Bob Gould, 2002

The People’s Choice: Electoral politics in 20th century NSW
A review

Source: Self-published pamphlet, August 10, 2002
Proofreading, editing, mark-up: Steve Painter

The People’s Choice. Electoral Politics in 20th century NSW. Three vols, edited by Michael Hogan and David Clune, published by the Parliament of NSW, 2001

These three volumes are extremely elegant. As a professional bookseller, who also happens to be a political agitator, researcher, writer and ideologue, I generally distinguish between books as objects and books for my own research and consumption. The elegant books as objects I tend to treat as tradeable commodities, and have quite a serious interest in them from that angle, but my own personal collection of 20,000 or so reading and working books, are anything but books as objects. I tend to keep the most battered copy for my own use, and flog the ones in better condition.

Occasionally, books come along that are useful for my core interests and beautiful objects as well. An earlier example of such a book is Shirley Fitzgerald’s useful history of Sydney. The three volumes I am discussing here, combine spectacular design, and visual excitement (achieved by incorporating a kind of cartoonists account of NSW politics, and other first class illustrations), with enormous utility, for the serious student of politics, class and social relations.

The only defect, as the editors acknowledge, is the lack of coloured maps of election results, and no list of results seat by seat, for which they refer the reader to the useful book by B.D. Graham. Maybe the Parliament of NSW could also afford a supplementary volume updating Graham, including detailed coloured maps of results. Such a four-volume set would be a delight indeed, and an enormously useful research tool.

The evolving demographics of NSW politics in the 20th century. Significant changes, but surprising elements of continuity, in the interface of social class, ethnicity and politics, particularly in the division between the Labor and conservative sides of electoral politics.

It is possible to get a vivid insight into the social dynamics of life and politics in NSW if one considers the evidence from the electoral sphere, assembled in The People’s Choice, alongside, and in dynamic interaction with, Shirley Fitzgerald’s and Peter Spearritt’s useful books on Sydney, and, most importantly, with the indispensable, and reasonably accessible Social Atlases and other summary statistical publications, produced for the past twenty years or so by the extremely efficient Australian Bureau of Census and Statistics. These bureau publications provide an insight into occupational structure, ethnicity, age, income and other issues.

It is necessary to bring all these factors into relationship with each other to get a useful picture. Happily, a serious overview of People’s Choice, the Fitzgerald and Spearritt books, and the Census Bureau publications makes such an inquiry possible.

In my view, most discussion of political trends, elections, and popular sociology, in the media, in universities, and on the political left, is hopelessly superficial. Such discussion almost never addresses the interface between the above network of factors.

A good example of this superficiality is the current discussion of the question of so-called “aspirational voters”. Demagogues like Mark Latham and Alex Sanchez, and their supporters in the bourgeois media, seize on an apparent phenomenon, which they claim is new, to justify proposals that would move the labour movement dramatically to the right.

However, when you examine the phenomenon of “aspirational voters”, it is in fact one of the oldest issues in NSW electoral politics, and is always, and has always been, mediated by occupational location, ethnicity, religion and other factors. In The People’s Choice the conservative electoral alignment of outer suburban areas, at some points in the political cycle, is documented for different periods.

In a number of illuminating tables, The People’s Choice also documents that these outer suburban areas have swung as solidly to Labor at other points of the cycle than they have to the Liberals in the 1996 and 2001 federal elections.

“Aspirational voters” in these outer suburban areas, actually swung to Labor in the most spectacular way of all in four very different contexts: the first Lang Depression election in 1930 (partly because of Lang’s populist radicalism), in 1941 (despite a three-way split in the labour movement), in the 1953 election, and in the “Wranslide” of 1981. These outer suburban seats largely swung to Labor in the 1999 state election, although not as sharply as in the other four instances.

Major electoral landmarks in NSW

The People’s Choice commences with the 1901 election. It is ironic that one major factor in this election was Australia’s involvement in a foreign war, that of British Imperialism against the Boers in South Africa. Jingos attacked one Labor leader, W.A. Holman, because of his opposition to the Boer War, but he held the country seat of Grenfell despite the “patriotic” brouhaha.

It is ironic that as we run up to the first state election of the new century, one of the background issues could conceivably be Australia’s involvement in the war, of the now dominant American imperialism, on Iraq. Tories in Australia have the dogged habit of sending Australians off to die in the foreign wars of the dominant world imperialism.

The People’s Choice also records the opposition of the Catholic Cardinal, Daniel Moran, to the Boer War, and the cutting edge that this gave to the then underlying sectarian divide in NSW politics, with Catholics on the Labor or Protectionist side, and Protestant bigots on the conservative side. The opposition of Protestant bigots to alcohol, gambling and most sports tended to sharpen these cultural divisions.

The striking thing about the 1901 election is the small size of the electorate. In 1901, NSW, and most Australian states, had a broader franchise than most other countries, and Australia was noted for its early extension of the franchise. Nevertheless, by modern standards, this franchise was still extremely limited.

Voting was confined to male British subjects older than 21. You had to go through a very elaborate procedure to get on the roll and claim a voting right, which operated against itinerant workers and many others. Voting was not compulsory. Of the 1,354,846 NSW population, only 298,673 were on the roll, and only 196,514 voted.

Labor got 36,427 or 18.68 per cent of the vote, to win 24 seats of the 125 available. The Labor votes were concentrated in overwhelmingly proletarian inner-Sydney seats, where nearly a third of the population were Irish Catholics, in mining seats in Newcastle and the Illawarra, and in some country areas, mainly inland wheat farming or pastoral areas (usually, again, places with an unusually high proportion of Catholics).

Census and Statistics Bureau information for this early period is rather sketchy, but Shirley Fitzgerald has documented the class composition of Inner Sydney. Her very valuable early book, Rising Damp is a mine of information on the class composition of Sydney, painstakingly gathered from a variety of sources.

In 1901, the Sydney CBD had an enormous residential population, and a very, very crowded one, mostly industrial workers in the light industries around the city, and many of them waterfront workers. There were ten electoral seats concentrated around and in the CBD, and they mostly had overwhelming Labor majorities.

Women’s suffrage

The electoral roll for the 1904 election more than doubled to 689,490, with the introduction of women’s suffrage. The Labor vote went up to 23 per cent, so there is no indication that women voted more conservatively then men.

The sectarian mobilisation of conservative Protestants continued, which pushed Catholics even more towards the Labor Party. The sectarian issue became even sharper in the 1907 election, with the Protestant wowsers mobilising against Labor under the rubric of opposing “Rum, Romanism, Socialism and Gambling”, but the Labor vote made another leap to 33 per cent.

Finally, in 1910, after a decade in which the new institution of industrial arbitration, and other factors, had contributed to a mushrooming of trade unions, to include a very large part of the working population, the Labor vote leapt to 49 per cent of an electorate of 867,695, despite non-compulsory voting, and the continuing Protestant mobilisation against Labor.

Labor still did not make much headway in the outer suburban areas (the “aspirational voters”), but it won the Illawarra and Newcastle mining seats, and it made gains in the country. In 1910 Labor was able to form its first NSW government, with a majority of one.

Trade union predominance in the ALP takes clear shape. The conscription split

From about 1910 onwards, the trade unions asserted themselves more and more in the ALP. From about 1915 they began to take effective control of the party. The 1915 ALP Conference in NSW was the arena for an intense conflict between the state Labor government of William Holman and the trade unions.

The bitter conflict over conscription brought all the other conflicts to a head, and the unions organised jointly to increase their power in the ALP and to discipline the Premier Holman and the Prime Minister Hughes, with the conscription issue at the cutting edge of this conflict. The initiative was taken by the Australian Workers Union which had emerged as the largest union.

The first big setback to Labor’s electoral rise was the conscription split of 1916, which cost Labor a NSW government as well as the federal government. In the subsequent 1917 NSW election, the split state election, Labor was defeated, being reduced to 33 seats.

