The Role of Cost-benefit Analysis and Economic Impact Analysis in Environmental Assessment: The Case for Reform

ABSTRACT

Cost-benefit analysis and economic impact analysis both provide the ability to assess projects’ economic impacts, but through different methodological approaches and perspectives. In Canada and elsewhere, cost-benefit analysis is often eschewed in favour of economic impact analysis in environmental assessment processes. This paper presents a criteria-based evaluation of the two methods and shows that cost-benefit analysis has numerous strengths relative to economic impact analysis but that both methods are useful. Consequently, we recommend that the environmental assessment process be reformed to include cost-benefit analysis complemented by economic impact analysis to provide decision-makers with more complete information.

KEYWORDS:

• cost-benefit analysis
• economic impact analysis
• methods of impact assessment
• benefits

Introduction

Environmental assessment (EA) seeks to inform decision-makers on the potential impacts of projects and to help them determine if a project is in the public interest. In Canada, the EA process seeks to ensure that projects are in the ‘public interest’ and that any ‘significant adverse effects’ of projects are ‘justifiable’ based upon consideration of environmental, cultural, economic, health, and social impacts (e.g., section 10, Alberta Oil Sands Conservation Act; section 52, National Energy Board Act; section 36, Impact Assessment Act).

Analysis of economic impacts is central to the assessment process.

A common method used in EA in Canada and elsewhere to assess economic impacts is economic impact analysis (EconIA).¹ EconIA relies on economic linkages and multipliers emanating from Keynesian income determination and Leontief input-output models to provide information on a project’s expected economic output, employment, and tax revenue. EA journals have published a variety of case studies of EconIA, e.g., for coal mining (Robertson et al. 2017), agriculture (Piper 2003), and habitat conservation (Prato and Hamed 1999).

While EconIA is helpful for forecasting the economic impacts of projects to support community planning (Gunton 2003), EconIA is typically misused to justify projects on the basis that the economic impacts are benefits that offset adverse environmental impacts (McDonald 1990; Lockie et al. 2008; Denniss 2012). Yet EconIA has serious limitations with respect to estimating project benefits because of its methodological inability to incorporate costs. EconIA’s underlying assumption of no constraints on labour or other inputs, and its omission of a range of impacts, such as government infrastructure costs, means that it is only capable of assessing a limited range of gross economic impacts and not a project’s contribution to the public interest (Grady and Muller 1988; Davis 1990; Gunton 2003; Williams 2019).

Cost-benefit analysis (CBA) is another method capable of assessing economic but also environmental and other types of project impacts, and consequently it is capable of determining a project’s net impacts. The basic steps in CBA are: (1) determine the scope of the analysis in terms of whose gains or losses will be considered, (2) predict benefits and costs over the life of a project, (3) convert any impacts not normally measured in monetary terms into such terms as feasible and appropriate, (4) discount monetized impacts, (5) compute net present value (NPV), internal rate of return, and/or benefit-cost ratio, (6) perform sensitivity analyses, and (7) interpret results (Boardman et al. 2018). Case studies in the EA literature cover energy (Shaton and Hervik 2018), water (Mohammed 2009), transport (Fischer 2006; Kolosz and Grant-Muller 2015), and waste (Manni and Runhaar 2014). CBA has a long history and holds a prominent place in project, program, and policy evaluation around the world including many developed countries by international lending institutions such as the World Bank (Hanley 2001; Rodrigo 2005; Browne and Ryan 2011; White and VanLandingham 2015).

