The interwar years are conventionally described as a period of stasis in Australian policy development. The energy that produced major social reforms at the turn of the century was lost, and this is frequently the period in which is located a transition from being a “social laboratory” to being a “welfare laggard.”Footnote ¹ While these were years of crisis, “hard times” as Peter Gourevitch called them, they did not produce a major realignment of forces such as he analyzed in the American New Deal, the best of his examples of an alliance between agrarian, labor, and some business groups leading to new possibilities. As he argued, “The big policy and political shifts in the thirties came where ‘deals’ or bargains were made—where diverse societal actors were willing to make major trades that took them away from their traditional political as well as policy attachments.”Footnote ²

There was no such fundamental shift in political alignments in Australia. Conservative governments ruled at the federal level, other than the tragic interlude of the Scullin Australian Labor Party (ALP) government, winning power just as the Depression hit and falling two years later. The ALP split in 1916 over conscription, and split again in 1931, over adherence to orthodox austerity policies. Each time, the party regrouped, and although weakened it continued. And each time, the non-Labor side of politics was the beneficiary, re-forming into new parties, first as Nationalists and then as the United Australia Party (UAP), with Labor defectors providing their leadership. Despite upheavals, these non-Labor parties were also continuous, as coalitions of conservatives, remnants of the social liberalism that was powerful at the turn of the century and Labor defectors. In addition, the Country Party had emerged to influence after 1922, representing rural and conservative interests, and was in a formal governing coalition for most of the period.Footnote ³ The major political forces had not fundamentally realigned during the interwar years, and in Gourevitch’s schema this might seem to explain the stagnation of policy. Yet there were significant forays into social policy development from these conservative governments, despite the modesty of their reforming zeal, and their failure tells a story both of their internal divisions and of the institutional inertia described by path-dependence analysis.

This article argues that these failures can be understood by extending Jacob Hacker’s characterization (following Paul Pierson) of path dependence as “developmental trajectories that are inherently difficult to reverse.” Hacker’s description of American policy showed how early choices shaped later options, and how a mix of public and private provision was fashioned by state policies rather than by default. He emphasized how “timing and sequence are crucial,” because past decisions establish institutions and build interests, lead to “long-lived commitments on which beneficiaries make choices,” and preclude other options. Hacker’s argument was not that change cannot occur, but that it is “channeled by the self-reinforcing mechanisms that propel the existing path of development.”Footnote ⁴ Similarly, Peter Baldwin’s description of European policy history noted how alliances between liberal and agrarian interests in Scandinavia meant that welfare developments began with the solidaristic, tax-financed character later—teleologically—associated with Social Democracy. Initiatives for aged pensions (Denmark in 1891 and Sweden in 1913) were funded from taxation, and targeted with means-tests because farmers objected to alternative proposals for an insurance model that they feared would increase their labor costs. As Baldwin put it, “The cornerstone of the unique Scandinavian welfare edifice was set in place already during the late nineteenth century.”Footnote ⁵ It was the subsequent abolition of targeting that most came to characterize the Nordic model, distinguishing it from social-insurance welfare states and from the Australasian system of tax financing with means-tests.

Hacker has argued that “path-dependent processes imply a strong element of institutional inertia. Once past a certain threshold of development, what exists is likely to persist.”Footnote ⁶ In Australia, the interwar period is book-ended by the initiatives at the turn of the century establishing wage arbitration and the aged pension, and then by Labor’s welfare state in the mid-1940s, establishing new benefits on the same model as the pension. In this light, it could seem that it was simply too difficult to depart from the existing trajectory, with this path reemerging with new vigor in the 1940s. But the attempts between the wars to shift from the existing model to a social insurance model were blocked less by an attachment to this existing path than by the presence of the neighboring institution of arbitration as a mechanism for setting minimum wages and for potentially shifting costs. Similarly, attempts to introduce Child Endowment, on the model of the maternalist benefits that became common in many European countries and in the American states, were prevented by conflicts over the same arbitration institution. In terms of path dependence, this requires attention to the interaction of different institutions with intertwining paths.

