Travel behaviour change evaluation procedures

Technical report

Print version: Travel behaviour change evaluation procedures (PDF, 1.57 MB, 98 pages)

2. Evaluation methodology

• 2.1   Context
• 2.2   Transfund allocation process
• 2.3   Selection of evaluation framework
• 2.4   Theoretical framework
• 2.5   Practice in TBhC evaluations
• 2.6   Conclusion

2.1   Context

A key requirement is for the TBhC evaluation procedures developed in this project to be consistent with the Transfund Allocation Process and not inconsistent with other existing Transfund evaluation and funding procedures including:

• Project evaluation manual
• Alternatives to roading procedures
• Simplified procedure for walking and cycling projects
• Public transport funding policies.

All of these procedures are based on a social cost benefit analysis framework although the Allocation Process also includes more detailed reporting of all project effects in a triple bottom line format rather than just a single economic efficiency indicator.

2.2   Transfund allocation process

Over the last year Transfund has been revising the allocation process that it uses to prepare the National Land Transport Programme to take account of the publication of the New Zealand Transport Strategy in 2002 and the passing of the Land Transport Management Act (the Act) in 2003. The Act broadened Transfund’s objective to allocate resources in a way that contributes to an integrated, safe, responsive and sustainable land transport system.

To achieve this the Allocation Process comprises the following six stages:

• Formulation
• Assessment
• Prioritisation
• Programming
• Approval
• Monitoring.

TBhC evaluation procedures are expected to primarily input to and support the assessment stage in the Allocation Process but will also be relevant to the requirements of the other stages.

A prominent requirement of the Transfund Allocation Process is an increased emphasis on the development of “packages” of inter-related and complementary activities, where appropriate. Transfund’s September 2004 document describing the Allocation Process states that these packages should be clearly related to specific transport issues, and be optimised to make the most efficient and effective use of resources in addressing those issues. Approved organisations will be encouraged to develop joint packages with other approved organisations where this is appropriate. Examples might include packages designed to ensure an integrated multi-modal approach to developing new infrastructure, or to reduce the need to build new infrastructure by developing complementary network supply and demand management initiatives. The Allocation Process document gives a number of examples of packages that include TBhC measures and it is likely that organisations will wish to include TBhC initiatives as part of packages.

The Allocation Process document includes a section describing information requirements. The reporting requirements specify the information that is required to assess the alignment of projects with the objectives of the New Zealand Transport Strategy and the other requirements of the Act. Reporting requirements are grouped in the following three categories:

• Seriousness and urgency
• Effectiveness
• Economic efficiency.

Effectiveness covers a range of considerations including integration, safety, sustainability, and responsiveness, and the impact of the project on economic development, safety and personal security, access and mobility, public health, and environmental sustainability, as well as the level of confidence that anticipated outcomes will occur. Economic efficiency is demonstrated by benefit/cost ratio (including assessment of uncertainty) and identification of non-monetised benefits.

The Allocation Process is being implemented as an Internet based system. On-line forms will provide guidance to project proponents about the information required and are supported by a detailed reference document.

Transfund has differing information requirements for the following three types of proposals:

• Generic
• Standard
• Complex.

A table of generic activity categories is listed in the Allocation Process document. These include various types of walking and cycling projects and various types of public transport funding proposals. The generic categories are used to reduce the information requirements in programme submissions and the depth of evaluation required to develop the National Land Transport Programme. Each has a standard profile developed with respect to the seriousness and urgency, and effectiveness factors. The profile is completed for each generic proposal by a project specific assessment of its economic efficiency.

TBhC projects are not on the current list of generic activities in the Allocation Process document but at least some of the more common TBhC project types are expected eventually to fit the definition for inclusion, which is to be activities that are sufficiently similar in nature and impact to allow the development of standard assessments and profiles for each category. In the meantime Transfund has indicated that TBhC projects should be evaluated as standard projects.

If TBhC projects are eventually categorised as generic projects they will be assigned a standard profile with respect to seriousness and urgency, and effectiveness. This means that it would not be necessary for project proponents to provide a detailed assessment and justification of each proposal against the requirements of the Act and relevant strategies since this would be standardised and covered by a standard profile for TBhC projects. This is discussed further in Section 5.0.