Nevertheless, despite the defection to the conservatives of state Premier Holman, and Prime Minister Hughes, Labor retained 43 per cent of the vote, even in te the jingoistic atmosphere of World War I. As was to become usual in swings against Labor, it lost most seats in the bush and the suburbs, but still held the Illawarra and Newcastle mining seats, and its inner city industrial seats, which still had a larger than average proportion of Catholics.

1920s Labor recovery

In 1920, the conservatives decided to introduce a system of proportional representation in the lower house, with electorates of five members in the city and three in the bush. Much to the surprise of the conservatives, this new system benefited Labor, because it evened out the effect of the large concentration of Labor votes in industrial areas.

In many industrial areas, Labor got three or four seats, and even in the outer suburbs and Sydney's North Shore, it got at least one. The Labor vote only went up about 1 per cent from 1917, but it achieved government under John Storey, with a majority of one, with the support of a Socialist Labor independent from Broken Hill, Percy Brookfield.

Brookfield set as the price of his support for the Storey government that the government had to find a formula for the release of the twelve IWW prisoners framed in 1917 by the conservative government, on the allegation that they were trying to burn down Sydney. Storey found a formula, and the necessary judge for a royal commission, and the IWW men were released to scenes of much labor movement emotion. See Sydney’s Burning by Ian Turner.

The configuration of NSW politics remained similar throughout the 1920s. Proportional representation was ditched by the conservative government elected in 1922. Labor got 38.5 per cent of the vote in 1922, 46 per cent in 1925 (producing the election of the first Lang government), and 45 per cent in 1927, when the Lang government was defeated.

Throughout the 1920s there was a further conservative Protestant mobilisation against Labor, with the added curiosity that Jack Lang, the state Labor parliamentary leader, was painted as a tool of both “Romanism” and the “Trades Hall Reds”, who had, after the success of the Russian Revolution, become the other bogey of conservative politics.

Compulsory voting

In 1929, a rationally self-interested combination between Labor and the Country Party led to the adoption of compulsory voting. By now the electorate had increased to 1,440,785. The electoral arrangements had become much less restrictive, and a common roll existed, without the necessity of getting a voter’s right even after enrolling.

The percentage voting shot up from 82.54 per cent to 94.94 per cent. From the introduction of compulsory voting, the standard voting return in NSW came to be between 95 per cent and 97 per cent, the highest in the world, in marked contrast with the United States, where about half those eligible are on the role, and of these only 70 per cent vote. In non-compulsory Britain the voter turnout is usually far lower than in Australia, and has dropped recently to the between 50 and 60 per cent.

The Labor vote in NSW shot up to 55 per cent in October 1930, and the second Lang government was elected in the teeth of the blizzard of the developing world depression. Labor won most of the outer suburban seats and many country seats.

Depression turbulence

Lang tried to defend and protect the people of NSW from the worst impact of the depression, but he did so in a limited electoralist way, being not inclined to lead the social revolution required in the circumstances of the depression. As a result he was chopped to pieces, politically speaking, in the turbulent political crosscurrents of the period.

He remained an enormously popular figure personally, however. In 1932 the Tory Governor, Phillip Game summarily dismissed Lang’s government, and in an orgy of conservative mobilisation Labor was defeated in the subsequent election. The pattern was similar to the 1917 conscription split election.

Labor lost all its seats in the bush and all its seats in the suburbs, but held on in the mining seats around Newcastle, the Illawarra, and the inner suburban industrial seats. Nevertheless, the Labor vote remained high. The Lang party got 40.2 per cent of the vote. The breakaway Federal Labor Party got 4.25 per cent, and the Communist Party got 0.92 per cent. The combined labor movement vote was 45.37 per cent.

Once again, the decisive swing against Labor was among “aspirational voters” in the suburbs. The 1935 and 1938 elections followed a similar pattern. The combined state Labor, federal Labor and Communist vote went up to 49.93 per cent in 1935, but slipped back to 39.28 per cent in 1938.

The 1941 McKell state election. The beginning of NSW Labor electoral hegemony, which continued until 1965

The 1941 election was one of the most interesting in NSW history. With the onset of World War II, the conservative forces in Australian politics were in disarray, and a popular mood developed for Labor governments.

At the moment of the state election, the Lang group, by now a minority in the ALP, was part of the Official ALP, but the majority of the state ALP executive, which had come under the organisational influence of the Communist Party, split away to form a new party in that NSW election, the State Labor Party.

In the federal election, which happened very close to the state election, there was actually a three-way split, in which Official Labor got about 40 per cent, Lang Labor got about 10 per cent and the Communist-led State Labor Party got about 4 per cent. Despite this three-way split, the Curtin federal Labor government was comfortably elected on the preferences of the other two Labor groups.

In the state election, one featur, was the careful selection, by McKell, the new state ALP parliamentary leader, of country Labor candidates, who were prominent figures in their local area, “horses for courses”. There was a massive swing to Labor.

In an enrollment of 1,684,781, Labor got 50.79 per cent, State Labor got 5.64 per cent, Independent Labor got 2.27 per cent, and in seats, after preferences from State Labor and Independent Labor, there was an overwhelming Labor parliamentary majority.

Labor won all the significant industrial seats throughout the state and a majority of country and suburban seats. A similar pattern was repeated in 1944 and 1947, although it was not quite so pronounced as in 1941.

In 1950 there was a swing against Labor, but it narrowly held on to government. The combined Labor, Independent Labor, and Communist vote was 50.5 per cent. In 1953, however, there was a dramatic swing back to Labor, with the Labor vote going up to 55.02 per cent and a Communist vote of 1.38 per cent. In The People’s Choice the following table is presented of the 1950 and 1953 elections, which is revealing.

Seats won by Labor
Region 1960 1953
Inner Sydney 9/9 7/7
Sydney suburbs 17*/39 24/41
Western country 9*/19 9/21
Central country 0/12 0/12
Mining-industrial 13/15 13/15
*Includes one Independent Labor

In the elections of 1956, 1959 and 1962, Labor hung on to office with between 49 per cent and 51 per cent of the preferred vote. These elections were affected by two new factors, the great Labor split, and the demographic impact of mass migration, which combined with the tendency of the Anglo and Catholic working class to move to the suburbs, including the outer suburbs.

This period also included the beginnings of the phenomenon of gentrification in inner city areas, which in Sydney commenced in Paddington. Shirley Fitzgerald’s Sydney book, Peter Spearritt's Sydney book, and the spin-off books on Sydney’s suburbs, all describe these processes well.

The Labor split in NSW was comparatively contained, by contrast with the massive split in Queensland and Victoria. The breakaway, Catholic-dominated Democratic Labor Party, got a smallish vote, between 2 per cent and 4 per cent. It appealed mainly to middle-class, upwardly socially mobile Catholics in suburban areas.

The DLP was dramatically different than previous ALP breakaways, such as Lang Labor and State Labor, most of whose preferences went back to Labor. The overwhelming majority of DLP preferences went to the Liberals, which was a small contributing factor to the closeness of some of the elections in this period, but not decisively so because the DLP vote was comparatively small compared with other states.

Only in the 1965 election did the DLP vote made the critical difference, bringing the ALP electoral hegemony in NSW to an end as the Coalition won very narrowly.

Changing demographics. The impact of mass migration on electoral politics, and the beginnings of inner urban gentrification

Post-war mass migration to Australia began in a substantial way in 1947. It included initially a large British component, but the new feature was the substantial Non English Speaking Background component. Post-war mass migration began to have a major impact on NSW electoral politics from about 1950, and the impact of migration on electoral politics has grown dramatically since the 1950s.

The first wave of NESB mass migration was from Greece, Italy, Malta, Holland, Germany, and other European countries. Greeks and Italians largely voted Labor Eastern Europeans largely voted Liberal, and most other nationalities split on similar lines to the existing electorate.