THE WEIGHT OF THE PAST

Path-dependence analyses are fundamentally about the weight of the past, showing how the existing architecture of policy must be either built around, renovated, or demolished. What has already been constructed shapes both what can be imagined and what can be achieved, and in this light policy failure can be as illuminating as success. The existing policy institutions that played the largest part in the story of the interwar years were the old-age pension and compulsory wage arbitration. Both were implemented at the turn of the century, following the New Zealand example, by the Australian colonies of Victoria and New South Wales (NSW). In 1900, Victoria and NSW competed to be first to establish an aged pension, means-tested and funded from consolidated revenue; by 1901, NSW had also instituted compulsory arbitration. The new federal (Commonwealth) government established in 1901 legislated for arbitration in 1904, and after some delay the Commonwealth Old Age Pensions Act (1908) superseded the state schemes; in 1910, invalid pensions were added.Footnote ⁷ Industrial arbitration and the idea of a “living wage” were part of transnational debate and policy exchange; similarly, aged pensions were part of larger discussions in Europe and Australasia.Footnote ⁸ The impetus for both initiatives owed much to the influence of social liberal ideas, with their possibility enhanced by the emergence of the labour movement into the polity. Marian Sawer has analyzed the influence of British liberalism on these developments, while Stuart Macintyre has argued that liberals devised arbitration as a means of “restoring harmony and order to class relations” by compelling labor and capital to negotiate.Footnote ⁹

The distinctive features of Australasian aged pensions were that they were means-tested to target them to the poor, and were funded from general revenue, setting a different policy trajectory to the social insurance models that became the norm in Britain, America, and much of Europe. When debating pensions in the 1890s, Australian policymakers considered the Bismarckian alternative of compulsory insurance for low-paid workers. Edward O’Sullivan, the Labor politician who wrote the key NSW report on pensions in 1896, argued that compulsory levies would be “objectionable to people of British origin”; the same sentiments were current in Britain, where The Times claimed that “the German is accustomed to official control … from the cradle to the grave.… Self-help and spontaneous growth are better suited to the Englishman.”Footnote ¹⁰ Racial conceits aside, O’Sullivan and others thought an insurance model would not work because the burden would be heaviest for low-paid workers, women without employment would be excluded, and an itinerant workforce would be too hard to administer if they crossed colonial borders (which was only a problem before Federation).Footnote ¹¹ The corollary of a pension paid from general revenue was means-testing to contain costs. There was vigorous debate about whether the means-test had “the taint of charity,” but the originators of pensions convinced themselves that they were leaving behind the nineteenth-century system of discretionary charitable relief. O’Sullivan advocated a means-test while declaring the pension “a free gift from the State in recognition of services rendered to it. The pension should be conceded as a right, not as a pauper dole.”Footnote ¹² When the Commonwealth pension was enacted, the NSW system provided the model, with little consideration of either social insurance or means-testing, of how it was financed or how it was targeted. Yet these were the aspects of the pension that were most objectionable to conservative/liberal governments in the social insurance debates in the interwar period.

The defining feature of the other major institution, compulsory arbitration, was the establishment of courts with the power to settle industrial disputes and determine wages. Arbitration was a central part of the doctrine of “New Protection” at the turn of the century; at its most minimal, it meant an implicit social contract by which employers gained tariff protection if they paid the fair wages established by arbitration. This idea of social protection became a cornerstone of the Australian welfare regime with the establishment, in 1907, of the minimum wage in the Harvester Judgment when H. B. Higgins, the eminently liberal president of the Arbitration Court, determined a minimum male wage sufficient to support, in “frugal comfort,” a man, his wife, and (he assumed) three children. Higgins argued that the law must override “the higgling [sic] of the market” and determine a wage based on human needs. In 1919, he followed the same gendered logic and set a female minimum wage at 54 percent of the male rate.Footnote ¹³

The place of arbitration in the history of Australian social policy rests on two accounts of its significance. First, establishing a minimum wage based on needs rather than market mechanisms was itself a major social policy initiative, contributing to the distribution of social goods and to protecting citizens against the vagaries of the market. And as welfare regimes can be usefully characterized as stronger or weaker versions of the breadwinner model of work, welfare, and family, in terms of how policies reflect cultural ideas about the breadwinner and his dependents, the basic wage ensured that Australia was one of the stronger versions of this model.Footnote ¹⁴ Second, the argument has been made that arbitration had consequences for the trajectory of social policy development, producing what Francis Castles called “Australasian exceptionalism” in a “wage-earners’ welfare state.” This was not an argument that wage earners were the only beneficiaries, but that later developments presumed the existence of fair wages. Though he does not use its framework, Castles’s argument is the most extended exploration of path dependence for Australia, arguing that the presence of arbitration “influenced subsequent welfare state development in a distinctive manner,” because social policy developments “required only a welfare safety-net for those outside the labour market.”Footnote ¹⁵ While it has been criticized for building a model rather than analyzing actual policy processes, the argument has merit and is considered further below.

