Australia was the first country in the world where a workers' party, the Australian Labor Party (ALP), was elected to executive office at both the national at state government level. By 1915 Labor had formed majority governments at the federal level and in every state apart from Victoria. When Labor split over conscription in 1916 the party was condemned to political marginality at the federal level where it won only one election in the interwar period. Nevertheless, Labor soon recovered at the state level. Labor claimed to represent lower-income earners, but were Labor governments distinctive in their record of public expenditure and taxation or did Labor's rhetoric merely serve to incorporate workers into the capitalist system? Rudolf Goldscheid argued in 1925 that 'the budget is the skeleton of the state stripped of all misleading ideologies'. (1) The experience of Labor governments in the six Australian states provides a unique opportunity to consider what labourism meant in practice. However, previous scholarship on Australian political economy in the interwar period has focused largely on economic regulations and the role of the federal government rather than on fiscal policy. The result has been that key aspects of what labourism meant in practice been neglected.
Public Expenditure, Taxation and Labourism
The growth of public expenditure and taxation has been the object of substantial academic enquiry. We can identify at least three divergent patterns of explanation. Some have argued that the growth of public expenditure is an inevitable result of modernisation and economic development, whether due to the social stresses that result from an urbanised and individualised society or the simple fact that higher living standards enable governments to impose higher levels of taxation. Public choice analysts argue that governments may adjust public expenditure to match the preferences of the median voter or to buy voter support during election campaigns. Another approach would point to the continued divergences between nation-states, as evidence that politics and culture do have an independent influence on levels of public expenditure and taxation. This approach might highlight the impact of democracy, which potentially empowered the poor and insecure majority, or the influence of political traditions potentially hostile to laissez faire such as social democracy or political Catholicism. (2)
In 1901 Federation established a national commodity market but there remained major divergences in the political economy of the Australian states apparent in varying wage levels, per capita incomes and taxation. Over the course of the twentieth century, however, the states converged. Landmarks in this process included the increasing dominance of Commonwealth government and its agencies in tax collection, wage fixation and social policy. Compared to the United States or the European Union, Australia became a uniquely centralised federation. (3) However significant divergences persisted in the interwar period and these were linked in part to the partisan composition of state governments.
Early analysts of Australian public policy, such as James Bryce, Frederick Eggleston and W.K. Hancock, considered that Australia, a democracy with a strong workers' party, seemed to represent the future of democratic governance. These authors focused on industrial arbitration and tariff protection, but they also referred to social programs such as pensions and highlighted the nexus between Labor governments and public employees, including white-collar public servants. (4)
Hancock popularised the idea of Labor as the 'party of initiative' as opposed to the conservative 'party of resistance'. In 1957 Jim Cairns interpreted the history of Australian social policy in this framework. He argued that the development of the welfare state was driven by those who benefited from its expansion; this was the working class and the Labor Party as its political vehicle. From this perspective the absence of substantial social policy innovation at a federal level in the interwar years was due to Labor's exclusion from federal office. Even Eggleston argued that the appropriate role for the non-Labor parties was often to impose a dose of reality on Labor's social policy initiatives. (5)
Taxation policy received little attention from these authors but the history of Australian taxation could be explained within the framework of 'initiative' and 'resistance'. Labor in government claimed to link taxation policy to social goals. The 1912 federal land tax, introduced by the Fisher Labor government, sought to break up of large land holdings. From this perspective the explanation of Australian taxation policy would require what Fritz Mann described as 'fiscal sociology'. (6)
The fiscal policy of Labor governments has received little attention from labour historians. Those on the right tended towards a Whig progressivism in which the egalitarian commitment of Labor governments was assumed rather than demonstrated. The preoccupation of more radical scholars has been with property rights and economic regulation. They have tended to identify 'socialist' policies with the extension of public ownership. Labor governments were initially enthusiastic about public ownership, and in 1911 and 1913 the federal Labor government sought unsuccessfully to amend the Constitution to enable nationalisation of monopolies. If these proposed amendments had been ratified it is likely that the Australian political economy would have evolved in a radically different direction. However by the mid-1920s Labor's enthusiasm for public enterprise had ebbed. To explain what was distinctive about Labor in government before 1940 we need to give more attention to fiscal policy, but political historians overseas and in Australia have tended to neglect fiscal policy. (7)
The idea of Labor as the party of initiative came under sustained attack in the 1950s from two flanks. Henry Mayer rejected the assumption that Labor was by definition progressive. He argued that the advocates of this position assumed what they sought to prove, for example by attributing policy innovations of non-Labor governments to Labor pressure. Mayer argued that a useful research project would be to compare the policy record of a Labor-dominated state, such as Queensland, with that of a non-Labor state, such as Victoria. Noel Butlin's model of 'colonial socialism' also challenged the idea of Labor as the party of initiative. In this model, Australian public finance had been most distinctive in the later nineteenth century, long before the rise of Labor. At this time, colonial governments had borrowed in London to finance infrastructure construction undertaken by government agencies, especially railway expansions. In 1982 Butlin developed this model further in Government and Capitalism along with Alan Barnard and J.J. Pincus. They argued that 'colonial socialism' had declined in the twentieth century as the focus of government expenditure shifted towards urban infrastructure and the provision of social and human services. (8) This analysis suggested that the divergences between the Australian states were economic rather than political, so that the more urbanised states such as New South Wales and Victoria would lead this trend towards modernisation.
