The pragmatic Australian electorate will be convinced by workplace results, argues Andrew Norton

Twenty years ago the then Prime Minister, Bob Hawke, dismissed the HR Nicholls Society, a forum for supporters of industrial relations deregulation, as ‘political troglodytes and lunatics’.[1] The centralised system of setting wages and conditions seemed secure. Through the 1990s, however, the tribunals that had historically set Australian wages and conditions were pushed to the margins of the industrial relations system. By 2004, wages and conditions for the vast majority of employees were at least partly determined through localised bargaining, with just 20% of workers still receiving award rates set by tribunals.[2]

With its unexpected Senate majority from July 2005, the Howard government is promising new industrial relations reforms. How far they will go remains to be seen. The government’s proposals to date include a new way of centrally setting a minimum wage, relaxing unfair dismissal laws for small business, and reducing third-party interference with agreement making. So far, there is no published polling on these plans. However, past opinion polls tell us something about how the public sees these issues, and where the political dangers for the government lie.

Long-term opinion change

Despite the controversy surrounding industrial relations reform, over time the public has substantially changed its mind about how wages should be set. In the mid-1980s, around the time of Hawke’s attack on the HR Nicholls Society, two polls gauged public sentiment on the arbitration system and its alternatives, and these can be seen in Table 1. Both showed clear, though varying, majorities supporting the then existing arbitration system. The 1984 National Social Science Survey (NSSS) found three-quarters of respondents favored arbitration, and a 1985 Gallup Poll found 60% support. The difference may be as much due to the questions asked as actual changes in opinion between the polls. The NSSS offered a ‘mildly favour’ option with 13% mildly favouring arbitration, and 6% mildly favouring employers and trade unions negotiating. This suggests about 20% of respondents had soft views that could be swayed either way. In the NSSS, 62% of respondents ‘very’ or ‘fairly’ strongly supported arbitration, bringing it very close to the pro-arbitration view expressed in the simple Gallup Poll two-option choice.

Polling in the early 21st century suggests that, at least among people in the workforce, a similar sized majority to the one that once supported arbitration now prefers decentralised wage setting. In the 2003 Australian Survey of Social Attitudes (ASSA) 64% of working respondents agreed that ‘employers and employees should be able to negotiate pay and conditions directly’. 17% disagreed, with a large group of undecideds. Two polls carried out by Irving Saulwick and Associates for The Age’s pre-2001 and pre-2004 election surveys, and shown in Table 2, similarly suggest that more than half of employees favour agreements between employers and employees.

 

Different systems for different workers—concerns about the minimum wage?

While a centralised system for all workers now has minority support, there is likely to still be a large constituency for some workers to have their wages set by tribunals. In the Saulwick Poll cited above, the pollsters note that people tend to support whatever system they are under, or at least think they are under.[3] If we examine ABS figures on wages, full-time employees under collective agreements earned an average of $993 a week in 2004, and those with individual arrangements earned an average $1,022 a week. For workers on awards, average weekly pay was $643.[4] It is not hard to see why those with negotiated wage deals thought they were doing OK, and do not want to go back to the old system. Similarly, workers on the lower award wages may infer that they have weak market power, and prefer the status quo to further change.

The polling therefore suggests that the public does not see methods of setting wages and conditions as either-or alternatives. Different systems should apply to different workers. Other polls also suggest this interepretation. In the Australian Section of the 2001 International Social Science Survey, significant and quite similar rates of support (45%-54%) were found for broad equivalents of all three broad types of wage setting.[5] Confusion may explain some of this, but is also consistent with seeing a role for each, depending on worker circumstances.

Concern about the market weakness of some workers is also evident in other polls. According to the 2003 ASSA a large minority, 43%, agree with the proposition that ‘employees will never protect their working conditions and wages without strong unions’. A corollary of this is that without union protection there is a place for tribunals to protect low-paid workers. When we take into account the nearly 40% of respondents in the Saulwick Poll who thought they were better off under awards, it is likely that the Australian electorate would take some persuading before it would support complete abolition of the award system—though this is a proposal that is not on the Howard government’s political agenda. Of announced policy intentions, it suggests that there will be political sensitivity on how minimum wages are set.

Job security—worries about unfair dismissal?

While the government and employers argue that unfair dismissal laws costs jobs, by increasing the cost and risk of taking on staff that might later prove unsatisfactory or unnecessary, job security is valued by already employed workers. Researchers using data from the Australian International Social Science Survey found that job security significantly affects satisfaction with pay and standard of living. Even added pay does not necessarily compensate for the fear of being without a job. In their trend data, perceived job security declined between the late 1980s and the mid-1990s, and an analysis of the ISSS estimated that a pay rise of 13% would be needed to compensate for job security lost in that time.[6] Rising unemployment levels, which increase job security concerns, have been shown in European studies to decrease national life satisfaction levels.[7]

Another potential political difficulty is that there is already a widespread perception that job security has declined over time, as Table 3 shows. This 2003 ACTU-commissioned Newspoll found that 64% of respondents believed that job security in Australia was getting worse, and a tiny 6% believed it was improving.

