Evidence to the Waitangi Tribunal. subsequently attached as evidence to the High Court. (February 1994) The additional evidence to the High Court was primarily rebuttal evidence. (August 1994) Extracts from it are appended to this report.
Keywords: Governance; Maori; Statistics;
1. I am a research economist and social statistician, and I work as a consultant.
2. I have been asked by the Maori Congress to assist the Waitangi Tribunal by presenting and reviewing some statistics on various expenditures by the Crown on matters similar to the that of informing and enabling Maori to enrol on the Maori roll.
3. The data I shall present comes from various – usually, ultimately, official sources. In no case do I believe the data I am presenting is so inaccurate that it of no use to the Tribunal, or that it makes my interpretation of it invalid.
4. When comparing the data with the situation under dispute -the amount necessary to inform and properly enrol the Maori under the Maori Enrolment Option (MEO) differences between these circumstances and those for the exemplar data should be taken into account. There are six general ones:
4.1 The complexity of the task: Usually enrolment involves going to a population with a long history and commitment to the electoral franchise, and who are receptive to being enrolled. In the case under consideration, the task involves a group who are already (usually) enrolled, explaining that under the new circumstances of MMP, and the associated changes to the treatment of Maori electorates, that they need to review their old practice (of enrolling on the general list), and consider a new one (of enrolling on the Maori list). I have attached as an appendix a note I prepared on this matter.
I would draw attention to the information kit, Maori Option 1994: Your Choice, dated 23 December 1993, with an introductory letter from the Ministers of Justice and Maori Affairs. It manifestly fails to draw attention to the situation whereby many of (albeit negative) reasons for a Maori enrolling in a general roll under the old (FPP) system no longer apply, whereas there are now new (positive) reasons for the Maori enrolling on the Maori roll under the reformed {MMP) system.
4.2 The urgency of the task: Only two months has been allocated for carrying out the enrolment task, a very short time given its complexity {see the previous paragraph).
4.3 The momentous nature of the task: The enrolment is for a new electoral system. The first occasion will set a pattern for future ones. Getting the system as right as possible on the first occasion is vital, to minimise the subsequent impact of errors, anc to reduce the need for further change.
In effect the decisions that the Maori will make when they make their enrollinJ decision will shape Maori and Maori/Non-Maori politics for the rest of the decade anc beyond (given the inertia in an established system). In such circumstances informec decisions are imperative.
4.4 The size of the target population: While it is natural to think about funding o the campaign in per capita terms, it is relatively more expensive to cover a smaller population relative to a larger one. For instance a campaign to inform and contact 1 million people might cost $10 per head, a campaign of equal effectiveness to ½ million people might cost $14 per head. In simple terms, electoral enrolment like many other population management issues, experiences economies of scale (i.e. falling average costs with increased size) so that larger populations can be contacted more cheaply on a per capita basis.
Subsequently I use the square root rule to adjust for these economies of scale. It is one of the simplest assumptions, which says total costs go up at the square root of the population increase. Thus if the population doubles, the total costs only increases 41 percent (square root of 2 = 1.41), so the cost per capita falls 30 percent (1/1.41 = .70).
4.5 The location of the target population: The costs of a marketing program depend upon the characteristics of the target population. For instance it is more expensive to deal with a geographically dispersed population than a compact one. While the Maori population is not dispersed as evenly through the country as the population as a whole (e.g. the much smaller proportion in the South Island, they are still more rural than the total population (20.9 percent of Maori live in rural areas in 1991 compared to 15.0 percent for the population as a whole[#]), and many of those rural populations are more isolated (Northland, East Cape, Central North Island, South West Taranaki). This adds to contact costs.
[# The Government Statistician defines rural, as living outside cities and towns of more than 1000 people.]
4.6 The nature of the target population: Maori is an oral culture.[#] This rules out some publicity methods even though they are cheaper, especially where the message is complex . A letter plus brochure might be sufficient for a university which wished to enrol its graduates into an alumni association, but would hardly be appropriate in the case of a population which was less dependent on text for information.
