Data and Statistics Bill

Members, we now come to Part 3. Part 3 is the debate on clauses 22 to 42, “Collection of data and statistical confidentiality”. The question is that Part 3 stand part.

I’m in something of a dilemma, because I’ve a very high regard for the Minister in the chair, Dr David Clark. That said, he may not be able to answer the questions that I’ve got. Before I go on to the substance of my question in Part 3, which may need to be repeated in a later call, can I just touch briefly on the ad hominem defence of the Governance and Administration Committee, to which Mr Seymour interjected he didn’t agree that—he didn’t have faith in the committee, and I interjected I agreed and I was a member of that committee. It’s not only because the committee has a Labour majority. It’s because making law is a very, very messy kind of sausage factory process, of which the committee of the whole House is a really important part.

So the fact that the committee did a good job should not detract from the questions and the analysis that we’re doing now, to which I would go to Part 3, clause 22, paragraph (c), because what we’ve established in Parts 1 and 2 from the Minister is, in his words, there is no mandatory ability for the Statistician to collect data for the purposes of research. He answered my question correctly in the hypothetical scenario of whether the Statistician, to populate the Integrated Data Infrastructure, popped along to the Ministry of Education and said, “We’d like your data on school attendance.”—the Ministry of Education is quite free, in the answer the Minister gave, to decline that request.

If that’s the case, why is clause 22 written in the way that it is? Because what it says is, at (a), “Data may be collected … by means of a request”, so you ask for it, or by means of a request under delegated authority—paragraph (b). But if those are unsuccessful and the Statistician believes that “the data is necessary or desirable to enable the Statistician to produce official statistics, or desirable for research”—and those are the key words: “or desirable for research”—then he or she can just go and ask for it. So not only did the Minister not give a clause in the bill that made it clear that data collection for research was not mandatory, but I now point to a clause, clause 22(c), that actually demonstrates the opposite: that the Statistician, if he or she considers it necessary or desirable for research under Part 5—the Statistician can go and get it. That to me is the very definition of mandatory. If you request it under (a) or (b) or otherwise, well, we’ll just ask for it because we think it’s desirable.

Now, we’ve established in Part 1 that the Minister and the Government’s goal here is to collect data for official statistics using mandatory means but not to collect data for research using mandatory means. My submission to the committee is that clause 22(c) says the opposite of that, which is why in Supplementary Order Paper (SOP) 197 I have a very straightforward amendment, and that is to simply delete the words “or desirable for research under Part 5”. That change will give absolute effect to what the Minister has told this committee is his Government’s intention, and that is mandatory collection for data for statistical purposes but voluntary collection for research purposes. Only, this clause says the opposite. So I commend SOP 197 to the committee but also ask the Minister whether or not he agrees with my assessment of (a) what he said and (b) what clause 22 says.

Thank you, Mr Chair. Firstly, I just want to agree with the member’s comments around the sausage factory, and the fact that the Governance and Administration Committee did a good job does not detract from the need to have a robust debate in the committee of the whole House. I think I said that previously, but I think it’s important to note that for the record. Certainly I did not want it interpreted that we shouldn’t be having a robust debate at this point. It is an important part of our democratic process, and valuable.

In respect of the member’s discussion around clause 22, there’s a very important distinction here, which is that collection does not equate to the ability to compel. That is the absolutely critical thing. Collection—so data may be collected—is not a compulsory power. It is possible to collect it. It may be collected. But the department can refuse that request if it does not believe—it is empowering but it’s not a compulsion. The clause clearly references sections 23 and 26, which explicitly reference collection of data for official statistics. Sorry, I’m jumping ahead. Clause 29—and I risk causing confusion by bringing those things in too early—is the clause that places the obligation to comply with a request. That clause clearly references sections 23 and 26, which explicitly references collection of data for official statistics. I think it’s really important to note that this all hinges on the words “may be collected”. It’s not a compulsion at this point. That’s the really important and subtle distinction. Subtle—you know, it’s legal terms. That’s the distinction that’s being made in this particular point in the sausage factory.

Point of order, Mr Chairperson. I seek the leave of the committee to correct the record for the voting on tabled amendment No. 2 by Debbie Ngarewa-Packer, to correct it to: Te Paati Māori, two votes in favour.

Leave is requested for that. Is there any objection? There is no objection. The vote will be corrected.

Thank you, Mr Chair. The comment the Minister made right at the end made reference to sort of the legal interpretation, and that’s the lens through which I’m looking at this. I don’t have any doubt that the Minister’s intentions have been made clear to this House, and that may even be helpful, but when we’re passing black-letter law and a judge or a lawyer is then required to interpret what the law was that was passed, we need to make really sure we’ve got it right. Because if there is a refusal to provide information, it seems to me that the statistician can—and I refer the Minister to his own reference, actually, in clause 29, which is entitled “Obligation to comply with request”. So if the statistician asks for it, it can’t be declined, because there’s an obligation to comply with the request in the manner—“must provide the data by the date, and in the manner and form, specified in the request”.

