Opeskin, Brian --- "Constitutions and Populations: How Well has the Constitution Accommodated a Century of Demographic Change" [2010] UTSLRS 7; (2010) 21(2) Public Law Review 109

	Home \| Databases \| WorldLII \| Search \| Feedback University of Technology Sydney Law Research Series

Constitutions and Populations:

How Well has the Australian Constitution Accommodated a Century of Demographic Change?

Brian Opeskin

Australia, like most Western countries, has undergone profound demographic changes since 1901. The most significant transformations have been the sizeable growth in population, declining fertility and mortality, substantial immigration, population ageing, and the spatial redistribution of people between the states and territories. This article examines how the Australian Constitution has dealt with the deep processes of demographic change that have reshaped the Australian population over the course of a century. Demography was in its infancy as a scientific discipline when the Constitution was drafted in the 1890s, yet the founders showed an understanding of rudimentary aspects of population dynamics. They anticipated population growth, interstate migration, the role of international immigration in shaping the Australian community, and the importance of population statistics. However, in other respects they introduced constitutional rigidities that have impeded the capacity of government to adapt to population change. This article charts the areas of demographic foresightedness and short-sightedness in the Australian Constitution. It concludes by considering how a constitution might be designed to provide a sound framework for governance that is responsive to population dynamics.

The English Poet Laureate, Alfred Lord Tennyson, once famously described the English common law as ‘That codeless myriad of precedent, That wilderness of single instances’.^[1] It is an apt description of the professional disposition of many public lawyers who are inclined to ask, in individual cases, how public law protects the rights of citizens or advances the interests of the state. Less commonly, public lawyers might inquire how the law protects the interests of specific groups such as indigenous peoples or migrants. Rarely do they ask how public law addresses the needs and interests of an entire population, conceived as an organic entity separate from the individuals who comprise it.

This article wades into these unchartered waters by asking how constitutions address, or might address, deep social transformations that arise from long term changes in a population. The question is worth asking because constitutions have the capacity to shape the evolution of populations over time and because they influence how effectively governments can respond to those changes.

To answer these questions it is necessary to see the world through the eyes of a demographer and then to inquire how the demographic landscape is influenced by institutions, processes and principles of public law. These are large questions and it is prudent to limit the inquiry in space and time. This article focuses on Australia’s federal experience since 1901 but it acknowledges that additional perspectives might be found by extending the time horizon to pre-federation demographic history or by adding other geopolitical entities (states, countries) for the purpose of comparison.

The principal argument of this article is that if constitutions are to provide a solid foundation for the governance of a population far into the future, they need to have population dynamics firmly in mind in their original design and in their subsequent development. Despite the rudimentary state of demographic knowledge in the 1890s, the drafters of the Australian Constitution did consider some aspects of population change. Whether by design or good fortune, the Constitution adopted a number of provisions that have been responsive to subsequent population change. However, in other respects the Constitution is disappointingly short-sighted with respect to population issues.

The article is structured as follows. Part 2 describes the principal population issues that are of interest to demographers and then charts the major demographic changes that have taken place in Australia since 1901. Part 3 examines the Australian Constitution to reveal what it says about population-level issues and how it has responded to a century of demographic change. These questions could also be considered from the perspective of the constitutions of the Australian states but this article focuses only on the constitutional framework at the national level. Part 4 examines issues of constitutional design and speculates about how constitutions can be more ‘population conscious’ by building in appropriate types of constitutional regulation and different degrees of adaptive capacity. Part 5 offers a brief conclusion.

Demography is the scientific study of human populations. While individual human behaviour and experience lie at the root of all demographic events, demographers prefer to study populations that are large enough to be unaffected by the idiosyncratic behaviour of individuals. They are especially interested in population change, and a temporal dimension thus figures prominently in most demographic work.^[2] The name of the discipline was coined by a Belgian statistician, Achille Guillard, in 1855 by combining the Greek words for people (demos) and writings in a branch of science (graphy),^[3] yet the origins of the discipline are nearly 200 years older.

The founder of demography is widely acknowledged to be John Graunt (1629–1674), an English cloth merchant with no scientific training, who in 1662 published a detailed study of mortality in 17th century London. Graunt’s seminal work, Natural and Political Observations Made upon the Bills of Mortality, derived its name from the weekly lists of deaths from the plague that parish priests were required to compile as a means of identifying and quarantining areas of infection.^[4] Graunt’s treatise discussed many important questions that remain central to demographic analysis today, leading one scholar to describe Graunt as ‘ranking among the great natural scientists of the early years of the Royal Society’, and to describe his treatise as a universally recognised ‘work of genius’.^[5]

Developments in demography came haltingly in the centuries that followed. Modern demographic analysis had to ‘wait for large-scale datasets, scientific interest in their analysis and sufficient developments in mathematics to allow that analysis’.^[6] Significant contributions to the field were made by mathematicians, actuaries and statisticians who were interested in calculating the probability of dying or surviving at different ages, and later by social scientists who were interested in explaining the causes and consequences of observed changes in mortality and fertility. An important theoretical contribution was made by Thomas Malthus who, in his Second Essay in 1803, postulated that population growth always had the potential to outstrip the capacity of resources to support the population, leading to positive checks (increased mortality and death) and preventive checks (postponed marriage and declining fertility).^[7] Echoing these views, concerns about the rapidly growing population of the developing world in the second half of the 20th century—popularised in Paul Ehrlich’s book, The Population Bomb^[8]—led to significant institutional strengthening for collecting and analysing demographic data through the United Nations, national governments, and private foundations.

Today demography retains its interest in the formal analysis of population data through mathematical and statistical techniques.^[9] It also has strong connections with related fields of population studies, social biology, human geography, and anthropology, making the boundaries of the discipline quite malleable. Insights from all these fields have practical applications in a range of human enterprise, including electoral arrangements, social infrastructure, urban planning, business, marketing and human resource management.^[10]

Modern day demographers are generally interested in five attributes of human populations, which are considered in subsequent parts of this article.^[11]

The following sections give an overview of the major demographic changes in Australia by mapping key population parameters from 1901 to the present day. The data used for this purpose come from two sources. The first is the extensive data collection of Australia’s principal statistical agency, the Australian Bureau of Statistics (ABS), which was established in 1905 and collects a large array of demographic data.^[12] The second source is the Australian Demographic Databank held at the Australian Demographic and Social Research Institute at the Australian National University. Conceived in the 1960s, the Databank is a collection of officially recorded vital statistics and census data that have been ‘cleaned’ to ensure the greatest possible internal consistency between recorded annual flows of demographic events and the stock of population measured at successive censuses. The first version of the databank included population data from 1921–1976; a second version carried the data forward over the period 1976–1981; and the current version extends the dataset back to 1901 and forward to 2003, but only for some demographic variables.^[13]

The Australian population has grown dramatically since federation, as shown by the shaded area in Figure 1. In 1901 the population stood at 3.79 million. This comprised indigenous Australians (whose numbers had been significantly reduced by European settlement),^[14] survivors of the 160,000 convicts who were transported to Australia from 1788–1867, free settlers, other immigrants who arrived predominantly during the gold rush, and locally born children of all these groups.^[15] By 1945 the population had nearly doubled to 7.39 million, and by 2009 it had swollen to 21.87 million—a near six-fold increase on the 1901 population.

However, the rate of population growth has been quite variable. This can be seen in Figure 1 in the slope of the boundary line of the shaded area, which shows a significantly steeper rise after 1945. It can also be seen in the black line, which records the annual percentage rate of population growth. This has been calculated as a five year moving average to smooth out extreme fluctuations, yet even so there is significant variation over time. It is only in recent years that the rate of growth has stabilised within the range of 1.0%–1.7% per year.

The future size of the population is frequently the subject of intense public debate in Australia. The discussion has been shaped, among other things, by concerns about the magnitude of Australia’s immigration programme and the environmental sustainability of a growing population in an arid continent. These concerns have led to a National Population Summit in 2002,^[16] major government reports (including the Australian Treasury’s intergenerational reports),^[17] and detailed demographic analysis of alternative population futures.^[18] Any projection of future population depends entirely on the population at the date of the projection and the assumptions made about the demographic processes of fertility, mortality and net migration into the future. The ABS has projected that Australia’s estimated resident population will increase to between 30.3 and 40.1 million by 2051^[19] and to between 33.7 and 62.2 million by 2101,^[20] depending on whether high or low growth assumptions are made.

It is an open question how changes of this magnitude can be accommodated within Australia’s constitution framework. We return to issues of population size and growth in Part 3.1 because they are relevant to determining the size of the House of Representatives, and to a lesser extent the Senate, under Australia’s system of representative democracy.

Figure 1: Australian Population 1901–2009

In demographic analysis, the term ‘fertility’ describes the propensity of women in a population to bear children.^[21] Used in this sense, fertility patterns are influenced by both biological processes and individual choice, and this differs from the popular meaning of the term with its emphasis on biological reproductive capacity. Much work has been done on identifying and measuring the proximate determinants of fertility. These include the onset of cohabitation or marriage, disruptions to a union, onset of permanent sterility through menopause, the duration of breastfeeding after giving birth, the probability of conception during the menstrual cycle, the use and effectiveness of contraception, and the incidence of induced abortion.^[22] The social determinants of fertility—particularly the patterns of cohabitation and contraceptive use—obviously alter over time with changing societal values and practices.

The fertility of a population can be measured in many ways.^[23] One of the most widely used measures is the total fertility rate (TFR), which is the average number of children a woman would bear during her reproductive life (assumed to span 15–49 years of age) if she experienced the patterns of age-specific fertility prevailing at that time. In rough terms, the TFR can be regarded as a measure of average family size. Australia’s historical experience of fertility is shown in Figure 2 for the period 1921–2009 (earlier national data are not available). Starting with 3.1 births per woman in 1921, the TFR declined to 2.1 births in 1933 during the Great Depression, rose steeply during the post-war baby boom to a peak of 3.5 births per woman in 1961, and fell away again sharply until the early 1980s. Since 1980, the TFR has fallen as low as 1.73 births per woman, but has now risen to 1.98 births—the highest level since 1977.^[24]

Figure 2: Total Fertility Rate 1921–2009

Australia’s historical fertility experience can be measured against two benchmarks. The highest reliably recorded fertility of any population is that of the Hutterite religious sect in North America between the two world wars. Practising ‘natural fertility’, in which the population does not consciously practise fertility restraint, the Hutterites averaged 12.44 births per woman.^[25] At the other end of the spectrum, a TFR of 2.1 is regarded as replacement level fertility in populations with low mortality: two births replace each woman and man in the next generation, with an extra bit of fertility being required to compensate for the mortality of children before they reach reproductive age. Figure 2 shows that Australian fertility has been below replacement levels (dashed horizontal line) since 1976.

Other important trends in fertility, not shown in the graph, are changes in the age pattern of births, with a shift towards greater fertility of women in their 30s; an increased detachment of childbirth from marriage; and the emergence of a strong negative correlation between childbirth and female labour force participation.^[26] We return to issues of fertility in Part 3.2 because they are relevant to the ability of the federal government to pursue pro-natalist policies using its taxation and spending powers.

The second component of population change is deaths. There are many ways to measure mortality, but a convenient summary measure is the life expectancy of a population at birth, namely, the average number of years a newborn can expect to live under the mortality conditions prevailing at that time. Figure 3 shows the life expectancy at birth of males and females in Australia from 1901–2007. Three features of the graph deserve comment: (i) the sustained rise in life expectancy of males and females (which corresponds to declining mortality rates); (ii) the higher life expectancy of females over males in every year, by a margin of about 5–10%; and (iii) a number of irregularities such as the short-term decline in life expectancy during the 1918 influenza pandemic and the slowing of mortality improvements, especially for men, during the 1950s and 1960s.

Improvements in life expectancy in Australia have been echoed to large degree throughout the industrialised world.^[27] In 1971 Abdel Omran postulated that this was due to an ‘epidemiological transition’ in which declining levels of mortality are associated with changing patterns in the cause of death.^[28] The first stage of the transition—the ‘age of pestilence and famine’—is characterised by high and fluctuating mortality, minimal population growth, and low life expectancy of 20-40 years. In the second stage—the ‘age of receding pandemics’—mortality progressively declines and life expectancy rises to about 50 years; while in the third stage—the ‘age of degenerative and man-made diseases’—mortality declines to a low and stable level and life expectancy rises further.
Figure 3: Life Expectancy at Birth, 1901-2007

Australia’s mortality experience in the 20th century is an apt description of the second and third stages of this transition, and the population can now be said to have moved to a fourth stage characterised by high life expectancy and delayed degenerative diseases.^[30] The main consequences of this reduced mortality have been more rapid population growth than would otherwise have been the case, and a changing age structure of the population, with larger proportions of people at older ages.^[31] We return to issues mortality in Part 3.2 because they are relevant to the Commonwealth’s legislative and fiscal powers in relation to public health. Mortality is also relevant to assessing the mandatory retirement of federal judges at 70 years of age, which was introduced into the Constitution by referendum in 1977.

