EUROPEAN COMMISSION

Brussels, 22.9.2020

SWD(2020) 183 final

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT

Accompanying the document

Proposal for a Directive of the European Parliament and of the Council

amending Directive 2004/37/EC on the protection of workers from the risks related to exposure to carcinogens or mutagens at work.

{COM(2020) 571 final} - {SEC(2020) 302 final} - {SWD(2020) 184 final}

1.Introduction: Political and legal context

2.Problem definition

2.1.What is/are the problems?

2.2.What are the problem drivers?

2.3.How will the problem evolve?

3.Why should the EU act?

3.1.Legal basis

3.2.Subsidiarity: Necessity and added value of EU action

4.Objectives: What is to be achieved?

4.1.General objectives

4.2.Specific objectives

4.3.Consistency with other EU policies

5.What are the available policy options?

5.1.Process for setting binding OELs and associated provisions under CMD

5.2.Description of the retained policy options for establishing binding OELs under the CMD

6.What are the impacts of the policy options?

6.1.Acrylonitrile

6.2.Nickel compounds

6.3.Benzene

7.How do the options compare?

8.Preferred options

8.1.Summary of the preferred options

8.2.Overall impact of the package or preferred options

9.How will actual impacts be monitored and evaluated?

9.1.Monitoring arrangements

9.2.Evaluation arrangements

Annex 1: Procedural information

Annex 2: Stakeholder consultation

Annex 3: Who is affected and how?

Annex 4: Analytical methods

Annex 5: National OELs

Annex 6: Relevant sectors, uses and activities

Annex 7: Route(s) of exposure, adverse health effects

Annex 8 – Process for setting binding OELs and associated provisions under the CMD

Annex 9 – Options discarded at an early stage

Annex 10 – Consistency and synergies with the REACH RegulationDisclaimer. Unless otherwise specified, the information and views set out is this report refer to EU28 aggregated data collected before the 1st February 2020, date of the withdrawal of UK from the EU.Disclaimer. Unless otherwise specified, the information and views set out is this report refer to EU28 aggregated data collected before the 1st February 2020, date of the withdrawal of UK from the EU.

Glossary

1.Introduction: Political and legal context

Cancer, irrespective of whether it is related to work or not, is the second leading cause of mortality in the EU countries after cardiovascular diseases, accounting for about a quarter of all deaths 1 . It is recognised as one of the major contributors to premature deaths in the European Union. It has an impact not only on individual health, family life, but also on the national health and social systems, the governmental budgets and the productivity and growth of the economy.

Stepping up the fight against cancer is therefore an urgent priority for the EU. To that end, as announced by European Commission President von der Leyen in her Political Guidelines 2 , the Commission will present before the end of 2020, a European plan to reduce the suffering caused by the disease and support Member States to improve cancer control and care in order to ensure more fair access to treatment across the EU.  

Fighting against occupational cancer is all the more necessary that occupational cancer remains the first cause of work-related deaths in the EU 3 , as shown in the figure 1 below. 52% of annual occupational deaths in the European Union are currently attributed to cancer, compared to 24% to circulatory diseases, 22% to other diseases and 2% for injuries. Addressing occupational cancer will also be an integral part of the Europe’s Beating Cancer Plan.

Figure 1: work-related deaths in the EU (2017)

The initiative comes in the backdrop of an unprecedented crisis for the EU and the world. The Covid-19 pandemic has major health, economic and social consequences that will need to be addressed. The pandemic also sheds light on the importance of health and safety considerations in workplaces, especially for those in the front line of the response to crisis. It gives yet another incentive to redouble the efforts to protect workers and societies from all possible occupational risks, thereby having a positive impact on employment and economy.

A strong social Europe calls for constant improvements towards safer and healthier work for all. As outlined in the Communication on “A strong social Europe for just transitions,” 4 measures for the protection of workers need to keep up with a wide range of social, economic and technological developments while at the same ensuring continuous protection from traditional risks. In this Communication, the Commission has already committed to review the occupational safety and health (OSH) strategy to address among others the exposure to dangerous substances, with a view to maintain European’s high OSH standards. The European Pillar of Social Rights 5 , jointly proclaimed by the European Parliament, the Council and the Commission at the Social Summit for Fair Jobs and Growth on 17 November 2017, enshrines workers' right to healthy, safe and well-adapted work environment, including protection from carcinogens. In its Communication on “Safer and Healthier Work for All,” 6 the Commission emphasizes that European Union must continue investing in occupational safety and health and has committed to step up the fight against occupational cancer through legislative proposals. The recent extension of the Roadmap on Carcinogens 7 covenant, which was signed in Helsinki on 28 November 2019, also proves that a significant number of stakeholders continue to be committed to improve the protection of workers from the exposure to carcinogenic substances.    

Reducing exposure to carcinogens and mutagens at the workplace by setting EU-wide occupational exposure limit values (OELs) effectively contributes to the prevention of cancer cases and deaths, as well as other significant non-cancer health problems caused by these substances. Consequently, it improves the protection of workers by increasing the length, quality and productivity of the working lives of European workers and ensuring a similar minimum level of protection across the EU, contributes to better productivity and competitiveness of the EU, and improves the level playing field for businesses.

In order to further contribute to a better protection of workers, this Commission continues its process of updating the Carcinogens and Mutagens Directive (CMD) 8 to keep abreast with the new scientific and technical developments and to take account of its stakeholders' views. This is in line with the Directive itself, which requires that OELs must be set for all those carcinogens or mutagens for which this is possible in the light of the available information. Consistency with the REACH Regulation 9 is ensured in this respect. The finalised REFIT Occupational Safety and Health evaluation 10 as well as the conclusions of the REACH REFIT evaluation 11 have underscored and fed into this process.

Updating and reviewing the CMD has now become a continuous process

Over the last few years, the Commission proposed three directives amending the CMD. These three proposals have been adopted by the European Parliament and the Council 12   13   14 . These three revisions, which addressed 26 substances, enabled among others to revise two existing OELs, introduce 22 new OELs and set a skin notation without any OELs for the remaining two.

The REFIT OSH evaluation 15 says that “following concerns raised by different stakeholders’ groups in the evaluation process and in the National Implementation Reports, the need to adopt limit values for more substances should be considered. These additional OELs should lead to a better chemical risk management in the future”. Therefore, the Commission will continue the process of updating and reviewing the CMD with the aim to propose new or revised OELs in Annex III to the CMD.

It is not realistic to set an OEL for every hazardous chemical that may be used at the workplace. Instead, it is appropriate to identify and target priority substances. This prioritisation of the substances is based on a broad consultative approach, including opinions issued by the tri-partite Advisory Committee on Safety and Health at Work (ACSH), agreement in its Working Party on Chemicals (WPC), and formal two-stage social partners’ consultation (SPC).

The selection of the specific three substances or groups of substances (hereafter “substances”) considered in this impact assessment, namely acrylonitrile, nickel compounds and benzene, folllowed the same path, including a formal two-stage consultation of the European Social Partners launched in July 16 and November 2017 17 , in accordance with Article 154 of the Treaty on the Functioning of the European Union (TFEU).

It was agreed by all relevant stakeholders, taking into account factors such as the potential to cause adverse health effects, degree of evidence of such effects, as well as their severity, potency and reversibility, that these three substances are of high relevance for the protection of workers. The Commission's intention to prepare for the establishment or revision of OELs for those priority carcinogens was confirmed and encouraged by all the stakeholders.

For the specific case of benzene, although an EU OEL of 3.25 mg/m³ already exists in the CMD since 2004, the most recent scientific and technical evidence indicates that this existing EU OEL should be updated. This is in line with the recital 13 of the CMD laying down that “limit values must be revised whenever this becomes necessary in the light of more recent scientific data”. Furthermore, the ACSH strongly recommended the Commission to adopt as soon as possible a revised OEL under the CMD for benzene.

In order to carry out its impact assessement, the Commission contracted a study 18 to COWI S/A (hereafter “COWI study”) in order to collect the most recent information for the three substances. According to the data arising from this study, more than 1 million EU workers are currently exposed to these hazardous substances. This impact assessment aims to assess whether there is a need to introduce new (for nickel compounds and acrylonitrile) or revised (for benzene) OEL for these substances, to address them with other legal or non-legal initiatives, or not to take any action.

Given the level of scientific and technical knowledge required to identify measures ensuring adequate protection of workers while being practically feasible for industries, the Commission bases its proposals in this area on opinions developed by the ACSH. The opinions of ACSH take into account the scientific basis, which is indispensable to underpin OSH legislation. In order to establish this scientific basis for the ACSH, the Commission sought advice from the Risk Assessment Committee (RAC) of the European Chemicals Agency (ECHA). This is in line with the REACH REFIT evaluation 19 in which the Commission proposed to “enhance the role of ECHA’s RAC, involving also social partners, to provide scientific opinions under the OSH legislation while respecting the role of the ACSH” with the aim to clarify the interface between REACH and other pieces of EU legislation.

The purpose of this impact assessment is to verify, on the basis of available socioeconomic data, the robustness of ACSH opinions and, eventually to consider some complementary measures, which could be proposed, based on further scientific information.

Member States authorities, employers' and workers' representative bodies within the tripartite ACSH strongly anticipate the legal clarity and increased protection which would be the result of new or revised OELs of these substances.

2.Problem definition

2.1.What is/are the problems?

As mentioned in section 1, 52% of annual occupational deaths in the EU are currently attributed to cancer. Furthermore, carcinogenic and mutagenic substances lead not only to cancers, but also to other important health problems. For example, exposure to benzene, in addition to leukaemia, also causes leukocytopenia, lymphocytopenia, neutrocytopenia and thrombocytopenia. More information about all the health effects for the three substances addressed in this initiative are available in table 1.

As mentioned in the previous section, more than 1 million workers are currently exposed to acrylonitrile, nickel compounds or benzene. If no action is taken at the EU level, these workers will continue to run the risk of contracting a cancer or other severe health problems.

Furthermore, ineffective prevention of the exposure to carcinogens entails negative consequences for business such as higher costs and reduced productivity due to absenteeism, lost expertise and distorted competition. For Member States, this also leads to increased social security costs and missed tax revenues.

The problem tree below summarises the main drivers behind the problem and the resulting consequences for workers, business and Member States:

Figure 2: Problem tree

2.2.What are the problem drivers?

1.

2.

2.1.

2.2.

2.2.1.Exposure of workers to carcinogens represents a significant risk to workers’ health

This section presents an overview of the estimated numbers of workers exposed to the substances subject to this initiative and a short explanatory summary for each substance. More detailed information is provided in annex 6.

Different sources compile different estimates of the total number of exposed workers. For the purpose of this impact assessment, for each substance the most reliable number has been taken forward for the baseline scenarios and for the cost-benefit assessments related to the retained options for establishing limit values.

Table 1: Summary of estimates taken forward for the assessment of options

Acrylonitrile

Acrylonitrile is widely used in the aircraft, defence, aerospace and automotive sectors. 21 This substance is particularly important to ensure a transition to technologies which are more environmentally friendly. More information is available at the annex 6.

As shown in the table 1, between 10,000 and 33,000 workers are estimated to be exposed to acrylonitrile at their workplace. Such an exposure can lead to brain, stomach, tongue, intestines and mammary gland cancer, but also to other health problems like nasal irritation. Although inhalation is considered as the primary route of exposure, workers can also be exposed through dermal contact.

Nickel compounds

Nickel compounds are used in a wide variety of sectors and are usually supplied in granule, powder or liquid form. More information about the uses of nickel compounds can be found at annex 6.

Table 1 shows that around 87,500 workers are currently estimated to be exposed to nickel compounds. This may have dramatic consequences as exposure to nickel compounds can lead to lung and nasal cancers but also pulmonary morbidity and miscarriage. The primary route of exposure is by dermal contact or by inhalation of aerosols, dusts, fumes of mists containing nickel.

Benzene

Benzene is produced in petroleum refinery and chemical plant processes and is used as an intermediate in the production of a wide range of chemical substances. Benzene is also used in the manufacturing of some types of rubbers, lubricants, dyes, detergents, drugs, and pesticides. 22 More information about the production and the uses of benzene are available in annex 6.

It is assumed that more than 1 million workers are still exposed to benzene and could therefore contract different hazardous diseases like leukaemia, as shown in the table 1. Benzene is readily absorbed by all routes (inhalation, dermal and oral). Inhalation is the most important route of occupational exposure.

The table 2 below shows the current and future burden of cancer and other health effects related to the occupational exposure to the three substances under consideration. Given the long latency period of the illnesses considered in this impact assessment, the future health burden is estimated over a 60 year period.

However, the disease burden on workers is likely to be underestimated due to several limitations of the data/calculations, which are further explained in the analytical challenges section of the annex 4. For instance, the COWI study does not quantify the burden of all cancers and other adverse health effects but only of those which are known to be caused by the lowest exposures (so-called most sensitive endpoints). When exposed to these substances, workers may develop additional types of cancer and other diseases which could be prevented by establishing an EU-wide OEL as mentioned in the RAC opinions for acrylonitrile 23 , nickel compounds 24 and benzene 25 . The table 2 specifies the health effects which have been quantified and not quantified. As these three substances fall under the CMD, in order to prevent the whole range of health problems, an OEL can only be established under this directive.

Furthermore, occupational cancers may develop decades after exposures – including during retirement – complicating the possibility of identifying a causal link, which could lead to underestimation of the disease burden.

For the specific case of benzene, table 2 shows that even with an existing EU OEL, a high number of workers can still develop leukaemia and other serious illnesses in the absence of further EU action.

Table 2 : Current and future disease burden related to occupational exposure to carcinogens (number of cases)

2.2.2.New scientific and technical evidence is available that could lead to updating of existing or establishment of new OELs

Under the CMD, employers must identify and assess risks to workers associated with exposure to carcinogens and mutagens, and must prevent exposure where risks occur. Substitution to a non- or less-hazardous process or chemical agent is required where this is technically possible. Where carcinogens cannot be substituted they must, so far as is technically possible, be manufactured and used in a closed system to prevent exposure. Where this is not technically possible either, worker exposure must otherwise be reduced to as low a level as is technically possible. This is the so-called minimisation obligation under Article 5 of the CMD.

For some chemical agents, the CMD establishes binding OELs at the EU level. The fact that OELs are established does not affect the obligations of the employer to comply with other rules, including to reduce the exposure of his/her workers to carcinogenic and mutagenic substances to as low a level as is technically possible (minimisation of the occupational exposure).

The existence of OELs provides clarity and OELs are very relevant benchmarks for employers enabling them to know exactly the levels above which exposure cannot occur. OELs also allow employers to determine the level below which their risk management measures should aim to comply with the obligation to reduce the exposure to as low a level as is technically possible. To this end, Safety Data Sheets (SDS), and more especially their section 8.1, are very important for downstream users to identify and apply appropriate measures in order to adequately control the risk of the chemicals at their site 26 .

OELs also support enforcement authorities in controlling that employers are putting in place the relevant risk management measures, including those that could contribute to lower the exposure below the OELs.

Under the CMD, the European Parliament and the Council shall set out limit values in Directives on the basis of the available information, including scientific and technical data, in respect of all those carcinogens or mutagens for which this is possible, and, where necessary, other directly related provisions.

For the substances covered in this impact assessment, the scientific advice has been provided by RAC which adopted its three opinions on 9 March 2018 27   28   29 . The tripartite ACSH also adopted opinions for all three on 4 June 2019 30 . It is therefore appropriate to consider updating the CMD based on the above-mentioned information. Further information on the scientific advice and ACSH opinions is provided in annexes 1 and 2, respectively.

Among the three substances addressed in this impact assessment, only benzene has already an EU wide OEL of 3.25 mg/m³ dated from 2004. Although this EU OEL already provides among others clarity and support enforcement by authorities as explained above, the most recent scientific and technical evidence indicates that it should be updated. This is in line with the CMD obligation to revise OELs whenever this becomes necessary in the light of more recent scientific data.

2.2.3.Diverging national OELs create different competing conditions and protection levels across the EU

The 2004 benzene OEL enabled to set a minimum common protection for all the workers dealing with this substance across the EU and limited the different competing conditions. The table 3 below, which lists among others the lowest and highest binding national OELs, shows that the Netherlands have set an OEL more than four times more stringent than the EU wide OEL. We can also deduce from this table that 10 countries adopted a level stricter than the EU OEL. Although differences in national OELs for benzene remain, this EU OEL limits the scope for divergences. If no EU OEL had been set, we cannot exclude that some Member States would not have established a national OEL yet.