The prior development of wage arbitration and aged pensions affected debates in the interwar years about Child Endowment and National Insurance, but a third policy institution was ignored. Like other combatant nations in World War I, Australia developed a large veteran’s system for disabled veterans and war widows.Footnote ¹⁶ The pension rates were much more generous than the aged pension, and this system cost more than aged and disability pensions until at least 1926, with substantially more people receiving veterans’ benefits than aged and disability pensions until the mid-1930s.Footnote ¹⁷ Stephen Garton is the only historian to systematically include the veterans’ system in the welfare state, and he suggested that its expense helps explain the stagnation of social policy. Without a social insurance system, “general revenue was a well from which many drew. … Repatriation was a significant drain on public revenue. … In the squeeze on resources that resulted, it was the civilian welfare system that suffered.”Footnote ¹⁸ While part of the context, it is notable that debates in the interwar years did not mention the cost of the veterans’ system, as though it was too politically sensitive to notice. Instead, the focus was on the aged pension and what Rob Watts calls “the mistake of 1908” in deciding to fund it from general revenue.Footnote ¹⁹

NATIONAL INSURANCE, 1923–1929

In August 1923, Earle Page, treasurer and leader of the Country Party, announced a Royal Commission into compulsory insurance against sickness, invalidity, old-age, and unemployment. In four closely argued reports, this commission provided the proposals for Page’s National Insurance Bill, introduced in September 1928, fiercely debated and then abandoned exactly a year later. Before assessing why it failed, it is worth considering why the attempt was even made. For its advocates, several interlocking considerations made the argument for social insurance.

The first was a fiscal case that the projected costs of the aged pension would absorb more and more of the Commonwealth’s revenue. At this stage, the Commonwealth levied limited income taxation compared with the states. National Insurance debates invariably began with a rehearsal of the rising costs of the pension, with Page describing them as “a heavy burden on the taxpayers”; he argued that compulsory insurance was a more “satisfactory and permanent basis, to remove altogether the taint of pauperism.” Insurance would mean “assistance to the aged and invalid can be placed upon a non-charitable basis within the financial capacity of the Commonwealth.”Footnote ²⁰ And because the aged pension was fiscally unsound, it had to include the objectionable means-test: “The inevitable result, since it was based upon no intelligent financial foundation, was that it became what it is now, a charitable dole, given only to those who cannot live without it. It is as much a charitable institution as the poorhouse or the soup-kitchen … based on pity rather than on logic, on benevolence rather than on thrift.”Footnote ²¹ Both terms in Page’s characterization of the pension were significant; the taint of pauperism was as important as its financial instability. This fiscal case for a contributory scheme would be even more prominent in the late 1930s.

The second consideration making social insurance attractive was more ideological. For Australian liberals caught in the antiprogressive trajectory of the Nationalist party, a contributory scheme might be their salvation, marrying progressive policy with thrift. They looked to the British example of Lloyd George’s 1911 system of compulsory insurance for low-paid workers, organized through “approved societies,” and providing sickness and disability payments, health-care costs, and limited unemployment insurance. As New Liberalism in action, National Insurance has been seen as the foundation stone of the British welfare state, though Pat Thane points to the objection among working-class organizations that it required them to pay for their own benefits.Footnote ²² For Australian liberals, Lloyd George held out the hope that liberalism still had a future as a movement of social reform. In Melbourne, The Age viewed him as the future of a liberalism that it worried was imperiled, and this most liberal of newspapers championed the same path. “It flatters us to hear our country described as a political social laboratory … we love to think that we lead, that the older nations watch us anxiously, then prepare to imitate us timidly. … Unfortunately, our record in respect of National Insurance, for instance, plays havoc with the theory.” With neither health nor unemployment insurance, Australia was lagging; for The Age, National Insurance was part of the trajectory of liberal social reform, “consistent,” as they always added, “with self-respect.”Footnote ²³

That concern with self-respect indicates a third, more moral argument made for contributory insurance—that it encouraged thrift by ensuring benefits were a return on investment rather than a “demoralizing” payment by the state. The Royal Commission had positioned National Insurance as the next stage of reform, with “beneficial results in encouraging thrift and mutual assistance, mitigating poverty, allaying social unrest, and furthering national efficiency.”Footnote ²⁴ Page, when his legislation faced defeat in 1929, insisted that the means-test “must discourage thrift and be repugnant to, even if inevitably accepted by, many deserving citizens.” Footnote ²⁵ He conceded that his new benefits were not generous, but “this is not a scheme for providing all that is necessary to secure comfort. … It is a scheme for encouraging, enabling, even compelling, all workers to make some provision for their dependants.”Footnote ²⁶

