EUROPEAN COMMISSION

Brussels, 28.7.2023

SWD(2023) 269 final

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT REPORT

Accompanying the document

PROPOSAL FOR A REGULATION OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL

on the safety of toys and repealing Directive 2009/48/EC

{COM(2023) 462 final}
{SEC(2023) 297 final}
{SWD(2023) 268 final}
{SWD(2023) 270 final}

Table of contents

1.Introduction: Political and legal context

1.1.Regulatory context: the Toy Safety Directive

1.2.Link with other initiatives

1.3.Economic context: The market for toys in the EU

2.Problem definition

2.1.What are the problems?

2.1.1.    Problem 1- Insufficient protection of children from harmful chemicals    

2.1.2.    Problem 2- A high number of toys on the Union market do not comply with the Toy Safety Directive and are unsafe    

2.1.3.    New risks in toys, in particular from digital technologies    

2.2.What are the problem drivers?

2.2.1.    Problem 1- Insufficient protection of children from harmful chemicals    

2.2.2.    Problem 2- A high number of toys on the Union market do not comply with the Toy Safety Directive and are unsafe    

2.3.How likely is the problem to persist?

3.Why should the EU act?

3.1.Legal basis

3.2.Subsidiarity: Necessity of EU action

3.3.Subsidiarity: Added value of EU action

4.Objectives: What is to be achieved?

4.1.General objectives

4.2.Specific objectives

5.What are the available policy options?

5.1.What is the baseline from which options are assessed?

5.2.Description of the policy options

5.2.1.    Policy options to strengthen the protection of children when playing with toys    

5.2.2.    Policy options to reduce the number of non-compliant and unsafe toys    

5.3.Options discarded at an early stage

6.What are the impacts of the policy options?

6.1.Policy option 1a – minimum changes to the Directive

6.2.Policy option 1b – Improved protection: generic bans of the most harmful substances with derogations

6.3.Policy option 1c – Maximum protection: generic bans of the most harmful substances without derogations

6.4.Policy option 2a: extending third-party conformity assessment

6.5.Policy option 2b: facilitation of market surveillance through digitalisation

6.6.Policy option 2c: Extending third party conformity assessment and facilitation of market surveillance through digitalisation

7.How do the options compare?

7.1.Effectiveness of the proposed policy options

7.2.Impacts of policy options

7.3.Summary of policy options assessment

8.Preferred option

8.1.Legal instrument: converting the Directive into a Regulation

8.2.REFIT (simplification and improved efficiency)

8.3.Application of the ‘one in, one out’ approach

9.How will actual impacts be monitored and evaluated?

ANNEXES

Annex 1: Procedural information

Annex 2: Stakeholder consultation (Synopsis report)

Annex 3: Who is affected and how?

Annex 4: Analytical methods

Annex 5. SME TEST Summary

Annex 6: Legal Context: Main Provisions of the Toy Safety Directive and Market Surveillance

Annex 7: Toys Market

Annex 8: Conclusions of the Evaluation of the TSD

Annex 9: EU Chemicals Legislation

Annex 10: The most harmful chemicals currently not addressed by the TSD

Annex 11: Non-compliant toys on the Union Market

Annex 12: Digital Aspects and Internet-Connected Toys

Annex 13: Digital Product Passport under the Proposal for an Ecodesign for Sustainable Products Regulation

Annex 14: Impacts of the Options

Annex 15: Bibliography

Glossary

1.Introduction: Political and legal context

Play is an essential part of growing up: children discover the world and develop their capabilities through play. However, toys have to be safe for children. They are regulated by Directive 2009/48/EC on the safety of toys (the Toy Safety Directive, TSD) 1 . This Directive lays down the safety requirements that toys must meet in order to be placed on the market in the EU, whether they are manufactured in the EU or in third countries. At the same time, the Directive aims at ensuring the free movement of toys within the Single Market.

The Commission Evaluation of the Toy Safety Directive 2 (the Evaluation) identified a number of deficiencies that have emerged during the practical application of the TSD since its adoption in 2009. In particular, the Evaluation identified certain shortcomings in ensuring a high level of protection of children from possible risks in toys, in particular from risks posed by harmful chemicals. The Evaluation also concluded that the enforcement of the Directive lacked effectiveness, in particular in the context of online sales, and there remain a high number of unsafe toys on the Union market 3 .

The Chemicals Strategy for Sustainability 4 (the CSS) called for extending the generic approach towards harmful chemicals (based on generic preventive bans) to ensure that consumers, vulnerable groups and the natural environment are more consistently protected. In particular, the CSS aimed at strengthening the TSD with regard to the protection from the risks of the most harmful chemicals and with regard to possible combination effects of chemicals. Although the Directive already prohibits generally substances that are carcinogenic, mutagenic or toxic for reproduction (CMRs) in toys, it does not refer to other substances of particular concern such as endocrine disruptors, substances affecting the immune, neurological or respiratory systems or substances toxic to a specific organ.

Finally, on 16 February 2022 the European Parliament adopted almost unanimously an own initiative report on the implementation of the Toy Safety Directive 5 . In its report, the European Parliament calls on the Commission to revise the Directive to strengthen the protection of children from chemical risks, ensure that risks posed by connected toys are addressed by EU law and improve enforcement of the Directive in particular in relation to online sales.

Given the objectives of this initiative of better protecting children from harmful chemicals, as well as reducing the number of unsafe toys on the Union market, it is expected to contribute most significantly to the United Nations Sustainable Development Goal (SDG) #3 for good health and well-being. In addition, it will contribute to SDG#9: industry, innovation and infrastructure, SDG #12: responsible production and consumption and SDG#6: clean water and sanitation.

1.1.Regulatory context: the Toy Safety Directive

The TSD is an EU product harmonisation measure which has the twofold objective (1) to maintain a high level of safety for children and protection against possible health threats from toys, while (2) allowing the free movement of toys in the internal market.

The scope of the Toy Safety Directive covers all ‘products designed or intended, whether or not exclusively, for use in play by children under 14 years of age’. 6 The Directive explicitly excludes certain products from its scope 7 . Moreover, Annex I to the TSD enumerates examples of products that are not considered to be toys but could be confused with toys 8 . In this Annex, the Directive excludes from its scope “electronic equipment, such as personal computers and game consoles, used to access interactive software and their associated peripherals…” and “interactive software, intended for leisure and entertainment, such as computer games, and their storage media, such as CDs”. Therefore, whilst a physical toy which has play value and incorporates software would be considered a toy within the scope of the Directive (for example, a talking doll or animal), a videogame or other similar software would not.

The TSD lays down the safety criteria (‘essential safety requirements’) that toys must meet to be marketed in the EU. The essential safety requirements are designed to ensure a high level of safety; they cover identified hazards related to the characteristics of the toy or to its performance. 9 Essential requirements are supported by harmonised standards, which set out the technical specifications to comply with the essential requirements and once referenced in the Official Journal of the EU, provide presumption of conformity with the essential requirements they aim to cover.

The essential safety requirements in the Toy Safety Directive cover:

-general risks: the health and safety of children, as well as other people such as parents or supervisors;

-particular risks: physical and mechanical, flammability, chemical, electrical, hygiene and radioactivity risks. In particular, as regards chemical risks, there is (i) a general prohibition of CMR substances subject to derogations, (ii) specific limit values for specific CMR substances (nitrosamines and nitrosatable substances), as well as for 19 ‘elements’ in different toy materials, (iii) prohibitions of certain allergenic fragrances and labelling requirements for others, and (iv) specific limit values in toys for children under 36 months or to be put in the mouth.

In order to keep pace with latest technical and scientific developments, the Commission can amend certain parts of the TSD via the Regulatory Procedure with Scrutiny (RPS). 10 The Commission may adapt Annex I that lists examples of products that are not toys (but may be confused with them), the list of prohibited allergenic fragrances and the list of allergenic fragrances to be labelled in Annex II. It may adapt the limit values for heavy metals and other hazardous metals in Annex II, and the warnings for toys in Annex V. In addition, the Commission may establish maximum limit values for any chemical in toys intended for children under 36 months of age and in all toys intended to be placed in the mouth, and it may also amend those limits (Appendix C to Annex II). Since its adoption the TSD has thus been amended 17 times 11 . Finally, the Commission may allow the use of chemicals that are carcinogenic, mutagenic or toxic to reproduction (CMRs), albeit only following a strict scientific-technical assessment including an independent Scientific Committee. 

The TSD requires that conformity assessment is carried out to verify whether a toy complies with the applicable essential requirements. It is to be carried out by the manufacturer or with the intervention of a third party – a ‘Notified Body’ test laboratory that has been previously designated at EU level for the purposes of such assessments under the TSD. The TSD requires the intervention of a Notified Body only in certain cases, in particular when harmonised standards do not cover all relevant safety requirements or when the toy manufacturer has not applied such harmonised standards 12 . It is the manufacturer, whether established in the EU or outside the EU, who is responsible for the compliance of the toy with the applicable requirements. When a toy is in conformity with the applicable requirements, the manufacturer must affix the CE marking to it.

Market surveillance to ensure compliance with the Toy Safety Directive is currently regulated by Regulation 2019/1020 13 . This Regulation replaced Regulation (EC) No 765/2008 14 and became fully applicable on 16 July 2021. This Regulation requires that in order to place toys on the Union market, there must be an economic operator established in the EU and responsible for a number of tasks set out in that Regulation to facilitate market surveillance.

More details on the current legal provisions of the TSD can be found in Annex 6.

1.2.Link with other initiatives

Revision of the CLP Regulation 15 - Regulation (EC) No 1272/2008 on hazard classification, labelling and packaging of chemicals (the CLP Regulation) is the core piece of Union legislation for the hazard assessment of chemicals (human health, environment and physical hazards). It sets out the hazard classification of chemicals and how to communicate those hazards to consumers and workers. The CLP Regulation follows the United Nations’ Globally Harmonized System (GHS) of classification and labelling of chemicals setting up criteria for classification and communication of physicochemical, health, and environmental hazards.  The CLP Regulation is currently under revision and the introduction of new hazard classes (such as endocrine disruptors) is considered. Hazard assessment is the starting point for risk-assessment and -management measures, which are in turn provided for in downstream legislation, such as the Toy Safety Directive. This means the hazards that chemicals present are established under the CLP Regulation: they may be carcinogenic, lead to skin allergies or to irritation in the eyes for example, in certain of their forms. The decision of how to address those hazardous substances in certain products are taken then in other legislation that apply to the product. For example, the Toy Safety Directive already prohibits the use of substances which are classified under the CLP Regulation as CMRs. Therefore, the future hazard classes to be introduced in the CLP Regulation would be relevant for defining the regulatory consequences in certain products, such as in the context of the current revision of the Toy Safety Directive. The measures proposed under options 1b and 1c in this Impact Assessment are based on banning chemical substances classified under certain hazard classes of the CLP Regulation. To fully implement the generic bans of endocrine disrupting substances in a revision of the Toy Safety Directive (proposed in options 1b and 1c), it is necessary that the CLP Regulation be amended to include the hazard class for endocrine disrupting chemicals. This has been done in the recent Commission Delegated Regulation (EU) 2023/707 16 . Any of the other hazard classes considered for generic bans under options 1b and 1c already exist under the CLP Regulation and therefore do not require any change to this Regulation.

Figure 1: Mapping of the pieces of legislation according to the different steps of hazard and risk assessments 17

Revision of the REACH Regulation 18  - The REACH Regulation on Registration, Evaluation, Authorisation and Restriction of Chemicals, together with the CLP Regulation, are the key Union legislation for the assessment and management of chemicals. In accordance with the CSS, a targeted revision of REACH will explore extending the generic risk approach to restrictions to most harmful chemicals while allowing their use only when essential. In particular, criteria for essential uses will be defined to ensure that the most harmful chemicals are only allowed if their use is necessary for health, safety or is critical for the functioning of society and if there are no alternatives that are acceptable from the standpoint of environment and health. Toys are within the scope of REACH as consumer articles where REACH addresses environmental or human health concerns. However, the Toy Safety Directive provides for a more comprehensive and targeted approach to toys as regards aspects relating to human health, in view of the vulnerable population group toys are addressed to. 

Figure 2: Hazard identification and risk management in EU legislation

The proposed measures under this Impact Assessment do not rely on any changes to be introduced in the REACH Regulation. The REACH revision could introduce generic bans for the most harmful chemicals themselves and for certain uses, in particular in consumer products. This may have an indirect impact on the functioning of the toys industry and the options presented. This is because alternative chemicals will be more likely to be available on the market, including for toy manufacturers, which would limit the negative impacts of options 1b and 1c proposed in the toy industry. However, these changes are not necessary for the revision of the Toy Safety Directive. Thus, the quantitative analysis of impacts of the different policy options in this impact assessment does not factor in the possible effects from the REACH revision.

Digital Services Act (DSA) 19 : The DSA includes a new set of horizontal rules to regulate the responsibility of online intermediaries, including online marketplaces. It will establish new obligations for online intermediaries inter alia in relation with how they handle all types of illegal content hosted on their websites including unsafe products. The DSA will reinforce actions to be taken on non-compliant toys sold online, such as mechanisms to counter illegal content online, new rules to trace sellers on online marketplaces and an obligation by online marketplaces to randomly check against existing databases whether products or services on their sites are compliant.

General Product Safety Regulation 20 : The recently adopted Regulation (EU) 2023/933 on General Product Safety replaces the General Product Safety Directive 21 . One of the objectives of the proposal is to offer better protection of consumers when shopping online, including on online marketplaces, and from dangerous products coming from the EU and outside. The General Product Safety Regulation, as the current Directive, will continue to address only products, risks and aspects which may not be specifically covered by the Toy Safety Directive. It is expected to include specific provisions on recalls, accident reporting and online marketplaces which would apply to all consumer products, including toys.

Artificial Intelligence (AI) horizontal framework 22 : The new legislative proposal for the AI horizontal framework (AI Act) will lay down harmonised rules for the placing on the market, the putting into service and the use of artificial intelligence systems (‘AI systems’) in the Union. It lays down specific requirements for high-risk AI systems, and it is expected to cover, as high-risk, AI systems intended to be used as safety components of toys covered by the TSD. Therefore, any risks posed specifically by the incorporation of AI in toys would be addressed by that Regulation. Nevertheless, the AI Act refers to the EU product legislation applicable to the product (in this case the Toy Safety Directive) to determine the conformity assessment procedure which needs to be followed, also with regards to the verification of compliance with the essential requirements concerning AI systems that are to be imposed by the AI act.

Delegated acts under the Radio Equipment Directive (RED) 23 : the RED establishes the possibility for the Commission to adopt delegated acts in relation to several aspects, including requiring protection of networks, personal data and privacy and protection from fraud for specific categories of radio devices. As all internet-connected wireless devices fall under the RED, the Commission adopted on 29 October 2021 a delegated act under that Directive strengthening the level of cybersecurity, personal data protection and privacy of certain categories of radio equipment. This delegated act applies to radio toys (protection of privacy) and to internet-connected radio toys (protection of networks and from fraud) and will become applicable as of 1 August 2024 24 . It will require that radio toys incorporate safeguards to ensure that the aforementioned assets are protected.

Circular Economy: The proposal for a Regulation on Ecodesign for Sustainable Products 25 will set a framework and a process through which the Commission will be empowered to progressively set out sustainability requirements for each product or group of products. Safety concerns will not be a criterion to include products under the ESPR; this inclusion will be based on sustainability concerns. The ESPR will not regulate the safety of products. Toys may be covered in the long-term by one of the delegated acts setting sustainability requirements, however they are not included in the list of high priority sectors identified in the Circular Economy Action Plan (CEAP) adopted by the Commission in March 2020. The ESPR will set out the principle of the Digital Product Passport (DPP) and will develop its technical design and operation. The technical design and operation of the DPP are already being developed, following the general principles of the proposed DPP. The ESPR also foresees that the reference of the DPP is included in a central registry managed by the Commission 26 , that this reference is declared at customs, and that this central registry is interconnected with the customs IT environment 27 . This will allow an automatic verification of the existence of the DPP for products presented at customs prior to their release for free circulation. It is expected that the DPP will become the vehicle to provide digital information on products in the future.

Customs Union Reform 28 : The Commission has recently proposed a revision of the Union Customs Code in order to strengthen the legal framework for customs and address recent challenges such as e-commerce and to protect the single market from non-compliant goods imported from non-EU countries, among others. This initiative is expected to complement the current revision of the TSD by strengthening risk-based controls at the border and improving the supervision of the supply chain. The exact requirements for toys entering the Union market and being released for free circulation need to be set out in the toy safety rules. 

1.3.Economic context: The market for toys in the EU

The European toy market size was estimated to be worth around EUR 20 billion in 2020 29 in terms of overall value, making it the third-biggest global market after North America and Asia. Online sales are increasing and reach over 1 in every 4 toys in some countries 30 . Market research estimates place the size of the global toy industry in the range of EUR 86-109 billion in sales in 2019, growing from around EUR 81.7 billion in 2016. 31 , 32 The European toys manufacturing market achieved turnover of at least EUR 8.3 billion in 2019, increasing from EUR 7 billion in 2016. 33 However, as for much of the global economy, 2020 also saw a downturn in turnover within the EU toys industry, to around EUR 6.6 billion (based on provisional Eurostat SBS data). In 2017, the EU was reportedly the biggest global importer of toys with EUR 7.2 billion worth of toys imported. The value of EU toy imports had grown by an estimated 70% during the preceding decade. Most toys imported to the EU come from Asia, with China the biggest supplier and the ASEAN countries, such as Thailand and Vietnam, increasing the volume of exported toy products. 34

Almost 50,000 people are working in the EU toy sector. The structure of the EU toy industry is complex and very heterogeneous, ranging from large world-wide operating companies to very small producers of certain specific kinds of toys. The number of enterprises has been progressively increasing from circa 5,332 in 2016, to an estimated 6,313 in 2020. Around 99% of European toy companies are SMEs, employing around two thirds of employees in the sector. Toy production is highly concentrated, with 96.2% of production value generated across ten countries and 88% of turnover concentrated in seven EU Member States 35 .

Online sales have grown significantly, accelerated by the global pandemic 36 . Consumers are increasingly buying toys online instead of in brick and mortar stores. For instance, according to a study by a market research firm, all national markets in the EU recorded growth in online toy sales in 2020-2021. At the global level, online sales of toys were estimated to have increased markedly, by an estimated 20%, in 2020-2021. 37 See annex 7 for a more detailed overview of the toys market.

2.Problem definition

2.1.What are the problems?

In the context of toy safety, two main separate problems can be currently identified. As the Evaluation concluded, the first problem is that the Directive does not ensure adequate protection of children when playing with toys from risks posed by hazardous chemicals: the Directive does not sufficiently address the risks of certain harmful chemicals such as endocrine disruptors (EDs) and CMR substances. The possibilities to amend the TSD in relation to limit values for toys intended for children over 36 months, or for certain problematic substances such as nitrosamines, are limited. In addition, there are a high number of non-compliant and unsafe toys on the Union market, which also put children at risk and create an uneven playing field for compliant companies. See Annex 8 for the main findings of the Evaluation.

2.1.1.Problem 1- Insufficient protection of children from harmful chemicals 

The objective of the TSD is to ensure that toys, including the chemicals they contain, do not jeopardise the safety of children. However, as the Evaluation concluded, the protection from most harmful chemicals in toys in the current Directive is not complete. Almost 60% of respondents (116 out of 196) to the public consultation believed that the TSD should set stricter requirements for chemicals: however, 47% representatives of business associations and companies (strongly) disagreed (42 out of 89), while all the environmental and consumer organisations agree to some degree with the statement (57 out of 61 - over 90%). Similarly, 90 % of public authorities (28 out of 31) agreed or strongly agreed. Concerning other risks from toys (for example, preventing suffocation of small children by requiring that toys for younger children do not contain small parts), the Evaluation concluded that the protection in the TSD is adequate and no evidence has appeared since its publication which would affect that conclusion 38 .

Chemicals with hazardous properties can cause harm to human health and the environment. Chemical substances may be classified under the hazard classes of the CLP Regulation on the basis of the hazardous properties they present to human health, the environment or their physical hazards. The classification under the CLP Regulation informs on the hazardous properties of a chemical substance, but does not provide for the management of the risk. Other legislation, such as the TSD, is to define whether and how the risks posed by certain chemical substances must be addressed in specific products. While not all hazardous chemicals raise the same concerns, certain chemicals are particularly harmful as they can cause cancers, affect the immune, respiratory, endocrine, reproductive and cardiovascular systems, weaken human resilience and capacity to respond to vaccines, or increase vulnerability to diseases. Exposure to these harmful chemicals is therefore a threat to human health. 39 Children, pregnant women, workers and the elderly are particularly vulnerable to risks arising from chemical exposure, and have higher probabilities of adverse health symptoms or diseases throughout their lives 40 .

