Report on the 2nd Privacy Round Table meeting

This research was undertaken as part of the 'SAFEGUARDS' project that CIS is undertaking with Privacy International and IDRC

In furtherance of Internet Governance multi-stakeholder Initiatives and Dialogue in 2013, the Centre for Internet and Society (CIS) in collaboration with the Federation of Indian Chambers of Commerce and Industry (FICCI), and the Data Security Council of India (DSCI), is holding a series of six multi-stakeholder round table meetings on “privacy” from April 2013 to August 2013. The CIS is undertaking this initiative as part of their work with Privacy International UK on the SAFEGUARD project.

In 2012, the CIS and DSCI were members of the Justice AP Shah Committee which created the “Report of Groups of Experts on Privacy”. The CIS has recently drafted a Privacy (Protection) Bill 2013, with the objective of contributing to privacy legislation in India. The CIS has also volunteered to champion the session/workshops on “privacy” in the meeting on Internet Governance proposed for October 2013.

At the roundtables the Report of the Group of Experts on Privacy, DSCI´s paper on “Strengthening Privacy Protection through Co-regulation” and the text of the Privacy (Protection) Bill 2013 will be discussed. The discussions and recommendations from the six round table meetings will be presented at the Internet Governance meeting in October 2013.

The dates of the six Privacy Round Table meetings are enlisted below:

1. New Delhi Roundtable: 13 April 2013
2. Bangalore Roundtable: 20 April 2013
3. Chennai Roundtable: 18 May 2013
4. Mumbai Roundtable: 15 June 2013
5. Kolkata Roundtable: 13 July 2013
6. New Delhi Final Roundtable and National Meeting: 17 August 2013

Following the first Privacy Round Table in Delhi, this report entails an overview of the discussions and recommendations of the second Privacy Round Table meeting in Bangalore, on 20^th April 2013.

Overview of DSCI´s paper on “Strengthening Privacy Protection through Co-regulation”

The meeting began with a brief summary of the first Privacy Round Table meeting which took place in Delhi on 13^th April 2013. Following the summary, the Data Security Council of India (DSCI) presented the paper “Strengthening Privacy Protection through Co-regulation”. In particular, DSCI presented the regulatory framework for data protection under the IT (Amendment) Act 2008, which entails provisions for sensitive personal information, privacy principles and “reasonable security practices”. It was noted that the privacy principles, as set out in the Justice AP Shah Report, refer to: data collection limitation, data quality, purpose specification, use limitation, security safeguards, openness and individual participation. The generic definitions of identified privacy principles refer to: notice, choice and consent, collection limitation, purpose specification, access and correction, disclosure of information, security, openness/transparency and accountability. However, the question which prevailed is what type of regulatory framework should be adopted to incorporate all these privacy principles.

DSCI suggested a co-regulatory framework which would evolve from voluntary self-regulation with legal recognition. The proposed co-regulatory regime could have different types of forms based on the role played by the government and industry in the creation and enforcement of rules. DSCI mentioned that the Justice AP Shah Committee recommends: (1) the establishment of the office of the Privacy Commissioner, both at the central and regional levels, (2) a system of co-regulation, with emphasis on SROs and (3) that SROs would be responsible for appointing an ombudsman to receive and handle complaints.

The discussion points brought forward by DSCI were:

• What role should government and industry respectively play in developing and enforcing a regulatory framework?
• How can the codes of practice developed by industry be enforced in a co-regulatory regime? How will the SRO check the successful implementation of codes of practice? How can the SRO penalize non-compliances?
• How can an organization be incentivized to follow the codes of practice under the SRO?
• What should be the role of SROs in redressal of complaints?
• What should be the business model for SROs?

DSCI further recommended the establishment of “light weight” regulations based on global privacy principles that value economic beliefs of data flow and usage, while guaranteeing privacy to citizens. DSCI also recommended that bureaucratic structures that could hinder business interests be avoided, as well as that the self-regulatory framework of businesses adapts technological advances to the privacy principles. Furthermore, DSCI recommended that self-regulatory bodies are legally recognised.