However, despite its long history and strong theoretical foundations as a method of project appraisal, CBA is not widely used in EA in Canada. One of the standard Canadian texts on EA, for example, does not reference CBA (Hanna 2016), another only notes that EA was developed to address CBA limitations (Noble 2014, 8), and no Canadian EA agency currently requires CBA. The lack of use of CBA may in part be due to the technical and philosophical critiques of CBA on such issues as how environmental impacts are monetized and how future impacts are treated through discounting (e.g., Sagoff 1988; Hanley and Spash 1993; Vatn and Bromley 1994; Atkinson and Mourato 2008). de Jong and Geerlings (2003) describe how the Netherlands temporarily stepped away from CBA due to such critiques and that fears persist of ‘econocracy’ where decisions are dictated by dogmatic cost-benefit analysts. Gillespie and Bennett (2015) note the disconnect between Australian policy calling for CBA and resistance on the ground with its use.

Others suggest that method usage patterns reflect EA community knowledge, cognitive bias, and/or strategic interests (Boardman et al. 1993; Green 1997; Navrud and Pruckner 1997; Adamowicz 2004; Hahn and Dudley 2007; Howlett and Lindquist 2007; Brueckner et al. 2014). Relying on EconIA in EA undoubtedly reflects routines, institutional inertia, educational backgrounds, and experience of overseeing authorities. Strategic interests of proponents can also be a factor; EconIA tends to support project approval by exaggerating project benefits because it measures gross economic impacts instead of net impacts, while CBA estimates more modest net impacts (Gunton 2003). CBA may also be unfavourable to project stakeholders because it can provide much greater resolution on the distribution of impacts than EconIA of which proponents may not wish to display.

There is little research systematically evaluating the role of CBA relative to EconIA in the EA context in particular. Williams (2019) advocates a broader method of project evaluation, which she calls ‘economic impact assessment’, that is intended to overcome the limitations of EconIA by evaluating projects’ capabilities with respect to UN sustainable development goals. James (1994) concluded that CBA can be used to examine questions of significance, distribution, and project value, and that while EconIA can help predict projects’ cumulative economic impacts, EconIA only assesses gross impacts. Crookes and Wit (2009) examined the two methods in terms of their contributions to cumulative effects assessment and found that CBA performed well. Other studies (Nardini 1997; Deakin et al. 2002; Gasparatos et al. 2008; Browne and Ryan 2011; Hoogmartens et al. 2014) focused on CBA’s capabilities as an alternative to EA and as a method of transport project appraisal and sustainability assessment. But although this research is helpful, none of it comprehensively evaluates the relative strengths and weaknesses of CBA and EconIA in EA.

This article addresses this gap in the literature by evaluating CBA and EconIA using nine criteria that define a sound method of impact assessment. We begin with a summary of the criteria, followed by the evaluation of the methods. We conclude with a discussion of the appropriate roles of the methods in EA.

Methods

The evaluation is based upon criteria developed in a previous study (Table 1; Joseph (2013)) that synthesize the ideal characteristics of impact assessment methods. These criteria are used to structure a review of the two methods based on findings in the literature and our own experience using these methods in EA studies in Canada and the US.

Table 1. Criteria defining a sound method of impact assessment

Criterion	Description	References
Suitability to Context	Method is appropriate to the role, types of data, impacts, information sought, and other characteristics of the analytical context.	1, 4, 7, 9, 12, 13, 15, 16, 19, 21, 22
Flexibility	Method performs well under changing circumstances.	1, 4, 7, 21
Scientific Robustness	Method’s steps are sensible, and method is logical, sensitive to inputs, produces consistent results with same inputs, produces results that compare well with results of other methods designed to answer similar questions, is not internally biased, and produces results that are consistent with theory.	1, 3, 4, 6, 7, 12, 15, 19, 20, 22
Minimal Subjectivity	Method requires few subjective inputs.	4, 7, 15, 17
Ease of Comprehension	Method is easily evaluated by potential users for applicability, and method’s underlying theory, key assumptions, and results are easy to understand. Method is easy to use, and it is easy to see how method works.	1, 3, 4, 7, 8, 12, 16, 21
Usefulness of Outputs	Outputs address questions being asked, facilitate understanding of the meaning of the project’s impacts, and are expressed in familiar and verifiable terms.	1, 2, 3, 4, 5, 7, 12, 14, 15, 21
Cost	Method is not costly and has low time, data, and personnel requirements.	3, 4, 10, 12, 21, 22
Validity	Method is widely viewed as appropriate and is compatible with external conditions and existing institutions.	3, 4, 6, 12, 13
Participative Qualities	Method is participatory and democratic in nature or can be adapted for participation.	3, 9, 11, 18, 21, 22