The Toy Safety Directive does not sufficiently protect children from risks posed by the most harmful chemicals. Children are exposed to harmful chemicals from different sources (indoor air, food, water, consumer products in general, for example, as well as toys) and it is difficult to establish a clear link between their exposure to these substances through one of these sources, such as toys, and immediate health consequences. The particular vulnerability of children to the harmful effects of hazardous chemicals is well documented 41 . There are physiological differences between children (especially those under 36 months of age) and adults, affecting the accuracy of certain exposure assessment methodologies (i.e. immature metabolic and immune system, proliferative tissues)  42 43 44 45 . Furthermore, the manner in which children interact with toys, in particular in their earlier years, makes them particularly exposed to any harmful chemicals that may be present in toys. Children exhibit specific habits and practices that may result in exposure scenarios not considered for other population groups. Under the age of 3, mouthing behaviour plays an important role regarding contact of children with toy materials 46 . As an example, the current migration limits for 19 substances or heavy metals in the Directive are based on the assumption that a child would ingest per day 100 mg of dry, brittle, powder-like or pliable toy material, 400 mg of liquid or sticky toy material, and 8 mg of scraped-off toy material. Studies show that children are currently exposed to known and unknown endocrine disrupting chemicals in toys 47 , or to other chemicals that pose risks to children 48 . See annex 10. D. for an estimate of the potential harmful chemicals that could be currently present in toys. In particular, it has been mentioned by industry that toys like balloons, paints, modelling clays, glue or slime could be affected by the policy options proposed in this impact assessment. 

CMR (carcinogenic, mutagenic or toxic for reproduction) substances are particularly harmful, and their effects can only be seen in the long term and are often irreversible. The Evaluation concluded that the current requirements in the Directive for CMR substances are not effective and they do not sufficiently protect children from these substances in toys. As will be explained below when describing the drivers of this problem (see section 2.2.1 below), the TSD contains generic bans for CMR substances subject to derogations, but these derogations still allow for too high a presence of CMRs in toys. Furthermore, the limit values currently set in the TSD for nitrosamines and nitrosatable substances, which are strong carcinogens, are deemed to be too high, and these limits cannot be amended by the Commission with its empowerments.

In addition, there are ongoing and emerging health concerns for other most harmful chemicals that are not addressed by the Directive. This is particularly the case for chemicals that affect the endocrine system, the immune, neurological or respiratory systems or chemicals that are toxic to a specific organ. In respect of these chemicals, the Fitness Check on Chemicals legislation excluding REACH 49 pointed to shortcomings in current legislation in meeting the objectives of protecting human health in particular as regards exposures to: neurotoxic substances; chemicals linked to cardiovascular and respiratory (CVR) disease; and endocrine disrupting chemicals. The possible risks associated with the combination of chemicals are also not addressed in the Directive.  There are no specific legal requirements to ensure that risks from simultaneous exposure to multiple chemicals are effectively and systematically taken into account in toys. While it may not be currently realistic or economically feasible to specifically assess and regulate an almost infinite number of possible combinations of chemicals, scientific consensus is emerging that the effect of chemical mixtures needs to be taken into account and integrated more generally into chemical risk assessments 50 . 

Endocrine-disrupting (ED) chemicals require specific attention; these chemical substances can alter the functioning of the endocrine (hormonal) system and negatively affect the health of humans or animals. Endocrine disrupting properties have been the focus of increasing scientific research, and the accumulated knowledge identifies endocrine disruptors as a concern to public health 51 52 . Given the essential role of the endocrine system during development, exposure to endocrine disruptors during vulnerable periods can induce long-lasting changes, with adverse effects in the short and long terms; some of these effects are expected at very low-doses 53 . Early exposure during critical periods of development can affect health at a later stage of life 54 . Many chronic health disorders have been clearly linked to endocrine disruptors 55 . These disorders include obesity and metabolic disorders, male and female reproductive disorders, reproductive cancers, thyroid disorders, neurodevelopmental disease and IQ loss  56 . It is not possible to quantify fully the impact on human health of chemicals with the most critical hazards. However, for illustrative purposes, a causal link of 70% to 100% between exposure to EDs and IQ loss and intellectual deficiencies was identified 57 . EDs are also suspected to cause male fertility troubles with a probability of causation between 40-69%. Researchers estimated that exposure to EDs leads to substantial health-related societal costs in the EU between 46 and 288 billion Euro per year 58 . The TSD does not currently have provisions that allow to address EDs in toys in general and limit children exposure wherever this is necessary. 

Chemicals that affect the immune, neurological or respiratory systems or chemicals that are toxic to a specific organ are also amongst the most harmful chemicals. It is equally well documented that neurotoxic chemicals are particularly harmful to the developing brain of children, which is inherently more vulnerable to toxic injury than the adult brain 59 . Adverse effects of exposure to certain chemicals on IQ scores are well established and other disorders where the role of chemical exposure is suspected, include ADHD and autism. Recent research has focussed on links to depression, bipolar disorder, OCD, psychoses, dementia, Parkinson and Alzheimer’s. 60 For example, it can be estimated that, in the EU, some 30,000 disability adjusted life years (DALYs) related to neurotoxic disease may be a result of chemicals exposure, with some 250,000 DALYs a result of chemicals exposure combined with underlying genetic predisposition. Developmental disabilities due to chemicals alone could cost the EU between €1.7 up to €4 billion per year directly and up to €14 – €33 billion per year taking into account chemical exposure and combined with genetic factors, depending on the valuation used 61 . Environmentally attributable childhood asthma (i.e. where the onset of asthma has been triggered by exposure to pollution and hazardous chemicals) was estimated in 2013 to cost the EU approximately €1.6 billion per year (inclusive of direct medical costs and indirect economic losses due to productive time lost by care for children with asthma). 62  The TSD has specific limit values for some of the chemicals in these categories, but it does not address them in a general manner and new limit values for other substances in this category cannot be generally introduced either. 

Sufficient protection needs to be provided to children from potential exposure to those risks for human health when playing with toys. Risks coming from harmful substances are often only visible long term and parents are generally unaware of these risks in toys or of their potential severity; as opposed to mechanical risks, they cannot be avoided by parents being more or less vigilant of their children playing with toys. Increasing the requirements in the TSD for harmful chemicals is not expected to lead to less surveillance of children by their parents when playing with toys (i.e. the Peltzman effect is not expected to materialise).

REACH already regulates the environmental aspects of toys and the chemical substances in toys that pose a risk to the environment. Furthermore, the ESPR will also apply to toys. Therefore, the extent to which harmful substances to the environment may be limited in consumer products (such as substances which are persistent, bioaccumulative and toxic – PBTs and vPvBs), including in toys, will continue to be addressed under horizontal legislation such as REACH and the ESPR and not in the revision of the TSD.

2.1.2.Problem 2- A high number of toys on the Union market do not comply with the Toy Safety Directive and are unsafe

Unsafe toys put children at risk and may lead to accidents that can even be fatal. Not all toys on the market can be subject to checks and, therefore, the exact share of non-compliant toys in the Union market cannot be quantified with precision. However, there are sufficient separate indicators that confirm that the number of non-compliant toys on the Union market is very high. Whenever market surveillance actions or inspections take place, the percentage of non-compliant and unsafe toys found is consistently high.

Figure 3: RAPEX notifications per year and toys notifications 

First of all, as regards notifications in the Safety Gate for alerts of dangerous products (RAPEX notifications), during the past years, toys were consistently one of the top categories of products with the highest number of notifications for dangerous products per year.

Overall, during the period 2016- May 2022, a total of 3579 dangerous toys were subject to an alert through the RAPEX system for exchange of information on dangerous products 63 . 64 Around 40% of these notifications concerned chemical risks, which means that exposure to dangerous chemical substances through unsafe toys is a matter of concern, while another 40% concerned choking risks for younger children that can lead to fatal injuries.

Market surveillance authorities are also required to enter into the information and communication system for market surveillance (ICSMS) information in relation to products made available on the Union market for which an in-depth check of compliance has been carried out, in addition to their obligation to alert of dangerous products under RAPEX. This obligation has become more explicit with the new Regulation 2019/1020 on market surveillance 65 , which is fully applicable since July 2021. Data on the period 2016- May 2022 reveals that 43% of the toys subject to in depth inspections were found to be non-compliant and more than half of these relate to substantive safety risks. Market surveillance authorities rely on risk management procedures to try and target their activities on toys with higher probabilities to be non-compliant. In-depth checks are carried out on toys which present such higher risks of non-compliance. While these data may show the effectiveness of such procedures, it still evidences the high number of non-compliant and unsafe toys that are present on the Union market.

Figure 5: non-compliant toys and total number of in-depth toy investigations per year in ICSMS

Other recent but more targeted market surveillance joint actions focused on toys also confirm the high rates of non-compliance, see annex 11 for detailed results on coordinated activities on the safety of products (CASP) 66 . Data from the Evaluation based on the reports on market surveillance submitted by Member States 67 also shows a significant rate of non-compliant toys found on the Union market; almost every third toy inspected in the reported period was non-compliant.

The high number of non-compliant toys means that children in the EU can be exposed to risks when playing with these toys. Unsafe toys can cause injuries and even fatal accidents, for example risks of choking or from very serious injuries such as from magnets, or button cells batteries that children can swallow. Children may also be exposed to dangerous chemicals, for example from phthalates in soft plastics or too high quantities of boron in slime. Unsafe toys may undermine consumer trust in the safety of toys available in the EU. They are also a risk to the environment as the harmful substances they contain are ultimately released into waste.

Non-compliant toys also lead to an uneven playing field for compliant companies. Complying with the stringent requirements is costly for companies, reputable manufacturers face the unfair competition of non-compliant toys from rogue traders who are able to sell their toys at lower prices. The value of the non-compliant toys market has been estimated for 2019 to span from EUR 248 million to EUR 1.65 billion 68 .

At the stakeholder workshop organised on 26 April 2020, all the intervening stakeholders (consumer organisations, industry and market surveillance authorities) expressed strong concerns about the high number of non-compliant and unsafe toys, in particular made available through online sales, and the urgent need to take action at EU level to address this.

2.1.3.New risks in toys, in particular from digital technologies

The Evaluation also concluded on the TSD not being able to address new risks in toys linked to the use of digital technologies. The scope of the Directive is focused on health and safety and does not cover other issues such as privacy or information security. As a result, there were cases in the past where risks posed by cybersecurity or privacy concerns on toys could not be addressed. The application of the RED delegated act mid-2024 will address the cybersecurity and personal data privacy issues identified in the Evaluation. This delegated act requires radio toys to guarantee protection of privacy and data of the users, through cybersecurity features. Cybersecurity should be strengthened by the Cyber Resilience Act as well, once adopted 69 . Additionally, the incorporation of AI in toys may also lead to new risks. The Commission proposal on a regulatory framework for artificial intelligence, once adopted, will ensure that any AI systems incorporated into toys comply with the rules that safeguard the functioning of markets and the public sector, and people’s safety and fundamental rights. Therefore, these risks have been addressed in legislation already adopted or under inter-institutional discussions and it does not appear that any significant legal gap remains that requires intervention through the revision of the TSD for the purposes of addressing cybersecurity, personal data privacy or AI.

2.2.What are the problem drivers?

2.2.1.Problem 1- Insufficient protection of children from harmful chemicals 

a) The Directive allows for too high presence of CMR substances in toys

The Directive includes a generic ban on the presence of CMRs in toys subject to a number of derogations. The Evalution concluded that these derogations do not ensure sufficient protection of children from these harmful subtances. In particular, the Toy Safety Directive currently allows three possibilities to derogate from the generic ban to CMRs in toys: i) the TSD tolerates the presence of CMRs in toys or their components up to the ‘relevant concentrations’ of the CLP Regulation for the classification of mixtures containing these subtances. ‘Relevant’ are either the specific concentration limits assigned to specific substances in Annex VI, table 3.1 of the CLP Regulation or, if no specific concentration limits are indicated in that table, the generic concentration limits in Annex I of the CLP Regulation 70 . In addition, the TSD allows for the presence of CMRs ii) if they are inaccesible to children, including via inhalation and iii) if the Commission takes a decision to allow their presence if it is considered safe following a rigorous scientific evaluation and if they are not prohibited in consumer articles under REACH. For CMR substances of categories 1A and 1B which are of most concern, no suitable alternatives must exist for a Commission to take a decision granting the derogation. (For CMRs category 2, no analysis of alternatives is necessary.)

The Evaluation found the first derogation inadequate 71 . These ‘relevant concentrations’ have been set for the purpose of classification and labelling of mixtures containing hazardous substances, with the primary aim to ensure that the hazards of such mixtures are properly identified and communicated. They do not take account of possible exposures and do not entail an assessment of risk related to the uses of a substance in an article such as a toy. Whenever risk-based limit values for certain CMRs were introduced in the TSD through the empowerments to the Commission, these have been 15 to 1,000 times lower than the ‘relevant concentrations’ in the CLP Regulation. However, in accordance with the empowerements to the Commission for amending the Directive, these limit values could only be introduced for toys intended for children under 36 months, or to be put in the mouth (Apendix C). The Scientific Committee for Health, Environmental and Emerging Risks 72 (SCHEER) also identified a number of problems with this approach, due to the fact that classification limits set for mixtures are applied to articles (as the toys should be considered)  73 . This inadequacy of the CMR derogation based on the concentration limits for classification of mixtures from the CLP Regulation was referred to by public authorities (70% of public authorities responding – 22 out of 31 -disagreed or strongly disagreed with having this derogation) and by consumer organisations (75%- 46 out of 61) in the public consultation 74 .

In addition, the Evaluation found that the limit values in the Directive for nitrosamines and nitrosatable substances were too high 75 . Certain nitrosamines may be genotoxic and very strong carcinogens (CMRs). Nitrosatable substances can be converted into nitrosamines in the human body. The TSD sets migration limits for nitrosamines and nitrosatable substances in toys intended for use by children under 36 months or in other toys intended to be placed in the mouth: 0.05 mg/kg for nitrosamines, 1 mg/kg for nitrosatable substances. 76 Germany however insisted on its lower national limits of 0.01 mg/kg for nitrosamines and of 0.1 mg/kg for nitrosatable substances in toys made of natural or synthetic rubber designed for children under 36 months and intended or likely to be placed in the mouth, which the Commission allowed 77 . Therefore, currently Germany has lower limits for these substances that do not correspond to the ones in the TSD. In their letter of April 2019 to the Commission, 78 11 Member States considered that there was an urgent need to lower the limits for nitrosamines and nitrosatable substances. The Commission is not empowered to amend these limit values in the TSD via its possibilities for adaptation (see article 46 of the TSD and the section below); these limit values can only be amended via a revision of the TSD by the co-legislators. In the open public consultation, 31% (28 out of 89) of industry representatives believed that these limits should be lowered and a further 25 % were neutral about it; while 77% (24 out of 31) of public authorities and 73% (45 out of 61) of citizens and consumer associations were in favour of lowering them. In practice, most manufacturers have indicated to comply already with the lower limit values applicable in Germany, but a problem with enforcing this limit remains 79 .

b) The possibilities to adapt the Directive to new scientific knowledge on chemical substances are too limited

Having an effective and flexible system for adapting the protection from possible chemical risks in toys is crucial because scientific knowledge about chemical toxicity is constantly changing. Even chemicals with a long-known toxicity may turn out to be ‘more toxic’ when new knowledge arises 80 . In order to keep pace with latest technical and scientific developments, the Commission can amend certain parts of the Toy Safety Directive. 81   82 In particular, the Commission may establish maximum limit values for any chemical in toys intended for children under 36 months of age and in all toys intended to be placed in the mouth, and it may also amend those limits (Appendix C to Annex II). However, the Commission is not empowered to introduce new limit values for any chemical in other types of toys. This does not ensure sufficient protection of children or that the Directive can be appropriately adapted to address emerging scientific knowledge 83 .

The distinction between (1) toys for children under 36 months and toys intended to be taken in the mouth, and (2) other toys, takes account of the oral exposure to chemicals. Indeed, children under 36 months take ‘everything’ in their mouth, and toys such as a toy flute or a toy harmonica are by definition played in contact with the mouth. However, older children may also be exposed to chemicals via the skin or via inhalation. Experts in the Expert Group on Toys Safety repeatedly raised the need that older children be equally well protected for certain of these substances 84 . Examples are the sensitising preservatives benzisothiazolinone, chloromethylisothiazolinone and methylisothiazolinone for which specific limit values have been inserted in Appendix C to the Directive while SCHEER recommends that they are not used in toys 85 . In addition, the risk from chemicals may not be much different when comparing children under 36 months and older children based on bodyweight 86 . Finally, outside of Appendix C, all other chemical limit values in the Toy Safety Directive apply to all toys for children of all ages. In their letter of 25 April 2019, the 11 Ministers also stressed their concerns about the lack of possibilities in the Directive to add limit values for toys for older children. Over 93% of public authorities (29 out of 31) and 65% of industry (61 out of 89) responding to the public consutlation believed that the toy safety rules should allow for setting limit values for any toy when new scientific knowledge emerges.

A further limitation to the effectiveness of the adaptations of the Directive resides in the legal form of the toy safety rules, a Directive, which requires transposition in the national legal systems of all Member States to become applicable. Only then can all provisions be concurrently enforced in all Member States. This makes the overall process for adaptation of newly set limit values through the empowerments to the Commission lengthy. In addition, delays in the transposition in one or several Member States would be detrimental to the protection of children in those States. The Evaluation of the Directive concluded that national transpositions of amendments to the Directive often turn out to be excessively burdensome and time-consuming 87 .

2.2.2.Problem 2- A high number of toys on the Union market do not comply with the Toy Safety Directive and are unsafe

a) Global supply chains: non-compliant toys from third countries and online commerce

Evidence shows that a significant proportion of non-compliant toys come from third countries and that they are particularly present online 88 89 . Only 4% of dangerous toys notified in RAPEX come from EU countries. Online shopping is continuing to grow in the EU 90 and toys are among the most popular categories of products ordered online. For example, the 2021 Eurostat survey showed that 20% of people who bought online ordered toys or childcare articles. 91  

The value of online sales in total (not limited to toys) increased by 20% between 2010 and 2020 92 . The purchase of goods and services online has also increased the availability of products from outside the EU, with almost one third (31%) of online shoppers having purchased goods or services from outside the EU. 93   94  

At the global level, online sales of toys were estimated to have increased markedly, by an estimated 20%, in 2020-2021. 95  In 2020, 44 % of large enterprises conducted e-sales, corresponding to an e-sales value of 27 % of total turnover in this size class 96 . E-commerce enables traders to introduce products that comply with the legal requirements of a non-EU country where they were manufactured, to be sold to other markets, where product safety and chemical legislation requirements may substantially differ 97 . New business models, including variety of intermediary services (e.g., social media, online marketplaces) that connect online sellers and buyers, increase the engagement in e-commerce of even more diverse players worldwide 98 . 99

EU consumers are able to buy from non-EU websites toys which were never intended by the manufacturer to be placed on the Union market, but that had been directed to it by other third parties through global distribution chains and no economic operator clearly makes sure that the toy is compliant with the Directive. Non-EU economic operators may not know the requirements applicable in the EU or they may choose to ignore them. Non-EU sellers may only rely on the information provided by the manufacturer that the necessary steps have been taken to ensure that the toys are compliant with the applicable requirements. Rogue traders can benefit from these channels to sell their products at low prices avoiding the costs of complying with the legislation 100 . It has been found that many toys which have been recalled from the Union market because they are unsafe can still be found for sale online, sometimes from different online sellers. 70% (43 out of 61) of consumers/environmental groups, 80% (25 out of 31) of public authorities and 67% of industry and industry associations responding (61 out of 89) to the public consultation believed that the lack of specific rules on online sales hamper the effectiveness of the TSD.

b) Limits to market surveillance checks and customs controls

The data on non-compliant products set out above illustrates the ability of market surveillance authorities to target with their inspections non-compliant toys; the fact that toys are constantly one of the top categories of products subject to RAPEX notifications shows that significant efforts are put in market surveillance activities for toys. However, certain issues remain. The high rates of non-compliance show the limits to the effectiveness of market surveillance in ensuring that only compliant and safe toys are placed on the Union market.

In particular, the Evaluation concluded that the compliance information (EC Declaration of conformity) is difficult to obtain and equally often incorrect or of questionable quality and/or only drafted when requested by authorities 101 . Similarly, the safety assessment and the technical documentation also appear to be often incomplete, incorrect, difficult to obtain and only prepared on purpose when the authorities have asked for them. Parts of the technical documentation can be missing or even be faked. Re-launching a request for the obligatory documentation and the follow-up can cause considerable delays. In many cases, it is not possible to link the documentation to the toy, resulting in the documentation being of limited (or of no) value 102 .

An effective way to ensure that unsafe or non-compliant toys are not placed on the Union market would be to detect such products before they are released for free circulation. Controls on products entering the Union market take place on a risk assessment basis and on a limited number of toys entering the Union market. The recent report on the current challenges for the Customs Union “Putting more Union in the European Customs” 103 notes that the explosion of e-commerce presents customs with an inflow of small consignments with new financial, counterfeit, compliance, safety and security risks. In 2019, the volume of international trade that was handled by EU customs offices, in terms of import, export or transit was of over 868 million items 104 . From July to December 2021, e-commerce represented more than twice the number of traditional customs transactions for only 0.4 % of the value. 105 For customs, e-commerce means an exponential and unmanageable flow of millions of small individual consignments to be controlled and checked for fiscal and non-fiscal requirements, such as checking toys for compliance with the Toy Safety Directive. Data from European countries show the relevant importance of toys in cross border trade in individual parcels (10 % of small value parcels concerns toys), as well as the very high increase over the last period 106 . Evidence also suggests that the probability that small consignments will contain non-compliant or dangerous goods is very high 107 .

Figure 7: Problems and drivers

2.3.How likely is the problem to persist?

In the absence of EU action, the protection of children from chemical risks by the Directive will continue to be insufficient, and they will continue to face risks from non-compliant toys.

