavour of how the quality of the entire package of public transport can be improved using this framework of analysis as shown in Table 1.

Table 1: Examples of Instruments that can be used to improve the journey experience

The above is a simple framework, with cases where an interchange is involved, this will increase the need for the “walk”, “wait” and “ride” elements.

Kittelson & Associates et al (2003) provides some of these questions under several broad categories as follows:

Availability

• Spatial availability: Where is service provided, and can one get to it?
• Temporal availability: When is service provided?
• Information availability: How does one use the service?

Capacity availability: Is passenger space available for the desired trip?

Comfort and Convenience

• How long is the walk? Can one walk safely along and across the streets to and from transit stops? Is there a functional and continuous accessible path to the stop, and is the stop accessible to people with disabilities?
• Is the service reliable?
• How long is the wait? Is shelter available at the stop while waiting?
• How comfortable is the trip? Will one have to stand? Are there an adequate number of spaces?
• How much will the trip cost?
• How many transfers are required?
• How long will the trip take in total? How long relative to other modes?
• Are the vehicles and facilities clean?

Service Delivery

• Reliability: how often is the service provided as promised?
• Customer service: what is the quality of direct contacts between passengers and staff and more importantly, the customers’ overall perception of service quality?
• Comfort: what is the passengers’ physical comfort level as they wait for and use transit service?

Safety and Security

This relates to the likelihood or more importantly perceived likelihood that one will be involved in an accident (safety) or become the victim of a crime (security) while using public transport. An example question in this category:

• Are there security concerns—walking, waiting, or in the vehicle?

Maintenance

Are the vehicles and facilities clean?

Other Types of Service Level Change

There are a number of other service level changes which have been identified by TRB (2003) and TRL (2004). These include:

1. Regular Interval & Clockface Timetables – The former involves the implementation of a timetable that schedules services to arrive at a station or bus stop, at regular intervals, e.g. every 20 mins. The latter schedules services to arrive at the station or bus stop at the same time past the hour, e.g. 10 minutes past and 30 minutes past the hour. The reasoning behind these service patterns is that they should be easy to remember and will, in some cases, help minimise transfer times (Shires et al, 2003). They are very prevalent in Switzerland, the Netherlands, Austria and Germany.
2. Combined Service Frequencies – This approach involves mixing stopping and express services on the same route to cater for different types of travel needs.
3. Reliability Improvements – If a service is experiencing reliability problems, the operator may change the service schedule to improve reliability. This may well involve building delay into the timetable.

Mode choice and quality in public transport services

When we consider public transport, except for particular specialised services and first class services on rail networks, passengers do not generally have the option of purchasing an additional quality element. The practice thus far in public transportation planning and management is that conventional planning tends to overlook service quality impacts (Litman, 2008). In particular quality factors are generally not analysed to any large extent in considering public transport costs even if some of them have been valued in monetary terms using techniques such as stated preference modelling to elicit choices of a sample of the population under hypothetical scenarios (Balcombe et al 2004).

Enhancing the quality of public transport broadly targets two groups of public transport users:

1. “Captive Users”: Those who do not have a car available for the journey desired or are not able to drive
2. “Choice Users”: Those who have a car available or are able to drive

These terms represent polar ends and it should be emphasised that the “captive market” should not be thought of as a market that will always use public transport. Hence authorities and operators should not be complacent that they will always be there.  This market is not static. It will more often than not shrink with changing socio-economic demographics, rising income levels etc. Experience indicates that these “captive” customers will in due course find alternatives. These “captive customers” are not to be thought as the equivalent of loyal customers. That loyalty can be gained through delivery of a service that meets the user’s satisfaction.

With regards to choice, public transport provision has to be competitive against the private car in meeting travelling needs.  As an example given in Litman (2008), drivers have the option and are often willing to pay considerable sums to include a variety of “optional extras” such as satellite navigation systems and leather interiors in the car.  Whilst the basic need of the car owner is a means of provision of transportation from point A to point B, the quality elements and more importantly the image associated with ownership and use of this carare clearly an element of importance in the decision making process of car ownership and purchase. If more car users are to be persuaded to choose the option of public transport instead, the entire package of public transport options available has to be improved considerably. When approaching this group, the objective of quality is designed to entice users to change their choice from private vehicle by ensuring that public transport caters both to their basic transportation needs and also to their implied needs.