The question, really, is: does paragraph (c) also apply, even though a request has not been made under section 23 or 26, and, if so, why do we need the clause? Because the way I’m reading this, and the way I think a lawyer would read this, is if the Statistician makes a request under paragraphs (a) and (b) of 22, clause 29 says there’s an obligation to comply with it when the request is made under section 23 or 26. But even if the request hasn’t been made under section 23 or 26, paragraph (c) of clause 22 says, “if the Statistician considers the data necessary or desirable to enable the Statistician to produce official statistics, or desirable for research under Part 5”. What the Minister has said is that the key word in that request is “may”, but the word “may” refers to the actions of the statistician, not the response by the person being asked for the data, right? So the Statistician may ask for it, but it doesn’t say anywhere that the person being asked may decline the request—in fact, quite the opposite. It’s very strongly implied that the Statistician may collect data under this clause. Despite Dr Clark’s repeated assertions that this is not a mandatory collection of data for research, my reading of clause 22 is very much that.

Now, and this is the point that I was making in Part 1, which is where is the clause that says data is not mandatory? I don’t think there is one. A way to fix that would be to delete the words “or desirable for research under Part 5”.

Again, just to clarify, clause 22 simply enables the statistician to collect data for official stats and research. A request isn’t made under clause 22. It’s not where the—it doesn’t describe the requests being made. It’s an enabling clause.

Just briefly, I couldn’t agree more with the Minister, but that’s not what I was saying. In fact, I did say that as well. This is a “may” for the statistician. The question isn’t whether the Statistician may collect data. The question is what is the obligation on the person being asked for the data? My reading of this clause, very much so, is that he may collect it for research purposes and it must be complied with. That’s the point.

I’m also interested in this point, which my colleague the Hon Michael Woodhouse is raising, around clause 22(c), and it’s this continuation of statistics and research. I note the Minister’s point that the obligation is—and clause 29 refers to sections 23 and 26, but in clause 23 the Statistician may request data from any individual, public sector agency, or organisation. This takes us back to the functions of the Statistician, which we didn’t get a chance to properly canvass, in my view, earlier regarding collecting data in accordance with this Act under clause 13.

Then, if we go back to clause 11, the Minister has functions and powers under this Act “to specify the Government’s data and statistical priorities:”. And then if we go back to clause 6—and a question I asked but wasn’t answered last night—in the interpretation section “data includes information”, and there’s the issue of: what does that mean? I mean, this suggests that the data the agencies can compel other Government departments to provide for research does not have to exist solely as defined in a structured dataset, but can be any information held by it, however loosely. And then that, of course, goes right back to the purpose of the Act, which is to “enable the trusted collection and use of data for the production of official statistics and for research;”, and there’s just this issue around the policy structure of this bill, where it’s focused on official statistics and research as opposed to official statistics and then allowing it to be used for research in limited circumstances and spelling out what those are.

I’m interested in why the Minister thinks the structure of this bill has got this balance right in terms of conflating research with official statistics all the way through, as we see again in clause 22(c), and this issue of the data including any information. Clause 22 sort of entrenches the shift from data collection that uses formally designed surveys in the census to using administrative data, it could appear, and I’d be interested in the Minister’s thoughts on that. By “administrative data”, I mean data gathered by Government agencies in the course of normal service interactions—for example, when a hospital provides healthcare to a person or a person applies for a benefit, a permit or a consent or any interaction with a Government agency can result in the associated data being pulled together in tools like the centralised Integrated Data Infrastructure.

This bill appears to give the Statistician the ability to request any data if they are of the view that it’s desirable for research. But then data itself, it means any information, in clause 6. So there seems to be a significant expansion of the dataset that can be used and, in fact, if data includes any information, then the individual public sector agency or organisation upon whom the request is made under clause 29 has an obligation to comply with that request.

So, as the Hon Michael Woodhouse said earlier, the Minister is making clear what his expectations are, but the issue is how this is going to be interpreted by lawyers and by the courts, and there’s just significant issues, as I read this, in the conflation through this bill with official statistics and data being requested for the purposes of research. So I’d appreciate it if the Minister could respond to some of those issues.

Thank you very much, Mr Chair. I note you closed the debate on the earlier part because you said that there was too much discussion about yourself. I’d like to point out for the committee that statistically there was very little debate about you, but I guess we all recall some things more than others—selectively.

