Flood prevention schemes: guidance for local authorities

SECTION 6: CHOOSING THE OPTION

Introduction

6.1 Flood prevention schemes made by local authorities and submitted to the Scottish Executive for grant assistance must provide value for money. The funds available for such assistance are set with reference to the schemes in the Scottish programme, and to the demands of other programmes. Consequently, the funds are not limitless.

6.2 Historically, the number of schemes has averaged less than 2 per annum. With such low numbers, and the comparatively low level of capital expenditure, it has been possible to manage the grant scheme within available resources. However, the flood defence programme has been expanding, thereby increasing the pressure to obtain best value for money for the programme as a whole.

6.3 A practical way of ensuring best value for money, which may be applied on a project-by-project basis, is to maximise the benefit-cost ratio of those schemes being funded while seeking to achieve an appropriate standard of protection for the type of land in question. The procedure for achieving this, the decision process, is set out below.

Standards

6.4 To qualify for grant assistance, it is expected that flood prevention schemes in Scotland will continue to be designed to withstand, at least, a 1 in 100 year flood event. As noted in paragraph 3.8, this guideline standard is not intended to prejudge or otherwise constrain the appraisal process. It does not preclude the adoption of a higher standard where justified, nor indeed a lower one. However, experience over many years in Scotland has shown that a 1 in 100 year standard is appropriate for the protection of non-agricultural land. Such a standard reflects reasonable public expectations in relation to residential and other urban property. Further, it lies within the return period range of 50-200 years, which is the indicative standard in England and Wales for the protection of intensively developed urban areas against fluvial flooding. Against that background, a 1 in 100 year standard is a reasonable starting point, and provides a practical benchmark to assist with the administration of the grant scheme. This does not negate the need to appraise different standards in addition to the guideline value.

The decision process

6.5 The objective of this section is to guide flood defence practitioners in the choice of the most appropriate scheme option. It is appropriate only for the consideration of alternative standards of defence. It assumes that alternative design options will be separately considered to determine the least whole-life cost to achieve any particular design standard. In addition, it should be noted that the method is dependent on the sensible determination of discontinuities in both the benefit- probability and cost-probability curves (section 4.3).

Steps in decision-making

6.6 The starting point is the concept of benefit-cost ratio, both overall and on an incremental basis. The maximisation of this ratio often indicates the most economically worthwhile scheme but should be considered in the following steps to take account of reasonable expectations of standards of protection.

(1) Examine the benefit-cost ratios of all options. If none is at least unity, reconsider the scope of the options and recommence this process, or abandon the proposal.

(2) Identify the option with the highest benefit-cost ratio. The standard of protection for this option will fall below, at or above the guideline of 1 in 100 years. Where it falls below, consideration can be given to alternative options to come closer to achieving the guideline standard provided the incremental benefit-cost ratio is robustly greater than unity. Consideration can be given to options up to an indicative maximum of 1 in 200 years, provided the incremental benefit-cost ratio is competitive with that likely to be achieved if the additional costs were invested in other schemes. Any further increase in standard should only be considered if the additional benefits by far outweigh the additional costs. Such considerations should follow the steps set out at (3)(a), (3)(b) and (3)(c) below.

(3) The aim of this step is to provide best value for money while achieving the most appropriate standard of defence.

(a) Scheme with maximum benefit-cost ratio lies below guideline standard of 1 in 100 years.

Consider the option with the next higher standard and determine whether its incremental benefit-cost ratio is robustly greater than unity. If this is the case, proceed to the next higher standard option. Continue the process until either:

• the incremental ratio is no longer robustly greater than unity, at which point the process ends; or
• until the guideline standard is reached. Then proceed to (3)(b).

(b) Selected scheme lies at or above the guideline standard but below the indicative maximum.

Consider the option with the next higher standard and determine whether its incremental benefit-cost ratio compares with other nationally funded schemes. If this is the case proceed to the next higher standard option. Continue the process until either:

• the incremental ratio is no longer comparable with other nationally funded schemes, at which point the process ends; or
• the indicative maximum standard is reached. Then proceed to (3)(c).

(c) Selected scheme up to this point exceeds the indicative maximum standard.

Consider a higher standard scheme only if this yields an incremental benefit-cost ratio that is exceptional and, certainly, significantly higher than the benefit-cost ratios of independent schemes competing for funds.

6.7 It should be noted that incremental benefit-cost ratios can be sensitive to the choice of interval and spuriously high or low individual values should not be allowed to mislead. In each case it is important to compare 2 viable options. Further, where the best scheme choice is dependent on incremental ratios that are marginally greater than unity, a sensitivity analysis might be required to assess whether the choice would be worthwhile.

6.8 The recommendations arrived at from the above process may well be the final choice. However, it is recognised that benefit-cost analysis is but one tool available to aid decision-making. An authority may be justified in proceeding with an alternative where other exceptional factors influence choice. These could include:

• uncertainty regarding the economic outcomes of a particular option which it has not been possible to incorporate adequately into the preceding analysis (see Chapter 6);
• environmental considerations for which it has not been possible to assign monetary values;
• development plan policies and other material planning considerations;
• availability of funds (affordability).

6.9 It should be emphasised that the exercise is not about the manipulation of figures. Rather, what is required is a reasoned interpretation of the information produced from the appraisal.

Sensitivity analysis and robustness testing

6.10 Within economic appraisal, the purpose of sensitivity analysis and robustness testing is to determine whether, within the reasonable bounds of confidence, and for the various assumptions made:

• the scheme is economically worthwhile;
• the economic return is likely to be achieved;
• the option choice is robust.

6.11 It is therefore important to determine those factors that would impact on any investment decision. If, for example, the benefit-cost ratio is highest for an option where there is significant uncertainty, it may be better to pursue an alternative with a lower, but more certain outcome. Chapter 6 incorporates a framework for evaluating such risks.

6.12 Having determined the most important factors, assessments of uncertainty should be made on the basis of experience and judgment. As a general guide, a range of possibilities should be considered for items such as:

• costs (whole-life capital, maintenance and management);
• threshold of flooding (many schemes will be sensitive to assumptions about the level, and hence frequency, at which flood damage commences);
• calculation of extremes and their probabilities;
• changes to major beneficiaries (for example, a major business in the benefit area could cease trading or relocate).

6.13 Even where it is not possible to quantify the uncertainty associated with each variable, it should be possible to identify a reasonable range of possibilities. All major risks should be considered both singly and in combination.

6.14 For major projects it is particularly important to identify switching points where a change in the assumptions would alter the option choice. Informed judgments can then be made of the relative likelihood of the different outcomes, to determine the best option.

Figure 6.1 The decision process