Australia’s demographic history has been substantially shaped by migration, which is the third component of population change. Mention has already been made of the population impact of convicts, free settlers, and gold rush immigrants during the 1850s. It took many decades, and the introduction of the first centralised passport system during the First World War,^[32] for the Australian Government to assert effective control over the movement of persons into Australia.

Figure 4 shows net overseas migration to Australia from 1901–2008. After federation, Australia continued to receive immigrants of largely Anglo-Celtic origin under the government’s ‘White Australia Policy’.^[33] International migration during this period was relatively modest in scale, especially when one leaves aside the distorting effect of troop movements associated with the First World War. There was a lull in migration during the Great Depression and the Second World War, but substantial changes came immediately after the war, with a boost in the volume and diversity of immigrants. During this period the Government regarded sustained high migration as fundamental to Australia’s national security and economic fortunes,^[34] taking waves of settlers from Britain and then Europe. As the White Australia Policy was gradually unravelled, and eventually dismantled, in the early 1970s, the nature of immigration also underwent a transformation. There was greater emphasis on skilled migration, family reunion, and humanitarian migration, and there was a major expansion in the number of migrants who came to Australia on a temporary basis as tourists, students and short-term contract holders.

Figure 4: Net Overseas Migration to Australia, 1901–2008

Post-war migration has been very substantial, but also variable. Unlike many demographic events that reflect relatively stable patterns of human behaviour and biology, annual migration movements are erratic because they are strongly influenced by changing government policies and economic conditions. Thus, Australian immigration policy has been likened to a boa-constrictor, ‘taking great lumps of new settlers followed by a period of quiescence and digestion’,^[35] which can be seen clearly in Figure 4. We return to migration in Part 3.2 because it is relevant to the Commonwealth’s legislative power over ‘immigration and emigration’. The federal government’s migration policy has been a powerful determinant of Australia’s population dynamics over the past 60 years, and will continue to be so in the future.

Demographers have a strong interest in the age-sex composition of populations because fertility, mortality and migration are all strongly influenced by these two variables. The age-sex structure of the Australian population in 1901 and 2009 is illustrated in Figure 5. These population pyramids show the percentage of the total population that falls into successive five-year age bands, with males shown on the left of the vertical axis, and females on the right. In 1901 the graph takes the form of a pyramid. This is typical of young and growing populations, where a large percentage of people is found is the youngest cohorts. Conversely, the proportion of the population in middle and older ages is relatively small. In 2009 the shape of the graph is quite different, with the pyramid being replaced by a beehive shape. The youngest cohorts form progressively smaller percentages of the population and undercut the graph, while the percentage of the population at older ages is relatively large.

The ageing of the Australian population is predominantly due to the declining birth rate and death rate: immigration has only a minor impact on the age-sex structure in the long term because migrants age like the rest of the population.^[36] The impact of these processes is that in 2010 the population aged 65 years and over comprised 13.5% of the total population, but this is projected to rise to 22.7% by 2050.^[37] These changes will put substantial fiscal pressure on governments as the cost of age-related payments and services rises, while the proportion of working age people falls.

Figure 5: Population Pyramids, Australia 1901 and 2009

The sex structure of the Australian population has also changed significantly over the past century as the population has become progressively more feminine. In 1901 there were 110.1 males for every 100 females. This is a high sex ratio by international standards. Most countries lie within the narrow range of 95–102 in the absence of special circumstances such as heavy war losses or heavy net migration, while a sex ratio below 90 or above 105 is viewed as extreme.^[38] In Australia’s case, the male bias in the population was the product of its particular immigrant history; a bias that attenuated only gradually as the country evolved into a settled population.^[39] By 1951 there were 102.1 males per 100 females, and by 2009 the sex ratio had declined to 99.1 males per 100 females.

This near equality of the sexes across the total population at the present time masks substantial differences in the sex ratio at different ages. It is a near universal phenomenon that more males are born than females—typically around 105 male births for every 100 female births—but the male advantage is offset over time because males have higher mortality than females at every age. The ratio of the sexes at different ages is thus determined by the balance between the natural processes of birth and death, together with any confounding effects of migration. In older ages, especially above 70 years, Australian females outnumber males by a sizeable margin.

The final issue to be addressed in this brief account of Australia’s principal demographic trends is the spatial distribution of the population. Australia’s population distribution is highly uneven and has changed significantly over time, making it ‘one of the most dynamic and policy-relevant dimensions of the nation’s contemporary demography’.^[40] Population distribution is influenced by many variables—including physical, social and economic factors—but historical patterns of colonial settlement have had a special significance as well.

Changes in the spatial distribution of a population can be viewed in many ways, depending on the purpose of the inquiry. This might emphasise long term movements from rural to urban locations, from one urban area to another, or from urban areas to lifestyle locations on the coast.^[41] The change that merits attention for present purposes is the long term net movement between states and territories because this has a particular bearing on the system of democratic representation established by the Constitution (see Part 3.3 below).

Figure 6 shows the proportion of the total population resident in each state and territory in 1901 and 2009. Today—as at federation—New South Wales is the most populous state and Tasmania the least populous. Yet there have been significant changes in relative populations, marked by a shift away from the south-eastern states to the northern and western parts of the country.^[42] In 1901 the four south-eastern states (New South Wales, Victoria, South Australia and Tasmania) contained 81.7% of the Australian population but this had fallen to 67.0% by 2009. Over the same period, Queensland increased its share of the population by 6.9% and Western Australia by 5.2%. These changes have altered the relative rankings of the states, with South Australia losing its position as the fourth largest state to Western Australia in 1983. In addition, the two internal territories now have a combined population greater than that of Tasmania.

We return to the spatial distribution of the population in Part 3.3. There we examine how the Constitution facilitates the movement of persons within Australia through the guarantee of freedom of ‘interstate intercourse’, and determines its political consequences through the spatial allocation of seats in the lower house of the Australian Parliament.

Figure 6: Distribution of Australian Population by State and Territory, 1901 and 2009

Having surveyed the major demographic trends in Australia since federation, this Part asks how ‘population conscious’ is the Australian Constitution, and how well has it been able to accommodate the deep demographic changes that the country has experienced over this period. These questions can be approached from different angles: one could look solely at the text of the Constitution as drafted and later amended; one could examine how the Constitution has been interpreted by the courts; or one could examine how the Constitution has been applied in practice in government administration.^[43] This Part is largely confined to the first approach but it is also necessary to say something about the major judicial decisions that have interpreted relevant constitutional provisions.

As a prelude to this investigation, it is important to temper one’s expectations of what might be found in the constitutional text. The Australian Constitution was not written as an inspirational document predicated on a grand vision for Australia’s future population, but rather as a practical working document that sketched the legal and institutional framework for uniting six colonies into an effective federal union. To this day, the Australian people appear to be unimpressed with grand rhetorical flourishes in the Constitution. A referendum to add a modest Preamble to the Constitution in 1999 was rejected by a majority of voters in every state and territory, and by 60.7% of voters overall.^[44] In addition, demography was in its infancy when the Constitution was drafted in the 1890s. While certain aspects of population change were foreseen, the drafters had neither the data nor the methods of analysis at their disposal to anticipate long term population patterns, whether globally or in their application to the Australian colonies. To this extent, the search for ‘population consciousness’ in the Australian Constitution is a search for nascent ideas about population around the time of federation, as well as a study of how subsequent demographic knowledge has been reflected in our constitutional understanding. The following sections examine how the Constitution addresses many of the demographic issues identified in Part 2.

Population size can exert a significant influence over governmental arrangements in a federation. At a fundamental level, it can be important in defining the territorial boundaries of the states that constitute the union. In the United States, for example, when the 13 original states united to form a Union following the Philadelphia Convention in 1787, their inhabitants numbered about 3.6 million but occupied only a small portion in the north-east corner of the continent.^[45] While the United States Constitution provided that Congress might admit new states into the Union, it left it to Congress to fashion the conditions of admission.^[46] The Northwest Ordinance 1787 addressed this issue by stipulating that a territory could petition for admission as a new state only once its population reached 60,000. This established a precedent (not always adhered to) by which the United States could expand westward into the interior of the continent through the addition of new states.^[47] By 1850 there were 31 states and by 1900 there were 45, with the last two states (Alaska and Hawaii) joining only in 1959.

This organic process of sculpting states out of territory and population was unnecessary in the case of Australia. When Australia’s 3.79 million people united in a federation in 1901 they already inhabited pockets of territory across the length and breadth of the continent. The colonial boundaries drawn up by Imperial authorities—although amended from time to time^[48]—already encompassed the whole continental land mass. Australia’s federal dialogue thus lacked the detailed attention given in the United States to the nexus between population and territory in defining the states. Nevertheless, federations by their nature are unions of polities identified by reference to their people and territory, and the existence and size of colonial populations were important background facts underpinning federation in 1901. As the population grew, it also became easier to refute the claim that the population was too small to sustain a federal system of government with its multiple tiers of representation and regulation.

More importantly for Australia was the role of population size in determining the form of democratic government in the federation. While political thought from the time of Plato and Aristotle had championed the model of direct democracy epitomised by the city-state, the practicality of this ideal was irreversibly challenged by persistent population growth and the rise of the nation state.^[49] It was inevitable that Australia, following the United States example, would adopt a system of representative democracy, and significant sections of the Australian Constitution were devoted to establishing the rules governing democratic representation in the Australian Parliament.^[50] The interplay of population growth and representative democracy can be seen by examining the constitutional provisions affecting the House of Representatives and the Senate in greater detail.

The impact of population growth can be seen most clearly in recurrent debates about the most appropriate size of the House of Representatives. As finally drafted, the Constitution made provision for 75 members of the House of Representatives in the first Parliament, distributed between the original states in accordance with the formula in s 24, but with a guaranteed minimum of five members per state. At a conference of statisticians held in February 1900, the population of the six colonies was reckoned as 3,717,700 so that, in aggregate, each of the 75 lower house representatives in the first parliament represented 49,569 persons.^[51]

There had been earlier attempts to establish a fixed ratio between the population and the size of the lower house, which was referred to as a ‘fixed quota’ system. The First Official Draft of the Constitution Bill 1891 provided that there would be one representative for every 30,000 people.^[52] This was similar to, but more rigid than, the formula in the United States Constitution, which provided that the number of representatives should not exceed one for every 30,000, and therefore permitted dilution of representation in Congress as the population increased.^[53] During the Australian Convention Debates the question was raised whether a fixed quota of 50,000 people was a more appropriate figure, but ultimately it was the very concept of a fixed quota that ran into opposition.

Underpinning these concerns, one convention delegate made a projection that, if Australia’s population in 1897 continued to grow in accordance with past rates, a fixed quota of 50,000 persons per representative would result in a total of 446 members in the lower house by 1941.^[54] This was a gross overestimate of Australia’s population growth in the following decades. This is not surprising given that modern demographic projection methodology did not become commonplace until popularised by the work of Frank Notestein in 1945.^[55] With the benefit of hindsight, the 1941 population of 7.1 million would have necessitated only 142 representatives. Nevertheless, the point had been made that a fixed quota, combined with the progressive increase in the population, could make the House ‘inordinately large and inordinately expensive’, while the Senate would become ‘weak and impotent’ in the absence of a nexus provision linking the size of the two chambers.^[56] The compromise was to adopt a provision that met a number of common objectives: there was to be no automatic increase in the membership of the House as the population grew, but nor was there any limitation on the power of the Parliament to increase the size of the House, provided the Senate was increased proportionately so as to preserve the two-to-one ratio between the chambers—the so-called ‘nexus provision’. Australia’s population growth was thus foreseen (if poorly predicted), and informed the drafters’ views about the appropriate arrangements for securing representative democracy in the new Commonwealth.