The most harmful chemical substances will continue to be harmful in the future. Their toxicity or hazard is not expected to decrease; on the contrary, as scientific knowledge of chemical substances progresses, new scientific evidence is expected to appear refining previous conclusions on the risks of harmful substances and identifying new substances presenting hazardous properties. As highlighted in the CSS, the sheer number of chemicals on the market represents an immense knowledge challenge, and the expected future rise in chemical production and use risks to further widen the ‘unknown territory of chemical risks’. Children will remain a vulnerable population with specific exposure to chemicals in toys, and the need for them to benefit from swift protection from the most harmful substances will remain. The TSD in its current form will continue to be unable to ensure sufficient protection of children from these risks. It would only be possible to ban new harmful substances in toys for children under 3 years or to be put in the mouth.

The high rates of non-compliant and unsafe products on the Union market are expected to remain. Market surveillance authorities will continue to perform checks on toys to remove non-compliant and unsafe toys from the Union market. However, e-commerce will continue to be used, probably increasingly, by EU customers 108 , in particular cross-border trade will continue its upward trend and the current pressure on customs and market surveillance authorities will continue to grow. From 2020 to 2021 alone the proportion of online sales coming from outside the EU increased from 20% to 23% 109 .

3.Why should the EU act?

3.1.Legal basis

The TSD is a ‘total harmonisation’ directive based on Article 114 TFEU (ex-Article 95 TEC) and follows the new legislative framework. As explained in Chapter 1, the TSD, like other EU product legislation, sets the ‘essential safety requirements’ which toys must satisfy to benefit from the free movement of products across the internal market.

3.2.Subsidiarity: Necessity of EU action

This initiative addresses the issues identified in the Evaluation of the TSD. The Evaluation concluded that the TSD is generally relevant, effective, efficient and coherent, and has EU added value, but that there was a need for specific improvements.

The TSD is key in ensuring protection of children in the EU. The main objective of the Directive is to ensure a high level of safety of children, and to allow the free circulation of toys in the EU. In particular, the TSD helps to reduce social costs by preventing accidents or harm to children that may be caused by the use of unsafe toys. A key rationale for an EU-level toy safety legislation is to provide harmonisation across Member States based on Article 114 TFEU. The TSD is a total harmonisation measure for the safety of toys, so Member States are not allowed to introduce additional safety requirements for toys; any changes to the scope or requirements of such a directive must be made at EU level.

3.3.Subsidiarity: Added value of EU action

In terms of toy safety and the creation of a large internal market for safe toys, the Evaluation has confirmed the EU added value of the Toy Safety Directive. In particular, without the Directive, Member States could set diverging limit values for chemicals, which would be to the detriment of the internal market. The Evaluation also concluded that all categories of stakeholders welcomed the existence of harmonised safety requirements across the EU, and companies valued the creation of a large market for toys and the simplification of trade as major achievements. Possibly diverging national rules were not considered as being more beneficial. Notified bodies in particular agreed that the Directive contributes to streamlining testing and standards, and public authorities welcomed the harmonisation of testing and standards and the opportunity to work together with authorities from other Member States.

A regulatory action at EU level would ensure coherent implementation of any new safety requirements for toys and thus a greater level of safety, and provide legal certainty and a level playing field for industry. Regulatory action at EU level should lay down EU-wide requirements for ensuring the health and safety of children when playing with toys, and allowing market enforcement at the national level.

4.Objectives: What is to be achieved?

4.1.General objectives

The revision of the Toy Safety Directive would pursue two general objectives, corresponding to the general objectives of the current Directive:

·General Objective 1: Ensure a higher level of safety of toys so as to guarantee a higher level of protection of children

Children are amongst the most vulnerable population in the EU; they must continue to benefit from a high level of protection when playing with or enjoying toys. This general objective will also contribute to achieve another objective of the CSS: achieving safe products and non-toxic material cycles and directly to SDG#3: good health and well-being.

·General Objective 2: Continue to guarantee the functioning of the internal market for toys, limiting the number of non-compliant toys and preserving the competitiveness of EU industry

The objective of the functioning of the EU internal market is natural to a “total harmonisation” EU product legislation like the TSD. Toys complying with the Directive can circulate freely within the single market without further requirements. At the same time, it is essential to limit non-compliant products in the Union market, which put children at risk and create an uneven playing field for reputable companies. Manufacturers producing compliant toys must be protected from unfair competition, especially in online marketplaces.

4.2.Specific objectives

The initiative pursues the following specific objectives:

·SO 1: Adapt the rules to ensure protection of children against the most harmful chemicals

·SO 2.: ensure the necessary flexibility to be able to adequately address emerging scientific knowledge and risks

The toy industry is very innovative as this is essential to maintain a competitive position. At the same time, scientific knowledge on chemical substances is constantly evolving. It is essential that the rules can easily adapt to emerging knowledge and risks.

·SO 3: reduce the number of non-compliant and unsafe toys on the Union market

There is a significant number of non-compliant and unsafe toys on the EU market. To ensure the safety of children and facilitate the development of the internal market there is a need to reduce the number of these unsafe or non-compliant toys on the market. Given the growing prominence of online sales of products coming from outside the EU, there is a risk of seeing toy manufacturers following the requirements of the TSD having to compete with manufacturers producing toys which are not in conformity with the Directive.

·SO 4: streamline enforcement of the TSD

The work of market surveillance and enforcement authorities is crucial in ensuring the reduction of the number of non-compliant toys on the market and the internal market functions properly. Rogue traders can have an advantage against reputable manufacturers by exploiting the difficulties market surveillance authorities have in enforcing the rules. As such, one of the objectives of the revision of the Directive should be to facilitate the work on enforcement by market surveillance authorities.

5.What are the available policy options?

5.1.What is the baseline from which options are assessed?

The baseline assumes no change in the current TSD. Other legislative developments taking place in parallel are nonetheless considered in the baseline. On chemical protection, the revision of the CLP Regulation is expected to define new hazard classes for harmful chemicals, such as endocrine disruptors. However, the introduction of new hazard classes under the CLP Regulation will not have any effects in terms of the protection of children under the TSD: these hazard classes are not addressed in the TSD and the Commission is not empowered to introduce new limit values in all toys. The revision of the REACH Regulation has not been presented yet. It is expected to introduce generic bans on substances for the most harmful chemicals. It is also expected to address the risk management of consumer products containing chemicals classified under the most harmful hazard classes identified by the CLP Regulation. The REACH revision is not expected to introduce an upfront generic ban of the most harmful substances in the legislation but rather to provide for the legal means for the Commission to introduce such generic bans of the most harmful substances in consumer articles. However, the extent to which it will apply to consumer articles including toys, the hazard classes it may cover as well as the timing is yet unclear. A work plan to define how the most harmful chemicals will be addressed in REACH as well as the priorities is being prepared. So far, REACH has not been able to fully address the protection of children when playing with toys described in the problem definition. The protection from harmful substances that it will offer is not expected to be as comprehensive as the targeted approach in the TSD for toys. Nevertheless, toys will remain within the scope of REACH also concerning the risks on human health and if a substance is restricted under REACH in consumer articles, this can continue applying to toys.

Under the baseline, the Commission will continue to be able to amend the TSD with its empowerment under article 46. These regular amendments are expected to continue at the same pace as since the adoption of the Directive, and will continue to require economic operators to adjust to those. Economic operators will continue to face business-as-usual costs from complying with the current obligations of the TSD and its regular amendments.

In terms of administrative burden, the total costs of the current TSD have been estimated at EUR 4000 per toy model both in case of large firms and SMEs in case of self-certification, and EUR 4500 in case of third-party certification. This includes drafting the technical documentation necessary to demonstrate compliance, the certification itself and labelling 110 . 

In terms of adjustment costs, the key costs considered are i) costs related to chemical substitution or withdrawal and ii) testing. The analysis on adjustment costs makes a number of assumptions but acknowledges that there are a number of uncertainties which do not make it possible to provide a precise estimate for the baseline or each policy option. These underlying assumptions and uncertainties concerning the cost estimates are explained in detail below 111 . In respect of the dynamic baseline, the uncertainties concern particularly i) the specific substances for which a new limit value may be introduced in accordance with the existing possibilities to adapt the Directive, ii) the number of toy models which would be affected by the new restrictions; iii) the precise costs of replacing the particular chemical substance which would be restricted; iv) the capacity of the EU industry to shift production to alternative toys; and v) in cases of product withdrawals, the extent to which demand will shift to other toys which remain on the market, or will rather shift to other non-toy products. This is why a range of economic impacts is provided for the baseline and each policy option. Adaptation and withdrawal costs are one-off costs but would occur over time with the regular amendments of the Directive; for the purpose of the analyisis a 10 year timeframe can be considered. 

As indicated above, in a dynamic baseline scenario the Commission will still be able to adapt the TSD’s chemical restrictions, and new limit values for chemical substances will continue to be introduced in the Directive in accordance with the provisions of Article 46. Such changes would lead to product adjustment costs stemming from chemical substitutions and potential market withdrawals that would mostly affect toys intended for children under 36 months or to be put in the mouth. These costs, presented below, are considered business as usual compared to which costs of the different policy options should be presented. Assuming that the market for toys intended for use by children under 36 months and toys intended to be placed in the mouth accounts for around 20% of the total toys market, any new restriction could only apply to these toys, but any new limit value would not apply to all toys in that category, only to toys in which the restricted substance is used. It is estimated that the baseline scenario would require one-off product adaptation or withdrawal costs in 0.6-1.2% of toy models 112 to comply with newly adopted limit values. Approximately 0.4-0.8% toy models would be subject to adaptation, while a smaller segment of around 0.2-0.4% could no longer be made available on the market. While there are a number of uncertainties around these cost estimates depending on the specific substances that will be limited 113 , adaptation costs to new limit values could range from EUR 6,700 per toy model produced by large firms and EUR 7,700 per toy model produced by SMEs to over EUR 15,000 per toy model 114 . Accordingly, the total one-off costs associated with product adaptation (including chemical substitution) across the EU toys industry in the baseline scenario could range from around EUR 2 million to EUR 44 million over the period considered  115 . It is expected that the costs are closer to the lower range as these mirror the costs for complying with the updates of limit values calculated in the Evaluation.

In terms of market withdrawals, although the ultimate impact would depend on the value of the toy models affected, it is possible to indicate an estimate based on the turnover achieved by the European toys industry. The baseline scenario could affect EUR 13.2 to 26.2 million worth of toys (based on the percentage of toy models of 0.2-0.4 % which could no longer be made available on the market because of new limit values and based on provisional EU industry turnover of EUR 6.56 billion for 2020) over the period considered. In addition, given the time provided to accommodate such changes, the impact of such product withdrawals would likely be mitigated by: i) the ability for producers to shift resources to the production and sale of alternative toy products; and ii) the purchasing decisions of consumers, who, instead of choosing not to purchase a product, will in many instances purchase an alternative product and still contribute to the toys market.

The Toy Safety Directive puts the responsibility for the compliance of the toy on the manufacturer, in line with all EU product legislation. Beyond one-off adaptation costs to find alternative chemicals to those which are subject to restrictions or generic bans, any new toy model will need to be tested by the manufacturer for compliance with the applicable requirements. The manufacturer being responsible for the compliance of the toy, cannot simply rely on information provided by suppliers of raw materials for the toy, as he would not be complying with its obligations under EU product legislation. Testing costs for new toy models in the future will increase, as toys will have to be tested for more chemical substances. Testing of toys as part of assessing their compliance is estimated to cost currently EUR 2,200 per new toy model, but this is not considered to be administrative burden (testing per year for the overall EU industry under the baseline would be estimated at EUR 189.25 million 116 ).

The regular amendments of the Directive should continue to introduce limit values for specific substances that will ensure better and targeted protection of children health but will be mostly limited to younger children or mouthing exposure in toys intended to be put in the mouth. This will lead to certain health benefits, targeted to those substances which have been limited, where there are specific and demonstrated risks. However, developing appropriate risk assessments for single chemicals has proven to be laborious, if not cumbersome in the past 117 . It will only be possible to introduce a reduced number of new limit values per year, as is the case now. The precise benefits on human health will only be realized after the introduction of such specific limit values and only for exposure to such substances. Furthermore, under the baseline, the increased protection of human health will only benefit younger children or children exposed to these substances through toys intended to be put in the mouth. This is because the current mechanisms to introduce new limit values in the Directive are limited to toys intended to be put in the mouth or intended for children under 3 years. 

Under the baseline, the Commission together with members of the Expert Group on Toy Safety will continue to produce guidance on the application of the Toy Safety Directive which is helpful to economic operators in understanding how to comply with the Directive. Regulation 2019/1020 on market surveillance started applying fully on 16 July 2021. It is expected to improve the efficiency of market surveillance, including for toys, through more coordination and powers for market surveillance authorities; and to facilitate checks of toys coming from third parties and in particular online by being able to contact an economic operator established in the Union and responsible for the toy. The Digital Services Act will increase the involvement of online platforms in ensuring product safety by establishing the “know your business customer” principle. This legislative measure, together with the Regulation on General Product Safety should facilitate that non-compliant and unsafe products are withdrawn from online marketplaces when they have been subject to market surveillance measures (e.g. recalls). These initiatives should have positive effects on the rates of non-compliance, in particular concerning products sold online, but a high number of non-compliant products is expected to persist, damaging competitiveness of reputable manufacturers and the single market. E-commerce will continue to be used increasingly by EU customers 118 , and the current pressure on customs and market surveillance authorities will continue to grow.

The number of inspections in toys undertaken annually is estimated at a total of 25,259 119 , and the average cost per inspection has been estimated at EUR 739 on the basis of the annual budget available. Customs authorities carry out controls on toys entering the Union market on a risk based manner. Amendments to the TSD will continue to take place on a regular basis, which will require the transposition of the new rules into national legislation leading to one-off costs per amendment of the Directive (EUR 6,174 to EUR 18,522 per Member State) 120 .

In line with the strategic foresight approach, the baseline also takes into account relevant megatrends 121 identified. Growing consumption is relevant for this initiative, and in particular new trends such as sustainable consumption and digital, as well as previous covered trends such as eco-friendly products and e-commerce. Consumers are increasingly buying online, which will strengthen the role this channel plays in toys distribution, with the challenges that it poses for toy safety. The overall trend of digitalisation should support the companies’ move towards digitalising product documentation. Other megatrends have also been factored in in the analysis, such as shifting health challenges and that new health burdens may appear and in particular the current trend towards a healthy environment. These megatrends strengthen the need to act as they show that the problem and the knowledge of hazardous substances will increase.

5.2.Description of the policy options

Three policy options were identified to address each of the problems 1 and 2. In order to address all problems that the initiative aims to tackle, option 1a, 1b or 1c (addressing problem 1 relating to strengthening the protection of children when playing with toys) would have to be combined with option 2a, 2b or 2c (addressing problem 2 related to the high number of non-compliant and unsafe toys). Addressing both different problems at the same time is essential, given that strengthening the requirements for toys will make compliance more costly, which will increase the incentives for rogue traders not to comply if the possibilities for detection are low. This risks damaging significantly competitiveness of industry if not accompanied with effective policy options to tackle non-compliance.

5.2.1.Policy options to strengthen the protection of children when playing with toys

Three policy options are presented, which are incremental, from minimum to highest level of intervention and with different approaches as regards risk management measures for harmful chemical substances. See annex 9 for the different risk management approaches in chemicals legislation and annex 9.4 for the description of the interaction between REACH and the TSD.

Option 1a – Minimum changes to the Directive

This option would keep the overall framework of the Directive unchanged, but would allow the Directive to adapt to emerging knowledge of chemical substances by imposing new limits for substances in any toy (and not only in toys intended for children under 36 months or to be put in the mouth, as in the current Directive). Therefore, more specifically, this option would:

·Empower the Commission to set, and amend, limit values for any chemicals and for toys for children of any age by introducing an additional Annex for other toys that are not for young children or to be put in the mouth. Substances would be scientifically assessed and banned from toys when they are deemed unsafe in a particular use (specific risk assessment). Therefore, the Directive would have the possibility of introducing limit values for any toy and to differentiate between toys intended for children under 3 years or to be put in the mouth (in the current Appendix C) and for toys for older children (in a new Appendix) if this would be warranted. This approach would not distinguish between chemical substances depending on the hazards they present, but would allow that any substance is limited in toys if an assessment of the substance’s hazards and the exposure to those hazards in toys concludes that they pose a risk to children.

·Reduce limit values for nitrosamines and nitrosatable substances to align them with those in force in Germany

·Require manufacturers to assess the known risks from the combination of chemicals in the toy, by providing guidance;

·Finally, a fast-track mechanism will allow the Commission in consultation with national authorities to take exceptional measures against unsafe toys presenting sudden risks not specifically foreseen in the TSD 122 .

This option would not revise the general prohibition of CMRs substances in toys nor the derogations to that prohibition; these would remain unchanged. However, specific limit values for such CMRs substances could be added for any toy. A differentiation in limit values for toys intended for children under 36 months/to be put in the mouth and toys for older children will be possible where this is warranted. Limit values for other chemical substances would also remain unchanged but could be amended. In this manner, this option would address all the problem drivers with minimum changes to the Directive.

Option 1b – Improved prevention: generic bans for the most harmful chemicals subject to derogations

This option would include the following elements:

·Extend the generic approach to risk management to other most harmful substances. This means that the existing general prohibition of CMR substances would be extended to endocrine disruptors (a new hazard class recently introduced in the CLP Regulation), substances affecting the respiratory system 123 , substances affecting the immune or neurological systems and chemicals toxic to a specific organ (currently under specific target organ toxicity – STOT - classes in the CLP Regulation). These substances would be automatically banned from toys as soon as they are classified under the CLP Regulation in these most harmful hazard classes. As these generic bans will concern the most harmful chemicals, they would apply to all toys without distinction, i.e. toys for children under 36 months or for older children as well as toys intended to be put in the mouth.

·Derogations to these generic bans would be possible. Derogations would be based on one of the current provisions for derogations to the CMR ban, and would apply when all the following conditions are met: i) when the use of the specific substances in toys has been assessed as safe for children, ii) when there are no suitable alternatives for that substance and iii) when that specific substance is not prohibited in consumer articles under REACH. In the current TSD, these derogations are adopted by the Commission following a scientific assessment and discussed with experts from Member States and industry. Under this option, the scientific assessment should be carried out by ECHA the European Chemicals Agency in Helsinki. ECHA would evaluate whether the substance, despite its hazard classification, would not expose the children to any health or safety risk in the manner in which it is used in toys. This chemical risk assessment would require the consideration of the inherent hazardous properties of the substance or a mixture and the extent of exposure to that substance or mixture in toys 124 . Derogations could only be granted where the scientific assessment concludes that the use of the substance in toys is safe and there are no alternatives. This is why derogations granted in this manner would ensure that the safety of children is not put at risk. The intention is to streamline this assessment as much as possible with other health and safety assessments done by ECHA for other EU safety legislation. The ECHA opinion would then serve as a basis for a Commission decision on the derogation request.

This option only retains one of the current derogations to the generic prohibition of substances classified as carcinogenic, mutagenic or toxic for reproduction (CMRs) and is therefore more stringent than the current system. Accordingly, derogations will only be possible if the scientific assessment concludes that the use of the substance is safe. This means that where there are uncertainties about its safety in toys, a derogation should not be granted.

·Other derogations to generic bans in the current TSD, such as the possibilities of harmful substances to be present in toys up to the relevant concentration limits for classification established under the CLP Regulation will be deleted. The Evaluation concluded that this derogation still allowed for too much presence of CMRs in toys, posing a risk to children. The derogations proposed in this option 1b would not allow for this exposure. In addition, the derogation allowing for harmful substances in toys if they are inaccessible to children will be removed too. There were uncertainties in the past on the correct interpretation of this derogation. If they are inaccessible and they are safe, this should be scientifically assessed in the context of the remaining derogation. Lower limits or limits of quantification and/or detection would be set for the purposes of verifying compliance.

·In addition, this option would include all the elements of option 1a and would allow the introduction of specific limit values for specific chemicals if these are not covered by the generic bans.

With the introduction of generic bans for the most harmful substances and the removal of the derogation allowing the presence of such substances up to the ‘relevant concentrations’ of the CLP Regulation, this option is expected to address the problem drivers in a more preventive manner. As in option 1a, whenever scientific knowledge would point to the Directive still allowing for too high presence of other chemical that is posing a risk to children, specific limit values for such substance in any toy could be introduced 125 . A differentiation in limit values for toys intended for children under 36 months/to be put in the mouth and toys for older children would be possible where this is warranted. However, in view of all the substances that would be covered by generic bans, these specific limit values should be less frequent than with the current TSD or option 1a.

Option 1c – Maximum protection from harmful chemicals: generic bans for the most harmful chemicals without derogations

This option would include all the elements in option 1a, it would also include the general bans of the most harmful substances in toys, as soon as these substances are classified under the CLP Regulation as presenting hazardous properties for human health in the most hazardous classes, as in option 1b. However, this option would not allow for derogations to the generic bans for the most harmful substances. As toys are intended for a vulnerable population group, this option would offer zero tolerance for the most harmful chemicals in toys. This would ensure the most preventive approach for children, and if new scientific knowledge emerged indicating that these harmful substances are even more harmful than considered before, exposure of children to those would have been prevented. As in option 1a and 1b, whenever scientific knowledge would point to a certain chemical that is posing a risk to children in toys and is not included among the hazard classes covered by generic bans, specific limit values for such substance in any toy could be introduced. A differentiation in limit values for toys intended for children under 36 months/to be put in the mouth and toys for older children would be possible where this is warranted. However, in view of the number of substances covered by generic bans, these specific limit values should be less frequent than under the current TSD or in option 1a.

