THE history of the census in New Zealand dates back to the days prior to the establishment of responsible government—indeed, some crude form of census-taking appears to have been brought into operation in the early “forties,” shortly after the proclamation of British sovereignty over the Islands of New Zealand.

The volumes of statistics of New Zealand covering each year from 1841 to 1852, prepared (in triplicate) in manuscript on forms with printed headings, contain for each year from 1842 onwards statements of population (male and female separately) for each of the settlements of New Zealand. These appear to have been compiled in each settlement by the local Resident Magistrate, and it would appear that it was the custom to take a complete count of the population in the settlements themselves, which in those days were few in number and circumscribed in area. The population is seldom given in round numbers, and, where it is, an explanation is added—as in the case of Nelson in 1842, where an estimate of 2,500 is given, with a footnote to the effect that actual figures were unobtainable on account of the Resident Magistrate having been killed by Maoris. There seems to be little doubt that these early annual enumerations are entitled to rank as censuses, but most of them are palpably incomplete in that they cover only the definite settlements and apparently do not include settlers, whalers, . &c, at remote places.

What is usually regarded as the first general census of New Zealand took place in December, 1851, under the provisions of the Census Ordinance, 1851 (No. VIII of Session XI). This census, besides being the first taken by law, was more comprehensive than earlier enumerations in that information was collected not only as to sexes of the population, but also as to age and degree of education of each person. Information was also obtained concerning numbers of scholars at day and Sabbath schools, as well as regarding live-stock and crops.

Censuses should have been taken in 1854 and 1857, but for some unexplained reason the requirements of the Ordinance do not appear to have been complied with in all the provinces into which New Zealand had then been divided. A question asked on the point during the first session of Parliament (in 1854) elicited the information that censuses had been taken in Wellington, Canterbury, and Nelson in 1854, but had been accidentally overlooked in Auckland, while the Government appeared to be in doubt as to the position elsewhere. Auckland, New Plymouth (now Taranaki), Wellington, and Canterbury Provinces all took censuses for March, 1857, and Otago either took a census or was otherwise able to supply complete figures as to numbers and sexes for December, 1857. Wellington took another census for March, 1858.

Judging from remarks made in the House and in official documents, the position was not regarded as at all satisfactory, and steps were taken in 1858 to place the census-taking on a proper footing by the passing of the Census Act of that year, which will be referred to later on. A census of the whole country was taken in 1858, the forerunner of a long series of censuses which have followed at intervals of three, four, or five years. The list of dates of all general censuses from 1851 onwards is as follows:—

Tenders will be received at this office till Saturday next, the 13th instant, from persons willing to undertake the duty of leaving the census schedules with the several householders in the undermentioned districts, and calling for the same when filled up in accordance with the provisions of the Census Ordinance, Session II, No. 1.

For further particulars apply at this office.

John Poynter,

Resident Magistrate.

The history of the census in New Zealand reveals a steady widening in the ambit of the inquiries involved in so far as they relate to population characteristics. The gradual tendency in this direction is indicated by the progressive changes revealed by the successive census enactments already referred to.

Copies of the schedules used in connection with the census prior to 1858 cannot be traced, but the early census-takings appear to have been very little more than mere enumerations of population. The census of 1851 obtained details of sex, age, and degree of education, live-stock and crops being also enumerated. It was not, however, until the passing of the Census Act, 1858, that anything like a comprehensive inquiry into the domestic and social condition of the people was instituted.

The remarks of Mr. Stafford (who seems to have been in advance of most of his contemporaries in appreciating the desirability of the existence of accurate statistical information) in moving leave to introduce the Census Bill of 1858 are worth quoting:—

“It is highly desirable that a perfect picture of the moral, social, and commercial position of the colony should be obtained. The importance of this matter is now beginning to be recognized by every civilized State, so that an eminent historian has truly said, that government is now a government of statistics; while in bygone ages Governments were content to legislate on mere hypothetical knowledge in this direction, a course which has led to many serious errors.”

The chief innovations as regards the scope of the census introduced by the Act of 1858 were in respect of enumeration of live-stock and land in cultivation, including the number of acres under various crops and the total area fenced. The Act was not to apply to Natives except in so far as the Governor in Council should direct to the contrary.

The population portion of the census was to embrace the domestic and social condition of the people, including the number and description of the houses, relationship of occupants to heads of households, conjugal condition, age, sex, occupation, extraction and place of birth, religion, date of entering colony, whether British subject or alien, ability to read and write, and schooling (distinguishing Government, public, and private schools, and day and Sunday schools). Details were also collected in connection with places of worship, hospitals, dispensaries, and other charitable institutions, savings-banks, friendly and building societies, literary and scientific societies, and manufactories and works.