In the 1980s the terms of Butlin's analysis were reversed by the proponents of radical 'political economy' approach to Australian federalism. These scholars divided Australian states into two groups; a rust-belt core economy of New South Wales, Victoria and South Australia, based around manufacturing industry, and a periphery of Western Australia and Queensland (Tasmania was largely ignored). The peripheral state economies were considered to be increasingly subject to multinational resource companies. The core state governments clung to the welfare state but in the periphery the focus of public sector activity was increasingly the provision of subsidised infrastructure for multinational capital. In this analysis the political cultures and partisan alignments of states reflected their position in the emerging 'Pacific Rim' economy. (9)
These analyses reflected the polarised political climate of the late 1970s in which radical scholars cast Coalition governments as the agents of a radical new right, unchallenged by a timorous Labor Party, that sought to dismantle social protection and subordinate the Australian economy to multinational capital. The more consensual climate of the 1980s, together with Labor's embrace of economic restructuring and its support for public expenditure restraint, encouraged the emergence of a new critique of labourism. Francis Castles argued that Australia was a wage-earners welfare state that sought to ensure full employment by tariffs and fair income by wage regulation. The result was a delayed development of an Australian welfare state. The focus of Castles' analysis was on federal politics, but Stuart Macintyre and Colin Forster argued that loan-financed developments bolstered rural support for political labour and demand for unskilled labour. Advocates of Castles' approach distinguished between social democracy and labourism. Social democracy was defined as policies of redistribution via taxation and direct government payments, whereas labourism was defined as redistribution by economic regulation. Labourism focused on the rights and conditions of male wage-earner; social democracy was directed towards a broader constituency of the disadvantaged, including those outside the paid workforce, such as women and children. (10)
Taxation policy received little attention from the 1980s critics of labourism. However some interwar writers had argued that Labor's egalitarian commitment only influenced taxation policy at the margin. Pragmatic criteria of political acceptability were considered to have encouraged governments to employ less visible indirect taxes, as exemplified by a Federal Labor government's introduction of a national sales tax in 1930. The history of Australian income taxation provided evidence for this case. The early proponents of income taxation in the Australian colonies were liberals and radicals but the acquiescence of conservative Legislative Councils in the late nineteenth century was a pragmatic response to budgetary deficits. Tax policy followed expenditure policy rather than being a conscious vehicle of redistribution. The major analysis of Labor taxation policy completed in the 1980s was that of Rob Watts. He read Labor's fiscal and social policy revolution of the 1940s, the introduction of uniform taxation and the foundations of the national welfare state, as a pragmatic stabilisation of a wartime capitalist economy at the expense of the working class. (11)
In the 1980s Castles' concept of the wage-earners welfare state informed radical critiques of the Hawke government. The government's (early) commitment to wage indexation and tax cuts to safeguard working-class incomes was evaluated as labourist and gendered. Critics argued that that Labor's commitment to tax restraint required public sector retrenchment with negative impacts for the disproportionately female constituencies of public sector workers and social welfare beneficiaries. Labor's support of centralised wage-fixation also froze existing gender-based wage inequalities. (12)
The radical critique of corporatism ebbed as Labor's policies became increasingly neo-liberal but the late 1980s saw the emergence of a new feminist critique of historical labourism. This critique echoed the anti-corporatist arguments of the mid-1980s. Marilyn Lake highlighted the focus of labourism on the rights of the male breadwinner at the expense of women. She argued that Australian feminists had campaigned on a 'non-party basis' for a 'materialist welfare-state' centred on policies such as child endowment and health services for mothers and children. This model implied that Labor governments would not be associated with distinctive patterns of social expenditure; indeed their commitment to male working-class living standards and the institution of the 'family wage' suggested they would oppose them. Lake's emphasis on the non-party origins of the maternalist welfare state was taken up by Marian Sawer with a different inflection. Sawer emphasised the congruence between feminism and social liberalism and suggested the existence of a cross-party consensus in Australia around support for the 'ethical state'. (13) Her analysis suggested that there would be little distinctive about the record of Labor in government, at least before Whitlam.
The conclusion we would draw from the recent scholarship on labourism is that state Labor governments in 1911-40 would have been expected to prioritise the interests of male wage-earners, through industrial arbitration and loan-financed public works programs, and would not be associated with high levels of social expenditure.
Australian State Politics, 1911-40
The second decade of the twentieth century was a period of fundamental transformation in Australian politics. In all states and the Commonwealth a two-party politics emerged that pitted Labor against a united liberal-conservative bloc. From the early 1920s the new Country Party quickly found its place on the political right, with occasional exceptions in Victoria. Majority Labor governments were elected in all states and the Commonwealth in 1910-15 apart from Victoria. The Commonwealth Constitution, under the so-called 'Braddon clause', provided that for the first decade after federation at least three quarters of Commonwealth customs revenue was to be paid to the states. The expiry of the Braddon clause enabled a federal Labor government in 1910 to legislate the Surplus Revenue Act, which replaced the payment of a fixed portion of Commonwealth revenues by fixed per capita payments. This Act was a landmark in the emergence of Commonwealth financial dominance. The Commonwealth had already entered the sphere of direct income support with old-age pensions in 1908 and in 1912 it continued this expansion with the introduction of the maternity bonus. In 1915 the Commonwealth introduced a national income tax. (14)
As demonstrated in Table 1, Labor's fortunes at a state level recovered fairly rapidly from the party split of 1916. In Queensland the party remained in office through the split and lost only one election during this period. In the other states, except Victoria, Labor was back in power within two elections after the party split. Even in Victoria Labor enjoyed its first prolonged periods in government during the 1920s. Labor also became more ideologically divided during this period. In Queensland and Western Australia the party organisation and parliamentary party were dominated by moderates aligned with the Australian Workers Union. In New South Wales an alliance of left-wing unions secured control of the party organisation and allied with parliamentary leader Jack Lang. In Victoria, South Australia and Tasmania severe division emerged between the party organisation and parliamentarians during the early 1930s that led to major splits in the party.
The institutional and cultural underpinnings of Australian politics in this period suggest that political parties would be well-placed to influence public policy. The administrative apparatuses of the states were professional bureaucracies. The interests of many voters were directly linked to the state, whether as employees, pensioners or holders of government debt. Michael Mann suggests that in Europe the overall economic role of governments declined during the nineteenth century as a major increase in social and infrastructural expenditure was outweighed by a decline in military expenditure. In Australia however the colonies were largely dependent on the imperial government for defence, and Australian levels of public expenditure grew steadily in the later nineteenth century. The adult male franchise came early to Australia and turnout levels were high even before the introduction of compulsory voting. The integration of low income earners into the political process distinguished interwar Australia. In Europe electoral participation was often low in the new democracies of the 1920s. In the United States, racial disenfranchisement reduced the influence of low-income earners, and the weak party system and divided government structure discouraged turnout among poor white voters. (16)
Most Labor governments faced Legislative Councils controlled by non-Labor parties. Only in Queensland, where Labor controlled the Legislative Council from 1920 and abolished it in 1922, was Labor's legislative power unconstrained. In New South Wales Labor enjoyed periods of Council control in 1925-26 and 1931-32 but these were short-lived due to internal party divisions. In the other states the Councils remained under conservative control. Their existence certainly confined the fiscal autonomy of Labor governments, but their power was less impressive in practice than theory, public expenditure was popular and the requirements of budgetary balance often forced Councils to acquiesce in taxation increases.