References: 1: Chicken (1994), 2: Diez (2001), 3: Green (1997), 4: Gunton and Vertinsky (1990), 5: Gunton et al. (2004), 6: Hanley and Spash (1993), 7: Hochschorner and Finnveden (2003), 8: Hollick (1981), 9: ICPGSIA (2003), 10: Kok et al. (2006), 11: Laird (1993), 12: McAllister (1982), 13: Nash et al. (1975), 14: Noble (2014), 15: Park et al. (2006), 16: Rosenhead (2005), 17: Sadar (1996), 18: Samset (2003), 19: Soumelis (1977), 20: Tiruta-Barna et al. (2007), 21: Warner and Preston (1974), 22: Wood (2003).

Evaluation results

Suitability to context

A fundamental objective of impact assessment is to estimate the net impact of a project to determine whether a project is justified and in the public interest. CBA explicitly concerns itself with estimating net impacts by considering the full range of economic, environmental, social, and other types of impacts and by considering what otherwise would happen without the project (e.g., the opportunity costs of labour and other project inputs). No other method is capable of assessing such a wide range of impact types within the same methodological framework. In contrast, EconIA only considers a subset of impacts and ignores opportunity costs. This is illustrated in the example summarized in Table 2, which shows that the CBA estimates a net social cost of $2.5 billion (2012 Canadian) based upon the full range of project benefits and costs, while the EconIA estimates hundreds of billions in gross domestic product (GDP), employment, and government revenue ‘benefits’ based on the value of production without any deduction for project costs. The CBA examines environmental and other costs, while the EconIA provides no estimate of overall net project costs and benefits because it omits consideration of all of project costs from the analysis.

Table 2. Economic impacts of the Enbridge Northern Gateway pipeline project from the alternative perspectives of CBA and EconIA.^1.

Impact Category	CBA^2	EconIA^3,4
Project benefits	Net social cost of $2.5 billion (base case) based on net of project benefits and costs.	$311.5 billion in GDP ‘benefits’ plus employment and government ‘benefits’ (see below)
Employment	No incremental employment or labour income benefits due to opportunity costs of labour (base case).	$69.9 billion in labour income and 907,069 person-years of employment across Canada
Environmental impacts	$1.4 billion (base case) in air pollution, greenhouse gas, ecosystem service costs, passive use damages, and other unmonetized impacts.	n.p.
Government	$0.4 billion (base case) in royalties and taxes.	$98.3 billion in royalty and tax benefits
Other	Various unmonetized ‘sociocultural’ impacts including conflict and opposition, loss of subsistence use, and psychological stress	n.p.
Sensitivity analysis	Net social costs ranging from $7.4 billion to $0.7 billion.	No sensitivity analysis.

Notes: 1. 2012 Canadian dollars. 2. Net present value. 3. ‘n.p.’ means not provided as outside of scope of EconIA. 4. Undiscounted. Source: Broadbent (2014).

A second key feature of CBA is that it provides the means to assess project financial rationality. It is quite common to assume that major projects generate large economic benefits (Lockie et al. 2008; McLeod-Kilmurray and Smith 2010), yet this gospel of major project development is often based on exaggerated and/or misinterpreted estimates of benefits and underestimation of costs (Flyvbjerg et al. 2003; Gunton 2003; Joseph 2013). By definition, CBA assesses costs and benefits, but within the analysis one can test private financial viability (e.g., private internal rate of return) not just net social value. EconIA has no capacity to assess a project’s financial viability.