Discussion on the draft Privacy (Protection) Bill 2013

Discussion of definitions and preamble: Chapter I & II

The second session began with a discussion of definitions used in the Bill. In particular, many participants argued that the term ´personal data´ should be more specific, especially since the vague definition of the term could create a potential for abuse. Other participants asked who the protection of personal data applies to and whether it covers both companies and legal persons. Furthermore, the question of whether the term ´personal data´ entails processed and stored data was raised, as well as whether the same data protection regulations apply to foreign citizens residing in India. A participant argued that the preamble of the Bill should be amended to include the term ´governance´ instead of ´democracy´, as this privacy legislation should be applicable in all cases in India, regardless of the current political regime.

Sensitive Personal Data

The meeting proceeded with a discussion of the term ´sensitive personal data´ and many participants argued that the term should be broadened to include more categories, such as religion, ethic group, race, caste, financial information and others. Although the majority of the participants agreed that the term ´sensitive personal data´ should be redefined, they disagreed in regards to what should be included in the term. In particular, the participants were not able to reach a consensus on whether religion, caste and financial information should be included in the definition of the term ´sensitive personal data´. Other participants argued that passwords should be included within the scope of ´sensitive personal data´, as they can be just as crucial as financial information.

Information vs. Data

During the discussion, a participant argued that there is a subtle difference between the term ´information´ and ´data´ and that this should be pointed out in the Bill to prevent potential abuse. Another participant argued that ´sensitive personal data´ should be restricted to risk factors, which is why unique identifiers, such as passwords, should be included in the definition of the term. Other participants argued that the context of data defines whether it is ´sensitive´ or not, as it may fall in the category of ´national security´ in one instance, but may not in another. Thus, all types of data should be considered within their context, rather than separately. The fact that privacy protection from several financial services already exists was pointed out and the need to exclude pre-existing protections from the Bill was emphasised. In particular, a participant argued that banks are obliged to protect their customers´ financial information either way, which is why it should not be included in the definition of the term ´sensitive personal data´.

Exemptions

Several exemptions to the right to privacy were discussed throughout the meeting. A participant asked whether the right to privacy would also apply to deceased persons and to unborn infants.  Another participant asked whether the term ´persons´ would be restricted to natural persons or if it would also apply to artificial persons. The fact that children should also have privacy rights was discussed in the meeting and in particular, participants questioned whether children´s right to privacy should be exempted in cases when they are being surveilled by their own parents.

Discussion of “Protection of Personal Data”: Chapter III

Following the discussion of definitions used in the Bill, the meeting proceeded with a discussion on the protection of personal data. A participant emphasized that the probability of error in data is real and that this could lead to major human rights violations if not addressed appropriately and in time. The fact that the Bill does not address the element of error within data was pointed out and suggested that it be included in draft Privacy (Protection) Bill. Another participant recommended an amendment to the Bill which would specify the parties, such as the government or companies, which would be eligible to carry out data collection in India. As new services are been included, the end purpose of data collection should be taken into consideration and, in particular, the ´new purposes´ for data collection would have to be specified at every given moment.

Data Collection

In terms of data collection, a participant emphasized that the objectives and purposes are different from an individual and an industry perspective, which should be explicitly considered through the Bill. Furthermore, the participant argued that the fact that multiple purposes for data collection may arise should be taken into consideration and relevant provisions should be incorporated in the in Bill. Another participant argued that the issue of consent for data collection may be problematic, especially since the purpose of data collection may change in the process and while an individual may have given consent to the initial purpose for data collection, he/she may not have given consent to the purposes which evolved throughout the process. Thus, explicitly defining the instances for data collection may not be feasible.

Consent

On the issue of consent, several participants argued that it would be important to distinguish between ´mandatory´ and ´optional´ information, as, although individuals may be forced by the government to hand over certain cases, in other cases they choose to disclose their personal data. Thus participants argued that the Bill should provide different types of privacy protections for these two separate cases. Other participants argued that the term ´consent´ varies depending on its context and that this should too be taken into consideration within the draft Privacy (Protection) Bill. It was also argued that a mechanism capable of gaining individual consent prior to data collection should be developed. However, a participant emphasized upon the fact that, in many cases, it is very difficult to gain individual consent for data collection, especially when individuals cannot read or write. Thus the need to include provisions for uneducated or disabled persons within the Bill was highly emphasized.

