Market Economics - Memo

Latest news & views from market economics / March 2013

The cost of physical inactivity
M.E have recently completed a study on the full cost of physical inactivity on behalf of Auckland Council, Greater Wellington Council and Waikato Regional Council. To find out more on this research that calculated a cost of $1.3 billion to the New Zealand economy, click here.

Kiribati EIA
M.E is carrying out an economic impact assessment of the Marine Training Centre in Kiribati for the Ministry of Foreign Affairs and Trade. The Centre trains students from across the Pacific Islands in a range of marine based occupations.

Alcohol sales policy
M.E has successfully tendered for a study for Auckland Council on the economic impacts of local area policy for alcohol sales, under the Sale and Supply of Alcohol Act. This project is now underway.

Planning for growth in schools
M.E is underway on a new contract with the Auckland office of the Ministry of Education. We are developing a model that estimates growth in school age demand on the existing school network in four Auckland catchments and translates this into a 20 year strategy for allocating new classroom capacity.

Invite to Canberra
This month Garry joined a NZ delegation led by Treasury who were invited to meet with the Geoscience Australia Economics Team working on the Critical Infrastructure Projection Modelling and Analysis Programme.

M.E is again pleased to be contributing to a number of education courses this year. The Directors are each contributing lectures and practical labs to Stage 2 and 3 Human Geography courses at the University of Auckland. Doug continues to be a member of the External Reference Group for the University of Auckland’s School of Environment. Greg will be co-presenting a NZPI training course on the economics of development later in the year.

Future Auckland - growing out vs growing up
Tensions have soared again between Auckland Council and the Government over Auckland’s future urban form, with Housing Minister Nick Smith vowing “to break through the stranglehold the legal metropolitan urban limit has on land supply…” and identifying land prices as the biggest factor putting home ownership out of reach of many Aucklanders. Such a move would run directly counter to the Council’s strategy for consolidation rather than continued outward expansion. This major clash of philosophies demands a strong evidence base to properly inform the debate.

Staff development
Last month two M.E staff had the opportunity to attend a course on the use of General Algebraic Modelling Software (GAMS) held at the University of Western Australia and given by Bruce McCarl, specialist in Applied Optimisation from Texas A&M University. This proved to be an excellent chance to refine mathematical programming techniques and learn directly from a practitioner who has a very long history in the development and application of optimisation software.



Costs, benefits and decision making

The words ‘cost’ and ‘benefit’ have been getting some serious attention lately. This has been driven largely by the Resource Management Reform Bill which (among other changes) drafts amendments to Section 32, giving greater emphasis to economic and employment opportunities, as well as quantification (where practicable) when carrying out an assessment of costs and benefits in evaluation reports. Submissions are now closed and at M.E we await with interest how the Section 32 amendments will evolve. In the meantime, we thought it was timely to look a bit closer at the issue of cost and benefit assessment. Specifically, this issue provides some insight into the principles of cost and benefit analysis, explores the appropriateness or otherwise of formal Cost Benefit Analysis (CBA) in a RMA Section 32 context (keeping in mind that the amendments do not specify that a formal CBA is required), provides an example of how a formal CBA has been applied in a policy context, and describes an alternative methodology that could also assist with RMA-based decision making.

Principles of cost and benefit analysis

Cost Benefit Analysis (CBA) in strict economic terms requires a quantification of what people value – both negative and positive – with these quantified values then monetised in dollar terms, so a comparison of relative values is possible. The theoretical foundation of formal CBA rests on benefits being defined as increases in human wellbeing (utility) and costs defined as reductions in human wellbeing (Pearce et al., 2006). Some costs and benefits are straightforward to measure, others are not so easy to discern.

The issue of valuation is central to cost benefit accounting methods, and economic methods have developed to measure value, with techniques evolving and continuously improving. Value is a property of an object (physical or abstract), representing its degree of importance. Economic valuation has its roots in utilitarianism, expressing the degree to which a good or service satisfies individual (or societal) preferences. However, people in different cultural settings communicate their sense of value in multi-layered ways (O’Connor, 2002 & 2007).