5.2.2.Policy options to reduce the number of non-compliant and unsafe toys

Three policy options are presented, with a different focus in its intervention:

Option 2a – Extending third-party conformity assessment

To reduce the number of non-compliant toys on the European market, pre-marketing conformity assessment (third-party conformity assessment) would be extended for certain toys. The Directive currently requires third-party conformity assessment only under very specific circumstances. With the extension of third-party conformity assessment, the intervention of a notified body could be required for a) those types of toys which present higher risks, as well as b) for toys for which higher rates of non-compliance are found. As shown in section 2 above, the notifications of toys intended for children under 36 months which do not comply with the requirement not to have small parts which may cause a child to choke constitute around 40% of unsafe products notified under RAPEX. Notifications of toys non-compliant with chemical risks constitute around 40 % of notifications, too. While these may concern chemical risks on any toy, toys which are chemical substances or mixtures can lead to higher exposure of children to any harmful chemicals in them. Therefore, with this option, third-party conformity assessment would be extended to:

·Toys marketed at children under the age of 36 months old or designed to be put in the mouth. These toys are subject to the specific requirement not to include small parts which may cause the child to choke.

·Toys which are chemical mixtures or substances (i.e. such as slime, modelling clay or finger paint).

Option 2b – Facilitation of market surveillance checks and customs controls through digitalisation

To reduce the number of non-compliant and unsafe toys on the European market option 2b would include post-marketing facilitation of market surveillance checks. It would require the introduction of a digital product passport (DPP) containing the EU declaration of conformity that could be immediately consulted by market surveillance authorities. In addition, the reference of this DPP would be presented at customs so that its presence can be automatically verified when the toy enters the Union market. Release for free circulation of such toys would not be granted if it is absent. This would be based on the Digital Product Passport under the proposal for Ecodesign for Sustainable Products Regulation 126 , in order to avoid multiplication of digitalisation systems but this option is independent from the regulation of sustainability of toys under the ESPR. While toys are not currently included in the list of high priority sectors identified in the Circular Economy Action Plan, sustainability requirements under the ESPR may be imposed in the medium or longer term. Furthermore, it is expected that the DPP will become the vehicle to provide digital information on products in the future. The ESPR already foresees that the DPPs would be registered in a central registry which will be interconnected with the EU Customs Single Window Certificate Exchange System (EU CSW-CERTEX), which is the central module of the EU Single Window Environment for Customs 127 . Relying on the DPP under the ESPR and its registry for the revised toy safety rules will guarantee that these efforts are not duplicated. By means of this interconnection, when toys coming from third countries are presented at the borders, customs authorities will check that the DPP is referenced in the central registry. This will apply to any toy, either sold online or distributed through more traditional supply chains, and therefore cover also small or individual parcels containing toys and coming into the Union market from third countries. In order to be registered, the DPP for toys will have to include the EU declaration of conformity. 

All commercial goods imported into the EU require a customs declaration for release for free circulation. The main benefit of the DPP as proposed in option 2b will be that a reference to it should be included in the customs declaration for every toy, and this includes online purchases by individual EU customers from third countries. The reference of the DPP, which for toys would include the product compliance information, will also have to be included in a central registry managed by the Commission, which will be interconnected with the IT customs environment to allow for its automatic verification. This will allow that any toy which does not indicate a valid DPP reference in the customs declaration is automatically stopped at EU customs and is not released for free circulation. Customs authorities will be able to carry out this automated control on all products entering the Union market, rather than controls on a risk-based manner on a limited amount of products as is the case today. Automated controls on every toy cannot take place today; a legislative change is necessary in the Toy Safety Directive to improve controls on toys in this manner.

With this option, the revised Toy Safety rules would require that the compliance information on the toy is presented at customs when it enters the Union market. The proposed measures under option 2b can take place independently from the revision of the UCC but they would support the general objectives of the revision of the UCC, including improved access to data to improve risk management at customs, to increase the capacity of customs to detect non-compliant toys. When both are adopted, there will be better controls at the border for toys.

Option 2c – Extension of third-party conformity assessment and facilitation of market surveillance checks and customs controls through digitalisation

This option would combine options 2a and 2b.

Figure 8: Intervention logic

5.3.Options discarded at an early stage

Several potential options were discarded at an earlier stage:

Introducing essential requirements for Internet-connected or other digital aspects of toys – The Evaluation had concluded that the issue of the security of internet-connected toys and the related protection of privacy (cybersecurity) emerged as a concern: the security threats that new technologies (including toys) pose cannot be addressed by the Directive in force, because of its limited scope, which focuses on health and safety, but not on privacy and security issues. Nevertheless, since the publication of the Evaluation, the RED delegated act 128 establishing essential requirements for the personal data protection and privacy as well as cybersecurity was adopted, covering radio toys within its scope. Similarly, a proposal for a Regulation on Artificial Intelligence 129 has been put forward and should address the specific risks that the integration of AI may pose on toys. Cybersecurity should be strengthened by the Cyber Resilience Act as well, once adopted 130 . With these initiatives, the risks for privacy, those concerning cybersecurity as well as artificial intelligence should be addressed and no remaining gap has been observed in the field of toys.

Derogations for “essential uses” to the generic bans of the most harmful chemicals

The Chemicals Strategy for Sustainability committed generally to extend the generic approach to risk management (preventive bans) to the most harmful chemicals in consumer products, while still allowing their use where proven essential for society. While the concept of essential use is still under development, including for its application under REACH and possibly other downstream legislation, it is not presented as an option or criterion for derogating from the generic bans under the Toy Safety Directive. First of all, toys will remain under the scope of REACH in the future, both for human health and environmental considerations. If a harmful substance would be banned under REACH in consumer articles (be it due to generic risk to human health or to the environment), for instance because it has no essential use, it can be prohibited for toys. If this is not (yet) the case, substances under the most harmful hazard classes will be covered by the generic bans under the TSD and subject to its derogations. As the main objective of the TSD is ensuring children safety, the assessment of the derogations under the TSD will remain based on the safety of the use of the substance in toys. Consequently, in PO1b presented in this Impact Assessment, derogations to the generic bans are considered possible if (i) the use of the chemical is safe, (ii) there are no alternatives to the presence of the chemical in the toy, and (iii) the specific substance is not prohibited for use in consumer articles under REACH 131 . The latter condition would continue to ensure consistency with REACH.

Creating an injury database

The Evaluation indicated that data on injuries or accidents caused by toys was not readily available. Nonetheless, addressing this issue identified in the Evaluation through the maintenance of a European injury database has not been considered as part of the options. First of all, since the publication of the Evaluation, Regulation 2019/1020 on market surveillance has become fully applicable. This Regulation requires market surveillance authorities to share information on all the in-depth checks of compliance that they perform, and the results of those investigations, in addition to any notification of dangerous products they may need to perform through RAPEX – Safety Gate. This information should provide a comprehensive picture of non-compliance of toys in the EU including on the specific type of non-compliance and the percentage of toys inspected which are not compliant. This information can also provide information on new risks emerging in toys. As such, these data provide more thorough information on the extent to which in the Union there are unsafe toys which do not comply with the applicable requirements of the legislation, and which may be putting children at risk. In addition, the General Product Safety Regulation includes a provision on mandatory accident reporting, which would be applicable to harmonised products such as toys. This obligation is primarily for manufacturers to notify accidents caused by products, within two working days from the moment they know about the accident, to the competent authorities of the Member State where the accident has occurred. This should already ensure greater availability of data regarding the accidents related to toys. Setting up and maintaining an injury database would be costly, and it would not provide sufficiently reliable information on toy safety in the EU. Information should be filled in by doctors when accidents happen, this may not be done consistently or in a comparable manner, or in a way that identifies the product at stake precisely enough and even when accidents concern toys, they may not relate to a safety issue of the toy or to a malfunction, as many accidents can be caused by safe toys (for example, injuries caused by falling from toy scooters). Furthermore, such an injury database should be addressed generally for all products in a manner that could provide additional sources of evidence for policy monitoring, but creating it only for specific toys injuries would not be efficient.

6.What are the impacts of the policy options?

The following sections provide estimations of the costs based on certain assumptions (see annex 14 for the detailed estimates) and address in particular economic impacts for economic operators, for public authorities, competitiveness and innovation, human health and the environment. Regarding fundamental rights, none of the policy options is expected to have significant impacts, but the first three policy options should generally assist in the rights of children and their possibility to play. While the objectives of the revision of the TSD are focused on strengthening the protection of children health, the different policy options should have a limited positive impact on the environment. As such, the policy options devised respect the ‘do not significant harm’ to the environment principle but do not address it specifically. In addition, the policy options do not have a relevant impact on the climate-neutrality objective of the Union. PO2b could, by means of digitalisation, lead to increased energy needs in terms of IT servers or cloud services by toy manufacturers, but these should be covered by the overall digitalisation of industry and balanced by energy efficiency efforts by the underlying IT infrastructure. On the other hand, it should reduce the use of paper.

6.1.Policy option 1a – minimum changes to the Directive 

Economic impacts - Impacts on economic operators

In terms of administrative burden, this option would not entail a direct increase of the administrative burden reported in the baseline since it does not involve any specific change in terms of information obligations or certification. In terms of adjustment costs, the lowering of limit values for nitrosamines and nitrosatable substances to the levels already applicable in German law would have a minimal impact, as manufacturers are already complying with these lower values. Similarly, most manufacturers are already assessing and addressing the risks stemming from combination of chemicals 132 and therefore the need to consider known risks of combination of particular chemicals was not considered to lead to increased costs. The primary adjustment costs anticipated form this option are costs of chemical substitution and redevelopment resulting from new limit values which would apply to all toys (in contrast to the baseline in which new limit values only apply to toys intended for children under 36 months or to be put in the mouth). Should suitable chemical alternatives not exist, or product redesign not be possible, it may no longer be possible to make certain toys available on the market. The precise costs of product adaptation and product withdrawal will depend on the specific limit values that could be progressively introduced in the TSD and as such are difficult to quantify at this stage. Also, the scale of impact could differ based on the make-up of a manufacturer’s portfolio and to what extent they manufacture toys using the restricted substance.

Potential impacts could be estimated to require adaptation or withdrawal of 2.4-4.8% of toy models on the market in addition to the baseline 133 . In particular, it could be estimated that 1.6-3.2% of toy models would require redevelopment to identify and use safe, alternative chemicals, while the remaining 0.8-1.6% could no longer be made available on the market if no alternatives to the restricted chemicals are available 134 . Based on these estimates and the cost estimates for adjustments per toy model detailed in the baseline above which indicate that adaptation costs to new limit values could range from EUR 6,700 per toy model produced by large firms and EUR 7,700 per toy model produced by SMEs to over EUR 15,000 per toy model 135 , the total incremental one-off adjustment costs associated with product redesign and redevelopment to adapt to greater restrictions over the period considered could range from EUR 8.18 million to EUR 176.3 million compared to the baseline 136 . As this option would be similar to the current system of the TSD but allow that limit values are introduce for all toys, costs are expected to be closer to the lower range as these mirror the costs calculated in the Evaluation.

With limit values being added for new substances in toys, new toy models will need to be tested in order to ensure compliance. The increased complexity of testing for lower limit values would reportedly lead to increases in testing costs from EUR 2,200 to around EUR 3,300 per toy model when toys need to be tested for the substances subject to new limit values 137 . In view of the fact that products subject to adaptation efforts will be subject to these increased testing costs, yearly testing costs are estimated to incrementally increase compared to the baseline by around EUR 1.89-3.79 million 138 .

These are overall estimates for the EU market, given that the specific number of substances that will be addressed with future limit values for chemicals is not currently known. Depending on which substances are addressed by new limit values, companies will have to determine which products in their portfolios are affected and need to be redesigned. This is why precise cost estimates per type of company could not be calculated. Nevertheless, SMEs account for a fourth of turnover in the sector so they will bear a proportionate amount of these costs; they are also expected to have higher costs per new toy model than larger firms, as they face higher unit costs. As it was reported in the Evaluation, while large companies have had adaptation costs of EUR 6,700 per toy model, these cost are of EUR 7,700 per SME. SMEs were concerned about facing higher unit costs than large manufacturers in particular for testing 139 . Specific differences per size of SMEs were not available either.

In terms of product withdrawals, although the ultimate impact would depend on the value of the toy models impacted, it is possible to indicate an estimate based on the turnover achieved by the European toys industry. Accordingly, this option could affect EUR 52.47 to 104.9 million worth of toys in addition to the baseline (based on provisional EU industry turnover of EUR 6.56 billion for 2020). As detailed under the baseline, these impacts would be mitigated by the time provided to adjust to the updated rules, which would allow producers to shift resources to the production and sale of alternative products, and the fact that, in many instances, consumers would choose to purchase an alternative toy product, rather than not purchasing anything, thereby ensuring the revenue to the toys market remains.

Impacts on public authorities

It is likely that there would not be any additional costs for national authorities. The budget available is expected to remain the same, as well as the number of inspections. Even if the number of substances subject to limit values would increase, market surveillance authorities are expected to conduct testing only for certain types of products inspected, as is currently the case.

Impacts on innovation, competitiveness and the internal market

It has been estimated that for a number of restricted chemicals, alternatives may not be found and certain toys could no longer be made available on the market (around 0.8-1.6%). This would be mitigated in terms of the impact on the toy industry overall as consumers may instead purchase alternative products that remain on the market and that meet more stringent limit value thresholds, as well as the possibility for producers to adapt production through a sufficiently long transition period. The impacts on competitiveness will depend strongly on the specific substances concerned by a change in the limit values. Regarding the impacts on innovation, no evidence was identified that there are alternative substitutes to many chemicals used in toy production and therefore, if major changes to limit values are introduced, this could incentivise producers to search for alternatives, which could spur innovation.

This option is not expected to have net impacts on employment, as it is expected that companies will require more man hours dedicated to research of alternative chemicals and product redesign. Overall, complementing this option with another option to effectively address the high number of non-compliant toys on the Union market was considered essential by industry to preserve its competitiveness 140 . There would be positive impacts on the single market for having uniform limit values for all substances in toys, and responding to consumer expectations regarding the safety of chemicals in toys.

Social impacts - Impacts on human health

The available evidence on potential human health impacts of PO1a was limited mainly due to the lack of detail on which limit values would be updated or introduced (apart from nitrosamines). Most manufacturers already comply with the lower limits for nitrosamines 141 , but this option would ensure enforceability of those lower limits if this may not be the case and thus lead to better protection from these strong carcinogens. For other specific limit values, this option will lead to better and targeted protection of children health by ensuring that only those cases based on scientific evidence of hazards and risks would be addressed, with due account of mouthing exposure by younger children, as appropriate. This will guarantee a targeted approach for the protection of specific and demonstrated risks. However, developing appropriate risk assessments for single chemicals has proven to be laborious, if not cumbersome in the past 142 . The time between the identification of a hazard for children or from the classification of a chemical as hazardous under the CLP Regulation until it is banned in toys is likely to be too long (around 2 years in past instances), leaving children unprotected for a certain period of time and requiring prioritisation of substances to be tackled in toys. For these reasons, it may only be possible to introduce a limited number of new limit values per year. The precise benefits on human health will only be realized after the introduction of such precise limit values.

In terms of consumer welfare, as this option is only expected to have limited impacts on product redesign and adaptations, it should not lead to lower choice of products or increased prices. A certain improvement of toy safety and trust could be expected but, like the benefits for human health, these are expected to be limited.

Environmental impacts

PO 1a will not address chemical substances that are harmful to the environment as such. It will address specific chemical substances that would pose a risk to children’s health. Nevertheless, this PO could still have a minor positive effect on the environment, as it would lead to less harmful substances from toy materials once they are banned in toys. This means that less harmful chemicals would reach the environment by means of end-of-life toys in waste or recycling.

Stakeholders’ views

92% of consumers/environmental groups (56 out of 61), 94% (29 out of 31) of public authorities and 68% of industry (61 out of 89) agreed or strongly agreed that the Commission should be empowered to amend the Directive to set new requirements for chemicals in any toy.

6.2.Policy option 1b – Improved protection: generic bans of the most harmful substances with derogations

Economic impacts - Impacts on economic operators

Similar to PO1a, this option would not entail a direct increase of the administrative burden since it does not introduce any specific change in terms of information obligations or certification. Nevertheless, if industry intends to request a derogation from generic bans, it will need to submit a dossier with this request, justifying that the specific substance is safe and that there is no alternative. This cost is considered as administrative burden even if it will only be incurred if manufacturers want to continue using substances subject to generic bans in toys. This administrative burden will depend on the specific derogations that will be requested. In the past, requests for derogations have been made by industry associations, which means that the administrative cost for requesting a derogation could be expected to be shared among industry rather than born by a specific company. Past derogation requests have been estimated by the toy industry at roughly EUR 50.000 per derogation request (per industry association, not per company), including the work that manufacturers incur in verifying whether a derogation appears necessary, formally requesting the derogation through industry associations and commissioning a study on the scientific aspects. Other industries which need to request derogations for continued use of chemicals under generic bans, through similar procedures, have indicated a rough estimate of EUR 150.000 to EUR 200.000 per derogation request in the most complex cases. Based on this information, it could be estimated that the cost per derogation request could range from EUR 50.000 to EUR 150.000 per derogation request, and that there could be a maximum of two derogation requests per year, which would represent a total administrative burden for the overall industry of EUR 100.000 to EUR 300.000 per year. Derogations would be requested by companies as a means to mitigate the adjustment costs.

As for PO1a, the primary adjustment costs anticipated under PO1b are the costs of chemical substitution or product redesign and redevelopment resulting from greater restrictions on the use of chemical substances in toys. Should suitable chemical alternatives not exist, or product redesign not be possible, or derogations are not possible, certain products could no longer be made available on the market. The challenge in providing specific cost estimates is the uncertainty about the precise number of substances that might be included in extension of the generic bans 143 . The specific substances that may be covered in future hazard classes of the CLP Regulation is not yet known. Furthermore, manufacturers will need to assess the extent to which their portfolio is affected. In addition, if similar generic bans are applied to substances themselves under REACH, it will be much less costly for the toy industry to comply with this option as chemical substitution will already be carried out by chemical producers. It could be estimated that the number of substances covered by generic bans under PO1b might increase by about 10-30%. This could affect a significant number of toy models, but derogations would limit the toy models that would need to be subject to product adaptations or which could no longer be made available. A total of 8.4-12.8% of toy models additional to the baseline may be impacted under PO1b and for which a derogation may not be possible 144 , with 4.6-7.2% subject to product adaptation efforts (including chemical substitution efforts) and 3.8-5.6% could no longer be made available on the market if no alternatives to the restricted chemicals are available. Using the same data on cost estimates for product adaptation and redevelopment detailed above for the baseline (from EUR 6,700 for large firms and EUR 7.700 for SMEs per toy model to over EUR 15,000 per toy model), the estimated impact on 4.6-7.2% of all EU new toy models could result in total incremental one-off adjustment costs associated with product redesign and redevelopment of EUR 23.5 to 396.66 million compared to the baseline 145 . With more substances being subject to general bans, as well as limit values added for new substances in toys, new toy models will need to be tested in order to ensure compliance with such limit values. Due to the need to have more complex and sensitive testing for the toy models that may be impacted by the new requirements for chemical substances, the costs of testing may increase from EUR 2,200 to EUR 3,900 per toy model. In view of the products subject to adaptation efforts will be subject to these increased testing costs, yearly testing costs are estimated to incrementally increase compared to the baseline by around EUR 7.31-11.70 million 146 .

As in PO1a, these are overall estimates but the specific costs of this option and the precise impacts depending on the size of the company could not be quantified. Nevertheless, as mentioned for PO1a above, SMEs are expected to have higher costs per new toy model than larger firms, as they face higher unit costs. Also, as SMEs account for one fourth of the turnover in the sector, they are expected to bear costs in proportion to that.

In terms of product withdrawals, although the ultimate impact would depend on the value of the toy models impacted, it is possible to indicate an estimate based on the turnover achieved by the European toys industry. This option could thus affect EUR 249 to 367 million worth of toys in addition to the baseline 147 . As mentioned throughout, this would not lead to a direct market contraction of that size, given manufacturers will be provided with an appropriate transition period in which they will be able to assess the viability of existing products and, if needed, shift resources to the production and sale of alternative toy products. Moreover, consumers will in many cases simply purchase an alternative toy product rather than not purchase anything.

There would be additional costs for companies to request a derogation insofar as they will need to prepare a dossier justifying the request. These costs could not be quantified with precision at this stage, as it will depend on the specific derogations that will be requested and on whether a number of derogations are requested at the same time, in particular in the initial application of GRA. In the past, derogations have been requested jointly by industry, and sometimes for a number of substances at the same time, which helped generate savings.

Impacts on public authorities

As in PO1a, it is likely that there would not be any additional costs. The budget available is expected to remain the same, as well as the number of inspections. Even if the number of substances subject to limit values would increase, market surveillance authorities are expected to conduct testing only for certain types of products inspected, as is currently the case.

