EU Directive on Cybersecurity agreed
Filip Van Elsen
Catherine Di Lorenzo
06 January 2016
On 18 December 2015, the agreed text of the Network and Information Security Directive (the NIS Directive) was released. With cybersecurity firmly established as a key business risk, the introduction of specific laws in this area across the European Union will have a significant impact.
The NIS Directive is the culmination of a process that began in 2013 when the European Commission adopted its EU cybersecurity strategy and proposed a new draft Directive. The NIS Directive is a key component of the overall strategy to prevent and respond to cyber disruptions and attacks. The Commission recognised that cybersecurity incidents are increasing in frequency and magnitude, and becoming more complex and cross–border in nature. As such incidents can cause major damage to safety and the economy, the Commission considered that efforts to prevent, co-operate on and be more transparent about cyber incidents should improve.
The aim of the NIS Directive is therefore to ensure a high common level of network and information systems (or cyber) security across the EU by requiring Member States to increase their preparedness and improve co-operation with each other, and by requiring operators that provide essential services in relation to certain infrastructure and providers of particular digital services to adopt appropriate steps to manage security risks and to report serious incidents to national competent authorities. This is seen by the Commission as vital to ensuring a secure and trustworthy digital environment throughout the EU.
What began as a broad reform proposal by the Commission has been watered down to some extent by the European Parliament and Council, which led to extensive Trilogue discussions between those institutions. This delayed agreement of the NIS Directive, and has led to compromise.
There was considerable disagreement regarding the Commission’s inclusion of “internet enablers” or “digital service platforms” within the scope of the NIS Directive in its initial proposal. This inclusion was rejected by the European Parliament as well as a considerable number of Member States. The ultimate inclusion of such entities as cloud platforms, online marketplaces and online search engines is therefore significant.
While Directives always lead to an element of national differences in implementation, the NIS Directive is particularly prone to inconsistencies in application, as much of it sets out actions to be taken by the individual Member States, leaving those Member States to determine the detail. In particular, it remains to be seen how Member States will approach the co-operation requirements under the NIS Directive to create co-ordinated responses to incidents, as Member States currently have different approaches.
In this alert, we set out some of the key features of the NIS Directive and how they might affect businesses.
Requirements for Member States
Establishment of national frameworks
Under the NIS Directive, Member States must ensure they have in place a minimum level of national capabilities by establishing or designating, and adequately resourcing:
•a Network and Information Security strategy that defines the strategic objectives and appropriate policy and regulatory measures with a view to achieving and maintaining a high level of security of networks and information systems;
•one or more national competent authorities to monitor the application of the NIS Directive in their territory and to contribute to its consistent application throughout the EU – Member States may designate this role to an existing authority or authorities;
•a national single point of contact on security of networks and information systems to liaise between its Member State and the co-operation group and Computer Security Incident Response Teams (CSIRTs) network – Member States may designate this role to an existing authority or authorities; and
•one or more CSIRTS, which are responsible for handling risks and incidents.
Co-operation between Member States and their CSIRTs
The NIS Directive also requires enhanced co-operation between Member States by establishing:
•a co-operation group composed of representatives from the Member States, the Commission and the European Network and Information Security Agency (ENISA), whose function is broadly to circulate and exchange information amongst its members and co-operate to counter cybersecurity threats and incidents; and
•a network of the national CSIRTs, composed of representatives of Member States’ CSIRTs and CERT–EU (the Computer Emergency Response Team for EU institutions, bodies and agencies) with the Commission participating as an observer, to promote swift and effective operational co-operation by, amongst other things, exchange of information and supporting Member States to address cross-border incidents on a voluntary basis.
Requirements for operators of essential services
Who is an “operator of essential services”?
The NIS Directive imposes obligations on both public and private “operators of essential services”. These services are of the type referred to in Annex II (see below) in the fields of energy, transport, banking, financial market infrastructures, health, drinking water supply and distribution, and digital infrastructure, and which fulfil all the following criteria:
•the entity provides a service which is essential for the maintenance of critical societal and/or economic activities;
•the provision of that service depends on network and information systems; and
•an incident affecting the network and information systems of that service would have significant disruptive effects on its provision.