Third, CBA is concerned with impacts accruing to all individuals, which supports the democratic underpinnings of EA. EconIA, in contrast, has no particular concern for how all individuals may be affected by a project.

Fourth, CBA has important capabilities with respect to assessing sustainability, which is typically the key broader concern of EA. CBA is able to track environmental losses (Barbier et al. 1990; Williams 2019), including ecosystem constraints (Roan and Martin 1996), and assess the costs of depletion of natural capital (Knowler 2005). However, CBA is limited with respect to a ‘strong sustainability’ view which demands that at least some critical natural capital be sustained (Ayres et al. 2001; Gasparatos et al. 2008). EconIA provides no perspective on a project’s sustainability implications.

Fifth, CBA’s outputs relate directly to the commonly used EA criteria of impact significance, project justification, and public interest. The monetized magnitude of environmental impacts helps measure the degree of significance, and NPV results based on an aggregation of project costs and benefits inform justification and public interest decisions. As illustrated in Table 2, EconIA can only describe gross economic activity, which is of limited use in understanding the broader public interest considerations of a project.

However, CBA is not capable of assessing incremental economy-wide impacts that might be caused by a massive project, and so in such situations CBA (like EconIA) should be paired with methods of general equilibrium analysis (Pearce et al. 2006; Vickerman 2007). CBA is also unable to assess impacts on non-utilitarian values, such as impacts affecting human rights, but neither is EconIA. Another potential limitation of CBA is that it may conflict with the precautionary principle, a commonly accepted foundation of EA, as a positive NPV result implies that a project should go ahead despite risks. However, this can be addressed by a number of techniques such as using probability-adjusted values (expected values), developing alternative scenarios to test the impact of risk, and using variable discount rates (Almansa and Martínez-Paz 2011; Kolosz and Grant-Muller 2015). EconIA is unable to accommodate the precautionary principle or risk.

Flexibility

CBA is flexible in that there are many techniques and forms of the method that can be used depending on circumstances. For example, multiple account benefit-cost analysis (Shaffer 2010) and cost-effectiveness analysis can be used when certain impacts are difficult, controversial, and/or inappropriate to monetize, and non-market valuation techniques native to CBA can be used to assess particular impacts of projects (such as carbon emissions) without summing all of a project’s impacts into a NPV. Even qualitative CBA – in which the principles of CBA are applied without quantification – can be used in situations of short timeframes, limited budgets, or to frame EA thinking. EconIA, in contrast, has no particular variations.

Scientific robustness

CBA and EconIA both follow a series of logical steps that fit well with the rational model of decision-making and overall EA process. The internal logic of CBA rests on perhaps the strongest theoretical foundation of any method of impact assessment, perhaps even greater than that of EA itself (McAllister 1982; Pearce et al. 2006). CBA originates from the mid-1800s, first came into practical use in the 1930s in US water management, and by the 1950s had developed a strong theoretical and practical foundation (Major and Frederick 1997; Pearce 1998; Hanley 2001). This foundation can explain why some view CBA as an alternative to EA. EconIA also has a long history, with its origins in Keynesian economics and Leontief input-output modeling and, like CBA, has a strong theoretical and practical foundation (Davis 1990).

Over the past several decades, the limitations of CBA and EconIA have been debated in a variety of disciplines and refinements have been made to address limitations.

Many critiques of CBA revolve around its reliance on individuals’ willingness to pay (WTP) as a measure of value. While a hallmark of modern, neoclassical economics, WTP is critiqued because: (a) people have a willingness to pay for harmful things (e.g., tobacco), (b) WTP is compromised by peoples’ unfamiliarity with ecological goods and services, (c) the monetization underlying WTP encourages a consumer and thus individualistic perspective, (d) WTP is biased towards the wealthy’s greater ability to pay, (e) there is a divergence between WTP and people’s willingness to accept compensation for losses, and (g) WTP relies on individuals’ valuations today of the future as a reflection of future generations’ preferences (McAllister 1982; Sagoff 1988; Jacobs 1997; Hanley 2001; Horowitz and McConnell 2002; Gowdy 2004; Pearce et al. 2006; Revesz and Livermore 2008; Green 2009; Vatn 2009; Boardman et al. 2018). Some of these issues with WTP indirectly carry over to EconIA given this method’s reliance on money as a measure of project impacts.