In inner city industrial suburbs, like Marrickville, Surry Hills, Leichhardt and Bankstown, the new migrant populations largely replaced the Anglos and Irish Catholics who were moving to the outer suburbs, but these inner suburban seats remained safe for Labor because most of the migrants who settled in them were as loyal to the Labor side of politics as the people they replaced.

In the Australian context, new migrants rapidly became unionised, and their social location at the bottom of the economic tree predisposed them to the Labor side of electoral politics in a similar way to the Irish Catholics of previous generations.

The 1950s, 1960s, and 1970s were the high point of Australian trade unionism, and trade union density peaked in the late 1970s at more than 50 per cent of the workforce. For this whole period recent migrants were a large proportion of the unionised working class, although they had less representation and impact in union leadership above the level of shop steward. Migrants were an increasing proportion of shop stewards.

In the 1950s, 1960s, and 1970s the pattern developed that inner suburban areas of dense recent NESB settlement were also areas with a higher than average proportion of trade unionists, and were overwhelmingly safe Labor seats.

The beginnings of inner urban gentrification

The first gentrification in Australia began in the inner Sydney suburb of Paddington, at that time largely inhabited by industrial workers, who lived in pleasant old terrace houses originally built for the middle class in the 1880s.

The gentrification in Paddington took its toll on the Labor side of politics, and the colourful waterfront tally clerk Tom Morey eventually lost his seat of Bligh to the Liberals as the waterfront and other workers in Paddington sold their pleasant terraces to rich professionals, the beginning of a trend that was to eventually have considerable demographic and electoral impact in other inner city areas. Tom Morey (a mate of mine) actually hung for on an election or two against this trend through the force of his personality and his colourful campaigning.

This early gentrification took place in a suburb very convenient to the city, where the housing stock largely consisted of down-at-heel, but basically very functional and potentially elegant and terrace houses. (One of the oddball ironies of Sydney history is that the wonderful old terrace house in Heely Street, Paddington, which was for his whole period in parliament both the home and electoral office of Eddie Ward, the redoubtable leader of the left in the Labor Party in the federal parliament, is now the home and office of the born-again right-wing ideologue Keith Windschuttle.)

At this early stage of gentrification the professionals who bought the houses were overwhelmingly traditional Liberal voters, and the wide extension of university education into sections of the working class had not seriously begun. The development of a statistically important section of tertiary educated people who supported Labor or the Greens electorally, came later and began with the radicalisation arising from opposition to the Vietnam War in the middle 1960s. In the earlier period the impact of gentrification was to swing seats away from Labor to the Coalition.

Liberals win the 1965 election

The 1965 elections were the end of an era. In the atmosphere of jingo hysteria at the beginnings of the Vietnam War, the Liberals narrowly won the 1965 election. Nevertheless, the Labor vote remained reasonably high, at 43.3 per cent, the Communist vote was 0.64 per cent, the DLP vote was 2.11 per cent and Labor still held the inner suburban industrial seats, the Newcastle and Illawarra mining seats, and a number of suburban and country seats.

The number on the roll was now 2,256,568. The result in the 1968 election was almost exactly the same as 1965, percentage-wise. The 1971 election was a cliff-hanger, with the Labor vote rising to 47 per cent, but the Liberals scraped back into government.

In 1973 Labor slipped back to 43 per cent, but a new phenomenon was the emergence of the Australia Party, the predecessor to the Democrats, which got 4.21 per cent, and a large proportion of its preferences went to Labor. The basic NSW and Australian electoral pattern reasserted itself in this period.

The impact of mass migration changed the form slightly, but not the fundamentals of this pattern. All the areas of high concentrations of industrial workers and unionists, in Sydney, Newcastle and Illawarra tended to be safe Labor seats, with the new element that these seats in Sydney and the Illawarra were coincidentally seats with a high concentration of NESB migrants. Labor also held a number of suburban and country seats. The swing seats tended to be a couple of areas of inner city gentrification and outer suburban seats — “aspirational voters” again.

It’s worth noting that during this whole period the Labor vote remained quite high, even with Labor in opposition, and it’s not difficult to detect that a pattern was emerging in which two factors combined to shore up the Labor vote. One was the tendency of the last group of migrants just off the plane to vote Labor, and the fairly rapid growth of a substantial section of the, then exploding, tertiary educated section of the population to vote Labor, Democrat and, later, Green.

This period brought the demise of the DLP as the ostensible centre party of Australian politics, directing its preferences to the Liberals, and its replacement by the Democrats, and eventually the Greens as the ostensible centre parties, with the majority of their preferences going to Labor. The Greens, in particular, have skilfully achieved the situation where they get a large vote to the left of the ALP and some votes from people who seem to locate the Greens between Labor and Liberal.

Using a modernised Marxist sociology to illuminate electoral patterns

The People’s Choice is an extraordinarily useful book in every way that pertains to the history of electoral politics in NSW, but it obviously does not try to go too deeply into the sociology of this electoral history. To properly interpret and understand the broader implications of this narrative it is useful to have a sociological standpoint, and to bring to bear the Bureau of Census and Statistics information on the electoral pattern.

In another essay, polemicising against the “New Class” rhetoric of some right-wing pundits, I develop a rudimentary sociological model of modern Australian society.

The two most useful census documents for this inquiry are Australian Social Trends 1999 and the Social Atlas for each capital city, and I will use the 1996 Sydney Social Atlas as my working example. On page 83 of Social Trends 1999 you get the ABS classification of qualifications, which divides post-school qualifications into: “bachelor degree and above”, “undergraduate diploma”, “associate diploma”, “skilled vocational qualification”, “basic vocational qualification”.

For purposes of describing people who have a university degree or equivalent, it seems sensible to group the first two together as representing a university degree. In the first census that counted degrees, 1966, 1.5 per cent of the population older than 15 had degrees. In 1976, 3 per cent had degrees. By 1996, Katharine Betts gives the figure of 10.2 per cent, but she seems to be wrong, as the bureau gives 12.8 per cent. In addition, the bureau gives a figure of 8.8 per cent for people with undergraduate diplomas and associate diplomas together.

For simplicity’s sake, we may assume half the 8.8 per cent for each category, which means that in 1996, according to the bureau, about 17.2 per cent of the adult population had a university degree. By 1998, according to the bureau, the figure had become 14.5 per cent plus 7.9 per cent, which takes the number with a university degree to 18.4 per cent of the adult population, a very high figure indeed.

Another framework useful concerning the educational qualifications of the population are the figures for the raw number of tertiary students. In 1912, when there were 4.5 million Australians, there were a tiny 3672 tertiary students. In 1938, when the population was about 6.5 million, students were still a tiny 12,126. In 1966, when the population was 11.7 million, the number of students had risen to 91,272. Thirty years later, when the population had increased about 50 per cent to about 18 million, the number of tertiary students had soared sevenfold to 634,094.

Women predominate among graduates in the health, education, and society and culture fields

In the Census Bureau’s documentation there is a very detailed breakdown of “people with post-school qualifications, by type of qualification” by both age and sex. They reveal a very sharp increase in the number of women with university qualifications, who now number about the same as men, and are concentrated in areas such as teaching, the health industry, social work and also, to some degree, in commerce and business. The number of female primary teachers rose between 1988 and 1998 from 71.7 per cent to 77.5 per cent. The number of female secondary teachers went up from 48.3 per cent to 53.5 per cent, and the number of women teaching in higher education increased from 27.3 per cent to 35.1 per cent.

In 1996 227,000 people had bachelor degrees or higher in business and administration and 35.7 per cent of them were women. In health, 213,600 had university degrees, 66.2 per cent of them women. in the delightful ABS classification called Society and Culture, defined as “economics, law, behaviour, welfare, languages, religion and philosophy, librarianship, visual and performing arts, geography, communication, recreation and leisure, and policing” 357,800 had university degrees and 54.8 per cent of them were women. In engineering, however, with 120,100, only 8.4 per cent were women.

