Introduction • All impacts in money terms • No impacts in monetary terms • Some impacts in money terms • Risk & uncertainty

No impacts in monetary terms

When none of the impacts have been presented in monetary terms, the choice of preferred option requires some degree of judgement. To retain transparency in decision-making it is important that these judgements are recorded. There are a number of recognised techniques available that allow judgements to be made and recorded when selecting between options whose impacts are presented in different units. Two methods are described here:

• Pairwise comparison: this provides an approach that is best used where there are only a small number of options and criteria across which a decision has to be made; and
• Ranking techniques: these techniques provide an approach that can be used across all levels of decision, but are most useful where there are a larger number of options and/or criteria to compare.

Pairwise comparison

Pairwise comparisons are often used as a means for conveying information to decision makers on the degree to which one option outperforms another across a range of decision criteria. No attempt is made to incorporate any judgements as to the relative importance of different magnitudes of impact or of the different criteria.

The first stage in undertaking pairwise comparisons involves listing the criteria or impacts and comparing options in pairs against each of these, indicating a preference for one option over another. The results are then recorded in a table, (see below for example), to illustrate which alternative performs better or worse for each of the criteria. An overall preference is then identified, or the information is used to highlight the trade-offs involved in selecting one option over another. Ultimately, the information is provided to decision makers who must make a judgement on the relative importance to be assigned to the different criteria and, thus, to determine the best option.

Although this approach is readily applied to problems with only a few options or criteria, undertaking the comparisons and ensuring consistency becomes increasingly complex as the numbers of criteria and options increase. It is unlikely to be useful, therefore, above the ‘management’ decision level.

Ranking involves the ordering of options or impacts into ranks using verbal, alphabetical or numerical scales and provides an indication of relative performance. Value judgements (e.g. expert or a decision maker’s opinion) are used to decide on the order of preference for different options or impacts. So, for example, if there were five options and a numerical scale was being used, the ‘best’ option would receive a ranking of 1 and the ‘worst’ a ranking of 5.

This method obviously provides a simple means of evaluating the performance of different options over a range of different criteria. However, when used on their own, they provide little information on the degree or magnitude of any differences in impact between options. They, therefore, hide any uncertainty that may exist as to the extent of such differences. In addition, when there are several options under consideration, it may be difficult to select a preferred option. This latter problem has led to the tendency for people to add ranks (or trends) together, a mathematical operation which is invalid unless it is assumed that: the decision makers will place an equal value on impacts falling under the various criteria (i.e. that impacts on consumers are equally important to changes in environmental risks); and that all trend scores or ranks reflect proportional changes in level of impact (i.e. +++ is three times better than +).

Such methods must, therefore, be backed up by further descriptive information if decision makers and others are to be provided with an accurate picture of the implications associated with alternative saltmarsh management options or, more broadly, flood and coastal defence options.

Ranking may be more useful at the ‘restoration’ and ‘realignment/compensatory habitat’ levels of decision, although it may become complex where the options provide very different levels of benefits.

Click here for an example of a ranking exercise (based on an Appraisal Summary Table) for three different options for restoration works.

For the above example each option is ranked. The outcome of which is that Option 1 is not better than Options 2 and 3 for most categories. Only in terms of ‘soil provision’ does it outrank Option 3. This means that the decision will be between Options 2 and 3. Option 3 only outranks Option 2 in terms of ‘knowledge’ (environmental monitoring, educational resource, and natural science research). Option 3 is ranked equal to Option 2 for ‘distant appreciation’, ‘social development’ and ‘global life support’.

At this point, it is necessary to consider the costs. Option 2 has costs of £1.5 million while Option 3 has costs of £1.0 million. This means that the decision-maker has to decide if the additional benefits that Option 2 presents over Option 3 are ‘worth’ £0.5 million. However, Option 3 has additional benefits over Option 2 in terms of ‘knowledge’. Therefore, it is also important to consider which the most important benefits are. In this case, it is known from discussion with local stakeholders that ‘ecosystem services’ are most important feature (and are the driving forces behind the decision to ‘do something’). Option 2 would provide benefits over Option 3 in terms of ‘flood and erosion control’, ‘soil provision’, ‘pollination’, and ‘habitat provision’. However, for the purpose of this example, the decision maker determines that the additional benefit of Option 2 over Option 3 is uncertain and does not consider this to be worth the additional £0.5 million. Hence, Option 3 is selected as the preferred option.

Where no impacts have been presented in money terms, it is not possible to select an option based on numeric ratios, so judgments have to be made. All such judgments should be justified and a record of the justification should be included with the preferred option.