In quantifying costs and benefits, CBA needs to encompass both market and non-market effects. Formal CBA often seeks to apply strict valuation methods to derive shadow prices for the non-market values. The key to the (formal) CBA approach is monetary reductionism, that is, to reduce all costs and benefits of a project or policy to a dollar, which equals ‘net position’. When prices have been identified for market values and non-market values, CBA can then help identify the optimal choice from the range of possible resource allocations or scenarios. Many goods and services are able to be exchanged on a market, which automatically reveals their value. Market valuation is the price at which an asset or service would trade in a competitive auction setting, and is generally appropriate for private goods and services. When goods are transacted in this way, the market prices can be used to discern value.

However, market valuation is problematic for common pool resources and public goods where externalities are usually not included. The market is generally capable of revealing only one component of the total economic value, known as ‘direct use’ values. Other components of value include indirect use values (deriving from personal use but not traded on a market), and non-use values (including bequest and existence values). The nature of public goods (such as ecosystem services) means that although people do value them, there is little incentive from an individual’s perspective to pay to maintain that good, nor usually a mechanism to do so.

These characteristics of goods and their values mean it is not always possible or feasible to assign monetary values to costs and benefits, and it may be more appropriate to describe the perceived costs and benefits and leave them in qualitative form. This is especially so when externalities account for a significant share of the total benefits or costs. The problem is that qualitative assessment does not fit well into a monetised CBA, particularly if a cost benefit ratio is being sought for decision making.

Complicating matters is the reality that for many of the components which make up the four wellbeings (social, cultural, economic and environmental) there is not (yet) sufficient information to support the science and so support adequate quantification, let alone monetisation. This applies to each of the wellbeings per se, and is especially evident when the inter-relationships among the wellbeings have to be considered – which is a core requirement for RMA (and LGA) evaluation because the Act requires consideration of all of them. There is uncertainty over environmental effects in particular, given the potential for unintended consequences to natural systems which are unknown to decision makers and experts without significant scientific research.

Dr Catherine Murray, 09 915 5517 or

RMA Reforms: costs and benefits in s32

Assessment of costs and benefits is the basis of all decision-making (that is how humans think: will I/we be better off or worse off if I do this?). Hence there is intuitive comfort with the underlying principles and the inclusion of these general cost and benefit terms in Section 32 of the RMA.

However, because Cost Benefit Analysis (CBA) has been developed as a formal methodology and primarily applied by economists and engineers, it is commonly viewed as a specific methodology with associated rules and conventions to do it ‘properly’. This relates especially to quantitative CBA, where the relevant costs and benefits are monetised or at least quantified.

CBA works by comparing numeric values. If the total of measured benefits is greater than the total of measured costs we are better off. The benefit of 10 is always greater than cost of 9. Generally, formal CBA works best where the costs and benefits are direct, and are localised in space, and as between few affected parties. However, as the scope of effects widens (in space, through time, across affected parties) then things become fuzzier. More factors have to be taken into account: indirect, flow-on and feed-back effects become important. There is less ability to quantify and compare. Hence, the sum of the benefits is less precise, so is the sum of the costs. The benefit of 10 may now be somewhere between 8 and 12. The cost of 9 may now be between 7 and 11.

Many decisions in the RMA context fall into this more holistic (social, cultural, economic and environmental) and therefore fuzzier category, where traditional CBA usually struggles. The risk in the RMA reforms is that “placing greater emphasis on the need for quantitative assessment of costs and benefits” will be viewed in the simplest and narrowest context, where the RMA-mandate is seen as backing to ‘cut through the proverbial crap’ and follow a narrower quantified approach – where it fails to take into account the softer aspects (e.g. landscape effects) and by default places too much emphasis on quantifiable aspects (e.g. economic effects).

This is not a new problem and there is lots documented about the shortcomings of formal CBA – but the risk is that it will be ignored in the enthusiasm to get a ‘new’ approach into the RMA. The practical risk is that TLAs respond to the intent of the reforms by defining an “appropriate” cost and benefit analysis methodology (for all projects and policies) which is not able to take into account all the important matters.

