PATRIOTISM, IMMIGRATION AND THE 1996 AUSTRALIAN ELECTION

Katharine Betts

In March 1996 there was a gap between political candidates and voters on the question of immigration and (with the exception of Coalition candidates) on the question of pride in Australia's history. On both questions Coalition candidates' opinions were closer to those of the voters. Concern about immigration is unlikely to have cost Labor the election by itself, but this concern is linked to feelings of national pride. Both attitudes are strongly associated with a vote for the Coalition and may well have swung the tide against the Keating Government.

After the 1996 election the Australian Election Study (AES)¹ canvassed the views of a random sample of voters on a range of topics: how did they vote, what influenced their vote, and what were their opinions on a range of other political and social questions. A previous article drew on this study, as well as a range of other surveys, in an examination of public attitudes to the immigration program since 1961.² That examination showed that around two thirds of Australians have been opposed to current levels of immigration since the mid 1980s, but that this opposition has not been uniformly spread across the population.

Two groups stand apart from the general trend. People with a university education are less opposed to immigration than other educational groupings and so are people who were born overseas in non-English-speaking-background (NESB) countries.³ In contrast, less educated people are in general more likely to express opposition to the intake. In the 1996 AES survey just under 63 per cent of respondents thought that the current intake was too high but the group most opposed were people with trade qualifications (or their equivalent). Here 73 per cent thought the current intake was too large, as did 68 per cent of the group with no post-school qualifications at all.

For most of its time in office Labor had pursued a high immigration policy. It also supported multiculturalism in the sense of cultural maintenance (and, of course, in the 'access and equity' sense of the term as well). Unfortunately the AES has few direct questions on multiculturalism and none on attitudes to structural pluralism. But it is possible that some of the feeling about immigration, rather than expressing an unequivocal opposition to population growth, also reflects unhappiness with the idea of emphasising cultural differences. The AES did, however, ask a range of questions about the pride respondents took in aspects of Australia's achievements and heritage which permit an exploration of the link between patriotic feelings and voting patterns.

On March 2 in 1996, after 13 years of government, the Labor Party, led by Paul Keating, lost to the Liberal-National Party Coalition led by John Howard (leader of the Liberal Party, the senior partner in the Coalition). People who are unskilled or who have trade qualifications were the group most dissatisfied with the Labor Government's immigration policy; they might also be presumed to constitute a large section of the Labor Party's natural constituency. Did some of them desert Labor at the poll and, if they did, was this desertion linked to immigration policy?

On the face of it, a causal link between immigration and Labor's defeat seems unlikely. As far as the two major political groupings were concerned (Labor and the Coalition), immigration was not an election issue.⁴ The Coalition and the Labor Party did not offer the voters a choice on immigration: both sides said they would keep the program much as it was, thus maintaining the long tradition of political bipartisanship between the major parties on immigration.⁵ And, in contrast to the heated exchanges sparked by Pauline Hanson's maiden speech in September this year,⁶ immigration was not widely debated during the campaign. When the AES asked its respondents to say how much various issues had affected their vote, immigration was, as Table 1 shows, a fair way down the list.

Nonetheless it was not an unimportant issue. It outranked 'privatisation' which, with the proposed sale of Telstra, was an election issue, as was the question of links with Asia. Table 2 shows that, although the two major parties appeared to offer no change on immigration, voters did believe that there was a difference; pro- or anti-immigration, 66 per cent of voters did find some dissimilarity.

Data from the 1996 Australian Candidate Study⁷ are now available. These illustrate the degree to which people who ran for parliament shared the feelings of those whom they hoped to represent. They also show the extent to which attitudes to immigration (and other questions) differed between groups of candidates affiliated with different parties. Table 3 shows that there is a wide gulf between the attitudes of the voters and those of the candidates: voters were nearly five times more likely to say that the numbers of migrants had gone much too far. But while all candidates were much less concerned about immigration than were the voters, candidates standing for the Coalition were closer to the voters' feelings than Labor. (However, the two minor parties, the Democrats and the Greens, were closer still.) The gap between the attitudes of the Labor candidates and those of the voters is striking: here voters were 10 time more likely to say that the intake had 'gone much too far' than were candidates from the former governing party.