To these the collection of agricultural and pastoral statistics was added by the amending Act of 1867, stocks of grain on hand at the date of census being added in 1870, and stocks of flour in 1877.

The Act of 1877 made little alteration in the scope of the census except that it required details to be furnished as to persons suffering from sickness, accident, or infirmity.

By 1890 the modern distinction between census proper and statistics had emerged, and by the amending Act of that year the question as to the amount of land fenced was omitted from the census, as were the details in respect of stocks of grain and flour on hand as at the date of the census. These statistics were from that date collected not through the medium of the census, but through the instrumentality of the Agriculture Department.

The chief innovations in connection with the census introduced by the 1910 Act were in respect of the inclusion of details as to duration and issue of marriage and grade of occupation, and' authority was granted for the inclusion of other details to be prescribed from time to time.

It has already been mentioned that there has been a gradual movement in the direction of limiting the scope of the census to a population inquiry entirely, and to treat the collection of statistical information other than concerning the population as quite apart from the census. In this connection it may be pointed out that until recently along with the population census were taken censuses of industrial manufacture and agricultural and pastoral production. These inquiries are now conducted annually as part of the general routine work of the Census and Statistics Office, so that at present the only extraneous matter taken in conjunction with the population census is an inquiry as to churches, chapels, and other places of worship, and libraries; in addition the census schedule is still regarded as the best medium for obtaining information as to poultry and bees. On the other hand, while formerly the population census dealt entirely with individuals, recently its scope has been extended to cover the family or household, its composition and its home.

A copy of the schedule used in connection with the 1916 census is reproduced.

Collection of the Census.

It has always been the practice in New Zealand to enumerate the de facto and not the de jure population—that is, to enumerate with each locality the persons actually residing there at the date of the census, not those habitually residing there.

In the early censuses, before the abolition of the provinces, appointments of Enumerators for the several provinces appear to have been made by the Governor, on the recommendation of the respective Superintendents, who were requested to nominate the persons most competent in their judgment to discharge the duties satisfactorily. Sub-Enumerators were selected by the Enumerators, so that the machinery for the collection of the census in each province was virtually organized by its own Provincial Government.

In connection with the census of 1878 the size of the Enumerators' districts was reduced. For the provinces were substituted for this purpose twenty-five Enumerators' districts. It was claimed at the time that “while the total remuneration for Enumerators was somewhat larger, there was a saving in the cost of clerical labour by Enumerators, and the local cost of collecting by Sub-Enumerators could be more closely watched and kept within reasonable bounds.” As a result of this innovation the expense of the census per head of population was reduced from 7d. in 1874 to 5¾d. in 1878.

The census of 1916 witnessed an important innovation in respect of the mode of distribution and collection of the census schedules, resulting in a considerable saving of expense. This was the substitution of the machinery of the Post Office for the old system of special Enumerators.

Postmasters acted as Enumerators, while specially selected letter-carriers performed the functions of Sub-Enumerators. In the few cases where no letter-carriers were available the Enumerator made arrangements with some outside person for the collection.

Not only was the work performed more economically (the cost of collection was approximately £20,600 in 1911 and £17,500 in 1916), but the schedules were in much better condition than at any previous census, the proportion of incomplete entries being infinitesimal, and the necessity for queries being reduced to practically nil.

A point in connection with the 1916 census was the increase in the number of Enumerators’ districts—ninety-five, as compared with fifty-nine in 1911—and it is probable that the consequential reduction in the average size of the districts made for increased efficiency.

Briefly, the system of census-collection, which has not altered except in only minor details during the past forty years, is as follows:—

Each Enumerator is supplied with two maps of his district, and instructed to mark on them his proposed subdistricts (for each of which a Sub-Enumerator is appointed), and to send one of them to the Government Statistician for approval, keeping the other by him for his own use.

The maps returned by the Enumerators are forwarded, on the divisions being found satisfactory, to the Lands Department, where smaller maps are prepared from them, one for each Sub-Enumerator.

On the maps are coloured lines defining the boundaries of all local-governing districts, such as counties, ridings, road districts, boroughs, and town districts. Boundaries of provincial districts, electoral districts, and goldfields are also indicated.

The Sub-Enumerator’s principal duty is to deliver a household schedule at every dwelling in his subdistrict before census night, and to collect it as soon as possible thereafter. He is required to keep in a book provided for the purpose a note of all schedules delivered and collected, with particulars of the number of persons, distinguishing the two sexes, returned on each schedule.