However irrespective of whether TBhC projects are assessed by Transfund as standard or generic activities it would still be necessary to undertake an assessment of the economic efficiency of each specific TBhC project to complete the Transfund Allocation Process information requirements. The next section discusses the potential frameworks for assessing economic efficiency.

2.3   Selection of evaluation framework

As part of the review of New Zealand and international literature we examined existing TBhC evaluations and evaluation procedures. Key objectives were to investigate what problems had been encountered in applying existing Transfund procedures to TBhC projects, why they were considered unsuitable for TBhC projects, and how, if at all, other countries/jurisdictions had overcome these issues.

Perceived problems and issues raised in relation to TBhC evaluations include the following:

• There are no applicable existing simplified procedures for TBhC projects
• The Project Evaluation Manual or Alternative to Roading Evaluation Procedures are for infrastructure projects or projects that change generalised cost for existing users
• TBhC projects do not usually change the actual service level or price (and hence generalised cost) of any mode
• Effects are diverse and dispersed and hence difficult (and/or costly) to measure, estimate and incorporate accurately in an evaluation - effort involved in this is not justified by project cost as TBhC projects tend to be smaller projects
• Uncertainty and apparent variability of the diversion rates and other effects (eg lengths and timing of avoided trips)
• Diversity of TBhC projects means that analysts have to adapt existing procedures and cannot just “follow the formula”
• Newness of TBhC means analysts have to make assumptions and projections on limited experience which results in inconsistent treatment of various aspects in evaluations.

Our literature review examined a number of evaluations and did not find a common approach on many of these issues. It is apparent that a significant degree of approximating, aggregating, and averaging will need to be accepted if appropriate simplified evaluation procedures are to be developed for most small to medium scale TBhC projects in New Zealand. The current public transport funding policies, and simplified procedures for walking and cycling are existing examples of Transfund procedures using averaging so it would not be inconsistent for TBhC evaluations to use broad averages in a similar way.

Most evaluations attempted to deal with the above problems and issues within a cost benefit analysis framework. A basic distinction in evaluation frameworks is that between quantitative and qualitative evaluation. Quantitative evaluation (such as cost benefit analysis) involves assessing the amount of change, for example, the reduction in vehicle kilometres travelled, associated with a project. A monetary unit value is then, generally, applied to this change. Under a qualitative framework (such as multi-criteria assessment) the measurement/estimation of the quantity of change is not required, only a subjective judgement is made about the relative magnitude of the impact and whether it is positive or negative.

It is understood that TBhC measures are likely to be funded from a separate activity category and that it is not an absolute requirement for the measure of project “value for money” to be comparable with benefit cost ratios and efficiency ratios for projects in other categories. It would be possible to consider using a multi-criteria assessment approach for TBhC evaluations even if this is not the approach used for other types of projects.

However, the literature review and the consultation undertaken for this project identified the following issues that support the adoption of a form of social cost benefit analysis as the most appropriate evaluation approach for determining the economic efficiency of TBhC projects as required by the Transfund Allocation Process.

Firstly, many of the benefits of TBhC projects are of the same types as those of other projects, eg travel time savings for remaining road users, and reduced environmental externalities. Unit values for these benefits already exist. The difficulty in applying them in the case of TBhC projects is mostly due to the difficulty of measuring the effects/changes that these unit values are applied to. Greater consistency with other Transfund procedures is likely to be achieved if effort is put into improving estimates of effects (even if these are still subject to large uncertainty) and then using common benefit unit values than if scoring systems are used to measure both effect and unit value of TBhC projects.

Secondly, the investigations to date indicate that:

• some types of TBhC projects are more effective if implemented together with complementary traffic restraint measures,
• many TBhC projects are likely to include items of infrastructure, and
• many councils are likely to want to propose TBhC initiatives as part of integrated packages of measures that combine road, alternatives to roading, and walking/cycling projects, and possibly pricing measures.

It is likely to be more difficult to compare packages if different evaluation methods are used for different package components. Other approaches such as multi-criteria assessment may be less suited to prioritisation which is a required stage of the Transfund Allocation Process.

Thirdly, most of the evaluations and evaluation related papers reviewed for this study adopted a cost benefit analysis framework so it appears that such an approach would not be inconsistent with general practice. TBhC projects are transport interventions like any other and in principle should be amenable to a social cost benefit analysis evaluation. Having said that, it is noted that the number of TBhC evaluations that have been undertaken to date is relatively small, and these were in New Zealand, Australia and the United Kingdom (an indicative evaluation in Smarter Choices), which tend to prefer more quantitative approaches.