The great numerical explosion of people with university degrees was a product of the Whitlam period educational reforms.

The extremely useful Australian Social Trends 1999 has a detailed breakdown of the age composition of people with university degrees. Part of this table is reproduced here.

Voters with degrees, 1996
  Age Associate/undergrad diploma Bachelor/higher
Males 15-24 4.2 2.4
  25-34 14.6 5.9
  35-44 16.2 6.9
  45-54 13.5 7.0
  55+ 6.6 4.4
Females 15-24 6.8 4.6
  25-34 16.8 8.5
  35-44 15.3 9.5
  45-54 10.7 8.6
  55+ 3.6 4.5
  Total 10.1 6.9

The extraordinary increase in both men and women with degrees in the 25-54 age group clearly illustrates the magnitude of the explosion of tertiary education from about 1974 onwards. This forcefully underlines that this was the period when women soared from being a very small portion of university graduates to rough numerical equality with men. It is fascinating to note the rage of conservative mysoginists like Michael Thompson against the Whitlam period of free education. Possibly the rough equality in educational achievement gained by women in this period is one of the features that infuriates them.

What emerges most strikingly from these statistics is the enormous growth in the proportion of the adult population with university degrees. The very size and diversity of this group makes an absurdity of the conservative rhetoric that they comprise, as a whole, an elite “new class”.

It is important to bring to bear other available statistical information to get a picture of Australia's class formation at the moment and how this vastly increased group of university graduates fits into it. This is where an investigation of the information contained in the Social Atlas comes in, particularly if you superimpose on this information the statistics of electoral behaviour in federal and state elections.

The Social Atlas shows that people with degrees are heavily concentrated in Sydney on the North Shore, most of the Eastern Suburbs, and in a belt in the Inner Western Suburbs. There are smaller concentrations in the Sutherland shire, the Georges River area, and the Blue Mountains.

The useful category provided in the 1991 Social Atlas, “Managers and Administrators”, coincides almost exactly with the map of “high income earners”. Both these maps, however, coincide only in part with the map of people with university qualifications.

Most of the people in the southern part of the Eastern Suburbs and in the Inner Western Suburbs, with university degrees, are thus neither “managers or administrators” or “high income earners” as defined by the ABS. It is likely that these graduates are by and large the ones working in teaching, health, social work, etc.

Coincidentally, the divide in political voting behaviour is on almost exactly the same geographical lines among graduates as the apparent geographical divide between “high income earners” and “managers and administrators” and the rest of the population.

The southern Eastern Suburbs and the Inner West vote overwhelmingly Labor or Green etc. The North Shore, Wentworth, the Georges River area, etc, all vote solidly Liberal. Any serious investigation of all these statistical tools shows that a real economic, political, and class division exists within the ranks of university graduates, not between graduates and the rest of the population.

Census information, combined with election results, gives a rough but informative insight into the current Australian class structure.

The Social Atlas provides a wealth of useful information. There are maps of the distribution of migrants of different backgrounds, and these maps are very informative. Most migrants of non-English-speaking background are concentrated in the Eastern Suburbs, the Inner Western suburbs, and the middle western suburbs.

The pattern of people with trade qualifications is the obverse of the pattern of people with university degrees. Many people with trade qualifications are concentrated in the southern part of the Eastern Suburbs and the farther Western suburbs, but quite a few are also concentrated in the Sutherland shire and areas like Hornsby and the Northern Beaches. An overview of all the statistical information gives a breakdown of the class structure of the Sydney population on broadly the following lines.

Even a cursory overview of the correlation between the information provided in the census publications and electoral results confirms the general thrust of the above breakup and analysis. This four-level description of Australian society is realistic and useful for a variety of purposes.

In my view, the dominant class division in Australian society is between the ruling class, with enormous economic and political power, which exercises very great ideological influence and hegemony over the “high income earner” and “managers and administrators” Statistical Group One, and the rest of the population.

This four-level division of Australian society holds for all the major capital cities and for the Illawarra, Newcastle, Whyalla, Launceston and Geelong, with the qualification that the smaller capitals and the provincial towns have a much lower NESB component in Statistical Group Four, the blue collar section of the working class. Rural and provincial Australia contains some elements of this division, but a concrete analysis of rural and provincial Australia has to incorporate a number of other factors, and I will deal with rural and provincial Australia elsewhere.

From 1976 to 2002

This was a period of significant changes in Australian society that have a bearing on NSW electoral history. This period, from the fall of the Whitlam government until now, brought the working out of a number of trends.

The ethnic composition of Australia changed dramatically. Australia as a whole, in 2002, has the following ethnic profile. Our 19.5 million people break down roughly as follows:

The total of the above is about 11 million. There are also about 8.5 million people of British, Scottish or Welsh cultural identification, and they are now a minority. The Middle Eastern and Asian components are the product of the last twenty-five years of mass migration. Intergroup ethnic intermarriage is a major feature of modern Australian society. Some groups such as Australians of Greek, Italian and Middle Eastern origin tend to marry within their own communities, even into the second generation, but intermarriage among all the other groups is increasing, particularly intermarriage between Aboriginal people and Europeans, and intermarriage between Asian people and Europeans.

The rapidly changing ethnic mix has had quite a considerable bearing on electoral politics. Anecdotal evidence suggests that many North Asian professional people swung to the Liberals in the 1996 federal election, but swung dramatically against the Liberals in the 1998 federal election because of the Liberals’ adaptation to the racism of Pauline Hanson.

Most recent NESB migrants, however, repeat the historic pattern of the underprivileged, and vote solidly Labor, and this has been reflected over the whole period from 1976 until now. The middle inner west suburban electorates, with a high concentration of NESB migrants, and trade unionists, are still safe Labor seats.

Areas of older second generation NESB migration, like the St George area, and Earlwood, are still reasonably safe Labor seats, although their income profile is somewhat higher than the mid inner west seats, and it seems that the well established second generation Greeks, Macedonians and Italians generally vote Labor, although some are swinging voters. Even a cursory overview of electoral trends, cross-checked with the ABS statistics on migration and ethnicity, shows a strong identity between migrant ethnicity and Labor voting throughout this period. Another feature of electoral politics is an overwhelming Labor vote among Aboriginals and other people of colour from the South Pacific.

NESB migrant ethnicity and the republic referendum

A similar identity between NESB migrant ethnicity and republicanism shows up in analysis of the result of the republic referendum (about which I have written a short analysis). In the recent election, dominated by hysteria over the Tampa crisis and terrorism, the biggest swings against Labor were in areas of predominantly Anglo ethnicity.

The demographic pattern of NSW is now quite complex. Sydney and the urban Illawarra have a disproportionately high concentration of NESB-background Australians, whereas the rest of NSW has a far smaller NESB component. These days, Sydney and the Illawarra between them, have about 4.7 million of the total 6.3 million population of NSW, a far greater proportion than in 1900.

If you take the NSW population as a whole, non-Anglo-identified people are more than half the population, but they are nudging 60 per cent in the Sydney region and the Illawarra, and are only about 20 per cent elsewhere. Rural Australia now has about a 10 per cent migrant component, but the other 90 per cent in most non-coastal rural areas, are still divided between Catholics and Protestants, roughly the way they were in the 19th century, with some inland areas like the South West Slopes, Bathurst and Bourke, having higher proportions of Catholics, and other areas like New England having higher proportions of Protestants. In times of swings to Labor in the bush, often the areas with fair-sized Catholic concentrations are more likely to swing to Labor. Ethnicity and religion, in the modern period, have a complex and subtle impact on electoral politics.

Another feature of the recent electoral period is the changed composition of the coastal country seats in NSW. For about the first 60 years of the century, those areas, then mainly made up of small farmers, few of whom employed much labour, with a below-average percentage of Catholics, were rock-hard electoral strongholds of the conservatives. In the past 30 years, the rapid growth of cities like Lismore, Kempsey and Grafton, with people who have moved from Sydney and Brisbane, and the settlement of city people on the land in rural areas, combined with the growth of university colleges in some towns, has dramatically changed the electoral pattern. The far North Coast seats, from Grafton to the Queensland border, now alternates between the conservatives and Labor, and the margin between the conservatives and Labor is narrowing in other coastal rural seats.