The moral case behind persistent criticisms of the means-test in the aged pension was not so much that it demoralized individual recipients, but that it demoralized society as a whole; it discouraged self-reliant and thrifty behavior while rewarding improvidence and dependence. It was a powerful moral case with a lineage running back to nineteenth-century Poor Law discourses about charity and pauperism, with their persistent worry that assistance for the poor undermined self-reliance and encouraged dependence.Footnote ²⁷ It resonated strongly, as it always had, with core Protestant values and narratives. It was not that the advocates of National Insurance were particularly godly but rather, as Judith Brett demonstrates, that Australian liberalism had been infused with particularly Protestant themes.Footnote ²⁸ Improvidence was not just waste, it meant squandering the gifts of Providence; thrift was not only saving the rewards of work, it denoted the self-respect that indicated one was not among the fallen. Frequent references in these debates to self-respect, thrift, and the demoralizing means-test were not just questions of policy design; they invoked an inner moral world imbued with faith. These resonances brought social reform and self-reliance together in National Insurance, and they explain the commitment of some to the contributory model.

Finally, there was some significant support for social insurance from women’s groups, such as the National Council of Women, the Feminist Club in Sydney, and (from 1929) the United Associations of Women. Jessie Street, for example, was a firm advocate of a universal social insurance scheme. Her proposal was that all men and women, working or not, and rich or poor, would pay equal contributions and be covered for sickness, unemployment, old age, and invalidity; but most important, social insurance would pay a “marriage and child endowment,” which would contribute to women’s financial independence. That also involved a position in the parallel debate in the interwar years about child endowment and the male basic wage. Street’s proposal was that “whether the husband was working or not the payment of marriage and child endowments would continue, since the cost of maintaining a wife and child would not be included in the basic wage.” Footnote ²⁹ In the event, the National Insurance schemes debated in the late 1920s and late 1930s were nowhere near this radical; they were limited to lower-paid workers only, which meant that women who worked would accrue benefits, but married women who were not working would only be entitled to benefits as dependents of insured husbands. Nevertheless, some feminists supported a more diluted social insurance scheme because it seemed to be one mechanism for extending welfare benefits.

Page’s legislation proposed compulsory contributions by workers and employers, with the Commonwealth contributing only for superannuation benefits, which it would initially fund in full. All wage and salary earners with incomes up to roughly twice the basic wage were included, though “the aboriginal native of Australia” was excluded, just as they (and “Asiatics”) were excluded from the aged pension. Flat-rate benefits would cover sickness and disability, allowances for children of those on these benefits, and allowances for the widows and orphans of those insured; the scheme would also establish a parallel superannuation scheme, to eventually phase out the existing pension. The National Insurance Bill had no provision against unemployment, and that absence was the only substantive criticism the ALP voiced. Labor had some interest in the contributory model in the 1920s, particularly for unemployment. In 1922, a Queensland Labor government had introduced compulsory unemployment insurance with equal contributions by workers, employers, and the state, directed to seasonal rural workers; in the late 1920s, the Victorian Labor government attempted to do likewise but was rebuffed three times by a conservative upper house. A decade later, Labor would insist welfare should be funded from tax revenue, but in the 1920s they were not opposed to the contributory model.Footnote ³⁰

But although there was little controversy within parliament, there was significant opposition outside, from the friendly societies and then from employer groups. Compulsory insurance would have meant additional business for the friendly societies, who were to be the main providers as in Britain, but they saw their voluntary ethos threatened: “If it became law,” one official stated, “friendly society membership must cease, and it would develop a race dependent on somebody else.”Footnote ³¹ Another declared that the bill would “destroy the spirit of thrift and independence characteristic of and inculcated by the great Friendly Society movement.”Footnote ³² This opposition was ironic given that National Insurance encoded into policy, and made compulsory their values of thrift and self-reliance, but they chose what they saw as self-interest ahead of the opportunity to universalize their values.