[# The Waitangi Tribunal’s adaption of its hearing protocols is witness to the proposition.]
5. In summary, we would expect the per capita cost of a campaign to enable Maori to decide whether to enrol or not on the Maori roll to be more expensive than the larger and simpler exemplar campaigns we have described here, if it is of equal effectiveness.
6. We need some means of comparing campaigns over different populations. Even though there are economies of scale, I propose to use a per capita basis. but this still requires some definition of the reference population. Here I describe what I have done for the target population for the campaign.
6.1 There are a number of potential population groups, including:
– the Maori enrolled on the general roll;
– the Maori enrolled on either the general or Maori roll (since there may be reason for some of those currently on the Maori roll to switch to the general roll;[#]
– all adult Maori, including those not currently enrolled, since the unrolled still have the option of enrolling and expressing their preference under the legislation. Indeed there may be good reasons for some who chose not to enrol under the old system to enrol under the new one.[##]
[# For instance they may feel the general MMP system by itself will offer the Maori the political benefits they require, even without Maori seats (or their enrolment in them). Many would deny this, but the legislation provides that every Maori has the right to reconsider their enrolment.]
[## For instance, one option for those in a safe seat under an FPP system is not to bother enrolling, since one’s vote does not matter. This is not true under a MMP system.]
6.2 In the circumstances it seems to me sensible to take the intention of the legislation and the campaign is for as large a reference population as possible, so I shall work with the entire population of adult (over 18 years of age) New Zealanders who report some Maori ancestry (i.e. the definition in the statute).
Statistics New Zealand (the Department of Statistics) reports that, according to the 1991 Census, 287,250 New Zealanders over the age of 18 said they had Maori ancestry .I have projected this figure, using known demographic trends, to March 1994 as 308,000.
7.0 A point of contention is the government has provided $150,000 (inclusive of GST) to run an education program for the Maori to assist them in deciding whether to use the Maori role option. Thus the government provision amounts to 49 cents an adult Maori.
In addition the government will spend resources on its own behalf to assist the enrolment. (The information kit Maori Option 1994: Your Choice has already been mentioned, and I understand that the intention is to mail out material to every known enrolled Maori.) I have not been able to find an expenditure figure for this.
8.0 In order to assess whether this amount is reasonable, I have examined a number of other similar campaigns, although alas there are no exact parallels.
8.1. Maintenance and Update of Electoral Rolls: According to information obtained from the Justice Department, the government has spent in the three years to 1993/94 (i.e. one electoral cycle) a total of $47 million for the purposes of the maintenance and updating of the electoral rolls. That amounts to an average of $18.80 on an adult population of about 2.5 million.
The usefulness of this comparison is moderated by the figure covering activities other than education, which may have been a low component of the total expenditure, because enrolment is much simpler than choosing between two different rolls. In addition the educational expenditure itself would be targeted into the small proportion of the total population (especially those who were coming of the age eligible to vote).
8.2 The Electoral Referendum Campaign: The Electoral Referendum Commission provided $320,000 for the purpose of educating the Maori on the implications of the choice between FPP and MMP .This is slightly more than twice the figure proposed for this campaign. However, in addition there was a further general expenditure of $4,366,000, directed at the population as a whole. On a pro rata basis the Maori could be allocated around $540,000 of the general campaign fund. In total, over 5 times as much was spent on the Maori for the referendum education than is proposed to be spent on the MEO campaign. In per capita terms it is $2.79 an adult Maori, and $1.99 per person for the adult population, including Maori.
It could be argued that the referendum campaign involved more complex issues than the options campaign. On the other hand the economies of scale effect means that a much larger amount than $860,000 would have had to be spent to attain the same understanding by the Maori if there had been no synergies between the general referendum campaign and the Maori one.
Moreover the government funded campaign was complemented by very active media discover plus two public lobby groups. The total spending on publicity of all kinds was certainly double, and possibly quadruple, the government funded element.