Social Expenditure: Definitions and Patterns (17)
The analysis of social expenditure in this article focuses on two aspects: firstly, specific payments to individuals in return for which neither goods nor services are supplied, such as pensions (this includes some payments in kind in particular Depression-era food relief); and secondly, social services, primarily health and education. My analysis of social services expenditure is restricted to recurrent expenditure, mostly wages and salaries but also purchases of goods and services. Capital expenditure, on works such as hospitals and schools, is excluded because the available statistical records are incomplete.
Over these 30 years per capita levels of Australian social expenditure in 1911 prices rose from 1.32 [pounds sterling] in 1911 to 4.06 [pounds sterling] in 1940, an increase of 208 per cent--substantially more than Butlin's estimate for GDP growth in 1911-39 of 48 per cent. (18)
The most rapidly increasing area of public expenditure as shown in Figure 1 was direct income support, such as pensions, particularly from the Commonwealth. State government direct payments remained low until the introduction of child endowment and widows' pensions in New South Wales and then the upsurge in unemployment relief payments during the Great Depression.
Another way of looking at public expenditure is displayed in Figure 2. Here I have distinguished programs by client group and adopted Theda Skocpol's distinction between a maternalist and a soldiers' welfare state. In the maternalist category I include child endowment, widows' pensions and the maternity bonus and the very small direct expenditure on women's and children health services. Repatriation was overwhelmingly a federal government responsibility. The tentative steps Australia had made towards a maternalist welfare state after the maternity bonus of 1912 were swamped by the rise of a soldiers' welfare state. (19) The upsurge in maternalist expenditure from the 1920s was due to the introduction of child endowment and widows' pensions by a Labor government in New South Wales. Falling family incomes during the Depression made many more families eligible for child endowment in New South Wales and led to an increase in expenditure.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
Once the immediate fiscal crisis of the early 1930s was past, governments reduced direct unemployment relief (paid from recurrent expenditures) in favour of relief works largely financed by loan expenditure. In part these programs substituted for the public works programs of the 1920s which had paid award wages. The changing patterns of unemployment relief expenditure during the 1930s largely indicated how quickly states returned to the loan market after the alarms of the 1930s rather than their generosity towards the unemployed. For this reason I have excluded unemployment relief from the subsequent analysis of social expenditure. In the early 1930s when states were largely excluded from the loan market New South Wales, as shown by Figure 3, was the most generous in unemployment relief. In 1931-32 NSW unemployment relief per unemployed person at current prices was 28.6 [pounds sterling] compared to a national average of 21.8 [pounds sterling]. (20)
[FIGURE 3 OMITTED]
Population Dispersal, Commonwealth Grants and Taxable Capacity
From 1909 the Commonwealth made special grants to Tasmania and later to Western Australia and South Australia, nominally to compensate them for the impact of Commonwealth legislation, in particular tariffs and arbitration, but also to help bring state government services up a level comparable to other states. Tasmanian taxation had more than doubled from 1901 to 1910 as it struggled to bring its public sector up to mainland standards. In 1933 the Commonwealth Grants Commission (CGC) was established to rationalise the allocation of Commonwealth grants. The CGC adopted the principle that special grants were justified when a state was unable despite a 'reasonable' effort in revenue collection 'to function at a standards not appreciably below that of other States'. In 1940 7.7 per cent of South Australia total revenue, 17.5 per cent of Western Australian total revenue and 17.4 per cent of Tasmanian total revenue was made up of Commonwealth grants. (22) The system of Commonwealth grants redistributed income from taxpayers in the wealthier core states to those in the periphery.
A difficulty in the comparison of the states is the higher per capita cost of the provision of social services in states with a more dispersed population. From 1937 the CGC introduced 3 per cent allowances for South Australia and Tasmania and 7 per cent for Western Australia. (23) These were increased further in the post-war period, but for the purposes of the further analysis in this article as displayed in Figures 4 to 6 I have adjusted South Australian, West Australian and Tasmanian expenditure downward by the 1937 percentages to provide a more accurate representation of their commitment to social expenditure.
Any analysis must also take into account the fact that more prosperous states would be expected to have higher expenditure levels. Estimates of the size of the state economies are primitive for this period but there was a tendency towards convergence between the states. In 1909 the Commonwealth Bureau of Census & Statistics estimated that the per capita value of output in Western Australia was 49 per cent above the national average, but in 1940 the most prosperous state was Queensland with a per capita value of output only 3 per cent above the national average. However the convergence of state economic performance was never as complete as some had predicted before Federation. The CGC argued that any comparison of taxation levels between the states should take into account the divergent 'taxable capacities' of the states. (24)
[FIGURE 4 OMITTED]
The overall pattern of social expenditure is illustrated in Figure 4. In 1911 there were still significant divergences between the states in social expenditure. Western Australia's high rating suggests that perhaps 3 per cent was an inadequate allowance for dispersed population, but the economic boom generated by the gold industry would also have enabled higher levels of expenditure. There are three notable periods of upheaval, first the post-war inflationary boom that increased wage costs (which were then maintained at this higher level by industrial arbitration) combined with the influenza epidemic, then the emergence of New South Wales as a social policy leader in the late 1920s, and finally the upheavals of the Depression. The high level of expenditure in New South Wales was also in part a product of higher wage costs, due to a shorter working week and a union-friendly state arbitration system. In 1933-34 the CGC found that average male public service clerical wages in New South Wales were about 20 per cent higher than the all states average. (25)
Over these 30 years there was only limited convergence between the states in social expenditure, as illustrated in Figure 5. This was despite the redistributive role of direct Commonwealth grants and the economic convergence between the states.