Each Member State will identify those entities which meet the definition of “operator of essential services” with an establishment, meaning the effective and real exercise of activity through stable arrangements (branches and subsidiaries will be included within this definition), on their territory according to the above criteria. This may be achieved by adopting a list enumerating all operators of essential services or adopting objective quantifiable criteria (e.g. output of the operator of essential services or number of users) allowing determination of the entities to which the NIS Directive will apply and those to which it will not. When determining the significance of a potential disruptive effect, a Member State will take into account sectoral factors and at least the following cross-sectoral factors:
•the number of users relying on the service provided by the entity (i.e. its establishment in the territory);
•the dependency of other sectors referred to in Annex II on the service provided by the entity;
•the impact that incidents could have, in terms of degree and duration, on economic and societal activities or public safety;
•the market share of the entity;
•the geographic spread with regard to the area that could be affected by an incident; and
•the importance of the entity for maintaining a sufficient level of the service, taking into account the availability of alternatives for the provision of that service.
Types of entities that might be “operators of essential services”
Annex II of the NIS Directive lists or cross-references the types of public and private entities in the designated sectors that may be “operators of essential services”. These include the following:
•digital infrastructure – internet exchange points, top level domain name registries, and domain name system service providers;
•energy – electricity/gas suppliers, distribution system operators, transmission system operators, storage system operators, LNG operators, and operators of oil and natural gas production, refining and treatment facilities;
•transport – air and maritime carriers, traffic management control operators, airports, railways, road traffic management control and intelligent transport system operators;
•banking – credit institutions in accordance with the Capital Requirements Regulation (575/2013);
•financial market infrastructure – stock exchanges and central counterparties;
•health – healthcare providers (including hospitals and private clinics); and
•drinking water – entities in the drinking water supply and distribution sector.
Security and incident reporting
If an operator of essential services falls within the scope of the NIS Directive and is established in a Member State, it must:
i.take appropriate and proportionate technical and organisational risk management measures, including measures to prevent and minimise the impact of incidents that affect security of the networks and information systems used with a view to ensuring continuity of those services; and
ii.comply with a reporting scheme, to be established by the Member State in question, under which it must notify “without undue delay” to the competent authority or CSIRT “incidents having a significant impact on the continuity of the essential services they provide”.
“Incidents” are those events that have an actual adverse effect on the security of networks and information systems. The NIS Directive establishes parameters to be taken into account when assessing the “significance” of the impact of any incident:
•number of users affected by the disruption of the essential services;
•duration of the incident; and
•geographical spread with regard to the area affected by the incident.
It should be noted that the model on which the NIS Directive is based is the EU’s existing Framework Directive for electronic communications (2002/21/EC), which currently requires telecoms companies to adopt risk management steps and to notify regulators of major breaches of their network security. Telecoms companies are therefore excluded from the scope of the NIS Directive.
Where an operator of essential services relies on a third party digital service provider (see below) for the provision of its services, the obligation to notify regarding an incident having a significant impact on the continuity of the essential service shall be on the operator rather than on the digital service provider.
Enforcement and sanctions
Competent authorities may require operators of essential services to:
i.provide information needed to assess the security of their networks and information systems, including documented security policies; andii.provide evidence of effective implementation of security policies, such as the results of a security audit carried out by the competent authority or a qualified auditor and, in the latter case, make the results thereof, including the underlying evidence, available to the competent authority.
Following the assessment of either the information provided by an operator of essential services or the results of a security audit, the competent authority may issue binding instructions to the operator of essential services to remedy its operations.
Requirements for digital service providers
What is a “digital service”?
The NIS Directive imposes obligations on providers of “digital services”. These are an Information Society service (as defined by Article 1(b) of Directive 2015/1535) of the types described below.
Annex III of the NIS Directive (together with the corresponding recitals) lists the types of digital services which fall within the scope of the Directive. These include the following:
•online marketplace – the Directive covers services allowing online consumers and/or traders to conclude online sales and service contracts. Online services which compare the price of particular products or services from different traders, and then redirect the user to the preferred trader to purchase the product are outside of the scope of the Directive;
•online search engine – services which allow the user to perform searches of, in principle, all websites or all websites in a particular language on the basis of a query are covered by the Directive. The scope of the Directive does not extend to the provision of search functions that are limited to the content of a specific website; and
•cloud computing service – there are a number of different models for the provision of cloud computing services. The Directive is intended to apply to services that enable access to a scalable and elastic pool of shareable computer resources. This means cloud computing services which can respond to an increase or decrease in demand for resources or processing power from multiple users accessing the service in different geographical locations, but where the processing is carried out separately for each user, although the service is provided from the same electronic equipment.
