Final Draft. To be published on the web after the ad hoc expert meeting on further reductions of VOC emissions on 13 May 2002.

Transcription

1 Final Draft To be published on the web after the ad hoc expert meeting on further reductions of VOC emissions on 13 May THE COSTS AND BENEFITS THE REDUCTION OF VOLATILE ORGANIC COMPOUNDS FROM PAINTS Prepared by DIRECTORATE-GENERAL ENVIRONMENT Air and Noise Unit 2 May 2002

2 CONTENTS EXECUTIVE SUMMARY 1. INTRODUCTION UNIT COSTS OF THE REDUCTION OF VOCS FROM PAINTS Increase in production costs implied by a shift from SB to WB paints Increase in production costs implied by a shift within the paint category Cost increase due to additional research and development Cost increase due to additional investment Summary of cost increases Unquantified effects Costs according to the National Emissions Ceilings Directive COSTS OF REDUCING VOC CONTENT IN PAINTS BENEFITS OF REDUCING VOCS FROM PAINTS COMPARISON OF COSTS AND BENEFITS CAVEATS CONCLUSIONS... 9 ANNEXES 1. Changes in costs due to reduction VOC content in paints 2. Production and consumption of paints in the EU 3. Requirements of the reduction of VOC content in different paint categories

3 EXECUTIVE SUMMARY This study analyses the costs and benefits of the reduction of VOCs from paints underpinning the draft proposal for a directive. The basis for the analysis are mainly on Decopaint (2000) as well as AEAT (forthcoming). The overall annual reduction in VOC emissions resulting from the proposal is estimated to be 279 kilotonnes in This is estimated to cost between 108 and 157 million per annum in The average cost for reducing VOC content in paints is thus estimated to be between 387 and 563 per tonne of VOC reduced. The difference in the costs stems from the uncertainty of the additional costs for exterior paints. This cost (from 387 to 563) per tonne of VOC reduced is smaller than average of cost-effective options to reach the objectives of the National Emissions Ceiling (NEC) directive ( 687 per tonne of VOC abated) The total benefits of reducing 279 kilotonnes of VOCs in the EU is estimated to bring annually health related benefits of 582 million. It should be noted that some benefits were not monetised. These unmonetised effects are the improved health of painters (due to less solvent exposure) and the benefits to ecosystem (due to lower ozone concentrations). The benefits of the proposed directive were estimated to be four to five times higher than the costs. The benefits were estimated to be higher than costs in all Member States. In sum, the proposed directive passes the benefit cost test, even if the costs were somewhat underestimated or the benefits somewhat overestimated.

4 1. INTRODUCTION This study summarises the analysis of costs and benefits of the reduction of Volatile Organic Compounds (VOCs) from paints being proposed by the European Commission. The basis for the analysis of costs is based largely on Decopaint (2000) while the benefits are derived from the analysis of the National Emissions Ceilings directive and the subsequent refinements to the estimates of marginal damage of VOCs. In section 2 the costs associated with reducing the VOC content of paints is assessed. Section 3 provides estimates of the costs implied by the proposed Directive. In section 4, the benefits of reducing VOC emissions in different Member States are reported. In section 5, the costs and benefits are compared. All prices used in this study refer to year 2000 price level. 2. UNIT COSTS OF THE REDUCTION OF VOCS FROM PAINTS According to the Decopaint (2000) study switching from solvent based (SB) to water based (WB) paints would imply and increase in costs. This is mainly due to more expensive inputs for WB paints, but also due to some additional research and development and investment costs. By contrast, reducing the solvent content of existing WB paints is usually fairly straight-forward and the additional cost is rather low. However, in some cases the lowering of solvent content of a paint requires a significant technological shift and the costs would then be of the same order of magnitude as in the case of shifting from SB to WB paint Increase in production costs implied by a shift from SB to WB paints On switching from SB to WB paints the composition of the paints will change and the raw material for WB paints are more expensive. However, there would be a reduction in the use of white spirit during the application of the paint. It is estimated that to reduce the VOC content of paint by switching from SB to WB paint costs 698 per tonne in trim paints 1. Annex 1 gives the details of the calculation. The quality requirements for trim paints are higher than for interior wall and ceiling paints. Therefore, the costs of replacing SB paints by WB paints for trim paints is likely to be higher than in the case of interior wall and ceiling paints. However, there are no estimates of the costs for interior wall and ceiling paints, and so we have assumed (conservatively) that the cost is the same as for trim paints. 1 Trim paints are used for painting door windows, frames, doors etc.

