The Scottish Government

Contents |

Listen

CHAPTER 8: ECONOMIC ASSESSMENT METHODOLOGY

Introduction

8.117 Detailed economic assessments were carried out for 33 of the Category B schemes. A summary of the results is available in Appendix C. Once detailed modelling had been carried out, an economic appraisal of the scheme was undertaken for each of these schemes. The economic appraisal assessed the schemes in terms of economic performance and value for money.

8.118 The methodology employed used existing tools to carry out an economic appraisal for each scheme. The 'Modelling and Decision Support Framework' ( MDSF) was used to derive damages in a consistent procedure using the Flood Hazard Research Centre ( FHRC) depth damage curves ¹⁷. The economic appraisal was carried out using the DEFRA Flood and Coastal Defence Project Appraisal Guidance ( FCDPAG) ¹⁸ and standard calculation tables adapted for use in Scotland.

MDSF methodology

8.119 The 'Modelling and Decision Support Framework' ( MDSF) system was used to determine the economic damages for the pre-scheme (undefended) and current post-construction (defended) cases.

Introduction

8.120 The MDSF is a support tool that facilitates a common approach and tools that can deliver efficient procedures in aspects of floodplain mapping and economic analysis. The MDSF was originally developed for DEFRA and the Environment Agency to assist the Catchment Flood Management Plans and other medium to large scale flood management studies. The primary use of the MDSF for this study was to generate an assessment of the direct economic damages for the pre-scheme and defended modelled scenarios for a range of flood return periods (5, 10, 25, 100, 200, 1000 years as a minimum).

8.121 In order to derive an indication of how each scheme performs, damages for the undefended case have been compared against those for the defended case.

Datasets

8.122 Certain core datasets are required by MDSF in order to carry out the analysis. Standard datasets have been used throughout in order to maintain a consistent approach. However, in some cases, due to the small scale analysis and availability of data, improved information has been utilised: this is summarised in the sections below.

Digital Elevation Model

8.123 The MDSF relies on a Digital Elevation Model ( DEM) to provide ground elevations with which to derive flood outlines and property flood depths. The NEXTMap 5 m grid Digital Terrain Model has been used for most sites as it is available nationally and provides a relatively accurate ground level.

8.124 LiDAR data has been used where available as this provides improved elevation accuracy. This was used in two sites in the Glasgow area (Crosslee and Brock Burn and Levern Water).

8.125 In some cases where applicable to do so, the DTM was disregarded and a new elevation grid derived from survey data. This was usually only carried out at very small sites were the DTM showed significant error and sufficient data was available to derive a new grid. This approach was undertaken at a number of sites including; Port Logan, Skiprunning Burn, Bridge of Allan and Millfield of Cupar.

Property dataset

8.126 A property dataset required for use in MDSF must contain as a minimum a grid reference and a code to link the property type to the depth damage curves. The MDSF typically uses the National Property Dataset which does not include properties in Scotland. The property data for each site were therefore derived manually based on the AddressPoint dataset, surveyed threshold levels and property type and valuation compiled from the Scottish Assessors Association website ( SAA.gov.uk).

8.127 The MDSF application links the derived depth data with the FHRC's Multi-Coloured Manual ¹⁹ ( MCM) data to provide a damage estimate for a given return period. This is done automatically for the return periods selected. The use of MCM data based on properties in England and Wales was deemed appropriate to Scotland in absence of any comprehensive Scottish data.

8.128 This damage estimate is refined by using the surveyed threshold level of properties, in order for the difference between the DTM ground elevation and the property threshold to be calculated. Should these values be omitted, then the depth of property flooding is assumed to be the same as that in the modelled depth grid. In the case where threshold levels are omitted, the property floor levels have been raised by a standard value (typically 300 mm) in order to limit overly conservative damage estimates.

8.129 MDSF simplifies the damage calculation for residential properties by grouping all residential properties into the same code and calculating damages using a single depth-damage curve. This is an average depth damage curve for all properties that does not distinguish between property ages, types or size. This is a simplification typical of a broad scale approach and may result in differences when compared to previous analysis.

8.130 Additional fields in the property dataset that improve the calculation of commercial and industrial properties are 'PropertyArea' and 'Valuation'. The area of properties is used to improve damage calculations for commercial properties as these can vary widely. Without the inclusion of this parameter, the national average values are used by the MDSF. The 'Valuation' field is used to provide a market valuation for each non residential property. This is obtained to make a comparison against the total damages avoided. If the total damages avoided are greater than the market value then, they are capped at this value as recommended by Defra FCDPAG guidance.

Calculation methods

8.131 The MDSF can derive damages from a number of flood level dataset types including:

• Modelled flood elevations - Flood elevations as supplied by HEC- RAS, ISIS and InfoWorks RS and CS modelling packages. These provide flood levels for each cross section for a range of return periods.
• Flood depth grids - Flood depths as supplied by the JFLOW 2-dimensional floodplain modelling package. These results from JFLOW can be exported as depth grids and imported directly into MDSF.
• Flood outlines - This approach was not used for this study.