There may be additional risks for Transfund in trying to develop TBhC evaluation procedures based on multi criteria assessment that are consistent with other Transfund procedures, particularly given the overlaps between TBhC and other project types. For example, Transfund would not be able to tell whether a cycle path included as part of a TBhC project was being attributed the same benefits as one assessed using the existing simplified procedure for walking and cycling. This could lead to projects which were not economic in one category still being funded in another.

Given that Transfund currently uses a cost benefit analysis framework for its evaluation procedures, a similar framework for TBhC may also have the advantage that analysts already have a working knowledge of the framework. The use of pre-determined unit values provides a common measure for benefits. A cost benefit analysis framework should provide outputs that are more easily comparable between different projects and possibly be less subjective (although this might be questioned in early years).

A further advantage of using a cost benefit analysis framework for TBhC projects is that if such projects are as effective as some experience to date indicates they may be, the use of evaluation procedures that enable them to be compared directly with other types of transport intervention will help to establish their credibility more quickly both within the proponent councils and with national policy and funding agencies.

Based on these considerations a cost benefit analysis framework was adopted for evaluation of TBhC projects.

2.4   Theoretical framework

A conventional theoretical framework for cost benefit analysis of transport projects involving switching between modes (of which TBhC projects are a subset) assesses the benefits as the sum of the following:

1. Resource benefits to people already on mode which is improved (generalised cost change for that mode)
2. Perceived benefits to mode switchers (behaviour changers)
3. Benefits from avoidance of unperceived costs associated with previous behaviour of switchers, comprising:
   1. resource cost adjustments for switchers themselves (including monetary (eg non-fuel variable vehicle operating costs) and non monetary (eg accident trauma))
   2. other resource cost impacts (externalities) on other transport system users or of the transport system (eg decongestion, environmental, and accident externalities)
4. Unperceived costs associated with new behaviour of switchers, comprising:
   1. resource cost adjustments for switchers themselves (including monetary (eg public transport fare payments) and non monetary (eg health benefits of cycling and walking))
   2. other resource cost impacts (externalities) on other transport system users or of the transport system (eg environmental, accident, and health externalities (to the extent that costs of less health were being incurred by society other than the behaviour changer)).

Category (A) benefits are the benefits to existing users of the mode that is improved by the infrastructure project or public transport service improvement. Benefits to existing users are changes in generalised cost and usually include mainly aspects of cost, time and comfort.

If people change mode in response to an infrastructure project or public transport service improvement (ie a “supply side intervention”) their benefits (B) are valued at half the unit benefits to existing users (A). For the mode changers we also add the resource cost adjustments (C(i) and D(i)). This represents the additional resource cost savings to the behaviour changer themselves resulting from replacing a car trip with a public transport trip (or cycle/walk trip) that are not included in the perceived benefit or rule of a half benefit. Finally we add the other resource cost impacts on other transport system users or of the transport system (C(ii) and D(ii)). These include decongestion and net environmental externalities.

This is the approach used for Alternatives to Roading evaluations in New Zealand (although the resource cost adjustment items C(i) and D(i) have tended to be ignored). It is also the approach used as the basis for Transfund’s public transport patronage funding policy. This approach is consistent with established transport evaluation theory as discussed in an authoritative paper by Neuburger. The approach and principles are also similar to those used for evaluating user benefits in situations where there is induced traffic.

TBhC projects primarily involve “soft” measures such as marketing and information that aim to change perceptions and knowledge about different travel options and choices rather than changing generalised cost. Therefore category (A) benefits are typically zero for TBhC projects. Some types of TBhC projects, eg school travel plans, may involve some infrastructure improvements that change generalised cost for people already using that infrastructure and this may still need to be quantified in some cases if significant. Also it could be argued that the more accurate perception of costs that is achieved by the TBhC programme is a benefit to existing users even if they do not change mode or they already use environmentally friendly modes. This effect is ignored as it has been decided previously that we are mainly interested in actual behaviour change, not simply changed “travel awareness” without change in behaviour.