While the friendly societies had a marginal place in the existing system of social protection, the presence of arbitration was much more significant. It is pertinent that the 1920s was the decade when the arbitration system became fully operational. The Commonwealth Court had established supremacy over the state tribunals, minimum wages were close to the Harvester standard, industrywide awards were developing, and union recruitment had risen substantially.Footnote ³³ In these circumstances, employers worried that workers would ensure that their contributions were counted in determining wages; as Kewley noted, they “feared that through the process of wage fixation by industrial tribunals, the contribution of employees would ultimately be added to the weekly wage.”Footnote ³⁴ In September 1928, the president of the Victorian Chamber of Manufacturers warned that Page’s scheme should not proceed; he pointed to “the very high basic wage” and worried that workers’ insurance contributions would lead to even higher wages.Footnote ³⁵ The chorus was louder in mid-1929, when the Associated Chambers of Manufacturers declared “an emphatic protest against national insurance, because the burdens for the successful financing of the scheme must eventually be borne by industry, which could not afford to carry additional taxation.”Footnote ³⁶ That arbitration was now so well established, with interests organized around its presence, meant that it was part of the policy architecture in assessing a new proposal; institutional arrangements for determining the basic wage meant there was a potential mechanism for shifting costs. By September 1929, the Nationalist government accepted defeat, pointing to the looming depression as its pretext; as Page said in parliament, the time was not right: “Expenditure upon schemes of this character falls ultimately upon the industries of the country, and the burdens so imposed are immediately reflected in the cost of living of the nation, while the benefits to the community are delayed.”Footnote ³⁷ Two weeks later, he quietly announced that the bill would be postponed, and nothing was heard of National Insurance for another decade.

CHILD ENDOWMENT, 1925–1929

While proposals for contributory insurance came from the conservative/liberal side of politics, the strongest pressures for Child Endowment came from Labor and from women’s groups. The maternalist politics that were prominent transnationally in the interwar years argued that the state should direct assistance to women and children.Footnote ³⁸ Marilyn Lake has shown how maternalist policy advocacy in Australia represented choices made by women’s groups, positioning the family as an asset of the nation and arguing the rights of mothers as an alternative to the male citizenship of soldiers or workers. Feminist groups such as the National Council of Women, as well as Muriel Heagney within the ALP, argued for Child Endowment as securing an independent income for mothers.Footnote ³⁹ Child Endowment could be attractive to both left and right; for the left, it was part of income distribution, and for both left and right it was part of pro-natalist policies.

Child Endowment was being debated during the same years as Page’s National Insurance, but it too stalled because it was enmeshed with the arbitration of a breadwinner wage supporting a family. What was this family? Higgins had assumed three children, and most wage courts followed his lead, but NSW and Western Australia assumed two children. The problem lay with these inconsistencies in the size of the family considered, and with the fact that male workers without children received the same basic wage as those with many. Albert Piddington, the president of the NSW Industrial Commission and a strong advocate of Child Endowment, liked to point out the anomalies; the basic wage was the same for single men, who were paid for “phantom” wives and children, while large families were not adequately provided for. His preference was that minimum wages should be calculated for the needs of a couple, with children supported through Child Endowment. That was the view adopted by the National Council of Women, though Street went further and argued that the minimum wage should be based on a single man, with additional payments for mothers and children. But both Child Endowment and arbitration were fraught. The Nationalists insisted that they would only support Child Endowment if the anomalies in the basic wage were settled, feminists argued against the family wage and supported Child Endowment as at least a step forward, unions worried that conservatives really intended a wage cut for single men, and Labor struggled to justify both Child Endowment and a wage that supposedly included the needs of two, or three, children.Footnote ⁴⁰

The entanglement of Child Endowment with wage arbitration was clear in the NSW Labor government’s Family Endowment Act of 1927. It provided a weekly payment directly to mothers for each child, with a means-test at the level of the basic wage. Jack Lang, the populist Labor premier, wrote in his frequently misleading memoirs of his surprise that union officials “regarded it as a thin wedge that would result in the reduction of the basic wage.”Footnote ⁴¹ Although he reassured them, his government passed parallel legislation to do exactly what Piddington wanted, prescribing the family unit for the NSW basic wage as a couple without children.Footnote ⁴² Millicent Preston-Stanley, a Nationalist in the NSW parliament, and an advocate for maternal and child health reforms, clearly articulated the maternalist rationale for Child Endowment: “The most important function performed by any person for or on behalf of the State is that performed by the working mothers of the community in bringing into the world and rearing children.” But she also pointed out that in none of the European nations and American states that implemented family allowances was this “superimposed upon an already fair living wage.”Footnote ⁴³ It was an astute point, and remained the nub of the problem. The maternalist politics of Child Endowment clashed with the politics of wage-fixing, and this contested politics meant that Child Endowment would remain deadlocked at the national level.