Although there is still no EU wide OEL for acrylonitrile, 21 Member States have already set a national OEL in their own legislation while 7 Member States have no OELs yet. These national OELs range from 0.5 mg/m³ in Latvia to 7 mg/m³ in Slovenia and Slovakia.

In the absence of any EU wide OEL for nickel compounds, the table below shows that only a few Member States have already set an OEL. However, 19 out of the 23 Member States with no single national OEL for nickel compounds have a mix of OELs for some or all of these compounds. Moreover, the strictest OEL for nickel compounds is eight times lower than the highest limit value.

Table 3 : National OELs in EU Member States

As shown in the table 3, the disparities between the existing national OELs for the three substances are high, which leads to different level of protection of the EU workers. In addition to this main consequence, such differences also cause complications for businesses operating in different Member States and bring confusion to cross-border companies which have to deal with highly diverging OELs. To a lesser extent, these disparities could also provide potential incentive for companies to locate their production facilities in Member States with the lower standards, which would prevent the smooth functioning of the internal market. Such disparities may also be a brake to the freedom of movements of workers, as they could renounce to accept a job in another Member State, for fear to receive weaker health and safety protection. Annex 5 provides an overview of all national OELs in EU Member States for the substances considered under this initiative.

2.2.4.Modern production technologies allowing lower exposure to carcinogens or mutagens are not fully exploited

Although modern production technologies allow to reduce further the occupational exposure to carcinogenic and mutagenic substances in the workplace, their implementation is not yet generalised in all the companies dealing with the three substances addressed in this initiative. One possible reason is that decisions of business are often influenced by short-term cost assumptions rather than long-term benefits. 

For example, as regards acrylonitrile, although a significant part of the companies have already a closed system in place, some companies having laboratories using this substance mainly use hoods as ventilation system. Moreover, it is assumed that companies operating at low concentration levels usually have full enclosure local exhaust ventilation (LEV) while the others have partial enclosure or open hood LEV 31 . Companies working at higher levels of exposure should therefore invest in full enclosure LEV where possible.

According to a consultation survey carried out in the framework of the COWI study, companies indicated in which types of risk management measures (RMMs) they would invest to comply with OELs of 0.01 and 0.03 mg/m³ for nickel compounds. As an example, 30% of the companies that replied to this survey would make investments in partial hoods to comply with an OEL of 0.03 mg/m³. This tends to prove that companies do not always use the most modern production technologies allowing to reduce the exposure to carcinogens or mutagens.

A similar consultation survey was conducted for companies dealing with benzene. During this survey, several companies indicated that it was possible to reduce further the exposure of workers to benzene, meaning that they do not use the most modern production technologies yet.

2.3.How will the problem evolve?

In the absence of EU action, it is estimated that workers exposed to these substances will continue to face the risk of contracting occupational cancer or other adverse health effects. Estimations on the numbers of cancer and other hazardous diseases cases and their associated health costs over a 60-years period in case no action is taken are contained in the table 4 below (baseline scenario).

The general obligations set by the CMD, employers' actions and measures adopted by Member States contribute overall to lowering exposures. Exposure levels have generally been decreasing in the past years and this positive trend could continue in the future. Substitution may be possible for some carcinogens in the future, also the numbers of workers in the industries using these carcinogens may change, and technological developments could facilitate lower exposure concentrations.

Future forecasts in this area are however far from certain due to scarcity of relevant data and the fact that market forces such as raw material and energy prices, developing technology, as well as regulatory changes can drive decreases or increases in use which are not easy to predict. Even if trends were overall positive, as explained above, the existing employers' practices as well as protective measures at Member State level do not always reflect available scientific and technological knowledge. For that reason, assumptions are based on the legal provisions contained in the CMD, including the minimisation requirements, but also on other data including information gathered from the stakeholders. The objective has been to define a baseline scenario as close as possible to the future situation. However, it is very challenging to anticipate all the developments over such a long period. Further demographic changes increase the life expectancy of workers exposed and, therefore, the chances to develop the illnesses mentioned in table 2. More information about the past and future trends regarding the number of workers exposed and the exposure concentrations are available in Chapter 6.

Member States usually do not inform the Commission on their intentions to revise existing or determine new OELs in their national legislation. However, national administrations represented in the ACSH are aware of the preparatory work at EU level and therefore it is likely they will await its results in order not to duplicate efforts.

Table 4 - Estimated number of expossed workers, expected number of cancers and other hazardous diseases cases and related health costs in case no action is taken (baseline scenario), over a 60 years period

Source: COWI study (2019)

3.Why should the EU act?

3.1.Legal basis

Article 153 TFEU empowers the EU to support and complement the activities of the Member States as regards improvements, in particular of the working environment to protect workers' health and safety and to adopt, by means of directives, minimum requirements for gradual implementation, having regard to the conditions and technical rules obtaining in each of the Member States. On the basis of this provision, Article 16 (1) of the CMD provides a specific legal basis for action, allowing for adoption of limit values in respect of those carcinogens or mutagens for which this is possible, having regard to the available information, including scientific and technical data.

3.2.Subsidiarity: Necessity and added value of EU action

Scientific knowledge about carcinogenic chemicals is constantly developing and technological progress enables improvements in protection of workers. In order to ensure that the mechanisms for protecting workers from carcinogenic chemicals established in the CMD are as effective as possible, the Directive needs to be kept up to date with those developments. Updating the CMD to take account of newer scientific evidence is an effective way to ensure that preventive measures would be updated accordingly in all Member States.

The ex-post evaluation of the European Union occupational safety and health Directives 36 (REFIT evaluation) emphasizes that chemicals classified as carcinogens and mutagens continue to be manufactured across the EU and workers in manufacturing but also downstreams users continue to be exposed to them. The main conclusions of this evaluation indicates that the CMD is considered as of high relevance. Following concerns raised by different stakeholders’ groups in the evaluation process and in the National Implementation Reports, the need to adopt limit values for more substances should be considered. These additionals OELs should lead to a better chemical risk management in the future.

This initiative is the fourth update of the CMD aiming to set additional OELs for all those carcinogens or mutagens for which this is possible in the light of the available information, as required by the article 16 of the CMD. In the future, the Commission will continue to evaluate the need to propose additional OELs in the CMD and to propose additional amendments when needed, in collaboration among others with the ACSH.

Amending the CMD can only be done by action at EU level and it presents an EU added value in several respects:

Improved clarity and enforcement

Establishing new OELs for acrylonitrile and nickel compounds will provide common reference points that are used as a practical tool by employers, workers and enforcers to assess compliance with the general CMD requirements. OELs can also be used by process plant and machinery designers when planning new or considering alterations to existing process plants.

Clear support for establishing OELs for the substances subject to this initiative has been expressed from key stakeholders as it clearly results from the two phases of the consultation of the social partners and the opinions of the tripartite ACSH.

Ensuring a similar minimum level of protection across the EU

The CMD aims to improve workers’ health and safety by protecting them against risks arising or likely to arise from exposure to carcinogens or mutagens at work. This objective is in line with the European Pillar of Social Rights and the commitment of President von der Leyen to step up the fight against cancer in the context of a European plan against cancer.

Limit values under the CMD are an important component of the general arrangements for the protection of workers 37 . As laid down in Article 16 of the CMD, limit values should be established for all those carcinogens and mutagens for which the available information, including scientific and technical data, makes this possible. Such OELs should also be revised whenever this becomes necessary in the light of more recent scientific data 38 .

As mentioned in annex 8 related to the process for setting OELs under the CMD, it is not realistic to set an OEL for each hazardous chemical that may be used at the workplace. Therefore, the Commission identified acrylonitrile, nickel compounds and benzene as priority substances, following the procedure detailed in annex 8. This list of substances to be addressed in the 4th revision of the CMD has been approved by the Social Partners in the context of the formal two-stage consultation for which more information is provided in annex 2.

As highlighted in section 2.2.3., the difference between the national OELs may be very high. For instance, national OELs for acrylonitrile range from 0.5 mg/m³ to 7 mg/m³, meaning that some Member States have set limit values fourteen times lower than others. And in addition to that, 7 Member States have no OELs yet to prevent workers’ exposure to this substance.

Lack of EU action will most likely mean that there will remain Member States where no limit values exist for certain carcinogens or where those values are too high to ensure adequate worker protection. A minimum standard across the EU will not be ensured, to the detriment of worker protection.

With regard to the specific case of benzene, although an EU OEL already exists, its revision will ensure a more appropriate minimum level of protection across the EU.

Contribution to a level playing field

Employers' organisations stressed in their response to the social partner consultation that setting EU OELs helps to provide a level playing field for industry. The costs of complying with lower national levels are generally higher and entail, therefore, a competitive advantage for enterprises operating in markets with no or less stringent national OELs.

Setting EU OELs will not completely eliminate the differences between Member States. Indeed, they retain the possibility to adopt lower limit values as shown for benzene in table 3. However, it will limit (acrylonitrile and nickel compounds) or further limit (benzene) the scope for divergences and enhance certainty that there is a core definition and/or enforceable exposure limit for all concerned carcinogens in all Member States.

Assuming burdens at EU level related to derivation of limit values

The process of establishing limit values is very complex and requires a high level of scientific expertise. An important advantage of setting OELs at EU level is that it eliminates the need for Member States to conduct their own scientific analysis with likely substantial savings on administrative costs. These resources saved could instead be dedicated to improve further the occupational health and safety policies in each Member State.

4.Objectives: What is to be achieved?

4.1.General objectives

The main general objective of this initiative is to ensure to the workers the right to a high level of protection of their health and safety at work, as laid down in the principle 10 of the European Pillar of Social Rights 39 , and to prevent death caused by work-related cancer and other health problems as mentioned in the Commission Communication on “Safer and Healthier Work for All” 40 .

4.2.Specific objectives

The specific objectives are:

·To further improve protection from occupational exposure to carcinogens and mutagens in the European Union;

·To increase the effectiveness of the EU legal framework by updating it on the basis of scientific expertise;

·To ensure more clarity, facilitate implementation, and contribute towards a better level playing field for economic operators by adopting minimum requirements at the European level. The existence of OELs provides clarity by determining among others the maximum level of exposure with which employers must comply. However, complying with the OELs does not prevent the employers from their other obligations pursuant to the CMD, including the reduction of the exposure as low a level as is technically possible.

4.3.Consistency with other EU policies

Charter of Fundamental Rights of the EU

The objectives of the initiative are consistent with Article 2 (Right to life) and Article 31 (Right to fair and just working conditions) of the EU Charter of Fundamental Rights.

REACH Regulation

The REACH Regulation 41 , entered into force in 2007, establishes among others two distinct EU regulatory approaches that are restrictions and authorisations. Restrictions enable the EU to impose conditions on the manufacturing, placing on the market and/or use of substances, in mixtures or in articles, and authorisation is designed to ensure that risk of substances of very high concern (SVHCs) is properly controlled while promoting progressive substitution by suitable alternatives that are economically and technically viable 42 .

The applicable provisions of REACH authorisation and/or restriction of the chemical substances under consideration in this assessment, are as follows 43 :

·Restriction: all uses of benzene and its mixtures with some exemptions, the use of nickel and its compounds in jewellery and articles which are intended to come into contact with the skin, the use of acrylonitrile and its mixtures supplied to the general public.

·Authorisation: none of these substances are subject to authorisation under REACH.

More information about the REACH restrictions for the three substances is available in annex 10.

Having an OEL under the CMD plays an important role in the reduction of worker exposure to benzene. Industrial processes covered by other EU legislations which set up specific quantity of emission of benzene, are indeed not subject to the REACH restriction for this substance 44 .

Setting an OEL for nickel compounds under the CMD will also enable to reduce the exposure of workers to this group of substances.

As the REACH restriction for acrylonitrile does not concern the industrial uses, the setting of an OEL for this substance under the CMD will improve the protection of workers as they are not covered by the REACH restriction.

Europe’s Beating Cancer Plan

In her political guidelines for the next European Commission 45 , President von der Leyen recognized that there is much more the European Union can do about cancer and committed to put forward a European plan to fight cancer. On 4 February 2020, the Commission launched an EU-wide public consultation on Europe’s Beating Cancer Plan on the occasion of a conference entitled “Europe’s Beating Cancer Plan: Let’s Strive for More”hosted in the European parliament.

Europe’s Beating Cancer Plan will be structured around four pillars: prevention, early diagnosis, treatment and follow-up care. This proposal is fitted with the objectives of the prevention pillar such as measures to reduce environmental risk factors, for instance pollution and exposure to chemicals, and in particular the reduction of the exposure to carcinogens in the workplace.

5.What are the available policy options?

5.1.Process for setting binding OELs and associated provisions under CMD

A simplified outline of the process for the development of EU OELs for carcinogens and mutagens is set out here. A more detailed description is provided for in annex 8.

As mentionned in Chapter 1, the selection of the specific three substances considered in this impact assessment was based on a consultative approach, including opinions issued by the tripartite ACSH and a formal two-stage consultation of the social partners.

It was agreed by all relevant stakeholders, taking into account factors such as the potential to cause adverse health effects, degree of evidence of such effects, as well as their severity, potency and reversibility, that the three substances are of high relevance for the protection of workers. The Commission's intention to prepare for the establishment of OELs for those priority carcinogens was confirmed and encouraged by all the stakeholders.

Figure 3: simple representation of typical EU OEL setting procedure (Source: ECHA website)

5.2.Description of the retained policy options for establishing binding OELs under the CMD

The reference limit values per substance are presented below.

In addition to the baseline scenario, options for OELs have been considered at the level proposed by the ACSH, including the transitional OELs, and at additional reference points (e.g. the strictest limit value observed among Member States, OELs derived by the RAC). It needs to be noted that the most stringent national OEL might not always be feasible as an EU standard for the following reasons: firstly, the substances subject to this proposal are used in many different industries, and for some industries it might be difficult to comply with strict OELs due to their specific production processes. Member States with the strictest OELs might not host the industries having problems to comply with the strictest OEL. Secondly, industries are at different stages in their maturity and use varying technologies and processes. Thus, in Member States and industries with more advanced and automated production processes it would be easier to reach a low OEL. However, this cannot prevent the European Union to set a similar and adequate minimum level of protection across the EU. These considerations are taken into account in the analysis substance by substance.

All the OELs recommended by the ACSH, including the transitional limit values, as well as the OELs derived by the RAC, are covered by the different scenarios for each substance and are mentioned in the table 5 below. The OELs unanimously recommended by the three Interest Groups of the ACSH (Employers, Workers and Governments) have been established through scientific, technical and socioeconomic discussion. However, the costs and benefits of the combination of a targeting OEL and a transitional OEL have not been assessed in the COWI study. Therefore, it has not been possible to consider them as individual policy options. However, it is reasonable to consider that accompanying an OEL with transitional measures would have limited impacts on the benefits and relatively higher impacts on the costs. Furthermore, the OELs recommended by the ACSH as transitional OELs are covered by the policy options.

For acrylonitrile, an OEL at the level of the lowest limit value in the Member States and another OEL two times higher the OEL recommended by the ACSH and the RAC have been retained as policy options . These two policy options for acrylonitrile are reference points aiming to establish whether the OEL recommended by the ACSH is appropriate to follow.

As in the previous impact assessments, several other options than the retained policy options mentioned above have been discarded as they were considered as disproportionate or less effective in reaching the objectives of this initiative. More information about these discarded options is available in annex 9.

Table 5: Options matrix of OELs 46

6.What are the impacts of the policy options?

Different policy options presented in the options matrix in table 5 have been compared based on the methodology outlined in detail in annex 8. Other options than those supported by the ACSH in its opinions, such as the OEL recommended by RAC, are presented as reference points for the assessment of the ACSH options, to establish whether these are appropriate to follow.

Analytical methodology

The introduction of an OEL is expected to result in a reduction in the occupational exposure to the carcinogen concerned. The extent of such reduction depends on the current levels of exposure, as well as on the projected future levels of exposure in the absence of the proposed measure, i.e. the “baseline scenario”.