I just want to ask the Minister about clause 15, and, for that matter, 19. These are clauses that say that principles of engagement by the Statistician with Māori “must begin early and be meaningful”, for example, and I wonder why that’s there. Statistics New Zealand, funnily enough, has categorised people into a series of ethnic groups: European New Zealanders, Māori, Pacific, Asian New Zealanders, and also MELAA—Middle East, Latin American, and African, which I’ve always found an odd statistical grouping. It seems strange that throughout this Act, including in Parts 1 and 2—although we weren’t able to debate it—the Data and Statistics Bill repeatedly emphasises that Statistics New Zealand and the Statistician must engage in special ways with Māori, which turn out not to actually be that special: “must begin early and be meaningful”.

Well, I would have thought that the objective of Statistics New Zealand would be to engage with all New Zealanders early and meaningfully. I just wonder if the Minister could explain why, for example, he hasn’t said that Statistics New Zealand or the chief statistician should engage early and meaningfully with Asian New Zealanders. The Minister, I’m sure, will be aware that Statistics New Zealand is the one Government department which has Asian New Zealanders as the largest ethnic group employed, according to the Public Service website. I’m not entirely surprised. If you look at the academic results, Asian New Zealanders have been doing very well in maths and stats for a long time. But, you know, it’s clearly a group that is engaged with the subject, that cares about it, and, actually, I wonder why we aren’t including Asian New Zealanders as a group of people that should be engaged with.

Over the weekend, I had the pleasure of watching the film Whina. I highly recommend it. One part of the film depicts Dame Whina Cooper leading the Māori Women’s Welfare League to do their own census of how Māori were living in Auckland after urbanisation in the 1960s. Because Statistics New Zealand didn’t seem to care enough to collect—in fact, they were mainly using administrative data, and the whole problem was that Māori, not being engaged with public services, weren’t part of the administrative data. So they did a survey from outside the State sector, by the Māori Women’s Welfare League. I also note that during our country’s vaccination roll-out during COVID-19, we failed to reach a lot of people because we use healthcare administrative data rather than Statistics New Zealand data to identify the population to vaccinate. There were some people who had been caught by Statistics New Zealand but not the healthcare system.

I make these points because while it was critical that engagement with Māori was improved in order that we didn’t miss things—like things had been missed that Whina Cooper uncovered with her census. Today, as someone who’s been responsible for charter schools, for example, there’s a lot of people in the Pacific community that could probably do with better engagement by Statistics New Zealand. There’s a lot of people in the Pacific community deeply disadvantaged. There’s a lot of Asian New Zealanders who have different needs, culturally. There are European New Zealanders who are having a rough time of things. And, of course, there are Middle Eastern, Latin American, and African New Zealanders, if we’re going to use that slightly odd Statistics New Zealand categorisation of humans by ethnic groups.

So Mr Speaker, Madam Chair, Mr Chair—see, you’re not that important! I want the Minister to stand up and explain why he has clauses such as 15 that say that the Statistician should have early and meaningful engagement with Māori; why that’s needed: is he saying that he doesn’t want to engage meaningfully and early with every other ethnic group? If he does, why doesn’t he put those in? And if he’s going to start listing every category of human by ethnicity, why don’t we just say to hell with it, take it out, and just say we’re going to engage meaningfully and early with all New Zealanders?

I’m not sure I’m going to get a response out of the Minister to these comments, but I’m working very hard on the possibility that I have misunderstood clause 22. I know that possibility is remote but I do draw the Minister’s attention to the fact that in support of my contention that the collection of data for research purposes is mandatory, clause 33(3) says, “No individual or organisation may specify conditions on which data is provided by the individual or organisation to the Statistician as described in section 22(c)(i) for the production of official statistics.” , and “(4) No public sector agency may specify conditions on which data is provided”. So we’ve got all of these clauses around the conditions, the ability for information for research to be supplied, with conditions, and basically the Statistician can turn around and say that there no conditions. We still don’t have a clause that says it’s voluntary—which is what the Minister told the committee. It would be a really simple amendment. He could support my amendment to just simply say it’s voluntary.

I want to move to Subpart 4, and this is a very important part of Part 3 of this bill, because it goes to the census in 2023. This needs a little bit of a preamble, Mr Chair, so I’ll trust you’ll indulge me before I get to point I’m going to make. And that is that when the Minister came before the select committee in February, I think it was 13 February, I asked him about whether the Government was thinking about delaying Census 2023, and his response, which is on the official transcript, was almost theoretical: “Oh, yeah, well, you know, we might give that some thought.” What we now know from Official Information Act requests is that they have been thinking about that for 2½ months, maybe longer, and there was a very high level of concern that the debacle of 2018 was going to be repeated and that they wouldn’t be ready for a 2023 census and how they are going to give themselves that power. Because this bill basically said when it was introduced that a census had to take place. It originally said that Statistician must take a census of population and dwellings in 2023 and every fifth year after that.