Another issue with CBA has to do with discounting, which allows aggregation of project impacts no matter when they occur in time. This is important because future impacts are worth less than those today due to peoples’ time preference and ability to earn interest off investments. In this respect CBA has unique capability – EconIA ignores the fact that people discount, and we’re not aware of any other method of impact assessment that explicitly takes timing into account. There are challenges in discounting, such as from sustainability and intergenerational standpoints (Chichilnisky 1997), and sensitivity analysis using different discounting approaches and rates can lead to wide variation in CBA results. For example, the estimate of the overall project net benefits of a recent Canadian oil project was found to be highly sensitive to discount approach and rate: while the base case NPV was -$4.6 billion, the use of a uniform 3% discount rate resulted in an NPV of +$30 billion (Joseph 2018). Various resolutions of discount rate issues are proposed in the literature (e.g., Almansa and Martínez-Paz 2011; Weitzman 1998; Gowdy 2004).

A third issue with CBA concerns its focus on NPV and the notion that net benefits alone are sufficient to justify project approval. CBA relies on the Kaldor–Hicks criterion, which holds that a project is worthwhile if some people can be made better off if they are able to compensate those made worse off, but compensation is not required. Thus, if CBA outputs are not framed properly by assessing the distribution of costs and benefits – which is proper practice – decision-makers may be oblivious to important distributional consequences.

The key critiques of EconIA from the point of view of scientific robustness are its assumption that there are no constraints and thus no opportunity costs to project inputs, its reliance on economic impacts as an implicit indicator of project benefits, its omission of environmental and other types of impacts, and its tendency to ignore uncertainties in input parameters (Grady and Muller 1988; Gunton 2003; Kinnaman 2011; Williams 2019).

Minimal subjectivity

Subjectivity can be controlled through methodological directives contained in texts and guidance manuals, but many subjective judgements must still be made. In CBA, key subjective judgments concern who gets standing and what discount rate is appropriate (Stirling 1997; Broome 2008). Cognitive and interest biases may influence these judgements (Huszar 1998; Gunton 2003; Kammen and Pacca 2004; Nelson 2006; Flyvbjerg 2007). For example, optimism bias may affect judgements of key input parameters (e.g., project output prices or carbon damage costs) and special interests may seek to manipulate results by excluding standing to certain groups. However, these problems are present throughout impact assessment. EconIA is often blamed for excessive estimates of project benefits for promotional purposes (Matthews 1975; Hollick 1981; Hanley 2001; Anex and Focht 2002; Williams 2019). Government guidelines can help minimize bias by directing analytical decisions, but the scope of subjective judgments that arise in impact assessment is wide, and those who write guidelines can themselves inject bias. Impact assessment under all methods must be subject to close scrutiny (Ascher 1993).

Ease of comprehension

The essence of CBA is easily understood – add up the benefits and subtract the costs – and CBA’s reliance on the common metric of money helps, but beyond this CBA is not simple or easy to understand. Fundamental concepts in CBA like how project inputs and impacts are measured (opportunity costs and WTP, respectively) are not easily understood by people who have not studied economics (McAllister 1982). It is also difficult for many people to understand how one could generate a single final NPV result by examining a wide range of impact types. CBA is often thought of as a ‘black box’ (Hanley 2001); Gowdy (2004) contends that most economists do not fully understand it. Added to this is are various methodological debates related to discounting, treatment of uncertainty, valuation of non-market goods and services, distributional impacts, and sustainability that complicate the application of CBA and interpretation of CBA results, especially when there are a large number of scenarios based on differing assumptions and methodological options.