[FIGURE 5 OMITTED]
A clearer comparison is available if we consider how the relativities between the states changed from 1911 as displayed in Figure 6. The leading position of New South Wales is clear but Tasmania also advanced relative to its 1911 position along with South Australia. The laggard performance of Victoria remains notable although Western Australia also lost ground relative to other states, as its economic boom ebbed. The depression of the 1890s hit Victoria especially hard and constrained its level of social expenditure, so that at Federation it lagged behind most other states, but post-Federation Victorian governments were slow to make up lost ground in social policy. (27) Victoria may have been the last stronghold of Deakinite liberalism but its social policy record calls into question the assumption of cross-party social liberal consensus. The increase in Tasmanian social expenditure is an example of the 'fly-paper' effect; the tendency of governments to expend lump-sum grants on increased expenditure rather than taxation reduction. (28)
[FIGURE 6 OMITTED]
In large part the history of state social expenditure is a story of New South Wales and Victorian exceptionalism. NSW exceptionalism emerges with the first Premiership of Jack Lang in 1925-27. New South Wales provides an example of the 'class politics' paradigm identified by Castles. Although Labor governed infrequently its policy record was more radical, and fear of Labor radicalism impelled conservatives to acquiesce in Labor's social policy advances, contrary to the argument of some of Lang's critics. The distinctiveness of New South Wales and Victoria was noted by contemporary observers from the 1920s and led to the CGC excluding New South Wales from its standard of comparison until 1937. Apart from New South Wales exceptionalism the specific impact of Labor governments may be most apparent in short-term responses to economic crisis. In the early 1930s state Labor governments in New South Wales, South Australia and Victoria presided over expenditure increase and their conservative successors strove with some success to reverse this trend. In the other states Labor governments came to power from 1932 onwards and presided over expenditure increases. (29)
There is evidence from the American states in the interwar period that higher levels of political competitiveness contributed to higher levels of public expenditure. Peter Lindert found a statistically significant relation between frequency of executive turnover and levels of public expenditure in a comparison of 30 countries over the period 1880-1930. The nation that demonstrated the most rapid increase in social expenditure during the 1920s was Weimar Germany as governments beleaguered by coup attempts from the left and right sought to buy support. (30) In the period under study New South Wales was the state with the highest level of executive turnover as displayed in Table 2.
'the revenue ... is the spring of all power' (Edmund Burke) (32)
Taxation was not the only means by which governments could extract resources from the private sector. In the early 1930s governments borrowed on a large scale to support recurrent expenditure. In 1931-32 state governments required 23.15 million [pounds sterling] in short-term bank financial support. However in this article I focus on taxation as the most contentious mode of the transference of resources to the private sector. (33)
Australian Labor was slow to develop a distinctive position on taxation. The Australian colonies relied heavily on customs revenues and land sales. The first proposals for the introduction of income tax long predated Labor's emergence as a political force but income taxes were not introduced until the 1890s. In New South Wales Labor supported George Reid's free-trade government, which legislated a state income tax in 1894. However the introduction of income tax was Reid's priority, not Labor's, and the 1892 NSW Labor platform made no reference to taxation. Instead it pledged support for measures to 'secure for the wage-earner a fair and equitable return for his or her labour'. (34)
European socialists initially shared a similar indifference towards taxation. Karl Kautsky and Otto Bauer argued that in a socialist society public revenues would be drawn from the surplus generated by nationalised industries. In both Europe and Australia the entry to government of social democratic parties led them to adjust their position on taxation. By the early 1920s Kautsky could argue that in the period of transition towards socialism there would be a role for direct taxation to finance public services such as education. He contrasted this taxation policy, which would tax surplus value in the hands of the wealthy with the reliance of capitalist governments on indirect taxation, which depressed wages below the value of labour-power. Australian Labor politicians occasionally evoked the image of a self-financing public sector but in practice relied on taxation as a source of revenue. In the 1920s state enterprise receipts, particularly for railways, fell markedly due to wage increases, road transport competition and the burden of uneconomic developmental lines. By 1933 of all Australian public transport systems only Victorian tramways made a positive contribution to state government revenue. Only authoritarian governments of the left or right have been able to rely on state enterprise profits for revenue and even in the Soviet Union taxation was a major source of government revenue. Frederic Eggleston, in an anticipation of the argument of 1990s Labor supporters of privatisation, argued that unprofitable state enterprises discouraged innovation in social policy due to their drain on revenue. (35)
As Labor governments confronted the need for income tax they adopted the rhetoric of equity. The most substantive exposition of Labor taxation policy in this period came from Queensland railways minister James Larcombe. He defended the principle of 'ability to pay' and argued for steeply progressive tax rates and exclusion of as many wage-earners as possible from income tax. The policies of the first generation of Labor governments were consistent with this approach: they increased thresholds, established higher thresholds for the heads of families and introduced steeply scaled tax schedules. (36)
However the extent to which the application of the principle of 'ability to pay' actually leads to income redistribution depends on how much has to be paid. The ultra-rich are a small minority and very high tax rates on them raise limited revenue. Such taxes are a form of populist political symbolism. At the 1933 Australian census only 11.1 per cent of those who reported an income claimed to have received over 260 [pounds sterling] in the preceding year. (37)
By the 1930s income tax rates were most progressive in Labor-dominated Queensland as shown in Table 3 but the fiscal crisis of the Depression ended Labor's effort to exempt workers from taxation. In the early 1930s all states increased income taxes or established special unemployment relief taxes, and together these drew most wage earners into the income tax net for the first time. (38)
Most states tended to increase their reliance on income tax over this period as shown in Figure 7, but Queensland particularly favoured income tax.
[FIGURE 7 OMITTED]
Over time tax levels across the states converged more than social expenditure. The shift of financial power to the Commonwealth from 1910 forced the states to increase taxation as displayed in Figure 8. Commonwealth grants meant that Tasmania did not have to rely on its resources alone to reach mainland levels of social services. By 1940 Labor-dominated Queensland and conservative Victoria had emerged as the two extremes. In Victoria conservative electoral campaigns stressed their commitment to low taxation. (39) In the early to mid-1930s however Queensland taxation levels had been surpassed by New South Wales and South Australia, in the former governments struggled to finance Labor's social programs whilst South Australia was hard hit by the crisis of the wheat industry.
