Data and Statistics Bill

We come first to the Data and Statistics Bill and Part 1. This is the debate on clauses 3 to 9 and Schedule 1, Preliminary provisions. The question is that Part 1 stand part. I call the unusually frequent attender, the Hon Dr David Clark. You’re always there!

Thank you, Mr Chair. When I know you’re in the Chamber, Mr Chair, I take great pleasure in engaging in the committee stage debates, and I’ve got two lined up in the current session, so I look forward to take some quality time engaging with the debates. [Hon Jenny Salesa takes the Chair] And it is with some disappointment that I see you moving to another place in the Chamber, Mr Chair. Your presence is always welcomed. But I also delight in Madam Chair being back in the Chamber here to steward our debate on this incredibly important bill. And I do have some introductory comments to make, which I hope will outline a response to some of the issues that have been raised in the select committee stage, that have been raised in subsequent public debates, and which I suspect will want to be worked through carefully here in this Chamber, because this is very important legislation and there is a very important debate to have around it.

Data and statistics and the insights generated from them are critical to effective democracy, decision making, and accountability. Timely and accurate data and statistics support equity by informing public policy decisions, resource allocation, and service design and delivery. The bill we are debating repeals the Statistics Act 1975 and replaces it with a modernised and futureproofed Data and Statistics Act in line with international best practice. It enables a coherent, system-wide approach to addressing data gaps and improving the quality of data used for official statistics and research, under the leadership of the Government Statistician. As Minister of Statistics, I continue to be responsible for specifying the Government’s data and statistical priorities, or the “what” of the statistical production. This mirrors the Minister of Statistics’ role in the current legislation. The statistical independence of the Government Statistician is preserved, and that is what I would describe as the “how” of statistical production.

Statistics will continue to be produced and released free from political interference, a fiercely guarded democratic convention and an incredibly important one, I would stress. As an example of statistical independence, it will be the Statistician who determines the timing of the next census. The primary consideration for deciding whether the next census is in 2023 or 2024 is the viability of the census operation and the corresponding quality of census data. Considering whether to proceed with the census as soon as all relevant information is available is part of due diligence and should be done before investing time and money in activities that cannot be recouped, such as launching publicity campaigns and printing millions of forms.

More importantly, it respects the time, effort, and goodwill of people and communities that come together to support census delivery, and who rightly should expect that Government has considered whether a census is likely to be delivered successfully. Given the pandemic and the lessons from the Canterbury earthquake sequence, frankly, I think it would be irresponsible not to give the Government Statistician the necessary flexibility. Contingency planning is an important part of effective governance.

The bill clarifies the Statistician’s ability to delegate their functions under the Public Service Act, with enforcement actions unable to be delegated outside the Public Service. Any delegate must comply with all requirements in the bill, and the Statistician continues to be both responsible and accountable for actions taken. The bill in its current form removes Officers of Parliament, the Office of the Clerk of the House of Representatives, and the Parliamentary Service from the definition of public sector agency and makes it clear that the Statistician cannot make mandatory requests for data from these bodies. These changes address concerns raised by the Ombudsman and the Auditor-General in their submissions to the select committee.

To support transparency, the bill requires the Statistician to publish information about delegations outside of Stats NZ, including what informed of any such decision. I note that these requirements were added in response to the Privacy Commissioner’s late submission at the select committee. Like the Act it replaces, the bill only allows Stats NZ to collect data for official statistics and research. To be clear, the Statistician can only make mandatory requests for data for official statistics, not for research. It doesn’t expand the scope of information-gathering powers across Government. In New Zealand, sharing of personal information between Government agencies is regulated through agency legislation or agreed information-sharing agreements and information-matching programmes under the Privacy Act.

The bill provides a robust framework for enabling safe and appropriate access to data for research to benefit New Zealanders. This is an internationally recognised function of national statistical officers set out in the implementation guidelines for the UN’s fundamental principles of official statistics. Research is an important output of statistical agencies globally and incredibly important for the betterment of peoples. It’s one of the reasons why, of course, the Data Iwi Leaders Group has argued in the Mana Ōrite agreement that there should be data collected to fill gaps and so that we produce high-quality research that betters the population outcomes. So removing research from the Act would be against what the UN fundamental principles suggest and against what the Data Iwi Leaders Group are arguing for in their agreement with the Government. So I will challenge in due course that particular Supplementary Order Paper put forward by the Māori Party. To remove research from the Act, I think, is short-sighted.

Like the Privacy Act, the bill enables access to data for research, subject to strong confidentiality requirements ensuring protection of personal privacy. But the bill goes much further by setting out a comprehensive framework that the Statistician must use when making decisions about access for research. Data may be accessed for the purposes of research and only in accordance with the enforceable confidentiality requirements in the bill. Stats NZ will continue to hold the data, with researchers only having access through a secure portal to the data they need for their research.