How a perceived reduction in job security plays in the electorate will depend partly on the proportion of people making political decisions according to their own situation, and the proportion who are also concerned with the situation of others. A common feature of public opinion is that people believe that their own situation is better than that of others or society a whole. At any given time, most people, for example, believe that their own job is safe, as figure 1 shows. Nor has there been any long-term decline in perceptions of personal job security. Polling on this goes back to 1975, and only recessions seriously affect perceptions. The most recent ‘job safe’ perception of 79% is only slightly below the peak of 82% recorded in 1977, 1984 and 1989.

If people believe that their employer needs them, they are not likely to fear for their jobs, regardless of the law relating to dismissal. On a 0-10 scale in 2001, the most easily sacked employees, casuals, rated their satisfaction with job security at an average of 6.77, compared to 7.92 for ‘permanent’ staff.[8] During the 1998 waterfront dispute, when Patrick Stevedores sacked union workers, a Morgan Poll found that 84% of employed respondents believed that they had ‘no chance’ of being dismissed that way, while only 14% thought they might be sacked that way.[9] There is not an endemic fear of arbitrary dismissal.

However, in my analysis of the politics of trade protection I found evidence that the main reason protectionism remained popular was the belief that it preserved jobs. Opinion was nearly identical whether the respondent believed their own job was secure or not.[10] This suggests that policies that reduce job security may be unpopular, whether or not voters fear for their own jobs.

However, data on objective job security, shown in Figure 2, suggests that many more people worry about losing their job than in fact do lose their job in any given year. The retrenchment rate counts those whose employer went out of business or who were sacked or laid off; the job loss rate includes retrenched workers, those whose seasonal or temporary work had ended and those who had stopped work due to ill-health or injury. Comparing this to Figure 1, about three times as many people in 2004 thought that they might lose their jobs as did actually lose their job. It is difficult to know to what extent concerns about job loss were realistic, and to what extent unfounded concerns were created by other factors, including perhaps general assumptions that jobs are less secure than in the past.

 

One trend in the opinion poll data suggests that now is the time to change unfair dismissal laws. A Figure 3 shows, unemployment has all but slipped out of the three issues voters nominate as being most important for the federal government, with it being suggested by only 9% of respondents to the 2004 Roy Morgan survey. However, the rapid spike during the early 1990s recession suggests that the current low political priority given to unemployment could change quickly if the economy weakens.

 

 

Union power and third-party interference with agreements

The government is signalling that it wants to reduce the interference of third parties, including unions, with agreements. In this, they can no longer tap into the strong anti-union sentiment that was once a reliable feature of Australian public opinion. Typically over the last 30 years around two-thirds of poll respondents have agreed that unions have too much power (figure 4 starts in 1987 with the Australian Election Survey series, but earlier one-off polls showed similar views). A plurality of poll respondents still believe that unions have too much power, despite a very large reduction in strikes. But it is no longer a majority view and is trending down. By contrast, the proportion of people saying that unions do not have too much power (impliedly, about right or too little) though still less than one-third, has doubled since 1987, as has the share of voters with agnostic views on unions. Together, they are a majority.

 

Consistent with the trends seen in figure 4, the proportion of electors supporting stricter regulation of trade unions is also in decline, dropping from over two-thirds in 1990 to 44% in 2004, as seen in figure 5. However the shift is more toward a neutral position—neither agreeing nor disagreeing with stricter regulation—than to disagreeing with stricter regulation.

 

The softening of views toward unions is paralleled with a hardening of views about big business, as seen in figure 6. In the late 1980s, around half of those polled thought that big business had too much power. By 2004 that proportion was nearing three-quarters. Though the total proportion of respondents believing that business had too much power was stable between 2001 and 2004, the proportion strongly agreeing increased from 32% to 44%. So while most workers prefer to have their wages and conditions set between employers and employees, there seems to be a growing perception that big business has too much power, and that unions are needed to counter-balance it.

 

 

Conclusion

The political acceptability of further industrial relations change will not turn on abstract or ideological ideas about appropriate systems. Public opinion research confirms the conventional wisdom that the Australian electorate is unideological.[11] The mixed views in evidence in the polls cited above support this interpretation of public opinion—the workers who are doing well support decentralised wage fixing for themselves, but not necessarily for others, and generally positive views about wage setting methods co-exist with negative views about job security and about the power of big business.

To achieve further industrial relations reform at minimum electoral cost, the government will need to focus not just on the benefits of further change, in lower unemployment and higher productivity, but also on allaying concerns about perceived negatives. On the evidence of past polling, the major concerns are with the protection of workers with little market power and with job security. With no information in the polls on the ‘right amount’ of power for trade unions or big business, it is difficult to tackle these perceptions directly. With job security, however, there is good objective data showing that concerns about this are likely to be exaggerated. More obvious objective evidence in strong wage growth and low levels of strikes probably explains pragmatic acceptance of less centralised industrial relations. The level of job security is much harder to perceive from daily experience, and so there needs to be a more concerted effort to show that results here are good too. But results are the key, which suggests that if the government is going to make major changes it should do so early in its parliamentary term, so that it has some good news to tell by the time of the next election.

Andrew Norton is a Research Fellow at the CIS and Editor of Policy. He is currently writing a book on public opinion about economic reform.