The democratisation of tertiary education and inner city gentrification as electoral factors

Almost all the inner urban areas in Australia have now been gentrified in the sense that the housing stock has been substantially renovated and housing values in these areas have dramatically increased with the acquisition of the houses by people with higher incomes.

In addition, the constant tendency of the last 50 years for the number of people in each dwelling to decline, has been offset in inner city areas by a dramatic urban intensification, with the building of additional home units and apartments. If you take a line between Cronulla and, say, Homebush, across Sydney — to the East of this line, almost any vacant or underused site you can swing a cat on now has units, and this trend is continuing.

Another factor that has some impact on electoral politics is Sydney housing prices. Sydney houses are now more expensive than any other city in Australia. This is of great significance in a social situation in which between 60 per cent and 70 per cent of residences are owned by the occupants, or are being paid off.

The gentrification pattern in the old industrial area of the Illawarra, south of Sydney, is of considerable interest. The northern suburbs of Wollongong are the old mining villages, which were the site of many class battles in the 19th and early 20th centuries.

The Port Kembla steelworks, built in the late 1920s, south of Wollongong, had their heyday from the 1950s to the 1970s, and the whole region grew from about 100,000 people to about 300,000 today during the first wave of mass migration from non-English-speaking countries, which included Italy, Greece, Macedonia and Latin America.

In the 1990s the workforce of the steelworks was cut to a quarter of its peak, and employment in the area shifted into service industries, education, building, tourism, etc, but there was also quite high unemployment. This can be seen visually from Bulli Lookout, the traditional place where you overlook the South Coast, driving from Sydney — the old pall of smoke that used to dominate the dramatic landscape is vastly reduced. (This environmental effect is even more spectacular in the other old industrial area, north of Sydney, Newcastle, where the steelworks closed completely. On a recent visit to Newcastle to a book fair, my friend and I could not work out what was different, until driving out of the city. On a clear winter’s day, the air was completely clear, an extraordinary contrast with past visits to that grimy old industrial city.)

From Helensburg to central Wollongong, the visually exciting mining villages, and almost all the land below the escarpment, have been built on and gentrified. Much of the population in this area now works in Sydney, commuting every day, mostly by fast train. The professional part of the population in the Illawarra now lives largely in the northern suburbs. The Census Bureau statistics show that incomes in the northern suburbs of the Illawarra are higher than in the southern suburbs.

These suburbs have remained fairly safe Labor seats, but with an enormous new-social-layer Green vote, as well, a similar demographic to the inner-western suburbs of Sydney. The densely populated industrial suburbs south of Wollongong, where the NESB migrants of the 1950s-70s and their children live, remain overwhelmingly safe Labor seats, with a much smaller Green vote than the northern suburbs.

The gentrification of the northern suburbs of Wollongong, and the consequent rise in the Green vote, is the underlying demographic reason for the drop in the Labor vote recently for the Mayor of Wollongong, and in the seat of Cunningham, which takes in the northern suburbs.

In rural areas, the relationship between ostensible income levels and electoral behaviour is not as straightforward as in most urban areas

Paradoxically, there is no longer a simple straight line between ostensible incomes, education and voting behaviour. It seems clear from voting patterns in the recent period that the poorer people in the areas of high recent NESB, and high trade union concentration, continue to vote Labor. In rural areas, it’s not quite so straightforward.

As Phil Raskall’s useful research about postcodes underlines, there are postcode areas in rural NSW with, according to taxation records, the lowest incomes, which vote for the National Party, and show high One Nation votes. Partly this is explained by the fact that farming families, which have significant assets in land and equipment, often legitimately end up with low incomes for taxation purposes, and those postcode areas contain a substantial number of rural unemployed and poorer retired people. Ostensible income statistics at the bottom end are a fairly good guide to Labor voting electoral behaviour in urban areas, but in the bush they are sometimes associated with non-Labor voting patterns.

Despite similar incomes, and comparable educational profiles, people on the Lower North Shore, and in the Inner West of Sydney tend to vote differently

At the other end of the income and occupational spectrum — the end of people with higher incomes and more tertiary education — voting behaviour has become more complex. It’s paradoxical that the Inner West, and the inner Eastern Suburbs, have quite a similar income profile to the Lower North Shore, and quite similar occupational profiles, with both regions having some high-income, dense-tertiary-education hot-spots, and a generally higher income and educational profile than the more strongly Labor seats where recent NESB voters predominate.

Yet in the Inner West and the Inner Eastern Suburbs, there is a pronounced general electoral inclination towards the Labor Party. On the other hand, the Lower North Shore exhibits a pronounced electoral bias towards the Liberals. Both areas throw up fairly substantial Green and Democrat votes.

What seems to be at work here is a certain residential self-selection, among what, for convenience, one can call the new social layers. There is a whole segment of modern society, often tertiary educated people, who, in the strictest sense of Marxist sociology, are not capitalists, because they sell their labour power, and don’t own much productive capital, but in terms of ideology, often don’t identify with the working class as an organised social force.

These people frequently seem to self select in terms of where they live. Often the ones who work in the finance industry, and up and coming, rather underpaid, but ambitious juniors in law offices, accounting firms, etc, etc, who are Liberal voters, find the Lower North Shore more congenial, whereas people who work in education, health and the public service, and are often a little “hipper”, and who tend to vote Labor or Green, find the inner West and the inner East more congenial.

Those differences, which are rather subtle, show up in divergent electoral behaviour some of the time. These days the “aspirational voters” who, as I’ve established, are an old phenomenon, not a new one, are, in fact, divided amongst themselves, in a number of directions. On questions like racism, migration and the republic, the Liberal-voting ones, and the Labor and Green voting ones, have similar, more civilised attitudes, which infuriates atavistic conservatives in Australian society, but on economic and social issues they have quite divergent views, which very frequently leads to opposed electoral behaviour.

A considerable amount of confusion has arisen in recent discussion of electoral trends and electoral politics, and the opposite voting behaviour of the Inner West and the Lower North Shore lies at the heart of some of this confusion. These days much electoral comment relies heavily on polling and focus groups made up of people selected in swing seats, whose flucturating moods and attitudes are charted over longer or shorter periods.

The problem with both polls and focus groups, which are possibly a reasonable guide for changes in mood in swing seats, is that they don’t usually incorporate a consideration of the underlying demographic social, ethnic, religious and income factors which also bear on electoral behaviour. This tendency, of polls, and superficial observation of electoral moods (considered without addressing underlying social and demographic factors) to produce an inaccurate picture, has reached a peak several times lately.

After several recent Labor electoral defeats, so-called exit polls were used by the mass media to suggest a dramatic swing in which a majority of blue collar workers had allegedly voted for the Liberals. (This confused and inaccurate information was seized upon by some hopeful sectarians on the far left, who took it up uncritically as support for their totally metaphysical perspective of themselves presenting an effective electoral alternative to Labor and the Greens.)

The problem with the superficiality of this kind of analysis of electoral politics is obvious. In the elections that Labor lost, it still comfortably held the core seats of inner suburban Sydney and the Illawarra, where blue collar and NESB migrants predominate, and it still held the Newcastle seats. If the blue collar workers had swung in their majority against Labor, where did the Labor votes come from? This polling-focus group method of interpreting electoral behaviour often gets it wrong,because of a neglect of underlying empirical sociology.

The history of NSW elections in The People’s Choice, studied carefully, shows that major swings between the Labor and conservative side of politics tend to go up and down the economic and social scale. In major swings against Labor, the outer suburban seats fall to the conservatives, and in major swings to Labor, like for instance, the “Wranslides” of 1978 and 1981, they fall to Labor.