For the first time in New Zealand’s history our statistics legislation recognises the relationship between Māori and the Crown and the Government’s obligation to work with Māori to ensure that data and statistics meet the needs of Māori, iwi, and hapū. I do want to acknowledge my associate, the Hon Meka Whaitiri, for the work that she is leading in that regard on behalf of the Government. Closing data gaps and improving data quality will ensure we make the right decisions to improve outcomes for Māori, iwi, and hapū. This revised approach will empower Māori, iwi, and hapū to make decisions that benefit their communities.

I’m very mindful of work under way such as work on developing a Māori data governance model. I did consider a proposal put forward to me by the Data Iwi Leaders Group to include a review provision in the bill but, ultimately, considered this undesirable as I have the ability to review or amend the legislation in my portfolio at any time, as does any future Minister of Statistics. Indeed, stipulating the timing of a review may unnecessarily curtail productive engagement between Crown and Māori, possibly forcing a less than optimal outcome. But I do want to be clear, and I have corresponded with the Māori Data Iwi Leaders Group, that I am committed to reflecting the outcomes from the Māori data governance work in legislation as appropriate. That work is incredibly important work and I think, when it is completed, it is entirely appropriate that that is then considered in the context of the new legislation.

Given the careful and thorough work of officials and robust consideration by the Governance and Administration Committee, I am confident that the bill will ensure high-quality, impartial, and objective official statistics; provide for trusted collection and use of data for producing statistics and research; and protect the interests of people and organisations through privacy, confidentiality, and security measures. I want to thank the officials for their careful and thoughtful work on the bill. They been at it since the Hon Craig Foss was the Minister of Statistics. That is when this bill had its genesis. Early decisions were taken under his careful eye. And then I do, also, want to acknowledge the careful stewardship of the Hon James Shaw, too, through his time as Minister of Statistics. This bill has been a long time in the making. Obviously, the 1975 bill that we’re replacing has been around for a while and many of the provisions are carried over. The modernising was necessary, and it’s taken eight years and officials have worked very hard on it. I thank them for their efforts. I think they have produced a bill which will carry us into the future.

I want to thank the Governance and Administration Committee for their thorough consideration of the bill, as well, and for recommending changes to strengthen and clarify provisions. I want to thank, also, all those who made submissions. Again, as I said at the outset of my comments, I think the debate we’re having around this bill is a very important one.

Thank you, Madam Chair. I think the Minister might need to go and have a chat to his ministerial advisers, because it seems as though they’ve given him his second reading speech to deliver again, exactly 10 minutes long and traversing just about every part of the bill. I’m not going to do the same. I’m going to focus—and I think it’s necessary to some degree, perhaps not to that degree, that the Minister does traverse a wider range in his opening remarks, and he certainly did that.

I will leave him with some homework in respect of Part 1, but I want to touch on a couple of things that he did say. One is: the fact that we’re here is because the Act is old. As a Minister who was born two years before the Act was born, I’m sure he’ll agree—and one who, I think if memory serves, is probably only a few months away from a certain milestone birthday—that something approaching 50 years old isn’t necessarily out of date. His children might disagree, but I’ll go into bat for him on that point.

The Statistics Act, though, is different. And as we have talked about, both in committee and previous readings, the environment for collection of data that existed in 1975 is light years away from where we are now. And I think it’s that very modern, very ubiquitous internet-based speed of data environment that has led to a lot of the concerns that the Minister and I, as spokesperson, and a number of other colleagues have received from the likes of the Council for Civil Liberties—legitimate concerns they are, although I think they’ve gone over their skis a bit on a couple of the issues. But it’s certainly necessary to update this legislation, more so than any other 47-year-old Act.

But I do want to just challenge one thing that the Minister said in his opening remarks. They relate to Part 3, but I think it’s important to signal to him that we need to have the discussion about the role of research in this. Because I think he tried to reassure the committee, in his opening remarks, that research was not a primary purpose and that the collection of data is only for the publishing of statistics. But that’s at odds with the words used in clause 22(c), which I’ll come to and we will have a more reasoned debate about it in Part 3.

Part 1, though, lays out the purpose of the Act, and, as he mentioned several times, the high-quality, impartial, and objective official statistics production is the goal. We have received a number of submissions to say that the purpose is much broader than it should be, including from a previous Chief Statistician and the Council for Civil Liberties. So I wonder if the Minister, in Part 1, could provide something of a commentary and potentially a reassurance that the protections that are set out in clause 3 of Part 1 are robust enough to give those who are concerned about that sufficient comfort so that people can volunteer data that’s often very, very personal, and be reassured that it will be protected both from disclosure to the wrong agents—that it will be kept private, confidential—and free from scrutiny from those who are not, frankly, in their view, desirable to access.