[FIGURE 8 OMITTED]
If we compare how the relativities between state tax levels changed from 1911, as displayed in Figure 9, it was New South Wales that showed the largest increase in taxation:
[FIGURE 9 OMITTED]
The Impact of Commonwealth Grants
The high levels of taxation in Queensland attracted unfavourable comment but were partly enforced by the state's exclusion from Commonwealth grants. Figure 10 reveals that it was Tasmania whose revenue base was most dramatically increased due both to higher taxes and the receipt of Commonwealth grants. Victoria remained the outlier.
[FIGURE 10 OMITTED]
State taxation levels and the total of taxation and grants converged in the late 1930s more than ever before as Figure 11 displays.
[FIGURE 11 OMITTED]
At the level of a simple bivariate analysis it is difficult to identify a clear relation between Labor political success, levels of taxation and the overall resources claimed by governments. The ability of state governments to increase taxation reflected not only political will but also differing levels of taxable capacity. From 1911 to 1938 if we discount the increases in state taxation and grants by the taxable capacity of the states it is still Tasmania that stands out for its commitment to increasing the resource base of government but Labor's bulwark of Queensland came second.
The pattern of state taxation was overlaid by the redistributive role of federal finance. As a result of federal grants South Australian, Western Australian and Tasmanian Labor governments faced less pressure to increase taxation, although Tasmania still had to make substantial sacrifices to approach mainland service levels. The core states had more difficult political choices. Victoria opted for the low revenue and expenditure road, Queensland Labor increased taxation and was able to carry the political cost in the 1930s, but in New South Wales Lang's Labor governments tended to spend first, and establish a constituency for social programs. Conservative governments in New South Wales had to substantially increase taxation in 192728 and 1932-33 despite the complaints of many of their supporters and imposed a heavier burden on the wealthy than in Victoria. In 1930 a NSW Labor government found that it could only finance continuation of its social programs by extension of the reach of income taxation to wage earners and despite grumbling from many Labor supporters it secured trade union acquiescence by the promise of favourable industrial legislation. (42)
Types of Expenditure
The model of labourism proposed by Castles suggests that Labor governments would prioritise public works financed by loans to generate jobs for male manual workers and that their taxation policy would strive to exclude manual workers from taxation. By the 1930s state Labor governments had moved away from this model towards a more social democratic position that implied a broader tax base and a commitment to social expenditure. In the previous sections of this article I have focused on the level of recurrent expenditure but beyond this there is the broader question of the balance between expenditure financed by taxation and that financed by loans. In this section I broaden the analysis to include total government expenditure, including that financed by loans.
In all states the significance of social expenditure increased relative to the nineteenth century staples of infrastructure, but a division did emerge between core and peripheral states as displayed in Figure 12. Particularly from the late 1920s, Tasmania, New South Wales and Victoria devoted more resources to social expenditures. New South Wales and Victoria were distinctive in their levels of taxation, and hence the size of their public sector, but they did allocate similar shares of public expenditure to social programs. Here I would qualify Barnard's argument that from the 1920s the dynamic element of public expenditure shifted towards social expenditure. New South Wales accounted for much of this shift in aggregate terms and the states differed in their commitment to social expenditure. (43)
[FIGURE 12 OMITTED]
Class Politics or Social Democracy?
Despite the increasing dominance of the Commonwealth government in these years the Australian states remained distinctive in their financial policies. Convergence was limited and much of it was accounted for by Tasmania's struggle to bring its nineteenth century economy into the twentieth century. Victoria, where Labor was politically marginal, adopted a distinctively conservative position on fiscal policy, but the records of state Labor governments reveal differing approaches. Politically hegemonic Queensland Labor was able to implement a high taxation strategy whereas NSW Labor, once Lang had emerged as the dominant figure in the party, tended to prioritise expenditure. For nearly all of the 1930s New South Wales and Queensland led in per capita social expenditure. This reflected not only Lang's personal style but the strength of radical elements in the NSW labour movement together with the influence and political skills of Labor feminists such as Kate Dwyer who campaigned for a maternalist welfare state. Queensland came close to exemplifying a social democratic labour movement strategy; New South Wales had more resemblance to the class politics paradigm. (45)
The analysis in this article calls into question the assumption of Australian distinctiveness, at least at the level of working-class political mobilisation. Certainly interwar Labor governed within an environment notably divergent from that of Europe during this period or that of Australia in later decades. However the ALP remained the political representative of a subordinate social class. The working class was not the ruling class in interwar Australia. Labor worked within the institutions of the Australian political economy, such as White Australia, tariffs and industrial arbitration, but it did not create them. Labor's first politics of initiative sought to drastically expand state socialism by the nationalisation of monopolies and the establishment by state governments of major industrial enterprises. This project was politically contentious, it evoked strong opposition from liberal protectionists such as Alfred Deakin, and it remained unfulfilled. The second generation of Labor government elected in the states from the 1920s had to respond to this policy vacuum. Their response was a turn towards social democracy. The extent to which this shift was effective varied from state to state. The next stage in this research project will be the application of more sophisticated statistical techniques to distinguish the varying influences of economy, demography and politics and to estimate the overall redistributive impact of public expenditure and taxation. (46)
(1.) R. Goldscheid, 'A Sociological Approach to Problems of Public Finance' (1925), in R.A. Musgrave & A. Peacock (eds), Classics in the Theory of Public Finance, Macmillan, London, 1958, p. 203.
(2.) C. Webber & A. Wildavsky, A History of Taxation and Expenditure in the Western World, Simon & Schuster, New York, 1986, pp. 570-72; R. Mule, Political Parties, Games and Redistribution, Cambridge University Press, Cambridge, 2001, pp. 33-35; P.H. Lindert, Growing Public: Social Spending and Economic Growth Since the Eighteenth Century, Vol. 1, Cambridge University Press, New York, 2004, pp. 20-36; H.L. Wilensky, The Welfare State and Equality: Structural and Ideological Roots of Public Expenditure, University of California Press, Berkeley, 1975, p. 5
(3.) S. Hansen, Globalization and the Politics of Pay: Policy Choices in the American States, Georgetown University Press, Washington, DC, 2006, pp. 6-9; P. Conceicao, P. Ferreira & J. K. Galbraith, 'Inequality and unemployment in Europe: the American cure', in J.K. Galbraith & M. Berner (eds), Inequality and Industrial Change: A Global View, Cambridge University Press, Cambridge, 2001, pp. 109-11.