For them, public transport needs to deliver at least a journey experience rivalling, or better than, the private car. Hence measures improving the image of public transport, enhancing safety and personal security, reducing overcrowding are targeted at this segment of the market.

Expected response to increased service levels

The table below shows the expected response to increased service levels, the responses can be assumed to be opposite if service levels are reduced.

Response	Reduction in road traffic	Expected in situations
		An increase in service levels is likely to reduce peoples’ schedule delay and excess wait time and so departure time. This will represent a better quality service as the reliability element of quality will be improved.
		Unlikely to affect people's routes.
		Unlikely to affect people’s choice of destination except where improvements in service level make a given destination more attractive.
		An increase in service levels is likely to generate more bus trips from existing users and new users.
		An increase in service levels is likely to make bus a more attractive mode of travel and so attract car users.
		May affect decision on car ownership.
		But probably most likely to move house for other reasons such as job change.

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

= No response

Short and long run demand responses

If service levels are reduced then responses can be assumed to be diametrically opposite to those presented here.

	-	1st year	2-4 years	5 years	10+ years
	-
	-
	-
	Total number of journeys likely to increase
	Switch to Public Transport
	Possible factor influencing car purchase decision
	-

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

= No response

Supply impacts

Increased service frequency increases capacity of public transport, and in addition to having potential for increasing demand, may reduce overcrowding on popular services. Extended network coverage could increase capacity if it is not done at the expense of frequency of services. The feasibility of increased capacity relies on availability of stock (carriages or buses).  

Financing requirements

The financing of an increase in service levels will tend to come from two main sources, the fare box and support from local/transit authorities. The make up of the funding will largely be decided by who makes the decision to increase service levels and the rationale for that increase. If the decision is made by the transport operator on purely commercial grounds then the operator will expect fare box revenue to cover the financing of the service increase. If a decision is taken by a local/transit authority to increase service levels on social grounds (e.g. to reduce social exclusion) then the costs of providing those services are likely to be met by a combination of fare box revenue and operating subsidies from local/transit authorities. These costs are also likely to vary according to the degree and/or type of regulation that is in place.

Altering service levels will have a direct effect on the fixed, the semi-variable and the variable costs experienced by transport operators. These are outlined in Whelan et al (2001) below.

1. Variable costs are costs that vary directly and immediately with output. For example, fuel costs vary directly with vehicle kms operated, crew costs may vary directly with vehicle hours etc.
2. Semi-variable costs are costs that only vary partially with output. For example, vehicle maintenance is partly related to the extent that vehicles are utilised but there is some element of maintenance that will need to be undertaken irrespective of how intensely the vehicle is utilised. Similar arguments hold true for vehicle depreciation.
3. Fixed costs are costs that do not vary immediately with output. That is, they cannot be varied in the short run. These costs include buildings and general administration.

Transport services are not highly divisible and so neither are the costs associated with them. For example, if a bus operator wished to increase a bus service from 4 buses per hour to 5 buses per hour in the morning peak period (7am till 9 am) it would not simply be a case of hiring an additional bus and driver for that two hour period. The additional bus would have to be either purchased or leased on a permanent basis and the driver hired as either a part time or full time employee. The fixed cost element of an increase in service would therefore be substantial, i.e. the bus purchase costs, vehicle insurance, vehicle taxes, depot costs, maintenance costs. In practice a bus operator might choose to increase the service level throughout the day on one particular route, or may increase the service level of one route during the peak and another during the off-peak.
The same problem of indivisibility is faced by heavy and light rail operators to an even greater extent because each vehicle is many times more expensive than a bus. This means that the ability of public transport operators to change service levels, in terms of frequency, are restricted. It is much easier to extend hours of operation since existing vehicles can be utilised.