This could occur because councils assume the RMA dictates that formal CBA is the method, yet formal CBA may well struggle with complex situations. The scope of evaluations would tend to get limited to those aspects which formal CBA is able to deal with (i.e. easily quantified effects that are direct, short term, and localised). The implication is that the matters which are taken into account in decision-making are reduced in scope. This might create more ‘efficient’ but poorer quality decisions.

Avoiding such an outcome depends on defining and adopting ‘an assessment of costs and benefits’ which is able to deal with the wider, longer, indirect, and less readily quantified matters. This is easier said than done, even though the challenges of doing so have been long recognised.

Some practical hurdles exist. Many are not familiar with CBA. Treasury has published a CBA primer (how to do it) and that is likely to be a first port of call for many. The Treasury primer has some useful points, but a CBA carried out in accordance with the Treasury approach is not likely to be adequate for RMA-related assessment, especially regarding the evaluation of proposed provisions. The good points of the Treasury primer are: identify all relevant criteria, make sure they are included, quantify where possible, evaluate. However, there are also weaknesses in the Treasury approach, such as short evaluation periods, high discount rate, and a focus on the national level ‘overall’ outcome, without requiring consideration of the distribution of effects within the community or economy. At a more general level is the implicit underlying assumption that important matters and effects can be accurately quantified at all, and expressed in simple number terms.

There are options. Other methods also take into account the costs and the benefits in a systematic manner – i.e. they are generic cost and benefit approaches, though not using the same mould as the formal CBA method (see the following MCA article for an example). M.E’s submission on the Reform Bill reflects these wider ‘implementation’ concerns. While registering our support for the strengthening and consistent application of the evidential base, we have also recommended that the explanatory notes underlying the amendments first specifically identify that there are a number of methodologies which are appropriate to achieve s32(2) which include but are not limited to ‘CBA’ (assuming it is applied in a sufficiently comprehensive manner), and second, we specified an expectation that the Bill require councils to support the s32 amendments through clear guidelines as to how the s32(2) assessment may be undertaken. We believe that the Ministry for the Environment could play a role in providing high-level guidance to TLAs on this matter.

To view M.E’s full submission on Section 32 amendments, click here.

Dr Douglas Fairgray, 09 915 5514 or

Applying CBA to air quality policy options

Working with Auckland Council staff, M.E has developed a formal Cost Benefit Analysis (CBA) model (the ‘Model’) to compare different policy options (combinations of four packages and three implementation paths) for managing Auckland’s domestic air emissions. Domestic air emissions, from open fires and inefficient wood-burners contribute about three quarters of Auckland’s particulate matter emissions in winter which in turn impact on the health of the community – especially those with respiratory problems, infants and the elderly - and puts pressure on the health system in terms of costs.

The Model runs off Council’s Domestic Fire Emissions Prediction Model which projects burner numbers and translates these into PM10 (particulate matter that is less than 10 microns in diameter) concentrations across the Auckland region. The Model then translates these into health effects of domestic emissions associated with PM10 and expresses the number of cases in monetary terms.

A key challenge for the application of CBA was how to monetise each cost and benefit component. In this case, it is fortunate that some guidelines exist. In terms of the health effects, the costs per health case were monetised using the values estimated in the Health and Air Pollution in New Zealand (HAPINZ) study. This study was commissioned by the Health Research Council of New Zealand in partnership with the Ministry for the Environment, Ministry of Transport, New Zealand Transport Agency with in-kind support from the Ministry of Health and the legacy Auckland Regional Council. This work formed a key building block of the Model because it linked sources of pollution, impacts on human health and the cost of these impacts.

Using these established costs made it possible to estimate the ‘cost effect’ of changing the population’s health. The total cost forms the baseline from where any change could be measured. If, for example, a policy measure reduces the number of burners operating in the region, then the PM10 concentration should decrease. In turn this should reduce the total health effects, avoiding health related costs. If some costs are avoided, then the net reduction in total cost can be viewed as an indicator of benefit. In other words, we used the costs and benefits of a policy change compared to the outcomes without the change, assuming a baseline or status quo.