Impacts on innovation, competitiveness and the internal market

The extension of the GRA to other harmful chemicals could have certain negative impacts on competitiveness. It has been estimated that if alternatives to restricted chemicals cannot be found, 3.8-5.6% of toy models could no longer be made available on the market. The impacts of market withdrawals may be limited to a certain extent by transition periods, shift of production to alternative toys and consumers purchasing an alternative safer product. While some industry stakeholders noted that most toy manufacturers already adopt a risk management approach centred on exposure risk and do not use the substances that would be subject to restrictions under the extension of the GRA, others highlighted the potential scale of the substances that could be banned and stated that there would be a potentially significant impact on their product portfolio 148 . There are a number of concerns among industry (especially SMEs) regarding adverse impacts on competitiveness, as SMEs have less resources to allocate for research and innovation 149 . These negative impacts may be limited by the discussed implementation of the GRA in REACH. If the same hazard classes and substances are banned under REACH, it will be much easier for the toy industry to be compliant as chemical manufactures themselves will need to adapt their practices. Innovation could be stimulated e.g. through investment in R&D&I for alternative chemical substitutes. In order to preserve competitiveness of EU industry, measures under the second set of policy options to address the high number of non-compliant products are essential and which would harm even more compliant manufacturers which face these adaptation costs. PO1b is not expected to have net impacts on employment, as it is expected that companies will require more man hours dedicated to research of alternative chemicals and product redesign and that the toys that may no longer be made available on the market should not lead to loss of turnover for the overall sector. In terms of working conditions, it could lead to certain benefits in terms of reduced exposure to harmful substances of workers for toy manufacturers.

Social impacts - Impacts on human health

The human health impacts of an extension of the GRA and the removal of the derogation up to the concentration limits of the CLP regulation would bring high benefits for the protection of children health. The health benefits would be immediate for children, as the generic bans will apply as soon as substances are classified in one of the relevant most harmful hazard classes in the CLP Regulation. The study analysing the extension of the GRA in REACH 150 is looking into the overall damages that endocrine disruptors in particular are causing. It is not possible to establish the precise contribution of toys to exposure to such harmful substances. However, certain indicative health benefits are presented from the introduction of generic bans based on the value of the avoided health damage from exposure to harmful chemicals in toys, in particular to endocrine disrupting chemicals, taking into account currently identified health outcomes from exposure to these chemical substances. The estimate considers a number of assumptions and relies on the study analysing the extension of the GRA in REACH. As indicated, the precise chemical substances that would be covered under the generic bans in the TSD, as well as their current presence in toys are not known at this stage. Therefore, the precise exposure that children have today to these harmful chemicals in toys is not known either. Due to these uncertainties, the value of the health damage is estimated for endocrine disrupting chemicals only, as it is known that children are particularly vulnerable to the harmful effects of those. Exposure during childhood and puberty, even at low doses, has been found to lead to significant health damage that can only be observed at later stages. Even if the contribution of toys to the exposure to harmful chemicals is not known with precision, specific limit values in the current TSD are based on the fact that toys should only contribute to 10% of tolerable daily intakes of harmful chemicals, and 5% for particularly toxic metals 151 . Furthermore, the limit values are also based on the assumption that a child would ingest per day 100 mg of dry, brittle, powder-like or pliable toy material, 400 mg of liquid or sticky toy material, and 8 mg of scraped-off toy material. These assumptions are based on a number of studies on children exposure scenarios and mouthing behaviour and have been validated by the relevant scientific committee 152 .

On this basis, it is considered that from 1 up to 5% of health damage can be avoided due to reduced exposure to endocrine disruptors in toys. For other classes of harmful substances, the precise contribution that toys could have to the health damages and the benefits that could be expected from the proposed policy options were considered more uncertain and could not allow for reasonable estimates. Even if only 1-5% of the exposure could be attributed to toys, banning those substances from toys would have considerable health benefits (EUR 240 million to EUR 1.2 billion per year)  153   in terms of avoided health damage. This is compared to limited benefits under the baseline in terms of reduced exposure to specific substances in toys intended for children under 3 years or to be put in the mouth and which will only happen after a certain time after the substance has been found to pose a risk to the endocrine system and a limit value for it has been added to the Directive. It is worth noting though that comparing them with the cost estimates is very difficult as the benefits would accrue over the life time of a child exposed (or not exposed) to endocrine disruptors now which means that the time span can be over several generations far exceeding standard appraisal periods of 20-30 years, while industry will face adaptation costs after the entry into force and start of application of the TSD and thus in an earlier timeframe 154 . As derogations will be allowed only when it has been considered that the use of that chemical in toys is safe, derogations to general bans should not compromise the health benefits of this option. In addition, by removing the derogation that allows substances subject to generic bans to be present in quantities below the concentration limits for classification under the CLP Regulation, this minimum exposure will also be eliminated. Requiring regular reviews of the derogations granted, it will ensure that any new scientific knowledge that may have emerged is duly considered. In addition, as is required by the current derogation procedure, derogations granted should be reassessed every five years to ensure that their use remains safe. Product withdrawals estimated would result in corresponding reduction of exposure to the most harmful substances. This should lead to direct health benefits, in particular as these substances may cause harm at very low doses, and to a vulnerable population which is particularly sensitive to such harmful effects.

In terms of consumer welfare, this option may have certain impacts on the choice of products or increased prices, though these should be moderate, as the possibility to have derogations should limit the impacts on categories of products. Improvements in toy safety and trust could be expected, in line with the significant benefits for human health that can be expected from this option.

Impacts on the environment

PO1b would address substances that are hazardous for human health. It would introduce generic bans on substances classified as hazardous for human health under certain hazard classes in the CLP Regulation, as well as additional limit values for other substances that may pose a risk to children health. However, this option would have certain benefits for the environment too. The CLP Regulation includes different hazard classes for hazards to human health or the environment. However, harmful substances can be classified under several such hazard classes at the same time. Substances which can be harmful both for human health and for the environment, such as EDs and CMR would not be present in toys, as they would be subject to generic bans due to their hazardous classification for human health. This would lead to certain benefits for the environment. If fewer harmful chemicals are included in toys, there would be less harmful substances that would reach the waste stream or recycling. Derogations to generic bans will still be possible under this option, limiting these benefits on environmental protection.

Stakeholders’ views

Over 85% (between 50 and 55 out of 61, depending on the specific harmful substance) of consumer groups and environmental associations would like to see the most harmful substances banned preventively. Also while 58% of public authorities (18 out of 31) would like to see certain of endocrine disruptors preventively banned by the TSD, 40% (13 out of 31) considered that generic bans should also cover other harmful substances such as substances affecting the immune or neurologic systems. Only 10% (8 or 10 out of 89, depending on the substance) of industry supported this option.

6.3.Policy option 1c – Maximum protection: generic bans of the most harmful substances without derogations

Economic impacts - Impacts on economic operators

Concerning administrative burden, as for PO1a and PO1b, this option would not entail direct costs because it does not affect information obligations, labelling or certification. In terms of adjustment costs, the removal of all derogations under PO1c would implement further adjustment costs on industry stakeholders. In particular, this option could lead to a significant number of toys that could no longer be made available on the market. Indicative estimates of the potential adjustment costs related to PO1c could lead to a total of 19.4-28.8% of toy models in addition to the baseline 155 that will be impacted under PO1c, with 9.6-14.2% subject to product adaptation efforts (including chemical substitution) and 9.8-14.6% that could no longer be made available on the market. Using the same data on cost estimates for product adaptation and redevelopment detailed above in the baseline (from EUR 6,700 for large firms and EUR 7.700 for SMEs per toy model to over EUR 15,000 per toy model), the estimated impact on 9.6-14.2% of all EU toy products could result in total incremental one-off adjustment costs associated with product redesign and redevelopment of EUR 49.1 – 782.3 million compared to the baseline 156 . As in PO1b, due to the need to carry out more complex and sensitive testing, testing costs would increase from EUR 2,200 to around EUR 3,900 per new toy model. In view of products subject to adaptation efforts will be subject to these increased testing costs, yearly testing costs are estimated to incrementally increase compared to the baseline by around EUR 14.62-21.94 million 157 .

In terms of product withdrawals, although the ultimate impact would depend on the value of the toy models impacted, an estimate based on turnover would imply that this option could affect EUR 642 to 957 million worth of products (based on provisional EU industry turnover of EUR 6.56 billion for 2020). As mentioned throughout, this would not lead to a direct market contraction of that size, given manufacturers will be provided with an appropriate transition period in which they will be able to assess the viability of existing products and, if needed, shift resources to the production and sale of alternative toy products. Moreover, consumers will in many cases simply purchase an alternative toy product rather than not purchase anything. However, the lack of derogations will render it very difficult to find alternatives for certain categories of products. Derogations under option 1b would set out permitted uses in toys which have been considered safe for prohibited substances. As such, they would be included in the legislation itself, benefitting the whole industry and not only the applicant. Since the adoption of the Toy Safety Directive including a generic ban for CMR substances, only 3 derogations have been requested. Even if the number of derogations requested under PO1b is not expected to be high, the absence of specific derogations for certain substances may mean that an entire category of toys could no longer be available in the Union market. For example, nickel is classified as a carcinogenic substance and as such is not allowed in toys in accordance with the provisions for generic bans of CMRs of the current Toy Safety Directive. Nevertheless, it is allowed in toys and toy components made of stainless steel or in those toy components which are intended to conduct an electric current, and it actually prevents corrosion of metals. Nickel is allowed in toys because it has been scientifically assessed in the past that nickel present in toys as set out above does not pose a risk for children 158 . According to the scientific opinion of the predecesor of SCHEER, nickel only causes tumours in the respiratory tract after inhalation exposures to nickel-containing dusts and fumes, but not after oral intake. Since inhalation of nickel from toy materials intended to transmit the electric current is extremely unlikely, SCHER concluded that a tumor risk due to nickel exposure when handling toys is not present 159 . Nickel in these uses has thus been considered to be safe and that it does not expose children to the hazardous properties of the substance. Nevertheless, without a derogation to allow for the presence of nickel in the materials described above, electric toys would not be allowed on the Union market. This is why the adjustment costs and product withdrawals in PO1c are significantly higher than in PO1b.

As in PO1a and PO1b, the number of specific substances that will be covered by the generic bans, as well as the precise number of products impacted and the availability of alternatives are not possible to quantify with precision. Thus, data on the detailed impacts on the different sizes of company is not available. However, as in PO1a and PO1b, SMEs are expected to have higher costs per new toy model than larger firms, as they face higher unit costs and, as they account for a fourth of the turnover in the EU market, they are expected to bear costs in that proportion.

Impacts on public authorities

It is likely that there would not be any additional costs for national authorities. The budget available is expected to remain the same, as well as the number of inspections. Even if the number of substances subject to limit values would increase, market surveillance authorities are expected to conduct testing only for certain types of toys inspected, as is currently the case.

Impacts on innovation, competitiveness and the internal market

This option would have a negative impact on competitiveness due to the inability of manufacturers to rely on derogations to generic bans. If there are no viable alternative substitutes, it has been estimated that a number of products could no longer be made available on the market (9.8-14.6%). This would be unlikely to be fully mitigated as consumers would not be able to purchase alternative products not containing these chemicals if they are necessary for production in the specific product concerned across the whole product category. There could therefore be loss of revenue for the overall toy sector. The removal of the derogations related to the use of nickel (which is a CMR substance) in certain forms for example would lead to a significant withdrawal of toys across different product lines, as its use is necessary in toys and toy components made of stainless steel and toy components which are intended to conduct an electric current 160 . Therefore, the non-availability of a specific derogation for this substance would have impacts on an entire category of toys.

The vast majority of SMEs responding to the public consultation disagreed with the option not to have derogations to general bans (67.4 % - 31 out of 46). SMEs view this option as having significant adverse impacts on firm and sectoral level competitiveness, as the impacts of the option on product categories will be significant and SMEs would have less resources available to invest on research and development of products which would not contain these chemicals. This option could have some positive impacts on innovation. As with the other options, PO1c should be accompanied by measures under PO2 to address the high number of non-compliant toys, and which would harm even more competitiveness of compliant manufacturers which face these adaptation costs.

This option could lead to certain negative impacts on employment. Even if, as in the previous PO1a and 1b, it is expected that companies will require more man hours dedicated to research of alternative chemicals and product redesign, the extent of the restrictions could lead to loss of turnover for the overall sector. In terms of working conditions, it would limit exposure of workers for toy manufacturers to harmful chemicals, and this limitation would be higher than in PO1b in view of the lack of derogations.

Social impacts - Impacts on human health

The extension of the GRA and the removal of derogations on CRM substances would have strong benefits to human health. Exposure of children to such harmful substances would be significantly reduced, similar to PO1b. The health benefits would be immediate for children, as substances would be banned from toys as soon as they are classified in one of the relevant most harmful hazard classes in the CLP Regulation. The product withdrawals estimated would be due to the presence of these harmful chemicals in such toys. Their withdrawal will result in a corresponding reduction of exposure of children to the most harmful substances (only toys containing these harmful substances will be withdrawn from the market). The measures envisaged in this option to address chemical substances will lead to direct health benefits, in particular as these are the most harmful substances that generally cause harm at very low doses, and exposure will be reduced for a vulnerable population which is particularly sensitive to such harmful effects. If the criteria for the derogations under PO1b are sound and the risks can be clearly limited with the conditions to grant such derogations, PO1c would not provide significant additional health impacts but similar health impacts as PO1b (EUR 240 million to 1.2 billion EUR per year) 161 , based on the assumptions and estimates as explained under that policy option. Derogations should only allow uses of substances subject to generic bans for which safety can be scientifically concluded. Generic bans will cover hundreds of substances while derogations are expected to cover only a limited number of them and for limited uses. Derogations under PO1b would be of general application to allow certain uses of substances in toys, but only where this has been considered to be safe for children in terms of exposure to the substances. In addition, derogations should be reassessed every 5 years.

The estimated health benefits stem from the introduction of generic bans for the most harmful chemicals. Generic bans are proposed in both options 1b and 1c and will apply to a significant number of substances and will protect children from exposure to these harmful chemicals. Based on the explanations above, derogations should not allow for exposure to the harmful effects of these chemicals and this is why the estimates for health benefits in both option 1b and 1c should be quantitatively similar. If derogations are only granted for a use that has been evaluated as safe and there are no alternatives, it should not diminish the overall estimated health benefits achieved by reducing exposure of children to these substances. The overall health benefits in option 1c should not be significantly higher than in PO1b. However, derogations could affect entire categories of products; see the example of nickel explained above which could affect all electric toys on the Union market. This is why, while the health benefits of PO1c are expected to be substantially the same as in PO1b, the economic impacts in terms of adjustment costs in PO1c are expected to be much higher than in PO1b.

In terms of consumer welfare, this lack of possibility for derogations is expected to limit significantly the choice of products and possibly lead to higher prices. Improvements in toy safety and trust could be expected, but comparable to those under PO1b.

Impacts on the environment

Similar to PO1b, PO1c would introduce generic bans on harmful substances classified as hazardous for human health under certain hazard classes in the CLP Regulation, as well as additional limit values for other substances that may pose a risk to children health, this option would have certain benefits for the environment. The CLP Regulation includes classification of harmful substances as hazardous for human health or the environment. Chemical substances can be classified under several such hazard classes at the same time. Where substances are considered as hazardous not only for human health but also to the environment and are subject to generic bans in toys, this option will have positive impacts on the environment. This option will bring more benefits to the environment than the previous options, given that substances which can be harmful both for human health and for the environment, such as EDs and CMRs, would not be present in toys at all. If there are no derogations to the generic bans, this should mean that less harmful substances will be present in the waste stream, which would thus have a positive impact on the environment. Less harmful substances would be present in recycled materials from toys, which could then be used more safely. In addition, this option may have other indirect positive impact on the environment; by leading to higher prices of toys, there may be less consumption and waste.

Stakeholders’ views

A number of respondents to the public consultation believed that there should be no derogations to the general bans: most support came from consumer and environmental groups (50% - 31 out of 61- of respondents) while only 35% (11 out of 31) of public authorities supported that there would be no derogations. 73% (66 out of 89) of industry respondents disagreed with removing derogations from general bans. Consumer and environmental groups believed that maximum protection should be offered to children, while public authorities and industry were concerned about the impacts on competitiveness and choice of toys that the lack of derogations would have.

6.4.Policy option 2a: extending third-party conformity assessment

Economic impacts: Impacts on economic operators

This option would lead to additional costs in terms of administrative burden for companies. The proportion of new toys undergoing third party conformity assessments would increase from currently 3% to about 20% 162 . The current administrative burden has been estimated at an average of EUR 4,000 per new toy model in case of self-certification and EUR 4,500 in case of third party conformity assessment. Applying this extra cost to the percentage of additional toy models which would be required to go through third party conformity assessment, it has been estimated the yearly increase of the administrative burden for the EU market would amount approximately to EUR 7.8 million 163 . No differences in the administrative burden were identified depending on the type of company, as these costs are associated with the amounts that Notified Bodies charge for their services. There would be no adjustment costs stemming from this option.

For notified bodies, the number of EU-type examinations would be expected to increase, leading to benefits. Notified bodies may incur costs relating to the increase in the testing capacity required to undertake these additional tests, since only around 3% of the toys in the EU market are currently subject to third-party testing. 164 These costs are not expected to be borne by notified bodies but rather to be passed on to manufacturers who will pay for the additional EU-type examinations.

Impacts on public authorities

It can be estimated that this option will reduce the time necessary to collect information for market surveillance authorities, as the time needed to collect information from a notified body is significantly lower than economic operators. It can be estimated that PO2a would result in a 5% time saving for the average inspection by market surveillance authorities which would translate into a 5% increase in the number of inspections 165 , as it could be assumed that the budget allocated to market surveillance activities in toys will remain the same and that the efficiency gains will be redirected to additional toy inspections. The number of inspections in the EU is expected to rise to a maximum of an additional 1000 per year (from 25,259 to 26,522) leading to a reduction of the cost per inspection from EUR 738.52 to EUR 703.35 166 . There would also need to be more efforts, and costs, to control conformity assessment bodies’ quality by public authorities.

Impacts on innovation, competitiveness and single market

The main negative impacts on competitiveness were considered by industry to include: the higher costs of mandatory third-party conformity assessment for toys, and increased lead times to market. In particular, while the costs of lead times to market are difficult to quantify, mandatory third-party conformity assessment could add an additional 4-6 months to the process of placing a product on the European market, with an adverse impact on competitiveness. The reduction of non-compliant toys on the single market would lead to positive effects on competitiveness for EU and other compliant industry, as they will face less competition from rogue traders selling non-compliant products, often at cheaper prices. However these positive effects on competitiveness will not be very significant, as the improvements on the number of non-compliant toys are expected to be limited. This is because this option will facilitate enforcement by market surveillance authorities to a limited extent, and therefore it will not significantly increase the incentives to comply for rogue traders. It may be expected that those manufacturers that already comply with the requirements of the TSD will also comply with the obligation to perform third party examination, but those rogue traders which already ignore the rules will continue to do so. Nonetheless, these limited positive effects will likely not offset the costs for EU industry (in particular SMEs) who will continue to face unfair competition from rogue traders to a significant extent. SMEs were particularly concerned by this option, as they were of the opinion that this would adversely affect their competitiveness.

This option is not expected to have impacts on employment by toy manufacturers, but a slight increase in employment by Notified Bodies would be likely, as they would need to ensure additional capacity to provide conformity assessment services.

Social impacts - Impacts on human health

The available evidence suggests that only a limited improvement in the number of illicit toys is to be expected from this option, as explained in the previous sections on impacts on public authorities and on competitiveness, innovation and the single market. Accordingly, the risks of illicit toys would only decrease slightly and this option would only have limited benefits on children’s health. Similarly, as the reduction of non-compliant toys is expected to be limited, this option is not expected to have significant impacts on consumer welfare and trust.

Impacts on the environment

This option is not expected to have positive impacts on the environment. Illicit toys could be more harmful to the environment as they could contain more dangerous substances, therefore, a reduction of the number of such toys would mean less pollution. Less harmful substances would be present in recycled materials from toys, which could then be used more safely. However, it is expected that this option will only lead to minor reductions on the number of non-compliant toys, thereby yielding limited benefits to the environment.

Stakeholders’ views

In the public consultation, 83% (51 out of 61) of consumer/environment groups and 75% (23 out of 31) of authorities supported the extension of third-party conformity assessment while 75% of industry was against it (67 out of 89). Industry believes that it will impose additional costs without bringing a significant improvement to the problem of a high number of non-compliant toys.

6.5.Policy option 2b: facilitation of market surveillance through digitalisation

Economic impacts - Impacts on economic operators

This option requires the introduction of a digital passport that would include the EU declaration of conformity so that it is digitally accessible (via a machine-readable code) and the registration of the reference in a central registry. This will be based on the Digital Product Passport proposed under the ESPR. Toys may be covered in the longer term by one of the delegated acts setting sustainability requirements, however they are currently not included in the list of high priority sectors identified in the Circular Economy Action Plan (CEAP) adopted by the Commission in March 2020. This is why the impacts for economic operators are considered under PO2b and not part of the baseline. In terms of administrative burden, this option can lead to one-off costs (such as development of required IT tools, new procedures for the preparation of the product documentation, and potentially additional IT services) and recurrent costs (e.g. the FTE required to collect and provide electronically the required documentation). For the calculation of the overall market costs, SMEs have been assumed to incur in yearly costs of EUR 3,000 per company, while for larger companies, up to EUR 140,000 167 , but the specific breakdown of these costs into one-off and recurring costs for the toy sector was not available. Previous research in other markets has found a split of 80% one-off vs 20% annual costs for large firms, and 62% one-off vs 38% annual costs for SMEs 168 . Applying this breakdown to the estimates for toys 169 would suggest a total one off cost of EUR 18 million in addition to an annual recurring cost of EUR 10.5 million. It is therefore expected that companies will incur in significant initial costs, but that these would be lower in subsequent years thanks to economies of scale and the fact that after the systems are set up and most initial data entered, there are only expected to be additional costs related to updating and maintenance costs.