So we quizzed the officials about what’s going on. We got some fairly muted responses, I have to say, until a draft amendment to the bill turned up at the end of March, and it basically said—a simple change—that the Statistician must take a census of populations and dwellings in New Zealand in 2023 or 2024. That’s all it said—no real fanfare. The ability to just kick the can down the road for 12 months or more, as much as 20 months, was just sort of slipped into the bill.

Colleagues will recall that after the February Canterbury earthquakes in 2011 it was necessary for the previous Government to move quite quickly to delay the census because it simply wasn’t possible to conduct a census in those conditions. What did that previous Government have to do? It had to change primary legislation. It had to come back to the House and lay out its case for why the census would need to be delayed, and it needed to get the support of Parliament. And it needed to appropriate some funds, because delaying the census comes at a not insignificant cost.

What we have in clause 34(1) is completely different, and I don’t accept that the committee of the whole House is scrutinising the conditions that could lead to a delay in Census 2023 in the manner in which it occurred 10 or 11 years ago. This is very important because Official Information Act documents that were provided to me definitely showed that Stats NZ staff were very worried about this and also were trying to find a way in which a legal power could be granted. Simply putting an amendment into the bill was probably a eureka moment—for an official to say, “All we have to do is change the bill, and we’ll slip it through, and Bob’s your uncle.”

They gave other options, which was to have a more belts and braces enduring ability to suspend a census in certain circumstances, and those certain circumstances were laid out in information provided to the Minister, and they’re actually contained in my Supplementary Order Paper (SOP). There are four: a national state of emergency, an unreasonable level of risk to public health and safety—which, of course, a pandemic may constitute—an unreasonable level of risk to Stats New Zealand staff, and civil unrest. There is none of that in the bill. So it was thought about and officials basically said, “Oh, we’d need to do significant policy work.” Well, I’m not so sure that’s the case. They came up with four very good reasons why you might want to delay the census. I’m not quite sure how much more policy work needs to be done. It seems like a solution by a contractor looking for a problem.

So there are two really key things to my SOP, and I’ll just remind the committee of what one is, and that is the Government Statistician should not be given the power to just make that call, and that’s what the bill we’re considering will do—it doesn’t have any caveats whatsoever. It just says that for next year or the year after, the Government Statistician makes that call. I don’t think that’s a decision that should be delegated to the Statistician. I think it should be given to the Minister, and the Minister should make a comprehensive consultation with the public about why that’s necessary. It’s the Minister who decides, it’s the Minister who is part of the Government that is voted in by the public and is accountable to the public. The Statistician isn’t. Also, a delay should occur following public consultation and should be done only for one or more of the following reasons: a national state of emergency, an unreasonable level of risk to public health and safety, an unreasonable level of risk to the health and safety of staff employed by Stats New Zealand, or civil unrest. That’s it.

Here’s what’s not on that list: risk of political embarrassment, and the risk that six months out from the 2023 election we have a debacle the way we did with the ham-fisted 2018 census. They can say whatever they like; the documents that I’ve seen and the public statements that the Minister has made suggest to me that he is going to take a paper to Cabinet recommending the delay of the census until 2024 and that that will be a decision the Government Statistician makes, if this bill is passed. So the Minister will be able to do two things: say, “That’s the Statistician’s decision, not mine.”, and then blame the public. Because all of the public statements he’s made on the question of Census 2023 say, “We may not have it, not because the Government’s not ready but because the public aren’t ready.” They weren’t able to do a dry run in the Eastern Bay of Plenty in the midst of a pandemic, and that’s fair enough. There was probably quite significant reluctance on behalf of the public to having strangers knocking on doors and chatting to them, even with masks on. But that was a long time ago, and we’re still nine or 10 months away from Census 2023. I don’t believe that’s going to be a reason for the census to be delayed.

Definitely, political embarrassment six months out from an election is not a reason for the census to be delayed, so I’m commending Supplementary Order Paper 156 to the Minister—in fact, I’m doing the Ministers’ job. The Stats New Zealand staff did—they came up with a better framework than simply just giving the Government Statistician a chance to kick the can down the road. We know any delay will come at a significant cost, so this House does need to be involved at some point. Maybe a ministerial statement would be sufficient, but he’ll have to appropriate for that change, and we need some transparency around what the cost of that will be. But we just can’t have a public servant going, “Nah, I don’t think we’re ready. We’ll just delay it for a year.” That’s not what robust public policy-making should be, and if colleagues agree, they must support SOP 156.