8.3 The Maori Voter Enrolment Campaign. A special grant was made to INCO Services to fund a program to increase Maori enrolment. They reported expenditure of around $100,000 for a recruitment of 12,310 (including some non-Maori), at an estimated cost of $8.12 per enrolment. This figure does not include the contribution from a background national campaign.
8.4 Population Census Publicity and Public Relations Campaign. Statistics New Zealand kindly supplied me with their figure for their publicity and public relations campaign for the 1991 Population Census. In total this was $996,000, excluding the statutory advertisements and 0800 phone lines. The total is equivalent to $0.41 per adult.
In addition many of the activities of the enumerators were a form of publicity, especially at the person-to-person level. Total expenditure on enumerators was $6.0 million, which is $2.46 per adult.
In 1991 the Department had a small program oriented towards the special requirements of the Maori, but there are no separate costings for this. I understand there is the intention to have a larger publicity program aimed at the Maori in 1996.
8.5 Health Reforms Publicity .There are various estimates of the costs of the 1993 health reforms publicity. A figure supplied by the Ministry of Health for television and press advertising, an information brochure, and communications and quantitative research (but excluding the 0800 service), came to $2,870,000, or $1.15 per adult.
9.0 A summary of the per capita costs of the publicity campaigns I have examined is shown in the attached table. As well as the per capita figure for each campaign I have also given the cost for a similar campaign which was targeted on 308,000 people, adjusting for economies of scale by the inverse square rule (para 4.4).
| Campaign | Cost $/head |
Omissions | Extras | Cost for Campaign of 309,000 $000 |
| MEO Proposed |
.41 | Governemt Campaign? |
None | 150 |
| Electoral Enrolment (National) |
18.80 | Enrolment Process |
16,500 | |
| Electoral Enrolment (Maori) |
8.12 | National Campaign |
Enrolment Process |
NA |
| Electoral Referendum (national) |
1.99 | Media, lobby groups |
None | 1,740 |
| Electoral Referendum (Maori) |
2.79 | National Campaign |
None | 2,450 |
| Population Census (narrow costing) |
.41 | 0800 line, Statutory advertising, enumerators |
None | 360 |
| Population Census (wide costing) |
2.87 | 0800 line, Statutory advertising, enumerators |
enumeration survey |
2,520 |
| Health Reforms |
1.15 | 0800 line | None | 1,010 |
10. Before coming to a general conclusion there are three matters I need to tidy up.
10.1 I have not tried to calculate a sum based upon the individual costs of an itemised education program. Such programs can be gold plated or simple. I have not the expertise to judge the effectiveness of the various possibilities. Neither has the government, which explains the practice of contracting a group (e.g. INCO Services) with a set of objectives rather than a specific program. (This is the method legislated in the Public Finance Act between government and its departmental agencies.) However I would make the point that this campaign is likely to be more expensive than some of the previous Maori education ones insofar as the target population widens from the Maori on the electoral roll, to all Maori because informing those on the general roll is likely to be more expensive.
Nor have I been able to obtain the costs other government agencies are proposing to spend on the activity. I know that they are proposing a mail out, and there is the information kit. To what extent they are funding a free phone calling service or broadcasting and other advertising I do not know.[#]
[# The information kit mentions an 0800 service, but apparently only for obtai a parliamentary enrolment card.]
10.2 One side effect of a high profile MEO campaign will be a non-Maori interest Almost certainly the campaigning Maori will be involved with the media dealing with non-Maori queries which while having a low effect on the objectives of the Maori campaign, while use resources. This may have a positive nation-wide benefit insofar as it encourages non-Maori enrolment and improves non-Maori understanding of Maori (and of Maori politics). This suggests that the contract between the Crown and INCO Services (or whomever) should make specific provision for these activities, and an allowance in the resourcing of the MEO campaign.
This may be especially important if television is used. It is a medium with a powerful but imprecise reach. So while communicating with the relevant Maori, it will also be informing non-Maori viewers too.
10.3 The question of whether the MEO should include a enrolment campaign of Maori on to the rolls involves constitutional issues outside my expertise. However the following remarks may assist by providing a background.