In time, as the Australian population grew, Parliament did think it advisable to increase the number of representatives in the lower house—first in 1948 and again in 1983. The constitutional mechanism through which this was achieved was the expansion, by legislation, in the size of the Senate. This is because the nexus provision in s 24 of the Constitution requires the House of Representatives to have ‘as nearly as practicable’ twice the number of senators. When introducing the necessary legislation in 1948, the Government justified the increase by reference to the doubling of Australia’s population since federation and to the great expansion in the responsibilities of government following two world wars and the Great Depression. Australia was seen to have ‘altogether outgrown’ the small parliamentary bodies with which it was equipped in 1900.^[57] In 1983 reference was once again made to the doubling of the population (from 7.5 million in 1947 to over 15 million) since the last major increase in the size of the Parliament, and to the increasing number of electors in each constituency.^[58] However, the argument for change was now more sophisticated. The volume of legislation had grown fourfold in the past 60 years; a robust committee system was necessary for legislative oversight of the Executive and demanded more parliamentarians; and the increase in the size of the ministry by 70% since 1949 simultaneously eroded the pool of parliamentarians available to perform committee tasks. In the government’s view, the Australian people now had much higher expectations about what their representatives should achieve, and meeting those expectations required more hands on deck in managing a complex national economy.

Figure 7 shows the effect of these parliamentary changes on the average number of persons represented by each member of the House of Representatives from 1901 to 2009. As noted above, at federation there were 49,569 persons for each of the 75 representatives, but by 2009 there were 145,833 persons for each of the 150 members—nearly triple the original ratio. Thus, while the size of the House has increased with Australia’s population, the increase has been less than proportionate.^[59] This has been a common phenomenon globally, and the reasons for it are not difficult to find. As Dahl and Tufte remarked in their seminal work on size and democracy:

‘As the membership of a parliament increases ... discussion becomes more burdensome; participation in debate must be more and more severely restricted; delegation of authority to committees creates problems of coordination and collective control. The parliament, in short, becomes less and less capable of functioning as an assembly. As a consequence, parliaments cannot be allowed to expand indefinitely, or even in strict proportion to population.’^[60]

Figure 7: Population per House of Representatives member, 1901–2009

Although changes in the size of the House of Representatives provide the clearest example of the influence of population growth on the arrangements for representative democracy, changes in the size of the Senate are also revealing. At first sight this seems like an odd claim because the Senate was originally intended to represent the states, not the people.^[61] This was made evident by the requirement that each original state was to be equally represented in the Senate because they were ‘equal in constitutional power and status, although not necessarily equal in territory or population’.^[62] Thus, the Senate was initially established with 36 senators—six from each state—and this was increased to 10 from each state in 1948 (a total of 60 state senators) and 12 from each state in 1983 (a total of 72 state senators).^[63]

Yet there was another change to the Senate of special relevance here. In 1973 the Senate was increased from 60 to 64 members with the addition of two senators from each of the internal territories.^[64] The Government’s justification in proposing the change was that ‘all parts and all the people of Australia should be represented in both chambers’ by giving senatorial representation to the territories.^[65] Population issues loomed large in the Government’s explanation, in two opposing respects. On the one hand, the populations of the Northern Territory and the ACT were said to be growing at a much greater rate than those of the states, and this bolstered the case for senatorial representation. On the other hand, those populations were still small (94,800 and 162,000 respectively in 1972), so there was never any suggestion that the territories should be granted as many senators as the states. This was made possible because the Constitution authorises Parliament to allow representation of the territories in either House of Parliament on whatever terms it thinks fit (s 122).^[66] One senator per territory might have been sufficient to satisfy the call for representation, but it was thought prudent to have two senators so that both major political parties could be represented in each territory, and the balance of power would therefore not be affected by the expansion of the Senate. In this way, population size and growth have also been relevant to the composition of the Senate.

The three components of population change—births, deaths and migration—are the major drivers of demographic change. The capacity of government to influence these events, and the support that a constitution gives to government in doing so, is thus central to the subject matter of this article. The issue has poignancy for federal states because effective influence over the components of population change generally requires regulation at the national level. This is because sub-national policies with respect to fertility and mortality are susceptible to being undermined by interstate migration, which is itself constitutionally protected within the boundaries of the federation.^[67] This section examines what the Australian Constitution says or implies about each of the components of population change. As we shall see, the role of the Constitution in this area has been largely facilitative, delineating spheres of federal legislative competence within which different population polices can be developed and pursued.

The regulation of international migration assumed great significance in constitutional debates in the 1890s. This reflected two important and related sentiments of the time: the desire to define the Australian community by excluding members of ‘coloured races’, epitomised by the Chinese; and the desire to wrest control over immigration from Imperial authorities so that immigration might become an instrument of Australian policy rather than British foreign policy.^[68]

The outcome of the debates was s 51(xxvii), which gives the Commonwealth Parliament power to make laws with respect to ‘immigration and emigration’, bolstered by other provisions dealing with ‘naturalization and aliens’ (s 51(xix)) and ‘the people of any race ... for whom it is deemed necessary to make special laws’ (s 51(xxvi). The migration power was not exclusive to the Commonwealth Parliament and for some years the Australian states continued to regulate aspects of international migration, for example by issuing travel documents.^[69] Nevertheless, the new Parliament moved quickly to use its migration powers, enacting the Immigration Restriction Act 1901 (Cth) in its first year to give immigration officers broad discretionary powers to decide who could enter or stay.^[70] The ‘dictation test’, in particular, was used as an indirect vehicle for imposing a colour bar and excluding many non-British citizens, to whom there was a great deal of hostility from labour unions.^[71]

Australia’s migration policy has gone through many twists and turns since federation, yet the ability of the federal Parliament to control the arrival and presence of ‘non-citizens’ through use of the migration and aliens powers is still the overarching concern of the principal legislation, the Migration Act 1958 (Cth). This is understandable for, as Helen Irving has observed, ‘A nation, to be sovereign, must be able to regulate its own community, in part through its immigration laws’.^[72] The legislative powers conferred by the Constitution have thus enabled the federal government to exert substantial control over the size and composition of immigration to Australia. This impact that has been felt most keenly since the Second World War and still dominates contemporary discussion about desirable population futures for Australia.^[73]

By contrast, the legislative power over emigration has been largely untouched. Historically, many nations have been less concerned about who leaves their territory than who arrives in it. Moreover, in the post-war era, international human rights law has placed restrictions on the power of States to regulate the exit of persons from their territory, whether by prohibiting exit or requiring it. Today, individuals enjoy the right ‘to leave any country’,^[74] while the power of a State to expel foreign nationals is limited to well-recognised circumstances such as where a foreign national has entered in breach of immigration laws, engaged in criminal activity, become involved in undesirable political activities, or otherwise threatened national security.^[75]

The role of the Constitution in regulating fertility is as important as its role in regulating migration, though somewhat less obvious. As noted in Part 2.3, fertility is often the subject of individual choice in contemporary Western societies and is therefore susceptible to influence by government policies that affect the costs and benefits associated with that choice. In recent years public discussion has focussed on the decline in fertility, which has been below replacement level since 1976. The concerns expressed about the social and economic consequences of continuing low fertility have prompted governments in Australia, as elsewhere, to ask how public policies can support couples in achieving their desired family size, while overcoming barriers to having children.^[76]

The social value of children as ‘tomorrow’s citizens and tomorrow’s workers’ may be readily acknowledged,^[77] yet Australian government policies that support families, and hence fertility, have been described as incoherent. In Peter McDonald’s assessment, the current system represents a mish-mash of five decades of ‘bolted-on’ polices whose outcome is ‘an illogical, inefficient, complex jumble of conflicting principles’.^[78] Many solutions have been proffered, including family-friendly workplaces, child-care reform, taxation changes, and paid parental leave.^[79] This is not the occasion to debate complex issues of family policy but rather to identify the broad range of powers at the disposal of the Australian Parliament, which provide the foundation for a potentially coherent national policy on fertility.

The first provisions to note are those relating to marriage (s 51(xxi)) and the ‘custody and guardianship of infants’ in relation to divorce and matrimonial causes (s 51(xxii)). The demographic significance of these legislative powers comes from the fact that marital status remains one of the most important proximate determinants of fertility.^[80] The inclusion of these powers in the Constitution comes as something of a surprise given the drafters’ goal of giving the Commonwealth only those powers necessary for national government, while preserving ‘domestic’ matters for the states. What brought these matters into the federal arena in 1901 was the desire for national uniformity in an area already subject to considerable colonial diversity, especially in light of the view that marriage concerned ‘not merely the individuals who are party to the contract ... but also those who are to come afterwards’.^[81]

Beyond the powers over marriage and divorce, successive federal governments have been able to pursue pro-natalist policies through the taxation power (s 51(ii)) by allowing tax deductions for children and for certain child-related expenses. These policies are designed to promote horizontal equity between couples with children and those without, bearing in mind the substantial costs imposed by child rearing. Such policies were championed by Robert Menzies many years ago,^[82] and have been reflected more recently in programs such as the Family Tax Benefit (introduced in 2000 to help with the cost of raising children) and the Child Care Tax Rebate (introduced in 2005 to help families meet the high cost of child care).^[83]

The social security system has been another important source of fertility-related policies at the federal level. A key amendment to the Constitution in 1946 gave the Commonwealth power to make laws with respect to maternity allowances, child endowment, family allowances, and benefits to students (s 51(xxiiiA)). An example of the use of these powers is the ‘baby bonus’, which was introduced in 2004 by Treasurer Peter Costello with the famous exhortation that every Australian couple should have ‘one [baby] for your husband and one for your wife and one for the country’.^[84] Originally introduced as a $3,000 lump sum payment upon the birth of a child, it has risen to $5,185 per birth (paid in 13 fortnightly instalments), and now costs in the order of $1.4 billion annually.^[85] In addition to temporary effects on the timing of births,^[86] the baby bonus has been credited with producing a significant upturn in Australia’s fertility, which is now the highest it has been for a generation. Although some writers have questioned whether the policy was necessary in the first place,^[87] there is little doubt that the Constitution has provided the structural mechanism by which government can affect Australia’s population dynamics through the tax-transfer system.

The role of the Constitution in regulating mortality is the least obvious of the three components of population change. Ultimately, this can be distilled to the capacity of governments to promote public health, provide appropriate health services, and support the well-being of Australians as they age. The federal Parliament does not have a general power over health, and only a few legislative powers are directly related to public health.^[88] These include power with respect to quarantine (s 51(ix)), and the provision of ‘medical and dental services’ and ‘pharmaceutical, sickness and hospital benefits’ (s 51(xxiiiA)). The Parliament also has power to make laws with respect to ‘invalid and old-age pensions’ (s 51(xxiii)), and the income support provided to individuals under this power since 1908 can be seen as one means of promoting the health and well-being of the elderly population.

By far the most significant role of the Commonwealth in relation to health, and therefore mortality, stems not from its legislative powers but from financial control exercised by making grants to the states (s 96) and direct appropriation for Commonwealth programs (s 81). Since 1998 the Australian Government has entered into five-year funding agreements with each state and territory through which it has provided funding to assist the states and territories in maintaining free hospital services to the public.^[89] For example, in 2009–10 the Commonwealth paid $50.1 billion to the states for specific purposes under tied grants, representing 14.8% of total Commonwealth expenditure. Approximately 24% of these grants ($12.1 billion) was for health related purposes, including preventive health measures designed to address the rising prevalence of lifestyle related chronic diseases.^[90] Recently, the Australian Government proposed a major reform of the health funding arrangements, by which it would directly cover 60 per cent of the ‘efficient price’ of every public hospital service provided to public patients. This would be funded by retention of about one-third of the revenue obtained from the Goods and Services Tax (GST), which is currently levied by the Commonwealth under s 51(ii) of the Constitution and distributed to the states and territories under a Commonwealth Grants Commission formula.^[91]

There is another constitutional issue that merits discussion in connection with mortality. In 1977 the Australian people were asked to approve a constitutional amendment that would introduce a compulsory retirement age for federal judges. For justices of the High Court of Australia this was to be 70 years of age; for judges of other federal courts, 70 years was to be the maximum age but Parliament could set a lower limit if it chose to do so. Until that time, the Constitution made no mention of the length of judicial tenure other than providing that a judge could be removed from office on the ground of ‘proved misbehaviour or incapacity’.^[92] In 1918 the High Court interpreted this provision to mean that federal judges enjoyed life tenure in the absence of removal on these limited grounds.^[93]

The Government’s rationale for proposing the constitutional amendment can be seen in the Second Reading speech of the Constitution Alteration (Retirement of Judges) Bill 1977. In the opinion of the Attorney-General there was an almost universal practice that holders of public office retire on attaining a maximum retirement age, for reasons that are ‘well known’ and ‘do not need to be spelt out here’.^[94] He noted that a fixed retirement age had been adopted in all state Supreme Courts and that it was now appropriate to make similar provision for the growing number of federal judges. The proposed change was said to safeguard judicial independence because it applied only to future appointments and prevented any alteration to an individual judge’s term of office after his or her appointment.