The Digital Product Passport will contain the compliance documentation that economic operators must possess and be able to present upon request in any case. This option is also expected to lead to savings as compliance information that economic operators currently need to create will only be provided digitally. The digital passport under option 2b will replace the EU declaration of conformity that manufacturers have to produce currently for every toy. Previous studies have suggested a benchmark of about 10-15% savings in administrative costs from the full digitalisation of product information (not only compliance but also user-related information) 170 . These studies also estimate the cost of the provision of compliance information at 0.4% of turnover of the sector. 171  Accordingly, savings could be estimated at around EUR 2.62 to EUR 3.93 (EUR 3.275 on average) million per year  only from moving to the digital provision of compliance information. 

In addition, there would be savings from the immediate availability of information when dealing with market surveillance authorities for specific inspections. These are independent from the savings on the digitalisation of compliance information that are detailed above; these savings will happen when toys are subject to inspections. The Evaluation provides an indication of the average cost – per new toy for an average importer or distributor – to “ensure that the toy is accompanied by the required documents”. 172 The average cost of such activities per new toy model are estimated to be between EUR 548 and EUR 651. This can serve as a basis to estimate the amount of time and relative cost a company would incur in retrieving all the necessary information for a toy model and supplying it to market surveillance authorities when inspections are conducted 173 . With the implementation of the DPP, this cost would likely disappear for companies, since relevant documentation would be available online for market surveillance authorities, as well as the supply chain information. Since the annual number of inspections in Europe is calculated to be around 25,000 (and may rise to 30,000 in case of implementation of the DPP), and each inspection would concern a specified toy model, it can be estimated that the average annual saving for EU companies could be between EUR 13 million and EUR 16 million in case inspections remain at the same levels or increase slightly, or even up to EUR 20 million in case of increased number of inspections.

It is reasonable to assume that in the long term the internal costs for companies to implement the DPP will diminish, and the benefit of less burdensome inspections for compliant companies will lead to a visible economic benefit in terms of reduction of administrative burden. Efficiencies are expected from having all the information about a given toy available via a single QR code or a similar machine readable item, which is how the DPP will be implemented in practice.

Impacts on public authorities

While it is difficult to assess the exact scale of the reduction by this option in the time needed to undertake inspections, it is estimated that it could range between 10% to 20%, which could translate into an equivalent percentage increase in the number of inspections which could be undertaken with the resources available 174 . The number of inspections in the EU is expected to rise to a maximum of an additional 2500 to 5000 per year (from 25,259 to between 27,785 and 30,311) assuming that the overall budget for market surveillance authorities for toys will remain the same and that efficiency gains will be reinvested in additional toy inspections, thus lowering the cost per inspection from EUR 738.52 to between EUR 671.37 and EUR 615.42 175 . There would be one-off costs relating to market surveillance authorities adapting their information systems and procedures to the digitalised information but this would already be covered by the ESPR 176 ; additional costs are not expected to be attributed to the revision of the TSD. The ESPR also foresees that the reference to the DPP will be included in a central registry, managed by the Commission, as well as the interconnection with the underlying IT infrastructure for customs purposes, in order to allow for the automatic verification of the existence of a DPP at the time of import. In terms of customs controls, this option would entail the submission of the required documentation by the importer at a stage before the actual shipping of the toy, which will apply to any toy entering the EU. This would lead to potential savings for customs with less shipments on hold at the border 177 . Customs authorities would check through automated systems the provision with the imported shipment of the DPP, thereby ensuring a minimum automatic check of the existence of the DPP on any toy entering the EU, either in small consignments or in higher volumes. The actual content of the DPP and the toy itself could still be subject to checks by the relevant national market surveillance authorities. Carrying out automatic controls would allow customs to reduce their costs and likely allow more effective and targeted controls.

Impacts on innovation, competitiveness and single market

This option is expected to lead to significant benefits for the competitiveness of companies. Firstly, the digitalisation of product information (including regulatory compliance aspects) could lead to greater efficiencies and cost savings. Once toys are subject to the ESPR, relying on the DPP already for compliance under the toy safety rules will lead to important savings for the toy industry. In addition, the information that could be mandatorily required under the ESPR in the DPP could bring synergies if it helps contribute both towards meeting toy safety compliance requirements, for example, and information on specific chemicals. Furthermore, this option is expected to lead to important reductions in the number of non-compliant toys in the EU. This is because, first of all, it will allow market surveillance authorities to be more efficient in their inspections and target products without the DPP which may be more likely non-compliant. In addition, currently toys entering the Union market from third countries are subject to controls at the border on a risk- based approach. Under PO2b the presence of the DPP in toys will be automatically verified at customs when toys are presented for release for free circulation. These will apply to any toy entering the Union market, either bought online or through other distribution channels. Toys without the DPP will be prevented from being released for free circulation in the Union market. These automatic controls at customs are expected to have significant impacts in reducing the number of non-compliant toys entering the Union market. This will prevent that toys not intended for the Union market reach EU customers. It will only be possible to create the DPP if a valid EU declaration of conformity exists for the toy. All toys imported into the Union market will have the EU declaration of conformity, and this will be easier to verify when market surveillance authorities inspect them.

The overall reduction of the number of non-compliant toys will have a positive impact on the competitiveness of compliant manufacturers by improving the level playing field and curtailing illicit competition. A shift in demand to compliant products in the EU can also occur. SMEs were concerned about the impacts of this option on their competitiveness as they are less able to invest in digitalisation compared with large firms. Despite this, this option was still seen very positively by industry, both by large firms and SMEs 178 . Digitalisation of product information was therefore welcomed by industry as a means of modernising the provision of regulatory compliance and supply chain information, despite certain concerns on costs by SMEs. This in turn could improve the functioning of the single market as digitalisation of certain information is being actively considered and is close to being introduced in other EU legislation (e.g. fertilisers, chemicals and detergents) and other global jurisdictions that have also embraced digital labelling.

This option could have certain positive impacts on employment, as toy manufacturers will need to dedicate resources to ensure the digitalisation of information. There might be a shift in demand towards IT skills and qualifications.

Social impacts - Impacts on human health

The introduction of a DPP could have significant impacts on the number of illicit toys and in protection from the health risks of those. Controls at customs of the presence of the DPP would be rendered automatic in the first place, facilitating that a significant percentage of non-compliant toys are prevented from being released for free circulation in the Union market. In terms of consumer welfare, it will lead to positive impacts on product safety and consumer trust, and should not affect the choice or price of toys.

Impacts on the environment

As explained in PO2a, non-compliant and dangerous toys are more harmful to the environment, therefore, a reduction of the number of such toys would mean less pollution. As this option is likely to significantly reduce the number of non-compliant toys, the benefits on the environment may also be relevant. Less harmful substances would be present in recycled materials from toys, which could then be used more safely. Additionally, the digital passport for toys could help reduce the number of papers used by economic operators and national authorities; however this impact could not be quantified.

Stakeholders’ views

59% of consumer/environmental groups (36 out of 61), 96% of public authorities (30 out of 31) and 52 % of industry (47 out of 89) supported that toys should have a DPP with compliance information to improve enforcement.

6.6.Policy option 2c: Extending third party conformity assessment and facilitation of market surveillance through digitalisation

Economic impacts - Impacts on economic operators

The costs and savings of this PO2c will be the sum of costs and savings estimated for PO2a and PO2c. No additional adjustment costs are expected.

Impacts on public authorities

While it is difficult to assess the exact scale of the reduction in the times needed to undertake inspections under PO 2c, the effects of option 2a and 2b are expected to be aggregated. As such, the time savings for inspections would be in the range of 15% 25% 179 . The average number of inspections in the EU is expected to rise to a maximum by an additional 3500 to 6000 per year (from 25,259 to between 29,048 and 31,573) assuming the budget allocated for market surveillance in toys will remain the same and that the efficiency gains will be reinvested in additional toy inspections thus lowering the cost per inspection from EUR 738.52 to between EUR 642.18 and EUR 590.83 180 . Other impacts for market surveillance and customs were already highlighted in the analysis of PO2b.

Impacts on innovation, competitiveness and single market

The impacts on innovation, competitiveness, the single market and employment were already highlighted in the analysis of PO2a and PO2b. However this option will lead to a stronger reduction of the number of non-compliant toys thereby strengthening the competitive position of compliant manufacturers.

Social impacts: Impacts on human health

This option will have strong impacts on improving human health, as it will lead to significant reductions of the number of non-compliant toys. The two PO2a and 2b combined in PO 2c will reinforce each other. A real improvement in market surveillance could be achieved with more certainty on the assessments coming from third party conformity assessments and the information more easily available with the DPP. The impacts on consumer welfare in terms of increased product safety and trust will also be high.

Impacts on the environment

Both measures combined would lead to less non-compliant toys in the EU market, and thus more benefits for the environment. Less harmful substances would be present in recycled materials from toys, which could then be used more safely. Digitalisation should lead to less paper consumption.

7.How do the options compare?

7.1.Effectiveness of the proposed policy options

The following table presents the effectiveness of the proposed options against the relevant specific objectives.

Legend: +- no relevance; + limited effectiveness; ++ effective; +++ very effective

Introducing generic bans for chemicals classified under the most harmful hazard classes 181  as proposed in options 1b and 1c is the most preventive approach for regulating chemical substances. This is because substances are preventively banned in toys as soon as they are classified without having to assess whether their use in toys poses a risk to children (as required in option 1a). These specific risk assessments in toys have proven to be laborious and time consuming in the past, although ensuring tailor-made protection. The time between the identification of a hazard for children or from the classification of a chemical as hazardous under the CLP Regulation until it is banned in toys is likely to be too long (around 2 years in past instances), leaving children unprotected for a certain period of time and requiring prioritisation of substances to be tackled in toys. This is why PO1a is not considered as effective in ensuring protection of children from harmful chemicals.

PO1a will ensure that the TSD can adapt to emerging scientific knowledge though in a less streamlined manner than PO1b and PO1c. This is because under PO1b and PO1c, as soon as substances are classified in one of the most harmful hazard classes under the CLP Regulation, they will be banned in toys. PO1a will require the identification of specific substances that are found to pose a specific risk in toys, substance per substance, and carry out a detailed risk assessment of their use in toys, before they are actually regulated.

PO 1b and PO1c will be similarly effective in ensuring SO1 and the protection of children from the most harmful chemicals, as substances will be preventively banned in toys as soon as they are classified under the most harmful hazard classes in the CLP Regulation, and derogations will only be possible in PO1b if substances are safe for use in toys. The possibility for derogations proposed under option 1b would constitute the main possibility for costs mitigation. If there would be an absolute ban of the most harmful substances without the possibility to have any derogation, the consequences for the toys industry and for entire categories of toys would be extremely severe, without corresponding benefits in the protection of children. For example, as explained above, nickel is classified as a carcinogenic substance. Nevertheless, it is allowed in toy and toy components made of stainless steel or in those toy components which are intended to conduct an electric current, and it actually prevents corrosion of metals. It has been scientifically assessed in the past that nickel present in toys as set out above does not pose a risk for children 182 . Without a derogation to allow for the presence of nickel, electric toys would not be allowed on the Union market.

As regards the objectives to reduce the number of unsafe toys and facilitate compliance and enforcement, PO2a will be less effective in reaching the objectives than PO2b, as PO2a is expected to allow a moderate increase in the number of inspections, while option PO2b is expected to lead to more significant efficiency gains and be more effective in increasing inspections. All commercial goods imported into the EU require a customs declaration for release for free circulation. The main benefit of the DPP as proposed in option 2b will be that a reference to it should be included in the customs declaration for every toy, and this includes online purchases by individual EU customers from third countries. The reference of the DPP, which for toys would include the product compliance information, will also have to be included in a central registry managed by the Commission, which will be interconnected with the IT customs environment to allow for its automatic verification. This will allow that any toy which does not indicate a valid DPP reference in the customs declaration is automatically stopped at EU customs and is not released for free circulation. Customs authorities will be able to carry out this automated control on all toys entering the Union market, rather than controls on a risk-based manner on a limited amount of toys as is the case today. Automated controls should have a significant impact on the number of non-compliant toys reaching the Union market. It will prevent that toys not intended for the Union market reach EU customers. All toys imported into the Union market will have the compliance information, and this will be easier to verify when market surveillance authorities inspect them.

Furthermore, PO2a will not be as effective in addressing rogue traders who deliberately ignore the rules to gain a competitive advantage, as it will not increase the probability of getting caught and these traders will probably not carry out the third-party conformity assessment. PO2c is however the most effective as it will cumulate the impacts of PO2a and PO2b in achieving the objectives. Nonetheless, most of the effectiveness in PO2c is attributable to PO2b.

7.2.Impacts of policy options

The following table summarises the costs and benefits, for each policy option, quantified and presented in section 6.

Costs & Benefits summary table 183

Summary of incremental costs per policy options. See annex 14 for details.

The aim of the assessment is to provide ranges of the magnitude of potential impacts generated by each policy option. While costs are borne by industry and the benefits of PO2b and 2c are also savings for industry, PO1b and 1c present overall health benefits for citizens and society. The following table summarises the quantitative and qualitative assessments, presented in section 6, on the potential effects of policy options to industry, consumers, internal market, and competitiveness as well as on the environment.

Summary of Economic, social and environmental impacts

Legend: +- no / neutral impact; + minor positive impact; ++ positive impact; +++ significant positive impact; - minor negative impact; -- negative impact; --- significant negative impact

The scale of the negative economic impacts will increase with the level of intervention. In contrast, however, the policy options with high levels of intervention are anticipated to have potentially significant positive social impacts on the protection of health of children.

PO1a would have certain negative impacts for industry and competitiveness but slight positive impacts on innovation the functioning of the internal market and the environment. It would also have limited impacts on the protection of children. PO1b and 1c would have strong positive impacts on the protection of children due to the more preventive approach towards the most harmful substances in toys. These options would entail important costs for industry, higher in PO1c than in PO1b. While PO1b would mitigate to a certain extent the negative consequences in costs and competitiveness for companies by allowing derogations to the generic bans, PO1c would lead to very important number of product withdrawals which will lead to high costs for industry and hinder innovation. If derogations are granted only where safety and protection of children is not compromised, there will not be substantial differences in terms of benefits for human health between PO1b and PO1c.

PO 2a would have some benefits for the internal market and minor impacts on competitiveness and innovation. However, it would have important costs for the industry while not clearly yielding very high impacts on reducing the number of non-compliant products in the EU. Under policy option 2b, a certain economic impact is to be expected to adjust to digitalisation, with a potential for savings over the longer run in producing compliance information as well as in simplifications of contacts with the market surveillance authorities. The incremental benefits of option 2c in comparison with 2b alone are not evident, while this option represents higher costs for the reputable manufacturers who are complying with the TSD. PO2a and 2c would be particularly burdensome for SMEs.

Policy options face greater opposition from the industry as the level of intervention increases, while receiving greater support from consumers.

Stakeholder support

Legend: +- no support; + limited support; ++ support; +++ strong support

In particular, industry stakeholders supported that new limit values could be added in the TSD for all toys, but voiced strong opposition in particular to removing derogations to generic bans. The main concern was that removing derogations completely would have strong consequences by requiring the removal from the market of a significant number of toys (for example electric toys). Industry was also opposed to extending third party conformity assessment in PO2a, but supported the digitalisation of compliance information. Member States expressed clear support for revising the TSD and strengthening the chemical requirements, both with specific limit values and with additional generic prohibitions for certain substances. There was also support to digitalising product information as well as for extending third party conformity assessment but to a lesser extent. Consumers favoured the options with stricter chemical requirements for products for children and the more limited derogations.

7.3.Summary of policy options assessment

The table below summarises the assessment presented so far, providing an overview of the effectiveness, efficiency and coherence with the EU law for each of the policy option analysed.

Summary table

Legend: +- no / neutral impact; + minor positive impact; ++ positive impact; +++ significant positive impact;

For the first set of policy options, PO1b and PO1c will more effective in strengthening the protection of children (SO1), as they will introduce generic bans to the most harmful chemicals. While all three PO1 will allow for adaptations to new scientific knowledge (SO2), PO1b and PO1c will be more effective as substances will be banned in toys immediately after they are classified under the most harmful hazard classes in the CLP Regulation, while in PO1a a specific assessment for each substance will be required. In achieving both objectives, PO1b will be the most efficient. PO1a will require resources to analyse each substance that may pose a risk in toys, and this has proven in the past to be a laborious process. By not allowing for derogations, PO1c will be less efficient as it will achieve the same results as PO1b with higher costs for industry.

For the second set of policy options, PO 2a will be the least effective and efficient in achieving the related SO: it will lead to limited improvements in the number of non-compliant toys (SO3) and limited facilitation of enforcement (SO4), but with significant costs for compliant industry. PO2b and 2c will be more effective in achieving the reduction of non-compliant toys in the Union market, including for online sales, as they will prevent that any toy which is presented at customs and does not have a DPP is placed on the Union market and lead to significant efficiency gains for market surveillance authorities. PO2c will be more effective in facilitating enforcement and lead to higher efficiency gains for market surveillance authorities, as it will cumulate the benefits of PO2a and PO2b. However, it is not clear that PO2c would lead to significantly lower levels of non-compliance than option 2b, while having higher costs for industry. PO2b is considered to be the most efficient, as it will achieve the objectives in an effective manner without disproportionate costs on industry.

PO1a and 1b are considered to generally be coherent with wider EU policy and regulatory developments, in terms of future and ongoing regulatory actions following the Chemicals Strategy for Sustainability. PO1b will introduce generic bans for the most harmful hazard classes, coherent with other measures EU chemicals legislation and in particular those considered under the REACH Revision. Option 1c would be the least coherent, though, as it would not allow for derogations to generic bans which are currently possible in all chemicals legislation where generic bans are present. Derogations to generic bans are expected to continue to be possible under EU chemicals legislation in the future.

PO2a is coherent with the New Legislative Framework and other EU product legislation which requires the intervention of a notified body for assessing the compliance of certain products. PO2b is also be coherent with priorities and current trends towards digitalisation by default, the conclusions regarding digitalisation of product information in the Evaluation of the New Legislative Framework, which also covers the TSD as well as the recent proposed ESPR. Considering that the digital passport will contain the documentation that economic operators must possess and be able to present upon request in any case, it will be coherent with the existing relevant obligations under EU product and market surveillance legislation. As PO2b follows the current and future trends towards digitalization, it is scored as more coherent than PO2a. For the same reasons, the coherence of PO2c is considered to be high as it contains the measures included in both PO2a and PO2b.

8.Preferred option

Based on the comparative assessment of policy options above, the preferred combination of policy options consists of PO1b and PO2b. They scored well across a range of criteria (positive social and environmental impacts, with less important economic costs, effective, efficient and coherent). PO2b also responds to the “digital-by-default” principle, by providing for a move towards the digital provision of compliance information.

Impacts of the combination of PO1b and PO2b

Economic impacts – Impacts on economic operators: The economic impacts will correspond to the sum of PO1b in terms of product adaptations and withdrawals and relevant administrative costs (in respect of applications for derogations), as described above, and PO2b in terms of administrative burden and savings for the implementation of the DPP.

The number of substances covered by generic bans under PO1b might increase by about 10-30%. A total of 8.4-12.8% of products additional to the baseline may be impacted under PO1b and for which a derogation may not be possible, with 4.6-7.2% subject to product adaptation efforts and 3.8-5.6% could no longer be made available on the market if no alternatives to the restricted chemicals are available. The estimated impact on 4.6-7.2% of all EU toy products could result in total incremental one-off adjustment costs associated with product redesign and redevelopment of additional EUR 23.5 to 396.66 million. Yearly testing costs are estimated to incrementally increase compared to the baseline by around EUR 7.31-11.70 million. In terms of product withdrawals, this option could affect EUR 249 to 367 million worth of products 185 , which should be mitigated by an appropriate transition period in which manufacturers will be able to assess the viability of existing products and, if needed, shift resources to the production and sale of alternative toy products. Moreover, consumers will in many cases simply purchase an alternative toy product rather than not purchase anything. Administrative burden incurred if derogations are requested could amount to EUR 100.000 to EUR 300.000 per year. Derogations would be requested as a means to limit the adjustment costs that the restrictions on chemical substances could lead to. 

For the introduction of the DPP under PO2b it is estimated that the cost for the EU manufacturers could be around EUR 18 million of one –off costs and EUR 10.5 million per year due to the incidence of initial costs, and decrease after the systems are set up and most initial data entered, there are only expected to be additional costs related to updating and maintenance costs. PO2b could lead to savings of around EUR 2.62 to EUR 3.93 million per year only from moving to the digital provision of compliance information. In addition, savings from no need to prepare for dealing with inspections by market surveillance authorities could range from EUR 13 million up to EUR 20 million per year.

Impacts on public authorities: PO2b will lead to significant efficiency gains for market surveillance authorities (if reinvested in toy inspections, the number of inspections is expected to increase by a maximum of 2,500 to 5,000), that are not expected to be lowered by the increased number of chemical substances banned in toys under PO1b. This is because market surveillance authorities do not necessarily test all toys inspected, nor for all substances concerned. In many cases they conduct documentary checks only, in other cases they identify the toys that would need to be subject to testing, as well as the chemical substances are most relevant for the specific toy at stake.