The baseline or "no policy change" option includes all relevant EU-level and national policies and measures which are assumed to continue being in force in the absence of further EU action. The baseline also takes into account as much as possible how the problem would evolve, considering legal provisions contained in the CMD but also all relevant societal, economic and technical developments that would probably occur in the following decades. For instance, the expected decrease of the use of petrol due to the gradual change to electric vehicles is taken into account. It also includes the existing national OELs, the current number of workers exposed and its evolution over time, the current and future exposure levels, the current risk management measures, the voluntary industry initiatives, the development of new technologies/growing use of substitutes, the future use of the substances and any other relevant factors.

For a given reduction in exposure levels, it is then necessary to estimate the expected decrease in the incidence of cancer and non cancer cases over a given timeframe to the carcinogen in question. This requires estimates of the risks of carcinogenicity and other adverse health effects, which can be derived from the existing toxicological and epidemiological literature, as well as information about the current level of worker exposure (number of workers, level, duration and frequency of exposure).

The health benefits of avoided cancer and non-cancer diseases registrations and deaths can then be expressed in monetary terms by applying standard evaluation methods 47 , in line with the Better Regulation Toolbox guidance. These health benefits of implementing new or revised OELs are calculated in terms of the costs of ill health avoided.

These monetised health benefits can in turn be compared to the expected monetary costs that would have to be incurred in order to comply with the proposed OEL. The estimate of the costs was made based on a literature research and data obtained from stakeholder contacts and take into account the following factors: the RMMs needed to comply with the proposed OEL, the costs of these RMMs for each company, the life span of the RMMs and the number of companies. The monetised health benefits are also compared to the other costs, including costs for public authorities, as well as to the potential effects on the market for the substance by the imposition of an OEL.

The benefits and costs of possible OELs are measured against the baseline, meaning that only marginal costs and marginal benefits are taken into account (for example, additional costs added to the current costs to comply with REACH regulations).

Given the complementarity between the regulation under REACH and the minimum requirements established through the CMD, the past and present REACH measures have been specifically taken into account in assessing the baseline and the impacts of the proposed CMD measures for all substances in this report.

More information about the analytical methodology, but also some analytical challengens arising from this cost-benefit analysis, are further discussed in annex 4.

6.1.Acrylonitrile

Baseline

Approximatively 10,000 to 33,000 workers in the EU are exposed to acrylonitrile, depending on the sources of data used for the calculation 48 .

The sectors using acrylonitrile that have been examined for this impact assessment are industrial manufacturing, manufacture of textiles, leather and fur, manufacture of bulk, large scale chemicals (including petroleum products), manufacture of fine chemicals, formulation of preparations and/or re-packaging (excluding alloys), manufacture of rubber products, manufacture of plastics products (including compounding and conversion), manufacture of computer, electronic and optical products, electrical equipment, building and construction work, health services, professional use, scientific research and development.

Acrylonitrile is not subject to authorisation according to Annex XIV of REACH but its use is restricted under Entry 28 of Annex XVII under the REACH Regulation 49 . However, this restriction only concerns the general public and not workers. More information about this restriction can be found in annex 10 of this impact assessment. There is not yet any OEL for acrylonitrile at the EU level.

It is assumed that the number of exposed workers remained constant in the past and will remain unchanged in the future. Annual declines of 3% 50 and 0% 51 in exposure concentrations are respectively assumed for the current and future disease burdens. In addition to that, a 5% staff turnover is taken into account for calculation purposes. The future disease burden only reflects cases occurring as a result of future exposure, leading to underestimated numbers of cancer 52 .

In the absence of any further action, up to 12 new brain cancer and 408 nasal irritation cases resulting from future exposure are estimated to occur in the coming 60 years, which would cost respectively up to €5,692,000 and €1,754,000. With regard to brain cancer, a 30 years latency period and a mortality rate of 80% are assumed. It should be borne in mind that only two health endpoints (brain cancer and nasal irritation) have been used in this impact assessment, which could lead to underestimate even further the exact number of cancer cases and other adverse health effects.

Table 6: Baseline scenario over 60 years for acrylonitrile

Impacts of the policy options

The table 7 below shows the multi-criteria analysis, summarizing both the monetised impacts as well as those that are assessed qualitatively.

In its opinion 53 , RAC derives a limit value of 1 mg/m³ (0.45 ppm) which assumes a mode of action-based threshold for the carcinogenic effects of acrylonitrile. RAC also recognises that “there may be occupational tasks at industrial sites presenting a short term acute exposure risk”. For that reason, It also derives a short-term exposure limit (STEL) of 4 mg/m³ (1.8 ppm). The costs and benefits for STEL values could not be monetised for acrylonitrile. However, the three Interest Groups in the ACSH opinion recommended the same OEL and STEL as RAC accompanied with a 4-years transition period starting from the entry into force of the Directive revising the CMD.

Although the main route of occupational exposure to acrylonitrile is by inhalation of the vapour, RAC noted that dermal exposure is also possible. Acrylonitrile can indeed readily penetrate the skin. For that reason, RAC recommended to set a skin notation. This skin notation has also been recommended by the ACSH.

Compliance cost levels for companies generally depend on the extent to which exposures are currently close to the required OELs. The higher the difference between the current exposure levels and the required OEL is, the higher the investments in RMMs will be. Since the exposure levels in the different sectors using acrylonitrile are already very low, the compliance costs for companies should be very limited.

Complying with an OEL of 1 mg/m³ as recommended by the RAC and the ACSH would cost to the employers €28,000,000 over a 60 years period, which corresponds to between 0.02% and 0.07% of the turnover, depending on the current average exposure and the size of the company. This is not considered to be significant enough as to force companies to cease operating. However, these compliance costs have to be entirely considered as one-off costs which will be dedicated to investements in LEVs. A transition period could therefore be necessary to mitigate the adverse effects of such upfront effort by enabling companies to anticipate the changes, more gradually introduce improvements and plan the necessary investments. In the event of the introduction of an OEL of 0.5 mg/m³, compliance costs would represent up to 0.12% of the turnover and could, therefore, have more consequences for companies.

Some of the costs included in the estimated compliance costs concern RMMs that may be required in any case to control exposure to other substances, leading to a possible compliance costs overestimation.

Given their lower turnovers, SMEs usually encounter more difficulties to comply with lower OELs than the large companies. For an OEL of 1 mg/m³, it is estimated that SMEs will have to invest up to 0.07% of their 60-years turnover while larger companies could only need to invest 0.02% of their turnover. The estimated costs for this scenario therefore remain relatively low and no SMEs are expected to close down. However, as mentioned above these costs will be one-off costs and a transition period could be needed for companies, especially SMEs that might need more time to manage the investments. In the event that the OEL would be set at 0.5 mg/m³, SMEs could face compliance costs of up to 0.12% of their turnover.

The impact of introducing any of the OELs on competitiveness are estimated to be relatively modest. Only a very limited number of companies interviewed mentioned that the introduction of an EU OEL could have a moderate or even significant impact. With regard to research and development (R&D) expenditures, even if the estimates of the costs arising from the implementation of the different OELs represent a relatively small percentage of overall turnover, they still represent an increase in costs compared to the current situation, which may slightly put the R&D expenditures under pressure.

As no closures have been estimated for any of the OELs suggested in this impact assessment, it is likely that there would be low or very limited impacts on workers in terms of employment. With regard to environmental aspects, the impact of setting an OEL for acrylonitrile should also be very limited. People living close to acrylonitrile productions are indeed exposed to very low levels of acrylonitrile in the air and the introduction of an EU-wide OEL should not lead to higher levels. In addition to that, the overwhelming majority of the companies dealing with acrylonitrile already uses waste water treatments to avoid any contaminations by water. 54

Setting an OEL of 2 mg/m³ or above would have very limited benefits for the workers and their families 55 and no benefits for the companies and the public sector. The benefits associated with an OEL of 1 mg/m³ or 0.5 mg/m³ would be exactly the same for the workers and their families as well as for the companies. The benefits for the public sector associated with an OEL of 0.5 mg/m³ would be slightly higher than with an OEL of 1 mg/m³.

Administrative costs 56 for Member States will depend on whether they already have a national OEL lower or equal to the proposed EU OEL for acrylonitrile. As mentioned in the table below, all Member States except Latvia would have to revise their national OEL in the event of option 2 or 3. According to the estimates, administrative costs would be of 50,000€ per Member State. In the event of option 4, only Latvia, Poland and Czech Republic would not need to transpose the new EU OEL.

Table 7: Multi-criteria analysis on acrylonitrile (all impacts over 60 years and additional to the baseline)

6.2.Nickel compounds

Baseline

Around 87,500 workers in the EU are estimated to be exposed to nickel compounds in the sectors considered in this impact assessment, namely oil refineries, pigments, frits, catalysts, glass, metals and alloys, metal surface treatment 57 , batteries, materials recovery and welding. Some sectors have not been considered in this impact assessment for the following reasons:

·Lack of source of information: agriculture;

·Marginal exposure or activity in the EU: mining and metal ores, manufacture of basic pharmaceutical products and pharmaceutical preparations, sewerage;

·Considered as covered by other sectors: manufacture of other inorganic base chemicals, manufacture of abrasive tools.

Lung cancer is the main cancer endpoint due to an exposure to nickel compounds. The exposure to this group of substances can also lead to other adverse health effects like pulmonary morbidity and miscarriage. Exposure to nickel compounds can also lead to sinonasal cancer. However, this endpoint is supposed to occur at concentrations above those at which lung cancer occurred, and has not been considered in this impact assessment.

Nickel compounds are not subject to authorisation according to Annex XIV of REACH. However, as indicated in Annex XVII of REACH (entry 27), the use of nickel and its compounds in jewellery (including watches) and articles that come into contact with the skin is restricted. More information about this restriction can be found in annex 10 of this impact assessment.In addition, all the nickel compounds and their mixtures cannot be supplied to the general public due to the restriction entry 28 of Annex XVII to REACH, similar to the previous case of acrylonitrile.

The exposure levels used to estimate the current and future burdens of disease vary from a sector to another 58 . It is assumed that the number of exposed workers remains constant both for the current 59 and future 60 burdens of disease. A mean latency period of 5 years for lung cancer (with a mortality rate of 80%) has been used for the calculations while it is assumed that there was no latency and a zero worker mortality rate for pulmonary morbidity and reproductive toxicity (miscarriage). The calculations enabling to estimate the number of miscarriage cases have been based on the workforce adjusted to represent those of reproductive age.

In the absence of any further action, 149 lung cancer, 718 pulmonary morbidity and 90 miscarriage cases could occur over the next 60 years, which would correspond to a total health cost between 53 and 125 million EUR.

Table 8: Baseline scenario over 60 years for nickel compounds

Impacts of the policy options

In its opinion on nickel and its compounds 61 , RAC derives an OEL for respirable fraction of 0.005 mg/m³ (to protect against lunger cancer) and an OEL for inhalable fraction of 0.03 mg/m³ (to protect against sinonasal cancer).

RAC also recommends a skin and respiratory sensitisation notation as occupational exposure to nickel compounds may result in contact sensitisation and in rare cases also sensitisation of the respiratory tract.

In the ACSH opinion, the three Interest Groups support the notations and, after considering also feasibility and socio-economic aspects, recommend both an OEL of 0.01 mg/m³ for the respirable fraction (which would apply from January 2025, after a transitional period) and an OEL of 0.05 mg/m³ inhalable fraction (with a transitional OEL of 0.1 mg/m³ for the inhalable fraction which would apply until January 2025).

The multi-criteria analysis indicating impacts and stakeholders affected is summarised in the table 9 below. Unless otherwise specified, the limit value mentioned in this impact assessment is based on the inhalable fraction. Due to limited availability of relevant data, the impact of changes related to the respirable fraction could not be fully assessed.

The compliance costs represent the needed investments in RMMs to comply with the proposed OEL. Complying with an OEL of 0.03 mg/m³ would cost €11 billion to the companies over the next 60 years. These compliance costs would decrease to €4.3 billion and €947 million for an OEL of 0.05 mg/m³ 62 and 0.1 mg/m³ 63 , respectively. The costs of compliance with an OEL of 0.05 mg/m³ and above represent a sustainable percentage of overall turnover, even for SMEs. Small companies operating in the oil refineries, welding and metals sectors would be the most affected and would have to invest 0.18%, 0.15% and 0.12% of their turnover over the next 60 years, respectively. However, these compliance costs would have to be entirely considered as one-off costs which would enable to invest in new RMMs. In the event of an OEL of 0.03 mg/m³, these compliance costs for companies in the oil refineries, welding and metal sectors would represent 0.37%, 0.31% and 0.31% of their turnover over the next 60 years, respectively.

To prove their compliance with an OEL, companies have to monitor the exposure of their workers. In order to comply with the OELs recommended by the ACSH, the companies could be required to also measure the exposure of the respirable fraction. The average cost of measurement for both the inhalable and respirable fractions, irrespective of the OELs level, would be of €152,000 every five-years over the next 60 years for large companies, €76,000 for medium companies and €38,000 for small ones. For small companies operating in the metal surface treatment sectors, this could represent more than 0.3% of their turnover. Therefore, it is important to take into account both the compliance and monitoring costs when assessing the potential impacts of each scenario.

As mentioned above, small companies operating in the oil refineries, welding and metals sectors would be the most affected in terms of compliance costs, as they would have to invest 0.18%, 0.15% and 0.12% (one-off costs), of their turnover in new RMMs to comply with an OEL of 0.05mg/m³, respectively. In addition to these compliance costs, these companies would also have to deal with monitoring costs representing 0.02%, 0.081% and 0.108% of their respective turnover over the next 60 years. Overall, the impact of both the compliance and monitoring costs may be higher for small companies in comparison with medium and large ones, especially as one-off costs will represent more than two thirds of the total costs. SMEs could therefore face problems when making the initial investments to comply with the new OELs. As requested by the Social Partners in the context of the two-phase consultation, socioeconomic considerations should be taken into account when adding or revising OELs. Therefore, the introduction of a transition period with a temporary higher initial value could be necessary to enable companies to anticipate the changes, gradually introduce improvements and plan the necessary investments.

The number of companies which might be forced to fully or partially cease activities varies with the level of the OEL. The stricter the OEL is, the higher the number of closures would be. At the strictest OEL of 0.03 mg/m³, around 90 companies (out of around 60.000) could be forced to close down, among them 80% would be small companies. The main sector impacted would be oil refineries where 61 companies (out of which 51 are small companies), representing 6% of the number of companies operating in this sector, could not be able to make the necessary investmets to comply with this OEL. With OELs of 0.05 mg/m³ and 0.1 mg/m³, this number will be much lower, or even close to zero.

In a letter 64 to the European Commission dated 7 October 2019, the Nickel Institute 65 encourages the Commission to follow the ACSH recommendation. Even if they recognised that these values (both inhalable and respirable fractions) are stringent and can be challenging, they remain committed to supporting their implementation.

As indicated in the Better Regulation Tool #21, “all compliance costs divert resources from other purposes, potentially including research and innovation”. Although these compliance costs represent a small percentage of the overall turnover, they might still lead to a decrease in the R&D expenditures. This could especially be the case for catalysts and batteries sectors that would need to invest substantial amounts of money in order to position themselves in the Li-ion battery value chain. However, the companies of both these sectors should face low compliance costs compared to their turnover.

Looking beyond the EU borders, the majority of the EU competitor countries have OELs for nickel compounds that are higher than 0.05 mg/m³ (including Australia, Canada, Japan South Korea, India, Brazil, USA and China for insoluble nickel compounds). The companies located in these countries could face lower costs due to the less strict measures to protect workers from the exposure to nickel compounds. Companies which also have plants in non EU countries with less stringent legal provisions could be tempted to relocate some of their operations located in the EU in these countries. However, these companies are often large companies for which complying with a new OEL for nickel compounds should be less challenging. Furthemore and as shown in the COWI study, the EU’s share in the global market for nickel metal dropped from more than 30% to less than 20% between 2008 and 2017 for the benefit of the Far East and particulary China. This seems to indicate that the level of the EU OEL will not be the only driver for the future trends for this global market.

The impacts on consumers in terms of price should be very limited. Only the metal sector, which would face higher compliance costs, could pass on a part of these costs to the price. In the event of companies passing on a part of the increase of their costs to the prices, this could have an impact on consumers which could try to buy these products from outside the EU.