More revealing are the comments made during the debate in the House of Representatives, namely, that judges are not immune from the geriatric processes of mental decay, and that the proposal would lead to a younger body of judges who are ‘closer to the people’ and have ‘current day sets of values’.^[95] An unstated reason for the broad parliamentary support may have been the fact that Sir Edward McTiernan was by then Australia’s longest serving High Court judge, and his great age (he was then 84 years) and apparent feebleness are said to have shocked parliamentarians when he swore in new members of the Senate in his role as Acting Chief Justice.^[96]

The notion that there should be a mandatory retirement age for federal judges appears to have been quickly accepted at the time, but there was virtually no discussion in Parliament about the appropriateness of selecting 70 years as the maximum age. Even the report of Senate Committee, on which the Bill was based, said little on the subject, noting only that 70 was the ‘retiring age most commonly established for judges of State and territory Supreme Courts’, and was therefore appropriate for the High Court.^[97] Other federal courts, in its view, should have the same constitutional maximum but a lower statutory maximum of 65 years.

Faced with bipartisan support for the referendum proposal—and in the absence of counterarguments in the official ‘no case’ distributed to voters^[98]—the people gave the proposal their overwhelming support. The referendum passed in all six states and was approved by 80.1% of the population—the third most successful referendum in Australian electoral history. Shortly afterwards the Parliament exercised its new power over the judicial retirement age by providing that judges of the Family Court must retire at 65 years of age.^[99] This was reversed in 1991 in an effort to improve the status of the Family Court in the eyes of the public by giving its judges the same tenure as judges of the Federal Court.^[100] Since that time, all federal judicial appointees have been subject to compulsory retirement at 70 years of age.

The demographics of ageing provide an important subtext in any consideration of the constitutional changes brought about in 1977. In that year, the life expectancy at birth was 70 years for males and 77 years for females (see Figure 3). More importantly, a male who had already reached 70 years of age could expect to live another 10.5 years (to age 80.5), while a female who had reached 70 years of age could expect to live another 13.7 years (to age 83.7).^[101]

Figure 8 shows that the life expectancy of a 70 year old male had changed very little over the period 1901–1977, as indicated by the flat portion of the curve. However, the late 1970s marked the beginning of a significant upswing in the life expectancy of elderly people, due largely to the decline in mortality from heart disease and stroke.^[102] Today a 70 year old male can expect to live another 14.9 years, to age 84.9, while a 70 year old female can expect to live another 17.5 years, to age 87.5. Furthermore, over the past ten years there has been a general increase in the proportion of older Australians reporting their health to be excellent or very good,^[103] suggesting that there is considerable untapped judicial experience in those retirement years. There is also a good deal of expense in taxpayer-funded judicial pensions, which provide retired federal judges with 60% of the relevant judicial salary for the remainder of their life, and then 37.5% of that salary for the remainder of the life of their surviving spouse or partner.^[104]

In 2009 these matters were agitated again during an inquiry into the role of judges by the Senate Legal and Constitutional Affairs References Committee. Once more the Committee recommended that a compulsory retirement age should be mandated for federal judges. However, in contrast to its position in 1976, on this occasion the Committee concluded that 70 years was too low and should be increased to at least 72 years or possibly 75 years, citing the broader social trends of increased life expectancy and later retirement.^[105] The Committee also acknowledged that further increases in life expectancy may bring the matter to the fore again in the not-too-distant future, and that s 72 of the Constitution should accordingly be amended at the next referendum to allow Parliament to fix the retirement age for federal judges.

Figure 8: Life expectancy at age 70 years, 1901–2007

This is clearly a sensible solution which shows a consciousness of population issues that was largely absent during the debate in 1976. The point is not that 70 years of age is necessarily too young for mandatory judicial retirement but that it was a mistake to entrench in the Constitution a specific age that implicitly reflected the mutable demographic experience of the day. The details should have been left to Parliament, and the Constitution should have been confined to articulating enduring principles, such as those that protect judicial independence. This was the method adopted by the Constitution’s original drafters, who on no less than 24 occasions made detailed provision in the Constitution only ‘until the Parliament otherwise provides’. It is noteworthy that, since the 1977 referendum, age discrimination legislation and related policy changes have effectively removed compulsory retirement ages for most workers,^[106] while the qualifying age for the aged pension is set to rise to 67 years between 2017–2023 to reflect improvements in life expectancy and respond to the long-term cost of an ageing population.^[107]

The Australian Constitution has many things to say, explicitly or implicitly, about the geographical or spatial distribution of the population. The power in s 122 of the Constitution to acquire territories, and govern them, has been the means by which Australia has become responsible for a number of ‘far flung island outposts’, including external territories with small populations in the Pacific and Indian Oceans.^[108] The constitutional mandate in s 125 that a seat of government be established within the state of New South Wales ‘not less than one hundred miles from Sydney’ paved the way for the establishment of the national capital, Canberra, with its present population of around 350,000. And the declaration in s 92 of the Constitution that ‘trade, commerce, and intercourse among the States ... shall be absolutely free’ has been interpreted as a guarantee of freedom of movement, thus providing the legal foundation for large scale interstate migration to Queensland and Western Australia, as discussed in Part 2.5.^[109] This guarantee has been extended by statute to the movement of persons between the territories and the states under the Acts granting self-government to the internal territories.^[110] It is also supported by the prohibition of discrimination against out-of-state residents in s 117 of the Constitution which has, among other things, facilitated the practising of professions on a national basis.^[111]

Perhaps the most significant example of the relevance of population distribution to the constitutional framework is again in the area of representative democracy. This might be expected given the role of the Constitution in uniting ‘the people’ of six geographically dispersed colonies ‘in one indissoluble Federal Commonwealth’.^[112] A proper understanding of the Constitution’s spatial dimensions is aided by distinguishing between three principles that might potentially guide the architectural arrangements for representative democracy in a federal system of government.

The first is a federal principle, which determines representation on the basis of the sovereign equality of the polities that united to form the Commonwealth. I have already discussed the operation of this principle in relation to the Senate, where each state was given equal representation—first six, then ten, then twelve senators—regardless of its population. The least populous states are most favoured by these arrangements: thus Tasmania returns the same number of senators as New South Wales, which has 14 times Tasmania’s population. This was an explicit compromise that was ‘one of the terms of the federal bargain’.^[113] Its importance is such that the provision for amending the Constitution (s 128) expressly prohibits any diminution in the proportionate representation of any state in either chamber unless a majority of electors in that state approve the law. In practice, this power of veto will guarantee equal state representation in the Senate for as long as the federation exists.

The second principle is a territorial principle, which determines representation on the basis of territory—measured for example by land area. Nowhere does the Constitution endorse such a principle, but the notion emerges from time to time in discussions on democratic representation. An illustration in the federal sphere was the parliamentary debate on whether, and on what terms, the Northern Territory and the ACT should be granted representation in the Senate. In introducing the Bill, the Minister argued that the territories were ‘significant parts of Australia’ and that the Northern Territory ‘comprising an area of 520,280 square miles, is a vital part of Australia which should be represented in the Senate.’^[114] Similar arguments are sometimes heard as justification for malapportionment or ‘rural weighting’ in electoral systems in which rural electorates are favoured in comparison with their urban counterparts—in essence, a triumph of geography over population.^[115]

The third principle is a population principle, which determines representation on the basis of the number of people residing within a geographically bounded area. In contrast to its stance in relation to the Senate, the Constitution expressly adopts a population principle in determining representation in the House of Representatives: under s 24 ‘the number of members chosen in the several States’ must be ‘in proportion to the respective numbers of their people’. This was honoured in the first Parliament, where the original 75 members were allocated between the six states in accordance with their populations determined by a conference of statisticians in 1900.^[116] For subsequent elections the population principle was to underpin the quota mechanism set out in s 24, and in this way electorates would be lost in some states and gained in others as their relative shares of the total population changed over time.

There are, however, a number of factors that distort a pure application of the population principle in allocating lower house seats. First, every original state is guaranteed a minimum of five seats because a lesser number was thought at federation to be ‘insignificant representation’.^[117] In 1901, this elevated Tasmania’s representation from four seats to five, and Western Australia’s representation from three seats to five. To this day, this guarantee continues to favour Tasmania, while Western Australia’s expanding population now earns it more seats than the guaranteed minimum. Secondly, the representation of territories in the House is entirely at Parliament’s discretion (s 122) and this power has been used to set the number of territorial seats at levels that are sometimes at variance with the quota mechanism. For example, the Northern Territory was guaranteed two seats in the 2004 federal election after it fell short of its second quota by a meagre margin of 295 people.^[118] Thirdly, the quota mechanism has to confront the problem of ‘lumpiness’ that arises from dividing the Australian population into a discrete and relatively small number of electorates, while still observing state and territorial boundaries in the allocation of seats.^[119] The problem is addressed in s 24 by rounding fractional quotas up or down, and thereby giving greater or lesser representation to states and territories than would have been justified by the population principle alone.

The impact of the population principle, and its qualifications, can be seen by reference to the determination of the Electoral Commissioner with respect to the 2010 federal election.^[120] Using the latest official statistics to ascertain the Australian population (as required by the Commonwealth Electoral Act 1918 (Cth)) and the quota mechanism (as set out in the Constitution), the Commissioner determined that 150 seats will be contested in the 2010 election, with New South Wales losing a seat to Queensland in the process.^[121] The strength of the population principle in the allocation of lower house seats can be seen in Figure 9 by comparing the proportion of the Australian population in each state and territory in 2009 (column B) to the proportion of electorates allocated to each jurisdiction under the electoral determination (column D). The resulting ratio (column E) indicates over representation to the extent that the ratio is greater than 1.0 and under representation to the extent that the ratio is less than 1.0. For five of the six states, there is a high degree of fidelity to the population principle, with only small variations from the ‘ideal’, but there are marked variations for the other three jurisdictions (column F). Tasmania is over represented by 43% due to the constitutional guarantee of five seats; the Northern Territory is over represented by 29.6% due to the lumpiness problem, and the ACT is under represented by 17.3%, also due to the lumpiness problem. However, the latter three jurisdictions account for only 9 of the 150 electorates and therefore the problem is not large in absolute terms.

Figure 9: Relationship between population and federal electorates, 2009

Finally, it should be noted that the issue of population distribution has had legal relevance beyond the text of the Constitution in the unsuccessful attempts to establish an implied constitutional guarantee of ‘one vote one value’. Inspired by United States jurisprudence requiring equality of population in electoral divisions,^[122] the matter was first agitated in 1975 in McKinlay’s Case, where it was argued that there was an implication in s 24 of the Australian Constitution that federal electoral divisions should have equal numbers of people or, alternatively, electors.^[123] A majority of the High Court rejected that proposition, but it was canvassed again nearly 20 years later in McGinty’s Case.^[124]

On this occasion the challenge was to state electoral laws that created significant differences in the number of electors in metropolitan and non-metropolitan electoral districts in the lower house of the Western Australian Parliament. The greatest disparity was between the rural electorate of Ashburton (8,587 electors) and the urban electorate of Wanneroo (35,529 electors)—a four-fold difference.^[125] It was argued, among other things, that the Australian Constitution implied a system of representative democracy with approximate equality of voting power. However, a majority of the High Court declined to overturn McKinlay’s Case, holding that the Constitution did not require ‘one vote one value’. As Gummow J remarked (quoting John Stuart Mill), it is necessary to distinguish between the essence of representative government and the particular forms in which the idea has been clothed by accidents of history or current fashion.^[126] In the result, the Court’s decision leaves the choice of electoral arrangements to the democratic process itself through laws made by elected representatives. In large measure the most egregious aspects of rural weighting have now been dismantled.^[127] Nevertheless, the judgment is a judicial rejection of the notion that population distribution has a broader relevance in determining electoral arrangements under the Australian Constitution.