(4.) J. Bryce, Modern Democracies, Vol. 2, Macmillan & Co., London, 1921, pp. 181, 189, 231, 258, 281-83, 355; W. Hancock, Australia, Ernest Benn, London, 1945 (first ed. 1930), pp. 140, 179; F.W. Eggleston, 'Political parties and their economic policies', in D.B. Copland, (ed.), An Economic Survey of Australia (The Annals of the American Academy of Political and Social Science, vol. 158), Philadelphia, 1930, p. 250.
(5.) Hancock, Australia, pp. 185-199. J.F. Cairns, The Welfare State in Australia: A Study in the Development of Public Policy, PhD thesis, University of Melbourne, 1957, p. vii; F.W. Eggleston, Reflections of an Australian Liberal, F.W. Cheshire, Melbourne, 1953, p. 151.
(6.) Royal Commission on Taxation, Fourth Report, Commonwealth Parliamentary Papers (CPP), 1923-24, vol. 3, para. 613, 680; F. Mann, 'The sociology of taxation', Review of Politics, vol. 5, no. 2, 1943.
(7.) S. Sherlock, '"Good-bye the state's progress": state enterprise and Labor's plan for a north Queensland steel industry, 1915-20', Labour History, no. 90, 2006; R. Fitzgerald & H. Thornton, Labor in Queensland: From the 1880s to 1988, University of Queensland Press, Brisbane, 1989, pp. 69-137; B. Jansson, The Sixteen-Trillion Dollar Mistake: How the U.S. Bungled Its National Priorities from the New Deal to the Present, Columbia University Press, New York, 2001, p. 2.
(8.) H. Mayer, 'Some conceptions of the Australian party system 1910-1950', in M. Beever & F. B. Smith (eds), Historical Studies: Selected Articles, Second Series, University of Melbourne Press, Melbourne, 1967, p. 234; N.G. Butlin, 'Colonial socialism in Australia 1860-1900', in H.G.J. Aitken (ed.), The State and Economic Growth, Social Science Research Council, New York, 1959; N.G. Butlin, A. Barnard & J.J. Pincus, Government and Capitalism: Public and Private Choice in Twentieth-Century Australia, George Allen & Unwin, Sydney, 1982, pp. 10-48.
(9.) R. Gerritsen, 'State Budgetary Outcomes and Typologies of the Australian States', in B. Galligan (ed)., Comparative state policies, Longman Cheshire, Melbourne, 1988, p. 149. H. McQueen, Gone Tomorrow: Australia in the 1980s, Angus & Robertson, Sydney, 1982, pp. 112-16. B. Galligan, Utah and Queensland Coal: A Study of Micro Political Economy of Modern Capitalism and the State, University of Queensland Press, Brisbane, 1989, pp. 12-14.
(10.) R. Catley & B. McFarlane, Australian Capitalism in Boom and Depression, Alternative Publishing Co-operative Ltd, Sydney, 1981, pp. 221-24; M. Simms, A Liberal Nation: The Liberal Party and Australian Politics, Hale & Iremonger, Sydney, 1982, pp. 143-165; F.G. Castles, The Working Class and Welfare in Australia and New Zealand, Allen & Unwin, Sydney, 1985; C. Forster, 'The economy, wages and the establishment of arbitration', in S. Macintyre & R. Mitchell (eds), Foundations of Arbitration: The Origins and Effects of State Compulsory Arbitration 1890-1914, Oxford University Press, Melbourne, 1989.
(11.) S. Mills, Taxation in Australia, Macmillan & Co., London, 1925, p. 263. J. H. Gilbert, The Tax Systems of Australasia, University of Oregon Monographs, Eugene, Oregon, 1943, pp. 23, 30. P.A. Harris, Metamorphoses of the Australasian Income Tax: 1866 to 1922, Australian Tax Research Foundation Research Study No. 37, 2002; pp. 17, 45, 63, 130, 141; R. Watts, The Foundations of the National Welfare State, George Allen & Unwin, Sydney, 1987, pp. 88-103.
(12.) S. Macintyre, 'The short history of social democracy in Australia', in D. Rawson (ed.), Blast, Budge or Bypass: Towards a Social Democratic Australia, Academy of the Social Sciences in Australia, Canberra, 1984, pp. 12-13; C. Bulbeck, 'A review of the Accord's first three years', Thesis Eleven, no. 14, 1986; P. Beilharz, Transforming Labor: Labour Tradition and the Labor Decade in Australia, Cambridge University Press, Melbourne, 1994, pp. 132, 148-55.
(13.) M. Lake, 'The politics of respectability: identifying the masculinist context', in S. Magarey, S. Rowley & S. Sheridan (eds), Debutante Nation: Feminism Contests the 1890s, Allen & Unwin, Sydney, 1993; M. Lake, Getting Equal: The History of Australian Feminism, Allen & Unwin, Sydney, 1999; M. Sawer, The Ethical State? Social Liberalism in Australia, University of Melbourne Press, Melbourne, 2003.
(14.) A. Barnard, Australian Government Finances: A Statistical Overview, 1850-1982, Australian National University Working Paper in Economic History, no. 59, December 1985, p. 25.
(15.) This does not include the one-day Fuller Nationalist government in New South Wales on 20 December 1921.
(16.) M. Mann, The Sources of Social Power, Vol. 2: The Rise of Classes and Nation-States, 17601914, Cambridge University Press, Cambridge, 1993, pp. 360-81, 479, 499-500; L.E. Davis & R.A. Huttenback, Mammon and the Pursuit of Empire: The Political Economy of British Imperialism, 1860-1912, Cambridge University Press, Cambridge, 1986, pp. 20 22, 122, 162; Bryce, Modern Democracies, Vol. 2, pp. 195-97; Lindert, Growing Public, pp. 80-86.