In 1978 and 1981, even some of the lower North Shore seats, Manly and Wakehurst, fell to Labor, and North Sydney was captured by a slightly left of centre Independent. The subtle self-selection and differentiation among the new social layers, between the Inner West and the Lower North Shore, that I note in this article, are by no means absolute, and they don’t fundamentally alter the hard spine of a class division in electoral politics, which the electoral history in People’s Choice makes quite apparent, right up to the present.

Religious self-identification, age, and sexual orientation, as factors in electoral behaviour

These three factors intertwine with the occupational and educational factors just discussed. In recent years polling, after all elections has constantly shown a pattern in which the oldest cohort of the population, those older than 65, seem to have by far the strongest tendency to vote conservative.

This is not surprising, when you consider that this cohort is ethnically less diverse than younger cohorts. This over-65 cohort has a far lower component of tertiary educated people, for the obvious reason that the explosion of tertiary education, only commenced in the mid-1960s. It is also a well-known demographic fact that people who have been industrial workers for most of their working lives have a shorter lifespan than the middle and upper classes, which tends to skew the composition of the cohort over 65 in the direction of people who have been conservative voters all their lives.

It is also clear that many people in this cohort, having commenced their lives, and had their life experience moulded, in the Menzies years, have a certain nostalgia for the conservative values of that era. This strong bias of the older cohort towards the conservative parties, is, in the medium and long term, bad news for the conservatives, as this cohort will be reduced over time by natural causes. Nevertheless, longer life spans mediate the disappearance of this older cohort (of which the writer has just become part), and this reduction will take place gradually, so their electoral impact, though decreasing, will continue.

It is important to note also, that the impact of age on electoral patterns is different in Sydney and the rest of NSW. There is a pronounced trend for young people from country areas to move to the city for job opportunities, and there is also a tendency for older Australians to retire to non-metropolitan coastal areas, and both these trends show up in electoral patterns.

Sexual orientation and electoral behaviour

The impact of sexual orientation on electoral behaviour is harder to pinpoint for the obvious reason that there are almost no statistics, and that polling in this area is a bit difficult to say the least, so little polling has been done. Nevertheless, it is quite clear that there are areas of inner suburban Sydney with a high concentration of gay men and lesbians, and a lower concentration in suburban and country areas for the obvious reason that residual prejudice in country and suburban areas tends to influence gay people to migrate to areas where acceptance of their sexual orientation is greater.

Areas with a high concentration of gay men and lesbians seem to coincide with those areas of tertiary educated, relatively well paid people who vote Labor or Green. However, some evidence suggests that there is a distinct group of gay males who vote conservatively. Other anecdotal evidence suggests that gay men and women are a significant part of the vote for Clover Moore, the independent for Bligh.

Changes in religious identification as an electoral factor

In 1901, census figures showed 70 per cent self-identification with at least nominal Protestant religion, 25 per cent Catholics and 5 per cent Jews, other non-Christians, non-stating or atheists. By the year 2001, census figures showed a most dramatic collapse in Protestant nominal religious identification, from 70 per cent to 30 per cent.

The Catholics have gone up to almost 30 per cent. Orthodox, Lutherans and other non Anglo Protestant Christians are about 5 per cent. Non-Christian religions are about 10 per cent, and no religious belief, or religious belief not stated, has gone up to a rather large 25 per cent of the population.

Like age, ethnicity and educational level, with which, to some extent, it is associated, religious identification diverges dramatically between Sydney and the rest of NSW, and even between the inner city and the outer suburbs. Protestant religious identification is far higher in the outer suburbs of Sydney and rural NSW. Catholics, Orthodox and non-Christians are proportionately much higher in the middle-western seats of Sydney, with a pronounced Labor voting tendency, and in the Illawarra. Non-believers are heavily concentrated in the Inner West and inner East, and to some extent the lower North Shore of Sydney. Religious unbelief seems to be strongly associated statistically with tertiary education.

Associating all these factors, a pattern begins to emerge. It seems highly likely that in the modern era there is a natural Labor-Green electoral majority among Catholics, Orthodox, non-Christians and non-believers, and this underlying demographic reality is an enormous difficulty for the conservative parties. This is part of the underlying background to how Labor bounced back so quickly after the defeat of the unpopular Keating government in 1996, and why the Labor vote held up relatively strongly in the recent federal election, despite the chauvinist hysteria of the conservatives, about the September 11 terrorist attack, and about refugees. This is also the underlying demographic reality behind the universal hegemony of Laborites and Greens in all the eight state and territory governments.

The electoral pattern from 1976 until now

In the early 1970s, the right-wing state Labor machine machine, led by John Ducker, deliberately organised the election to state Labor parliamentary leadership, of a secular, small-l liberal figure, barrister, Neville Wran, on the principle that he was likely to be a charismatic figure.

This tactic was spectacularly successful, and just a few months after the devastating defeat of Gough Whitlam in the December 1975 election, the Wran Labor government was elected, by one seat, in the May 1976 NSW election, to the surprise of most political pundits. The old NSW electoral pattern reasserted itself, the Labor vote went back up to 49.8 per cent, and Labor won the Inner West industrial seats, the Newcastle and Illawarra seats, and some suburban and country seats.

This was an unusual election in that there were almost no electoral alternatives to Labor and the conservatives, and non-Labor and conservative candidates got a tiny 4 per cent of the vote. The Communist Party and the DLP both disappeared electorally.

The 1978 and 1981 “Wranslides”

Both these elections registered an enormous swing to Labor. In 1978 the Labor vote was 57.7 per cent, the Communist vote was 0.3 per cent, the Socialist Workers vote was 0.2 per cent, and the Australian Democrats vote was 2.7 per cent, giving Labor a 60 per cent preferred vote, and 63 seats out of 99. On this occasion

Labor won most of the outer suburban Sydney seats. In 1981 there was a similar result, Labor won 55.73 per cent, and the Democrats 2.43 per cent. Despite a slight drop in the Labor vote, Labor’s number of seats increased to 69, out of the 99 available. Again Labor won almost all of the outer suburban seats. The enrollment in 1981 was 3,212,657 and the percentage who voted was a very high 97.96 per cent.

There was a swing against Labor in 1984. The Labor vote dropped back to 48.75 per cent, the Democrats got 2.85 per cent and independents got 5.08 per cent, but in terms of seats, Labor still won comfortably, holding a number of the outer suburban and country seats, as well as its core inner suburban seats, and the Newcastle and Illawarra seats.

The 1988 election. Labor defeated by a big swing. The “Greiner effect” and the gun issue

In 1988, the state election results were disastrous for the ALP. Wran retired as premier shortly before the election for reasons that are still not entirely clear. He may have been fearful of the electoral consequences of an Independent Commission Against Corruption inquiry into certain matters, in which he was ultimately completely exonerated.

He retired at an ALP state conference, which gave rise to the rather spectacular vision of the various powerful figures in the dominant ALP faction debating who his successor would be under the Sydney Town Hall stage, the traditional place for factional negotiation at ALP state conferences.

The forceful, powerful, but electorally uncharismatic secretary of the Labor Council of NSW, Barrie Unsworth, asserted his claim to ALP parliamentary leadership, and was successful. But from that point on, the 1988 elections were all downhill for the ALP. For Unsworth to become leader, Brian Bannon had to stand down for him in the previously safe Labor seat of Rockdale, but in the necessary by-election for Rockdale, Unsworth only won the seat in an agonising cliff-hanger by a handful of votes.

The Wran and Unsworth governments had antagonised the core industrial base of the ALP by savagely cutting back workers compensation rights, leading to the expulsion of popular long time South Coast leftist George Petersen from the ALP for opposing the legislation in the parliament.