Part 2.1 of this article noted that demographers are interested in population data because demographic analysis is predicated on the existence of relevant, timely and accurate statistical information. Population data potentially come from four sources. The first source is a population census, which is a complete enumeration of a population at a specified point in time. Examples of censuses stretch back to antiquity but their modern incarnation—the periodic census—dates from 17^th century Europe.^[128] A second source of population data is a register of vital events, which typically captures details about births, deaths, marriages and divorces. These began as parish records of religious events (christenings and funerals) but they are now maintained on a secular basis by government agencies throughout the world. A third source is administrative collections maintained by government departments primarily for other purposes (immigration control, healthcare claims, school enrolments), which contain useful information about demographic attributes of a population, especially when analysed in conjunction with data derived from other sources. Finally, many European countries keep population registers, which are complete and continuous records of vital events experienced by every individual belonging to the population. Population registers have never been adopted in Australia due to ongoing concerns about privacy and data protection.^[129]

Modern constitutions have recognised the importance of population data to some degree. The drafters of the United States Constitution anticipated the need for population statistics and mandated that a census of the American people be conducted within three years of the first meeting of Congress, and every ten years thereafter.^[130] Writing in the Federalist Papers in 1788, Alexander Hamilton and James Madison described the purpose of the census as being (i) to readjust, from time to time, the number of congressional representatives elected from each state to the number of inhabitants of each state, and (ii) to augment the total number of representatives as the population grew.^[131] The constitutional drafters thus demonstrated a concern for both the size of the population and its spatial distribution. A decennial census was the agreed means of adapting the principles of representative democracy to a dynamic population, although the drafters shied away from estimating the future size of the population—leaving this to those with a more ‘prophetic spirit’.^[132] The use of census data for recalibrating congressional membership was an important and forward thinking development, yet by today’s standards it was also quite limited. The use of population data for broader social purposes, such as planning the delivery of government services, still had to await the Keynesian economic revolution and the advent of the welfare state.

The Australian Constitution did not follow the United States model in mandating a periodic census. Instead, the Parliament was given power in s 51(xi) to make laws with respect to ‘census and statistics’. This clause was borrowed from the 1867 Canadian Constitution, with the important difference that the Canadian Parliament was granted exclusive power over this subject matter whereas the power of the Australian Parliament was concurrent with that of the states.^[133] Despite the absence of an express mandate for a periodic census, the Australian Constitution might be taken to imply a similar requirement. On three occasions the Constitution makes reference to using ‘the latest statistics of the Commonwealth’—twice in determining the composition of the House of Representatives (s 24) and once in determining the manner in which the Commonwealth may take over state debts (s 105). A plausible implication is that the Commonwealth is obliged to procure reasonably up-to-date population data for these identified purposes, whether it be from censuses or other sources.

By the time of federation, the idea of periodic censuses was well accepted in Australia. The colonies had conducted sporadic ‘population musters’ from the early days of European settlement. Regular censuses had commenced in New South Wales in 1828 and had spread to most other colonies by the 1850s.^[134] The first simultaneous census of the Australian colonies was taken in 1881 as part of a census of the British Empire, and this process was repeated in 1891 and again in 1901. It is not surprising then that the new Australian Parliament moved relatively quickly to establish a national statistical agency—the Commonwealth Bureau of Census and Statistics, later renamed the Australian Bureau of Statistics (ABS)—and that this agency undertook its first national census in 1911.^[135] Other censuses followed at ten-year intervals (interrupted by two world wars) until 1961, and at five-year intervals thereafter.

The census is not the only source of population data about which the Constitution makes provision. The Commonwealth has also assumed responsibility for collecting information on international migration flows for administrative purposes, based on the incoming and outgoing passenger cards that are familiar to Australian travellers. The collection of this data is supported by the power over immigration and emigration (s 51(xxvii)) and the power over census and statistics (s 51(xi)), while the simultaneous collection of additional information at the border is supported by the grant of legislative power over quarantine (s 51(ix)) and customs (s 90).

By contrast, primary data on births and deaths is collected not by the Commonwealth but by state and territory governments through their registries of births, deaths and marriages.^[136] This can create difficulties in compiling national statistics because of variations in the policies and practices of different registries. For example, the lag between the occurrence of a birth and its registration is nearly three times longer in Queensland than in New South Wales, leading to discrepancies in national data on annual births.^[137] It is likely that the federal legislative power over census and statistics would provide constitutional support for a federal law covering the field of vital registrations, thus excluding state laws on that topic. However, the Commonwealth has not sought to do this. The ABS thus continues to perform a coordinating function among state and territory registries, which is facilitated by formal administrative arrangements between federal and state executives.^[138]

There is one further feature of the Australian Constitution that has significance for population data. As originally enacted, the Constitution excluded Indigenous Australians from being counted in estimates of the Australian population. Section 127 provided that ‘in reckoning the numbers of the people of the Commonwealth, or of a State or other part of the Commonwealth, Aboriginal natives shall not be counted’. The purpose of the provision is obscure as there was little discussion about it during the conventions leading up to its inclusion in the Constitution. One explanation is that s 127 was thought relevant to calculating the per capita contribution of each state to federal expenditure, and to determining the distribution of surplus revenues to the states, based on their respective populations.^[139] Another explanation is that s 127 was a means of discounting Indigenous Australians in the process of allocating House of Representatives seats among the states.^[140] In support of the latter view, ss 24 and 25 of the Constitution expressly use the language of s 127 in referring to people of the Commonwealth and people of a state in determining lower house seats.

Consistently with s 127, every census taken from the time of federation until 1967 (when s 127 was repealed) included a question on race for the purpose of excluding ‘Aboriginal natives’ from official population counts. In practice, this uncertain term was treated as referring to any person with more than 50 per cent Aboriginal blood, while Torres Strait Islanders were given differing treatment over the years.^[141] In 1967, s 127 was struck from the Constitution in the most successful referendum in Australian history. Commentators have noted the gulf between the rhetoric that has come to treat the 1967 referendum as a turning point in Aboriginal affairs and the reality of the meagre achievements that followed it.^[142] Yet, on the more limited topic of population statistics, the referendum successfully removed an impediment to counting Indigenous Australians as part of the Australian population. Since 1971, respondents to census questionnaires have not been asked to state their degree of Aboriginal descent but only whether or not they are of Aboriginal or Torres Strait Islander origin. The rapidly rising number of people who now identify themselves in this way is thought to be the product of changing social attitudes, improvements in census coverage, and a broader definition of aboriginality.^[143]

There has been a burgeoning interest among legal scholars in issues of comparative constitutional design. In recent years this has been prompted by the transition of authoritarian regimes to democracy, and communist economies to market economies—largely in Eastern Europe. These changes followed significant constitutional reforms that took place during post-war reconstruction in the 1940s and decolonisation in the 1950s and 1960s.^[144] Yet most of the literature on constitutional design is focussed on a narrow set of interrelated questions about the constitutional protection of human rights—what they should be, how they should be enforced, and how they should be interpreted.^[145]

These questions have been especially pertinent in ‘divided societies’, which Sujit Choudhry has aptly described as societies marked by ethnic, linguistic, religious or cultural diversity, where that diversity is politically salient because it is a basis for political mobilisation.^[146] Demographic issues are relevant to constitutional design in these societies because population composition may provide the basis of the distribution of power in its various forms. These arrangements may vary from the informal—such as Canada’s constitutional convention that francophone Quebec will have adequate representation on the Supreme Court^[147]—to the highly formal, such as Fiji’s constitutionally enshrined reservation of a specified number of House of Representatives seats for Fijians, Indians, Rotumans and others.^[148] However, Australia is not a divided society in this sense.

When issues of rights protection in divided societies are set aside, there is very little to be found in the literature on constitutional design that addresses core issues of population change. Accordingly, I approach the issue from basic principles, drawing together the Australian examples described above and extrapolating from them to suggest broader principles of general application. The role that constitutions can play—positively and negatively—in addressing population issues can conveniently be discussed under three headings: the foundation principles established by the constitution with respect to population issues; the capacity of constitutional rules to adapt to population dynamics; and the degree to which those principles are entrenched for future generations. These are discussed below.

Long term changes in the demographic characteristics of a population are merely the aggregate effects of a multitude of personal experiences. Individual decisions to have children, to adopt healthy or unhealthy lifestyles, or to migrate, lie at the heart of the three demographic processes that shape a population over time—fertility, mortality and migration. Constitutions potentially have a role to play in protecting the capacity of individuals to make decisions about these demographic processes.

The Australian Constitution provides limited safeguards for this type of demographic decision making. In Part 3.3 we noted that the freedom of ‘interstate intercourse’ in s 92 has been interpreted as a guarantee of free movement from state to state, thus underpinning decisions about migration within the federation, particularly when combined with the prohibition against discrimination against out-of-state residents in s 117. Beyond this, the Constitution says very little. There is an ongoing debate about whether Australia should adopt a Charter of Rights, what rights should be protected, and whether this should take legislative or constitutional form.^[149] Without canvassing these larger questions, a population perspective on these issues suggests there should be legal protections for rights that have demographic relevance. These would include a right to have a family; a right to adequate health care; a right of nationals to reside in a country or leave it; and perhaps a right to political equality, reflected in the ‘one vote one value’ slogan. A number of these rights are already reflected at the state and territory level in the Charter of Rights adopted in Victoria and the ACT.^[150]

The absence of these constitutional protections does not mean that Australians do not already enjoy some of these rights to a significant extent. However, history provides many illustrations of population policies that might have been avoided by a constitutional commitment to individual rights. Examples include the one-child policy in China, compulsory sterilisation in India,^[151] and the mass expulsion of religious or ethnic minorities from various countries.

Beyond individual rights, a constitution might also craft a set of principles that are important for the protection of institutions of government in the face of persistent population change. Examples from the Australian Constitution include the principle that states shall be equally represented in the Senate (s 7); that the lower house shall be twice the size of the upper house (s 24); and that electorates shall not be formed out of parts of different states (s 29). These principles may constrain the capacity of governments to respond to population change, yet their importance lies in the articulation of competing constitutional values, and in the establishment of a hierarchy for resolving conflicts between values. For example, differential rates of growth of state populations since federation have not seriously challenged the federal value implicit in equal state representation in the Senate, despite moving progressively further away from a ‘population’ ideal.

With the benefit of hindsight, contemporary demography provides quite sophisticated tools for describing and analysing population changes in the past, and for theorising about the modernising processes that have generated long term population trends such as the ‘demographic transition’ and the ‘epidemiological transition’. Yet the capacity to predict population change far into the future is still very limited. This reflects the unpredictability of human behaviour, the complexity of population dynamics, and the unknowable population impacts of processes such as climate change and events such as disease and war. This uncertainty is reflected in the very broad range of population projections produced by most national statistical agencies, such as the projection that Australia’s population in 2101 will be 33.7–62.2 million (see Part 2.2 above). The drafters of the Australian Constitution can hardly be criticised for their failure to anticipate, in the 1890s, the significant demographic changes that occurred in Australia in the first century of federation—all the more so because the data and methods at their disposal were so rudimentary. What is predictable, however, is that change will be a persistent feature of any country’s population future. Constitutions should therefore be designed so that government policy can adapt to mutable population dynamics.

The constitutional provisions considered in this article suggest three types of adaptive capacity, which I shall call self-adaptive, potentially adaptive, and non-adaptive. A constitutional rule is self-adaptive if it contains an automatic internal mechanism for responding to population change. An example is s 24 of the Australian Constitution, which provides that the ‘number of members [of the House of Representatives] chosen in the several States shall be in proportion to the respective numbers of their people’. As the spatial distribution of the population changes, so the allocation of seats between the states also changes by constitutional fiat, subject only to the qualifications discussed in Part 3.3. Similarly, the requirement in the United States Constitution that a census shall be held every ten years falls into the class of self-adaptive rules: population changes are automatically recorded because a process of periodic enumeration is mandated by the constitution.^[152] However, self-adaptive rules are quite uncommon.

A constitutional rule is potentially adaptive if it relates to a population issue but leaves it to the legislature, or another branch of government, to make whatever changes are deemed necessary from time to time to address that issue. The Australian Constitution contains many examples of this type of rule. Prominent examples include the power of Parliament to increase or diminish the number of members in either House of Parliament (ss 7, 27); to make provision for representation of territories in either House of Parliament (s 122); and to make laws with respect to immigration and emigration, taxation, family and health benefits, and census and statistics (s 51). In each case, the constitution embodies adaptive capacity, but the adaptation itself takes place through a separate governmental measure. Not all such measures will be seen as responding appropriately to population change since it is commonly thought that ‘our Statute Books are filled with monuments of the deficient wisdom of our legislators’.^[153] Nevertheless, this approach creates a degree of flexibility that would be absent if the population policy were itself locked into the constitution.