The intractability of the problem was demonstrated in the one action Stanley Melbourne Bruce’s Nationalist government took, establishing a Royal Commission that reported in March 1929. It seemed to confirm the critics’ view that the supremely conservative Bruce only appointed inquiries to bury an issue, with the commission split between a majority implacably opposing Child Endowment and a minority supporting it. The minority report was written by John Curtin, former journalist and socialist, and future Labor prime minister, and Mildred Muscio, a feminist and president of the National Council of Women.Footnote ⁴⁴ His biographer records that during the commission Curtin was depressed and drinking heavily; it did not take a lot to drive Curtin to the bottle, but the Royal Commission must have been a dispiriting experience.Footnote ⁴⁵

The majority found many reasons for disliking Child Endowment; it would involve “paralyzing” taxation and lead to a litany of moral hazards. It would treat “a mother as a salaried servant of the State,” and “where distress among children is due to drinking, gambling or other extravagant habits … the receipt of money would accentuate the evil.” Fathers would “slacken off in their efforts, and … pass into a condition of careless acquiescence in a system which made their children … wards of the State.” Some of this was ludicrous, but the set-piece battle in the commission was about the relationship between wages and Child Endowment. The majority report reviewed the inconsistencies between federal and state systems and argued wages and Child Endowment were not, as advocates claimed, separate: “[apart from recent New South Wales legislation,] the two have been blended into one, every basic wage containing elements of child endowment.” Their view was that “the allowance [for children] should be struck out of the wage,” but without any Child Endowment, as neither employers nor the state should pay for children; that was “an obligation of the father.”Footnote ⁴⁶

To untangle the connections between arbitration and Child Endowment, the minority report proposed means-tested payments for families with an income about one and a half times the basic wage, with arbitration tribunals being required to adopt a common standard of a family of two children. Their Child Endowment proposal would provide only for third and subsequent children, and this represented some compromise by Muscio, whose organization favored Piddington’s solution. Muscio and Curtin left to one side the issue of the single worker with “phantom” dependents. But more important for future debates, they articulated how to pay for Child Endowment by levying a higher income tax on high incomes, and by lowering the taxable threshold to below the basic wage for “persons without dependants”; this would be “income tax levied in proportion to individual capacity to pay. The broadest backs should bear burdens carried in the interests of the community as a whole.”Footnote ⁴⁷ Lowering the tax threshold was a relatively radical idea to extend income tax deep into the working class, and it assumed the Commonwealth took a greater part in income taxation; both policies were acted on after Curtin became prime minister in 1941.Footnote ⁴⁸ They also considered the issue of the means-test, describing it as a “very difficult problem.” They had been urged by witnesses that universal payments were simpler and “avoid[ed] any stigma of charity,” but had concluded that available funds should be spent “on families where the need of the children is probably greatest.” Means-testing accorded with “the fundamental principle … that all children in the community should … have the possibility of the minimum standard of living implied in current wage awards.”Footnote ⁴⁹ Both these principles, a progressive tax system with a lower threshold and means-tested benefits, would come to the fore in the second debate on National Insurance as positions Labor adopted under Curtin’s leadership.

That the Royal Commission could not devise a way for a “family wage” to coexist with Child Endowment indicated why family assistance was stillborn for over a decade. Bruce’s Nationalist government would not act unless it had central control over wage-fixing; Bruce had lost a constitutional referendum in 1926 to give the Commonwealth that control, and the states were unwilling to hand over the power. In 1929, after endorsing the majority view of the Royal Commission, he told the state premiers that Child Endowment could be introduced only if wages were reduced for workers without children.Footnote ⁵⁰ Child Endowment touched too many suspicions and too many unresolved disputes, about what was a fair wage, about whether state or Commonwealth tribunals should dominate, and about who, if anyone, should support the children. It would take until 1941 before Menzies’s UAP government introduced a universal Child Endowment payment excluding the first child, and it was clearly devised as an anti-inflationary strategy to convince the Arbitration Court not to increase the basic wage.Footnote ⁵¹

NATIONAL INSURANCE, 1935–1939

As Australia climbed out of the Depression, the idea of reconstructing social policy on contributory lines was revived. The Lyons UAP government had a tepid commitment to social reform, and had been inactive and austere during the Depression. But Richard Casey, the treasurer who carried the burden of this second attempt, believed a contributory scheme was essential to a modern, rationally organized society; his biographer describes him as “quite comfortable with the notion of state planning.”Footnote ⁵² Casey’s own formulation was that he wanted “a more kindly state in the background to succour those whose lot in life, for various reasons, requires amelioration.”Footnote ⁵³ Watts put it more skeptically: “For Casey, a noblesse oblige indicated a duty to remedy the lot of the unfortunate; social insurance would ensure that the class most likely to benefit from this reform would also be paying for it.”Footnote ⁵⁴