The introduction of an OEL is estimated not to change the total environmental releases of nickel compounds significantly and is estimated not to have any significant environmental impact. However, two sectors in particular play a considerable role in the development of greener alternatives to fossil fuels, namely catalysts and batteries industries. Both these sectors require considerable R&D investments to satisfy, for example, the growing demand of Li-ion batteries for electric cars. The cost of complying with new OELs for nickel compounds at the levels below those recommended by the ACSH (i.e. option 3) might jeopardise this development.

Setting an OEL of 0.1 mg/m³ and 0.05 mg/m³ would enable to reduce the number of lung cancer cases by about 56% and 89% over the next 60 years, the number of pulmonary morbidity cases by about 72% and 98% and the number of miscarriages by about 57% and 89%, respectively. The setting of any OELs would bring marginal benefits to the employers and the public authorities as indicated in the table 9 below. The impact on other cancer endpoints could not be quantified, resulting in an underestimation of the benefits 66 .

Only five Member States, namely Bulgaria, Finland, Germany, Poland and Romania, have one single OEL for all nickel compounds. Nine Member States have no OEL at all: Cyprus, Czech Republic, Estonia, Greece, Italy, Luxembourg, Malta, Portugal and Slovenia. The fourteen remaining Member States have a mix of OELs for different nickel compounds, which can complexify the transposition of a single EU OEL for nickel compounds in their own legislation. According to the estimates, Member States with one single already existing OEL for nickel compounds should not face major administrative costs in case they would have to revise their own limit value. Member States with an existing mix of OELs would have to invest €20,000 to transpose the EU wide single OEL in their own legislation while Member States with no OEL would have to incur costs of €50,000.

Table 9: Multi-criteria analysis on nickel compounds (all impacts over 60 years and additional to the baseline)

6.3.Benzene

Baseline

Approximatively 1,000,000 workers in the EU are exposed to benzene in the sectors that are included in this impact assessment, namely upstream and downstream petroleum industry, coking plants, petrochemical industry distribution of petrol, retail and petrol stations, maintenance and repair of motor vehicles, foundries, laboratories and other sectors such as paint, adhesive, traffic, fires, etc.

Benzene is not subject to authorisation according to Annex XIV of REACH. However, according to entry 5 of Annex XVII of REACH, some uses of benzene are restricted. More information about these restrictions is available in annex 10.

It is assumed for the current disease burden that the number of exposed workers in the relevant sectors has been decreasing in the past by 3% 67 per year and the exposure concentration have been decreasing by 7% per year. These past trends are partly explained by the effects of the introduction of an EU OEL for benzene in the CMD in 2004. Annual declines of 2% for the number of workers exposed to benzene and 0% in exposure concentrations are assumed for the future disease burden. This null trend for the future exposure concentrations is based among others on information gathered from the relevant stakeholders in the context of the COWI study. The 2% annual decrease is explained by an estimated decrease in the use of petrol due to the gradual change to electric vehicles. For the estimates of the current and future burdens of disease, a mean latency of 5 years for leukaemia and 0 years for leukocytopenia as well as a mortality rate of 80% for leukaemia and 0% for leukocytopenia have been used. Without any medical treatment, leukocytopenia can lead to weakened immune defence and increased risk of contracting infectious diseases.

The main cancer effect of benzene is leukaemia, a cancer of the blood system. Benzene also causes other hazardous effects on the bone marrow. However, due to lack of data, only leukocytopenia has been taken into account in the calculation of the burdens of disease. Other effects such as lymphocytopenia, neutrocytopenia and thrombocytopenia could not be quantified and monetised.

In the absence of any further action, up to 300 cancer cases and 237 leukocytopenia cases are estimated to occur in the coming 60 years. This has been estimated to have a health cost between 202 and 331 million euro.

Table 10: Baseline scenario over 60 years for benzene

Impacts of the policy options

In its opinion on benzene 68 , RAC derived an OEL of 0.05 ppm (0.16 mg/m³) under which there should be no significant residual cancer risk or other adverse effects. It also recommends a skin notation for benzene as the dermal route can be an important contributor to total benzene exposure in certain situations.

The table 11 below shows the multi-criteria analysis, summarizing both the monetised impacts as well as those that are assessed qualitatively.

Compliance cost levels for companies generally depend on the extent to which exposures are currently close to the required OELs. The higher the difference between the current exposure levels and the required OEL is, the higher the investments in RMMs will be. For the purpose of adequately assessing the economic feasibility for companies to comply with the different OEL scenarios, the annualised compliance costs in percentage of the turnover were calculated.

Overall, complying with an OEL of 0.05 ppm would cost to the employers 0.17% of their turnover every year over the next 60 years. Companies operating in the upstream petroleum sector, distribution and foundries would be the most impacted by facing higher compliance costs of about 0.24%, 0.64% and 0.29% of the turnover respectively. The average compliance costs related to an OEL of 0.2 ppm 69 across all the sectors would be of about 0.07 % of the average turnover and relatively higher for the upstream petroleum sector (0.08%), as well as the distribution (0.35%).

For the specific case of the foundry sector, employers would face relatively higher one-off compliance costs. Companies operating in this sector would have to initially invest 3.41% and 1.1% of their turnover to comply with an OEL of 0.05 ppm and 0.2 ppm, respectively. However, as mentioned in annex 3, one-off costs would represent one third of the total costs for companies. As requested by the several employers and workers’ representatives in the context of the two-phase consultation, socioeconomic considerations should be taken into account when adding or revising OELs. The introduction of a transition period with a temporary higher initial OEL could enable companies to anticipate the changes, gradually introduce improvements and plan the necessary investments, softening in this way the economic impact on them.

As this industry is confronted with global competition, companies will not be able to pass on these costs to the users further down the value chain or consumers. At the strictest OEL, it is likely that companies operating in this sector would lose competitiveness, especially compared to China which represent the largest global suppliers in this sector (45% of the global market) and where companies have only to comply with an OEL of close to 2 ppm. Any other OELs equal or above the limit value recommended by the ACSH should not have any negative impacts on the competitiveness of this industry.

The Member States with the highest number of companies operating with benzene across the EU are Germany, Italy, France, Poland and Spain. More information about the share of companies by Member States in each sector is available in annex 6.

The setting of a stricter EU wide OEL for benzene will have an impact on companies’ cost. The companies’ costs increase arising from the needed investments in RMMs in order to meet a stricter OEL could potentially weight negatively on the R&D expenditures. These impacts would depend on the level of the OEL and the sector in which the company is active.

The impact of introducing any of the envisaged OELs on employment is very limited. Only the foundry industry, which could lose global market shares at the strictest OEL, could be concerned by potential consequences on the employment. With regard to the environmental impacts due to the introduction of a stricter OEL, it is estimated that it would result in positive impact due to reduced emissions to the environment by reducing the fugitive and diffuse emissions from a number of the main sources such as upstream and downstream petroleum sectors and the coking sector.

Setting an OEL of 0.5 ppm 70 and 0.2 ppm would enable to reduce the number of leukaemia cases by about 30% and 60% over the next 60 years respectively and the number of leukocytopenia cases by about 43% and 80% respectively. Concerning the two other envisaged scenarios, they would enable to reduce further the number of leukaemia and leukocytopenia, although to a lesser extent than for the two first scenarios discussed in this paragraph. The setting of any OELs below the current OEL (baseline scenario) would bring marginal benefits to the employers and the public sector as indicated in the table 11 below. A number of non-cancer health endpoints could not be quantified, resulting in an underestimation of the benefits 71 .

Although an EU wide OEL already exists for benzene, many Member States would need to revise their national OEL whatever the scenario. As indicated in the table 11, all the Member States would need to revise their OEL in the event of an EU OEL at the level of option 2. The Member States which would not need to revise their limit value for the two other scenarios are referenced in the table below. Considering that all Member States have already an OEL for benzene, the cost of transposing a revised OEL would be lower compared to the costs of transposing a new OEL. This cost has been estimated to €10,000 per Member State that would have to revise the national OEL.

Table 11: Multi-criteria analysis on benzene (all impacts over 60 years and additional to the baseline)

7.How do the options compare?

The main objective is to balance health considerations against economic impacts, by proposing OELs that are still economically feasible while protecting a maximum number of workers.

This impact assessment assessed the different OEL levels without any transitional period. However, the ACSH identified sectors that might find it initially difficult to comply with the OEL proposed, and thus suggested to include transitional periods. The duration of these periods was established by the governments, workers and employers, as they have the expert knowledge about technological development in different sectors. It is assumed that these transitional periods are necessary for companies to develop their production processes to be able to comply with the OELs proposed. They should attenuate short-term negative impacts, as the necessary investments in protective measures would be spread over a longer period of time.

The aim is therefore to ensure a balanced approach and to prevent industries from closures or severe disadvantages in particular Member States due to e.g. adopting the most stringent OELs while providing an adequate protection of the workers at the EU level.

The comparison tables used to compare the different options against the baseline scenario in terms of effectiveness, efficiency and coherence apply the following ranking symbols: '0' – baseline, '≈' – similar to baseline, '+' more efficient/effective or coherent than baseline; '++' – much more efficient/effective or coherent than baseline; '-' – less efficient/effective or coherent than baseline; '- -' – much less efficient/effective or coherent than baseline.

With regard to the effectiveness, the options are analysed among others from the perspective of the prevention of deaths and other adverse health effects.

With regard to the efficiency, the options are analysed among others on the basis of how balanced they will be between adequate protection of workers at the EU level and prevention of closures and other severe disadvantages for the industries. The efficiency is also analysed from the perspective of the costs/benefits ratios. However, although an option has the best costs/benefits ratio, it does not automatically mean that the protection it will bring to workers will be considered as sufficiently adequate or that the resulting costs for businesses will be disproportionate.

With regard to the coherence, the options are analysed on the basis of how coherent they are with other EU policies (including the Charter for Fundamental Rights, the EU Pillar of Social Rights, the Europe’s Beating Cancer Plan and REACH). Coherence with general EU priorities and policies, as well as with the Charter of Fundamental Rights, goes hand in hand with the level of the OELs. The lower the OEL, the more protective for workers’ health. Furthermore, coherence will be also based on to what extent the opinion of the ACSH has been taken into account. Indeed, as laid down in the Council decision 72 setting up an ACSH, this Committee “shall have the task of assisting the Commission in the preparation, implementation and evaluation of activities, in the fields of safety and health at work”.

3.

4.

5.

6.

7.

Acrylonitrile

With regard to effectiveness, options 2 and 3 will have the same impact on prevention of deaths and other adverse health effects compared to the baseline and the other options. Option 4 brings less benefits compared to options 2 and 3.

With regard to efficiency, all the options except option 1 have a costs/benefits ratio higher than 1. Option 4 has the lowest costs/benefits ratio, followed by option 3. Option 2 has a relatively high costs/benefits ratio compared to the previous two. However and as mentioned in introduction of this Chapter, the aim is to ensure a balanced approach between providing an adequate protection of workers at the EU level while preventing industries from closures or severe disadvantages. By recommending OELs and transitional measures through scientific, technical and socioeconomic discussion, the tri-partite ACSH is probably the best guarantee of such a balanced approach. Therefore, although all the options - except the baseline scenario - have costs/benefits ratios higher than 1, option 3 is considered as representing the best the balance between the adequate protection of workers and bearable costs for the companies. Nevertheless, as unanimously agreed by the ACSH, this balance would require a transition period of 4-years to make sure that companies could make the necessary investments.

With regard to coherence, establishing an OEL following options 2 and 3 also increases the coherence of the CMD with other EU policy objectives, including the Charter for Fundamental Rights, the EU Pillar of Social Rights and the Europe’s Beating Cancer Plan. It increases complementarity with REACH, as outlined in section 4. It will also provide legal clarity and a common reference point that can be used as a practical tool by employers (particularly important for SMEs), workers and enforcers to assess compliance with the general CMD requirements, and will also contribute to a level-playing field for businesses across the EU (see section 3). For the specific case of acrylonitrile, the strictest OEL (option 2) does not bring more benefits to workers and their families and to employers than the ACSH’s recommended OEL. Based on the criteria explained above, option 3 is the most coherent as it ensures the same coherence with the other EU policies as option 2 while being coherent with the opinion of the ACSH.

Option 3, with a STEL of 4 mg/m³ (1.8 ppm), a skin notation, and with a 4-years transitional period starting from the entry into force of the new Directive is therefore the preferred option.

Table 12: Comparison of options for acrylonitrile (mg/m³)

Nickel compounds

With regard to the effectiveness, option 2 and 3 would enable to drastically reduce the risks of work-related cancer and non-cancer diseases. Option 4 would result in a substantial reduction of the number of lung cancer, pulmonary morbidity and miscarriage cases.

With regard to the efficiency, the costs outweight the benefits for all the envisaged options, except option 1. If option 4 has the lowest costs/benefits ratio, costs associated with option 3 are much lower than those that would result from adopting option 2. Furthemore, all the Interest Groups in their ACSH opinion 73 strongly recommended the Commission to adopt a revised OEL, with a preference for option 3. This option will ensure a balanced approach between adequate protection of workers at the EU level and prevention of closures and other severe disadvantages for the industries. However, a transition at the level of option 4 has been considered as necessary to enable companies to more swiftly make the necessary investments for the purpose of reaching such a level of workers’ protection. For all these reasons, option 3 could be seen as an medium-term objective with a transitional value at the level of option 4. Option 2 is considered as less efficient.

With regard to coherence, option 2 would be the most coherent with other EU policy objectives, including the Charter for Fundamental Rights and the EU Pillar of Social Rights, followed by option 3 and option 4. All the options increase complementary with REACH, as outlined in section 4. Taking into account the opinion of the ACSH (option 3 with a transition period at the level of option 4) increases also coherence as mentioned in introduction of this Chapter. Furthermore, as mentioned in the ACSH opinion, nickel compounds and chromium VI compounds are frequently occurring in the same sectors and, often, in the same processes. Therefore, aligning the measures for nickel compounds to those for chromium VI adopted in the context of the first revision of the CMD and that will apply in January 2025 would bring more coherence, both in terms of investments for companies and between the updates of the CMD.

Option 3 (0.05 mg/m³ - inhalable fraction), with a corresponding OEL for respirable fraction of 0.01 mg/m³, a skin and respiratory sensitisation notation, and with a transitional value until the 17th January 2025 included at the level of option 4 (0.1 mg/m³ - inhalable fraction), is the preferred option.

Table 13: Comparison of options for nickel compounds (mg/m³ - inhalable fraction)

Benzene

With regard to the effectiveness, option 2 will have the most positive impact on prevention of occupational exposure-related leukaemia and leukocytopenia compared to the baseline scenario. However, any other envisaged OELs would significantly reduce the number of leukaemia and leukocytopenia cases..

With regard to the efficiency, even if an EU-wide OEL for benzene is already set in the CMD, all the Interest Groups in their ACSH opinion 74 strongly recommended the Commission to adopt a revised limit value. Option 3 has been considered as the most balanced option between adequate protection of workers at the EU level and prevention of closures and other severe disadvantages for the industries. However, the ACSH considered as important to foresee a transitional value at the level of option 4 from 2 up to 4 years after the entry into force of the Directive in order to give enough time to the companies to make the needed investments. Option 4 has the lowest costs/benefits ratio, followed by option 3. Option 2 is the less efficient option with much higher compliance costs than options 3 and 4. All these options have a costs/benefits ratio higher than 1.

With regard to coherence, option 2 will ensure the highest coherence of the CMD with other EU policy objectives, including the Charter for fundamental rights and the EU pillar of Social Rights, followed by option 3 and option 4. All the options increase complementary with REACH, as outlined in section 4. Coherence also depends on to what extent the opinion of the ACSH is taken into account, as explained in introduction of this Chapter., The ACSH agreed on an OEL at the level of option 3 as well as a transitional value at the level of option 4 which would enable companies to make the necessary investments.

Option 3, with transitional value at the level of option 4 from two up to four years starting from the entry into force of the new Directive and the confirmation of the skin notation, is therefore the preferred option.

Table 14: comparison of options for benzene (ppm)

8.Preferred options

8.1.Summary of the preferred options

It has been shown in the previous sections that the impacts of the considered measures for the protection of workers vary significantly for the different substances assessed in this report.The table below summarises the preferred options.

