Once options have been identified in broad terms, or a shortlist of possible low-regret options has been identified, it is possible to use appraisal to assess them in more detail. This can include the analysis of the costs and benefits of options. This is particularly relevant when moving to the project level. However, the type of decision-support tools for adaptation vary with the application. The key difference is between immediate no- and low-regret options, and those that involve longer periods and thus uncertainty.
No- and low-regret options. In methodological terms, no- and low-regret options can be assessed using conventional economic appraisal, as they are focused on options that have almost immediate benefits (and thus discounting is less of an issue or not relevant). They can therefore be considered using similar approaches to conventional fisheries and aquaculture economic appraisal. The most commonly used methods are CBA, cost-effectiveness analysis (CEA) and multicriteria analysis (MCA). These are often applied in fisheries management (Activity 3.3: Management option evaluation and selection, FAO EAF toolbox [FAO, 2011-2019]) – as summarized in Figure 3 and Box 4
These tools vary in the way they address the methodological issues raised in Chapter 3 relating to discounting, equity, uncertainty and non-monetary values. These conventional decision-support tools are in particular relevant for early adaptation, where the decision lifetime is short and the focus is on delivering early benefits. However, as highlighted in Chapter 3, many fisheries and aquaculture adaptation options are associated with socio-economic or non-market benefits, or involve areas that are more challenging to quantify in terms of benefits (e.g. capacity building). For this reason, they are often considered using extended cost-benefit approaches, or with decision-support methods that can include qualitative as well as quantitative aspects, such as MCA. It is also possible to use conventional CBA and test for unknowns by using switching values,14 for example, to assess how large the benefits would need to be to justify the costs of the intervention, and then to assess qualitatively how likely it is that the project or investment could achieve this benchmark.
Longer-term decisions. For options that involve longer-term decisions, i.e. beyond the no- and low-regret options above, a more detailed set of appraisal methods are applicable. These are often termed decision-making under uncertainty (DMUU). These methods are more focused on options appraisal, particularly at the project level, and they involve a set of more formalized approaches to address the uncertainty issues identified in Chapter 3. The main approaches are summarized in Figure 4, with more information included in Box 5.
Decision-making under uncertainty
Adaptive management is an iterative cycle of monitoring, research, evaluation and learning that is used to improve future management strategies.1 It is a process rather than a tool. The approach is relevant for adaptation given the high uncertainty and the long lifetimes, and is sometimes referred to as an adaptation pathways approach. One variation of the approach is to use thresholds (biophysical or policy) that trigger changes in adaptation options or policy (adaptation-tipping points).2 These are sometimes presented as adaptation route maps,3 also termed dynamic adaptation policy pathways.4 These approaches are not formal economic methods, but they can include extended cost-benefit analysis. These adaptive management approaches have very high relevance for the marine fisheries sector, where there are likely to be shifts over time, but where uncertainty is large.
Real option analysis quantifies the investment risk with uncertain future outcomes. It is useful to consider the value of flexibility over the timing of an investment, or the adjustment over time in a number of stages, in response to unfolding events. This allows for consideration of flexibility, learning and future information (option values). In the adaptation context, it can be used to assess whether there is a value to waiting for (climate) uncertainties to be resolved to avoid negative outcomes, of whether it is beneficial to invest in more flexible adaptation solutions that can be changed later. It involves extended cost-benefit analysis, but does require probabilistic-type information to work. The approach can be applied as a formal economic method for adaptive management.
Robust decision-making is a method premised on robustness rather than optimality.5 It involves testing options or strategies across a large number of plausible “futures” to identify which perform well over the range, rather than optimally to one central scenario. It can be used in cases of deep uncertainty (when there is no probabilistic information). Some studies test options against climate change only, while others examine wider futures that also consider socio-economics and policy.6 The approach does not involve economic analysis per se, but most studies include costs, and some cost-benefit analysis. It has potential for fisheries, although there do not appear to be applications to date in the literature.
Decision scaling is an approach that links bottom-up vulnerability assessment with multiple sources of top-down climate information.7 It identifies performance indicators and acceptable thresholds, and assesses the performance of the performance indicators to the current climate to develop climate response functions. It then uses multiple futures (multimodel climate information) to stress-test performance. The approach has been used for adaptation, in particular for water and hydropower investments. It does have some applicability for fisheries investment decisions, notably through the use of key fisheries performance indicators, although to date there have been no applications.