Estimating category (B) benefits is problematic with TBhC projects. Normally the benefits to mode switchers can be valued at half of the unit benefits to existing users (Category A), but as noted above in the case of TBhC projects such benefits are often zero. The benefit to mode switchers cannot be zero or people would be indifferent about changing behaviour. The explanation is that TBhC programmes change the information available to households and individuals and, partly as a result, their perceptions about alternative travel modes and choices - even where there are no changes to the system itself.

In the case of TBhC projects, people make changes because the new information:

• corrects an information gap or misperception and they realise that the alternative actually is more attractive on balance than the private car trip that it replaces, or
• changes their attitude so that they are willing to accept the disadvantages of the alternative mode because they feel that it is the right thing to do, eg they are being more environmentally responsible. This is still a valid benefit.

For TBhC projects category (C(i)) and (D(i)) benefits are the conventional ‘resource cost corrections’, reflecting differences between cost changes as perceived by the user and actual total resource cost changes associated with the user. Category C(i) benefits will include non-perceived costs of car use (car maintenance and other non-fuel variable costs, parking subsidies, part of accident costs, health costs etc), and D(i) will include public transport fares, cycle costs. This is discussed further in Section 3.0.

An alternative cost benefit analysis approach that has been used in several evaluations in Australia and is similar to the approach used for most road projects as outlined in the Transfund Project Evaluation Manual is to attempt to assess changes in resource costs directly rather than measuring perceived cost changes and making resource cost corrections. This is discussed further below.

2.5   Practice in TBhC evaluations

Previous studies have identified two approaches for valuing the benefits associated with behaviour changers in TBhC projects (New Zealand evaluations using the Alternatives to Roading procedures have not explicitly assessed behaviour changer benefits):

• Resource cost approach (used by Ker and James in evaluation of South Perth individualised marketing in 1999, and also more recently by Victorian Department of Infrastructure)
• Change in perceived costs approach (based on a paper by Ray Winn at Australasian Transport Research Forum, 2004)

Resource cost approach

This approach is described in a paper by Ker and James which outlines the evaluation framework, methodology, measurement of benefits, and estimated socio-economic returns of applying Individualised Marketing for the whole of the City of South Perth. The City of South Perth is an inner suburban municipality with a population of 35,000 residents. The Victorian Department of Infrastructure has recently used a similar method for an economic evaluation of a TravelSMART community programme demonstration project for two suburbs each having populations of 25,000 residents.

Some aspects of Ker and James’ methodology are not clear but the basis of the approach is to estimate benefits directly, as decreases (or increases) in resource costs compared with the base case (no TBhC programme). They do not explicitly follow the theoretical framework discussed in the previous section and in particular do not appear to distinguish between perceived and resource costs or to include resource cost corrections.

The direct first-year impacts on the individual, other than travel time, for South Perth, as calculated by Ker and James, are set out in Table 1. Note that they do not appear to have distinguished between resource and perceived costs, perhaps assuming that resource cost can be taken as equal to perceived cost (but clearly this is not the case for public transport fares).

Table 1: Private (user) benefits ($/year)

Impact	Value
Private vehicle operating costs	+A$3.53M
Public transport fares	–A$0.62M
Cycling costs	–A$0.05M
Walking costs	Not estimated
Health and fitness (mortality)	+A$0.58M
Perceived cycle/walk injury risk	–A$0.90M
Total	+A$2.54M

The paper notes that travel time related items were treated separately because of uncertainty about the real value of small increments of time. It was estimated that on average behaviour changers in South Perth would experience a four minute per person per day increase in travel time.

Ker and James noted that the conventionally derived value of time savings for road project evaluation is A$7.33/hour. Using this value, the loss to individuals due to the increased travel time of 4 minutes per person per day was calculated at A$4.73 million in the first year. However, if this were the true value, users would not have changed their behaviour because they would experience a net disbenefit of $2.19M ($4.73 -$2.54M).

For the observed behaviour change to have occurred Ker and James concluded that either:

• the behavioural value of time (for the behaviour changers) is much less in this case than has conventionally been assumed; or
• there are benefits to the individual over and above those that have been quantified here, (or both).

Ker and James state that either the value of time has to be factored down by at least 46% (to a maximum of $3.94/hour - this would result in a first-year travel time disbenefit of $2.54M and hence a zero net benefit in Table 1) or there are additional unquantified benefits equivalent to at least 86% of the value of those currently quantified. The impact on the overall (net) level of benefits and the expression of socio-economic benefits relative to costs is the same whichever basis is used, since both affect the same part of any expression and only that part.