(17.) This article draws on contemporary statistical publications and two later compilations. The major contemporary reports are: 1) the Finance Bulletins of the Commonwealth Bureau of Census & Statistics; 2) after 1933 the reports of the Commonwealth Grants Commission (CGC) and 3) the analyses produced for the Premiers' Conferences of the early 1930s. In 1975 Andrew Podger used Finance Bulletins and the CGC reports to complete a database of Australian government social expenditure from 1901: A. Podger, 'Social welfare expenditure 1900-1970', in R. Mendelsohn, The Condition of the People: Social Welfare in Australia 1900-1975, George Allen & Unwin, Sydney, 1979 (hardback edition). In 1985-86 Alan Barnard published a series of Australian National University Source Papers in Economic History based on the Finance Bulletins that summarised Australian state and Commonwealth government expenditure and taxation from the first available records up until the early 1980s: Preliminary Statistics of N.S.W. Government Finances, 1850-1982 (No. 8, December 1985); State and Local Government Finances in Queensland, 1859-1982 (No. 9, December 1985); Tasmanian Government Finances, 1953 to 1982: A Statistical Survey (No. 14, December 1986); Finances of Western Australian Governments, 1851-1982 (No. 16, December 1986); The South Australian Budgetary Record 1850-1982 (No. 10, December 1985); Victoria's Fisc, State and Local: Preliminary Statistics (No. 15, December 1986). My analysis of public expenditure is based on Podger's collection and that of taxation on Barnard's work. There are some curious fluctuations in the figures reported by both scholars but a check of state-level statistical publications reveals that these were the totals advised by state governments at the time. In most cases I have expressed taxation and expenditure in 1911 prices based on a combination of the all-states 'A' and 'C' price indexes of the Commonwealth Bureau of Census and Statistics. I have then calculated per capita taxation and expenditure using the most recent Australian Bureau of Statistics historical estimates of state populations. The price indices are from W. Vamplew (ed.), Australians: Historical Statistics, Fairfax, Syme & Weldon, Associates, Sydney, 1987, tables, PC 10-15, PC 16-20. The population estimates are from Australian Bureau of Statistics, Australian Historical Population Statistics (Catalogue No. 3105.0.65.001), 2004. I have expressed monetary values in decimals rather than the archaic and confusing terminology of shillings and pence.
(18.) N.G. Butlin, Australian Domestic Product, Investment and Foreign Borrowing, 18611938/39, Cambridge University Press, Cambridge, 1962, table 13.
(19.) T. Skocpol, Protecting Soldiers and Mothers: Political Origins of Social Policy in the United States, Harvard University Press, Harvard, 1995.
(20.) G. Snooks, 'Government unemployment relief in the 1930s: aid or hindrance to recovery?', in R.G. Gregory & N.G. Butlin (eds), Recovery from the Depression: Australia and the World Economy in the 1930s, Cambridge University Press, Melbourne, 1988; B. Stevens, 'A Treasurer's problems', Australian Quarterly, no. 1, 1929, p. 23; Commonwealth Grants Commission (CGC), Second Report, the Commission, Canberra, 1935, p. 68; CPP, 1934-37, vol. 4, pt. 2. Record of the Conference of Commonwealth and State Ministers held...28th January to 5th February 1932, p. 29 (National Archives of Australia, Canberra, AS954/69, Item 2390/16).
(21.) Expenditure data are from Snooks, 'Government unemployment relief in the 1930s'. Unemployment levels per state are from the 1933 census scaled for 1930-32 and 193440 by the rate of trade union unemployment from Labour Reports.
(22.) CGC, Third Report, p. 75, CPP, 1934-37, vol. 4, pt. 2; May, Financing the Small States, table 4; Smith, 'Redistribution and federal finance', p. 297.
(23.) R.J. May, Financing the Small States in Australian Federalism, Oxford University Press, Melbourne, 1971, p. 91.
(24.) May, Financing the Small States, table 6 (b); J. P. Smith, 'Redistribution and Federal Finance', Australian Economic History Review, vol. 42, no. 3, 2002, p. 294.
(25.) Barnard, Australian Government Finances, p. 42; CGC, Second Report, p. 149.
(26.) Variance is the standard deviation divided by the mean.
(27.) Mendelsohn, The Condition of the People, p. 94.
(28.) B.E. Dollery & A.C. Worthington, Fiscal Illusion and the Grantor Government in Australia: An Indirect Test of the Flypaper Hypothesis, Federalism Research Centre, Australian National University, Discussion Paper No. 25, Canberra, 1995.
(29.) Castles, Working Class and Welfare, pp. 1-9; J. Hagan & K. Turner, A History of the Labor Party in New South Wales 1891-1991, Longman Cheshire, Melbourne, 1991, p. 138; CGC, Equality in Diversity; History of the Commonwealth Grants Commission, AGPS, Canberra, 1995, para. 5.54; J. Nimmo, 'The consumption standard', in F.W. Eggleston et al, Australian Standards of Living, Melbourne University Press, Melbourne, 1939, p. 181.
(30.) Webber & Wildavsky, Taxation and Expenditure, pp. 448-50; Lindert, Growing Public, p. 175; J.T. Patterson, The New Deal and the States: Federalism in Transition, Princeton University Press, New Jersey, 1969, pp. 162-63.
(31.) I do not count as a change of government the replacement of the Victorian Peacock Liberal ministry after the 1917 election by Nationalists who had been known as the 'Economy Party', but I do count the minority Victorian Labor governments formed after the 1924, 1927 and 1929 elections. The one-day Nationalist government of 20 December 1921 in New South Wales is also excluded. Information on changes of government is from: C. Hughes & B.D. Graham, A Handbook of Australian Government and Politics 1890-1964, Australian National University Press, Canberra, 1968.
(32.) E. Burke, Reflections on the French Revolution, Penguin, Harmondsworth, 1969, p. 351.
(33.) Record of the Conference of Commonwealth and State Ministers...10th August 1931 to 12th September 1931, Commonwealth Parliamentary Papers, 1929-32, vol. 2, p. 401. Taxation statistics are extracted from Barnard's compilations.