5 2.2. Increase in production costs implied by a shift within the paint category In most WB paint categories, it is assumed that the cost of reducing VOC content would not incur additional raw material costs. Thus, only additional research and development costs would be incurred for interior matt walls and ceilings, interior/exterior trim varnishes and woodstains as well as one pack performance coatings. The same is assumed to be the case for the following SB paints, interior/exterior trim varnishes and woodstains, primers, one pack and two pack performance coatings. However, for WB primers, two pack reactive performance coatings, and multi-coloured and decorative coatings, it is assumed that the assumed that the reduction of VOC content would require a technology shift. A similar technology shift is assumed to be required for SB of multicoloured and decorative effect coatings. For these paint categories, it has been assumed that the price in crease would be of the same order of magnitude as for switching from SB to WB paint (i.e. 698 per tonne of VOC abated). In two paint categories, (i) exterior walls of mineral substrate (ii) interior/exterior trim and cladding paints for wood and metal, two assumptions were made for both WB and SB paints. In the optimistic case, it was assumed that no additional raw material cost would be required. In this case it was thus assumed that there would be no requirement for a technology shift to reduce the VOC content variants of these paints. In the pessimistic case it was assumed that the additional raw material costs were 698 per tonne of VOC abated due to the technology shift Cost increase due to additional research and development Decopaint (2000) did not assess explicitly the increase of costs of reduction of VOC content of paint in different paint categories. It did, however, give an estimate how much research and development costs would need to increase in order to reduce VOC in all paint. Based on the study we have estimated that this additional expenditure is about 56 per tonne of VOC reduced for all paint categories. Annex 1 gives the details Cost increase due to additional investment The additional investment costs are estimated to occur when there would be a switch from SB to WB paints. The proposal for the directive would call for this to occur in practice in interior wall and ceiling paints. Thus, the estimated additional investment cost of 500 per tonne of VOC is used in interior walls and ceilings category. (Annex 1 gives the details.) 2.5. Summary of cost increases Table 1 describes exactly what the increases would mean for each product category when the results of Annex 1 are applied. 2

6 Table 1: Cost increases due to the abatement of VOCs in paints Average cost of reduction per tonne of VOC reduced Product Subcategory Optimistic Pessimistic Interior matt and glossy walls and ceilings WB to WB SB to WB *) Exterior walls of mineral substrate WB to WB SB to SB Interior/exterior trim and cladding paints for wood and metal Interior/exterior trim varnishes and woodstains Primers and binding primers WB to WB SB to SB WB to WB SB to SB WB to WB SB to SB One pack performance coatings WB to WB SB to SB Two pack reactive performance coatings for specific end use WB to WB such as floors SB to SB Multi-coloured and decorative effect coatings WB to WB SB to SB Source: See Annex 1 for details. *) Note: It has been assumed that the additional investment cost is added only to those paint categories where SB paints are likely to be replaced by WB paints more or less completely (i.e. interior walls and ceilings) Unquantified effects There are different estimates of the effect of the application time of WB and SB paints. Some claim that WB paints are easier, and thus faster, to apply, while others claim that SB paints are faster. It is recognised that the price of paint is a rather small component of total painting costs (whether professional or DYI). However, as there is no evidence of the effect of application times due to the switch from SB to WB paint, this (either positive or negative) effect is not included in these calculations. In addition, the stranded assets argument has been raised, i.e. if the application of the directive would cause some paint factories to close prematurely. It should be noted that the average technical lifetime of investment in paint factories is between 10 and 20 years, while the commercial lifetime is about 10 years. Given the proposed entry into force of Phase II of the directive in 2010, the transitory period is long, and regarded sufficient to not to cause assets to be stranded. The application of SB paints is dangerous at work places causing 'chronic toxic encephalopathy (CTE)' syndrome, which disables painters from working. For instance in 2000, in the Netherlands 25 painters were registered having CTE. In Scandinavia (where WB paints are more widely applied) cases of CTE are quite low. These positive health impacts of the reduction of SB paints are not included in these calculations. In sum, the unquantified effects of the switch from SB to WB paints are implicitly assumed to balance out each other. 3