Mr Chair, thank you. I’ll just make the obvious point in response to David Seymour’s last contribution, that clause 15 is in a part of the bill that’s already been passed. I appreciate he was probably making a political point and was just wanting to resurrect that debate, so I won’t respond further to that at this stage.

But the Hon Michael Woodhouse, I think, raises a very good and important debate to have in this part of the bill, albeit a conspiracy I don’t subscribe to. It is very much the case that the Government Statistician is independent and proudly independent and rightly independent, and that they produce statistics every week that may risk embarrassing a Government. Happily, sometimes they are also cause for celebration—the record-low unemployment figures we have at the moment, for example—but the Statistician produces those without fear or favour and this bill carries across the provisions that protect that independence, and I think all members of this House really appreciate that.

But what I don’t subscribe to is singling out this one particular set of statistics in saying that somehow the chief statistician will be influenced in a way that he or she might not be influenced every other day of the week when they’re putting statistics out, and I do not see any reason there could be—to pick up another of the member’s points—political embarrassment. I mean, the last census was an obvious example of a framework that was put in place, and not particularly well thought through, by the previous National Government. So if it was an embarrassment to anyone, it was certainly an embarrassment to the previous National Government. If the opportunity arises—and I hope we get it—to put out a census prepared for under the current Government’s time where we can contrast with the kind of census that was put out under the framework created by the prior Government, if anything, it’s an opportunity to draw a contrast between a competent Government and the one that went before.

All levity aside, I will repeat the comments I made, admittedly, at the start of my contribution in Part 1—so I do think it’s worth canvassing them again because they are directly relevant to this part on the census itself. I, in my introductory comments, made reference to the fact that the census is a good example of the Statistician’s independence. The primary consideration for deciding whether the next census is in 2023 or 2024 is the viability of the census operation and corresponding quality of the census data. Considering whether to proceed with the census as soon as all relevant information is available is part of due diligence, and it should be done before investing time and money in activities that can’t be recouped. Those kinds of activities are things like launching publicity campaigns and printing millions of forms, all of which have a cost to the taxpayer. That is something that the Government Statistician has to weigh up—whether we are in a position to run a successful census.

Importantly, such decisions respect the time, effort, and goodwill of people and communities that come together to support census delivery and who, rightly, should expect Government to have considered whether the census was likely to be delivered successfully. Given the pandemic and the lessons of 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.

So with those comments, I push back, and the message which the Government Statistician, from what I’ve heard, has put out there and from what I believe from my interactions are that at this stage—you know, if it was my personal view, and odds on, I would be expecting that census to go ahead. But if there were another major strain in a pandemic or an earthquake of the magnitude of the Canterbury or the Kaikōura sequence on top of challenges from a pandemic, and so on, the Statistician may make other decisions. But, ultimately, that is the Government Statistician’s decision—and rightly so—and the idea that we would put a political overlay to that is something that I think is quite unnecessary.

Thank you, Mr Chair. This debate was going so well until the Minister started talking about which Government was competent—inadvertently, some would say, incompetently admitting that his Government ran a census with a framework that wasn’t fit for purpose, which is something of an own goal. Look, I mean, nobody’s been asked—

💬 Hon Peeni Henare: It was already set by the previous Government.

Hey, Peeni Henare is saying something that I can’t hear behind his mask. I mean, Peeni Henare, nobody’s been asking about the statistics for the importation of mountain bikes or the number of times people move house in this debate, so let’s just keep it civil, folks.

I actually do want the Minister to address the question I asked about clause 15. Why does this law say that the principles of engagement by the Statistician for Māori say that engagement with Māori must begin early and be meaningful. It’s not good enough to say—

💬 Hon Dr David Clark: Point of order, Mr Chairperson. This has already been canvassed that this part of a previous part of the bill. The member continues to raise clauses that are not in the part being debated—

CHAIRPERSON (Hon Jacqui Dean): Thank you—thank you. David Seymour, I’ve just assumed the Chair, so if he could direct his comments and his questions to the Minister on Part 3 of the bill, which is the part we are considering.

Yeah, well, the only reason for raising it is that we were unable to get on to that topic in the previous part. If the Minister doesn’t want to answer it, then he doesn’t have to. But it is part of the bill, and it would be helpful for him to address it for the goodwill of the public and their understanding of the law being made, even if we weren’t able to debate it earlier.

Thank you, Madam Chair. In response to the Minister’s assessment that I was somehow undermining or challenging the independence of the chief statistician, nothing could be further from the case. He goes on to say, “Well, there are always statistics that are being produced that may cause political embarrassment.” I agree with that. That’s not the point. The point of the delay of the census is that there will not be adequate data collected in the census that will enable any sort of conclusions to be drawn, much less political embarrassment to be reached.