If the quality of public transport is to be improved the availability element of public transport would also increase. Availability is considered under the following (Kittleson Associates et al, 2003):

• Spatial availability: The network of services would be expanded in terms of coverage.
• Temporal availability: The services would also run more frequently.
• Information availability: The quality of information required for making a trip by public transport would improve.
• Capacity availability: New vehicles would be added to runs so that overcrowding is minimised
  

Increase in service levels: Expected impact on key policy objectives

	= Weakest possible positive contribution,		= strongest possible positive contribution
	= Weakest possible negative contribution		= strongest possible negative contribution

= No contribution

Increase in Service Levels: Expected impact on problems

	= Weakest possible positive contribution,		= strongest possible positive contribution
	= Weakest possible negative contribution		= strongest possible negative contribution

= No contribution

Expected winners and losers

Group	Winners / losers	Comment
Large scale freight and commercial traffic		High value freight journeys – less time spent in congestion the greater the vehicle utilization, however a relatively small proportion of the journey distance is in urban conditions. Service increase reduces traffic congestion so is beneficial. This depends upon the size of the service cross elasticities between car and bus.
Small businesses		Service increase – encourages trips to non local areas.
Cyclists including children		High incomes associated with high value of time and thus continued car use for high value journeys. These journeys will benefit from reduced congestion. A service increase reduces traffic congestion so is beneficial. This depends upon the size of the service cross elasticities between car and bus.
People at higher risk of health problems exacerbated by poor air quality		Unlikely to have car access. An extension of the service will increase the range of services, goods and opportunities open to them whilst an increase in frequency will reduce the generalised cost of travel by public transport.
High income car-users		If changes in service levels are restricted to existing services then no impact. However, if new services are implemented serving different areas then a very positive impact.
People with a low income		Service increase – will lead to reduced generalised costs of travel (e.g. reduced waiting and overcrowding) & more opportunities to travel if the service is extended.
People with poor access to public transport		Service increase - they may benefit from reduced congestion and improved or increased public transport supply.
All existing public transport users		Reduced generalised costs of public transport and reduced congestion will result in valuable time savings. A service increase reduces both so is beneficial.
People living adjacent to the area targeted		Reduced congestion will result in valuable time savings. A service increase reduces both so is beneficial. This depends upon the size of the cross elasticities between car and bus.
People making high value, important journeys
The average car user

	= weakest possible benefit,		= strongest benefit
	= weakest possible disbenefet,		= strongest possible disbenefit

= neither wins nor loses

Barriers to implementation

Barrier	Scale	Comment
Legal	Rail Based Deregulated bus Franchised bus Publicly owned bus	This will vary between local authorities and by type of mode. In most countries altering rail based public transport is extremely complicated, for technical and legal reasons, and takes several months to implement. Technically it is much easier to alter bus service levels. The legal barriers will, however, differ depending upon the local transit authority in charge of providing/regulating the bus service. In the UK (outside of London) bus services are provided and planned by the operators using a commercial criterion. Operators can change the level of bus services simply by notifying their local traffic commissioner of the intended change a few days beforehand. In most of Europe the planning of bus services tends to be the responsibility of the local transit authority. Any changes to service levels tend to involve detailed discussion amongst affected parties and will take several months.
Finance	Rail Based Deregulated bus Franchised bus	Altering service levels has a large impact on costs. The key issue is whether such changes are self financing or not. In the case of deregulated bus industry such changes will tend to be self financing and the cost implications will fall upon the passenger and not tax payers. In Europe the financial burden will tend to fall on both passengers and tax payers. It will therefore be a lot more difficult to finance any increase in services and politically very difficult to reduce services. This is also the case for rail based transport in the UK and abroad.
Governance		Governance will be more straightforward where services are run by public sector organisations. Where there is a combination or where services are provided by private organisations but under contract to public sector then governance arrangements may be complicated. Where services are deregulated governance could be straightforward however there may be political and public concerns about lack of network coverage.
Political acceptability	Reduction Increase + Subsidy Increase + No Subsidy	Political pressures, following a change in service levels, are likely to bear down from a number of sources. A reduction in service levels will not prove popular amongst the general public, but might be seen as politically necessary for budgetary reasons. Alternatively, an increase in service levels would receive support from the general public but not from all political parties if it means an increase in government subsidies, e.g. some politicians might prefer government support to be used elsewhere, e.g. hospitals.
Public and stakeholder acceptability		Increased services and extension of network coverage can gain public support.
Feasibility	Heavy Rail Based Bus and light rail	Changing the service levels of rail based public transport modes would appear to face substantial technical barriers in that any single change has a significant network impact. For bus based modes technical barriers do not exist and changes in service levels are relatively straightforward.

= minimal barrier,

= most significant barrier