It is possible that some voters, while they lacked explicit assurances, hoped (or feared) that the Coalition would adopt a more restrictive immigration policy and tailored their vote accordingly. It is also possible that these feelings were reinforced by the demeanour of some candidates, despite the lack of an overt policy difference between Labor and the Coalition. Of course, the fact that the Australians Against Further Immigration Party (AAFI) received a very low vote speaks against such a theory.⁸ If immigration had been a burning issue which the two major groupings had decided to ignore, the AAFI vote would surely have been higher. But if immigration had been linked to a range of other issues concerning national feeling, could it have swayed some votes?

The AES study of the voters provides some help with this question but, though it was a based on a random sample of people on the electoral role, the AES was a self-reported mail questionnaire. The research instrument consisted of a substantial booklet of 27 pages and, of the 3000 posted out, only 1797 (60 per cent) came back. While this is a very good response rate for research of this kind, it is likely to reflect a bias in favour of people who have some spare time and who are not averse to reading, reflecting and answering questions. University-educated people are over-represented in the sample (they make up 19 per cent of respondents, as opposed to eight per cent in the general population aged 15 plus) as are Coalition voters, but to a much lesser extent. In the sample 50 per cent voted Coalition in the House of Representatives and 35 per cent voted Labor. But in reality the Coalition's votes was somewhat weaker than this (47 per cent), and Labor's was stronger (39 per cent).⁹ The over-representation of university graduates and of Coalition voters means that we should be careful about making direct generalisations from the sample to the wider population of voters but it does not invalidate comparisons within the sample. We can learn from results which show that some sub-groups are more or less likely to vote in a particular direction without needing to make direct extrapolations to the general population.

Table 4: Voting in the House of Representatives by education, birthplace and attitudes to immigration (percentages)
	Voting, whole sample and by education			Immigration too high, voting in total sub-group and by education
	Total	Trade...	Uni...	Total too high	Trade...	Uni...
Coalition Labor Dem.+ Green	50 35 9	52 34 8	45 38 11	55 30 8	55 31 7	59 22 12
Total % N	100 (1,797)	100 (524)	100 (343)	100 1,113	100 376	100 137
	Immigration OK, voting in total sub-group and by education			Voting by birthplace
	Total OK	Trade...	Uni...	Australia	Overseas	NESB
Coalition Labor Dem.+ Green	42 43 10	46 40 11	36 49 11	50 33 10	49 40 5	46 44 5
Total % N	100 (652)	100 (140)	100 (200)	100 (1,342)	100 (384)	100 (203)
	Immigration too high, voting by birthplace			Immigration OK, voting by birthplace
	Australia	Overseas	NESB	Australia	Overseas	NESB
Coalition Labor Dem.+ Green	55 29 8	57 31 6	59 27 7	42 41 13	41 48 4	36 56 3
Total % N	100 (878)	100 (198)	100 (88)	100 (440)	100 (180)	100 (110)
Notes: People who did not answer the question on voting for the House of Representatives or who voted for other parties or informally (8 per cent of total sample) are not shown separately. 'Education' is highest post-school qualification, 'trade' includes equivalent non-trade qualifications, and 'uni' includes both undergraduate and postgraduate degrees. 'Immigration too high' consists of the responses immigration has 'gone too far' or 'much too far'. 'Immigration OK' consists of the responses immigration is 'about right' [the majority in this category] or has 'not gone far enough' or has 'not gone nearly far enough'. 'NESB' is people born overseas in non-English-speaking-background countries. 'Dem.' is the Australian Democrats.

Tables 4 and 5 provide an overview of the votes recorded in the AES by education, birthplace and attitudes to immigration. They show only two educational groups: those with university degrees (postgraduate as well as undergraduate) and those with trade (or equivalent non-trade) qualifications. This selection has been made to simplify the presentation of the data and these two educational groups were chosen because, in almost every case, they represent the two extremes. In both the House of Representatives and the Senate vote, people with trade qualifications were the educational group with the highest Coalition vote and the graduates were the group with the lowest. The Labor vote showed less variation by education, but in the House of Representatives it reached 33.6 per cent both for those with trade qualifications and those with non-university diplomas (the lowest figure) and at 37.6 per cent, was highest amongst graduates. (In the Senate, graduates' support for Labor was reduced by their support for the Democrats and the Greens and this meant that Labor's Senate vote varied little by education.)