Table 15: Summary of the preferred options, by assessment criteria

*From the date of entry into force of the Directive amending the CMD

**The limit values for the respirable and inhalable fractions shall apply from 18th January 2025. Until then, a limit value for the inhalable fraction of 0.1 mg/m³ shall apply while no limit value for the respirable fraction shall apply.

8.2.Overall impact of the package or preferred options

Although the recent coronavirus pandemic is expected to have substantial socioeconomic consequences, it is still too early to assess with any degree of certitude the magnitude of these impacts on the business, especially at the sectoral level. According to International Monetary Fund’s (IMF) World Economic Outlook of April 2020 75 , the EU-27 Gross Domestic Product (GDP) growth could decrease from 7.1% in 2020 compared to 2019 and could rebound to 4.7% in 2021. However, IMF mentions that “there is considerable uncertainty about what the economic landscape will look like when we emerge from this lockdown”. Given the large degree of uncertainty, these impacts have not been included in the impact analysis below.

8.2.1.Impact on workers

The retained option package for the three substances or groups of substances considered in this impact assessment (henceforth “the preferred option”) should result in benefits in terms of avoided work-related cases of cancer and other serious illness, and related monetised health benefits including avoidance of intangible costs such as the reduced quality of life, the suffering of the workers and their family, the pain, etc.

This impact assessment is limited to assessing the most sensitive cancer endpoint and the most sensitive other adverse health effects for each substance. Nevertheless, the chemical agents under consideration pose a range of other occupational hazards for which the available data is not sufficient to estimate the magnitude of the global related benefits for workers. As a result, and taking into account general estimates of costs related to these diseases, benefits for workers and society are likely underestimated.

The greatest assessable benefits are expected in relation to nickel compounds and benzene. The retained option would indeed result in:

·Acrylonitrile: up to 12 brain cancer cases, 408 nasal irritation cases prevented and a monetised health benefit of €440,000-€5,800,000.

·Nickel compounds: 133 lung cancer cases, 702 pulmonary morbidity cases and 80 miscarriage cases prevented, and a monetised health benefit of €72-92 million.

·Benzene: 182 cases of leukaemia and 189 cases of leukocytopenia prevented, and a monetised health benefit of €121-198 million.

The costs and benefits resulting from the combination of several OELs for one substance have not been assessed in the COWI study. For that reason, the benefits mentioned above are those calculated for the preferred option for each substance without any transition measures. However, it is reasonable to consider that accompanying an OEL with transitional measures would have limited impacts on the benefits and relatively higher impacts on costs. It is all the more true that the duration of these transitional measures will be limited and it is likely that some companies have already started to make the necessary investments in anticipation of the new or more stringent OELs to come.

8.2.2.Impact on business

As regards costs incurred by enterprises for risk reduction measures, the preferred option will affect operating costs for companies which will have to put in place additional protective and preventive measures. This will be particularly the case for nickel compounds and benzene, where the total costs (compliance and monitoring costs) to industry of the retained option over the next 60 years are estimated to be about €6.5 billion and €7.61 billion respectively. Companies dealing with nickel compounds could indeed face additional monitoring costs for measuring the respirable fraction, which is not the case at the moment. However, these figures have to be put into perspective. Concerning benzene, companies, irrespective of their size or sector, will need to dedicate 0.07% of their annualised turnover each year. With regards to nickel compounds, it is estimated that the total costs per company will be of €108,000 over 60 years. In addition to that, in a letter 76 to the European Commission dated 7 October 2019, the Nickel Institute encourages the Commission to follow the ACSH recommendation. Even if the Nickel Institute recognizes that the ACSH recommended values are stringent and can be challenging, they remain committed to support their implementation by the time provided by the legislation. For these reasons, the cost per company in relative terms for both substances is expected to be sustainable, although some SMEs concerned by the use of nickel compounds might face more difficulties to comply with the preferred option as discussed in the section dedicated to the impact on SMEs. In addition to that, transitional measures are foreseen for acrylonitrile, benzene and nickel compounds so that companies would have more time to make the necessary investments while already improving the protection of workers. Indeed, although the annualised cost per company is expected to be sustainable, most investments in RMMs would take place early in the 60 year period considered in the calculations. For the specific case of nickel compounds, the transition period that ends in January 2025 will ensure alignment with the measures adopted for chromium VI, since both group of substances (nickel compounds and chromium VI compounds) are frequently occurring in the same sectors and, often, in the same processes. As unanimously recommended by the ACSH 77 , reducing the exposure to nickel compounds and chromium VI compounds must be coordinated and can benefit from synergies.

As discussed in section 6.3, in the case of benzene, companies operating in the foundry sector will face relatively higher one-off compliance costs compared to their turnover than in the other sectors. As shown in table 70 in annex 6, these companies are mainly located in Italy (18,7% of all the EU foundries), Germany (13,6%) and Czech Republic (12,9%).

Impact on SMEs

For nickel compounds and benzene, SMEs represent a large proportion of the relevant industries while companies using acrylonitrile are mainly large companies. For the specific case of nickel compounds, 95% of the companies are even small companies. Therefore, SMEs specificities, their limitations and particular challenges have been duly taken into account in the overall analysis presented in section 6.

Many of the RMMs required to meet the OELs involve capital expenditure, and SMEs might face higher cost of finance compared to large companies. However, the analysis has shown that in most cases, costs which will be incurred by SMEs dealing with acrylonitrile and benzene are not significant. Some SMEs concerned by the use of nickel compounds and operating in the oil refineries, welding and metals sectors might face more difficulties to comply with the preferred option. As shown in table 60 in annex 6, more than 1,000 companies dealing with nickel compounds in these three sectors are small companies while 200 are medium companies. Therefore, transition periods with higher initial values are necessary to ensure that it is possible for companies to anticipate the changes, gradually introduce improvements and plan the necessary investments.

Moreover, introduction of an OEL will only have a significant impact on companies which have not yet made the investments to protect workers either through closed systems or substitution of the substances where technically feasible.

Impact on competition and competitiveness

The preferred option would have a positive impact on competition within the internal market by decreasing competitive differences between firms operating in Member States with different national OELs and providing greater certainty concerning enforceable exposure limit across the EU. Regarding the specific case of nickel compounds, the capital expenditures required for start-ups to ensure compliance with the EU OELs could make their entry into the market more difficult.

Even if the retained option has often more stringent OELs compared to the European Union’s main competitor countries, it should not have a significant impact on the external competitiveness of EU firms. The detailed assessment provided above shows indeed that in most cases additional compliance costs per firm are not significant although some SMEs operating in the oil refineries sector might face more difficulties to comply with the preferred option, as discussed above.

It has however to be noted that the foundry sector dealing with benzene, which will face relatively high one-off compliance costs with the retained option and operates in a global market, could face more difficulties. The European foundry sector currently holds a global market share around 10-15% and has to compete with China (which holds 45% of the global market), India and the USA. While India and the USA have OELs that are lower than or similar to the current EU OEL of 1 ppm 78 , in China, companies have to comply with an OEL of 1.9 ppm. European companies in this sector could therefore face a greater competition due to the retained option. However, these possible impacts could not be estimated. Moreover, the preferred option has a transition period which will give more time to the companies to adjust to these changes.

It should also be noted that OELs established in the non EU countries cannot necessarily be compared to the EU limit values. OEL setting methods and the implementation of OELs differ substantially across jurisdictions as a result, for example, of different approaches to whether and how socioeconomic factors may be taken into account, differences in legal enforceability or expectations regarding compliance, use of scientific evidence and analytical method, industrial relations and roles played by industry, worker representatives, and others. As a result, caution should be exercised in making comparisons and drawing conclusions regarding values which may not be directly comparable.

8.2.3.Impact on environment

The impact of the package of the preferred options on environment depends on the substances. Setting an EU-wide OEL for acrylonitrile and nickel compounds will not lead to higher releases in the environment and have therefore no impact. With regard to benzene, lowering the existing OEL at the EU level will even reduce the fugitive or diffuse emissions in some sectors as explained in section 6. Revising the limit value at the level of the preferred option for benzene would therefore have positive impacts on the environment. The introduction of the preferred option for acrylonitrile will not lead to further installation of LEVs that might lead to increased emissions in the air. None of the potential RMMs to comply with the preferred option for acrylonitrile are expected to lead to significant changes in the releases of acrylonitrile to water. The introduction of the preferred option for nickel compounds should not significantly change the total environmental releases of nickel compounds.

In addition to the direct impacts on environment, the introduction of an OEL could also lead to indirect impacts. As mentioned in section 6, nickel compounds play a considerable role in the development of greener alternative to fossil fuels. The Nickel institute considers that the need for nickel and its compounds will keep growing as they support “the development of solutions and innovations to address some of the most pressing challenges, such as energy storage and supply, sustainable, recyclable products and green transport” 79 . The compliance costs arising from lowering the OEL to the level of the preferred option should however not have any impacts on the innovative capacity of the sectors using nickel compounds to develop these solutions and innovations.

While an EU OEL for acrylonitrile should not lead to higher releases in the environment and have therefore no significant impact, the acrylonitrile industry also has a role to play in developing greener technologies for some sectors. It is however difficult to say at this stage whether this positive impact will be significant. For example, this substance is widely used to produce high performance elastomers/rubbers, structural resins and plastics, which are essential for the automotive sector’s transition to lightweight, fuel efficient and electric vehicles 80 . Acrylonitrile is also essential to produce water purification coagulants 81 and is key to produce carbon fibres used in wind turbine blades’ manufacturing.

8.2.4.Impact on Member States/national authorities

Member States with established OELs at the level of the retained option will be less affected than those having higher or no OEL in place. Each multi-criteria analysis above lists, for each option, the Member States that will have to enforce a stricter OEL. More details and national OELs for all substances are provided in annex 5. The preferred option should contribute, although not significantly, to reduce the costs related to occupational diseases for social security systems. Additional administrative and enforcement costs might be incurred by enforcing authorities. These costs are not quantifiable as the granularity of Member States' reporting of enforcement activity is not sufficient to distinguish costs related to a particular OEL.

However, it is not expected for the costs to be significant. OEL enforcement will take place according to already existing mechanisms for compliance improvement and enforcement, including informal conversations with employers as well as formal correspondence and legal enforcement action.

Usually, national inspectors organise visits in companies to ensure the employers’ compliance with several OSH provisions (for example, workplace transport, slips and trips, machinery safety, stress) rather than only checking the conformity with the OELs. Specific reporting would only be the case where Labour Inspectorates undertake targeted chemical carcinogen enforcement activity and OEL campaigns. Costs will therefore be generally affected by Labour Inspectorate resourcing, prioritisation and targeting. No assumption may be made that enforcement, a Member State competence, will receive (or demand) greater resourcing and priority as a result of an OEL being set.

At the same time, establishing OELs, and other explicit references to a given substance in the CMD brings clarity regarding legal requirements, and so facilitates the work of inspectors by providing a helpful tool for compliance checks. Setting OELs at EU level would also limit the need for national administrations to carry out the costly and burdensome necessary work for establishing those limit values at national level.

8.2.5.Impact on fundamental rights

The impact on fundamental rights is considered positive, in particular with regard article 2 (Right to life) and article 31 (Right to fair and just working conditions which respect his/her health, safety and dignity).

8.2.6.Subsidiarity, proportionality and REFIT

In view of the available scientific evidence it is necessary to establish both new OELs for a number of substances for inhalation exposures and information on other routes of exposure (e.g. dermal) which could contribute significantly to the overall body burden of the workers. The protection of workers health against risks arising from exposure to carcinogens is already covered by EU legislation, in particular by Directive 2004/37/EC (CMD), which can be amended at EU level after a two-stage consultation of the social partners. The preferred option takes into account long and intensive discussions with all stakeholders (representatives from workers’ associations, representatives from employers' associations, and representatives from governments), including consideration of socioeconomic feasibility.

Updating the CMD to take account of newer scientific evidence is an effective way to ensure that preventive measures would be updated accordingly in all Member States, providing a uniform level of minimum requirements designed to guarantee a better standard of health and safety. Action taken by individual Member States in response to available technical data would risk increasing divergences between Member States with potential competition on the basis of OELs set at different levels. Business would therefore continue to compete on an uneven playing field, which would hamper the operation of the internal market. Updating the CMD therefore complies with the principle of subsidiarity.

The proportionality principle is fully respected as the preferred option is limited to setting out the limit values for three additional agents by amending Annex III to the Directive on the basis of the scientific and technical data available, as provided by Article 16 (1) of the CMD. This initiative aims to make a step forward to achieve the objectives set to improve living and working conditions of workers.

With regard to the preferred option, socio-economic and feasibility factors have been taken into account after intensive discussions with all stakeholders within the ACSH (representatives of workers’ organisations, representatives of employers’ organisations, and representatives of governments). It also includes measures for mitigating burdens and supporting compliance (including transition periods) which have also been discussed by the relevant stakeholders. These transitional measures contribute to the proportionality of the preferred options by ensuring a more appropriate temporal margin for businesses to adapt.

As shown above, the costs outweight the quantified benefits for all the preferred options. However, the selection of the most appropriate option cannot be done merely on the basis of a costs/benefits ratio comparison. Indeed, the main general objective of this initiative is to ensure to the workers the right to a high level of protection of their health and safety at work. This objective is in line with the European pillar of Social Rights, which enshrines workers’ right to healthy, safe and well-adapted work environment, and with the upcoming Europe’s Beating Cancer Plan. As mentioned in Chapter 7, the aim of this initiative is to ensure a balanced approach, i.e. to prevent industries from closures or severe economic disadvantages while providing an adequate protection of the workers at the EU level. The preferred options are considered balanced and justified in light of the accrued and longer-term benefits in terms of reducing health risks arising from workers' exposure to carcinogens and saving lives. Despite their high costs, the preferred options have the support of the key stakeholders in the area of OSH, namely the employers, employees and national governments representend within the ACSH. It also needs to be noted that, when considering the costs per company and, in particular when comparing the expected costs to the turnover of the average company, the estimated burden is overall sustainable.

Furthermore, the preferred options also offer a certain margin of flexibility to Member States. In accordance with Article 153(4) of the TFEU, setting OELs at the EU level does not prevent Member States from maintaining or introducing more stringent protective measures (i.e. lower limit values). However, Member States cannot set higher limit values than the EU OELs set in the Annex III of the CMD.

Finally, regarding the simplification and the efficiency improvement of the existing legislation, the preferred option eliminates the need for Member States to conduct their own scientific analysis to establish OELs for the three substances and brings clarity regarding the acceptable levels of exposure, facilitating the work of inspectors by providing a helpful tool for compliance checks. Employers also benefit from the simplification in ensuring legal compliance, particularly those operating in different Member States.

9.How will actual impacts be monitored and evaluated?

9.1.Monitoring arrangements

The table below presents the core indicators for each operational objective and the data sources for the monitoring of the core indicators.

Table 16: Indicators and monitoring arrangements/data sources

A two-stage compliance assessment (transposition and conformity checks) will be carried out by the Commission for the transposition of the limit values. At workplace level, there is an obligation for employers to ensure that the exposure does not go above the limit values set out in Annex III to the Directive. The monitoring of application and enforcement will be undertaken by national authorities, in particular the national labour inspectorates. At EU level, the Committee of Senior Labour Inspectors (SLIC) informs the Commission regarding problems relating to the enforcement of Directive 2004/37/EC.

While collection of reliable data in this area is complex, the Commission and EU-OSHA are actively working on improving data quality and availability so that the actual impacts of the proposed initiative could be measured in a more accurate way and additional indicators could be developed in the future (e.g. in relation to mortality caused by occupational cancer). Ongoing projects include cooperation with national authorities on the European Occupational Diseases Statistics (EODS) data collection.

Legislative action needs to be followed up through effective implementation at the workplace. In this context, EU-OSHA has carried out a Healthy Workplaces Campaign on dangerous substances in 2018-2019 pursuing several objectives, including raising awareness of the importance of preventing risks from dangerous substances, promoting risk assessment, heightening awareness of risks to exposure to carcinogens at work or increasing knowledge of the legislative framework. Under the framework of this campaign, EU-OSHA has provided a broad range of tools, information and good practices that will support the implementation of this Directive.