7 2.7. Costs according to the National Emissions Ceilings Directive For purposes of comparison, we can look at the estimated costs of VOC abatement that were used for the National Emission Ceiling (NEC) directive. In the cost-benefit analysis carried out for the NEC directive 2 it was established that the benefits of the directive were higher than the costs. In addition, it was established, that in order to achieve the limit values on VOC abatement the average (weighted) abatement cost for the EU would be 687 per tonne VOC abated. 3. COSTS OF REDUCING VOC CONTENT IN PAINTS Decopaint (2000) reported the production (table ) and consumption (table 4.2.4) of about 90% of paints in each Member State. As data on production and consumption of each paint category are not available in each Member State, it has been assumed that all paint produced in one Member State is also consumed there. Thus, it is possible that there are some inaccuracies concerning the production and consumption numbers but these are not assumed to be important. Further, the possible inaccuracies cancel each other out at the EU level The production and consumption data were used for estimating the amounts of VOC that paints contain. For the purposes of this analysis, it was assumed that the consumption of paints would increase by 1% per annum up to For those categories excluded from the Decopaint (2000) the Commission estimated the data on a pro-rata basis. Annex 2 contains the production and consumption data that were used and details of the estimation of consumption up to Annex 3 documents the changes in the VOC contents of implied by the proposal for the directive. These are slightly different from the Decopaint (2000). The total costs per product category and Member State are given in tables 2 and 3. Overall, the reduction of VOCs from paints as a result of the proposal for a directive is estimated to be 279 kilotonnes per annum in This is estimated to cost between 108 and 157 million per annum. The average cost for reducing VOC content in paints is thus estimated to be between 387 and 563 per tonne of VOC reduced. The difference in the costs stems from the uncertainty of the additional costs for paints for (i) exterior walls of mineral substrate and (ii) interior/exterior trim and cladding paints for wood and metal. This cost (from 387 to 563) per tonne of VOC reduced is smaller than average of cost-effective options to reach the objectives of the NEC directive ( 687 per tonne of VOC abated) The cost-benefit analysis of the Directive 2001/81/EC on National Emission Ceilings for certain air pollutants, OJ L 309, , p.22, can be found at and See Section 2.6 4

8 Table 2: Reduction of VOCs and related costs per paint category in 2010 Kilotonnes of VOC abated in 2010 Cost of reduction optimistic ( million) Cost of reduction (pessimistic) ( million) Interior matt walls and ceilings (Gloss<25@60º) WB 85,5 4,8 4,8 SB 34,6 43,3 43,3 Interior glossy walls and ceilings (Gloss>25@60º) WB 18,5 1,0 1,0 SB 34,5 43,3 43,3 Exterior walls of mineral substrate WB 27,8 1,6 21,0 SB 4,3 0,2 3,3 Interior/exterior trim and cladding paints for wood WB 3,0 0,2 2,3 and metal SB 35,0 2,0 26,4 Interior/exterior trim varnishes and woodstains WB 2,9 0,2 0,2 SB 17,1 1,0 1,0 Primers WB 1,7 1,3 1,3 SB 0,0 0,0 0,0 Binding primers WB 1,2 0,9 0,9 SB 0,0 0,0 0,0 One pack performance coatings WB 0,3 0,0 0,0 SB 1,2 0,1 0,1 Two pack reactive performance coatings for specific WB 1,6 1,2 1,2 end use such as floors SB 0,0 0,0 0,0 Multi-coloured coatings WB 0,6 0,5 0,5 SB 3,8 2,9 2,9 Decorative effect coatings WB 1,3 1,0 1,0 SB 3,8 2,9 2,9 Sub-total WB 144,4 12,6 34,1 SB 134,3 95,6 123,1 Total 278,7 108,3 157,2 Average cost of reduction per tonne 388,40 563,9 Table 3: Costs of reducing VOCs per Member State in 2010 Kilotonnes of VOC abated in 2010 Cost of reduction optimistic ( million) Cost of reduction (pessimistic) ( million) WB SB Total WB SB Total WB SB Total Austria 3,0 2,2 5,2 0,3 1,4 1,6 0,7 1,9 2,6 Belgium 4,1 3,1 7,2 0,4 1,9 2,2 0,9 2,6 3,6 Denmark 2,7 2,0 4,8 0,2 1,3 1,5 0,6 1,7 2,4 Finland 1,7 1,3 2,9 0,1 0,8 0,9 0,4 1,1 1,5 France 13,1 36,5 49,6 1,4 35,4 36,8 4,5 39,3 43,8 Germany 35,6 26,5 62,1 3,0 16,3 19,4 8,0 22,7 30,8 Greece 5,4 4,0 9,4 0,5 2,5 2,9 1,2 3,4 4,7 Ireland 2,3 1,7 4,0 0,2 1,0 1,2 0,5 1,4 2,0 Italy 16,3 12,1 28,4 1,4 7,5 8,9 3,7 10,4 14,1 Luxembourg 0,2 0,1 0,3 0,0 0,1 0,1 0,0 0,1 0,1 Netherlands 8,3 6,2 14,5 0,7 3,8 4,5 1,9 5,3 7,2 Portugal 5,4 4,0 9,5 0,5 2,5 3,0 1,2 3,5 4,7 Spain 25,9 19,3 45,2 2,2 11,9 14,1 5,8 16,5 22,4 Sweden 3,2 2,4 5,6 0,3 1,5 1,7 0,7 2,0 2,8 UK 17,2 12,8 30,1 1,5 7,9 9,4 3,9 11,0 14,9 Total 144,4 134,3 278,7 12,6 95,6 108,3 34,1 123,1 157,2 5