With reference to the cost-side, Council’s costs to implement and enforce the policy as well as private costs to install clean heating devices were included. The private costs include installation and appliance costs as well as the consenting costs. Comparing the costs and benefits of the different policy options provided insight into the relative (cost-benefit) efficiency of the policy options, enabling a wider understanding of the relationships between the community benefits and the private costs.

During the study we encountered a range of methodological and practical issues – some were associated with the CBA technique in general and others were due to the study field. The main issues were:

Equity and distribution – traditional CBA techniques do not necessarily address questions of distribution across communities and tend to focus more on efficiency considerations. This means that questions around ‘who benefits’ and ‘who pays’ are typically left unanswered.

Monetising cost and benefits – this CBA compares the positive and negative outcomes of Council’s policy. The HAPINZ work provided the monetised values for the health effects needed for the Model. However in many cases such quantified or monetised values are not readily available and then expressing costs and benefits in dollar terms (if that is the aim) needs further research and analysis. This can be done using different valuation methods e.g. stated preference methods and/or revealed preference methods. Stated preference methods include ‘Willingness to Pay’ and Willingness to Avoid’ approaches whereas the Revealed Preference methods include ‘Travel Cost methods’ and ‘Hedonic Pricing methods’

Current value of future benefits and costs – the benefits and costs are distributed over different time periods and this requires choosing a discount rate which determines the value of future costs and benefits relative to current ones. The selected discount rate can make a big difference to the cost-benefit ratio (and implied desirability) of policy options if costs and benefits occur over long periods.

For this study, it was appropriate to undertake a sensitivity analysis to determine how the cost benefit ratio (CBR) responded to changes in the assumptions underlying the Model – in particular the variables associated with the costs, benefits and discount rates.

Key Findings: The CBA evaluated each policy option (package and implementation path combination) in terms of its CBR, benefits and costs and expressed these in discounted and undiscounted terms. Treasury’s default discount rate of 8% was used. The discounted CBR of the four packages and the three implementation approaches are compared below.

Package 1 (Point of Sale & Replacements) implemented using a Bylaw approach had the highest CBRNPV of 3.51. Package 3 (Point of Sale & Open Fire Ban) implemented with the contested appeal approach yields the lowest CBRNPV of 2.06. Importantly, the CBR only compares the benefits and cost of each package and does not consider the scale of benefits and costs associated with each package – these also need to be considered.

Package 4 (Point of Sale, Open Fire Ban & Replacements) had the highest benefits (i.e. health effects). In discounted terms this Package is projected to have health benefits of $430m (between 2012 and 2031). Package 2 (Open Fire Ban & Replacements) is projected to generate the second highest benefit with a NPV benefit of $ 370m, some $60m less than Package 4.

Combining these observations showed that either Package 2 or Package 4 is the most favourable policy option for Auckland Council due to the comparatively large benefit (and health effect) and the CBRs for these two packages are comparatively high. Package 1 was not an option due the relatively low benefits associated with it in spite of a favourable CBR.

This study, currently being updated by M.E, is a good example of how formal CBA can assist with local government decision making in a policy context. The issue was one of funding and investment and subsequent cost savings in the health sector, hence it was relevant to express the cost-benefit outcomes in monetary terms. The scope of costs and benefits considered was narrow: limited to Council, household and health costs and the benefits to reduced incidence of health effects, and importantly, is only a component of the wider (four well-beings) framework that would be required for assessing costs and benefits for RMA section 32 decision making.

Lawrence McIlrath, 09 915 5523 or

MCA contributed to the Auckland Plan

The establishment of a single Council for Auckland necessitated the development of a single planning direction for Auckland’s way forward, bringing together aspirations for better transport, environmental protection, improved land uses, housing growth, and economic development in one spatial plan. The Council anticipated that the Auckland Plan would need to stand up to scrutiny from a range of stakeholders because of its significant influence in shaping the role of Auckland’s future landscape, and needed to be developed in a robust and transparent manner with an evidential basis for decisions about the preferred way forward.