9.2.Evaluation arrangements

In accordance with Article 17a of Directive 89/391/EEC, every five years, Member States are required to submit a report to the Commission on the practical implementation of the EU OSH Directives, including Directive 2004/37/EC. Using these reports as a basis, the Commission is required to evaluate the implementation of Directive 2004/37/EC and, to inform the European Parliament, the Council, the European Economic and Social Committee and the Advisory Committee on Safety and Health at Work of the results of this evaluation and, if necessary, of any initiatives to improve the operation of the regulatory framework.

Given the data challenges explained earlier, it is suggested to make use of the next relevant evaluation exercise, after the end of the transposition period, to define the baseline values (benchmark) that will allow assessing the effectiveness of further amendments of the CMD. Evaluation of the practical implementation of the proposed amendments could possibly be based on the following period (2023-2027).

Annex 1: Procedural information

1.Lead DG, Decide Planning/CWP references

Lead DG: Directorate-General Employment, Social Affairs and Inclusion, Unit B/3 Health and Safety.

2.Organisation and timing

A first consultation of the Occupational Safety and Health Inter-services Steering Group (OSH ISG) on the draft Impact Assessment Report (IAR) was launched on 27 February 2020 with a deadline for comments on Friday 20 March 2020. Twelve services (SG, SJ, BUDG, GROW, ENER, ENV, RTD, CNECT, EAC, SANTE, JUST, ESTAT) as well as EU-OSHA have been consulted. Most of the comments provided by the services have been addressed in the revised draft IAR. A table summarising the key comments and the way DG EMPL addressed them in the revised IAR was circulated.

A second consultation of the OSH ISG on the revised draft IAR was launched on 30 March 2020 with a deadline for comments on Tuesday 14 April 2020. This second consultation was carried out with the same services than for the first consultation. DG EMPL took most of these comments into consideration in the revised version of the draft IAR. This revised version of the draft IAR was sent to the Regulatory Scrutiny Board (RSB) on 24 April 2020.

3.Consultation of the RSB

The RSB opinions for the first three CMD amendments have been taken into account when carrying out this Impact Assessment. The draft IAR for this initiative was submitted to the RSB on 24 April 2020 and the meeting with the RSB has taken place on 27 May 2020. Following this meeting, the RSB gave a positive opinion with reservations.

The table below summarises the RSB comments as well as the revisions introduced in response to them:

4.Evidence, sources and quality

Risk Assessment Committee’s Opinions

The assessment of health effects of the carcinogens subject to this proposal is based on the relevant scientific expertise from ECHA’s Committee for Risk Assessment (RAC).

RAC prepares the opinions of the European Chemicals Agency (ECHA) related to the risks of substances to human health and the environment. RAC examines among others the proposals for harmonised classification and labelling, evaluates whether the proposed restriction on manufacture, placing on the market or use of a substance is appropriate in reducing the risk to human health and the environment, and assesses the applications for authorisation of chemicals. Moreover, opinions from RAC also support Union regulatory activity in the field of occupational safety and health. More information about what this committee does can be found on the website of ECHA 83 .

RAC develops high quality comparative analytical knowledge and ensures that Commission proposals, decisions and policy relating to the protection of workers’ health and safety are based on sound scientific evidence. Based on a Service Level Agreement (SLA) signed by DG EMPL and ECHA, this Committee assists the Commission delivering scientific evaluations, upon request, on the toxicological profiles of each of the selected priority chemical substances in relation to their adverse health effects on workers. These scientific evaluations shall, where appropriate, include proposals for Occupational Exposure Limit values (OELs), biological limit values/biological guidance values and/or notations. Based on such opinions, the Commission will propose occupational exposure limits for the protection of workers from chemical risks, to be set at Union level pursuant to Council Directive 98/24/EC and Directive 2004/37/EC of the European Parliament and of the Council.

Members of RAC are highly qualified, specialized, independent experts selected on the basis of objective criteria. They provide the Commission with Recommendations and Opinions that are helpful for the development of EU policy on workers protection.

For the purpose of this initiative, the Commission services have used the relevant chemical agent-related RAC opinions which are summarised in the following table:

Table 17: Summary of the RAC opinions

Studies performed by external consultants

The Commission launched a call for tender on 9 May 2018 an open call for tender 89 in order to carry out an assessment of the social, economic and environmental impacts of a number of policy options concerning the protection of workers health from risks arising from possible exposure to a certain number of substances at the workplace, including acrylonitrile, nickel compounds and benzene.

The contract started on 3 September 2018 and lasted 11 months. The outcome of this study provides the main basis for this Staff Working Document and is summarised in the relevant sections of this document.

Annex 2: Stakeholder consultation

The following consultation activites have been performed :

1.Social Partners Consultation: as requested by the TFEU Article 154, a formal two-stage consultation of the social partners at EU level is required prior to submitting proposals in the social policy field. Such a two-stage consultation has been performed in 2017. The first phase of social partners’ consultation closed on 30 September 2017 where 3 substances have been identified for this initiative. The second phase consultation closed on 22 December 2017 and confirmed these 3 substances as to be addressed in this initiative. More information about these two-stage consultation is provided below in this annex 2.

2.Tripartite consultation (ACSH): the tripartite Advisory Committee on Safety and Health (ACSH), composed of three full members per Member State, representing national governments, workers' and employers' organisations, is consulted on regular basis. It gives, taking into account the input of the RAC as well as socio-economic and feasibility factors, opinions which are used to prepare the Commission's proposal. More information about this tripartite consultation is provided below in this annex 2.

3.Consultation of other stakeholders (e.g. industry of employees associations specifically concerned): These consultations have been carried out in the context of the COWI study in order to collect detailed information on the potential impacts of establishing or revising OELs under the CMD that is not available in published literature and internet searches.

In line with the previous three amendments of the CMD, no public consultation on this initiative has been launched for the following reasons:

·A broad consultation of various stakeholders, social partners and Member States’ competent authorities has been carried out in view of this initiative.

·This initiative concerns a very technical topic for which the general public does not have sufficient expertise. For that reason, a more targeted consultation has been considered as a more proportionate approach.

·In the context of the scientific opinions carried out by RAC, stakeholders were allowed to express their views and concerns in the early phases of developing the scientific reports on occupational exposure limits for acrylonitrile, nickel compounds and benzene.

Due to the exceptional circumstances related to the coronavirus pandemic and the delay caused in the legislative procedure concerning this initiative, no inception impact assessment has been published.

1.Social Partners Consultation

1.1.Results of the first phase of the Social Partners consultation.

The first phase of Social Partners consultation closed on 30 September 2017.

The Commission consulted the Social Partners among others on the establishment and/or revision of further binding OELs in Annex III to the CMD.

Following a process described in more detail in annex 8 of this report, the Commission identified a first proposed list of priority substances for a subsequent amendment revision of the CMD in the first phase consultation document 90 , as follows:

a)Nickel compounds under the scope of the CMD

b)Acrylonitrile [CAS No 107-13-1]

c)Benzene [CAS No 71-43-2]

Workers' organisations

Three trade unions replied to the consultation: the European Trade Union Confederation (ETUC), European Confederation of Independent Trade Unions (CESI), European Federation of Building and Woodworkers (EFBWW). They all acknowledged the importance of the existing legislation and a need for further action.

The workers' organisations agreed, broadly, with the issues described in the consultation document and confirmed the importance they attach to protecting workers from the health risks associated with exposure to carcinogens and mutagens.

Concerning the approach regarding the fourth amendment, ETUC and EFBWW agree with the list of 3 priority substances identified by the Commission. CESI considers that the latest available data need to be used when revising the CMD.

As regards the other substances to be added to Annex III, while CESI suggests that they should be identified on the basis of sound and independent scientific research, ETUC and EFBWW insist that the target of binding OELs for 50 substances has to be achieved by 2020. ETUC has proposed a priority list of such substances. After 2020, the process of setting OELs should continue on a dynamic way in order to include most of the substances at the workplace. In the enclosed annex of its priority substances list ETUC has indicated it as a candidate for the fourth amendment.

With regard to Annex I to the CMD, ETUC considered important to include all processes generated substances for which IARC monographs are available.

CESI and EFBWW considered that legislative initiatives should be complemented by other measures, for example, fostering preventative health-oriented behaviour and information on best available technology.

The workers do not want to enter into negotiations under Article 155 TFEU concerning the third and fourth amendment of the CMD and urge the Commission to make progress on this.

Employers' organisations

Four employers' organisations replied to the consultation: BusinessEurope, the European Association of Craft Small and Medium-sized Enterprises (UEAPME), the European Chemical Employers Group (ECEG) and the Council of European Employers of the Metal, Engineering and Technology-based industries (CEEMET).

They supported the objective to effectively protect workers from occupational cancer, including by setting OELs at EU level. However, they also raised some concerns about the approach taken when setting such limit values.

The employers in principle supported further revisions of the CMD, subject to certain conditions. In their opinion, binding OELs should be set for priority substances only. The process of OELs setting should be based on sound scientific evidence, technical and economic feasibility, socioeconomic impact assessment and opinion of the tripartite ACSH. While employers considered that the Commission’s criteria for prioritising substances are relevant, they suggested that the criteria of technical and economic feasibility should also be included. BusinessEurope and CEEMET emphasized that proposing a series of substances on the basis of unofficial lists should be avoided, as should setting an arbitrary numerical target of additional binding OELs without clear criteria of prioritisation. UEAPME and CEEMET stressed the need to assess impact on SMEs and consider sectoral differences. Employers also highlighted a need to ensure coherence with other EU chemicals legislation and suggested that guides, examples of good practice and other tools can assist in implementing this Directive.

For subsequent amendments BusinessEurope stressed that inclusion of specific substances should depend on whether they meet the conditions / criteria mentioned above and whether the preparatory work has been completed. ECEG and CEEMET supported the overall process for developing and adopting binding OELs as long as the above criteria and processes are correctly applied. UEAPME, noted that without having seen concrete proposals for OELs it is not possible to take a complete position. They further suggested that the latest available data need to be used when revising the CMD (supported by CEEMET) and that too restrictive OELs could be very burdensome for employers leading to a risk of non-compliance.

The employers do not want to enter into negotiations under Article 155 TFEU as the existing preparatory procedures already involve Social Partners. Although, they are open to discussing in an informal way relevant issues.

1.2.Results of the second phase of the Social Partners consultation

The Commission launched a second phase consultation of the Social Partners which closed on 22 December 2017. In this second phase consultation, the Commission indicated among others that a first candidate list of the following substances is given consideration:

·Nickel compounds that are carcinogens as defined in the Directive

·Acrylonitrile [CAS No 107-13-1]

·Benzene [CAS No 71-43-2]

·Diesel Engine Exhaust Emissions (DEEE). At least two approaches are being explored – to address this mixture as a process generated substance or to take a component-specific approach 91 .

Workers’ organisations

Three workers' organisations replied to the second phase consultation: the European Public Service Union (EPSU), the European Trade Union Confederation (ETUC) and the European Federation of Building and Woodworkers (EFBWW). They all recognised the importance of further improving the existing legislative framework in line with the proposed Commission action and beyond.

Concerning Annex III, the workers' organisations agree that the first candidate list for the fourth batch is appropriate, as it includes substances that are among priority carcinogens. However, they underline that the list must be extended to reach the objective of setting 50 OEL by 2020. As regards the setting of OEL, ETUC points the need to define a consistent and more transparent methodology, asking the Commission to consider its priority substances list. EFBWW underlines that the OELs should be health based only, while ETUC stresses that socioeconomic considerations might be relevant for adapting Annex III, although not for Annex I. As a final point, EPSU considers that including some cytotoxic drugs in Annex III would be possible, but other actions are needed to tackle the exposure to these substances.

The workers' organisations agree that the amendment of the CMD should be part of a global strategy to prevent occupational cancer in Europe and urge the Commission to adopt a roadmap combining legislative initiatives (e.g. the revision of the Asbestos Directive and the revision of the Optical Radiation Directive) with non-legislative actions and mainstreaming work-related cancer prevention in other EU policies.

The workers' organisations do not want to launch a negotiation procedure pursuant to Article 155 TFEU.

Employers' organisations

Four employers' organisations replied to the second phase consultation: BusinessEurope, the European Association of Craft Small and Medium-sized Enterprises (UEAPME), the European Chemical Employers Group (ECEG) and the Council of European Employers of the Metal, Engineering and Technology-based industries (CEEMET). They confirmed their support to actions aiming to effectively protect workers from occupational cancer, including the setting OELs at EU level but underlined the need to ensure values that are proportionate and feasible.

The employers' organisations support the procedures for considering substances and setting OEL, underlining the need that they are based on the latest scientific information, proportionate and measurable and highlighting the importance of the tripartite system. UEAPME and CEEMET point that the limit values have to be set in a way which reduces workers exposure whilst still allowing SMEs to comply. CEEMET also proposes transitional measures where the new OEL will adversely affect industry and points that an arbitrary target of the addition of 50 new exposure limits should not be set, as OEL should only be proposed on an evidence-based approach and not in line with the precautionary principle.

Regarding the scientific body to provide information for the setting of OELs, BusinessEurope thinks that there should be a thorough assessment before any decision is taken, while CEEMET believes that SCOEL should be the one setting the limit values.

BusinessEurope encourages the Commission to continue its preparatory work and consultations with the ACSH regarding the candidate list of substances for the fourth batch.

Finally, ECEG, BusinessEurope and CEEMET stress the importance of other actions to achieve worker protection in addition to legislation, such as guidance documents, best practices, voluntary product stewardship programmes by companies and sectors, or social partner agreements.

The employers' organisations do not want to engage into negotiations under Article 155 TFEU.

2.Consultation of the ACSH/WPC

The Advisory Committee on Safety and Health at Work (ACSH) has adopted opinions for acrylonitrile, nickel compounds and benzene, in the context of the fourth amendment of the CMD.

The ACSH is proposing as possible approaches for these chemicals one or several binding OELs with additional notations for all of them.

The opinions for all substances adopted by the ACSH are summarised below.

Acrylonitrile

The ACSH adopted its opinion for acrylonitrile 92 on 4 June 2019, putting forward the following values and transition measures:

·An 8 hours TWA limit value of 1 mg/m³ (0.45 ppm) and a STEL of 4 mg/m³ (1.8 ppm) that would apply after a four year transition period starting from the entry into force of the new Directive.

The ACSH strongly recommended the Commission to adopt as soon as possible a binding OEL for this substance under Directive 2004/37/EC.

The ACSH also agreed on the Biological Limit Value (BLV) as proposed by RAC 93 but note that, at present, BLVs are not proposed under the CMD. It suggested that this BLV is taken into consideration when developing the guidance on the practical use of biomonitoring.

Nickel compounds

The ACSH adopted its opinion for nickel compounds 94 on 4 June 2019, putting forward the following values and transition measures:

·Two 8 hours TWA limit values of 0.01 mg/m³ (respirable fraction) and 0.05 mg/m³ (inhalable fraction) that would apply from 17th January 2025. Until then, an 8 hours TWA limit value of 0.1 mg/m³ (inhalable fraction) would apply.

·The ACSH indeed recommended that the date of application ot the OELs for nickel compounds should ensure alignment with the date of application of the OEL for chromium (VI) compounds adopted in Directive 2017/2398/EU 95 . Both groups of substances (nickel compounds and chromium VI compounds) are indeed frequently occurring in the same sectors and, often, in the same processes. The ACSH is therefore of the opinion that actions to reduce the exposure to chromium VI and nickel compounds must be coordinated and can benefit from synergies.

The ACSH also added that in a limited number of sectors or processes, including specifically smelting, refineries and welding, face particular difficulties for complying with the proposed OELs. Therefore, the ACSH believes that in these sectors or processes, and possibly in other sectors, there may be need for using RPE to ensure that the workers are appropriately protected.

The ACSH strongly recommended the Commission to adopt as soon as possible OELs for this group of substances under Directive 2004/37/EC. It also agreed that it would be appropriate to introduce an OEL under the Directive 98/24/EC 96 for Nickel.

Benzene

The ACSH adopted its opinion for benzene on 4 June 2019 97 , putting forward the following values and transition measures:

·Two years after entry into force of the Directive, a limit value of 0.5 ppm (1.65 mg/m³) will apply. Four years after entry into force of the Directive, a limit value of 0.2 ppm (0.66 mg/m³) will apply.