Ker and James state that in their approach “the base case for the evaluation is a zero value of time, with sensitivity testing at $3.94 per person hour, but it should be recognised that this may reflect un-estimated benefits as much as the real value of travel time”.

There are two further possible explanations for the observed behaviour change that were not discussed by Ker and James. Firstly, everyone’s value of time is different. Average values are used in transport project evaluations for equity reasons and to keep evaluations manageable. In reality travel time values for the population are distributed along a curve. It seems likely that behaviour changers are more likely to come from the proportion of the population with lower than average values of time, rather than that the average value of time is overestimated (values of time are based on extensive research).

Another possible explanation which many people find intuitively plausible is that the value of time savings on public transport or walking/cycling is less than car driver time savings because time on alternative modes has other uses (reading/sleeping or meeting exercise needs respectively). If time is being used effectively the value of time savings, or the disutility associated with travel, will be lower. In this case the different value of time would apply to the whole trip, not just to the difference in travel time (increased travel time of 4 minutes per person per day in the above example). This means that the difference in travel time values does not need to be as great as suggested by Ker and James.

For example if a trip typically takes 30 minutes by car and 34 minutes by public transport then at the Ker and James standard value of time of $7.33/hour the “cost” of the trip by car is 7.33*30/60 = $3.67. From this we can calculate the value of time on public transport that results in the same travel time “cost” for the trip. This is given by public transport value of time = 3.67*60/34 = $6.47/hour. The difference (approximately 10%) seems much more plausible than a zero value of time.

In 2002 Transfund completed a major review of benefit parameter values used in its evaluation procedures. The review included extensive market research involving stated preference “willingness to pay” surveys to ascertain New Zealand transport users’ preferences and values for the benefits provided by road improvement projects and urban bus and passenger rail services. A Transfund Information Paper summarising the changes arising from the review reported values of time as shown in Table 2.

Table 2: New Zealand values of travel time

User category	Travel time value as at 2001 ($/hr)
	Commuting	Other non-work
Car drivers	7.80	6.90
Car passengers	5.85	5.20
Public transport occupants •    Seated •    Standing	4.70 6.60	3.05 4.25
Pedestrian	6.60	4.25
Cyclist	6.60	4.25

The report noted that the new values for public transport users are lower than those for car occupants and public transport projects will therefore obtain less benefits in comparison to road projects. Transfund’s Board decided not to adopt “equity” values, and instead adopted the mode-specific values produced by the New Zealand research. While this may appear to disadvantage conventional public transport projects, it appears consistent with the hypothesis that the value of time savings on public transport or walking/cycling is less than car driver time savings because time on alternative modes has other uses, as discussed above.

Using the New Zealand values from Table 2 in the example on the previous page would indicate a benefit for a mode changer notwithstanding the longer travel time. The “cost” of undertaking the trip as a car driver is $7.80*30/60 = $3.90 (assuming a commuting trip). Using the same assumption as before about the total trip time, the “cost” of making the trip by public transport instead is $4.70*34/60 = $2.66, so the net benefit is $3.90-$2.66 = $1.24 even though it takes slightly longer by public transport. The different travel time values obtained from the stated preference surveys may have been influenced by income differences between average car and public transport users although adjustments were probably made for this.

Another evaluation by Tisato and Robinson of travel blending in Adelaide found significant travel time savings to behaviour changers but this appeared to be due to an actual reduction in travel due to trip chaining and does not account for the longer travel time that occurs with mode changes. All other studies assumed that travel time for behaviour changers would be longer.

There is no clear basis for the assumption that Ker and James use in their central case, ie that the value of time be set to zero. This is convenient for the calculation but it gives no information about the value of “other” benefits to behaviour changers. The resulting benefit value is therefore arbitrary.

Change in perceived costs approach

This approach follows the conventional theoretical approach of using the observed behaviour of mode switchers (behaviour changers) to estimate perceived benefits to behaviour changers and then applying resource cost corrections to adjust for differences between perceived and resource costs to determine total behaviour changer benefits. The perceived benefits inherently incorporate all of the unquantifiable benefits discussed in the previous section including variations in value of time.