9 4. BENEFITS OF REDUCING VOCS FROM PAINTS The reduction of VOCs from paints improves the European air quality by reducing the amount of tropospheric ozone. Ozone has a negative impact on human health. Further, that there can be a yearly loss of roughly up to 10% e.g. for certain crops due to tropospheric ozone (NEGTAP 2001). The marginal damage per tonne of VOC emitted has the same basis as the same as in the benefit assessment of the NED directive (see footnote 2). In that analysis the damage both to health, crops and buildings (but not cultural heritage) were included. It needs to be emphasised that the marginal damage of VOC relates to a situation where the NEC directive would not be attained in Due to the attainment of the NEC directive, the damage from VOCs in 2010 is likely to be lower than estimated in the cost-benefit analysis of the NEC directive. The Commission is in the process of updating these numbers. On the other hand, the value of statistical life used as the basis of the analysis in the NEC directive did not take into account the latest evidence. It was possible to include the latter effect to the marginal damage estimates 5. Table 4 (column 3) shows the most recent estimates of the damage caused by one tonne of VOCs in different Member States (AEAT, forthcoming). Table 4: Benefits of VOC reduction in Member States in 2010 Kilotonnes of VOC reduced Benefit of 1 tonne of VOC reduced Monetised benefits million Austria 5, ,3 Belgium 7, ,6 Denmark 4, ,3 Finland 2, ,4 France 49, ,2 Germany 62, ,0 Greece 9, ,7 Ireland 4, ,1 Italy 28, ,5 Luxembourg*) 0, ,8 the Netherlands 14, ,9 Portugal 9, ,2 Spain 45, ,7 Sweden 5, ,8 United Kingdom 30, ,1 Total 278,7 **) 581,8 *)Damage from VOC emissions in Luxembourg is assumed to be the average of the damage in Belgium and Germany.Note: The (unweighted) average of the damage from VOCs is about 2000/tonne in the EU. Source: AEAT (forthcoming) 4 5 It should be noted that the baseline emission of VOCs were projected to be 7,1 kt in Because of the NEC directive these emission are now projected to be 6,5 kt in This reduction is not believed to affect the marginal damage of VOCs significantly. For details, see European Commission (2000) 6