The fiscal case about the cost of the aged pension was even stronger in the late 1930s; key administrators and advisers were committed to the insurance principle to rectify the “mistake of 1908.” In 1932, Lyons had stated that the nation could not afford pensions “on their present basis,” and a contributory system was “the only answer.”Footnote ⁵⁵ The government commissioned a visit for advice from British officials; there was remarkably little curiosity about the extensive development of the social insurance model by the late 1930s, in France, Austria, and Germany and the American Social Security Act.Footnote ⁵⁶ In 1937, British expertise was thought to be enough, and this limited range of research contrasts with the investigations at the turn of the century about arbitration and pension schemes. A report by Sir Walter Kinnear of the British Ministry of Health argued that the cost of the pension was the central problem and provided advice for replacing it with a contributory scheme. Watts, who traced the policy process closely, argued that Treasury’s approval was “crucial,” and once gained “it now became a matter of political judgement as to how to implement the proposals.”Footnote ⁵⁷

In May 1938, Casey introduced his legislation to the parliament. It was remarkably similar to Page’s scheme, with equal contributions from employers and workers, offering sickness, disability, widows and old age benefits, but it added health benefits and medical care provided by doctors on the British panel system. It excluded unemployment insurance and relied on the cooperation of the friendly societies to work. National Insurance, Casey said, was “a scheme which will assist the sick, the aged, the widow and the orphan, preserve the dignity of labour, and enable the Government to extend its benevolence, on a self-respecting basis, to a very largely increased number of participants.” That combination of “benevolence” extended, but “on a self-respecting basis,” nicely captured the ideological appeal of social insurance. He immediately moved to the fiscal argument. “In view of the impending liability of the existing pension scheme, I say quite frankly that, unless something is done to put these schemes on a contributory basis, no government of the future … could embark upon any worthwhile extension of our social services without seriously threatening the whole financial fabric of the Commonwealth.”Footnote ⁵⁸ That was the conventional Treasury position, and it assumed that no further taxation could be countenanced.

The moral case for social insurance was still just as powerful and permeated both government and bureaucratic advocates. J. B. Brigden, the economics professor who helped draft Casey’s legislation and was then appointed chairman of the National Insurance Commission, described contributory insurance as preeminently modern, “as comparable with the old-fashioned way as a modern motor vehicle is to a bullock wagon.” Shifting from “our customary approach” was necessary because currently “only the ‘needy’ can get the full benefit. The cause of being sufficiently needy may be misfortune, but it is just as likely to be reckless improvidence. It is notorious that the customary way of dealing with the problem is demoralising. The thrifty and careful people as a rule cannot benefit. The ‘means test’ is generally disliked, but national insurance is the only known way of getting rid of it.”Footnote ⁵⁹ This was the same moral argument as a decade before; means-tested benefits were socially “demoralizing” because they penalized thrift and probity. As the Taxpayers’ Association of Victoria argued, “Under the present system, thrifty and industrious people were discouraged, while the shiftless and thriftless got a premium for wasting what they had acquired.”Footnote ⁶⁰ Proponents such as Casey might not be so blunt, but this was the moral hazard they saw in the existing system.

The government coalition had a majority in both houses, and barely two months later Casey’s legislation had passed and was enacted by July 1938. The following day, the National Insurance Commission was established and began the administrative work of establishing contributions and benefits. The path of Australian welfare policy had been irrevocably altered, finally achieving what had been attempted a decade earlier, a shift from a regime based on means-tested pensions funded from revenue to a wider range of benefits based on the contributory social insurance model. Yet only nine months later, the whole scheme had been abandoned.

The few historians who have examined this debacle have come to imprecise conclusions about what happened. Watts points to the medical profession’s opposition and to growing public opposition, and Kewley ascribed its demise to internal divisions and growing concern about defense expenditures. The available biographies of two key players, Casey and Menzies, provide little illumination.Footnote ⁶¹ There were sources of opposition, though none was insurmountable. The medical profession argued that the proposed capitation fee was inadequate, but their opposition was neatly deflected into a Royal Commission charged with determining appropriate fees. Women’s groups criticized the proposal that female workers would pay lower contributions and receive a lower pension, and were successful in having this provision changed; many were in general in favor of the scheme.Footnote ⁶² The friendly societies were no longer fierce opponents, especially after the ALP adroitly moved an amendment excluding life insurance companies from being “approved societies”; they now had a near monopoly on what Timothy Alborn, describing the British equivalent, called “the market for self-help.”Footnote ⁶³ There were suggestions that financial institutions pressured the government to abandon its policy once they had been excluded, and this fed rumors about the government’s dependence on its financial backers, but Watts doubts they had sufficient influence.Footnote ⁶⁴