It is quite rich for the Minister, as Mr Seymour said, to score an own goal by admitting that six months after they took office, they went ahead with a census that they thought was not ready to go and then blamed the previous Government. That’s not good enough. Actually, the things that are a concern to the Minister now, and that is, in particular, social economic groups and ethnic groups—Māori and Pasifika particularly, who may be reluctant to participate in Census 2023—were actually reluctant to participate in Census 2018, and that had nothing to do with COVID. I think the Māori response rate to the census was 63 percent. That is far below the level that would be required to draw meaningful conclusions from the data collected, and it had to be backfilled with administrative data, which makes it inherently unreliable, in my view.

💬 Simon O’Connor: What percentage again?

Sixty-three percent, Mr O’Connor, was the rate. They’ve had five years to fix that, and there has been a pandemic—I grant the Government that—but we’re well past the point, I think, where we should be blaming COVID for everything. The Government needs to—

💬 David Seymour: They’re starting to blame the next disaster.

Well, that’s right—he’s already anticipating that. Look, that’s the point of my amendment—these things can happen. We know that because we had an earthquake to deal with. But the idea that the Government Statistician should be so independent as to commit the Crown to about $80 million of extra cost in his own decision to delay the census actually takes independence far too far. The strong convention—and the reason we’re even debating this in the House—is that when and how censuses are held is a matter set out in primary legislation. We’re debating the very point that the Parliament is giving the requirement to conduct a census next year, and to simply tag on “unless a public sector official”—who’s not accountable to the public—“decides that we’re not ready for it”, that’s OK!

I’m saying it’s not OK, and that’s not a slight on the Government Statistician; it’s a slight on the Government. Put the four criteria in, give the Minister the responsibility that he should have, and the House has, to make the decision based on the advice of the Statistician and engage the public on why. That’s what a sensible, responsible, and transparent Government would do.

Thank you, Madam Chair. Minister, Part 3, which we’re debating now, is concerning the collection of data, and something that this bill does is it diffuses the powers of the Government Statistician to others who the Statistician delegates those powers to. So not only the Government Statistician, but other chief executives to whom those powers have been delegated may request information. And so there’s quite a significant diffusion of those powers in this bill. Now, the second thing is that there’s a significant change in terms of the collection of data to produce official statistics or desirable for research. So these powers would be delegated to chief executives, and then they will need to determine whether they are requesting data, produce official statistics, or desirable for research which they have a power to do under clause 26. So a public sector agency “may request the data on behalf of the Statistician from an individual, public sector agency, or organisation, if the individual, public sector agency, or organisation is in a position to provide it.” My question is: what oversight is there going to be to ensure that this diffusion of powers and this broad change to request data for official statistics or for research is done appropriately, as these powers are delegated and the powers of this Government Statistician are quite broadly distributed across different public sector agencies?

Thank you, Madam Chair. The member again comes back to the delegations. I think this is the sixth time I’ve rehearsed these points. I still think they’re important, so I’m going to say them again. Look, the Government Statistician is accountable for every decision made by a delegate. The idea of diffusion of responsibility here somewhere is just not relevant. The bill hardwires on the legal obligations regarding independence and confidentiality, which apply to any delegate, and it is a criminal offence to wrongly obtain data while acting as a delegate of the Government’s Statistician, which is in clause 79 of the bill.

I repeat these things many times because it applies to all of these different sections and the member is, I appreciate, trying to test where the different parts of the Act may lead to a different outcome, but I think we have to fundamentally come back to that delegation principle, which is hardwired into the Act.

I repeat my question to the Minister, which he’s refused to engage with. He said he wouldn’t talk about clause 15 because it was in the wrong part. Clause 15 requires the Statistician to engage early and meaningfully with Māori. I simply asked why not do that with everybody—people of all backgrounds. As it turns out, engagement with Māori on census in clause 35, which is in this part, requires basically the same thing, just with respect to census. It’s not hard to find these clauses. They are in every part of the bill we’ve debated so far.

So given that the Minister has committed to debating sections that are in the part currently before the committee, can he please explain why we have clause 35 and why we’re not committed to work with and meaningfully engage with New Zealanders of all backgrounds, whether they’re Middle Eastern, Latin American, and African; whether they’re Pacific people, who face many social and economic challenges; whether they’re Asian New Zealanders, who have a distinct culture; or whether they’re European New Zealanders, many of whom face hard times and could do with rigorous statistical engagement? Why are we emphasising engagement with Māori above all others in this legislation? Why do we not require the statistics department to engage with all?

💬 Hon Members: Madam Chair.

No, I—the Minister was seeking a call, yeah?

I guess, just to cover off the member’s point again: the Treaty obligations that are included in the bill.