In addition to these values and transition measures, the ACSH agreed that the steel foundries sector faces particular difficulties for complying with the proposed OELs. Consequently, the ACSH believes that in this sector, and possibly in other sectors, there may be a need for using RPE to ensure that the workers are appropriately protected.

The ACSH strongly recommended the Commission to adopt as soon as possible a revised binding OEL for this substance under Directive 2004/37/EC.

The ACSH also agreed on the usefulness of biomonitoring (BLV) as proposed by RAC 98 but note that, at present, BLVs are not proposed under the CMD. The ACSH therefore suggested this BLV to be taken into consideration when developing the guidance on the practical use of biomonitoring.

3.Consultation of other stakeholders

In the context of the COWI study, consultation activites have been carried out to collect detailed information on the potential impacts of modifications to the CMD that is not available in published literature and internet searches. Although some information on OELs is available, limited information is available on concrete measures already in place and that would need to be implemented, should the OELs be modified. The information sought via consultation therefore included sizes of companies, sectors and processes that would be affected, number of workers exposed, current air concentrations of substances concerned (both 8-hour time weighted averages (8-h TWA) and 15-minutes reference periods), risk management measures currently in place, as well as risk management measures that would need to be implemented should the OELs be modified and associated costs.

Consultation carried out for the purposes of this study consisted of the following main activities:

•Questionnaires;

•Email requests (possibly in combination with questionnaires);

•Telephone interviews;

•Site visits;

•Face to face meetings;

•Workshops;

Mixed methods (combining e.g. questionnaire responses with telephone interviews and site visits) were adopted to ensure that a large number of organisations and individuals were able to provide data and provide their views within the time constraints and resource limits. Using mixed methods also enabled the study team to gather varying details of information and to explore information further where the need arose.

3.1.Targeted Online Questionnaires

Stakeholders were initially contacted via email. The e-mail provided an overview of the study and a link to the questionnaires. If stakeholders preferred to answer the questionnaire in a Word document (so that it could be shared among several colleagues, for example), the stakeholder was informed of the possibility to obtain a word-version upon request.

Three separate questionnaires were drawn up, each one created to gather information from different stakeholder groups:

•Questionnaire 1 was aimed at companies whose workers were exposed to nickel compounds, acrylonitrile or benzene;

•Questionnaire 2 was aimed at occupational health and safety experts; and

•Questionnaire 3 was aimed at Member State authorities.

The questions aimed to collect information on processes during which worker exposure to the substances in question is likely to occur, risk management measures that are already in place, current exposure concentrations, risk management measures that would need to be implemented should the limit be lowered, and any other impacts that could result from the introduction of EU level limits.

Although many of the responses provided a significant amount of useful information, many of them were not sufficiently detailed. Other methods of consultation, allowing experts to question and probe answers further (namely telephone interviews and site visits), were therefore required to obtain a more in-depth understanding of the potential impacts. This includes the above follow-ups.

3.2.Telephone interviews

Both national experts and substance experts were activated for the purposes of the telephone interviews. Telephone interviews were asked for in the online questionnaires as well as through direct email and phone contact.

The purpose of the telephone interviews was to gain more insight into the answers provided in response to the questionnaires. It enabled the collection of more detailed information on processes, to pinpoint exactly where exposure is likely to occur, investigating what types of risk management measures are already in place and how effective they are, as well as what risk management measures would be required if limits were lowered and other potential ramifications for the company.

3.3.Email requests

As supplement to the interviews various information was collected by email requests. The purpose and questions were similar to those explained above for telephone interviews.

3.4.Site visits

Companies whose activities are likely to be affected by the potential modifications to the CMD were also asked whether they would be willing to welcome members of the study team for a site visit. Companies to be visited, were identified via the questionnaire or the contact was established via EU trade associations.

The purpose of the site visits was to a gain a more operational understanding of the risk management measures currently in place to protect against exposure to the substances concerned, as well as of the risk management measures that would be needed should the CMD be modified.

Detailed notes from each site visit were drafted and sent back to the company to ensure that the information recorded was accurate. This process also enabled the company to add more detail and information to the study, where possible, and to confirm the level of confidentiality required to the information.

3.5.Face-to-face meetings

Face-to-face meetings were held with key EU associations for each of the substances. This aimed among other things to:

•facilitate the collection of more comprehensive data on the relevant sectors, and

•facilitate exchange of information collected as part of the associations’ own socioeconomic assessments.

3.6.Workshops

In order to collect information from stakeholders and receive feedback, the consultant attended workshops with industry and workshops of the Working Party of Chemicals as listed in the table below.

Table 18: Workshops attended

Source: COWI (2019)

3.7.Stakeholders targeted

The following table summarises information on stakeholder groups targeted and the interests represented. The table demonstrates that all relevant stakeholder groups have been reached out to.

Table 19: Stakeholders targeted and interests represented

Source: COWI (2019)

Annex 3: Who is affected and how?

1.Practical implications of the initiative

Acrylonitrile

Table 20: Practical implications of the initiative for acrylonitrile

Benzene

Table 21: Practical implications of the initiative for benzene

Nickel compounds

Table 22: Practical implications of the initiative for nickel compounds

2.Summary of costs and benefits

Acrylonitrile

The table 23 below summarises the benefits as calculated on the basis of Method 1, which relies on Willingness to Pay (WTP) values for avoiding a case of mortality and morbidity. A low-high range has been provided that represents the lowest and highest values estimated in this study based on the different Method 1 approaches and assumptions. Estimates on the basis of Method 2, which relies on monetised Disability Adjusted Life Years (DALYs), are of a similar order of magnitude at OEL levels of 1 mg/m3 inhalable.

Table 23: Overview of benefits for acrylonitrile

Note: Estimates are relative to the baseline as a whole (i.e. the impact of individual actions/obligations of the preferred option are aggregated together).

Table 24: Overview of costs for acrylonitrile

Note: Estimates are relative to the baseline as a whole (i.e. the impact of individual actions/obligations of the preferred option are aggregated together).

Nickel compounds

The table below summarises the benefits as calculated on the basis of Method 2, which relies on monetised Disability Adjusted Life Years (DALYs). Estimates on the basis of Method 1, which relies on Willingness to Pay (WTP) values for avoiding a case of mortality and morbidity, are approximately 70-80% of those under Method 1.

Table 25: Overview of benefits for nickel compounds

Note: Estimates are relative to the baseline as a whole (i.e. the impact of individual actions/obligations of the preferred option are aggregated together).

Table 26: Overview of costs for nickel compounds

Note: Estimates are relative to the baseline as a whole (i.e. the impact of individual actions/obligations of the preferred option are aggregated together).

Benzene

The table below summarise the benefits as calculated on the bases of Method 1, which relies on Willingness to Pay (WTP) values for morbidity. Estimates on the basis of Method 2, which relies on monetised Disability Adjusted Life Years (DALYs), were about 60% of the values calculated with Method 1.

Table 27: Overview of benefits for benzene

Note: Estimates are relative to the baseline as a whole (i.e. the impact of individual actions/obligations of the preferred option are aggregated together).

Table 28: Overview of costs for benzene

Note: Estimates are relative to the baseline as a whole (i.e. the impact of individual actions/obligations of the preferred option are aggregated together).

Annex 4: Analytical methods

1.Monetisation of the health impacts 

1.1.Health impacts

The introduction of an OEL is expected to result in a reduction in the occupational exposure to the carcinogen concerned. The extent of such reduction depends on the current levels of exposure, as well as on the projected future levels of exposure in the absence of the proposed measure, i.e. the 'baseline scenario'.

For a given reduction in exposure levels, it is then necessary to estimate the expected decrease in the incidence of cancer cases and other non cancer health effects over a given timeframe to the substance in question.

The current and future cases of ill health have been estimated for both cancer and non-cancer endpoints using the following inputs:

·The Exposure Risk Relationships (ERRs) and Dose Response Relationships (DRRs);

·The numbers of workers exposed;

·The exposure concentrations; and

·Trends in the exposed workforce and exposure concentrations.

On this basis, we can therefore calculate the health impact which can be defined as the number of persons (“cases”), either suffering from cancer and/or experiencing some noncancer health effects due to this occupational exposure.

It has to be kept in mind that

·the ERR only applies on the most critical cancer site, which is given by the assessment of the European Chemicals Agency / Committee for Risk Assessment (ECHA/RAC), and only comment qualitatively on further cancer sites, which may be linked to exposure to the respective substance;

·the DRRs are derived by referring to the most critical non-cancer effects quantitavely. The effects, which were regarded as the most critical ones by RAC are selected and we only comment qualitatively on further non-cancer effects, which may be linked to exposure to the respective substance

·as there is even less scientific consensus on the increase of effect severity with increasing exposure concentration and the respective data are often not adapted to the workplace exposure scenario, we focus on the fraction of workers affected at the different exposure levels, when we establish a DRR, without taking into account the increase of severity of effects.

Therefore, the calculated health impact (e.g., in terms of “number of estimated cases with health impairments”) is not identical to the “real” health impact, but is just an approximation, which may underestimate the full impact of the occupational exposure to the respective substances. However, there are other uncertainties leading to under- or overstimates. These are further developed further in this annex.

Exposure Risk Relationships

The starting point for a cancer risk impact assessment is the OEL proposed by RAC and the respective RAC opinion, together with the annexed background report. For the three substances, RAC found significant arguments for a mode-of-action based threshold and used this concept for proposing OELs. No conclusions regarding incidence and severity of effects above the OEL is presented in these reports.

The OEL proposed by RAC is used to define a zero response, i.e., 0 % of the exposed individuals are assumed to suffer from the respective health effect, if exposed for all of their working life time to this OEL. RAC does not derive any ERR for the range above the proposed OEL. As the ERR is essential for the benefit assessment of the current study, an ERR has been established for this impact assessment.

As part of this, it has been assessed which of the existing dose-response relationships is the most adequate, following the mechanistic conclusions in the RAC opinion. Based on RAC’s assessment we use, where available, sublinear exposure risk relationships of high quality, which follow a similar mode-of-action argumentation as RAC. They are used to describe cancer risks above the OEL.

If the ERR is not already provided for a working lifetime exposure scenario, the respective transformation has to be calculated: working life time is assumed to be 40 years, with work day exposure for 8 hours/day, 5 day/week in 48 weeks of a year. It is a conservative estimate based on the most critical cancer site (cancer risk associated with highest risk).

Dose Response relationships

For non-cancer endpoints, the RAC opinion as well as other recent evaluations and literature reports are scrutinized to identify the most relevant endpoints for humans. Relevance means that existing information makes it likely that effects might occur in humans at exposure levels above the OEL proposed by RAC, which are of relevance in European industries. Human data are preferred over experimental animal data. Experimental data are used as supportive information only, where sufficient human dose-response information is available to derive a DRR. Where insufficient human information is available, we have to rely on experimental animal data.

Data from original toxicological and epidemiological studies, referenced by RAC or national committees as being qualified and demonstrating a dose-response, are selected and searched for effect levels linked to a specific fraction of the exposed (humans or animals). If not contradicted by the overall weight of evidence, this slope reported in such a study is adopted for the DRR. If effects are reported on a continuous scale, this needs to be transformed to quantal data (i.e. the incidence of effects in the exposed population), which often requires certain assumptions.

In case animal data are used, data are transformed to the human situation, e.g. by applying allometric scaling factors in accordance with the ECHA Guidance on Information Requirements and Chemical Safety Assessment, R.8; Characterisation of dose [concentration]-response for human health 99 .

The number of workers exposed

It is important to calculate the number of workers potentially exposed to a substance in order to calculate the potential benefits of implementing any new measures. For each substance discussed in this Impact Assessment Report, several datasets have been identified.

As soon as these datasets have been identified, there is a need to select those which are the most representative of the current reality. For example, only datasets on acrylonitrile at the EU level have been selected. Indeed, it is likely that there are significant differences between the Member States in terms of numbers of workers exposed. Therefore, extrapolating the number of workers exposed at the EU level on the basis of figures at Member States level would lead to wrong estimates.

1.2.Monetisation of the health impacts

Specific guidance is provided in the Better Regulation (BR) Toolbox for health impacts (BR Tool #31). This is summarised in the table below.

Table 29: Better Regulation Toolbox on social impacts

Source: COWI (2019)

Focusing on the example of cancer, the costs of cancer can be divided into:

·Direct costs: These are the costs of healthcare, in other words, the medical costs associated with the treatment of cancer and other costs, including non-medical costs, that arise directly as a result of cancer, for example those related with care and the costs to employers. Healthcare costs are those associated with the treatment and services patients receive, including the cost of hospitalisation, surgery, physician visits, radiation therapy and chemotherapy/ immunotherapy. Depending on the structure of national health care provision, these costs may be borne fully or partially by the government (tax payers). Direct medical costs associated with cancer vary significantly by cancer type and also vary over time. Indeed, it has been noted that cancer costs are highest in the initial period following diagnosis and, among patients who die from their disease, at the end of life; they are lowest in the period between the initial and end of life periods, following a “u-shaped” curve (Yabroff et al., 2012) 100 . Other direct costs may be incurred by the patients (say the cost of transport to attend appointments) but also by their family/friends, for example, through providing unpaid care.

·Indirect costs: These are the monetary losses associated with the time spent receiving medical care, including productivity losses due to time spent away from work or other usual activities and lost productivity due to premature death. Employers might also bear costs indirectly through inter alia loss of output; payments related to sick leave; administrative costs related to a worker’s absence; additional recruitment costs; loss of experience/expertise; overtime working; compensation payments (although this may be covered by some form of employer’s liability insurance); and insurance premiums. Depending on the national structure of social security provision, the government (tax payers) may also bear the costs of any disability/social security payments and will also suffer losses through foregone tax receipts.

·Intangible costs: These include the non-financial ‘human’ losses associated with cancer, e.g. reduced quality of life, pain, suffering, anxiety and grief.

In economic impact terms, the total social costs 101 of ill health are measured by the costs borne for health care provision, together with lost output (including productivity losses), gross wage and non-wage labour costs of absent workers (such as loss of experience), administrative costs and the intangible costs. These represent the direct and indirect resource costs and the non-market ‘external’ costs of illness. The other costs listed above (e.g. insurance premiums) relate to what are commonly referred to as ‘transfer payments’, which do not give rise to net welfare effects. As a result, they are not considered in economic analyses, even though they may be important in financial terms to an individual worker or an employer.

1.3.The Model

1.3.1.Introduction

The following table provides a first overview on the expected key endpoints per substance.

Table 30: Carcinogenic and non-carcinogenic endpoints

Source: COWI (2019)

The key model inputs are summarised below. The inputs are those parameters whose variation changes the results and for which the model is run multiple times to derive a benefits curve.

Table 31: Key model inputs

Source: COWI (2019)

In addition to the inputs, the model is underpinned by a range of default assumptions regarding the onset of the disease and its effects. These assumptions differ by substance but do not change depending on the variations in the input data. Some of these assumptions are a simplification of complex real life scenarios or best estimates (where authoritative evidence could not be identified from readily available literature).

The key areas in which assumptions had to be made to enable the calculations are set out below.

Table 32: Key assumptions and their consequences for the sensitivity analysis

Source: COWI (2019)

The model provides a good approximation of the order of magnitude of the expected impacts and the core calculations are supported by sensitivity analysis. The outputs of the model include:

•The number of new cases for each health endpoint assigned to a specific year in the 60 year assessment period;

•The Present Value (PV) of the direct, indirect, and intangible costs of each case.

1.3.2.Key model inputs

I.Rx: estimate of the risk or fraction of workers affected

The estimate of the risk or fraction of workers affected:

·For cancer: Exposure-Risk Relationship (ERR) i.e. excess risk of developing cancer due to lifetime occupational exposure to a substance (taken here to mean 40 years); and

·For non-cancer endpoints: Dose-Response-Relationship (DRR), i.e. the proportion of workers that will develop an endpoint when exposed to a certain level.

II.ExW: Exposed workers

Several scenarios are modelled for the exposed workforce. It is not possible to take into account all the complexities of real life workforce changes and these scenarios are theoretical constructs/simplifications which are designed to provide order of magnitude estimates without the need to construct a very large number of scenarios to cover all the types of workforce dynamics.

