What is important cannot be measured?
According to a famous expression¹, the most important things cannot be measured. This seemed also to be true for biodiversity in cost-benefit analysis, in particular for the impact on non-economic benefits of biodiversity. However, this is not true anymore, as for over a decade in the Netherlands a methodology known as biodiversity points is being applied for this purpose (see Bos and Ruijs, 2021). Biodiversity points are quite similar to the quality-adjusted life years (QALY) used for cost-effectiveness analysis of health care treatments. Biodiversity points provide a quality-adjusted measure of the changes in the quantity of biodiversity. It is not based on the preferences and information of consumers or citizens, but is based in a standardized way on the expert-opinion of ecologists. The unit of measurement is not dollars or euros but is the number of biodiversity points.

For many cost-benefit analyses (CBA), properly assessing the welfare effects of a policy measure on biodiversity is important. This does not only apply to CBA on conservation or stimulation of biodiversity, but also to CBA on other policy areas such as mobility, agriculture and water safety, as the policy measures in these policy areas often have impacts on biodiversity. For example, a new road connecting two cities through a forest is good for mobility but has also impact on the value of biodiversity. The value of the forest for various economic uses may increase due to the reduction in travel time for visitors but the non-use value, e.g. the existence value and the value for future generations of the forest and the biodiversity of its species, may be affected severely by fragmenting the forest and by increasing traffic, pollution and visitors.

Biodiversity in Dutch CBA
The Netherlands has a long tradition in cost-benefit analysis, even going back to 1901². The way biodiversity has been incorporated in Dutch CBAs has changed drastically over time: from CBAs in which major impacts on biodiversity were not even mentioned to CBAs in which the impact on ecosystem services are valued as much as possible and effects on the non-use value of biodiversity are measured by biodiversity points.

Up to the 1970s, CBAs in the Netherlands mainly pertained to major investments in flood risk management. Examples include the 30 km long Southern Sea enclosure dam (Afsluitdijk) and the Deltaworks, each of which costed about 6 to 7% of GDP. The impact on nature was usually not included in these CBA’s. The 1901 CBA on the Southern Sea enclosure dam (Lely, 1901) looked at many different types of costs and benefits, but ignored the negative impact on nature. An exception was compensation for salt water fishers. Similarly, following the massive flooding of the southwestern part of the Netherlands in 1953, the CBA on the Deltaworks (Tinbergen, 1953) compared two alternatives to ensure sufficient flood risk safety: raising and strengthening dikes all along the waterways versus shortening the coastline by blocking the estuary mouths with barrier dams (Deltaworks). Many different types of costs and benefits were monetized, quantified or at least mentioned. But closing off the estuary mouths by barrier dams would turn tidal salt water areas into fresh water lakes like the IJsselmeer (the former Southern Sea); these substantial negative effects on nature were ignored in the CBA.

Reporting environmental effects of public investments, including those on nature, started in 1978. NEI and RIN (1978) presented an Environmental impact analysis (EIA) and a CBA comparing the extension of the port of Den Helder with alternative solutions in the ports of IJmuiden and Rotterdam. Basically, most effects on nature were given in physical terms that were presented next to monetary costs and benefits. Some effects on nature were monetized, in particular the foregone revenues of fishing, the loss of shell lime production and the loss of water cleaning capacity. The impact of a new port on five basic functions of nature (production function, intermediary and supporting function, informative function, regulatory function and conservation function) was specified and at least scored qualitatively (with minus and plus signs) for 13 different sub-functions. Nearly a decade later, an European Act³ made environmental impact assessments (EIA) obligatory.

Unlike a CBA, EIA does not translate positive and negative effects in nature into monetary terms and usually do not consider double counting, i.e. whether several environmental impacts lead to the same impact on welfare. But their information can be used as an input for CBA. For example, in the CBA on deepening the Westerschelde-waterway from the Netherlands to Antwerp, the EIA was used to claim that from a European perspective the environmental effects were negligible. This is still the role of EIA in most Dutch CBAs on transport infrastructure and spatial projects (see Annema and Koopmans, 2015).