10 The damage is obviously related to the number of people affected, and thus, the damage from a tonne of VOCs is higher in central parts of the EU. A reduction of 1 tonne of VOCs is estimated to reduce the damage (i.e. gives benefits) between 490 and 7200 per tonne of VOC. It needs to be emphasised that the monetisation of the benefits from VOC emission reduction is quite complicated due to the non-linearities in relation to ozone formation. The total benefits of reducing 279 kilotonnes of VOCs in the EU is estimated to bring annually health related benefits of 582 million. The benefits to ecosystem could not be estimated because of lack of quantitative information of the monetised benefits of VOC reduction on forestry and ecosystems at large. 5. COMPARISON OF COSTS AND BENEFITS The costs of reaching the VOC reductions were established in sections 3 and the benefits (reduced damage due to decrease of VOC emissions) were established in section 4. In table 5 these costs and benefits are compared and the net benefits are derived. Table 5: Costs and benefits of the proposed directive in Member States in 2010 Total benefits ( m) Costs ( m) Benefits-costs ( m) Benefits/costs ( m) Optimistic Pessimistic Optimistic Pessimistic Optimistic Pessimistic Austria 7,3 1,6 2,6 5,7 4,7 4,5 2,8 Belgium 21,6 2,2 3,6 19,3 18,0 9,6 6,1 Denmark 34,3 1,5 2,4 32,8 31,9 23,1 14,5 Finland 1,4 0,9 1,5 0,5 0,0 1,6 1,0 France 99,2 36,8 43,8 62,4 55,4 2,7 2,3 Germany 174,0 19,4 30,8 154,6 143,2 9,0 5,7 Greece 8,7 2,9 4,7 5,8 4,1 3,0 1,9 Ireland 5,1 1,2 2,0 3,9 3,2 4,2 2,6 Italy 79,5 8,9 14,1 70,7 65,5 9,0 5,7 Luxembourg 0,8 0,1 0,1 0,7 0,7 9,3 5,9 Netherlands 34,9 4,5 7,2 30,4 27,7 7,7 4,8 Portugal 14,2 3,0 4,7 11,3 9,5 4,8 3,0 Spain 39,7 14,1 22,4 25,7 17,4 2,8 1,8 Sweden 3,8 1,7 2,8 2,1 1,0 2,2 1,4 UK 57,1 9,4 14,9 47,7 42,2 6,1 3,8 Total 581,8 108,3 157,2 473,5 424,6 5,4 3,7 Overall the benefits of the proposed directive are much higher than the costs. In the optimistic case (i.e. assuming no major increase for exterior paints) the benefits are 5 times higher than the costs. Even in the pessimistic case, the benefits are almost 4 times higher than costs. Also, it seems that the benefits are higher than costs in all Member States. 7

11 6. CAVEATS The findings need to be interpreted with caution. The following caveats apply: It was not possible to include in this analysis all cost elements that might affect the reduction of VOCs in paints. The main cost elements, (raw material costs, research and development costs and investment costs) were included. Thus, we believe that the costs are representing the right order of magnitude, but there could be essential differences between some product categories or between some Member States. Some cost elements were not included in the analysis. The most important uncertainty relating to the cost-benefit analysis concerns the implications to productivity of painting due to the reducing the VOC content in paints. It is unlikely that there would be major changes if the solvent content is reduced within the same paint category (e.g. if the solvent content of a water based wall paint is reduced). The uncertainty relates mainly time that it takes to apply SB paints vs. WB paints. Here opinions vary and thus, it was assumed that overall there would be no change in the productivity of painting. It also needs to be emphasised that the health related benefits to the painters of reducing the amount of solvents in paints are not estimated as part of this cost-benefit analysis. Neither has it been possible to include the benefits of lower VOC emissions to ecosystems (apart from crops), apart from the effect to crop production. Table 6 summarises what has been included in and excluded from the analysis. Table 6: Issues included and excluded in the cost-benefit analysis Included in cost-benefit analysis If not included, likely impact Change in raw material use Yes - Additional research and development expenditure Yes - Additional investment Yes - Labour costs in paint manufacturing No Negligible Productivity of painters No Important issue, differences of opinion Quality of the product No Negligible (assumed to be the same) Health effects to population Yes - Health effects to painters No Probably large, positive Effect to buildings Effect to crop production Effects to ecosystem No Positive Effects of cultural heritage (old buildings etc.) No Positive Yes Yes 8