We’ve got some great words in clause 3, but it’s actually clearly not sufficient to satisfy a number of the people who submitted to the select committee and who have corresponded with us since the second reading.

Thank you, Madam Chair. I’d just like to ask the Minister a question about clause 3(b)(i), which says that the purpose of this Act is to: “enable the trusted collection and use of data for the production of official statistics and for research;”. I wondered if the Minister could succinctly explain what the differences are between collecting and using data for the production of official statistics versus research. In other words, how are official statistics different from research as a purpose of collecting data?

I’ll address some of the points that have been raised so far. I do want to acknowledge, as the Hon Michael Woodhouse has done, the correspondence we’ve received from groups concerned to make sure that the Act is doing, or will do, responsibly, those things we would expect it to do, to keep people’s data safe. One comment I would make from the outset is that the greatest part of the Act itself, from the old Act, is carried over. Though it is approaching 50 years, they’ve been 50 good years. And our statistics agency is respected globally. It is an agency that is in high standing, and rightly earned.

It’s also true that what the bill does is codify current best practice across Government. When an Act gets that old, a lot of what it stipulates is no longer directly applicable. The bill doesn’t mention data at all because it wasn’t a term that was being bandied around in 1975, and yet there is data collected across Government every day. Different departments collect data for different purposes, for research. The private sector collects your data every day. We’re dropping data all over the place, it’s the nature of modern society.

What this bill does is make sure there is an ethical framework that sits across the collection and use of that data, and it makes it much more explicit. The bill incorporates effectively the Five Safes framework; it was developed originally by the UK Office for National Statistics. The Australian Act and this bill draw on that Five Safes framework. Anyone who works in the area of statistics will know it by that shorthand, but, for the purposes of the committee, that is principles that must be considered as part of assessing a request for Government-held data. So when we’re talking about research, certain questions have to be asked: why the data is being used—the projects principle; who is using the data—the people principle; where the data is being used—the settings principle; what data is appropriate—the data principle; and how the results of the project are used—the outputs principle. Those things have to be done in the public interest, so there’s careful thought given in the context of any research that’s done.

Indeed, the data is produced so that we make good decisions as a country, so that citizens, you know, when they get GDP data, can analyse trends; from statistics that are produced, can research around those statistics and use those as reliable inputs into wider research; can look at the methodologies; the Consumers Price Index data that’s collected is used in a certain way and can be trusted and analysed, research can be done around that and people’s interaction. And I referred to this report Te Matapaeroa 2020, which has been in the media in the past week—research done using data collected within Government but done in an ethical way to ensure that we are improving the wellbeing, the lot of parts of our society, and improving opportunity for people.

So the bill reflects principles that guide ethical and culturally appropriate data use, including decisions on research access. It’s a bill designed to bring us into the modern age; to recognise that data is being collected; to make sure that we’re collecting and using it in a responsible way; and where it is used, or passed between departments for specific purposes, that that’s done in a way that is responsible.

Thank you very much, Madam Chair. Minister, I note that a number of commentators have raised some quite significant concerns about this bill—Sir Geoffrey Palmer, former Prime Minister; the Hon Anne Tolley, as chair of Transparency International New Zealand; the Hon Maurice Williamson, a former statistics Minister; Len Cook, former Government Statistician; the Library and Information Association of New Zealand; and the Chief Ombudsman.

I’m going to limit my questions just to some concerns raised, obviously, in this part of the bill. And I’m just looking at the purpose, noting clause 3(b)(i) is to “enable the trusted collection and use of data for the production of official statistics and for research”. And I note, for example, Transparency International Zealand, who have echoed the New Zealand Council for Civil Liberties’ concern, saying the bottom line is that this bill dangerously conflates collection and use of data for the purpose of official statistics with collecting and sharing data for unspecified research.

I’ll just note that in the initial statements at the beginning of this debate, the Minister referred to the United Nations Fundamental Principles of Official Statistics and said this was complying with that. I just need to raise a concern that the New Zealand Council for Civil Liberties has expressed, saying that principle 6 of the United Nations Fundamental Principles of Official Statistics says “Individual data collected by statistical agencies for statistical compilation, whether they refer to natural or legal persons, are to be strictly confidential and used exclusively for statistical purposes.” So I’ll just invite the Minister to explain how clause 3(b)(i), namely the use of data for the production of official statistics and for research is consistent with principle 6 of the United Nations Fundamental Principles of Official Statistics, in that it’s used exclusively for statistical purposes; no mention there of research. It would be quite helpful if we could have an explanation of how that is consistent.