(34.) Platform of the Labor Electoral League of New South Wales (1892) in R.N. Ebbels (ed.), The Australian Labor Movement 1850-1907: Extracts from Contemporary Documents, Cheshire-Lansdowne, Melbourne, 1965, pp. 215-17; B. Nairn, Civilising Capitalism: The Beginnings of the Australian Labor Party, Melbourne University Press, Melbourne, 1989 (first ed. 1973), p. 138.
(35.) K. Kautsky, The Agrarian Question, Vol. 2, Zwan Publications, London, 1988, pp. 427-26; O. Bauer, The Question of Nationalities and Social Democracy, University of Minnesota Press, Minneapolis, 2000, pp. 405-06; K. Kautsky, The Labour Revolution, George Allen & Unwin, London, 1925, p. 211;Cairns, The Welfare State in Australia, p. 244; Barnard, Australian Government Finances, p. 22. CGC, First Report, the Commission, Canberra, 1933, p. 106; Mann, 'The sociology of taxation', p. 225; Barnard, Australian Government Finances, p. 25. Mann, The Sources of Social Power, Vol. 2: The Rise of Classes and Nation-States, pp. 387-88; Eggleston, Reflections, p. 149.
(36.) J. Larcombe, Labour Government and Financial Administration, Government Printer, Brisbane, 1927(?), pp. 2-5; J.P. Smith, Taxing Popularity: The Story of Taxation in Australia, Australian Tax Research Foundation, Research Study No. 43, 2004, p. 52. Harris, Australasian Income Tax, pp. 160-66, 176, 180. K. Wiltshire, 'Public Finance', in D.J. Murphy, R.B. Joyce & C.A. Hughes (eds), Labor in Power: The Labor Party and Governments in Queensland 1915-57, University of Queensland Press, Brisbane, 1980.
(37.) M. Leff, The Limits of Symbolic Reform: The New Deal and Taxation, 1933 -1939, Cambridge University Press, Cambridge, 1984, p. 115; Census of the Commonwealth of Australia, 30th June 1933, Government Printer, Canberra, 1936, table 28.3.
(38.) Barnard, Australian Government Finances, p. 25; J.H. Gilbert, The Tax Systems of Australasia, University of Oregon Monographs, Eugene, Oregon, 1943, p. 23.
(39.) Barnard, Government Finances, pp. 27-28; C. Cleary, Ballarat Labor: From Miner Hesitancy to Golden Age, the author, Epson, Vic., 2007, pp. 69, 245.
(40.) Variance is the standard deviation divided by the mean.
(41.) The 1938 estimate of taxable capacity is from the CGC (CGC, Sixth Report, the Commission, Canberra, 1939, p. 75; CPP, 1937-40, vol. 4, pt. 3). For 1911 I have applied the estimate of taxable capacity in 1919 by Leslie Giblin which is the earliest available (D.B. Copland, 'Some problems of taxation in Australia' (1924), in W. Prest & R.L. Mathews (eds), The Development of Australian Fiscal Federalism: Selected Readings, Australian National University Press, 1980, p. 40).
(42.) T. Matthews, Business Associations and Politics: Chambers of Manufacturers and Employers' Federations in New South Wales, Victorian and Australian National Politics to 1939, PhD thesis, University of Sydney, 1971, p. 286; G. Robinson, When the Labor Party Dreams: Class, politics and policy in New South Wales 1930-32, Australian Scholarly Publishing, Melbourne, 2008, pp. 66-72.
(43.) Barnard, Government Finances, p. 7.
(44.) There are gaps in the series for 1917, 1918 and 1920.
(45.) K. Deverall, They did not know their place": the politics of Annie Golding and Kate Dwyer', Labour History, no. 87, 2004.
(46.) Some very preliminary conclusions on the redistributive impact of fiscal policy are in G. Robinson, 'Public finance and income redistribution in interwar Australia: towards a class analysis, in Kimber, J., Love, P. and Deery, P. (eds), Labour Traditions: Papers from the tenth national labour history conference, pp. 165-169, Australian Society for the Study of Labour History, Melbourne, 2007.
Geoffrey Robinson * Geoffrey Robinson is a lecturer in History and Politics at Deakin University, Geelong. His analysis of the 1930-32 Labor government in New South Wales has recently been published as When the Labor Party Dreams: Class, Politics and Policy in New South Wales 1930-32.
Table 1: State Labor Governments, 1910-40 Percentage of days Periods in Government in Government New South Wales October 1910 to November 1916 38.5% April 1920 to April 192215 June 1925 to October 1927 November 1930 to May 1932 Victoria Dec-13 July 1924 to November 1924 May 1927 to November 1928 13.9% December 1929 to May 1932 Queensland June 1915 to May 1929 June 1932- 72.6% South Australia June 1910 to February 1912 April 1915 to July 1917 32.1% April 1924 to April 1927 April 1930 to April 1933 Western Australia October 1930 to July 1916 April 1924 to April 1930 59.7% April 1933 - Source: C. Hughes & B. D. Graham, A Handbook of Australian Government and Politics 1890-1964, Australian National University Press, Canberra, 1968. Table 2: Changes of government, 1911-40 (31) NSW Vic Qld SA WA Tas All changes of 10 8 4 8 6 6 government Changes resulting from 6 4 4 6 4 4 general elections Table 3: Annual tax payable on personal exertion income, 1934 Annual income NSW Vic Qld SA WA Tas [pounds sterling] 100 0 0.1 1.2 0 0.8 1.7 150 5.5 2.1 3.7 2 1.6 2.9 200 7.5 2.9 5.4 5.4 4.3 4 250 9.8 4.6 10 9.7 6.3 5.7 300 12.4 6.1 12.8 14.5 8.3 7.5 400 19 10 19 24 16 12 500 26 15 33 34 23 18 700 43 32 57 59 35 31 1,000 70 60 117 91 56 54 Source: Commonwealth Grants Commission, Second Report, 1935, p. 118 Table 4: Labor governments and per capita state taxation, 1911-38 (adjusted for taxable capacity) (41) Increase state Increase state Percentage Increase in state taxation and grants Labor taxation (%) (%) NSW 43 578 177 Vic 15 178 72 Qld 80 668 250 SA 36 256 167 WA 57 389 118 Tas 38 600 313…