12 7. CONCLUSIONS The costs and benefits of the proposed directive to reduce VOC content in paints have been analysed based mainly on Decopaint (2000) as well as AEAT (forthcoming). The increase in the cost of paints was analysed and overall the cost increase was compared with the reduction of VOCs from products. The reduction of VOCs from paints was estimated to be 279 kilotonnes per annum. This is estimated to cost between 108 and 157 million per annum. The average cost for reducing VOC content in paints is thus estimated to be between 387 and 563 per tonne of VOC. The monetised benefits of reduced VOC were estimated to be 582 million per annum in Given that the 279 kilotonnes of VOC are estimated to be abated, the benefits would be per tonne of VOC abated. It was noted that some benefits were not monetised. These were the improved health of painters (due to less exposure to solvents) and the benefits to ecosystem and buildings/cultural heritage (due to lower ground level ozone concentrations). The benefit-cost ratio was estimated to be between 5,4 and 3,7. In other words the benefits of the proposed directive were estimated to be four to five times higher than the costs. The benefits were estimated to be higher than costs in all Member States. In sum, the proposed directive passes the benefit cost test, even if the costs were somewhat underestimated or the benefits somewhat overestimated. 9

13 REFERENCES AEAT and IIASA Economic Evaluation of a Directive on National Emissions Ceilings for Certain Atmospheric Pollutants available at AEAT. Forthcoming. Database of Externalities of Air Pollutants. Report prepared for the European Commission, DG Environment. Decopaint Study on the Potential for Reducing Emissions of Volatile Organic Compounds (VOC) due to the Use of Decorative Paints and Varnishes for Professional and Non-professional Use. Study made for the European Commission, DG Environment. Available at European Coatings Journal Eurostat basic indicators from Eurostat Yearbook European Commission Recommended Interim Values for the Value of Preventing a Fatality in DG Environment Cost Benefit Analysis. Results of a workshop on The Value of Reducing The Risk of Ill-Health or a Fatal Illness held in DG Environment on 13 th November The Interim values are available at Frontier Economics An economic impact assessment of proposed regulation on the EU decopaints industry. Final Report. A report prepared for the European Solvents Industry Group. March NEGTAP National Expert Group of Transboundary Air Pollution. DEFRA Contract EPG1/3/153. Available at 10

14 Annex 1 CHANGES IN COSTS DUE TO REDUCTION VOC CONTENT IN PAINTS 1. Change in raw material costs Trim paint is used as the example in all calculations. In order to make a WB paint to replace a SB paint, essentially two changes take place. First, alkyd resins are replaced by more costly acrylate resins and second white sprit is replaced by more costly glycol ether. However, the amount of solvent is reduced as well. This balances the cost difference. In addition the amount of pigment is slightly reduced in a WB paint. Overall there would appear to be no cost increase when moving from a SB paint to WB paint (the price per kg would increase from 1,58 to 1,59 per kg). However, the area covered by a kilogramme of WB paint is smaller than with a SB paint as the amount of dry matter in a WB paint is smaller. The correct comparison should thus be made on the dry matter implying a cost of WB paints being 22% (i.e. 0,37/kg) higher than for SB paints (Table A1). This represents the upper limit of the cost difference as the switch from SB trim paint to WB trim paint is more costly that the switch from a SB wall/ceiling paint to WB one. This cost increase is also in line with the analysis of Frontier Economics (2002). Table A1: Raw materials cost price comparison water vs solvent based trim paints price /kg Amount of raw material (kg) in a litre of paint Raw materials Water based Solvent based Alkyd resins, medium oil 1,30 0,000 0,467 Pure acrylate 50% 2,25 0,357 0,000 TiO2 (pigment) 2,40 0,260 0,338 White spirit 0,45 0,000 0,352 Glycol ether 1,25 0,106 0,000 Water 0,05 0,520 0,000 Total weight 1,24 1,16 Total cost 1,59 1,58 Total cost/kg 1,28 1,36 Solid content 0,62 0,81 Total cost/solid content 2,07 1,69 Increase in cost/solid content 0,37 /solid content Increase in cost/solid content 22 % Direct reduction of VOCs (a) 0,246 kg VOC Direct cost of VOC abatement 1,524 /kg Reduction in associated solvent use (b) 0,291 kg VOC Total reduction of VOCs (a+b) 0,537 kg VOC Total cost of VOC abatement 0,698 /kg Source: Commission estimates based on prices given in European Coatings Journal (December 1999 p. 701) In sum, when SB paint is substituted be WB paint, the VOC content is reduced by 0,246 kg 70% with a cost increase of 0,37/kg. The cost of abatement of VOC would be 1,524 per kg. White spirit is not used for cleaning/diluting purposes when WB paints are applied. Based on the Decopaint (2000), it is estimated that for each 1 kg of applied SB paint, about some 0,291 11