To improve the quality and consistency of CBA, national CBA guidelines for transport infrastructure were published in 2000. These guideline did not explicitly discuss the effects on nature of infrastructure projects. The valuation of effects on nature in CBA was separately addressed in a supplementary guidance (Ruijgrok et al., 2004) and an overview with key-figures for such valuation of nature (Witteveen & Bos, 2006). As a result, attempts were made to include e.g. effects of changes in nature on housing prices, health and recreation. Due to shortage of data and a lack of primary valuation studies, this resulted in many cases in arbitrary assumptions or token entries – indicating that the effect was relevant but that no reliable monetary value could be estimated. In other CBAs, some impacts were double counted or other errors were made in quantifying and valuing the welfare effects in terms of cost and benefits.

In CBAs since 2000, like Stolwijk and Verrips (2000) and Ebregt, Eijgenraam and Stolwijk (2005), for measuring the non-use value of biodiversity, ordinal scaling or quantitative measures were used, like the change in the number of hectares of high environmental quality. But no detail was shown and the rarity of the species in the habitat was not taken into account.

The biodiversity point method was developed in 2009 at PBL Netherlands Environmental Assessment Agency as a joint effort of economists and ecologists (Sijtsma et al., 2009). According to the Dutch guidelines on cost-benefit analysis (Romijn and Renes, 2013; Klooster et al., 2018), biodiversity points are an innovative and practical method to measure the impact on the non-use value of biodiversity.

In 2013, an updated CBA guideline was published by CPB and PBL (Romijn and Renes, 2013), which included a brief discussion of accounting for the impact on nature⁴.  This topic is covered more in-depth in the supplementary guideline on CBA and nature (Klooster et al., 2018).  The guideline on CBA and nature recommends the use of biodiversity points for measuring the impact on biodiversity. It also stresses the importance of providing clarity about the welfare effects of changes in nature. For example, most regulating services, like natural pest control and water purification, are intermediate services. They indirectly affect welfare as they are an input in the production function for final ecosystem services. Effects on these intermediate services are to be indirectly included in a CBA through their effects on the final services and the valuation thereof (see Boyd and Banzhaf, 2007). Biodiversity – the variety of genes, species and ecosystems – holds a special position in these guidelines. Biodiversity is important to guarantee a continued delivery of ecosystem services over the long term and for maintaining ecosystem resilience (Cardinale et al., 2012; Isbell et al., 2017).

How to measure biodiversity points?
Biodiversity points measure the impact of a policy measure on the amount and the quality of biodiversity in a standardized way. Biodiversity points is defined as:

Biodiversity points = Σⁿ_i=1 Area_i X Quality_i X Weight factor_i

With i Ε {1,…,n} the different types of ecotopes or nature types.

The biodiversity points are calculated by multiplying three components:

• The area of natural or semi-natural ecosystems affected (in hectares or square kilometers);
• The ecological quality of each area (0-100%);
• A weight factor per type of ecosystem, reflecting the contribution of the ecosystem to species richness at national, European or global level, which depends on the species present in the ecosystem and their threat level.

The ecological quality is measured by an intactness or robustness score, in a range from 0 to 1. This measure is determined for each of the relevant ecotopes based on the number of characteristic species present in the area relative to their presence in an intact ecosystem. These ecotopes and characteristic species are derived from the universal set of biodiversity indicators (Convention on Biological Diversity, 2004), the more detailed European set of biodiversity indicators (EEA, 2007) and the Mean Species Abundance used in UNEP’s Global Environment Outlook. For this, national reference lists containing the species in an intact ecosystem are available, e.g. for the Netherlands⁵. In Europe, for the Habitat regulation (EC, 1992) and the Water Directive (EC, 2000), each EU-member country had to assess the ecological quality of ecotopes in comparison to a healthy ecological condition. So, this is already roughly similar to providing intactness or robustness scores for ecological quality.

Environmental impact assessments (EIAs) generally provide the necessary information on ecological quality, before and after the policy intervention. This information per ecotype of the number of characteristic species in the area can then be translated into ecological quality scores, before and after the policy intervention. Multiplying the ecological quality scores for the different ecotopes by the acreage of their area gives the Ecological Quality Area score (EQA) per ecotope.