The Hawke federal Labor government was going through an unpopular phase, and to cap it all, Unsworth decided to introduce laws greatly restricting access to long -arrelled guns, which offended rural people and sporting shooters. The gun lobby, which was powerful in outer suburban seats, country seats, and even the traditional non-urban industrial seats in Newcastle and the Illawarra, campaigned strongly against Labor.

The swing against Labor was a massive 10 per cent and the Labor vote dropped to 38.49 per cent. The Democrats also slumped to 1.81 per cent. Labor was almost wiped out in the bush and was wiped out in the outer suburban areas and all the marginal suburban seats. It suffered severe reverses in the Newcastle industrial area, where a number of independents won hitherto safe Labor seats, clearly because of the gun and workers compensation issues. Nevertheless, Labor retained its core Sydney and Illawarra seats — 43 out of 109, and there were seven independents.

In 1991, there was a bit of a swing back to Labor. With 10 fewer seats in the state, Labor went up from 43 to 46. Labor won back the Newcastle seats that it lost in 1998. The Labor percentage went back up to 39.05 per cent, and the Democrats went up dramatically, to 5.36 per cent. The Greens began to emerge with 0.54 per cent, and a new right wing force, the Call to Australia Party, began to occupy the ecological niche previously occupied by the DLP, getting 1.19 per cent.

Throughout the early 1990s the Liberals were led by Nick Greiner, who because of his Hungarian migrant background and his mixed Catholic and Jewish origins, gave the Liberals a certain appeal for a short period, to people of NESB background, but this appeal disappeared rapidly after Greiner’s demise, and Howard’s leadership of the Federal Liberal Party reasserted the Anglo backwardness endemic in the conservative parties. The new Labor leader, after Unsworth, Bob Carr, wasn’t particularly charismatic, but surprisingly to some, the electorate warmed to him quite rapidly.

The 1995 election

In this election Labor bounced back to scrape in with 50 seats out of 99, and there were three independents. Labor got 41.26 per cent of the vote, the Australian Democrats got 2.85 per cent, the Greens got 2.57 per cent, the No Aircraft Noise Party got 0.95 per cent.

Of the two right-wing niche parties, Call to Australia got 1.44 per cent and Australians Against Further Immigration got 1.11 per cent. Independents got 4.69 per cent and won three seats. Labor won back a few more outer suburban and country seats to give it the majority it needed to govern.

The 1999 election

By the 1999 election, the number on the roll had grown to 3,736,079. Labor won a dramatic victory, with 55 seats against the Coalition’s 33, and five independents. The Labor vote only went up marginally, to 42.2 per cent.

The most notable feature of this election was that it took place at about the high-water mark of Pauline Hanson’s conservative populist One Nation Party, which got 7.5 per cent of the vote. Another right-wing group, the Christian Democrats, got 1.5 per cent. (Pollsters tended to the view that perhaps two thirds of this 9 per cent had previously voted Liberal and one third had previously voted Labor. The One Nation vote was overwhelmingly concentrated in the outer suburbs and non-Sydney areas.)

Optional preferential voting, adopted in 1979, helped produce an effect that the breakaway vote from the Liberals damaged the Coalition two-party-preferred vote. On the other side of the ledger, the Greens got 3.9 per cent, the Democrats got 3.3 per cent and Unity got 1.1 per cent, as well as independents getting 5.1 per cent.

A rather higher propostion of the preferences of these groups were counted, even despite optional preferential voting, and those preferences favoured Labor. (This larger flow-on of preferences from the Greens, Democrats and Unity strongly suggests a social circumstance, which can clearly be inferred in other ways, that the general educational level of Green, Democrat and Unity voters is somewhat higher than that of voters for right-wing populist parties because the Green, Democrat and Unity voters obviously understand the voting system better.)

In the 1999 election, which can be regarded as an average kind of election, in the new demographic and electoral circumstances in Australia, Labor won the Newcastle seats, the Illawarra seats and the inner suburban seats of Sydney, and won also a large portion of the outer suburban and country seats. Most of the independents elected won country seats from the Coalition.

The NSW Upper House (Legislative Council) from 1981 to 1999. Its transformation from Tory-dominated, reactionary, appointed chamber, to ultra-democratic house of the people

Neville Wran’s great contribution to NSW politics was the transformation of the NSW Legislative Council, the Upper House. An attempt to abolish that bizarre appointed institution by referendum in the 1960s was defeated because of a fierce campaign by the Tory media, and it remained a stronghold of reaction.

Much to the amazement of all political observers, Neville Wran, in a political role for which he will always be remembered with affection by democrats and progressives, used his political clout and charisma at the height of his authority due to the “Wranslide” of 1978 to democratise this curious institution.

Wran managed to persuade 82.6 per cent of the voters to vote Yes in a referendum to the question: “Do you approve of the Bill entitled a Bill for an Act to Provide for the Election of Members of the Legislative Council directly by the people.” This provided for the replacement of the appointed members over three elections by elected members, elected by a system of proportional representation, in blocks of 15, the elections to be held simultaneously with lower house elections.

An overview of Upper House elections from 1978 until 1999

In the first election for the Upper House in 1978, Labor got a massive vote of 54.91 per cent (to win nine seats), to the Liberals’ six. The Democrats got 2.78 per cent, and the Communist candidate, the charismatic media personality Jack Mundey, former leader of the Builders Laborers and the Green Bans, who was lucky enough to draw first position on the ballot paper, got 2.91 per cent and was in the hunt for the 15th position, although he just missed out. The Marijuana Party got 0.91 per cent, and the right-wing group Family Action Group got 1.31 per cent.

In the 1981 election, the Labor vote fell to 51.78 per cent, to give Labor eight seats, the Liberals suffered a dramatic drop to 33.77 per cent, to win five seats, and on the right of politics Fred Nile’s Call to Australia party got the whole of the Liberal drop, 9.11 per cent and one seat. The Democrats got 4.03 per cent and won one seat.

In 1984 the Labor vote dropped further, to 46.88 per cent, to give Labor seven seats. The Liberals went back up a bit to 42.61 per cent. The Democrats got 3.15 per cent, and Call to Australia 6.09 per cent (for one seat).

In 1988, the Labor vote fell dramatically to 37.51 per cent (for six seats). The Democrats got 2.7 per cent and won one seat, and small parties like the Humanists, Environment and Nuclear Disarmament groups, on the left side of politics, got 3.95 per cent. The Liberal vote was rather high at 46.15 per cent, and they won seven seats.

In 1991, Labor got 37.29 per cent (for six seats), the Democrats went up dramatically to 6.7 per cent (for one seat), the Greens, at their first appearance, got 3.32 per cent, and the No Toxic Incinerator group, on the left side, got 0.58 per cent. On the conservative side, the Liberals got 45.34 per cent (for seven seats), Call to Australia got 6.7 per cent (for one seat), and other right wing populists got 2.65 per cent.

In 1995, a new system was adopted. The number in the house was reduced to 42, and elections were held for half this number every four years to coincide with elections for the Lower House, which now had fixed four-year terms The effect of this was to reduce the quota to a bit less than 5 per cent.

In this election, the names of parties were listed on the ballot paper, with the possibility of an above the line vote, selecting the party ticket, which had the effect of casting preference votes from that ticket for the pattern of preferences registered by each party with the Electoral Office. As a result of these new arrangements, smarties on both sides of politics registered a pile of micro parties, with little membership, but with names designed to attract particular constituencies.

The constellations of party preferences produced some surprising results. As might be anticipated in such a situation, the ALP vote dropped a little further to 35.25 per cent and the Liberal vote dropped even more, to 38.49 per cent. The Greens got 3.75 per cent (and one seat), the Democrats got 3.21 per cent (and one seat), John Tingle’s Shooters’ Party, which had somewhat more tangible existence than most of the micro parties, got 2.84 per cent (for one seat), the Call to Australia vote dropped dramatically, its vote being scattered amongst other right-wing micro-parties, but it still got 3.01 per cent (and one seat). The most unpredictable result was a micro-party with the attractive name of A Better Future for Our Children, which got 1.28 per cent, but built up to a sufficient proportion of a quota by shrewd preference swaps to win the last seat.