On the Sub-Enumerators’ maps the whole territory is cut into very small blocks by means of dotted red lines. On the map, which is fixed inside his collecting-book, the Sub-Enumerator enters the number of people residing in each block, this being done primarily for the information of the Representation Commissioners, who are charged with the preparation of new electoral districts based on the population disclosed by each census enumeration.

The Enumerator’s duties consist in making all general arrangements for his district, and in keeping a close supervision over his Sub-Enumerators during the course of the distribution and collection of the schedules. To enable the Enumerator to keep himself better informed of the progress of the work, a Sub-Enumerator’s daily report form, showing clearly the actual stage reached at the end of each day, was introduced in connection with the 1916 census. As the collected schedules come to hand, after check by the Sub-Enumerator, they are again closely scrutinized by the Enumerator, and, if necessary, returned for collection or completion before despatch to Wellington, where the work of compilation is carried out.

For general statistical purposes it is usual to exclude from the population of New Zealand the Maoris and the residents (European and Native) of the Cook and other Pacific islands annexed in 1901, and in this report it will be understood that unless the contrary is stated all figures given are exclusive of these divisions of the population, which are dealt with in special sections at the end of the report.

At the census taken for the night of the 15th October, 1916, the population, excluding Maoris, of the Dominion proper was 1,099,449. The census of the Maori population taken at the same time showed the number of the Native race to be 49,776, including 3,529 half-castes living as Maoris, as well as 5 persons returned as belonging to the now practically extinct Moriori race. The population of the Cook and other annexed islands at the same date was 12,797. A summary of the total population is as follows:—

The census of 1916 was abnormal in that there were at that time some 43,000 soldiers absent from the Dominion on service abroad with the Expeditionary Force. The effect of the absence of so decided a proportion of the younger adult males will be noticed in the statistical comparisons made with former censuses throughout this report.

Between the dates of the last two censuses, that of 1911 and that of 1916, roughly five years and a half elapsed. Most countries take censuses at intervals of ten years. In Denmark, France, Germany, and New Zealand it has, however, been the practice to use the shorter interval of five years, and an agitation has for some time been afoot in the United Kingdom and elsewhere for the shorter interval to be adopted. The chief argument in favour of the quinquennial period is probably the increased accuracy with which intercensal estimates of population can be made. Such estimates are used as a basis for calculating birth and death rates, and the various per capita rates used to show alterations in the volume of trade, amount of indebtedness, taxation, &c. In all such computations, in the interval between two successive censuses it is clearly essential to interpolate values for intermediate periods, and, although the error in such interpolation may be small as regards the whole of a country, the value of such an estimate as a basis of local population, and therefore for electoral and allied purposes, must always remain somewhat problematical. This is especially the case in rapidly developing countries such as New Zealand, where changes in industrial conditions produce rapid alterations of the centre of gravity of population. The sole reason why other countries have refrained from more frequent census enumerations appears to be on the score of expense.

During the whole of an intercensal period therefore it is desirable to estimate the total population, and there are various methods which have been adopted to this end, such as calculations on the basis of an arithmetic (or, better, a geometric) progression. In New Zealand, however, the more accurate method of making quarterly intercensal estimates from the records of births, deaths, and migration has been adopted, with admirable results, as will be seen from the following table showing the estimated population as at 30th September, 1916, and at 31st March in each of the preceding seven census years, compared with the population as ascertained by the census in the same years. The comparatively large difference in 1916 is due in part to the difficulty experienced in keeping a reliable record of arrivals and departures since the outbreak of the war.

The migration returns are compiled from the records of the Customs Department, and the departures from the Dominion are checked by special returns furnished by the pursers of passenger-steamers, so that where persons who did not book their passages have been omitted the necessary additions can be made. The pursers' returns also serve to prevent the occasional omission of the full number of persons leaving by any one vessel, which sometimes happened prior to the introduction of this check.

The system of registration of births in force in New Zealand is one of the best in the world, and, apart from the compulsory provisions of the law, the people appear to appreciate the advantages of having their children's births duly registered. A few births are overlooked either from inadvertent omission or, in certain obvious cases, from a desire to avoid publicity. This is evidenced from the fact that every year a few of the births registered under the provisions of the Legitimation Act are found not to have been previously registered as illegitimate. Another proof that all births are not registered at the proper time is afforded by the registrations effected under the various Extension Acts passed from time to time to provide for the registration of births after the lapse of the statutory limit of two years. But most of these births date back many years, and the advantage taken of the provisions of the Extension Acts is in the main merely proof of the public's appreciation of the desirability of registration being effected. Probably unregistered births do not exceed 100 annually.