But many impact assessment methods have these same problems, especially those reliant upon quantitative models, including EconIA. All sophisticated methods have terminologies and degrees of opaqueness that impede understanding. Impact assessment is complicated and few methods are simple.

Usefulness of outputs

CBA is uniquely capable of assessing economic and other project impacts in a common metric, determining their relative significance, and synthesizing all of this information and interpreting its overall meaning to society in quantitative terms. CBA should not replace other methods of impact assessment that are typically used in an EA – impacts on flora and fauna, for example, should still rely on methods from the biological sciences – but CBA can be used to augment such methods and provide another perspective for decision-makers. McAllister (1982) questions any ‘grand index’ method of aggregating impacts on ‘apples and oranges’, but disparate impacts must be aggregated nonetheless, whether through CBA, opaquely through some internal exercise in decision-makers’ minds, or some other method (Price 2000). CBA at least provides one method of doing so. By comparing costs and benefits in a consistent methodological framework, CBA helps decision-makers understand and confront the trade-offs in project assessments. The outputs of EconIA, on the other hand, do not help decision-makers understand project trade-offs and make it difficult for decision-makers to reject a project: every project looks good through EconIA because EconIA simply tallies economic activity, including any activity linked to the project elsewhere in the economy, and calls these benefits while omitting project costs of economic or any other kind.

Second, though a CBA does not necessarily examine the distribution of impacts, good practice CBA entails doing so (Shaffer 2010). EconIA can also describe the distribution of impacts, but typically does so at a coarse scale in concert with the structure of the input-output models used. For example, input-output models used in EconIAs in Canada typically only disaggregate impacts from the national to provincial levels.

A third useful output of CBA is information on economic efficiency: NPV is what you get for what you put in. Efficiency matters because resources are limited, yet EconIA does not consider efficiency because it assesses outputs without the costs of inputs (Hanley 1992).

Despite all of this, a key criticism of CBA is that it does not actually address that which decision-makers are most interested (Crookes and de Wit 2002; Nelson 2006; Pearce et al. 2006; Vining and Boardman 2007; Atkinson and Mourato 2008; Boardman et al. 2018). Government decision-makers are typically concerned with local employment, regional development, stakeholders, and government budgets, and CBA as it is typically practiced only informs of these topics indirectly. In contrast, EconIA’s outputs provide decision-makers with information on economic development, regional employment, and government revenues (Kinnaman 2011).

Another common critique of CBA – but equally relevant to any method of impact assessment applied properly – is that a wide range of results may be generated when uncertainties such as discount rates are probed. Such variation can cause confusion and may be seen as a weakness (for example, see section 33 in JRP Frontier 2019), but the opposite is the case. Impact forecasting is filled with uncertainty, and any assessment that does not explore uncertainty does a great disservice to decision-makers and stakeholders by masking input uncertainty and the actual range of possible outcomes (Tennøy et al. 2006). EconIAs in Canada do not tend to explore the effect of uncertainty on results.

Cost

CBA can be expensive because it is complex and often difficult to do, especially if supplemental studies are required to estimate non-market costs and benefits. Higher cost may be an issue deterring its usage (Canter 1998; Crookes and de Wit 2002; Nelson 2006; Atkinson and Mourato 2008). In contrast, EconIA is relatively inexpensive as it typically involves gathering data from proponents on expenditures and labour needs and then contracting out input-output modeling to a government statistical agency for a relatively small fee.² Regardless, the value of a method of impact assessment lies in the usefulness of its results in relation to the costs to get the results. When one considers the information that CBA can provide and its low cost relative to the project being assessed, it may be a more cost-effective than other methods.