Impacts on innovation, competitiveness and single market: The impacts on innovation and the single market for both 1b and 2b will cumulate. However, PO1b and 2b will have synergistic impacts in improving competitiveness. While PO1b will impose costs on industry for the compliance with new requirements on chemical substances, it will be accompanied by effective measures under 2b to significantly reduce unfair competition from non-compliant toys. This will help preserving the competitiveness of compliant industry. Without PO2b, PO1b may lead to more rogue traders benefitting from selling non-compliant (and often cheaper) toys.

Social impacts – impacts on human health: Although PO1b and PO2b each address a distinct problem, they will complement each other in terms of protection of human health. The human health benefits estimated in PO1b (EUR 240 million to EUR 1.2 billion per year) will be reinforced by ensuring under PO2b that non-compliance will be significantly reduced.

Environmental impacts : While the protection of the environment is not part of the SO pursued, the preferred combination of options will lead to certain benefits for the environment, due to the reduction of harmful substances in toys as well as the significant reduction of non-compliant toys. It is also consistent with the objectives of the Climate Law 186 . This is detailed in the analysis of each of these options.

Stakeholders’ views: Industry stakeholders supported that new limit values could be added in the TSD for all toys in all policy options, but did not support the extension of generic bans to other harmful substances (PO1b). Industry supported the digitalisation of compliance information in PO2b. Member States expressed clear support for revising the TSD and strengthening the chemical requirements, both with specific limit values and with additional generic prohibitions for certain substances (PO1a and PO1b). There was also support to digitalising product information (PO2b) as well as for extending third party conformity assessment (PO2a) but to a lesser extent. Consumers favoured the options with stricter chemical requirements for products for children (PO1b) and the more limited derogations (PO1c). Consumers also favoured the introduction of the DPP (PO2b) as well as the extension of third party conformity assessment (PO2a and PO2c).

Conclusion on the preferred option: In terms of protection of children from harmful substances, PO1b strikes a more appropriate balance between the negative impacts for industry which should be limited by providing for appropriate derogations to generic bans, and significant reduction on children exposure to these harmful substances. It will also ensure that the toy safety rules can continue to adapt to new scientific knowledge. PO 2b will ensure that toys presented at customs without the declaration of conformity included in the DPP would be automatically prevented from being released for free circulation in the Union market. This, together with the efficiency gains for market surveillance authorities will have a potential to reduce the number of non-compliant toys. In this manner, PO2b better addresses the drivers of the problem.

The preferred option will contribute achieving Sustainable Development Goals: to SDG 3 due to its positive impacts on children health and well-being, to SDG 6 by reducing harmful substances in toys that may become waste, to SDG 9 by providing incentives for businesses to innovate in substitution of harmful chemicals and digitalisation and to SDG 12 by enhancing that toys and children health are protected in a more sustainable manner.

8.1.Legal instrument: converting the Directive into a Regulation

A directive leaves Member States to choose which means they will use to comply with the legislative objectives. As the TSD is a ‘total harmonisation’ directive it does not allow the Member States to impose more restrictive obligations. The existing legal instrument being a Directive means that every amendment to the text (of which there has been 17 to date) must be transposed by all Member States. Therefore, to increase efficiency, the legal text would fit better with a regulation approach rather than with a directive approach 187 . There was a broad consensus across all stakeholder groups on the benefits of this conversion. A large majority of public-consultation respondents (85% of all respondents including 80 % of public authorities, 82% of businesses and business associations and 88% of consumers, consumer organisations and NGOs) expressed support for this change. Finally, PO2b requires that the underlying legal instrument be a Regulation to ensure a harmonised enforcement of the Regulation in the Union and, in particular, at the EU external borders.

With a Regulation the costs of transposing any amendment to the Directive would disappear. The transposition of each amendment to the Directive is estimated to be in the range of EUR 6,174 to EUR 18,522 per Member State. The annual savings of the policy measure for public authorities in the EU would therefore be EUR 435,979 188 . A Regulation will also lead to savings for the industry and benefit the single market, as it will enter into force simultaneously across the EU, as well as any subsequent amendment to it.

8.2.REFIT (simplification and improved efficiency)

The Evaluation assessed the potential for simplification of the TSD and concluded that there was no potential for simplification on the substantial obligations and administrative burden of the TSD, as simplification entailing fewer obligations for economic operators would risk losing protection of children. Similarly, currently under the TSD there is no requirement to go through third-party conformity if there are harmonised standards covering all aspects of toys; this could not be simplified further.

The option to move to digital in the compliance information will lead to simplification and improved efficiency in the contacts between economic operators and market surveillance authorities. This is supported by outcomes from the public consultation for the Evaluation of the New Legislative Framework, 79.2% of the respondents (54.2% to a great extent and 25% to some extent) considered that digitalisation of the declaration of conformity / technical product information / technical file would improve the efficiency of the conformity assessment procedure, without hindering market surveillance activities. Stakeholders from all groups in that consultation agreed that digitalisation offers a potential solution for simplification of administrative obligation related to product information requirements and CE marking, which is also applicable to toys.

Finally, one aspect for simplification that was raised very frequently by stakeholders was the need for the warnings required by the TSD to be preceded by the word “warning” which needed to be translated into all languages required by the Member States in which the toy was going to be made available. Replacing the word “warning” by a generic pictogram would lead to simplification for the industry without compromising the protection of children. It will also lead to savings to the industry when producing the labels but these cannot be quantified with precision.

8.3.Application of the ‘one in, one out’ approach 

As indicated in the previous sections, the first set of policy options is only expected to lead to an increase in the administrative burden as compared with the business as usual costs of the baseline if derogations are requested to continue using in toys substances which have been banned. It could be estimated that the cost per derogation request could range between EUR 50.000 to a maximum of EUR 150.000 per derogation request, and that there would be a maximum of two derogation requests per year, which would lead to a recurrent cost of EUR 100.000 to a maximum of EUR 300.000 per year (with an average of EUR 200.000 per year). Option 2b would entail administrative costs for businesses and benefits. Based on current market structure and expected average production per enterprise, the overall additional administrative burden has been estimated at approximately EUR 18 million one-off and EUR 10.5 million recurrent, per year.

In terms of benefits, option 2b is likely to bring some reduction of the administrative burden on companies and authorities: the introduction of the DPP, while it would lead to initial costs for companies, it has the potential to reduce the administrative burden on public authorities, in particular customs, when importing toys from third countries since the DPP would allow for more automatic documentation checks and prevent the import of non-conform toys that would be held on border premises and subject to physical controls. PO2b could lead to savings for companies from moving to digitalised information of around EUR 2.62 to EUR 3.93 (EUR 3.275 on average) million per year. In addition, a cost reduction would also incur for manufacturers during inspections by market surveillance authorities but these are not savings that are accounted for offseetting under the ‘one in, one out’ apporach.

9.How will actual impacts be monitored and evaluated?

The Commission will monitor the application of the future Toy Safety Regulation and its implementation in the Member States, possibly 5 years after the start of application. It will monitor that when new substances are classified under the most harmful hazard classes under the CLP and covered by generic bans, all the relevant stakeholders are sufficiently informed and testing methods are available in advance. The Commission will also monitor the requests for derogations to these generic bans and their regular evaluation, and any derogation granted should be reassessed every 5 years, as is the case in the current TSD.

The Commission will continue to follow with experts emerging scientific knowledge in the field of chemical substances to determine whether additional or stricter limit values should be introduced in the future Toy Safety Regulation. The Commission will continue to produce guidance documents in cooperation with Experts in the Expert Group on Toy Safety, to ensure that stakeholders have the best possible understanding of the obligations under the Toy Safety Regulation so that compliance is facilitated.

The coordination of notified bodies in the relevant NB group for toys will inform of the types of toys for which an EU type certificate is requested, and risks that may appear in novel toys.

As regards the evolution of the rates of non-compliance and possibly the number of toys stopped at the border, the Commission will monitor the implementation by Member States of their obligation under article 34 of Regulation 2019/1020 on market surveillance to notify under ICSMS information on the inspections they conduct on toys, and on the results of those inspections. First of all, the Commission will monitor the number of inspections carried out by market surveillance authorities, which is expected to grow. The information on the results of investigations should report as far as possible if the toy is available online or not. This information, together with the notifications performed under the RAPEX-Safety Gate on dangerous toys, should allow for a more complete picture on the evolution of non-compliance for toys in the EU, and the percentage of toys inspected which are compliant or not, and the extent of the non-compliance including online sales. It is expected that the number and share of non-compliant toys will be reduced. Furthermore, information on the number of toys which are not released for free circulation because the digital product passport including the declaration of conformity was absent, will also allow for evaluating the success of that measure in preventing that non-compliant toys are placed on the Union market.

ANNEXES

Annex 1: Procedural information

1.Lead DG, Decide Planning/CWP references

Lead DG: DG Internal Market, Industry, Entrepreneurship and SMEs (DG GROW)

Directorate: Directorate F – Ecosystems I: Chemicals, Food, Retail

This impact assessment corresponds to the initiative with the Decide reference PLAN 2021/11623, for the revision of the Toy Safety Directive.

2.Organisation and timing

The inception impact assessment feedback period ran from 5 October to 2 November 2021.

The public consultation period ran from 2 March 2022 to 25 May 2022.

An inter-service steering group was convened and chaired by DG GROW. The following Directorates-General participated: SG, JUST, GROW, CNECT, SANTE, ENV, TAXUD. The ISSG met 8 times. The last meeting on the final draft impact assessment report was held on 23 September 2022.

3.Consultation of the RSB

The RSB was consulted in an informal upstream meeting on 22 February 2022. This impact assessment was submitted to the RSB on 28 September 2022. The meeting with the RSB took place on 26 October 2022.

Following the opinion of the RSB from 28 October 2022, changes were made to the IA in order to reflect the recommendations of the Board. The table below presents an overview of the RSB's recommendations and how these have been addressed.

4.Evidence, sources and quality

The Evaluation of the Toy Safety Directive 189 identified the key areas for the revision. It was supported by a study by an external contractor 190 .

This impact assessment is also supported by a study undertaken by another external contractor 191 , who carried out dozens of interviews, analysed data from public and targeted consultations and complemented this through desk research.

Annex 2: Stakeholder consultation (Synopsis report)

1.Consultation strategy 

The objectives of the consultation were to collect evidence and views from a broad range of stakeholders, on the identified problems and the potential solutions concerning the TSD. The activities included a 12-week dedicated public consultation (PC) concluded in May 2022, a Stakeholder Workshop held on 26 April 2022, discussions with Member States at the Administrative Cooperation formation of the Expert Group on Toy Safety and feedback collected in response of the Commission’s inception impact assessment. Also, as part of the impact assessment study, an external contractor organised interviews with 41 relevant stakeholders and an online targeted consultation for SMEs ran between 7 April 2022 and 15 May 2022. Consulted stakeholders included EU and national consumer associations; industry associations; economic operators; citizens; and national authorities.

Summary table on the numbers and type of activities

2.Inception Impact Assessment (IIA)

The Inception Impact Assessment for this initiative was launched in 2021, with a feedback period running from 5 October to 2 November 2021. There were a total of 34 responses, of which 12 from business associations, 6 from companies/business organisations, 5 from non-governmental organisations, 6 from public authorities, 3 from consumer organisations, 1 from an environmental organisation and 1 from an EU citizen. Responses from industry argued that the TSD is currently sufficiently protective and that there was no need to strengthen its requirements. Conversely, public authorities, consumer and environmental organisations strongly advocated for a revision of the TSD which would ensure that children are protected from the most harmful chemical substances.

3.Stakeholder Workshop

The Stakeholder Workshop on 26 April 2022 brought together approximately 120 representatives of EU-level consumer, business and industry organisations, manufacturers, standardisation and notified bodies, as well as Member State authorities. The objective of the workshop was to present the state of play on the impact assessment work and to discuss ways to adapt the TSD to ensure a higher protection of children from the most harmful chemicals, and to reduce the number of unsafe products in the Union. Consumer groups and public authorities considered that the requirements for chemical substances should be strengthened, while industry and business associations were more sceptical. There was consensus among all participants that it should be possible to add new limit values for toys for any children in the TSD, and not only for children under 36 months or to be put in the mouth, as is the case today. On the other hand, all participants agreed that it was necessary to tackle the high number of non-compliant toys on the Union market, in particular those sold online. While consumer associations and notified bodies were in favour of extending the third party conformity assessment under the TSD, the option to require the EU Declaration of conformity in digital form as part of a Digital Product Passport, to be presented at customs too, gathered the most support.

4.Public Consultation

The consultation was open during 12 weeks between 2 March and 25 May 2022, via the EU Survey online system in 24 EU languages. A total of 17 position papers were attached by stakeholders to the consultation.

4.1.Respondents by Stakeholder category

All in all, 196 responses were submitted, the biggest represented group (34%) was of companies and business organisations, but EU citizens (22%), public authorities (16%) and business associations (12%) were also represented. Albeit in smaller proportion, NGOs, environmental organisations and academic, research institutions also shared their insights on the topic. Out of the 31 responding public authorities 48% is national, 32% is regional, 13% is international and 7% is local.

Figure 1: Respondents by category of their organisation

4.2.Respondents by country of origin

The 196 respondents are coming from 24 different countries, most from EU Member States, but there are respondents from Mexico, Singapore and the United Kingdom too. From Germany, an outstanding number of answers were received (59), but Spain (21), France (19) and Belgium (19) can also be considered as well represented.

4.3.Setting stricter requirements for chemical substances

The participants of the consultation were asked whether they agree or disagree that the EU rules on toy safety should set stricter requirements for chemicals in toys. The sample shows a support for stricter rules with 43% strongly agreeing and an additional 16% agreeing with setting stricter rules. The other end of the scale appears to be much less committed, as those who strongly disagree only amount to 9%.

Figure 2: Ratio of respondents supporting the stricter EU rules on toy safety

If the type of organisation is also taken into consideration when analysing this question, it becomes evident that those who (strongly) disagree are representatives of business associations and companies (47%), while all the environmental and consumer organisations agree to some degree with the statement (over 90%). Similarly, 90 % of public authorities agreed or strongly agreed with the statement.

ØSpecific chemical substances and regulatory approach

The respondents were asked how the TSD should handle the different substances that pose health risks. The answers show that the option of the TSD not regulating the harmful chemicals presented was preferred to a much lesser degree. From the sample, it seems that the respondents were most worried about the substances that are endocrine disruptors or immune system disruptors and they would like to see these substances banned. Over 85% of consumer groups and environmental associations would like to see them banned preventively. Also over 85% of public authorities would like to see these substances addressed by the TSD, but the specific approach to address them (specific risk assessment or generic risk assessment) is balanced. Over 70 % of industry believed that they should only be banned after they have been scientifically assessed as unsafe in toys.

The respondents also had the possibility to give further comments on the subject. Several of them underlined the importance for the revised Toy Safety Directive to ban certain categories of chemicals (rather than individual substances), in line with the "Chemicals Strategy for Sustainability - Towards a Toxic-Free Environment" of the Commission published in October 2020, which foresees to use generic bans ("generic approach to risk management") not just for endocrine disruptors but also for CMRs, substances of very high concern (SVHCs) or sensitizing substances as default approach.

Industry argued that harmful substances could be regulated through REACH restrictions and that the form or specific route in which the substance has been classified (for example, causing cancer by inhalation) should be taken into account.

Figure 3: Ratio of the preferred action for the substances posing different health risks

ØUse of derogations

The participants were asked whether they agree or disagree that the Toy Safety Directive should, by way of exception, allow the presence of chemicals which are subject to current and new general bans. While a number of respondent think that there should be no derogations to the general bans (35% public authorities and 50% of consumer and environmental groups), 73% of industry respondents disagreed with this statement.

The use of derogations is perceived differently by the participants of the PC depending on the reason behind the given derogation. The biggest support is visible in those cases when the used chemicals are found to be safe for human health for that particular use in toys and there are no alternatives or when the used chemicals are found to be safe for human health (as evaluated by a scientific committee) for that particular use in toys (91% of industry, 68% public authorities and 39% of consumer and environmental groups).

The participants of the PC were asked about their opinion on the likely overall impact of introducing general bans for the most harmful chemicals with some limited derogations if necessary. The answers were collected in 5 different categories (in addition to “no opinion” or “I don’t know” answers), where the higher numbers represent an expected very positive impact and a lower number means very negative impact.

From the diagram below, it is visible that the respondents expect that the general bans will have a different impact on different aspects. While industry respondents indicate that it will affect very negatively (42%) or negatively (26% answers) the costs of companies which have to adapt to the new requirements and it will also have a predominantly negative impact (in the same percentages) on the administrative burden for businesses, it is expected to have a very good impact on the protection of children (84% of public authorities, 80% consumer and environmental groups, and 30% of industry).

Out of the different aspects, most respondents thought that the most positive impact will be on the protection of children and on the appearance of innovative products on the market. Positive impact is expected mainly on free movement of toys within the EU single market and on consumer demand. At the same time, industry and in particular SMEs were concerned about costs, administrative burdens and loss of competitiveness for SMEs.

Figure 4: The ratio of the respondents expecting different impacts for the different aspects

When prompted to explain their opinions, industry feared that the result will be more non-compliant toys in the market and less compliant toys offering for consumers.

ØRequirements for toys intended for children under 36 months or to be put in the mouth

Firstly, stakeholders were asked whether the toy safety rules should continue to allow different requirements to be set for chemicals in toys for younger children (under 3 years) compared to older children. Almost half of the respondents (90% industry, 50% public authorities and 48% of consumer/environmental groups) (strongly) agree with the statement. The same trends in the responses per different group appear when asked if the TSD should continue to allow for different requirements to be set for chemicals in toys intended to be put in the mouth.

When asked whether the toy safety rules should allow new requirements to be set for chemicals in any toy should new scientific knowledge emerge, 92% of consumers/environmental groups, 94% of public authorities and 68% of industry agreed or strongly agreed.

ØRequirements on nitrosamines

32% of industry, 77% of public authorities and 73% of consumer/environmental groups agree to some degree that limit values in the Directive for nitrosamines and nitrosatable substances should be lowered.

ØLabelling of chemicals in toys

When asked whether the toys should be labelled with their chemical composition or not, the answers are painting a quite divided picture where 53% thought it would be a good idea and 39% does not support it. Companies (50 answers) and business associations (18 answers) are those who predominantly oppose labelling the products. The vast majority of answering EU citizens (86%) however would like to see the chemical composition on the label.

4.4.Protecting children from other risks in toys

Consumer and environmental organisation were supporting that the TSD addresses privacy breaches (88%), cyber security (90%) and protection from psychological harm (83%). Public authorities were slightly less supportive of the TSD handling these aspects (54% privacy breaches, 58% cybersecurity and 54% for protection from psychological harm). Least of support came from industry respondents (28% for privacy, 27% for cybersecurity and 29% for protection from psychological harm).

Among the additional comments, it was mentioned by industry that as children usually have access to "other" (non-toy) digital equipment like mobile phones, tablet computers, video consoles etc. which potentially pose comparable risks for children. Therefore, cybersecurity and data protection need to be regulated horizontally and not be treated differently than other products used by children and already covered by GDPR, GPSR, RED, Cyber Resilience Act, AI etc (52% indicated this as their preferred option). In contrast, 51% of public authorities and 52% of consumer/environmental groups chose as their preferred option to ensure that the TSD would protect not only the physical but also the psychological wellbeing of children.

4.5.Single Market

ØFactors hampering the application of the TSD

The participants of the public consultation were also asked to what extent five different issues hamper the application of the directive. The most problematic issue identified by all different respondent groups was the lack of specific requirements for online sales (70% consumers/environmental groups, 80% of public authorities and 67% of industry). Industry considered to a much smaller extent that the need to transpose the adaptations of the Directive into national law could hamper the effectiveness of the Directive (37% to a (very) large extent and 27% to a moderate extent, or chemical values being set in different pieces of legislation (31% to a (very) large extent and 24% to a moderate extent). Public authorities and consumer/environmental groups rather referred to other factors such as the fact that conformity assessment is carried out without the intervention of a NB (58% of authorities and 70% of consumer/environmental groups considered this affected the effectiveness of the TSD to a (very) large or to a moderate (20% for authorities) extent). Similarly, the fact that documentation is provided only upon request affected to a (very) large extent for 58% of public authorities and 65% of consumer/environment groups and to a moderate extent for an additional 19% of authorities and 13 % of consumer/environmental groups.

ØThird-party conformity assessment

When asked whether the participants think the toy safety rules should extend the obligation of third-party conformity assessment to more toys (EU-type examination) the results are quite balanced depending on the interest group. 83% of consumer/environment groups and 75% of authorities support it while 75% of industry is against. Out of those who were in favour of extending this obligation, 44% of consumer/environment associations and 32% of authorities thought it should apply to all toys, while close to 40% of authorities and consumer/environment groups supported that it covers chemical mixtures and an additional 25% of public authorities and 37% of consumer/environment groups believed it should cover toys intended for children under 36 months.

In terms of impacts of the measures, consumer groups rather believed that it would have a very positive impact in the protection of children (63%) and in compliance of toys (50%). Similarly, public authorities considered it could improve the protection of children (54%) and the compliance of toys to a high or moderate extent (80%). Conversely, industry was concerned about the negative impacts on their costs and administrative burden (over 60%) and on price of toys (52%). Negative impacts on the competitiveness of SMEs was also raised as a strong concern (55%). In general, industry did not believe this option would have a strong positive impact on the protection of children or on compliance.