Can the Minister elucidate: is he saying that the Treaty of Waitangi requires the Government of New Zealand to put special provisions to engage with Māori over and above what they do for every other New Zealander who is not Māori?

Thank you, Madam Chair. That’s a very important question that David Seymour’s asked, but there’s also another important question which the Library and Information Association of New Zealand Aotearoa has asked the Minister in a letter that they sent on 25 July to Hon Dr David Clark, the Minister of Statistics. They are concerned, Minister, that this bill dangerously conflates two separate types of Government data: official statistics, and the unspecified data gathered as a result of Government agency activity.

The Library and Information Association of New Zealand Aotearoa, which we assume includes many, many professional people who work in statistics, who work in libraries—in fact, they represent 195 institutions and 1,125 library and information professionals, Minister—have asked for the bill to be withdrawn altogether because of its dangerous conflation of two separate types of Government data. What they’ve said is, “Look, please remove the powers to gather data for research”, covered in Part 3, Subpart 3. They said, “The bill should provide for updates to the current regime for official statistics, and then work on an exposure draft of a bill to govern the access and gathering of data for research purposes.” Because there are legitimate reasons why we’d want to research things going on in New Zealand, and why Government agencies which collect information—whether it’s about the environment, transport, or agriculture, for example—why it may be important to have access to that data.

So the Library and Information Association of New Zealand Aotearoa has legitimately asked, Minister, with respect to Part 3, Subpart 3, to remove the powers to gather data for research as a proposed amendment to the bill. They do point out, Minister, and I’d like you to respond to this, in 1976, when considering the Wanganui Computer Centre Bill, Parliament recognised the danger of intersecting data collections and required there be a policy committee to provide oversight and governance of the computer centre. The committee was to consist of Government agency CEOs, as well as external representatives from the New Zealand Law Society and the New Zealand Computer Society.

Now, the Library and Information Association of New Zealand Aotearoa believes that New Zealand’s gone backwards in understanding and mitigating the risks of access to this kind of data, at a time when other Governments have moved forward. So the question that the Library and Information Association of New Zealand Aotearoa has, and that the ACT Party has, is would you look at the Australian Data Availability and Transparency Act 2022 for an approach to data oversight? That Australian Act creates a commission and an advisory board to look at applications to use Government data for research. The Library and Information Association of New Zealand Aotearoa believes that’s an appropriate approach to data and research in New Zealand, and this organisation represents over 1,100 professionals and 195 institutions, including many people who work in Government, who work in agencies, Minister.

So could you please respond to that—to remove the powers to gather data for research, as suggested, and the concerns around the separation of research and statistics data. What provisions were previously made in 1976 under the Wanganui Computer Centre Act; why are those controls, those restrictions, those safety mechanisms not included in this bill; and what would the Minister say to simply taking on the best aspects of the Australian Data Availability and Transparency Act?

Thank you, Madam Chair. The burden of the first part of the member’s contribution covered Subpart 3 of Part 3, which relates back to the rather lengthy debate we’ve already have on clause 22, so I won’t traverse that again. But the question the member asked toward the end about the Australian framework, I think, is interesting and has not been canvassed at length here. Certainly, I’ve had a conversation with Sir Geoffrey Palmer, who encouraged me to have a look at the Australian Act. My research in that area has come back to the point that, actually, that Australian Act—the relevant legislation in New Zealand that covers off those same concerns as the Privacy Act. The sharing of data is governed by the Privacy Act here. So the Australians have put that in a different Act, but we have the equivalent protections.

I move, That the question be now put.

No, this is a substantial part of the bill. I am still hearing new material. There are a number of amendments and Supplementary Order Papers proposed, so I am keeping a very close eye on proceedings.

Thank you, Madam Chair. Before I talk about clause 39(2)(ba), I want to just touch back on a point that the Minister made around the independence of the chief statistician. I’m going to contrast that in the context of the potential for a delay to the census with two other clauses in this bill. One is on the delegation of powers. So the chief statistician can only delegate powers with the approval of the Minister. In joint collections, the chief statistician can only enter into an agreement with a public sector agency for the joint collection of data with the approval of the Minister. And so we’ve got this weird situation where the chief statistician is not so independent that he can’t, without the Minister’s approval, go into a joint collection agreement or a delegation of powers agreement, but he’s so independent that he can delay the census next year for up to 20 months, at a cost of between $80 and $100 million to the Crown, without the Minister’s role. So, if anything, the Minister’s comments about the independence—if he adopts the same approach on independence as he did with clause 17 and clause 24, he should support my Supplementary Order Paper 156.