8.132 In certain cases where few properties were at risk and did not warrant a full MDSF analysis, the calculations were undertaken manually. This was also used for properties that were known to flood but were outside the main area protected by FPS's, or cases where one analysis method was used predominantly but was unsuitable for isolated properties.

Outputs

8.133 Outputs for this analysis are an assessment of the number of properties at risk and damage estimates at each return period for the defended and undefended scenarios. The damage estimates are used in the project appraisal calculations to determine the damage avoided by the scheme and the benefit-cost ratio.

Project appraisal methodology

Introduction

8.134 The project appraisal methodology has utilised the methods provided in the Defra FCDPAG with adaptations to take into account the Scottish legislation and confirmed during the project by Scottish Executive. This is in line with recently published guidance from chapter 5 of the Flood Prevention Schemes Guidance for Local Authorities ²⁰. The work calculation sheets from Defra have been used as the standard templates for this study.

8.135 The method relies on the benefit-cost method of appraisal in order to take into account the quantifiable costs and benefits of the scheme and determines the ratio of the present value of benefits to the present value of the costs.

Assumptions

8.136 Some general assumptions have been made to the appraisal process in order to simplify the procedure and maintain a consistent approach to the benefit cost assessment. These include:

• The present value costs of the scheme ( PVc) have been updated to 2004 prices. The cost of maintenance has been discounted over the 100 year scheme life and added to the cost of construction to give Present Value of costs.
• Damages have been determined using the data issued in conjunction with the Multi Coloured Manual ( MCM) ²¹ (January 2003) issued for Defra by the Flood Hazard Research Centre ( FHRC).
• Allowance for the reduction in costs incurred by emergency service organisations as a result of the scheme has been taken into account as a percentage (10.7%) of the total economic property losses, as recommended by the MCM.
• Unless actual figures were available maintenance was based on the assumption that the annual maintenance costs are 1.5% of the cost of construction. Where replacement is required i.e. gabion replacement/repair or pump replacement, sums have been included at the appropriate interval (20-25 years).
• No allowance has been made in the benefit cost analysis for intangibles or indirect losses such as the cost of clean up, temporary accommodation and traffic related damages/disruption.
• Optimism bias has been set at zero as the schemes have already been built (this factor is used to offset overly optimistic cost estimates).
• Discounting of damages and scheme costs have been calculated using the revised Treasury discount rate of 3.5% for the first 30 years, 3.0% for the next 40 years and 2.5% thereafter, as recommended by the Green Book ²².
• The total benefits, measured as damage avoided, have been capped (if necessary) at the capital values for all properties (residential and commercial) within the undefended 100 year boundary.
• The total capital damage values have been derived by combining industrial and commercial property rateable values (updated to capital values using the equivalent yield), with the capital value of residential properties flooded by the 100 year event.
• Residential property capital values are assumed to be the median valuation for a property within either the town or city within which it lies or, if too small, within the Local Authority area within which it lies.
• No assumption has been made for the age of the scheme components, with all schemes assessed with the assumption that they will last the appraisal period of 100 years.

Scheme costs

8.137 The construction costs of 54 schemes were available and in order. These costs were updated to reflect the cost of the scheme construction today. These have been updated using the Construction Cost Index ²³ and to a lesser extent the Retail Price Index ( RPI), to establish an estimate of the present day cost of each scheme. It is thought that the former provides a better estimate as construction costs have outstripped the RPI for the last two decades ²⁴.

8.138 Construction Cost Indices have factors that must be taken into account that other indices such as the RPI (Retail Price Index) or CPI (Consumer Price Index) do not. These are to do with the nature of the construction industry, and include factors such as the heterogeneity of construction projects, non-qualified items, and the long duration of construction projects, which can be months or years.

8.139 Approximately £62 million was spent on these 54 schemes at time of construction, which updated to present day costs is equivalent to £82 million. The more expensive schemes such as Perth, Kilmarnock or the Kelvin have been undertaken recently and represent a much higher expenditure than previous schemes.

8.140 Table 8-1 shows a breakdown of local authority expenditure at time of construction and updated to present day (2004) values.

Table 8-1: Outturn costs updated to present day (2004) value by local authority

8.141 In order to compare the costs against the benefits, the Present Value of future maintenance costs has been applied to construction costs for the benefit cost analysis. The impact of these additional costs is shown in Table 8-2 below.

Table 8-2: Comparison of scheme costs for sites studied

Values based on the 54 schemes with costs available and the 39 schemes assessed in terms of properties protected.

8.142 Table 8-2 indicates that the net cost per property (the total costs divided by the total number of properties protected is in the order of £18,000. The individual costs per property for each scheme have been reported in each modelling report and are listed in A ppendix C, illustrating a significant variation in these values.

8.143 Figure 8-1 illustrates the number of schemes implemented and the total present value of scheme construction costs for each 5-year period since 1961. It shows a significant increase in the number of schemes constructed in the 1990's. The present value of scheme costs shows a significant increase in the latter half of the 1990's, representing schemes that protected more complex areas to greater standards of protection.

Figure 8-1: Schemes constructed and total cost of scheme by five year period

Contents |