Table 5: Voting in the Senate by education, birthplace and attitudes to immigration (percentages)
	Voting, whole sample and by education			Immigration too high, voting in total sub-group and by education
	Total	Trade...	Uni...	Total too high	Trade...	Uni...
Coalition Labor Dem.+ Green	44 30 15	46 30 14	38 29 23	50 25 13	48 27 15	54 16 19
Total % N	100 (1,797)	100 (524)	100 (343)	100 (1,113)	100 (376)	100 (137)
	Immigration OK, voting in total sub-group and by education			Voting by birthplace
	Total OK	Trade...	Uni...	Australia	Overseas	NESB
Coalition Labor Dem.+ Green	36 37 15	43 36 13	29 38 25	46 29 15	41 34 14	39 38 9
Total % N	100 (652)	100 (140)	100 (200)	100 (1,342)	100 (384)	100 (203)
	Immigration too high, voting by birthplace			Immigration OK, voting by birthplace
	Australia	Overseas	NESB	Australia	Overseas	NESB
Coalition Labor Dem.+ Green	50 25 13	49 26 13	53 22 14	38 35 21	33 41 14	30 51 5
Total % N	100 (878)	100 (198)	100 (88)	100 (440)	100 (180)	100 (110)
Note: People who did not answer the question on voting for the Senate or who voted for other parties or informally (11per cent of the total sample) are not shown separately. For other notes see Table 4.

The first panels in Tables 4 and 5 show the voting pattern for the whole sample and by education. They reveal high levels of support for the Coalition among people with trade qualifications (especially in the lower house) and low levels of support for the Coalition from graduates. They also show the low level of support for Labor amongst people with trade qualifications, but here the general pattern does not vary very much by education. If people with trade qualifications are unhappy with the Labor Party, they are not very much more disaffected than the rest of the sample. But the situation is different when we divide the sample into those opposed to immigration and those who are not opposed, and we look at voting by education within these two sub-groups.

Amongst the critics of immigration the vote for the Coalition is five to six per cent higher in both houses while amongst those who are comfortable with immigration or who want more, the vote for Labor is seven to eight per cent stronger in both houses. Unfortunately for the Labor Party, the people in the latter camp were outnumbered nearly two to one by those in the former. Nevertheless, university-educated people who were happy with immigration gave Labor solid support in the House of Representatives (their Labor vote was a full 14 per cent higher than the average for the sample as a whole). In the Senate graduates were eight per cent more likely to vote Labor than the sample as a whole and 10 per cent more likely to vote Democrat or Green.

What of the immigrants themselves? The fourth panel in Table 4 shows that migrants voting for House of Representatives' candidates are less likely to vote Democrat or Green than are the Australian-born and that this tendency boosts their support for Labor. But their general level of support for the Coalition varies very little from that of the Australian-born. However, the sub-group of immigrants who were born in NESB countries was less likely to vote Coalition and nine per cent more likely to vote Labor than the sample as a whole.

So far Tables 4 and 5 support the hypothesis that immigration (or some other issue closely related to it) did matter in the vote for the mainstream parties. If voters were making the assumption that the Coalition was likely to take a more rigorous approach to immigration control and that Labor was likely to be rather more relaxed, the direction of the vote follows. Groups who are known to be favourable to immigration are inclined to vote Labor and those who are known to be sceptical about immigration are inclined to vote Coalition. This is the broad overview. But the details contain some surprises. University-educated people do tend to favour immigration and to vote Labor, but graduates who are sceptical about immigration provide strong support for the Coalition. The same is true for NESB migrants. A majority of these migrants favour immigration and vote Labor but amongst the sub-group of NESB migrants who are opposed to current immigration policy, a large majority vote for the Coalition.

Both of these groups, university graduates and NESB migrants, have produced active participants in the immigration debate provoked by Hanson's speech. A majority within them are happy with immigration (and support Labor) but a substantial minority are opposed to immigration (and support the Coalition). Perhaps this underlying political polarisation within the two groups helps explain some of the passion? But it is also possible that Tables 4 and 5 provide an overly simple picture. Can it really be true that the immigration sceptics somehow guessed that the Coalition was on their side, even though they had not been told about it, and tailored their vote accordingly? This is, of course, a possibility. In 1993 the then Liberal leader John Hewson did argue for a smaller intake (breaking the post-war tradition of bipartisanship) and in 1988 John Howard had been the focus of an extraordinary assault by opinion makers in the press after he said that it would help social cohesion in Australia if Asian immigration 'were slowed down a little'.¹⁰ So even though the Coalition went into the 1996 election with the policy of maintaining the intake at about the current level, immigration sceptics may have hoped that it would set lower immigration targets than its opponents. But this hypothesis is undermined by the low vote for the AAFI.