15 kg white spirit would be needed for cleaning/diluting purposes. This element needs to be added to the calculations when the cost of VOC reduction is calculated: by substituting from SB to WB trim paint would reduce in VOC by 0,537 kg [0,246 kg (directly) and 0,291 kg (indirectly)]. The cost increase is therefore estimated to be ( 0,31/kg / 0,537 kg =) 0,698 per kg. For a measure to reduce the VOC emissions from SB to WB is estimated to cost 698 per tonne of VOC abated for trim paint. For wall/ceiling paints this is an overestimate but the cost difference is not known. Therefore, it is conservatively assumed that the cost difference is the same for wall/ceiling paints as for trim paints. 2. Research and development costs Decopaint (2000) analysed the impact of additional research and development costs for the following categories of paints: interior matt and glossy walls and ceilings, exterior walls of mineral substrate as well as interior/exterior trim and cladding paints varnishes and woodstains for wood and metal. The research and development costs of the reformulation of paints are estimated to be 2 million per annum (Decopaint, section ). For the reformulation of resins the research and development costs are about 4,5 million per annum (Decopaint, section ). The emissions of VOC would reduce by some 117 kt. Thus, the research and development costs reducing VOC from these paint categories is estimated 56/t VOC abated. This estimate is extrapolated to all paint categories. 3. Investment costs Additional investment is required to move from SB to WB paint production. Decopaint (2000) estimated that due to this additional investment the price of trim paints would increase between 1% and 1,5% (section ). The reduction of VOC content of trim paints was estimated to be 44,5 kt of VOC 6. Assuming that that the price increase of paint would be 0,125 per kg 7 the additional investment cost due to the switch form SB to WB paint is estimated to be 500/t VOC abated 8. This additional investment cost is added only to those paint categories where SB paints are likely to be replaced by WB paints more or less completely (i.e. interior walls and ceilings). Other paint categories are not assumed to require additional investment, because there would be no a shift from SB to WB categories because of this proposal for a directive. 4. Summary of costs Table A2 summaries the costs presented in the three previous sections. It shows two cases, optimistic and pessimistic. The difference between the two is the fact that it is uncertain if the prices of all categories of paints for exterior walls of mineral substrate, interior/exterior trim and kt of VOC x (35%-10%)) = 44,5 kt of VOC Decopaint (2000) assumed the average price of trim paint to be 10 per kg. This calculated as follows: For 178 kt of paint to be substituted the price increase to cover the additional investment cost is 0,125/kg (i.e. 22,25 m). With this amount the reduction of 44,5 kt of VOCs is possible. Thus, dividing the additional investment of 22,25m by 44,5 kt of VOC gives the reduction per additional investment per unit VOC reduction, i.e. 500/kg of VOC. 12

16 cladding paints for wood and metal would increase because of a technology shift which would increase, i.a. the price of raw materials. In the pessimistic case this increase is assumed to occur. The increase is assumed to be the same as shifting from SB to WB paint (ie. 698 per tonne of VOC abated). In the optimistic case the only additional cost is assumed to occur due to the increase of research and development costs. Table A2: Summary of different costs including the cost of abatement of VOCs Optimistic case Pessimistic case /t VOC abated Type of cost Applied to Applied in addition to Costs due to increased costs of raw materials Research and development costs SB to WB in interior and ceiling paints WB to WB in primers and binding primers and in two pack reactive performance coatings for specific end use such as floors All categories of multi-coloured and decorative effect coatings All categories of exterior walls of mineral substrate, All categories of interior/exterior trim and cladding paints for wood and metal All paint categories 56 Investment costs SB to WB in interior and ceiling paints