Territorial scope and regulatory responsibility
Any legal person providing a “digital service” of a type identified above, a “digital service provider”, will be subject to the relevant obligations of the NIS Directive if they offer services within any Member State.
Like the General Data Protection Regulation and its concept of the “One Stop Shop”, the digital service provider will be deemed to be under the jurisdiction of the Member State where it has its main establishment (i.e. the Member State of its head office in the EU). If the digital service provider is established outside of the EU but offers services within a Member State, it shall designate a “representative” established in that Member State and shall be under the jurisdiction of the Member State where the representative is established. There is a form of co-operation mechanism requiring other interested Member States (e.g. where the networks and information systems are located) to assist.
It will be interesting to see whether entities may choose to forum shop depending on the approach taken by different Member States, and whether they may find they have “main establishments” in different Member States for different regulatory obligations.
Security and incident reporting
If a digital service provider falls within the scope of the NIS Directive, it must:
i.take appropriate and proportionate technical and organisational risk management measures to manage risks. The measures should ensure an appropriate level of security taking into consideration: ◦security of systems and facilities;
◦business continuity management;
◦monitoring, auditing and testing; and
◦compliance with international standards.
ii.take measures with a view to ensuring the continuity of the services by preventing and minimising the impact of incidents that affect the security of the networks and information systems used; and
iii.comply with a reporting scheme, to be established by the Member State in question, under which it must notify “without undue delay” to the competent authority or CSIRT any “incident” that has a “substantial impact” on the provision of a digital service.
The NIS Directive establishes parameters to be taken into account when assessing the impact of any incident:
•the number of users affected by the incident, in particular users relying on the service for the provision of their own services;
•the duration of the incident;
•the geographical spread with regard to the area affected by the incident;
•the extent of the disruption of the functioning of the service; and
•the extent of the impact on economic and societal activities.
Enforcement and sanctions
Competent authorities may require digital service providers to:
i.provide information needed to assess the security of their networks and information systems, including documented security policies; and
ii.remedy any failure to fulfil the relevant requirements laid down in the Directive.
Digital service providers are to be subject only to reactive ex-post supervisory activities by the competent authorities, which should take action only when they have received evidence that a digital service provider is not complying with the requirements of the Directive. Such evidence may be provided by the digital service provider itself, by a competent authority, including a competent authority of another Member State, or by a user of the service. A digital service provider, unlike an operator of essential services, is under no obligation to provide evidence to the relevant competent authority that it is complying with the requirements of the Directive. It should be noted that the Commission does not intend for the Directive to create a general obligation for a competent authority to supervise digital service providers.
It is suggested by the Commission in the recitals that the security requirements for digital service providers should be lighter than those applied to operators of essential services as the degree of risk to the security of the service provided will be higher for entities in the latter classification. The obligations on digital service providers under the NIS Directive will not be applicable to micro- and small enterprises as defined in Commission Recommendation 2003/361/EC.
Cross-border information sharing
Any notification of an incident by an operator of essential services/digital service provider should also include information to enable the competent authority (or CSIRT) to determine any cross-border impact of the incident. Based on this information, the competent authority (or CSIRT) shall inform other affected Member States if the incident has a significant impact. The notifying party’s security and commercial interests, and the confidentiality of any information, will be preserved if any such disclosure is made.
Disclosure to the public
After consulting with the operator of essential services/digital service provider concerned, the notified competent authority or CSIRT may inform the public, about individual incidents where public awareness is necessary either to prevent an incident, to deal with an on-going incident. With respect to notifications by digital service providers only, the notified competent authority or CSIRT has an additional discretion, after consultation with the notifying party, to inform the public where such a disclosure is otherwise in the public interest.
To promote convergent implementation, the NIS Directive requires Member States to encourage the use of European or internationally accepted standards and/or specifications relevant to security of networks and information systems. These standards and/or specifications have not been set out in detail in the NIS Directive. However, it does provide that ENISA may collaborate with Member States to produce recommendations and guidelines regarding the technical areas which should be considered as well as already existing standards, including national standards in Member States. The recitals suggest that it might be necessary to draft harmonised standards to ensure a high level of security at an EU level.