(i) Insofar as there is an enrolment campaign in association with the Maori campaign, it will reduce the costs of future enrolment campaigns, the cost of the campaign should be discounted.
(ii) The statute in effect requires the unenrolled Maori to be treated as if they decided between the Maori and general rolls in the same proportion as those who are unenrolled. However while in the 1993 election only 40 percent (roughly) of the Maori chose to go on the Maori roll, of those who were recruited by the enrolment campaign over 70 percent chose to go onto the Maori roll. While it is difficult to think of a better assumption for legislative purposes, it may well be that the assumption in the statistics biases the proportion on the Maori roll downward, compared to that if there was complete enrolment.
(iii) I was asked to provide a separate estimate of the numbers of Maori who are not registered on the electoral roll. As mentioned earlier I estimated the total number of adult Maori (based on census responses) in March 1994 would be 308,000. The number of Maori registered on the electoral roll at the time of the 1993 election was about 248,000. {The number would be a little less – perhaps 600 – because of mortality since that date.) The gap of 60,000 might be taken as the number of Maori who are unenrolled.
However there may be those who report Maori ancestry for Census purposes but do not do so when they enrol on the electoral rolls. (and vice versa). If this was at all significant then the figure of 60,000 unenrolled Maori would be too high.[#]
[# The Population Census suggests there are around 50,000 adults of Maori ancestory who do not think of themselves having Maori ethnicity.]
11. The purpose of this affidavit is to provide factual material of use to the Tribunal in assessing the claim, together with some interpretative caveats. However I think it within my expertise to make some judgement of the material.
On the basis of the evidence I have been able to collect, and unless there is provision for a major campaign by another government agency, it would seem that a sum of $150,000 (inclusive of GST) is a small amount for a serious Maori Enrolment Option Campaign.
APPENDIX: NOTES ON ENROLLING ON THE MAORI REGISTER
This note draws attention to some of the traditional reasons why Maori voters did not register for the Maori electorates. These reasons are no longer valid under the revised MMP electoral law. They make a strong case why the 150,000 New Zealanders who describe themselves as Maori but are enroled in the general roll should consider transferring their allegiance to the new roll.
The reasons are not simple, nor easy to convey. The weaknesses of the old system means that many Maori, if they remain uninformed, will stay on the general role because of inertia. It is unfortunate that there is only two months for the Maori to become informed, and to transfer their allegiance, a misfortune compounded by the implicit assumption in the legislation that the ‘default’ option is the general roll (otherwise non-Maori would also have to enrol at the same time under the same conditions). Since the next general election will set the pattern for Maori politics for some years -even decades -to come it is imperative that the Maori make informed decisions about their voting allegiance.
Behind this are the revolutionary implications of MMP. It is no longer true that if a vote fails to elect a local member, it is wasted. Because each voter has a list seat vote, each will have the same influence on the overall pattern of parliament whether their local member is their choice or not; whether they are in a safe or marginal seat.
Reasons for changing to a Maori seat
The old Maori seat system was not a good one for the Maori, so many chose to register on the general roll. The new system abolishes some of the worst anomalies, providing positive reasons for all Maori to enrol on the Maori roll.
The old system produced single party seats. Those who supported parties which would not be elected had an incentive to switch to a general seat because their vote would be more effective. Under MMP all votes are equally effective, irrespective who wins the seat. It will no longer matter whether a Maori did or did not support the dominant Maori political party. (until 1990 it was Labour). Under MMP the effectiveness of the Maori voter will not be handicapped by being in a Maori seat.
The old system produced safe seats. Those who supported parties the dominant party (i.e. Labour) but were in marginal general electorates had an incentive to switch to a general seat because their vote would be more effective. Under MMP all votes are equally effective , irrespective of whether a seat is safe or marginal. Thus Maori who supported the dominant Maori political party but registered in a general seat for tactical reasons no longer have that reason. Under MMP the effectiveness of the Maori voter will not be handicapped by being in a Maori seat.