Validity with stakeholders

While CBA is widely accepted in the policy, planning, and legal communities, it is viewed less favourably by many stakeholders (Gunton 1992, 2003; Shiftan and Shefer 2009; Barget and Gouguet 2010; Shaffer 2010; Kinnaman 2011). Some feel that CBA is too flawed for use in its traditional form (McAllister 1982; Sagoff 1988); others simply don’t trust the method (Navrud and Pruckner 1997; Crookes and de Wit 2002; de Jong and Geerlings 2003; Adamowicz 2004; Atkinson and Mourato 2008). Negative opinions can be traced in part to the US experience where CBA has been used as an ‘anti-regulation weapon’ (Revesz and Livermore 2008; Sinden et al. 2009), even though it may also be a useful tool for pro-regulatory interests (DeMuth and Ginsburg 2010). Thus, despite CBA’s status around the world as an official method of evaluation, actual views on the ground are quite mixed and the method is used less in practice than one would expect (Adamowicz 2004; Atkinson and Mourato 2008).

In contrast, EconIA is widely accepted as evidenced in the method’s broad usage in EA processes around the world. In Canadian EA, EconIA is officially required (e.g., BC EAO 2015; CER 2020; ERC 1991; MVEIRB 2007).

Participative qualities

An undeniable strength of CBA is its grounding in individuals’ preferences, lending a democratic quality (Hanley 2001; Pearce et al. 2006; Shaffer 2010). The underlying principle is that each individual is best able to judge what contributes to their own welfare. This is an important foundation of CBA and this sets CBA apart from most methods in which stakeholders have no direct linkage to significance determination. CBA is conventionally a technocratic exercise, though, and its complexity is an obstacle for stakeholder participation. This does not mean that it is not possible to do CBA more democratically, as many have explored (e.g., Sagoff 1998; Soderholm and Sundqvist 2003; Howarth and Wilson 2006; Aslaksen and Myhr 2007; Beukers et al. 2015). We are not aware of any attempts to democratize EconIA, and most methods of impact assessment have no involvement of stakeholders beyond initial issue scoping.

Summary of evaluation

As a method of project evaluation by design, CBA is very capable analytically, provides the capacity to examine many of the issues raised in project-level EAs within one methodological framework, and is the most theoretically appropriate for the assessment of net economic impacts, but the method has some analytical limitations that demand caution and have been the subject of extensive critiques. In contrast, EconIA generates information that decision-makers find useful and in a format that is easily understood, and at relatively low cost, but EconIA’s capabilities are relatively limited in the EA context. A full summary of the evaluation results is presented in Table 3.

Table 3. Summary of evaluation of CBA and EconIA

Evaluation Criterion	CBA	EconIA
Suitability to Context	focused on net impacts capable of assessing many types of impacts and providing perspective on project rationale, distribution, sustainability, significance, public interest, and justification monetization of non-market impacts is challenging incapable of addressing non-utilitarian values capabilities to address risk, but results may conflict with precautionary principle	incapable of assessing net impacts scope limited to subset of economic impacts incapable of addressing non-utilitarian values little capacity to assess project rationale, distribution, sustainability, significance, public interest, or risk
Flexibility	multiple techniques and forms that can be used to address multiple analytical contexts	single form
Scientific Robustness	longstanding and perhaps greatest theoretical basis of any method of impact assessment critiques of WTP, the basis of valuation in CBA problems in discounting, but the only method concerned with time-value Kaldor–Hicks criterion ignores whether compensation actually occurs	long history of methodological development problematic assumptions, such as no constraints on project inputs
Minimal Subjectivity	both entail subjective judgments creating space for bias
Ease of Comprehension	both easy to understand on surface but a black box beyond this
Usefulness of Outputs	rare abilities to aggregate all impact data and evaluate a project as a whole, assess significance in quantitative terms, and offer a decision recommendation, though aggregation criticized despite fact that aggregation inevitable makes value judgments explicit informs of distribution of impacts, also unique ability to inform of economic efficiency standard CBA outputs not what decision-makers are typically interested in results can vary substantially depending on uncertain input parameters, like all quantitative methods	outputs are in terms useful to decision-makers provides some distribution information results would vary substantially depending on uncertain input parameters, like all quantitative methods, but uncertainty rarely explored
Cost	not inexpensive	relatively inexpensive
Validity with Stakeholders	CBA viewed as essential by some, rejected or viewed negatively by others governments mandate it in some contexts, ignore it in others	well accepted and widely used around the world
Participative Qualities	based upon values of all individuals that are given standing, but is typically carried out technocratically divorced from stakeholders participative forms of CBA are emerging	no participative qualities