Notwithstanding the “suspicions of defective counting” suggested in his “Principles” by the eminent American economist, Professor Taussig, in lightly passing over New Zealand's remarkably low death-rate, the system of registration of deaths in the Dominion makes it almost impossible for any death (other than odd cases of drowning, &c, where the death is not witnessed and the body not found) remaining unregistered. The duty of registration is imposed on the undertaker, and in the remote event of that individual overlooking the requirements there is a formidable array of checks in the form of notifications by police and certificates from medical men, Coroners, and clergymen, not to mention inquiries for proof of death which sooner or later arise from insurance companies, lawyers, &c.

The “missed” deaths are therefore practically nil, and even allowing the limit of 100 unregistered births per annum it is evident that only a small portion of the slight discrepancy between census and estimated population is thus accounted for.

Nevertheless it is probable that it is in the registration of births and deaths that one of the causes of the discrepancy lies, and this owing to the simple fact that the births and deaths taken for a quarter are of necessity those registered, and not those occurring, during the quarter. It is probable that on the average three or four weeks elapse between birth and registration, and several days between death and registration. Each quarter's registrations thus include a proportion of births and deaths which occurred in the preceding quarter, and in the census enumeration are included children whose births are not registered, while deceased persons not yet registered as such are excluded. The registrations for the next quarter, of course, include these, and though by next census they will be offset by the corresponding non-registrations at that time, which by the law of averages will not in the ordinary course of events differ materially, it is quite possible for exceptional circumstances to exert a disturbing influence. Moreover, in an interval of five years the number of births or deaths in a quarter, and consequently the number unregistered at the end of the quarter, may alter considerably.

Other sources tending to discrepancies between the census and estimated populations are, of course, the enumeration itself (which can never be perfect), and inaccuracies in the recording of arrivals and departures, while the taking of the census on any night except that between the end of one quarter and the beginning of the next is a factor in a country of increasing population. Then, the following points should not be overlooked: Officers and crews of vessels in port on census night are included in the population of New Zealand, and no record of arrivals and departures is furnished by pursers and others for such people in the same way as for ordinary passengers. This means that if at the date of the census there were many vessels in New Zealand waters the actual population at a subsequent date might, from this cause alone, differ in a comparatively material degree from that estimated. That the difference between one census and another in regard to this item is not a negligible one will be seen from the fact that while on census night, 1911, 4,865 persons were returned as being on shipboard, the 1916 figures were only 3,085. This difference is in part explainable on grounds which will appear later.

As regards accuracy, however, the New Zealand figures compare more than favourably with those of Australia and other countries. In the last Australian census (1911) a discrepancy of 70,000 between the actual census return and the estimate based on registrations and migrations emerged, while the corresponding error in New Zealand was only 1,707.

Towards the latter end of 1914 there was instituted a system whereby inter-Island migration was kept tally of by means of returns furnished monthly through the agency of Collectors of Customs. To what extent an accurate estimate of the population of each Island during intercensal periods can be made will not be apparent until the next census, but it can scarcely be anticipated that the result will be quite as gratifying as in the case of the estimate of the total population of the Dominion. Larger numbers, involving increased risks of error, are being dealt with than in the case of oversea returns, and there is more chance of inaccuracy arising from the non-recording of passengers travelling by private yachts and launches -from Island to Island, and from the overlooking of passengers when at most twelve or thirteen hours (and usually by night) are spent on shipboard.

The increase of population at successive census periods has been—

The population of New Zealand has shown a continuous though not a regular increase in each year since 1855, the first year in which accurate records of births and deaths were obtained and used in conjunction With the returns of immigration and emigration. The greatest increase in any year was in 1874, when 32,118 assisted immigrants were brought into the country in accordance with the immigration and public-works policy of the Government. The next greatest increase of population was in 1863, following the discovery of gold in Otago. The accompanying logarithmic diagram gives a good indication of the progress of population in the Dominion, and is further interesting as showing the gradual tendency towards an equalization of the sexes. The loss of men in the years 1914–16 due to the departure of troops has actually brought the estimated female population at 31st December, 1916, slightly ahead of the male.

The smallness of the increase in 1914 and 1915, and the actual decrease in 1916, are due in great part to the war, immigration being considerably restricted, while the emigration figures were swelled by the departure of troops for the front.

The principal natural divisions in New Zealand are the North, South, and Stewart Islands. These contain nearly the whole population of European descent, the Cook and other annexed islands being inhabited almost solely by coloured Natives.