The other point, Minister, I would note, and we probably could explore this later on, but it’s just clause 3(d)(i), namely the protection of the interests of people and organisations represented in, or by, data that is used for the production of official statistics and for research, by providing for appropriate privacy, confidentiality, and security. And just where that is shared overseas—and I do note, this is clause 52, so we can probably get into this in more detail later, but I’d just be interested in the Minister’s views on how data shared overseas will be protected when the Privacy Act 2020 does not apply overseas, and how people can be assured that information collected in New Zealand when shared overseas will be protected when those privacy provisions don’t apply to the use of that data overseas.

The other point, Minister, is clause 6, namely the definition of data which says data includes information. I just note, again, that the New Zealand Council for Civil Liberties has raised an issue here where they are saying that this definition suggests that the data that agencies can compel other Government departments to provide for research does not have to exist solely as a defined and structured data set, but can be any information held by it, however loosely. They’ve gone on to say that data that was gathered by an agency for one limited purpose—and which are provided by members of the public on the understanding that that is all that it would be used for—will be available for any research Stats NZ or another agency chief executive to whom the Statistician’s powers have been delegated considers desirable. There will be no obligation on that agency that receives the transferred data to delete it after the research has been completed. So I would just invite the Minister to address that concern that the definition of data in clause 6—“data includes information”—could be any information gathered by an agency for any purpose, however loosely, as opposed to a defined and structured data set.

I just want to address a point raised by the member Mr Mooney. The UN provisions—it’s laid out in the guidelines, the implementation guidelines for the UN’s Fundamental Principles of Official Statistics, the aspect that relates to research. You asked the question about clause 3(b)(i), which is the purpose of the Act, to “promote consistent and collaborative practices across government in order to—(i) enable the trusted collection and use of data for the production of official statistics and for research;”. That’s in the implementation guidelines laid out, associated with those. Just a clear fact to clarify for the member.

Madam Chair, I asked a pretty straightforward question—“What’s the difference between collecting data for official statistics and research?”—and would really like the Minister to give a succinct definition of those two concepts, because they’re fundamental to one of the most significant changes in this bill.

Hon MICHAEL WOODHOUSE (National): I support the request, but I know we’re approaching a time when we will be adjourning this debate until tomorrow morning, and I would encourage the Minister to have a chat with his officials to answer Mr Seymour’s question, because I think it’s going to be fundamental to the discussion we’re going to have on both Part 1 and Part 3—and the Minister is ready to answer; that’s great.

Hon Dr DAVID CLARK (Minister of Statistics): Yes, thank you, Madam Chair. Yes, sorry, I was just trying to make sure I had correctly understood the question—because it seemed frighteningly simple, and it is—I think the member’s asking. So official statistics are those statistics held by Stats NZ, produced by Stats NZ, and research done on the basis of those statistics by researchers. For clarification, I mean, the Act doesn’t enable sharing of data for research. The data continues to be held by Stats NZ and is accessed through a secure portal. So Stats NZ holds on to data for the purposes of ensuring that it is protected appropriately. I think I’ve already answered the member’s question, so I’ll try and be succinct in my answers.

Thank you, Madam Chair. That’s a fascinating answer and lies at the heart, I think, of some of the confusion of people who have submitted, both through the select committee process and afterwards, and the consequent concerns that they have about the use of data for research. Now, up until that answer, I expected to hear that there was a both proactive and reactive element to information or data, which is that the Statistician collects certain data for statistical purposes for the publishing of regular demographic and trend-based analysis, and that research would primarily but not entirely be academic—for example, where a researcher from the university might come along and say to Stats New Zealand, “We’d like to drill into some of the datasets that you’ve got in order to establish trends.” One of the very good examples of that is in our mutual home city, and that is the Dunedin multidisciplinary study, which has not only contributed to the statistics database but they’ve actually drawn on it for the purposes of academic research.

Now, the reason that is important is because if one of the primary purposes, as clause 33, I think it is, says, is that data can be collected for a primary purpose which is research, then that really opens up the scope of what the power of this Act would do in terms of collecting just in case. And one of the disciplines that is set out, as Mr Mooney said, in respect of UN guidelines and the disciplines that we’ve had over the last 47 years is that data is a taonga. It is collected for a specific purpose, and inferred in that is that we need to know what that purpose is. Subsequent to that, research could be done on it as a secondary purpose but not a primary purpose. I think if we are opening up the purpose of statistical legislation to be that broad, that potentially gives the power of the Statistician to go and collect anything just in case it might be researchable in the future. I think that’s what’s giving the Council for Civil Liberties and Transparency International New Zealand concern. But noting the time, I’m sure that’s a conversation we’ll continue tomorrow.

Members, the committee is suspended until 9 a.m. tomorrow morning. Pō mārie.

Sitting suspended from 10.02 p.m. to 9 a.m. (Thursday)