This approach is described in a paper by Winn which sought to address the appraisal of benefits to behaviour changers in community based TBhC projects by inferring perceived benefit values from the mode split model in the Melbourne Integrated transport model. Winn noted that there are some significant challenges with respect to the appraisal of user (behaviour changer) benefits and that the comparison of participants’ generalised costs of travel before and after any behavioural change (the approach used in previous section) is an inadequate and potentially misleading basis for the calculation of total behaviour changer benefits.

Winn’s paper estimates benefits perceived by behaviour changers (B) from the mode split relationships which are incorporated in the Melbourne Integrated Transport model to allocate trips between motorised and non-motorised modes and between car and public transport within the motorised category. These relationships reflect the change in mode shares between two modes that will result from changes in the relative generalised costs of the two modes.

Winn noted that, based on experience in Australia and internationally, a proposed community based TBhC programme was expected to increase the public transport share of motorised trips from 13% to 17%. Using the relationship in the Melbourne transport model it was estimated that a change in the generalised costs of public transport compared to car of around $2 would be required to achieve this mode shift. Applying the rule of a half resulted in value of $1.00 per trip for the perceived benefit to behaviour changers (B). A similar approach resulted in an equivalent benefit value of $0.75 per trip for switching from car travel to walking and cycling.

Winn’s paper avoids the need for Category C(i) resource cost corrections in respect of vehicle operating costs by noting “while the issue of the perception of vehicle operating costs is important and misperception implies an underestimate of the benefits, we proceed on the assumption that the TBhC programme provides sufficient information for participants to understand the full costs of the car alternative.” Thus there is no longer any resource cost misperception to correct (corrections may still be required for unperceived accident costs). Winn includes additional public transport revenue from the new users as a benefit. This is a form of resource cost correction to offset the fact that the fare is perceived as a cost in the behaviour changer benefit term (B), whereas fares actually represent a transfer and not resource costs. If the public transport operator incurs increased costs due to having to provide additional capacity this is included separately. Generally it is assumed that there is sufficient spare capacity on public transport to accommodate behaviour changers.

This approach offers a more thorough alternative to the first approach and it identifies a tangible estimate of the net travel time benefits/disbenefits and “other” benefits to the behaviour changers. However, it requires a number of assumptions about the perceptions of behaviour changers and careful analysis to avoid double counting (see next paragraph). The values for behaviour changer benefits may be lower in New Zealand cities.

An important assumption with this approach is that the TBhC program provides sufficient information for participants to understand the full costs of car travel. If the programme only corrects some of the resource cost misperception for car travel there would still be a C(i) component and the total benefits from people changing behaviour could be higher than calculated above. However another possibility is that the information provided by the TBhC programme also results in people more accurately perceiving some of the externality costs and benefits associated with car use and alternative modes such as environmental, health and accident benefits/costs. If this is the case, it will be reflected in the category B benefit term and a corresponding reduction in the C(ii) and D(ii) terms (external resource cost impacts of the transport system) may be required to avoid double counting.

The only apparent way to quantify these factors more robustly appears to be market research surveys of TBhC project participants to identify the relative importance of each contributing factor to their decisions to change. However this could be expensive and might still not provide conclusive results. Sensitivity testing would be worthwhile to see whether plausible variations in these factors have an influence on the evaluation result.

2.6   Conclusion

Both of the methods for estimating benefits of TBhC projects discussed in the previous sections involve some simplifying assumptions. The second approach, estimating perceived costs/benefits and making resource cost corrections, is considered to provide a more complete assessment of the differences in perceived costs and the variations in value of travel time and other benefits that appear to be the key drivers of travel behaviour change. The first approach does not appear able to adequately deal with these factors.

However, the second approach also involves some assumptions and issues including:

• Availability of mode split relationships applicable to New Zealand cities
• Assumption that TBhC programme results in car related costs being fully perceived
• Assumption that the TBhC programme does not cause some externality costs to be included in perceived costs.

Sensitivity testing was undertaken to determine the impact of variations in the above assumptions. Results of this are discussed in Section 5.5. Further research may be required in future to refine some of these values.

The change in perceived costs with resource cost corrections approach was identified as being the most theoretically defensible approach if it could be adapted for TBhC projects in New Zealand. Subsequent investigations have established perceived TBh changer benefit values and resource cost corrections so this approach is considered the preferred framework for evaluation of TBhC projects in New Zealand. Derivation of appropriate New Zealand values of TBh changer benefits and resource cost corrections is covered in Section 3.0.

 

< previous | next >

Page created: 28 October 2008