Further questions were raised in regards to the withdrawal of consent. Several participants argued that the draft Privacy (Protection) Bill should explicitly determine that all data is destroyed once an individual has withdrawn consent. Participants also argued that consent should also be a prerequisite to the collection, processing, sharing and retention of secondary users´ data, such as the data of individuals affiliated to the individual in question. A participant argued that there are two problematic areas of consent: (1) financial distribution (such as loans) and (2) every financial institution must store data for a minimum of seven to eight years. Having taken these two areas in consideration, the participant questioned whether it is feasible to acquire consent for such cases, especially since the purpose for data retention may change in the process. Participants also referred to extreme cases through which consent may not be acquired prior to the collection, processing, sharing and retention of data, such as in disastrous situations (e.g. earthquake) or in extreme medical cases (e.g. if a patient is in a coma), and suggested that relevant provisions are included in the Bill.

Data Disclosure

In terms of data disclosure, several participants argued that the disclosure of data can potentially be a result of blackmail and that the Bill does not provide any provisions for such extreme cases. Furthermore, participants argued that although consent may be taken from an individual for a specific purpose, such data may be used in the process for multiple other purposes by third parties and that it is very hard to prevent this. It was recommended that the Bill should incorporate provisions to prevent the disclosure of data for purposes other than the ones for which consent was given.

A participant recommended that individuals are informed of the name of the Data Processor prior to the provision of consent for the disclosure of data, which could potentially increase transparency. Many participants raised questions in regards to the protection of data which goes beyond the jurisdiction of a country. It remains unclear how data will be processed, shared, retained when it is not handled within India and several participants argued that this should be encountered within the Bill.

Data Destruction

In terms of data destruction, a participant emphasized upon the fact that the draft Privacy (Protection) Bill lacks provisions for the confirmation of the destruction of data. In particular, although the Bill guarantees the destruction of data in certain cases, it does not provide a mechanism through which individuals can be assured that their data has actually been deleted from databases. Another individual argued that since the purposes for data collection may change within the process, it is hard to determine the cases under which data can be destroyed. Since the purposes for data collection and data retention may change in time, the participant argued that it would be futile to set a specific regulatory framework for data destruction. Another participant emphasized upon the value of data and stated that although some data may appear to have no value today, it may in the future, which is why data should not be destroyed.

Data Processing

In terms of data processing, participants argued that privacy protection complications have arisen in light of the social media. In particular, they argued that social media develop and expand technologically constantly and that it is very difficult to regulate the processing of data that may be conducted by such companies. A participant emphasized the difference between (1) the processing of data when it is being read and (2) the processing of data when it is being analysed. Such a distinction should be considered within the Bill, as well as the use of data which is being processed. Many participants distinguished between the primary and secondary use of data and argued that the secondary use of data should also be included in the privacy statements of companies.

However, participants also pointed out that purposes for the collection of data may overlap and that it may be difficult to distinguish between primary and secondary purposes for data collection. A participant disagreed with this argument and stated that it is possible to distinguish between primary and secondary purposes of data collection, as long as companies are transparent about why they are collecting information and about the purpose of its processing. This argument was seconded by another participant who argued that the specific purposes for the processing of data should be incorporated in the Bill.

In brief, the following questions with regards to chapter III of the bill were raised during the meeting:

• Should consent be required prior to the collection of data?
• Should consent be acquired prior and after the disclosure of data?
• Should the purpose of data collection be the same as the purpose for the disclosure of data?
• Should an executive order or a court order be required to disclose data?
• At the background of national security, anyone´s data can be under the ´suspicion list´. How can the disclosure of data be prevented in such circumstances? Non-criminals may have their data in the ´suspicion list´ and under national security, the government can disclose information; how can their information be protected in such cases?
• An individual may not be informed of the collection, analysis, disclosure and retention of his/her data; how can an individual prevent the breach of his/her data?

• Should companies notify individuals when they share their (individuals´) data with international third parties?

In brief, the following recommendations with regards to chapter III of the bill were raised during the meeting:

• The data subject has to be informed, unless there is a model contract.
• The request for consent should depend on the type of data that is to be disclosed.
• Some exceptions need to be qualified (for example, in instances of medical patients different exceptions may apply).
• The shared data may be considered private data (need of a relevant regulatory framework).
• An international agreement should deal with the sharing of data with international third parties - incorporating such provisions in Indian law would probably be inadequate.
• If any country is not data-secure, there should be an approval mechanism for the transfer of data to such a country.
• India could have an export law which would monitor which data is sensitive and should not be shared with international third parties.
• The problem with disclosure is when there is an exception for certain circumstances
• Records should be kept on individuals who disclose data; there should be a trail of disclosure, so that there can be more transparency and accountability.
• Ownership of data is a controversial issue and so is the disclosure of data; consumers give up the ownership of their data when they share it with third parties and ergo cannot control its disclosure (or non-disclosure).
• ´Data ownership´ should be included in the definitions of the Bill.
• What is the ´quality´ of data? The definition for ´quality´ under section 11 of the Bill is not well defined and should be improved.