A constitutional rule is non-adaptive if it introduces rigidities that hinder the capacity of government to respond to changes in the size, composition, or spatial distribution of the population. The Australian Constitution contains three notable examples of non-adaptive rules: one of these dates from federation and persists to this day; the second dates from federation but was removed by referendum in 1967; and the third was introduced by a short-sighted amendment in 1977. Specifically, the guarantee that every original state shall have five seats in the House of Representatives (s 24) has inhibited the capacity of the Constitution to respond to the changing spatial distribution of the Australian population. As a result, Tasmania is over-represented in the lower house, while every other state is correspondingly under-represented (see Figure 9).^[154] The prohibition on counting ‘Aboriginal natives’ among the people of the Commonwealth (s 127) prevented Indigenous Australians, until 1967, from being included fully in the federal arrangements for democratic representation. And thirdly, the requirement that the term of appointment of federal judges expires when they reach 70 years of age (s 72) has been unable to take into account the declining mortality of the Australian population and the trend towards later retirement as the population ages.

Some constitutional provisions, such as the mechanism for constitutional amendment in s 128, provide a curious blend of self-adaptive, potentially adaptive and non-adaptive rules. As noted above, s 128 is an exercise in direct democracy by which proposals for constitutional change require approval by a double majority of electors. The requirement that the proposed law be approved by a majority of all electors voting provides the only significant occasion on which Australians vote as a single electorate.^[155] Its practical effect is to establish a self-adaptive rule that automatically responds to changes in population size, in circumstances that make spatial distribution of the population irrelevant. By contrast, the requirement that the proposed law be approved by a majority of electors voting in a majority of states is a non-adaptive rule because it is insensitive to population redistribution between states and between states and territories. Any three states can overturn a proposed amendment, even if their population share is small. Thus, since federation, five referenda have failed despite carrying more than 50 per cent of the popular vote. Finally, one aspect of s 128 is potentially adaptive. In 1901 the referendum procedure encompassed only electors in the states, but since 1977 it has also included electors in each territory that has representation in the House of Representatives. Since Parliament determines whether, and to what extent, territories are represented in the lower house (s 122), it is Parliament that ultimately determines the extent to which Territorians may participate in the democratic process of constitutional change.

A third attribute of constitutional rules relevant to population dynamics is the degree to which those rules are entrenched. All constitutional rules are entrenched in the sense that special requirements or procedures must be followed to validate a change. In the Australian Constitution, s 128 requires any proposed constitutional amendment to be passed by the Parliament and then approved by the Australian people in a double majority voting procedure, namely, a majority of voters overall and a majority of voters in a majority of states. These are high hurdles, which have been overcome on only eight occasions in the 44 attempts since federation.

Yet on some issues the Constitution imposes an even higher barrier to change. Any constitutional alteration that would diminish the proportionate representation of any state in either House of the Parliament, or the minimum number of representatives of a state in the House of Representative, must be approved by a majority of the electors voting in that State (s 128). As noted in Part 3.3, this gives a right of veto to any state that is adversely affected by the proposed change, and in practice this makes the constitutional provisions immutable.

This deeper level of entrenchment is relevant to one particular type of population issue, namely, the spatial arrangements that bear on representative democracy in a federal constitutional system. In effect, it prevents any change to the population principle in electing members to the House of Representatives from the states (s 24), or any change to the federal principle in electing members to the Senate from the states (s 7). It also prevents any alteration to the guarantee that every original state shall have a minimum of five seats in the House of Representatives. Ironically, therefore, this additional entrenchment protects from interference both the one self-adaptive provision in the Constitution (the allocation of lower house seats to states based on their population) and one of the non-adaptive provisions in the Constitution (the guarantee of at least five seats to each state regardless of population).

Given the high barrier to constitutional change imposed by the requirement of a double majority in s 128, it seems fanciful today to claim that further safeguards are needed to protect these features of representative democracy. Yet it does attest to the difficulties faced by the drafters of the constitution in the 1890s in establishing a federal union from sometimes reluctant states. Having struck a federal bargain as a result of careful deliberation and political compromise, the states went to great lengths to ensure the bargain could not be easily unravelled by rapid amendment after federation.^[156]

The concern that public lawyers have for vindicating rights in individual cases—Tennyson’s ‘wilderness of single instances’—is laudable and necessary, but it is not sufficient to meet the challenges of modern constitutional governance. Constitutions should also have regard to the population, conceived as an organic entity that is greater than the sum of the individuals who comprise it. In 1901 it was ‘the people’ who ‘agreed to unite in one indissoluble Federal Commonwealth’, and it is for ‘the people’ that the Constitution must continue to be an effective instrument of government.

Through the lens of a demographer, this article has identified some of the most important changes in Australia’s population since federation—in size, composition, and spatial distribution, as well as in the components of change that are the principal drivers of demographic dynamics. The article described the near six-fold increase in population since 1901; the decline in fertility and mortality; the substantial growth in net overseas migration in the post-war period; the increasing ageing and ‘feminisation’ of the population; and the spatial redistribution towards states in the north and west of the continent.

These demographic changes have become the focus of much recent public debate. There appears to be greater consciousness of population issues now than at any time in recent history, and this is reflected in a range of government initiatives on topics such as the baby bonus, the Future Fund, and increasing the qualifying age for the age pension. As further evidence of this concern, in April 2010 the Australian Government created a new office of the Minister for Population to help guide the development of policies to meet Australia’s future population needs. However, effective governmental responses to demographic challenges depend on the existence of sound constitutional foundations. Are there appropriate powers to respond to current population challenges and to shape future population dynamics? Are those powers at a level that permits effective responses within a federal system of government?

The Australian Constitution reveals a degree of population consciousness that is perhaps surprising for the era in which it was drafted. It contains one provision that adapts automatically to changes in the spatial distribution of the population, and other provisions confer a range of powers on the federal Parliament which give it the capacity to respond to population issues by altering policy settings as the need arises. Yet in other respects the Constitution—as drafted or later amended—shows a degree of population short-sightedness that has locked Australia into positions that are increasingly at odds with secular population trends.

It has been said that a written constitution ‘is designed to be an anchor in the past. It creates rules that bind until a supermajority of the living changes them.’^[157] In this way a constitution provides stability against fads and fashions, and the tyranny of the majority. Similarly, Quick and Garran described the demanding alteration provision in s 128 of the Australian Constitution as a safeguard ‘to prevent change being made in haste or by stealth, to encourage public discussion and to delay change until there is strong evidence that it is desirable, irresistible, and inevitable’.^[158]

Real-world populations are also anchored in the past because incremental changes in human behaviour take generations to work their way through a population whose individual members have a life span that potentially exceeds 120 years.^[159] Yet, over the long term, populations are anything but static: modernising social processes have resulted in profound transformations in all Western populations over the past century, including Australia’s. If constitutions are to provide a solid foundation for the governance of a people far into the future, they need to have population dynamics firmly in mind in their original design and in their subsequent evolution.

In 1967 Geoffrey Sawer—contemplating the lacklustre success of constitutional referenda to that point—described Australia as a constitutionally ‘frozen continent’.^[160] Surprisingly, of the eight successful referenda since 1901, three have had demographic significance.^[161] This is not a bad record overall, although only two of these changes can be regarded as positive from a population perspective. This suggests that the barrier to further constitutional advancement does not lie in the mechanism for amending the constitution but in our collective failure to imagine constitutions as providing supportive structures for the evolution of entire populations. As the Australian population marches inexorably into the 21^st century, it is time to reassess whether the constitution crafted in the last decade of the 19^th century is capable of meeting the demographic challenges of the future.

[1] Aylmer’s Field (1864). See Christopher Ricks (ed), The Poems of Tennyson (1969) 1172.

[2] John Caldwell, 'History of Demography' in Paul Demeny and Geoffrey McNicoll (eds), Encyclopedia of Population (2003) 216.

[3] Achille Guillard, Éléments de Statistique Humaine ou Démographie Comparée (1855); John Weeks, Population: An Introduction to Concepts and Issues (9th ed, 2005) 4.

[5] Peter Lasett (ed), The Earliest Classics: John Graunt and Gregory King, Pioneers of Demography Series (1973) 1, cited in Rowland, n 4, 15.

[7] Thomas Malthus, An Essay on the Principle of Population, Patricia James (ed) (first published in 1803, reprinted 1989).

[8] Paul Ehrlich, The Population Bomb (1968). See also Paul Ehrlich and Anne Ehrlich, 'The Population Bomb Revisited' (2009) 1(3) Electronic Journal of Sustainable Development 1.

[9] See Jacob Siegel and David Swanson (eds), The Methods and Materials of Demography (2nd ed, 2004).

[13] H. Brown and A. Hall, 'Australian Demographic Databank: Volume I Recorded Vital Statistics 1921-1976' (Australian National University, 1978); S. Krishnamoorthy and B. Derrick, 'Australian Demographic Databank: Volume III Recorded Vital Statistics, Population Estimates and Demographic Rates 1976-1981' (Australian National University, 1983); Australian National University, 'Australian Demographic Databank, version 3.2b' (Australian National University, 2005).

[14] John Caldwell, Bruce Missingham and Jeff Marck, 'The Population of Oceania in the Second Millennium' (Paper presented at the Conference on The History of World Population in the Second Millennium, Florence, 2001).

[15] For historical accounts see W. Borrie, The European Peopling of Australasia: A Demographic History 1788-1988 (1994); John Caldwell, 'Population' in Wray Vamplew (ed), Australians: Historical Statistics, Australians: A Historical Library (1987) 23.

[16] Steve Vizard, Hugh Martin and Tim Watts (eds), Australia's Population Challenge (2003).

[17] Barney Foran and Franzi Poldy, 'Future Dilemmas: Options to 2050 for Australia's Population, Technology, Resources and Environment' (CSIRO Sustainable Ecosystems, 2002); Australian Treasury, 'Intergenerational Report 2002-03' (Australian Government, 2002); Australian Treasury, 'Intergenerational Report 2007' (Australian Government, 2007); Australian Treasury, 'Intergenerational Report 2010: Australia to 2050: Future Challenges' (Australian Government, 2010).

[19] Australian Bureau of Statistics, 'Australian Demographic Statistics, Cat. No. 3101.0' (Australian Bureau of Statistics, 2009) Table 9.

[20] Australian Bureau of Statistics, 'Population Projections, Cat. No. 3222.0' (Australian Bureau of Statistics, 2008).

[22] John Bongaarts, 'A Framework for Analyzing the Proximate Determinants of Fertility' (1978) 4(1) Population and Development Review 105; John Stover, 'Revising the Proximate Determinants of Fertility Framework: What Have We Learned in the Past 20 years?' (1998) 29(3) Studies in Family Planning 255.

[23] Carl Haub, 'Fertility Measurement' in Paul Demeny and Geoffrey McNicoll (eds), Encyclopedia of Population (2003) 420.

[24] Australian Bureau of Statistics, 'Australian Demographic Statistics, Cat. No. 3101.0' (Australian Bureau of Statistics, 2009) Table 13.

[25] Ansley Coale, 'Factors Associated with the Development of Low Fertility: An Historic Summary' (Paper presented at the World Population Conference, Belgrade, 1965); W.C. Robinson, 'Another Look at the Hutterites and Natural Fertility' (1986) 33(1-2) Social Biology 65.

[26] Gordon Carmichael and Peter McDonald, 'Fertility Trends and Differentials' in S.-E. Khoo and P. McDonald (eds), The Transformation of Australia's Population 1970-2030 (2003) 40.

[27] James Riley, 'Estimates of Regional and Global Life Expectancy, 1800-2001' (2005) 31 Population and Development Review 537; John Bongaarts, 'How Long Will We Live?' (2006) 32 Population and Development Review 605.

[28] Abdel Omran, 'The Epidemiological Transition: A Theory of the Epidemiology of Population Change' (1971) 49 Milbank Memorial Fund Quarterly 509.

[29] L. Reed and M. Merrell, 'A short method for constructing an abridged life table' (1939) 30(2) American Journal of Hygiene 33.

[30] Jay Olshansky and Brian Ault, 'The Fourth Stage of the Epidemiologic Transition: The Age of Delayed Degenerative Diseases' (1986) 64(3) Milbank Quarterly 355.