17 Annex 2 PRODUCTION AND CONSUMPTION OF PAINTS IN THE EU Decopaint (2000) estimated the production of paints in each sub-category for It is assumed for the purposes of this analysis that there would be no shift between paint categories without additional policies. (Table A3). Decopaint (2000) did not give shares of the smaller categories and did not consider multi-coloured coatings, decorative effect coatings. It has been assumed that each of these sub-categories would have the same share of EU wide production (i.e. 0,5% each). Table A3: Production of paints by sub-category, share or each sub-category in 1997 (same shares are assumed to exist also in 2010) Product Subcategory %production EU Interior matt walls and ceilings WB 37,8 (Gloss<25@60º) SB 4,2 Interior glossy walls and ceilings WB 15,7 (Gloss>25@60º) SB 1,7 Exterior walls of mineral substrate WB 15,1 SB 3,3 Interior/exterior trim and cladding paints for wood and WB 2,3 Metal SB 8,9 Interior/exterior trim varnishes and woodstains WB 2,0 SB 3,2 Primers WB 0,5 SB 0,5 Binding primers WB 0,5 SB 0,5 One pack performance coatings WB 0,5 SB 0,5 Two pack reactive performance coatings for specific end WB 0,5 Use such as floors SB 0,5 Multi-coloured coatings WB 0,5 SB 0,5 Decorative effect coatings WB 0,5 SB 0,5 Total 100,0 Source: Decopaint (2000) No official estimates of consumption of paints in 2010 exist. For the purposes of the costbenefit analysis it was noted that the paint market is saturated in the EU-15. In other words, no major increase is expected in the consumption of paint. Disregarding the cyclical nature of paint use (i.e. the fact that if construction activity is booming, so is the use of paint) the amount of paint used per capita has been increasing by some 0.5-1% per annum. Eurostat (2001) projects EU population to increase by about 0.25% per annum up to Thus, for the purposes of this analysis is assumed that the consumption of paint will increase by 1% per annum from 1999 to This means that the estimated consumption of paints in Member States is 12% in 2010 than in Table A4. 14

18 Table A4: Consumption of paints in the EU in 1999 and in 2010 (estimated), kilotonnes Consumption in 1999 Estimated consumption in 2010 Austria 57,4 64,0 Belgium 79,1 88,2 Denmark 52,4 58,5 Finland 32,4 36,1 France 420,7 469,4 Germany 684,2 763,3 Greece 103,5 115,5 Ireland 43,6 48,6 Italy 312,8 349,0 Luxembourg 3,0 3,3 Netherlands 160,1 178,6 Portugal 104,3 116,4 Spain 497,3 554,8 Sweden 61,7 68,8 UK ,3 Total 2943,5 3284,0 Source: Decopaint (2000). Note: Consumption of paint is assumed to increase by 1% per annum. With tables A3 and A4 it was possible to estimate the consumption of different paint categories in different Member States in 2010, assuming that the average density of paints is 1,25 kg/l. Example: In Austria it was estimated that the share of WB interior wall (matt) paints would be 37,8% in As the consumption of all paints is estimated to be 64,0 kt of paint, the share of 24,2 kt of paint. 15

19 Annex 3 REQUIREMENTS OF THE REDUCTION OF VOC CONTENT IN DIFFERENT PAINT CATEGORIES Limit values g/l Product Subcategory Current practice Phase I (2007) Phase II (2010) Interior matt walls and ceilings WB (Gloss<25@60º) SB Interior glossy walls and ceilings WB (Gloss>25@60º) SB Exterior walls of mineral substrate WB SB Interior/exterior trim and cladding paints WB for wood and metal SB Interior/exterior trim varnishes and WB woodstains SB * Primers WB SB Binding primers WB SB One pack performance coatings WB Two pack reactive performance coatings for specific end use such as floors SB * WB SB Multi-coloured coatings WB * SB * Decorative effect coatings WB * SB * Note: Current practice estimates are from Decopaint (2000), except for multi-coloured and decorative effect coatings, which are Commission estimates. The limit values for Phase I and II are from the Commission proposal. The Decopaint (2000) did not analyse multi-coloured and decorative effect coatings. The Phase II limit value for SB Interior/exterior trim varnishes and woodstains was 500g/l. For One pack performance coatings Decopaint (2000) analysed two categories (with limit values of 500 g/l and 600 g/l respectively), which have been merged to one in the proposal. Example: Given that the consumption of WB interior wall and ceiling paint in Austria was estimated to be 24,2 kt (see Annex 2), the VOC content without new policies would be estimated at 125 g/l divided density (1,25 kg/l) implying 10% VOC content. The limit value of 30g/l would imply a VOC content of 2,4%. Given that the amount of paint is projected to be 24,2 kt, the reduction in interior wall and ceiling WB paints in Austria would be [(10%- 2,4%)x24,2kt=] 1,84 kt of VOC. 16