Discussion

CBA is capable of providing a comprehensive assessment of project costs and benefits in a consistent framework that allows for a comparison of trade-offs and determination of whether a project generates a net benefit to society. The ability to assess costs and benefits and to estimate the net benefit to society is a unique strength of CBA that other methods of impact assessment in EA lack. This can be traced to CBA’s origin as a method of project evaluation and not simply as a method of economic impact assessment.

EconIA forecasts the economic impacts of a project, and this information on regional growth can be used to support community planning. The problem, however, is that EconIA is often misapplied in EA as a method for assessing project economic benefits. EconIA does not provide an accurate measure of project benefits because its scope of impact coverage is limited and it ignores opportunity costs, leading to an exaggerated estimate of project economic benefits that is often inappropriately used to skew decision-makers’ in favour of project approval.

The key conclusion from this evaluation is that both CBA and EconIA have important but different roles to play in EA and should be viewed as complementary methods. Consequently, the current standard EA practice in Canada and other jurisdictions of relying primarily on EconIA to estimate project economic impacts needs to be reformed to include CBA so that a broader assessment of project costs and benefits is undertaken and a project’s justification and net benefits to society can be better assessed. Critics of CBA emphasize the method’s limitations, but considering the limitations of the mainstream method of EconIA, the appropriate course of action is not to reject a method and forgo the important information it provides but to address these limitations through the application of best practices. CBA should be viewed as a tool to provide information to decision-makers and not a rule that decision-makers are obligated to follow (Campen 1986). Furthermore, EconIA’s results must be understood for what they are: estimates of a subset of economic impacts and not a measure of overall project benefits. Both CBA and EconIA have strengths and limitations, and the limitations need to be addressed by applying best practices, but both should be used to inform EA.

Conclusion

The evaluation in this paper shows that both CBA and EconIA provide important information for EA. CBA can be used to help identify trade-offs and inform decision-makers and stakeholders with respect to a project’s contribution to the public interest, while EconIA can be used to estimate economic impact information necessary for regional planning. CBA and EconIA are complementary methods that provide different information to decision-makers. EA practice can benefit in many ways through the inclusion of CBA and its perspective on a project’s overall contribution to the public interest. CBA is most capable for assessing project benefits, and EconIA results must not be misinterpreted as a measure of project benefits. While both methods need to be applied following best practice guidelines to address their respective limitations, failure to reform EA practice by continuing to eschew CBA for EconIA risks forgoing essential information necessary for rational EA decisions.

Acknowledgments

This manuscript was initially developed during the lead author’s PhD research, and accordingly we would like to thank Sean Markey and Kevin Hanna for helpful comments. Thanks also to anonymous referees for helpful comments.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This work was supported by the Social Sciences and Humanities Research Council of Canada [Canada Graduate Scholarship – Doctoral 2006-2009].

Notes

1. We use the term ‘economic impact analysis’ to refer to the method of EconIA and the term ‘economic impact assessment’ to refer to the act of assessing economic impacts through EconIA or other methods of analysis. Likewise, we use the acronym EconIA to avoid confusion as the acronym ‘EIA’ is commonly used to refer to ‘environmental impact assessment’.

2. In 2019, the lead author contracted Canada’s national statistical agency to run their input-output model for an EconIA at a cost of only $2,100 CDN. This compares to budgets of $50,000 CDN or more for CBAs that the lead author has undertaken.