1999. The tablecloth ballot paper

In the 1999 Upper House election, the tendencies inherent in the new arrangements worked themselves out in the most spectacular way. The Labor vote held up well, at 37.3 per cent (for eight seats). The Liberal-National vote plummetted to 27.4 per cent (for six seats). A very large number of the voters on the conservative side of politics obviously voted for right-wing micro-parties with appealing names.

The ballot paper was so large that the media dubbed it the Tableclot. The largest new right wing party was One Nation, which got 6.3 per cent (and one seat). The renamed Call to Australia party, now the Christian Democrats, got 3.2 per cent (for one seat). Mainly right-wing micro-parties got about 10 per cent of the vote between them, and something called the Outdoor Recreation Party, elected one Upper House member on a tiny 0.2 per cent of the vote, by means of ingenious preference swaps. The Democrats got 4 per cent (and one seat), the Greens got 2.9 per cent (and one seat), and two leftist micro-parties, Reform the Legal System, and Unity, got 1 per cent each, and each elected one member by judicious preference swaps. The Shooters Party got 1.7 per cent, but missed out, and a grouping called the Progressive Labor Party, with almost no campaigning or membership, got a rather large 1.6 per cent, clearly on the basis of name confusion with the ALP, but missed out on a seat.

The complex nature of the 1999 election led to changes in the Electoral Act. A rigorous arrangement requiring parties to have at least 750 real members, tested by the Electoral Office, was introduced. The non-public registration of parties’ preferences was eliminated. The net result of these changes iwa to dramatically reduce the number of micro-parties registered for next year’s Upper House election.

An overview of the changes in electoral politics in NSW from 1901 until now

The first obvious change is the dramatic broadening and democratisation of the electorate, from 300,000-odd to 3.8 million odd, almost 13 times, in a period when the state’s population has only increased a little less than five times. Another dramatic change is the shift from country to city. In 1901 Sydney was only about one third of the state’s population, and today it is about two thirds. In this period, the voting age was reduced to 18 years, which generally favours Labor or the Greens because the youngest cohort in recent times has generally leaned to the progressive side in electoral politics.

All these changes have weakened the electoral influence of the conservative side of politics.

The nature of electoral campaigning has changed over the 100 years. The impact of public meetings has dramatically diminished. Elections now are fought largely on television, and to a lesser extent through radio and newspapers, with a a heavy focus on the leaders of the political parties, which tends to weaken the local element in electoral politics, although the local element comes back dramatically, even in these circumstances, from time to time.

The swing against Labor in 1988 produced a large number of independents, mainly in the Hunter region, and the swing against the Coalition in 1999, produced another cluster of local independents, mainly in the country.

Government funding for election campaigns has also had a significant impact. Political parties receive some government funding for every vote polled, which obviously tends to reinforce the campaigning possibilities of the major parties, in this age of expensive television advertisements.

The ethnicity of the electorate has been transformed by mass migration, but the broad class division between the two sides in electoral politics has remained

For quite a while, conservative pundits have repeated a mantra that Labor has lost its traditional working class vote. Right wingers in the labor movement have also taken up this dirge. Socialist sectarians, who’ve become stuck in the eccentric rut of opposing Labor and Green parties electorally, but who only get a derisory vote themselves, also grasp at straws attempting to convince themselves that the core Labor vote is no longer significant.

Certainly, the sociology of Australian society has changed dramatically in the period under discussion, but no serious empirical investigation of the interface between sociology and electoral behaviour of the sort in which I have just engaged gives rise to any reasonable conclusion that the core Labor working class vote no longer exists.

The character of the working class has unquestionably changed, with a relative decline of manufacturing industry, and the transfer of workers to more skilled service areas, coincidental with a massive increase in the number of workers who have tertiary education.

In the past 20 years the “footprint” of trade unionism has declined from a little over 50 per cent, to a little less than 30 per cent of the working population, partly for a conjunctural ideological reason, the prices and incomes accord and the amalgamation of unions into large monoliths more remote from their memberships, and partly because new areas of work are intrinsically a bit harder to organise and the people working in many of them are subject to an incessant ideological barrage stressing individualism, etc.

This drop in trade union density and influence has bottomed out. The strenuous efforts of the ruling class to push the unions out of the Labor Party underlines the importance and potential danger of the union-Labor Party link from the point of view of the ruling class. The battle to defend and maintain the ALP-trade union link, and thereby preserve the working class basis of the Labor Party, is still continuing.

The main new feature of this electoral situation is the emergence, primarily in the, now elected, Upper House, of the Greens, who are politically to the left of Labor (but achieve the neat electoral device of locating themselves both to the left of Labor and in the centre) and the Democrats, both of which parties get their votes overwhelmingly from tertiary educated people, who make up the new social layers. On the right of politics, the emergence of a, mainly rural-based, right-wing populist formation, One Nation, is an electoral problem more for the Coalition than it is for Labor.

From about 1910, electoral politics in NSW has been a relatively clear map of class. Industrial workers, trade unionists, poorer farmers, most Catholics and some sections of the middle class voted Labor in 1910. In 1999, NSW politics is still a map of class. Industrial workers, organised trade unionists, Catholics, non-believers, non-Christians and NESB migrants, in their substantial majority, vote Labor.

The main difference in electoral politics lies in the rapid and continuing growth of the intermediate group or, in shorthand terms, the new social layers. This group is primarily divided between the Labor and conservative side of politics, but a significant section of this group also vote for the Greens and the Democrats. From my standpoint, as a Marxist, primarily interested in working class mobilisation in the broader sense, the sociology of electoral politics in this state, up until now, underlines the importance of Marxists having a serious orientation towards the Labor Party-trade union continuum, and also, in a secondary way, towards the Greens.

Federal election results in NSW

It is possible to do a similar inquiry into federal election results in NSW over the period. The results of such an investigation are similar, although sometimes the circumstances are different because of speedy changes in the political cycle, and the fact that state and federal elections take place at different points in this cycle.

The most striking example of such a speedy change in the political cycle was the election victory of Wran in the 1976 NSW election only six months after the defeat of Whitlam in 1975.

Despite shifts in the political cycle, the underlying demographic and social trends to which I have pointed have also had a major impact on federal electoral politics in NSW.

Self-interested, or just plain ignorant, neglect of empirical sociology as a factor in commentary on electoral politics is rather widespread

On the conservative side of politics, and in the media, there is a pronounced bias towards neglect of serious empirical sociology in commentary on electoral politics. The same sort of stupidity is repeated by some on the far left, who grasp at straws to try to buttress electoral pretensions that, empirically speaking, have no future at all.

From a Marxist point of view, my primary objection to this myopic independent socialist electoral strategy is that it is a powerful obstacle to Marxists engaging in their main tasks, which ought to be exerting direct influence in the labour movement, the working class and among the leftward leaning section of the new social layers.

The ferocious exposure rhetoric directed both at the leadership and supporters of the two major electoral representatives of the progressive side of Australian society, Labor and the Greens, which flows from the desire to achieve the Sysyphean task of carving out an electoral space in opposition to Labor and the Greens, has become a powerful obstacle to those “Marxists” exercising any influence at all on Labor and the Greens.

The groups in the Socialist Alliance have used their smallish but quite effective machine, and their very considerable energy to become the only far left group registered for the NSW state elections. It is possible to predict, with reasonable certainty, the electoral result of those efforts. They will get a vote in the region of that achieved by the Democratic Socialist Electoral Alliance, the only far left group registered in the Upper House in 1995 — 0.25 per cent, maybe a little more, maybe a little less.

The three volumes of the People’s Choice are rather expensive, but they are well worth the money for anyone with an interest in electoral politics in Australia, from any point of view really, but particularly for socialists who take electoral politics seriously as one of the major indicators of the forces at work in society.