Two distinct issues are usually covered under the term ‘turnover’. Primarily, turnover refers to the natural turnover rate resulting from workers leaving their employer and new workers joining. In addition, it can refer to the turnover triggered by those that are absent from work due to illness and replaced by others.

However, turnover refers to the rate by which workers change employment type as it is considered that often workers when leaving one employer are employed for similar work by another employer. As consequence workers may be exposed for longer time in similar jobs that indicated by the average times workers are employed by the individual employer.

It is assumed that there is a turnover of 5% per year. The 5% per year is lower than the turnover ratios in most of the published literature and Eurostat, which are typically derived at the level of individual companies rather than sectors.

III.Cx: Exposure concentration

The method has been applied for all substances under CMD 4 as follows:

•Exposures are for each sector or sub-sector represented by log-normal distributions fitted to available data on e.g. median values and 95th percentiles.

•The exposure distributions are considered to represent the variation in the exposure of each worker and among workers;

•For all concentrations below the threshold (if any), the effect is 0;

•By introduction of the OELV, it is assumed that the 95th percentile should be reduced to a level at or below the OELV;

•As a result of the implemented measures, the general exposure levels would be reduced at a rate comparable to the reduction in the 95th percentile i.e. the entire distribution is changed. The new percentiles can be calculated as: percentilenew = percentileold * OEL / 95thold. By this, it is assumed that the variance of the lognormal distribution is the same before and after introduction of the OEL, but the AM is reduced with the same ratio as the 95th percentile.

•Some 5% of the exposure levels of the new percentile will be above the OEL. In some of the previous CMD projects it has been assumed that all exposure above the OEL is reduced to the OEL, but it follows from the 95th percentile model that some measured concentrations would still be above the OEL.

In order to reduce the possible overestimation of the effect of introduction of an OEL, to the extent the available data allow, specific distributions should be developed for subsectors, SEGs, specific work areas (less well defined than the SEGs), regional sectors (e.g. different distribution for MS with an OEL and MS without an OEL), etc. The challenge, however, have not been to obtain exposure concentration datasets for different SEGs, but to obtain detailed data on number of workers within the different SEGs. In practice, it has been the availability of data on workers by SEG or subsectors which has been determining for the necessary aggregation into sectors.

When establishing the various exposure distributions, the existing OELs in MS have been taken into account. Consequently, it has not been necessary to take the existing OELs into account when estimating the effects of introduction of an OEL. For nickel compounds, however, due to complexity and the fact that the national OELs address different fractions, it has not be possible to fully reflect existing OELs in the exposure distributions.

1.3.3.Key assumptions

I.MinEx and MaxEx – The minimum exposure duration required to develop the endpoint

The model assumes that no cases arise until the minimum exposure duration required to develop the endpoint (MinEx) has expired, see table 33 below. The default MinEx is two years for cancer, a standard assumption for a chronic condition. The minimum exposure periods in the table below have been derived using a precautionary approach that maximises worker protection. The MaxEx reflect the time needed to reach the maximum risk (i.e. after the MaxEx has been reached, the risk of effects do not increase).

Table 33: Minimum & maximum exposure duration to develop a condition (MinEx & MaxEx)

Source: COWI (2019)

II.Dist – the distribution of cases between start of exposure and Year 59

Valuing the cost of occupational illness involves applying discounted costs to future cases which requires that the estimated cases over a 60 year period are assigned to specific years. However, the ERRs and DRRs developed under this study are for 40 years of exposure.

‘Dist’ refers to the distribution of cases between start of exposure and Year 59, also taking MinEx into account. This differs between endpoints. The main difference is between cancer and non-cancer endpoints;

III.Cancer

For reasons of simplicity, the following approach is used to distribute the total 40-year cancer risk (i.e. not incidence but risk since incidence is delayed due to latency) over the 60 year period: It is assumed that no risk arises until MinEx has expired. It is assumed that, subsequently, the distribution is linear, i.e. 0% of the excess risk arises in year 2 and 100% of the excess risk arises by year 40. The annual number of cases after MinEX is then used as a proxy for the number of cases until 60 years.

Figure 4 - Brain cancer risk of exposure to acrylonitrile – distribution over time (source: COWI study (2019))

IV.Default for non-cancer endpoints

While the further calculation is straightforward for the cancer endpoints, the estimations are more challenging for the non-cancer endpoints.

Typically, the fraction affected achieves that predicted by the DRR as soon as MaxEx expires and remains constant over the 40 year DRR period (although the certainty of the ‘fraction’ estimated on the basis of the DRR increases towards the end of the period). As a default assumption, two years has been chosen as a conservative MaxEx.

With a MinEx of 0 year and a MaxEx of 2 years for cancer endpoints, the fraction affected that is calculated on the basis of the DRR is the same between 2 years and 40 years and increases in a linear manner between Year 0 and Year 2.

Figure 5 - Non-cancer endpoints – fraction affected over time - example with a MaxEx of 2 years (source: COWI (2019))

V.Latency

Cancer endpoints

Similar to the approach used in CMD 3, by way of simplification, a single latency value is used for the calculation of the core scenario. According to Rushton et al. (2012), all solid tumours are expected to have a latency of 10-50 years, meaning that the average latency is 30 years. A latency of 30 years is used as a default for cancer endpoints.

As noted in the methodological note for CMD 3, 40 years of exposure and 30 years of latency would translate into a 70-year assessment period which is longer than the assessment period used in the CMD 3 study and the current study. In order for the assessment not to underestimate the benefits (longer latency reduces the benefits since they are discounted at lower factors) the CMD 3 study and the previous Impact Assessments for the OELs under the CMD which relied on an assessment period of 60 years, used a latency period of 10 years.

In order to avoid underestimations of the current burden of disease and avoid inconsistences in the applied latency periods used for estimations of the current and future burden of disease, latency periods in accordance with those reported in the literature have been applied. Latency periods for the cancer endpoints are shown in the table below.

Table 34: Latency (Lat) periods of cancer endpoints

Source: COWI (2019)

Non-cancer endpoints

The estimated latency period for the non-cancer endpoints in this study is 0 years. There is very limited evidence for latency of the relevant non-cancer conditions.

Table 35: Latency (Lat) periods of non-cancer endpoints

Source: COWI (2019)

VI.ModEX - the modelled exposure duration

The ERRs and DRRs are for a 40 year period. The modelled exposure duration is thus 40 years.

Whilst it is unlikely that a single worker is exposed to a substance at a constant concentration throughout their whole working life, the 40 year period has been chosen in order to be protective to workers by assuming a worst-case scenario. The evidence used for the development of the ERR means that the greatest certainty about the ERR is at lifetime exposure, i.e. 40 years.

It is highly likely that the real exposure duration is shorter than ModEx (the modelled exposure duration) and this have been taken into account by use of the staff turnover for the estimations as described elsewhere.

VII.MoR – mortality rate

Mortality rate as a result of the relevant condition is important since different monetary values are applied to mortality and morbidity. The mortality rates used in the model are given below.

Table 36: Mortality rate (MoR)

Source: COWI (2019)

VIII.Treatment period

The treatment periods used in the model are given below. The end of the treatment period signifies either a fatal or illness-free outcome.

Table 37: Treatment period

Source: COWI (2019)

IX.Monetary value of the relevant endpoint

The approach to the monetisation of ill health effects is based on the following approach.

Table 38: Benefits framework

Source: COWI (2019)

Two approaches to the monetisation of intangibles have been adopted for the purposes of this study:

•Approach 1: Application of a single WTP value to each case; and

•Approach 2: Use of DALYs (Disability adjusted life year) and their monetisation.

1.4.Benefits assessment

The health benefits of implementing new or revised OELs are then calculated in terms of the costs of ill heatlh avoided.

Benefits to workers & families

The direct and indirect resource costs are estimated using market-based information, for example, data on health care costs, and estimates of lost output (i.e. the value of a day’s work).

Added to these are the ‘human’ or intangible costs associated with a case, which are measured in terms of an individual’s willingness to pay for the reduction in the risk of mortality or morbidity (Approach 1) or monetised DALYs (Approach 2).

Under Approach 1, the most commonly used means of estimating individuals’ WTP for a reduction in the risk of an illness is through the use of experimental markets and survey techniques (e.g. contingent valuation or contingent ranking studies) to directly elicit individuals’ WTP for a reduction in the risk of death or morbidity.

The key measures are the value of a statistical life – a VSL – and the value of a case of morbidity (value of cancer morbidity VCM or value of morbidity VM). The VSL is essentially a measure of a change in the risk of fatality, where this is found by determining individuals’ willingness to pay for a small change in risk which is then summed across the population at risk.

1.4.1.Benefits to employers

The benefits of introducing OELVs have obvious benefits for workers, namely in terms of their health but also, indirectly, on their earnings. Employers will also reap benefits from their employees being less at risk of occupational illness. Such benefits include:

•higher labour productivity resulting from reductions in absenteeism and associated production losses;

•reduced administrative or legal costs relating to employees who are ill; and

•reduced sick leave payments.

1.4.2.Benefit to employers and workers – lost earnings and productivity losses

Individuals will incur costs associated with their inability to work in terms of a loss of earnings, including losses linked to days of for treatment as well as days off due to illness. Luengo-Fernandez et al. (2013) developed estimate of the magnitude of such costs by Member State in terms of an average cost per fatal or non-fatal cancer. These included what are referred to as “productivity losses” due to early death and then lost working days due to morbidity effects. Across all cancers, an average figure of €5,047 is given for productivity losses and €1,118 for the costs associated with lost working days due to morbidity effects (with these based on lost wages as the measure of lost output).

There are difficulties in including the type of estimates generated by Luengo-Fernandez et al. (2013) for lost working days within the analysis carried out here due to the potential for double counting. It is not clear whether the figures adopted in this study to reflect the intangible or human costs of cancer mortality and morbidity (i.e. €4 million and €400,000 respectively) also include an element related to the loss of income. If they do, then to include a separate cost item to reflect lost income would result in a double-counting of impacts.

The decision has therefore been taken not to include an additional element for lost income for mortality effects. However, due to uncertainty as to what may be captured by the value adopted here for cancer morbidity, lost income due to lost working days is considered within this analysis.

This inclusion may result in an overestimation of the economic costs associated with cancer morbidity. However, the exclusion of lost output for cancer mortalities may also lead to an underestimation if these are not fully accounted for within the value of a statistical life figure used here to reflect the intangible or human costs of a cancer.

1.4.3.Benefits to the public sector – cost of healthcare

Cancer cases: key data from Luengo-Fernandez, et al (2013) 103 have been used for the calculation of the avoided healthcare costs of illness. EUR 7000 is used in the model as the average cost for “all cancers”.

Leukocytopenia: there is no available data on the clinical process for leukocytopenia. It is assumed that performing a diagnosis will require visits to the general practitioner followed by consultation with specialists. Probably not all patients would need consultations with specialist including hospital. Consultation at the general practitioner, and subsequent treatment, might cost a few hundred euro, while the costs of specialist consultation could be several thousand euro. On average the cost for getting the diagnosis and the subsequent treatment is estimated at €1,000.

Pulmonary morbidity: The cost of treatment for pulmonary morbidity is based upon costs for chronic obstructive pulmonary disease (COPD). COPD is largely caused by smoking and is characterised by progressive, partially reversible airflow obstruction, systemic manifestations (skeletal muscle dysfunction, depression, and secondary polycythaemia), and increasing frequency and severity of exacerbations. The main symptoms, which are usually insidious in onset and progressive, are shortness of breath and inability to tolerate physical activity 104 . As such, the UK Department of Health Reference costs 2015/16 105 provide a comprehensive estimate of the costs associated with the treatment of the different conditions. From these, the healthcare costs for pulmonary morbidity are estimated at €1,000 per year.

Reproductive toxicity (miscarriages): The cost of treatment for reproductive toxicity (miscarriages) is taken from UK’s National Health Service (NHS) Reference costs 106 where the cost of a “Threatened or Spontaneous Miscarriage, with Interventions” is £1,908 and that of a “Threatened or Spontaneous Miscarriage, without Interventions” is £580. From these, the healthcare costs for reproductive toxicity (miscarriages) are estimated at €1,000 per year.

Nasal irritation: no data has been identified for the cost of treatment for nasal irritation, €200/year has been assumed since different severities are covered (from mild which require no treatment to severe).

2.Cost model

2.1.Compliance cost assessment

2.1.1.Introduction

This section describes the methodology for compliance cost assessment used across the three substances. For benzene, some costs estimates has been based on sector-specific studies.

The compliance costs for companies are basically estimated on the basis of:

•Risk Management Measures (RMMs) needed for reducing the air exposure levels from the actual levels to the target level.

•The costs of the RMMs (on-off and recurrent) for each company and/or workstation.

•The life span of the RMMs.

•Number of companies and/or workstations.

The costs are calculated in a worksheet model. For some sectors, where non-specific data are available, a likelihood model is applied. The likelihood model calculates the costs for a group of similar companies incurred in reducing air exposure to a target OEL based on an assumed sequence of RMM implementation which is determined by suitability, effectiveness, and cost. The model is run several times to construct a continuous cost curve.

For other sectors, where the RMMs to be applied are more well-described, a more simple model is applied, but the same unit costs and life span parameters are used as in the more complex model.

2.1.2.Key model inputs and assumptions

I.Discount rate

The static discount rate is 4%: this is taken over the 60 year period. A dynamic discount rate is taken in the sensitivity analysis. The dynamic rates start at 4% for the first 20 years; it then decreases to 3% for the remaining 40 years.

II.Affected workers and worstations

Each company size was in the likelihood model assumed to have an average number of workers affected and associated workstations requiring adjustment, shown in the table below.

Table 39: Number of workers and workstations

Source: COWI (2019)

Three different costs, all present values for 60 years, are calculated: TOTAL, (one-off + recurrent) one-off costs, and recurrent.

III.RMMs considered

The likelihood model considers following types of RMMs:

•Local Exhaust Ventilation (LEV), extraction at source;

•Worker enclosures (WE), i.e. physical separation of workers in an enclosure or control room;

•Respiratory Protective Equipment (RPE);

•General Dilution Ventilation (GDV);

•Organisational & hygiene measures (OH).

For reduction of fugitive emission (e.g. in coke plants or in the petroleum sector) other types of measures such as improved valves, pumps, etc. are considered.

IV.RMMs effectiveness

Every RMM has a different level of effectiveness in reducing the workers exposure to the substances. The percentage reduction in exposure due to each type of RMM used in the analysis is shown below.

Table 40: Percentage reduction in exposure achieved with RMM

Source: COWI (2019)

V.RMMs costs

Table 41: Cost of various RMMs in €

Source: COWI (2019)

VI.Sectoral characteristics that determine suitability of RMMs

For sectors or subsectors, where specific information on RMMs to be applied is not known, the likelihood model determines the suitability of RMMs on the basis of sectoral characteristics.

The amount of exposure is split into work where the worker is exposed to the substance for less than an hour a day and for more than an hour a day. This also equates to exposure for more or less than 2.5 days/month. Many production activities only occasionally use the substances. Where the exposure is less than an hour a day, it is acceptable, and often more cost effective, to use personal protective equipment (PPE) such as masks with filters or breathing apparatus.

The form of substance to which workers are exposed varies considerably from dust and fibres to vapour, fumes, gas, mist and aerosol. Again, the form of substance has a direct bearing on the types of RMM that are suitable. For example, general dilution ventilation is not advised for removing dust as it tends to stir it up and spread it around. For this analysis, the substance form is split into two types: dust which also includes fibres; and gas which includes all the other types.

The extent of the spread is the final characteristic that affects the choice of RMM and this is split into three types: local, diffuse and peripheral. Local means the dust or gas is created around a specific machine and often means that highly targeted ventilation can effectively remove the chemical. Other processes spread the substance over a wider area and this is known as diffuse. In this case, dilution ventilation, workers enclosures or full enclosures are more suitable, the choice depending upon the decrease in exposure required. Peripheral means that the substance spreads more widely and cause exposure to workers beyond the area where the substance is being worked. This means that administrators, managers and sales staff may be exposed.

In the table below, the types of RMM that are suitable or not for each amount of exposure, form of substance and extent of spread are shown. These values were built into the cost model.

Table 42: Suitability of various RMMs to amount of exposure, form of the substance and extent of spread