The old system produced large seats: the four largest covering the whole country .Because more Maori win register for Maori seats there could be more seats. If so they will be smaller, with Maori MP better able to service the constituents and more constituents with in easy reach of their MP. More seats, means smaller seats, means better Maori representation.
The old system produced no incentive to register for Maori seats. Irrespective of the number registered there would be only four Maori seats. Now every 22,000 registered is another electoral seat. Moreover the more electorate seats, the more Maori there will be on the party lists. If there are only four seats, most parties are likely to allocate one place in every 15 on their list to a Maori. If there are eight seats, it is likely to be one in every seven or eight seats. The more who register, the more Maori seats, and the more Maori in parliament representing Maori perspectives in every party.
In summary many of the structural disincentive for the Maori to enrol in Maori seats have been removed. The task is to make sure every Maori understands this, and recognises that by registering in a Maori seat their vote will count equally with the non-Maori, so that Maori influence win more properly reflect the size and significance of the Maori in New Zealand.
*******************
EVIDENCE TO HIGH Court (excerpts)
2. I was asked by the Maori Congress to assist the Waitangi Tribunal by presenting and reviewing some statistics on various expenditures by the Crown on matters similar to the that of informing and enabling Maori to enrol on the Maori roll. My evidence to the Tribunal is attached as Exhibit A.
5. In addition I review some of the evidence already sworn to the Court.
Crown Expenditure on the Maori Option Enrolment Campaign
11. When I prepared my evidence to the Waitangi Tribunal I understood that the amount allocated by the Crown for the enrolment campaign was $150,000, as mentioned in the previous paragraph. The Tribunal in its report provided a higher estimate of $220,663 plus $100,000 for “salaries already committed” by Te Puni Kokiri (para 4.5). The figure is higher than that I used because it includes the costs of an information booklet ($23,663) and personal and services by Te Puni Kokiri ($47,000).
12. In his affidavit of 5 August 1994, Richard Carrington Wheeler provides an estimate of actual expenditure (para 8). For comparative purposes it is appropriate to look only at the items for 1994 directly associated with the Maori Enrolment Option Campaign. These are items G, H, and I of Mr Wheeler’s list, and sum to $340,663. According to the evidence of Holden Brent Hohia, sworn on 5 August, this includes “approximate costs of salaries” of $100,000 incurred by Te Puni Kokiri (para 36.1). The main difference between Mr Wheeler’s figure and that of the Waitangi Tribunal is $20,000 for the costs to Te Puni Kokiri for informational pamphlets and posters, commercials, newspaper advertisement and photocopying reported by Mr Hohia (para 36.4).
13. Had I know at the time of my evidence to the Waitangi Tribunal that the true amount of expenditure was going to be $340,663 (or $240,663 if the $100,000 of salary costs are omitted), I would have still concluded on the basis of the evidence before me that the amount was too small..
14. Thus nothing that has occurred since I prepared my evidence to the Waitangi Tribunal which would change my mind on the opinions I presented to it.
Unenrolled Maori
15. I was also asked to provide a separate estimate of the numbers of Maori who are not registered on the electoral roll. As mentioned earlier I estimated the total number of adult Maori (based on census responses) in March 1994 would be 308,000. The number of Maori registered on the electoral roll at the time of the 1993 election was about 248,000. (The number would be a little less – perhaps 600 – because of mortality since that date.) The gap of 60,000 might be taken as the number of Maori who are unenrolled.
16. My conclusion was broadly supported by Mr Hunt of the NZ Post Electoral Enrolment Centre who appeared at the invitation of the Crown, and from Statistics New Zealand. Such differences appear to arise from my willingness to make an attempt to estimate for those factors which Mr Hunt mentioned but did not include.
17. I did mention in my evidence that there may be those who report Maori ancestry for Census purposes, but do not do so when they enrol on the electoral rolls. (and vice versa). If this was at all significant then the figure of 60,000 unenrolled Maori could be too high.
18. The Tribunal came to the conclusion
“that the precise number of eligible Maori voters who are not enroled is not known but it is reasonable to assume the number is not less than 50,000 and may be between 50,000 and 60,000.” (p.37)