In 1858 the North Island had a larger population than the South, this position being reversed at the succeeding enumerations until 1901, in which year the North Island was found to have slightly the larger total, a position which it has since considerably improved upon. The Maori war which broke out in 1860 retarded settlement in the North, while a large area of land reserved for the Maoris was until quite recently a serious hindrance to the development of this portion of the Dominion. The construction of railways, roads, and bridges is now giving access to larger areas of new lands, which are being offered for selection in blocks intended for close settlement, and are quickly occupied. The South Island was practically free from Maori troubles, and settlement was more rapid, though much of the land was disposed of in large areas. The discovery of gold in Otago in 1861 and on the West Coast in 1864 attracted to these localities considerable numbers of miners.

In 1870 a policy of vigorous construction of railways and other developmental public works and of assistance to immigrants was inaugurated, resulting in a large increase to the population, including nearly 90,000 Government immigrants introduced between 1873 and 1879.

The population of the two Islands at successive censuses is given in the table following, together with the percentage that each Island bears to the total:—

Population of the North and South Islands, 1858–1916.

The figures for 1916 are not quite fair to the South Island. All the military and internment camps were in the North Island, and many of the soldiers, &c, in these on census night belonged to the South Island. The presence of the camps in the North Island had the further effect of temporarily attracting people from the other Island for business purposes, or to be near soldier friends and relatives. The troops abroad on census night (42,666 in number) may be regarded as having come from the two Islands in proportions approximately relative to population.

The populations of the various provincial of the 15th October, 1916, are as follows:—

The following table shows for successive census-takings since 1880 the percentage the population of each province has borne to the total population of New Zealand:—

With the solitary exception of Southland, it will be seen that each southern province has suffered during the period in respect of the proportion that its population has borne to the total, while without exception the northern provinces have gained.

The following table shows the distribution of the population at each quinquennial census since 1881 in (1) counties and (2) boroughs (which term, unless this presumption is negatived by the context, is hereinafter used as being inclusive of cities):—

The counties contain what is understood to be the rural population, but this is not strictly correct, as some of the towns not municipalized and forming parts of counties have considerable populations. On the other hand, the population of a few of the smaller boroughs should strictly be classed as rural.

A progressive increase will be noted in the proportions of the population living in the boroughs, an increase which many would be inclined to view with alarm when it is considered that New Zealand is, after all, a primary producing and not a manufacturing country. As compared with Australia, however, the rural exodus is not marked, as is shown by the fact that as at 31st December, 1915, 48.24 per cent, of the population of Victoria, 40.83 per cent, of New South Wales, 46.63 per cent, of South Australia, and 38.36 per cent, of Western Australia were to be found in the capitals alone, while 40.06 per cent, of the population of the Commonwealth resided in the six capital cities. The 1916 census figures for the New Zealand metropolitan areas are—

A list of counties and their population (exclusive of interior boroughs) as at the census of 1916 is given:—

Prior to the year 1900 there was no statutory limitation to the number of inhabitants necessary to constitute a borough, and consequently many small centres, the residents being mainly engaged in rural occupations, became municipalities. The Municipal Corporations Act now imposes a limit as to area, and provides that no new borough may be constituted unless the proposed area contains at least 1,000 inhabitants.

The name of each borough, with the number of inhabitants as at the census of 1916, is given in the next table:—

For various purposes (vital statistics. &c.) metropolitan areas have been created at each of the four chief centres, and suburban areas at nine of the more important secondary centres. Each area includes, in addition to the central borough and suburban boroughs, a considerable non-municipalized area adjacent to and contingent on the centre.

An objection to the old system of publishing statistics for the municipal areas only was that these were constantly being extended or otherwise altered, and no proper comparison could be made between one period and. another, and no satisfactory record kept of the social condition and progress of the community. The creation of fairly extensive areas fixed for a long period will remove these disabilities.

Population figures as at the census of 1916 are as follows:—

The several stages in the growth of a New Zealand town may be marked thus: The village, taxed by the parent county, and dependent upon the latter for all public works; then the legally constituted town district, still subordinate to the county in some matters, its affairs administered by a Board presided over by a Chairman. When its population exceeds 500 the town district may become independent of the county and control the whole of its revenues. When the population reaches 1,000 full municipal powers may be obtained, with the dignity of à Council presided over by a Mayor; and finally, if the number of inhabitants should reach 2O.000 or over, the title of “city” may be assumed, although the constitution remains unaltered. The name of each town district and its population at the census of 1916 are as follows:—

In a country such as New Zealand, which is still largely in the development stage, villages and townships occupy a comparatively much more important position than is the case in older countries. Steps are taken at each census to compile accurate statements of population for such centres, and, though difficulty is experienced in many cases on account of the scattered nature of settlements, the following figures of population of small centres are sufficiently reliable to give a fair idea of the relative importance of the various places from a population point of view. Only townships with 100 or more inhabitants have been included.