Finally, the EQAs of the ecotopes are multiplied with standardised weight factors that indicate the threat level to the ecotope. This threat level is related to the relative number of red list species in the ecotope. Extremely threatened ecotopes have the highest weight, while commonly occurring ecotopes with common and not threatened species have the lowest weight. As a result, an intervention in a highly threatened ecotope results in a higher score than a similar intervention in a non-threatened ecotope. For example, salt marshes have a weighting factor of 2.4, nutrient-poor peatlands and moist heather lands have a weighting factor of 1.2 and agricultural grasslands have a weighting factor of 0.4.

Determining the weighting factors is not fully straightforward and different methods and data sources are possible (see Sijtsma et al., 2009). However, most important is that these weights are standardized for each country and based on systematic ecologic data collection which is objective and transparent, e.g. similar to how CO2-equivalents are used to aggregate different types of emissions or how different health effects are summarized by the indicator Disability Adjusted Life years (DALY’s).

An advantage of the biodiversity point method is that decision-makers have a single objective measure to compare biodiversity effects of alternative interventions. For some questions, this is more useful than the range of impacts shown by an EIA. Where an EIA is useful to assess whether legal norms are exceeded, it is not very useful to compare e.g. an intervention impacting fish stocks with an alternative intervention impacting water quality in an adjacent area. The scarcity based weighting of the biodiversity points allows decision-makers to compare these incomparable impacts.

Biodiversity points in Dutch CBA
Two examples can illustrate the use of biodiversity points in CBA. The first example pertains to renovating the 33 km enclosure dam of the IJssellake. According to table 1, only some alternative interventions and options have substantial impact on biodiversity, e.g. the natural enclosure dam with some trees and vegetation in front of the dam, the 500 ha extra marshes and the fish sluice. They either result in larger areas of rare habitat types (with high weighting scores) or result in substantial quality improvements. The table also shows that the option natural enclosure dam has a clear positive effect on biodiversity: an increase of 1,600 biodiversity points. An interesting result was that nearly the same amount of biodiversity points (1,500) could be obtained by constructing a fish sluice in the enclosuredam, but at only a fraction of the costs: not 550 mln euro but 10 mln euro. Hence, fish sluices were much more cost-effective for improving biodiversity.

This CBA was well received by policy makers. The results were almost completely adopted in the final decision of the Dutch Cabinet. The option Natural enclosure dam was rejected and it was decided to construct a fish sluice. Subsequent political decision-making led to a much more advanced and fish-friendly, but also much more expensive fish sluice (35 mln euro).

Table 1.Renovating the enclosure dam: cost-effectiveness of various options for extra biodiversity (Wessels et al. 2011)

The second example is a meta-study about the cost-effectiveness of 175 defragmentation policy measures in the period 2004-2018 (see table 2). Four types of defragmentation alternatives were distinguished: ecoduct, viaduct, big faunatunnel and small faunatunnel. The major conclusions are:

• Large fauna tunnels and viaducts with share use of traffic and animals are more cost-effective for stimulating biodiversity than ecoducts (0.08 mln euro per biodiversity point versus 0.18 mln euro per biodiversity point); small faunatunnels are by far the least cost-effective, i.e. on average more than double as costly than ecoducts (0.38 mln euro per biodiversity point).
• The cost-effectiveness differs between ecoducts: the more nature areas are in the direct vicinity, the more cost-effective.
• Buying agricultural land and using this for nature purposes is about as cost-effective for biodiversity than a viaduct or big faunatunnel. However, ecological improvement of existing nature zones is even much more cost-effective (0.02 mln euro per biodiversity point).

Table 2. CEA meta study on infrastructure defragmentation (Sijtsma et al., 2020)

The results of this meta-study can also be compared with those of other biodiversity-studies. In terms of impact on biodiversity, one ecological connection results in somewhat more than 10 biodiversity points (146 connections, in total 1791 biodiversity points). This implies that e.g. that the fish sluice in the enclosuredam with 1500 biodiversity points is broadly comparable to 150 ecological connections. In terms of cost-effectiveness, a fish sluice in the enclosure dam (0.02 mln euro per biodiversity point) is comparable to ecological improvement of existing areas and much more cost-effective than ecological connections.