ØDigital tools

The participants to the open consultation seem to agree that it would be a mistake to have certain information available only digitally in the case of name and address of the manufacturer, instruction for use, safety information and information on chemical substances; there are no significant differences in the different responding groups. Industry supports that the EU declaration of conformity (56%) and EU-type examination certificates (54%) are enough if published digitally. For the EU declaration of conformity, public authorities agree that it can be provided digitally (35%) or which basic information on paper (61%). Consumer groups rather support that this documents are either made available on paper (EU declaration of conformity 44% and EU type certificates 31%) or with basic information on paper and more details provided digitally (EU DoC 47% and EU type certificates 49%).

Public authorities and consumer/environmental groups considered that this measure would have the most important impacts in terms of protection of children, compliance of toys with the TSD and efficiency of market surveillance. Similarly, industry respondents also considered that the measure could have benefits for the protection of children (40%) for the compliance of toys (47%) and for the efficiency of market surveillance (59%). However, they were also concerned to a certain extent about possible costs (43%) and administrative burden (45%). SMEs were concerned to a certain extent of the possible negative impacts on their competitiveness (43%).

ØCompliance and enforcement

The respondents were asked whether they would like to see the Toy Safety Directive converted into a Regulation. 85% agrees with the conversion without significant differences across respondent groups (82% industry, 80% of public authorities and 87% of consumer/environment groups.

From the additional comments, it turns out that some of the respondents from industry think that the conversion of the Directive into a Regulation should only be carried out with a view to clarifying a number of provisions in the current framework, and not to increase in any way the burden on manufacturers. It was also pointed out that the same rules must apply everywhere. Either way, a Regulation would allow for stronger enforcement of the provision and guarantee a harmonized implementation of the provisions across the EU, guaranteeing the same level of protection for all children.

When asked about their preferred measures to be included in the Directive to improve compliance and enforcement, the respondents could select multiple replies simultaneously. Public authorities were strongly in favour of a digital passport (96%) of converting the Directive into a Regulation (74%) and of extending third party conformity assessment (67%). Consumer/environmental groups were strongly in favour of converting the Directive into a Regulation and extending third-party conformity assessment (78% for each response) and also to have a digital passport (59%). Industry was also quite in favour of converting the Directive into a Regulation (67%) and of the digital passport (52%) but in a much smaller percentage of extending third party conformity assessment (19%).

5.Targeted Survey for SMEs

5.1.Number and basic characteristics of the respondents

A targeted online survey was carried out to collect as many views from the representatives of SMEs as possible. In total, 201 completed responses were received.

Most of the answers were received from the United States (65), the United Kingdom (21), Germany (15), Spain (14) and Italy (13). Out of the 32 represented countries, 19 were based in one the European Union Member State with 40% of the companies in the sample.

Figure 5: Ratio of the respondents’ States where they are based 192

In terms of the size of the organisations, most of the respondents (73.6%) were categorised as micro-organisations employing 1 to 9 employees. Small enterprises employing 10 to 49 employees and medium ones employing 50 to 249 employees were represented in a more balanced way, having 11.9% and 14.4% respectively.

5.2.Costs related to the revision of the TSD

ØAdministrative burden

In connection with bureaucracy and extent of requirements, the respondents stated that the biggest burdens are rooted in the testing obligations. Also, it is quite burdensome that companies have to continuously update the documents, especially the Declaration of Compliance. Others referred to costs in relation to keeping the technical documentation up to date as all data from purchased parts and raw materials (for each batch), all test reports (from both raw-materials as finished products), all the updating of DOC's and every additional document are required to have a complete and complying dossier. Furthermore, there are specific challenges such as the length of the custom process in certain countries; the strict fire safety testing requirements; or storing the documentation for 10 years.

Another administrative burden stems from the need for specific knowledge, skills and extensive human resources for following the changing requirements.

In terms of staff costs expressed in number of working days per new SKU, for the costs of making available the declaration of conformity on the website 59 companies assigned a greater number of working days than zero, the majority estimating it to be between 1 and 10. To adapt the labels indicating the chemical content of a product 56 estimations greater than 0 were received. The least resource intensive intervention would be to register the link to the DoC on the website, this only receiving 39 votes. The vast majority of respondents estimated that all of the interventions could be carried out in 1-10 working days per new SKU.

Table 1: The number of companies that estimated that each intervention would require the given number of working days

The outsourcing costs and equipment costs were also assessed by the respondents 193 . According to the answers, from the point of view of outsourcing, the costs of an EU type examination before introducing a new product would be the highest (most of the companies estimating that it would cost between 201 and 1 000 EUR per new toy model, while the lowest would be the costs of registering the link to the website in a database (around 30 000 EUR in total). From the point of view of the equipment costs, the picture is quite different as the costs of registering the link to the website in a database would require the largest sum (around 420 000 EUR) in total, together with the costs of replacing / redesigning a new product if the product is found to be not compliant (around 145 000 EUR in total) and the cheapest would be to make available the declaration of conformity on the website (around 13 000 EUR) in total.

Figure 6: The outsourcing and equipment costs according to the respondents

The respondents were also asked to assess the total costs of process in relation to the total expected revenues. In connection with the EU type examination, 16 companies claimed that they would lose anywhere between 11-50% of their expected revenues, with an additional 8 saying they would lose between 51-100% and another 8 claiming that it would seriously question the sustainability of their respective companies. The interventions connected to the websites are considered to cause the least loss from the expected revenues.

ØImpacts of potential policy measures

The first policy measure envisions to introduce new limit values for toys or lowering the limit values following the newest scientific evidence. The responding SMEs were asked to assess this policy measure’s impact on their competitiveness. The responses show that almost half (41,8%) claimed that the competitiveness will decrease as a result and only a mere 8% expected that as a result of this measure the competitiveness of their company will increase to some degree.

The second policy measure is the ban on specific substances. Based on the newest scientific evidence on chemical risks, some substances might be banned from toys altogether or only allowed using derogations. The ban on specific substances would have an adverse effect on their competitiveness. 44,8% considers that their competitiveness would deteriorate. The number of those who are unsure or do not want to share their opinion was relatively high, however, the ratio of those who consider that this measure will help their competitiveness is quite low (7,5 and 4% respectively). Each tightening of chemical requirements causes a significant increase in costs (tests, searching for materials, documentation etc.) and this is why they feel that they are losing competitiveness with companies that do not comply with these requirements, if enforcement of the TSD is not improved.

Another proposed policy option is to extend third-party conformity assessment for certain toys. Only 7,5% of respondents considers that their competitiveness compared to imports outside of the EU will increase and 5,5% thinks similarly optimistically in connection with large manufacturers, in particular if nothing is improved in relation to the enforcement of the Directive.

The last proposed policy option is the introduction of a digital product passport containing the EU declaration of conformity which can be presented at customs and used by market surveillance authorities could be based on the Digital Product Passport under the ESPR. This option is the most supported, as 10% and 5% of the respondents think that their competitiveness would increase thanks to this. Nonetheless, SME companies fear they will be less competitive compared to larger companies, since the administrative and economic burdens will be bigger for them.

6.Interviews

All in all 41 interviews were carried out with authorities, notified bodies, companies and business associations as well as consumer associations.

Annex 3: Who is affected and how?

1.Practical implications of the initiative

The preferred policy package would lead to policy objectives related to strengthening the protection of children from harmful substances and ensuring that the toy safety rules are flexible to adapt to updated scientific knowledge. Furthermore, it would achieve a significant reduction of non-compliant toys in the EU and facilitate enforcement of the rules. The TSD’s legislative revision would contribute towards the achievement of EU policy objectives in strategic areas for the future of the EU economy such as green and digital economy.

Impacts on businesses

The preferred PO will lead to adaptation costs of product redesign and product adaptations to substitute the harmful chemicals which will be subject to generic bans. There will be product withdrawals of products in which the chemical substances cannot be substituted but this should be mitigated to a certain extent by a shift in demand to other products that will remain on the market. With the preferred PO, businesses will be required to include a Digital Product Passport on their products containing the EU declaration of conformity. The reference of this Digital Product Passport will be included in a central registry and indicated at the time of import into the Union. While this will lead to costs for businesses, it will also lead to benefits in streamlining the provision of compliance information, updating such information and replying to requests from authorities. Furthermore, it will yield significant benefits in terms of competitiveness of reputable companies which now face unfair competition by rogue traders.

Impacts on consumers

The preferred PO will lead to significant health benefits in terms of strengthen protection of children from the most harmful substances. These are substances such as CMRs, endocrine disruptors or substances which are toxic for the immune, neurological and respiratory systems. These substances are particularly harmful for children who are a vulnerable population group, and can produce adverse effects sometimes at very low doses. In addition, a significant reduction of non-compliant toys on the Union market can be expected from relying on the Digital Product Passport, thus reducing also the possible risks that children face from non-compliant toys.

Impacts on public authorities

The preferred PO will lead to significant efficiency gains for public authorities in tackling non-compliant toys in the Union. Non-compliant toys come particularly from non-EU countries. Customs controls of the existence of the Digital Product Passport can be rendered automatic on all toys coming into the Union market, rather than risk-based, which will stop a significant number of non-compliant toys which do not have this Digital Product Passport from being placed on the Union market. In addition, checks by market surveillance authorities will be rendered more efficient by allowing immediate access to the compliance information of the toy via a machine readable code on the toy. This will allow for an increase in the number of inspections between 10 to 20%.

2.Summary of costs and benefits

3.Relevant sustainable development goals

Annex 4: Analytical methods

In order to support the Impact Assessment, a study was carried out using a set of dedicated data collection and analytical tools to maximise relevance of the results and minimise the limitations of the study.

See annex 11 for further details on the methodology for calculation of number of non-compliant toys in the Union, and annex 14 for details on the methodology for the assessment of the different impacts of the policy measures.

1.Overview of the method

The study was carried out between January 2022 and October 2022. The work was structured around three phases, each of them with various sub-tasks. This section summarizes the work under the key evidence-gathering and analysis activities.

Intervention logic and identification of the key problems

To prepare the work for the development of the policy options, the study first established the intervention logic and mapped and assessed the 1) key drivers, 2) problems and sub-problems, and 3) consequences of these problems to consumers, national regulatory authorities, the internal market and businesses. The intervention logic was regularly updated based on evidence gathered from the various data collection tasks.

1)Literature review

The study included a review of relevant documentation and literature. The review covered a broad catalogue from a variety of EU and Member State sources, including legal and policy documents, statistical data, studies and academic papers, position papers and other publications from relevant stakeholders.

2)Development and appraisal of the policy options

Based on the analysis carried out in the legal gap analysis, literature review and analysis of the consumer journey and on the feedback collected from stakeholders, 6 different policy options for the future of the TSD were identified:

Policy option 0: no change (baseline)

Policy option 1a: minimum changes

Policy option 1b: improved protection

Policy option 1c: maximum changes

Policy option 2a: extending the conformity assessment

Policy option 2b: facilitation of control through digitalisation

Policy option 2c: extending the conformity assessment and facilitation of control (combination of options 2a and 2b)

3)Consultation with EU and national-level stakeholders

The stakeholder consultation aimed to collect information and feedback for the assessment of impacts and on the different policy options. The stakeholder activities included:

·Scoping interviews with notified bodies, business associations and consumer associations to gather preliminary information on the topic and assess key issues and challenges;

·SME survey: the survey was shared with SMEs, SME associations and business associations to ensure that specific input from SMEs was gathered, especially on impacts;

·Semi-structured interviews with EU-level and national stakeholders, particularly from the following stakeholder groups: businesses and business associations, national regulatory authorities and notified bodies, and consumer associations. The key aim of the semi-structured interviews was to receive feedback from the stakeholders on potential improvements of the Toy Safety Directive, and insight to feed into the assessment of impacts and comparison of policy options;

·Open public consultation questionnaire: the PC ran for 12 weeks from the end of March 2022 to May 2022 and gathered information and views on how the Toy Safety Directive can better protect children, and to collect information on how to improve the Directive.

Table 2: Summary table on the number and type of consultation activities

2.Limitations encountered and mitigation measures

The key limitation of the study was the shortened timeline of the data collection which made the setup of the data collection tools extremely challenging and left no time for further probing into the evidence and reconnecting with stakeholders after the data collection had been done. Some limitation in the assembled evidence was therefore unavoidable and will be addressed as soon as possible:

·The overall number of third-party conformity assessments both in the baseline option and in option 2a was not available to stakeholders. Further discussions with stakeholders on this specific topic will continue to revise the estimate as appropriate.

·Cost estimates for all of the options under 1 but especially 1a were very uncertain as the details of the chemical substances may only be determined later on. It is not clear which limit values would be introduced or how many toy relevant substances would be affected by an extension of the GRA.

·Information on toy specific health risks were also scarce. On the one hand, information on illnesses or accidents with toys is patchy at best while, on the other hand, there is ample evidence of dangerous substances still present in toys on the EU market. The study could not fully resolve that gap in the data.

To mitigate these deficiencies, the study triangulated the information from the SME survey, the interviews and the literature to develop robust estimates for each of the options. For some topics knowledgeable stakeholders were asked for further input to fill the gaps.

3.Structure of the analysis

The analysis followed a step-by-step approach, where each step built on the outcome of the previous step. The steps mirrored the different elements mapped in the intervention logic, from the drivers to the problems and consequences and then to policy objectives and policy options. In terms of the impact assessment of the options, first a baseline assessment of the toys market was carried out. The policy options were then assessed with regard to their impacts and compared, based on feedback collected from stakeholders through the online survey. The following impacts were assessed:

·Administrative burden

·Cost for companies

·Impacts on competitiveness and the single market

·Costs for authorities and notified bodies

·Human health impacts

·Environmental impacts

4.Legal analysis - Methodology

A legal comparative analysis was conducted to see possible overlaps, duplications and gaps between the TSD and other relevant EU legislation (both legislation in force and proposed). The legal analysis did not intend to repeat the comprehensive legal analysis performed for the 2020 evaluation of the TSD, therefore it was focused on the legislation adopted or proposed since then. Nonetheless, to understand the overall approach to product safety, including toy safety, many of the older legal acts were reviewed as well. All legal acts were reviewed in the latest consolidated version.

The following legislation (in chronological order) was reviewed:

•    Directive 2001/95/EC of the European Parliament and of the Council of 3 December 2001 on general product safety, OJ L 11 of 15.01.2002

•    Regulation (EC) No 1935/2004 of the European Parliament and of the Council of 27 October 2004 on materials and articles intended to come into contact with food and repealing Directives 80/590/EEC and 89/109/EEC

•    Regulation 1907/2006, concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency, amending Directive 1999/45/EC and repealing Council Regulation (EEC) No 793/93 and Commission Regulation (EC) No 1488/94 as well as Council Directive 76/769/EEC and Commission Directives 91/155/EEC, 93/67/EEC, 93/105/EC and 2000/21/EC

•    Regulation (EC) No 765/2008 of the European Parliament and of the Council of 9 July 2008 setting out the requirements for accreditation and market surveillance relating to the marketing of products and repealing Regulation (EEC) No 339/93

•    Decision No 768/2008/EC of the European Parliament and of the Council of 9 July 2008 on a common framework for the marketing of products, and repealing Council Decision 93/465/EEC

•    Regulation (EC) No 1272/2008 on classification, labelling and packaging of substances and mixtures (CLP), amending and repealing Directives 67/548/EEC and 1999/45/EC, and amending Regulation (EC) No 1907/2006Directive 2011/65/EU, on the restriction of the use of certain hazardous substances in electrical and electronic equipment (RoHS)

•    Directive 2009/48/EC of the European Parliament and of the Council of 18 June 2009 on the safety of toys

•    Regulation No 1223/2009 on cosmetic products

•    Directive 2011/65/EU of the European Parliament and of the Council of 8 June 2011 on the restriction of the use of certain hazardous substances in electrical and electronic equipment

•    Directive 2014/53/EU of the European Parliament and of the Council of 16 April 2014 on the harmonisation of the laws of the Member States relating to the making available on the market of radio equipment and repealing Directive 1999/5/EC

•    Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC

•    Regulation (EU) 2019/881 of the European Parliament and of the Council of 17 April 2019 on ENISA (the European Union Agency for Cybersecurity) and on information and communications technology cybersecurity certification and repealing Regulation (EU) No 526/2013

•    Regulation (EU) 2019/1020 of the European Parliament and of the Council of 20 June 2019 on market surveillance and compliance of products and amending Directive 2004/42/EC and Regulations (EC) No 765/2008 and (EU) No 305/2011

The following proposals for legal acts (in chronological order) were reviewed:

•    Proposal for a Regulation of the European Parliament and of the Council on a Single Market for Digital Services (Digital Services Act) and amending Directive 2000/31/EC, COM(2020) 825

•    Proposal for a Regulation of the European Parliament and of the Council laying down harmonised rules on artificial intelligence (Artificial Intelligence Act) and amending certain union legislative acts, COM(2021) 206

•    Proposal for a Regulation of the European Parliament and of the Council on general product safety, amending Regulation (EU) No 1025/2012 of the European Parliament and of the Council, and repealing Council Directive 87/357/EEC and Directive 2001/95/EC of the European Parliament and of the Council, COM(2021) 346

•    Commission Delegated Regulation (EU) …/… of 29.10.2021 supplementing Directive 2014/53/EU of the European Parliament and of the Council with regard to the application of the essential requirements referred to in Article 3(3), points (d), (e) and (f), of that directive, C(2021) 7672

•    Proposal for a Regulation of the European Parliament and of the Council establishing a framework for setting ecodesign requirements for sustainable products and repealing Directive 2009/125/EC, COM(2022) 142

To put legal developments in context, several policy documents were analysed:

•    Chemicals Strategy for Sustainability: Towards a Toxic-Free Environment, COM(2020) 667

•    New Consumer Agenda, COM(2020) 696

•    Cyber Resilience Act, Call for evidence for an impact assessment (2021)

•    European Parliament resolution of 16 February 2022 on the implementation of Directive 2009/48/EC of the European Parliament and of the Council on the safety of toys (Toy Safety Directive), 2021/2040(INI)

The analysis of the legal documents was supported by the literature review of the relevant scholarship and other studies. All publications are referenced in the text of the report.

5.Global and European toys market

Descriptive statistics on the European toys industry draw mainly on Eurostat Structural Business Statistics (SBS), C32.4 – Manufacture of games and toys (NACE Rev 2). The analysis focuses on data covering the EU-27 and EFTA countries, while highlighting the respective shares of leading countries in the European toys industry. Detailed data tables presenting disaggregation for all Member States and EEA / EFTA countries where Eurostat SBS data is available are provided below.

It is important to highlight data gaps and shortcomings in the quality of data. The statistical analysis was partly hampered by data gaps in Eurostat SBS data in some Member States and EEA / EFTA countries, especially for some variables, such as turnover and the number of people employed. In some cases, there were insufficient data for Eurostat to estimate the total for the EU-27, while those estimates that are available are marked as low reliability by Eurostat. The most recent year for which data was available was 2020 for some variables, but there were provisional estimates only and remained gaps.

Regarding mitigation measures, the methodology focuses on reporting the sum of the available data across the EU-27 and EEA/EFTA countries, rather than the overarching Eurostat EU-27 estimates. This allows increased traceability and analysis between the overarching figures and the country data. To ensure this is transparent, it is clearly stated in each case the number of countries whose data is included in the estimate. Furthermore, where appropriate, data imputations have been calculated for certain missing data points based on average market changes year on year. More detail on the data gaps and the approach to mitigating these challenges is provided throughout the analysis of the different elements.

Supporting data tables with full disaggregation

This section contains supporting data with full disaggregation by country and year. It should be noted that in Eurostat SBS, the majority of 2020 data is provisional only – marked by a (p) below. Furthermore, the data does not cover all countries across all years and all variables; for instance, no data on the number of toy industry enterprises is provided in Ireland, Malta and Liechtenstein for the entire period 2016-2020, while no data on this indicator exists for Cyprus in 2016-2017. To mitigate this challenge, data imputations have been developed for certain countries and years, but only where they can be based on robust assumptions. The methodology for developing the imputations is detailed in the box below.

Each table first presents, where available, the Eurostat composite figures for the EU-27 and former EU-28. However, in most cases, these figures are marked by Eurostat as low reliability due primarily to the use of different definitions in some countries. As such, each table also provides the sum of the available data for the EU-27 and the EU-27 + EEA/EFTA countries. However, as data is missing for certain countries and years, these overarching figures are accompanied by notes detailing the number of countries for which data is available.

In the main body of the analysis, the combined EU-27 + EEA/EFTA figures for each year are presented; this is due to the limited reliability of the composite EU-27 figures and the increased traceability and comparability provided by the EU-27 + EEA/EFTA figures in combination with the country-specific data.

Table 3: No. of enterprises in the European toys industry (disaggregated among EU-27 and EEA/ EFTA countries)

Source: Eurostat SBS.

* Data imputations developed for the following data points and years: IT (2020), IS (2020), and CH (2019-2020).

Table 4: Turnover generated in the European toys industry in million EUR (disaggregated among EU-27 and EEA/ EFTA countries)

Source: Eurostat SBS.

Table 5: Production value in billion EUR

Source: Eurostat SBS.

Table 6: Number of persons employed among toy producers