Now, I want to go to Subpart 5, “Statistical confidentiality”, not so much to ask—and I’ll be relatively brief because I don’t have a question for the Minister, but I think it’s really important to put on the record for this committee what the Governance and Administration Committee was concerned about and what we reached, in the context of clause 39(2)(ba), which is data that may identify an organisation when they don’t want to be identified. That exercised the committee quite considerably, because at all times we agreed with the paramount importance of data confidentiality, but actually individual and organisational confidentiality as well.

So we had this clause which basically said the Statistician will take all reasonable steps to ensure that information isn’t published in a form that could reasonably be expected to identify an individual or organisation unless they agree to it, which is fine. But then this sort of late addition came in that says “data in a form that may identify an organisation (but could not reasonably be expected to identify any individual)” can be published. So it’s data that can identify an organisation, and we were pretty worried about that. That’s kind of the antithesis of what we’re trying to do here.

But I want to reassure the committee of the whole that the select committee scrutinised this at length and put the officials’ feet to the flames. They responded well, I think, as well as they could, in terms of providing scenarios where this would be necessary. The one example that I remember is of analysis on residential care, aged residential care in a particular mesh block or town, where it would be pretty obvious who was being referred to, even if the organisation wasn’t named, because there’s only one of them. So what do we do about that, if the rest home in Cromwell or Arrowtown or something like that was the only show in town and publishing data on the needs of aged care in that region might identify an organisation? I think we got there. But I just want to reassure this committee that the select committee paid very close attention to this and ended up satisfied that this was the right thing to do.

Thank you, Madam Chair. I just want to ask the Minister just to expand a little bit on the comments around the Privacy Commissioner and this bill. So I note that the acting Privacy Commissioner made the comment that the bill should more carefully spell out the relation between the bill and the Privacy Act. I’m just wondering what the Minister’s views are on how that’s expressed in this bill—that relationship between the Privacy Act and this bill, and the collection powers in this bill. I also note that the commissioner suggested that this bill fails to allow for redress when people are harmed by the release of personal information, and recommended that in such a circumstance the bill should provide a right for individuals to complain to the Privacy Commissioner. I’m just interested in whether the Minister thinks that should be in this bill, or, if it shouldn’t be, why not?

The right to complain to the Privacy Commissioner exists beyond this bill. So this bill does not restrict that. That right exists, and I think it’s important to put that on the record for the House, so I thank the member for that question. In terms of incorporating advice from the Privacy Commissioner, along the way there was a close working relationship between the officials at Stats NZ and the Privacy Commissioner, and many of the aspects and things that are included in the bill come from that fruitful engagement and mirror, in fact, ways in which the Privacy Commissioner operates, and mirrors their legislation.

The question is that the Hon Michael Woodhouse’s amendments to Part 3 set out on Supplementary Order Paper 197 be agreed to.

The question is that the Hon Michael Woodhouse’s amendment to clause 34 set out on Supplementary Order Paper 156 be agreed to.

The question is that Debbie Ngarewa-Packer’s tabled amendments to Part 3 be agreed to. All those in favour say aye, to the contrary no.

💬 Hon Members: No.

💬 Jan Logie: Aye. Ayes have it. Party vote.

CHAIRPERSON (Hon Jacqui Dean): I didn’t hear an aye.

💬 Jan Logie: Oh, sorry. Aye.

CHAIRPERSON (Hon Jacqui Dean): No, I’m sorry. I believe we’ve moved on.

💬 Jan Logie: Point of order, Madam Chair.

CHAIRPERSON (Hon Jacqui Dean): Hang on. Let me just—we’ll sort this. Thank you for your indulgence, committee of the whole House. The member may seek leave to cast a proxy vote for the Māori Party.

💬 Jan Logie: I seek leave to cast a proxy for Te Paati Māori.

CHAIRPERSON (Hon Jacqui Dean): Is there any objection. Appears to be none.

💬 Jan Logie: Point of order. I’m sorry. Thank you. Just my notes, I’ve got tabled amendment one and tabled amendment two, and I’m not clear what we’re voting on.

CHAIRPERSON (Hon Jacqui Dean): The question is, and I did put the question, and I will read it again. I’m not putting it again, but I’ll read it out for the member: that Debbie Ngarewa-Packer’s tabled amendments to Part 3 be agreed to. There is another, for the member’s information. There is another question to be put on behalf of that member. This is the first one of those.

💬 Jan Logie: Point of order. Thank you, Madam Chair. If you could help me in clarifying, I don’t have which parts amend—

CHAIRPERSON (Hon Jacqui Dean): No. No, that is not a point of order. I suggest the member just waits. Members, I’m going to put the question again, and for clarity, the question I’m putting, on behalf of the member, relates to deleting references to “research”.

The question is that Debbie Ngarewa-Packer’s tabled amendments to Part 3 be agreed to.

The question is that Debbie Ngarewa-Packer’s tabled amendments to clause 34 be agreed to.