[31] Heather Booth, 'The changing dimensions of mortality' in S.-E. Khoo and P. McDonald (eds), The Transformation of Australia's Population 1970-2030 (2003) 104, 104-105.

[32] Jane Doulman and David Lee, Every Assistance and Protection: A History of the Australian Passport (2008) 13-14.

[33] Freda Hawkins, Critical Years in Immigration: Canada and Australia Compared (2nd ed, 1991) 8-16.

[34] Bob Birrell, 'The management of immigration: patterns of reform' in S.-E. Khoo and P. McDonald (eds), The Transformation of Australia's Population 1970-2030 (2003) 129, 132.

[35] Borrie, n 15, 181.

[36] Donald Rowland, 'An Ageing Population: Emergence of a New Stage of Life?' in S.-E. Khoo and P. McDonald (eds), The Transformation of Australia's Population 1970-2030 (2003) 238, 239; Rebecca Kippen and Peter McDonald, 'Australia's Population in 2000: The Way We Are and the Ways We Might have Been' (2000) 8(3) People and Place 10.
[37] Australian Treasury, Intergenerational Report 2010, n 17, Table 1.4.

[38] Frank Hobbs, 'Age and sex composition' in Jacob Siegel and David Swanson (eds), The Methods and Materials of Demography (2nd ed, 2004) 125, 130.

[39] Caldwell, n 15.

[40] Graeme Hugo, 'Changing Patterns of Population Distribution in Australia' (2002) (September) Journal of Population Research and New Zealand Population Review (Joint Special Issue) 1, 1.

[41] Peter Newton and Martin Bell (eds), Population Shift: Mobility and Change in Australia (1996); Ian Burnley and Peter Murphy, Sea Change: Movement from Metropolitan to Arcadian Australia (2004).

[42] Hugo, n 40, 2.

[43] An example of the last approach is the role of demography in allocating Commonwealth grants to the states under s 96 of the Constitution. This takes place through the process of fiscal equalisation using population shares as a distribution baseline: Commonwealth Grants Commission, 'The Commonwealth Grants Commission: The Last 25 Years' (Australian Government, 2008) 3.

[44] Australian Electoral Commission, '1999 Referendum Report and Statistics' (Australian Electoral Commission, 1999).

[45] An enumeration of the population in 1790 revealed a population of 3.9 million, of whom 3.6 million resided in the 13 original states: University of Virginia Geospatial and Statistical Data Center, Historical Census Browser (2004) <http://mapserver.lib.virginia.edu/> at 20 January 2010 .
[46] United States Constitution, Art IV, s 3.

[47] Thomas Jefferson had formulated a radically different plan for territorial expansion, which was adopted in modified form by ordinance in 1784. The Confederation Congress rejected this plan when it adopted the Northwest Ordinance in 1787. The 1787 ordinance was ratified by the US Congress in 1789. See Peter Onuf, Statehood and Union: A History of the Nortwest Ordinance (1987); Bill Hubbard, American Boundaries: The Nation, The States, The Rectangular Survey (2009) 101-179.

[48] M.H. McLelland, 'Colonial and State Boundaries in Australia' (1971) 45 Australian Law Journal 671.
[49] Robert Dahl and Edward Tufte, Size and Democracy (1973) 4-16.

[50] One attribute of direct democracy remained in the procedure for constitutional amendment through popular referendum (s 128). However, this was tempered by the requirement that the proposal had to be initiated by the Parliament: Cheryl Saunders, 'The Parliament as Partner: A Century of Constitutional Review' in G Lindell and R Bennett (eds), Parliament: The Vision in Hindsight (2001) 454, 463-464.

[51] This differed from the ‘quota’ determined under s 24 (which was calculated as 51,635 persons) because the quota does not make adjustments for fractions or for the minimum seat entitlement of the states. See John Quick and Robert Garran, The Annotated Constitution of the Australian Commonwealth (1995 reprint ed, 1901) 450, 459.
[52] John Williams, The Australian Constitution: A Documentary History (2005) 141.
[53] United States Constitution Art I, s 2.
[54] Quick and Garran, n 51, 451.

[55] Brian O'Neill and Wolfgang Lutz, 'Population Projections and Forecasts' in Paul Demeny and Geoffrey McNicoll (eds), Encyclopedia of Population (2003) 808, 808. The innovation was the cohort component method of projection, which ‘survived’ each age group forward to the next age group according to plausible assumptions about each of the three components of change: fertility, mortality and migration. The method had been first proposed in 1895.
[56] Quick and Garran, n 51, 452.

[57] Commonwealth, Parliamentary Debates, House of Representatives, 16 April 1948, 962 (Dr Evatt, Attorney-General).

[58] Commonwealth, Parliamentary Debates, Senate, 16 November 1983, 2608 (Senator Grimes, Minister for Social Security), 2609.

[59] A proportionate increase in population and elected representatives would be shown in Figure 7 by a horizontal line that intersects the vertical axis at an agreed value, say 50,000 persons. This would be the equivalent of the ‘fixed quota’ rejected by the constitutional drafters in the 1890s.
[60] Dahl and Tufte, n 49, 80.

[61] A similar view has been expressed in the United States, where ‘one House was viewed as representing the people, and the other, states’: Immigration and Naturalization Service v Chadha, [1983] USSC 143; 462 US 919, 950 (1983) (Burger CJ).

[62] Quick and Garran, n 51, 414. This was reinforced by the subtle distinction that senators were to be ‘directly chosen by the people of the State’, whereas members of the House of Representatives were to be ‘directly chosen by the people of the Commonwealth’: Australian Constitution ss 7, 24.

[63] Representation Act 1948 (Cth); Representation Act 1983 (Cth). The changes took effect from 1950 and 1984, respectively.

[64] Senate (Representation of Territories) Act 1973 (Cth). The legislation was upheld in Western Australia v Commonwealth [1975] HCA 46; (1975) 134 CLR 201 and again in Queensland v Commonwealth [1977] HCA 60; (1977) 139 CLR 585.

[65] Commonwealth, Parliamentary Debates, House of Representatives, 22 May 1973, 2425 (Mr Daly, Minister for Services and Property), 2426.

[66] The same latitude applies to the representation of any new state admitted to the Commonwealth: s 121. This will be relevant to the Northern Territory if it becomes a state.
[67] See the discussion on freedom of intercourse (s 92) in Part 3.3 below.

[68] Helen Irving, To Constitute a Nation: A Cultural History of Australia's Constitution (1997) 114-115.
[69] Doulman and Lee, n 32, 37–50.

[70] Kathryn Cronin, 'A Culture of Control: An Overview of Immigration Policy Making' in James Jupp and Marie Kabala (eds), The Politics of Australian Immigration (1993) 83, 88-89.
[71] Eric Richards, Destination Australia: Migration to Australia Since 1901 (2008) 35-39.
[72] Irving, n 68, 114.
[73] Vizard et al, n 16.

[74] International Covenant on Civil and Politial Rights, opened for signature 16 December 1966, 999 UNTS 171 (entered into force 23 March 1976) Art 12(3).

[75] Richard Plender, International Migration Law (2nd ed, 1988) 120-121.

[76] Ruth Weston et al, 'It's Not for Lack of Wanting Kids: A Report on the Fertility Decision Making Project' (Australian Institute of Family Studies, 2004) xiv.
[77] Peter McDonald, 'Reforming Family Support Policy in Australia' (2003) 11(2) People and Place 1, 3.

[78] McDonald, n 77, 3.

[79] Peter McDonald, 'Low Fertility in Australia: Evidence, Causes and Policy Responses' (2000) 8(2) People and Place 6; Willem Adema and Peter Whiteford, 'Matching Work and Family Commitments: Australian Outcomes in a Comparative Perspective' (2008) 80 Family Matters 9; Australian Productivity Commission, 'Paid Parental Leave: Support for Parents with Newborn Children' (Australian Productivity Commission, 2009).

[80] John Bongaarts and Robert Potter, Fertility, Biology, and Behavior: An Analaysis of the Proximate Determinants (1983).
[81] Irving, n 68, 93, quoting Sir John Downer in the Sydney Convention Debates, 1897.
[82] McDonald, n 77, 4.

[83] Both programs were later amended. The rebate is no longer allowed as a tax offset under taxation legislation but is paid as a transfer under the social security system.

[84] The Treasurer, Commonwealth of Australia, Budget Lock-Up Press Conference, Transcript (11 May 2004). See A New Tax System (Family Assistance) Act 1999 (Cth) s 36.

[85] In 2008-09, a bonus of $5,000 was paid in respect of 283,000 births: Department of Families Housing Community Services and Indigenous Affairs, 'Annual Report 2008-2009' (Australian Government, 2009) 161. This greatly exceeds the original cost estimate of $2.5 billion over four years: Commonwealth, Parliamentary Debates, House of Representatives, 11 May 2004, 28196 (Mr Anthony, Minister for Children and Youth Affairs), 28197.

[86] Joshua Gans and Andrew Leigh, 'Born on the First of July: An (Un)natural Experiment in Birth Timing' (2009) 93 Journal of Public Economics 246.

[87] Ross Guest, 'The Baby Bonus: A Dubious Policy Initiative' (2007) 23(1) Policy 11.

[88] John McMillan, 'The Constitutional Power of the Commonwealth in Public Health' in Australian Institute of Health Law and Ethics (ed), Public Health Law in Australia: New Perspectives (1998) 105, 107.

[89] For a review of the origins of the Australian healthcare funding, see Gavin Mooney and Richard Scotton (eds), Economics and Australian Health Policy (1998).

[90] Australian Government, 'Budget Paper No 3: Australia's Federal Relations 2009-10' (Australian Government, 2009) 27, 33.

[91] Australian Government, 'A National Health and Hospitals Network for Australia's Future' (Department of Health and Ageing, 2010), 45-55.
[92] Australian Constitution s 72.

[93] Waterside Workers' Federation of Australia v J W Alexander Ltd [1918] HCA 56; (1918) 25 CLR 434 ('Alexander's Case').

[94] Commonwealth, Parliamentary Debates, House of Representatives, 16 February 1977, 147 (Mr Ellicott, Attorney-General).

[95] Commonwealth, Parliamentary Debates, House of Representatives, 17 February 1977, 203 (Mr Sinclair), 206 (Mr Bowen), 209 (Mr Hodgman), 220 (Mr Neil).

[96] Michael Kirby, 'Sir Edward McTiernan: A Centenary Reflection' (1991) 20 Federal Law Review 165, 181.

[97] Australian Parliament, 'Report on Retiring Age for Commonwealth Judges, Parliamentary Paper No. 414/1976' (Standing Committee on Constitutional and Legal Affairs, 1977) [50].

[98] The ‘no case’ used its allocated space to counter a proposal to require simultaneous elections in the House of Representatives and the Senate: Australian Electoral Commission, 'Australian Referendums 1906-1999, CD-Rom' (Australian Electoral Commission, 2000).

[99] Family Law Amendment Act 1977 (Cth) s 4. The law giving effect to the referendum had come into force on 29 July 1977: Constitution Alteration (Retirement of Judges) Act 1977 (Cth).

[100] Family Law Amendment Act (No 2) 1991 (Cth) s 3.

[101] The life expectancy of a newborn in a given year is invariably less than the expected life span of an individual who has attained a mature age in that year because the latter person has already survived many of life’s vicissitudes, including one of the most hazardous years of a person’s life, namely, the first year.
[102] Australian Institute of Health and Welfare, 'Australia's Health 2008' (AIHW, 2008) 293.
[103] Australian Institute of Health and Welfare, n 102, 293-294.
[104] Judges' Pensions Act 1968 (Cth) ss 6, 6A, 8.

[105] Australian Senate, 'Australia's Judicial System and the Role of Judges' (Legal and Constitutional Affairs References Committee, 2009) [4.16], [4.21]–[4.26].

[106] Age Discrimination Act 2004 (Cth). See Human Rights and Equal Opportunity Commission, 'Age Matters: A Report on Age Discrimination' (HREOC, 2000) 39-43.
[107] Australian Government, 'Budget Overview, 12 May 2009' (Australian Government, 2009), 23.

[108] Henry Burmester, 'Island Outposts of Australia' in W Bateman and M Ward (eds), Australia's Offshore Maritime Interests (1985) 54, 54. Australia’s current external territories and the dates of their establishment are: Norfolk Island (1914), Australian Antarctic Territory (1933), Ashmore and Cartier Islands (1933), Heard and McDonald Island (1953), Cocos (Keeling) Islands (1955), Christmas Island (1958) and Coral Sea Islands (1969). For a detailed history see Alan Kerr, A Federation in These Seas: An Account of the Acquisition by Australia of its External Territories, with Selected Documents (2009).