Portfolio analysis provides a quantitative way to maximize the return on investments using a portfolio. The principle is that spreading investments over a range of asset types spreads risks. Portfolio analysis highlights the trade-off between the returns on an investment and riskiness, and can maximize the expected rate of return and minimize the total portfolio variance. For adaptation, it can select combinations of options that together are effective over the range of possible future climates.8 It uses an extended costbenefit analysis framework. There are studies9 that look at restoration/regeneration of natural systems (forests), and it is possible to see similar applications for fisheries, notably with the risk of species migration (and uncertainty).
Rule-based decision support involves a set of decision rules or criteria that can be used for decision-making under uncertainty. These include:10 the minimax regret rule, which is a cautious decision-support criterion and approach where the decision maker aims to minimize the maximum regret; the maximax rule, which is an optimistic decision-support criterion and approach in which the decision maker opts for the option with the highest possible outcome; and the maximin rule, which is a pessimistic decision-support criterion and approach in which the decision maker aims to maximize the minimum outcome.
Recent years have seen the growing use of these methods for adaptation. The first application relates to when there are early decisions that have a long lifetime, and it is possible to include adaptation in this early decisions to reduce future climate risks. This is primarily associated with decisions such as infrastructure investment. In this case, the main objective is to make decisions under uncertainty. The second main application relates to the use of iterative approaches to address the long term (i.e. mid-century climate impacts) as part of adaptation pathways, enabling learning and changes over time in response to the evidence. Note that in this case the decision is being made later in time (unlike the first application). In some cases, these elements are combined, i.e. with methods that look at design as part of longer-term iterative approaches.
A review of the academic and grey literature (ECONADPT, 2017) identified about 50 DMUU adaptation studies that included economic analysis. However, to date, there has been little application of DMUU in the fisheries sector.
There is a large body of DMUU applications for coastal investment and protection. Groves and Sharon (2013) applied robust decision-making to planning coastal resilience for Louisiana, the United States of America. There have also been several applications of real options analysis (ROA) to coastal protection. For example, Scandizzo (2011) applied ROA to assess the value of hard infrastructure, restoration of mangroves and coastal zone management options in Mexico, concluding that ROA highlights the value of gradual and modular options. There have also been applications to port infrastructure, using adaptive management approaches, for example, with the International Finance Corporation port study in Cartagena, Colombia (IFC, 2011), and examples of flexibility in port design for the port of Avatiu in Cook Islands (ADB, 2014).
There have been some applications to fisheries directly. Wellman, Hunt and Watkiss (2017) undertook a cost-benefit study and used adaptive management (with some light-touch ROA) for seaweed farming in Zanzibar, the United Republic of Tanzania. This considered various adaptation options to address the problems of increasing sea surface temperature and impacts on near-shore seaweed farming productivity. The analysis considered three options, growing different species in deeper water using floating rafts, as well as a longer-term iterative programme to gather information on temperature changes around islands, for use in long-term strategic decisions based on likely climate scenarios. There has also been an application of ROA to better study climate information in relation to coral protection and regrowth options, in response to deep-water fishing and aragonite saturation horizon shoaling, and acidification, and their effects on the extent and quality of cold-water Lophelia reefs in the Northeast Atlantic (in their role in providing a highly productive habitat for a number of fish species). The ROA element comes from the potential learning over time in the decision-making process, and the prospect that new information will become available on the impacts, and benefits of options to address these impacts, for these reef systems (Jackson et al., 2013).
There are some fisheries studies that have used adaptation pathways thinking, including adaptation turning points. Werners et al. (2013) analysed fish stock maintenance under climate change with an adaptation pathway approach, looking at salmon reintroduction in the Rhine River (although this did not include valuation). It is highlighted that longer-term iterative adaptive management is considered highly relevant for the fisheries sector, including fisheries policy, because it allows a cycle of monitoring and research to help build the evidence base on emerging climate impacts, and in turn, to inform future fisheries management. This includes investment in biophysical monitoring (sea surface temperature, acidification levels, etc.) as well as monitoring of fish species and distribution, complemented with modelling analysis. It also includes early research into potentially major long-term impacts (acidification). Watkiss and Cimato (2019) undertook a very initial economic analysis using such an approach, looking at fisheries in the United Kingdom of Great Britain and Northern Ireland, which indicated positive benefit-to-cost ratios.
Applying DMUU, at least when using the formal methodologies, tends to be a time- and resource-intensive process, requiring significant technical expertise. These techniques are complicated to apply even where data are good, and thus very challenging to apply in the developing country context (see Bhave et al., 2016). Indeed, many of the applications of DMUU to date are theoretical in nature and involve stylized examples rather than real project investments. This limits their formal application to projects with the necessary resources (i.e. larger projects). However, the concepts of these approaches are extremely useful, and they can be used in simpler applications more generally. Indeed, there is a growing focus on developing “light-touch” versions of these methods for more general application.