Conclusions
This biodiversity points approach is increasingly being applied in Dutch CBAs, resulting in more and more insight in what would be reasonable costs for obtaining an additional biodiversity point. This is in particular useful to compare the cost-effectiveness of project alternatives with respect to their impact on biodiversity. However, biodiversity points are also useful for assessing the net benefits of projects in which the impact on biodiversity is a major co-benefit or trade-off. In Dutch CBA practice, the impact on biodiversity is still often included as a token entry with some qualitative assessment. As consequence, major impacts on biodiversity may be ignored in public decision-making. It also happens that the impacts on biodiversity are very limited, but are exaggerated in public debate about the project for strategic reasons. Biodiversity points can be an important method to avoid such misunderstanding about the size of the impact on biodiversity.

Biodiversity points are a practical and transparent method to quantify the impact of policy measures on biodiversity. They are especially useful for policy measures that have a major impact on ecosystems, such as nature policies or infrastructural works near nature or protected areas. They can be very helpful to formulate more nature friendly or cheaper policy alternatives, or to find more cost-effective compensation measures.

For assessing the net benefits of projects, biodiversity points provide a standardized quantitative summary measure for the impact on biodiversity. This biodiversity measure can be decomposed into its constituent parts, is based on acreage of the impact area, internationally standardized ecological quality indicators and nationally standardized threat weights and can be checked on its consistency of application for various CBAs. For assessing the overall effects of a project, this is more informative than qualitative (nominal or ordinal) expert opinions on a policy measure’s impact on biodiversity; these are generally not standardized and comparable for different CBAs and cannot provide an indicator of change in biodiversity per euro invested. Sensitivity analysis with a range of reasonable different cost-prices per biodiversity point can show net benefits fully in monetary terms.

The use of biodiversity points can be advanced by providing overviews of their costs per point for various types of nature at various locations and various types of policy measures. This overview can give concrete examples of relatively cheap interventions for improving or protecting nature (e.g. a fish sluice in the Afsluitdijk) and much more expensive ones. The overview can also discuss the factors determining these differences in cost-effectiveness. If such an overview is available, this would be a great help for assessing biodiversity points in another CBA or CEA.

Biodiversity points are quite similar to the quality-adjusted life years (QALY) used for cost-effectiveness analysis of health care treatments. The major merits of biodiversity points are:

• simple;
• transparent;
• relatively cheap;
• mostly uses information already in environmental impact analysis;
• linked to international classifications and lists of scarcity.

There are also various clear limitations:

• linear relationships, e.g. biodiversity points of 2 hectare of an ecotope is two times 1 hectare and no threshold or minimum size required;
• perfect substitution between size of the area, ecological quality and ecological scarcity;
• different methods of operationalization are possible;
• experts opinion on scarcity are used, not a valuation reflecting preferences of individual citizens. As a consequence, biodiversity points are in particular suited for measuring the non-use value of biodiversity.

The value of biodiversity points may also be investigated by surveying the willingness to pay for such points. The quality and weighting factors per ecosystem type can be replaced by monetary unit values that reflect peoples’ preferences over maintenance of these ecosystems, in this way reflecting the contribution of each ecosystem and the ecosystem services it provides to welfare. As argued above, this requires more insight into the welfare contributions of the various ecosystems. This can then be compared with the results from surveys on all kinds of nature development and landscape types (see e.g. Bateman et al., 2006 and Liekens et al., 2013) and the various cost-prices for biodiversity points.

Footnotes
¹By the statistician and management guru Deming.
²For more than a century, CBA is used to support decision-making on public investments in the Netherlands, see Bos and Zwaneveld (2017) and Bos (2008).
³Act on environmental impact assessment, Directive 85/337 EEC.
⁴Recently, also the discount rates to be used in Dutch CBAs have been changed. In 2016, it was decided to reduce the official discount rate from 5.5% to 3%. For nature, an annual relative price increase of 1% is prescribed (Werkgroep Discontovoet, 2015 and Koetse et al, 2018). As a consequence, the net effect for nature is a discount rate of 2%.
⁵Reference lists of ‘species pursued’ have been prepared for monitoring the Dutch nature policies and contain the pursued biotic and abiotic characteristics of each nature type. Using the reference lists, for each nature area measurable objectives can be set and monitored. They provide the basis for conservation planning and management and national and European .