As in the late 1920s, Labor criticized the absence of unemployment insurance, but Curtin, now ALP leader, also explicitly opposed shifting welfare onto a contributory basis: “The Labour Party expresses its utter condemnation of individual contributions as a principle in regard to invalidity, old-age and widows’ pensions. These services should be a charge upon the consolidated revenue of the Commonwealth. To impose special levies, either on workers or employers, is utterly unjust.” Curtin was positing both welfare and taxation as national responsibilities that encompassed all citizens. The legislation was “sectional” because only workers and, to a lesser extent their families, were covered and only workers and employers paid. Instead, “There should be equity in the treatment of our citizens, having regard to their needs, their rights and their obligations.”Footnote ⁶⁵ Curtin was formulating an explicit case about redistribution, in which those who could most afford to should pay increased taxes to generate resources that would be directed, through means-tests, to those who most needed support. It reprised what he had written in the Child Endowment minority report in 1929. It was not so much a model of class as of inequality. In a booklet rehearsing why they opposed National Insurance, Labor shifted the emphasis onto taxation. “It is regressive as a social measure because it ignores the cardinal principle of taxation—that those in the best position to pay should bear the cost; it imposes charges on the lower-paid workers through their contributions—charges which cannot be passed on as in the case of employers. … The act is designed to relieve the government of the necessity of increasing taxation on higher incomes and property to maintain existing services or to provide wider social services.”Footnote ⁶⁶ Similarly, Labor had no real qualms about means-tests, which had few of the moral resonances they had for liberals. Curtin and Muscio had written in 1929 that targeting was more effective, and pointed to the NSW Family Allowances to prove “payments of this kind assured by law are soon regarded as a right of citizenship, even when an income limit is set.”Footnote ⁶⁷ Labor’s opposition was certainly not what doomed Casey’s legislation, as they had little capacity to prevent it; but in the course of debating it, Labor clarified its commitment to the path first set at the turn of the century. The example of New Zealand’s Social Security Act (1938) reinforced this trend, when Joseph Savage’s Labour government had legislated for unemployment and sickness benefits, along with health and hospital coverage, all means-tested and funded from revenue.Footnote ⁶⁸

Casey had anticipated that National Insurance would again be entangled with arbitration, and his legislation included a provision to prohibit the Arbitration Court from including workers’ insurance contributions in calculating the basic wage. The Age noted that this provided employers with “protection from the risk of having in effect to pay both”; not to do so “would bring influential opposition to the measure from employers’ interests.”Footnote ⁶⁹ When this provision was debated, the UAP’s Senator Leckie stated the relationship bluntly. If workers’ contributions were “added to the basic wage the employers would pay the lot.”Footnote ⁷⁰ Labor opposed the provision as “a direct attack” on living standards “because it prevented wage-fixing tribunals from increasing wage rates to meet the additional imposts on the cost of living.”Footnote ⁷¹ The Taxpayers’ Association of NSW countered that including workers’ contributions in the basic wage would “result in the whole burden being carried by the employer.”Footnote ⁷² The skirmish illustrated again how the institutional presence of arbitration was shaping subsequent policy debates.

What unraveled the government’s own policy, and led to reversing what it had actually established, was division within UAP ranks, with some fiercely committed to this reform, while others had little taste for Casey’s “more kindly state.” In addition, there was vigorous opposition from the Country Party; Casey’s original scheme excluded the self-employed, so small farmers were excluded. But the Country Party also demanded that farmers who were employers must be exempt, meaning that rural industries would be outside the scheme. Their opposition is reminiscent of the opposition Baldwin described in Sweden and Denmark, where agrarian interests feared that social insurance would increase their labor costs.Footnote ⁷³ With fault-lines opening up in the coalition, the growing threat of war emerged as a convenient rationale for delay. Menzies, the deputy leader of the UAP and attorney general, made it clear that he would resign if National Insurance was abandoned, and given the enmity between him and the Country Party, this only reinforced their opposition. When the government in effect did abandon it, Menzies resigned, and a few weeks later, after Lyons’s death, he had been elected leader. But the tensions of the previous months also meant that the coalition was dead and given the intractable divisions that had opened up, National Insurance was also dead, for the second time in just over a decade.Footnote ⁷⁴