[109] R v Smithers; Ex parte Benson [1912] HCA 96; (1912) 16 CLR 99; Gratwick v Johnson [1945] HCA 7; (1945) 70 CLR 1; Cole v Whitfield [1988] HCA 18; (1988) 165 CLR 360; AMS v AIF (1999) 199 CLR 160.

[110] Northern Territory (Self-Government) Act 1978 (Cth) s 49; Australian Capital Territory (Self-Government) Act 1988 (Cth) s 69.

[111] See Street v Queensland Bar Association [1989] HCA 53; (1989) 168 CLR 461; Genevieve Ebbeck, 'The Future for Section 117 as a Constitutional Guarantee' (1993) 4 Public Law Review 89; Michael Mathieson, 'Section 117 of the Constitution: The Unfinished Rehabilitation' (1999) 27 Federal Law Review 393.
[112] Commonwealth of Australia Constitution Act 1900 (Imp), Preamble.

[113] Quick and Garran, n 51, 414; Murray Gleeson, 'The Shape of Representative Democracy' [2001] MonashULawRw 1; (2001) 27 Monash University Law Review 1, 6.

[114] Commonwealth, Parliamentary Debates, House of Representatives, 22 May 1973, 2425 (Mr Daly, Minister for Services and Property), 2427, 2428.

[115] Murray Goot, 'Electoral Systems' in Don Aitkin (ed), Surveys of Australian Political Science (1985) 179, 210-215; Murray Goot, 'Electoral Redistribution in Australia: A Comparative Analysis' (New South Wales Parliamentary Library, 1986) 18.
[116] Australian Constitution s 26; Quick and Garran, n 51, 456-460.
[117] Australian Constitution s 24; Quick and Garran, n 51, 455.

[118] Australian Parliament, 'Territory Representation, Parliamentary Paper No. 712/2003' (Joint Standing Committee on Electoral Matters, 2003); Commonwealth Electoral Amendment (Representation in the House of Representatives) Act 2004 (Cth) sch, cl 11. In the 2007 federal election the Northern Territory obtained its second quota in its own right on the basis of its population.

[119] Australian Constitution s 29 provides that electoral divisions ‘shall not be formed out of parts of different States’.

[120] Australian Electoral Commission, 'Commissioner Issues Federal Electoral Determination' Media Release (18 February 2009) <http://www.aec.gov.au/About_AEC/Media_releases/2009/index.htm> at 19 January 2010.

[121] This change precipitated electoral redistributions in both states: Australian Electoral Commission, '2009 Redistribution of Queensland into Electoral Divisions' (Australian Electoral Commission, 2009); Australian Electoral Commission, '2009 Redistribution of New South Wales into Electoral Divisions' (Australian Electoral Commission, 2009).
[122] Wesberry v Sanders, [1964] USSC 31; 376 US 1 (1964).
[123] Attorney-General (Cth) ex rel McKinlay v Commonwealth [1975] HCA 53; (1975) 135 CLR 1.

[124] McGinty v Western Australia [1996] HCA 48; (1996) 186 CLR 140.

[125] Greg Carne, 'Representing Democracy or Reinforcing Inequality?: Electoral Distribution and McGinty v Western Australia' (1997) 25 Federal Law Review 351, 354.
[126] McGinty v Western Australia [1996] HCA 48; (1996) 186 CLR 140, 272 (Gummow J).

[127] Nick Economou, 'What Happened to Rural Weightage?: Malapportionment and the Reform of Australian Electoral Systems under Labor' (Paper presented at the Australian Political Studies Association, Melbourne, 24-27 September 2007).
[128] Weeks, n 3, 109-114.

[129] These concerns are illustrated by the Australian Government’s unsuccessful attempt to introduce an ‘Australian Card’ in 1985, although the proposal was prompted by fiscal rather than demographic concerns: Roger Clarke, 'Just Another Piece of Plastic for Your Wallet: The 'Australia Card' Scheme' (1987) 5(1) Prometheus 29.
[130] United States Constitution Art I, Sec 2.

[131] Alexander Hamilton, John Jay and James Madison (eds), The Federalist (first published in 1788, reprinted 1964) No 58, 377. The second issue was a live one because there was popular concern that the original 65-member Congress was too small: The Federalist No 55, 359.
[132] Hamilton, Jay and Madison, The Federalist No 55, n 131, 363.

[133] Constitution Act 1867 (Canada) s 91(6). See Quick and Garran, n 51, 572. Censuses had already been employed in Canada in finding political accommodation of British and French interests in North America: Bruce Curtin, The Politics of Population: State Formation, Statistics, and the Census of Canada, 1840-1875 (2001).

[134] Australian Bureau of Statistics, 'Population Estimates: Concepts, Sources and Methods, Cat. No. 3228.0.55.001' (ABS, 2009) 1-2.

[135] For an official history, see Australian Bureau of Statistics, Informing a Nation: The Evolution of the Australian Bureau of Statistics 1905-2005, ABS Catalogue No. 1382.0 (2005).

[136] See, eg, New South Wales Registry of Births, Deaths and Marriages, <www.bdm.nsw.gov.au> at 15 February 2010.
[137] Australian Bureau of Statistics, n 134, 33-34.
[138] Census and Statistics Act 1905 (Cth) s 6.

[139] Brian Galligan and John Chesterman, 'Aborigines, Citizenship and the Australian Constitution: Did the Constitution Exclude Aboriginal People from Citizenship?' (1997) 8 Public Law Review 45, 51-52; Frank McGrath, The Framers of the Australian Constitution: Their Intentions (2003) 197.

[140] John Gardiner-Garden, 'The Origin of Commonwealth Involvement in Indigenous Affairs and the 1967 Referendum' (1996-97) Australian Parliamentary Library, Background Paper 11 <www.aph.gov.au/library/pubs/bp/index.htm> at 16 February 2010.

[141] Kate Ross, 'Population Issues, Indigenous Australians: Occasional Paper, Cat No. 4708.0' (Australian Bureau of Statistics, 1996) 5. Torres Strait Islanders were initially classified as ‘Aboriginal natives’ and thus excluded from the population count. In the 1947 census they were classified as Polynesian, while in the 1954 and 1961 censuses they were classified as South Pacific Islanders, resulting in their inclusion.

[142] Bain Attwood and Andrew Markus, The 1967 Referendum: Race, Power and the Australian Constitution (2nd ed, 2007) 65-68; John Summers, 'The Parliament of the Commonwealth of Australia and Indigenous Peoples 1901-1967' (2000-01) Australian Parliamentary Library, Research Paper 10 <www.aph.gov.au/library/pubs/rp/index.htm> at 16 February 2010, 67.

[143] Australian Bureau of Statistics, n 134, 3.

[144] Ran Hirschl, Towards Juristocracy: The Origins and Consequences of the New Constitutionalism (2004) 7-8.

[145] Sujit Choudhry, 'Bridging Comparative Politics and Comparative Constitutional Law: Constitutional Design in Divided Societies' in Sujit Choudhry (ed), Constitutional Design in Divided Societies: Integration or Accommodation? (2008) 3, 8-9.
[146] Choudhry, n 145, 4-5.
[147] Peter Russell, The Judiciary in Canada: The Third Branch of Government (1987) 167-168.

[148] Constitution (Amendment) Act 1997 (Fiji) s 51. For a discussion, see Benjamin Reilly, Democracy in Divided Societies: Electoral Engineering for Conflict Management (2001); Yash Ghai and Jill Cottrell, 'A Tale of Three Constitutions: Ethnicity and Politics in Fiji' in Sujit Choudhry (ed), Constitutional Design in Divided Societies: Integration or Accommodation? (2008) 287.

[149] See Andrew Byrnes, Hilary Charlesworth and Gabrielle McKinnon, Bills of Rights in Australia: History, Politics and Law (2009); Frank Brennan et al, 'National Human Rights Consultation Report' (Commonwealth of Australia, 2009).

[150] Charter of Human Rights and Responsibilities Act 2006 (Vic) s 12 (freedom of movement), s 17 (protection of families); Human Rights Act 2004 (ACT) s 11 (protection of the family), s 13 (freedom of movement).

[151] Veena Soni, 'Thirty Years of the Indian Family Planning Program: Past Performance, Future Prospects' (1983) 9(2) International Family Planning Perspectives 35, 36.

[152] As noted in Part 3.4, there is an implication in the Australian Constitution that current population statistics will be maintained through periodic enumeration.

[153] The quotation is from a manual prepared by Richard Baker for the 1891 Constitutional Convention, cited in John Williams, 'The Constitutional Amendment Process: Poetry for the Ages' in H.P. Lee and Peter Gerangelos (eds), Constitutional Advancement in a Frozen Continent: Essays in Honour of George Winterton (2009) 1, 6.

[154] The rule requiring equal state representation in the Senate (s 7) is also non-adaptive, but is arguably justifiable as an expression of federal values. See Part 3.3 above.
[155] Plebiscites, such as the vote on a preferred National Song in 1977, are another example.

[156] Williams, n 153, 4.

[157] Frank Easterbrook, 'Abstraction and Authority' (1992) 59 University of Chicago Law Review 349, 363.
[158] Quick and Garran, n 51, 988.

[159] The longest human life ever reliably recorded is 122 years: M. Allard, V. Lebre and J.-M. Robine, Jeanne Calment: From van Gogh’s Time to Ours. 122 Extraordinary Years (1998). There is an ongoing debate about whether the human life span is limited. Compare S. Olshansky, B. Carnes and D. Grahn, 'Biological Evidence for Limits to the Duration of Life' (2003) 4 Biogerontology 31 and J. Oeppen and J. Vaupel, 'Broken Limits to Life Expectancy' (2002) 296 Science 1029.
[160] Geoffrey Sawer, Australian Federalism in the Courts (1967) 208.

[161] As described above, these are the addition of the social security power in s 51(xxiiiA) (1946); the repeal of s 127, enabling Indigenous Australians to be counted (1967); and the introduction of a compulsory retirement age for federal judges in s 72 (1977).

AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/UTSLRS/2010/7.html

State or Territory	1901	2009	Change
	% of population	% of population	%
New South Wales	36.0	32.5	–3.5
Victoria	31.6	24.8	–6.8
Queensland	13.2	20.1	6.9
South Australia	9.5	7.4	–2.1
Western Australia	5.1	10.2	5.2
Tasmania	4.6	2.3	–2.3
Northern Territory	n.a.	1.0	1.0
Australian Capital Territory	n.a.	1.6	1.6
Population (%) Population (million)	100% 3.82 million	100% 21.87 million	– 18.05 million
Source: ABS, Australian Historical Population Statistics, Cat. No. 3105.0.65.001 (2008), Table 4.1; ABS, 3101.0 Australian Demographic Statistics (2009), Table 4.
Notes: The Northern Territory and the Australian Capital Territory did not exist until 1911, when they were excised from the territory of South Australia and New South Wales, respectively.

State or Territory	Population		Electorates		Ratio	Variation
	Number	%	Number	%		%
	A	B	C	D	E (=D/B)	F
New South Wales	6,967,271	32.6	48	32.0	0.982	–1.8
Victoria	5,297,567	24.8	37	24.7	0.995	–0.5
Queensland	4,279,450	20.0	30	20.0	0.999	–0.1
Western Australia	2,163,251	10.1	15	10.0	0.988	–1.2
South Australia	1,601,827	7.5	11	7.3	0.979	–2.1
Tasmania	498,163	2.3	5	3.3	1.430	43.0
Australian Capital Territory	344,744	1.6	2	1.3	0.827	–17.3
Northern Territory	219,950	1.0	2	1.3	1.296	29.6
Other territories	2,080	0.0	0	0.0	–	–
Total	21,374,303	100.0	150	100.0	–	–
Source: Australian Electoral Commission, Commissioner Issues Federal Electoral Determination, Media Release (18 February 2009), author’s calculations.
Notes: Under the Commonwealth Electoral Act 1918 s 45, the total number of ‘people of the Commonwealth’ excludes the people of any territory. For the purpose of calculating the quota in s 24 of the Constitution, the total population in 2009 was therefore 20,807,529.

University of Technology Sydney Law Research Series

Opeskin, Brian --- "Constitutions and Populations: How Well has the Constitution Accommodated a Century of Demographic Change" [2010] UTSLRS 7; (2010) 21(2) Public Law Review 109