M.E had experience developing Multi Criteria Analysis (MCA) frameworks (e.g. the assessment of the Franklin District Growth Strategy) which proved a useful means of assessing a raft of complementary but sometimes competing assessment criteria. We used a similar approach for Auckland Council to understand how different spatial distributions of growth would be likely to impact on various aspects of the future living and business environments.

Among the benefits of a MCA framework are that it accommodates a range of diverse criteria in an organised structure, which allows for criteria (and underlying elements, and outcomes) to be taken into account and weighted to reflect their overall importance. These criteria can be informed by a number of measures both quantifiable and qualitative (being assigned a relative numeric value to allow for comparison with other directly measured indicators), which are generally derived from technical assessments such as modelling or evaluation by experts. In this way, it can link the political and scientific dimensions within a generally neutral structure.

For Auckland Council, the requirement was to establish that the draft Auckland Plan would satisfy the social, economic, cultural and environmental well-beings, as the high-level criteria. Underlying these overarching well-beings, at the next level, is a range of elements that contribute to each of the well-beings (for example, one element of economic well-being is the distribution of business and infrastructure capacity in a pattern which contributes to efficient urban function). At the next level, contributing to each element, in turn, is a set of outcomes. Each outcome has a weighting which contributes to its element’s weighting, which contributes in turn to its criterion’s weighting. This meant that the effects of any policy - or in this case, spatial distributions of economic and residential activity – may be assessed to understand which patterns would produce the most positive outcomes for each high level well-being.

The other dimension of the MCA framework has the range of effects – for example, a measure of traffic congestion – which are related to one or more outcomes. This means that each outcome may be assessed according to one or more effects, just as each effect may contribute to several outcomes. This structure helps take into account the underlying complexity of the real-world, where specific activities and their effects typically contribute to a number of aspects which are important to the community. It also means that (causal) relationships which are known to exist can be taken into account even when there is insufficient science to quantify them.

In this way, the MCA tool could provide a transparent evidential basis to assess options, within a defined and agreed, and fairly comprehensive structure. The work is a good example of how MCA can utilise an RMA-based effects framework to compare and contrast such major matters as Plan directions for a large regional community and economy. It takes into account the relevant aspects, assigns them importance (usually from a community perspective, informed by technical experts), and evaluates projects/policies within that framework. It places greater emphasis on valuing effects (in a relative sense) without necessarily quantifying effects in a nominal sense, but can also accommodate quantified effects where these are available. This is useful when many effects are difficult to quantify or are inappropriate to express in monetary terms.

If done well, MCA is a strong tool. It captures the intuitive aspects of decision making (what is relevant, am I better off or worse off taking the relevant matters into account) and is better at taking into account the wider aspects even while recognising that the numbers are a bit uncertain. It can address the big issues. And the requirement to weight the evaluation criteria mitigates against its use by technical or interest-based individuals on their own as it usually requires several operators who are empowered to make the necessary decisions and choices. Experience from the challenges of undertaking a formal CBA, in a situation which (for RMA applications) requires consideration of many wider, flow on, uncertain and long term matters, is that decision-makers often lean in the direction of a more comprehensive (if less quantified) tool, such as MCA.

MCA like any methodology has its limitations. Weighting the importance of the criteria is often contentious and may vary depending on who is being consulted. The fall back position is to treat all elements and outcomes as equally important and to allow weighting scenarios to demonstrate how sensitive the results are to these weighting changes. Nevertheless, where it is possible to reach agreement on the importance of the well-beings, their contributing elements, and their contributing outcomes, and the range of effects which contribute to these, then it makes available an approach which is held constant across a range of evaluations, as an efficient and consistent decision making tool that reflects the aspirations and objectives of the local community.

Rebecca Foy, 09 915 5515 or


Market Economics Ltd
Level 5, 507 Lake Road,
Takapuna 0740, Auckland, NZ
+64 9 915 5510
+64 9 915 5513

© 2013 Market Economics.