Enforcement and sanctions
Member States are required to put in place “effective, proportionate and dissuasive” sanctions for failure by an operator of essential services/digital service provider to comply with the NIS Directive’s provisions regarding security requirements and incident notification. The security and notification obligations will apply to operators of essential services and digital service providers regardless of whether they perform the maintenance of their networks and information systems internally or outsource it.
It remains to be seen what sanctions regimes Member States will develop to correspond with the NIS Directive’s requirements.
What does the NIS directive mean for businesses?
Consider whether you are caught
Operators in certain fields whose services are “essential” or providers of particular “digital services” should be aware of the compliance obligations imposed on them by the NIS Directive to assess the risks they face and to adopt appropriate and proportionate cybersecurity measures to protect their network and information systems from compromise. Such measures will be required to meet European or internationally accepted standards specified by Member States, which are yet to be detailed.
It is important to realise that, under existing duties of care and as a matter of corporate responsibility, or (in the case of listed companies) stock exchange rules, businesses may already effectively be required to assess these risks and take appropriate measures, regardless of whether they will be caught by the NIS Directive. In any event, it is advisable for companies to implement or refresh their cybersecurity policies, procedures and checklists, possibly after consulting with the competent authority or CSIRT. This may also be required by a company’s insurers if it holds cyber-risks insurance, although we usually advise against accepting an insurance policy under which coverage is excluded in case of non-compliance with an internal policy or procedure.
Notifying security incidents
Operators of essential services/digital service providers should put in place procedures to assess the significance of any network and information security incidents against the criteria specified in the NIS Directive to determine whether a notification to the competent authority or CSIRT may be required and to undertake such notification.
Under the notification regime, an operator of essential services/digital service provider is not required to notify any other parties (such as customers, employees or law enforcement agencies). However, a national competent authority or CSIRT may inform the public, where public awareness is needed either to prevent the incident or resolve an on-going incident, or where this would otherwise be in the public interest. The notifying party will be consulted before a disclosure to the public is made, and its commercial interests and the confidentiality of information it has provided will, in principle, be preserved. Notification will not expose the notifying party to increased liability.
Entities falling outside the scope of the NIS Directive may, if they experience incidents having a significant impact on the services they provide, notify on a voluntary basis the relevant authorities in the Member State in which they are established. This could lead to a broadening by convention, or pursuant to regulatory or industry guidance, of the reporting requirements.
Potential for multiple reporting obligations
Operators of essential services/digital service providers should be aware that multiple reporting of incidents may be required. For example, under the new General Data Protection Regulation, security incidents that might concern a breach of personal data would need to be reported to data protection authorities as well as to national competent authorities or CSIRTs under the NIS Directive. Similarly, financial services regulation in many countries will already require the reporting of incidents compromising the integrity of client data or impacting the continuity of services. This creates a number of potential issues including managing different triggers for reporting of breaches, with different timing expectations.
The recitals recognise this potential for administrative burden where a security incident also involves a personal data breach and suggest that ENISA could co-operate with personal data protection authorities and assist in the production of guidelines to facilitate the reporting of incidents compromising personal data.
Further national requirements
As it is a minimum harmonisation directive, individual Member States cannot reduce the level of requirements set out in the NIS Directive. However, with reference to operators of essential services only, Member States may legislate beyond the NIS Directive and set higher standards within national legislation for network and information security. This may create differences in implementation of the NIS Directive between the Member States.
Member States will have up to 21 months from the publication of the NIS Directive (expected in the spring) to publish national laws implementing the NIS Directive and to bring them into force.
Best practice on incident reporting
As initiatives such as the NIS Directive make clear, businesses are increasingly expected to report security incidents, whether to sector regulators, to customers and other affected counterparties or to the general public. Good “house-keeping” for breach notification preparedness includes:
•implementing a company-wide NIS policy to ensure on-going IT and information security;
•proactively identifying vulnerable areas in IT networks/systems;
•preparing an NIS incident response plan, including co-ordination, communications, forensics/investigation, reporting and – last but not least – recovery plans;
•setting up an independent response team (identify members and back-up members and explain their respective roles, responsibilities and decision-making authorities);
•ensuring that suppliers (and their subcontractors) implement security measures and provide evidence to you on a regular basis that these measures are appropriate, effective and reflect technological developments; and
implementing training and awareness programmes to ensure that relevant employees and suppliers are aware of the NIS response plan, and will have the necessary resources to comply with it.