References
Annema, J.A. and C. Koopmans, 2015, The practice of valuing the environment in cost-benefit analyses in transport and spatial projects, Journal of Environmental Planning and Management, vol. 58: 1635-1648.

Bateman, I.J., B.H. Day, S. Georgiou and I.R. Lake, 2006, The aggregation of environmental benefit values: welfare measures, distance decay and total WTP, Ecological Economics, vol. 60(2): 450-460.

Bos, F., 2008, The Dutch fiscal framework; history, current practice and the role of the Central Planning Bureau, OECD Journal on Budgeting, vol. 8(1): 7-48.

Bos, F. and A. Ruijs, 2021, Quantifying the Non-Use Value of Biodiversity: the Dutch Biodiversity Points, Journal of Benefit-Cost Analysis,

Bos, F. and P. Zwaneveld, 2017, Cost-Benefit Analysis for Flood Risk Management and Water Governance in the Netherlands: an Overview of One Century, CPB Background Document August 2017, CPB Netherlands Bureau for Economic Policy Analysis, The Hague, The Netherlands.

Ebregt, J., C.J.J. Eijgenraam and H.J.J. Stolwijk, 2005, Kosteneffectiviteit van maatregelen en pakketten – kosten-baten analyse voor Ruimte voor de Rivier, deel 2 (Cost-effectiveness of measures and packages – cost-benefit analysis for More Room for Rivers, Part 2), CPB document 83, The Hague.

Klooster, J., G. Warringa and J. Ohm, 2018, Werkwijzer Natuur: Maatschappelijke Kosten-Baten Anlayses. Arcadis, CE Delft, The Netherlands.

Lely, C., 1901, Ontwerp van wet tot afsluiting en droogmaking van de Zuiderzee met toelichtende memorie, Text downloadable via www.gutenberg.org, eBook 36317.

Liekens, I., M. Schaafsma, L. de Nocker at al., 2013, Developing a value function for nature development and land use policy in Flanders, Land use policy, vol. 30(1): 549-559.

NEI and RIN, 1978, Een haven op het Balgzand? Een economische en ecologische afweging van de voor- en nadelen van een beperkte zeehavenontwikkeling ten behoeve van Den Helder. Rijksinstituut voor Natuurbeheer, Nederlands Economisch Instituut, Rotterdam.

Romijn, G. and G. Renes, 2013, General Guidance for Cost-Benefit Analysis, CPB Netherlands Bureau for Economic Policy Analysis / PBL Netherlands Environmental Assessment Agency, The Hague, The Netherlands.

Ruijgrok, E.C.M., R. Brouwer and H. Verbruggen, 2004, Waardering van Natuur, Water en Bodem in Maatschappelijke Kosten-batenanalyses, Aanvulling op de Leidraad OEI.

Sijtsma F.J., A.V. Hinsberg, S. Kruitwagen and F. Dietz, 2009, Natuureffecten in de MKBA’s van projecten voor integrale gebiedsontwikkeling. PBL Netherlands Environmental Assessment Agency, the Netherlands.

Sijtsma, F.J, E. van der Veen, A. van Hinsberg, R. Pouwels, R. Bekker, R. van Dijk, M. Grutters, R. Klaassen, M. Krijn, R. Klaassen, M. Mouissie and E. Wyminga, 2020, Ecological Impact and Cost-effectiveness of Wildlife Crossings in a Highly Fragmented Landscape, A Multi-method approach, Land Ecology.

Stolwijk, H. and A. Verrips, 2000, Ruimte voor water; kosten en baten van zes projecten en enige alternatieven, CPB Werkdocument 130.

Tinbergen, J., 1953, De Economische balans van het Deltaplan (Annex to the report of the Deltacommission).

Witteveen & Bos, 2006, Kentallen Waardering Natuur, Water, Bodem en Landschap; hulpmiddel bij MKBA’s, Ministerie van LNV.