Discussion of “Interception of Communications”: Chapter IV

The discussion on the interception of communications started off with a statement that 70 percent of the citizens in India are enrolled on “voice”, which means that the interception of communications affects a large proportion of the population in the country. A participant asked whether the body corporate in India should be treated as a telecommunications provider and whether it should be responsible for the interception of communications. Another participant argued that the disclosure of information should be closely regulated, even when it is being intercepted for judicial purposes. Many participants agreed that data which is collected and intercepted should not be used for other purposes other than the original purpose, as well as that such information should not be shared with third parties.

Questions were raised in regards to who should authorise the interception of communications and a participant recommended that a judicial warrant should be a prerequisite to the interception of communications in India. Some participants argued that the Bill should clearly specify the instances under which communications can be intercepted, as well as the legitimate purposes for interception. It was also argued that some form of ´check and balance´ should exist for the interception of communications and that the Bill should provide mechanisms to ensure that interception is carried out in a legal way. Several participants recommended that the Privacy Commissioner is mandated to approve the interception of communications, while questions were raised in regards to the sharing of intercepted data.

Discussion on self-regulation and co-regulation

The final session of the meeting consisted of a debate on self-regulation and co-regulation. Questions were raised in regards to how self-regulation and co-regulation could be enforced. Some participants recommended the establishment of sector regulations which would mandate the various forms of surveillance, such as a separate regulation for the UID scheme. However, this recommendation was countered by participants who argued that the government would probably not approve every sector regulation and that this would leave large areas of surveillance unregulated.

The participants who supported the self-regulation framework argued that the government should not intervene in the industry and that the industry should determine its own rules in terms of handling its customers´ data. Other participants supported the co-regulatory framework and argued that companies should cooperate with the Privacy Commissioner in terms of handling customers´ data, especially since this would increase transparency on how the industry regulates the use of customers´ data. The supporters of co-regulation supplemented this statement by arguing that the members of the industry should comply with regulations and that if they do not, there should be sanctions. Such arguments were countered by supporters of self-regulation, who stated that the industry should create its own code of conduct and that the government should not regulate its work.

Furthermore, it was argued that although government regulations for the handling of data could make more sense in other countries, in India, the industry became aware of privacy far sooner than what the government did, which is why a self-regulatory regime should be established in terms of handling data. Such arguments were countered by supporters of co-regulation who argued that the industry has vested interest in self-regulation, which should be countered by public policy. This argument was also countered by participants arguing that, given the high levels of corruption in India, the Privacy Commissioner in India may be corrupt and co-regulation may end up being ineffective. Other participants questioned this argument by stating that if India lacks legal control over the use of data by companies, individuals are exposed to potential data breaches. Supporters of co-regulation stated that the Privacy Commissioner should formulate a set of practices and both the industry and the government should comply with them.

Meeting conclusion

The second Privacy Round Table entailed a discussion of the definitions used in the draft Privacy (Protection) Bill 2013, as well as of chapters II, III and IV on the right to privacy, the protection of personal data and the interception of communications. The majority of the participants agreed that India needs a privacy legislation and that individuals´ data should be legally protected. However, participants disagreed in regards to how data would be safeguarded and the extent to which data collection, processing, sharing, disclosure, destruction and retention should be regulated. This was supplemented by the debate on self-regulation and co-regulation which concluded the meeting; participants disagreed on whether the industry should regulate the use of customers´ data autonomously from government regulation or whether the industry should co-operate with the Privacy Commissioner for the regulation of the use of data. Though a consensus was not reached in regards to co-regulation and self-regulation, the majority of the participants agreed upon the establishment of a privacy legislation which would safeguard individuals´ personal data. The major issue, however, with the creation of a privacy legislation in India would probably be its adequate enforcement.