An alternative hypothesis starts from the position that immigration itself may have been an issue of only medium importance for most electors. Opinions about immigration may have produced the strong patterns shown in Tables 4 and 5 because they were working as a part of group of cognate attitudes, including patriotic feelings, economic anxiety, a yearning for stability and a desire for a stronger sense of national community.

Previous research suggests than an outlook focussed on a positive view of the national community combined with doubts about the benefits of internationalism is more prevalent among less educated people. In contrast, a blasé attitude to national borders, and enthusiasm for economic globalism and cultural cosmopolitanism, are more prevalent among university graduates.¹¹ These are the people who help constitute Alvin Gouldner's 'new class'¹² (and Robert Reich's group of 'symbolic analysts').¹³ Some of them may feel themselves to be either insulated from global economic competition, or well placed to meet it, and a number are not only confident about their own abilities and their own values, they are openly contemptuous of their less sophisticated compatriots.¹⁴

Recent Australian research has provided further empirical evidence for this split in Australian public opinion by showing that people without a tertiary eduction are more concerned about economic protection than are graduates.¹⁵ Table 6 shows that the AES confirms this pattern. Less educated people continue to be more in favour of protection than university graduates but the difference between them on this question is not as marked as it is on the question of immigration.

Table 7 shows a high proportion of people prepared to say that they are 'very proud' of Australian sport, and quite a few with similar feelings about Australian science, literature and history. In general, less-educated people express more pride. This is consistent with the hypothesis that people who are less-educated hold a more parochial, Australia-centred view and that graduates are more attached to cosmopolitanism. Nonetheless, the rankings of the attributes are relatively uncontroversial. While the university-educated may be more restrained than the rest, they rank the attributes in more or less the same order. But there is one exception. One attribute in Table 7 is controversial and that is pride in Australian history.

Table 7: 'How proud are you of Australia in each of the following?', per cent saying 'very proud', total sample and by highest post-school qualification
Rank	Attribute of Australia	Total very proud	no post-school qualific-ation	trade and other non-trade	undergrad diploma or associate diploma	bachelor's degree or post graduate
1 2 3 4 5 6 7 8	Its achievements in sports Its scientific and technological achievements Its achievements in the arts and literature Its history The way democracy works Its social security system Australia's economic achievements Its political influence in the world	67 44 38 37 29 15 10 9	71 46 40 44 29 16 13 11	72 45 40 40 27 14 12 8	62 46 42 27 34 20 7 9	53 41 33 22 32 13 5 5
	N	(1797)	(706)	(524)	(140)	(343)
Note: People who did not answer the question on qualifications (n = 84) are not shown separately. (There were four possible responses: 'very proud', 'fairly proud', 'not very proud' and 'not proud at all'.)

If it were left to people with no post-school qualifications, or with trade certificates, Australian history would have ranked third (or equal third) among the aspects of Australia of which people were very proud. But graduates show only half their enthusiasm and if they alone had been questioned, it would have ranked fifth. John Howard has several times criticised what he refers to as the 'black armband' school of Australian history, where, in his opinion, too much attention is given to the darker episodes in Australian history at the expense of pride in our forbears' achievements.¹⁶

Table 8: Australian Election Study - pride in Australian history, voters and candidates (percentages)
	Total all voters	Total all candidates	Candidates by Party
			Coalition	Labor	Dem.	Green
Very proud Fairly proud Not very, or not at all, proud No response	37 41 18 4	22 37 38 3	45 43 9 3	26 44 31 -	10 42 46 2	3 17 73 7
Total % N	100 (1797)	100 (439)	100 (122)	100 (110)	100 (111)	100 (95)
Source: AES and Candidates' file, 1996 Note: The last two responses, 'not very proud' and 'not proud at all' have been combined.

While Keating made the running in the debate about the republic, his support for Mabo, for closer integration with Asia, and for multiculturalism, may have made it seem as if he had less respect for the old Australia. (Of course if patriotism were an uncomplicated issue the republic should have been a plus for Labor among more parochial Australians. The AES shows that 59 per cent of voters favoured a republic, but the issue does not seem to have played a part in the election.)¹⁷