Submission 167 Australian Council of Trade Unions Workplace Relations Framework Public inquiry

The problem of inequality

Among the lowest wage earners, however, the story is less salutary. Unlike the United States, Australia has largely escaped the blight of a large working poor population,⁷¹ but the increases in the Federal Minimum Wage has not kept pace with average earnings. In 2003 the ratio of the FMW to average full-time wages (AWOTE) was 48.2 per cent; by 2008 it had dropped to 46.9 per cent; and by 2013 it was just 43.3 per cent.

To place this in context it is worth reviewing the growth in wages inequality over the last 30 years (Figure ). Looking at employees at the bottom quartile, their real wages declined for most of the 1990s and did not return to the level of the 1980s until just after 2000. The gap between those at the bottom quartile and those on median wages began to diverge sharply during the mid 1990s. There was a brief closing of the gap in the early 2000s (among men) but the overall trajectory has been a steady divergence. The same dispersion has been evident at the top of the labour market with a sharp divergence between the median and the top quartile from about 2006 onward.

Figure : Inter-quartile dispersion in real wages, full-time employees, Australia 1982 to 2012

Wages have been adjusted by the CPI and indexed to 100 in 1982. Source: Unit record data from ABS Income Distribution Surveys from 1982 to 2012.
Figure highlights how much those at the top of the labour market have sprinted ahead and how much stagnation in real earnings has taken place in the bottom half of the distribution. For male employees in the top quintile the growth in real wages from 2003 onwards was quite striking. Among women there was a similar pattern, but much less pronounced. By way of contrast, in the two bottom quintiles—virtually the bottom half of the wages distribution—average real wages remained almost flat for the best part of 20 years before rising in more recent times.
Figure :Growth in real wages by quintile, full-time employees, Australia 1981 to 2012

Y axis shows real wages in 2012 dollars, indexed by the CPI. Source: Unit record data from ABS Income Distribution Surveys from 1982 to 2012.

Inequality measures such as these are vulnerable to a range of methodological caveats and the ABS Income Distribution Survey (IDS) data are no different. A number of researchers have raised concerns about using these data for analysing inequality over time, though some of these concerns relate to household income rather than to individual wages.⁷² More serious have been the criticisms of Roger Wilkins, who expressed scepticism at the size of the increases in inequality in the period between 2003 and 2006.⁷³ He pointed towards inconsistencies in ABS definitions and collection methods over time, such as in the treatment of salary sacrificed income and bonuses. Comparison of the HILDA data with the ABS IDS data does suggest that the size of the dispersion in wages is greater in the latter compared to the former, particularly for the male full-time workforce.⁷⁴ The main discrepancies, however, affect interpretations of the top half of the labour market, with the story at the bottom of the labour market largely consistent with the findings shown here.
Another ABS data source throws some light on earnings inequality at the top of the labour market. The survey of Employee Earnings, Benefits and Trade Union Membership provides data on both mean weekly earnings and median weekly earnings. Comparing the divergence over time between these two sets of measures provides a useful indication of dispersion at the top of the earnings distribution (since the more extreme skew in the earnings distribution is invariably towards the upper end). Figure shows data for the period from 1997 to 2013 and confirms that from the mid 2000s onwards these two measures began to increasingly diverge. There was some closure in the gap during the GFC, but the divergence resumed after 2011, particularly among male employees. In other words, inequality at the top of the labour market increased from the mid 2000s onwards, closed during the GFC, and grew again from 2011 onwards.
Figure : Growth in mean and median real earnings, Australia 1997 to 2013

Source: ABS Employee Earnings, Benefits and Trade Union Membership (6310) time series spreadsheets. Wages data adjusted for the CPI (2014 dollars) and indexed to 100 in 1997.

It is important to keep in mind that the top of the labour market here concerns wage and salary earners, those within the ambit of the industrial relations system. Excluded from this story are those with non-wage income, such as executives, the self-employed and those using trusts, partnerships and similar devices. Income from these sources is largely unregulated; including these sources in an analysis of inequality would most likely see the problem magnified considerably. One of the criticisms levelled at the Accord during the 1980s was that it failed to curb the excesses for those on non-wage incomes, particularly among executives and the self-employed.⁷⁵ It did however, rein in high wage earners, whilst augmenting the incomes of low paid workers with a substantial social wage. The flat-dollar increases associated with some of the decisions compromised long-standing relativities within the wages structure and led to discontent in some quarters. Nevertheless, within the wages system overall, inequality was not a problem during this period. The transition to enterprise bargaining from the mid 1990s onwards changed this, and inequality became an enduring feature of the wages system. The current industrial relations system is also unable to influence outcomes for non-wage incomes, and the taxation system is much less progressive today than it was in the late 1980s (leaving aside issues of tax evasion and avoidance). When it comes to wages income, enterprise bargaining has not been able to curtail wages inequality at either the top or the bottom. It has only been institutional arrangements, in the form of award safety net adjustments (that is, the Federal Minimum Wage), which have had any effectiveness in protecting the low paid from falling too much further behind.

Thus while the current largely-decentralised wage setting system provides great flexibility in quarantining high wage growth to the most profitable enterprises and sectors, it retains a built-in bias towards reproducing wage inequality. It is important to acknowledge that these two features go hand-in-hand and that only these institutional arrangements, in the form of awards and the Federal Minimum Wage, that have kept a lid on this momentum towards increased inequality.

Traditional arguments that promoting equity compromises efficiency have been overturned by a number of studies, including those from bodies like the OECD. There is increasing evidence that the best economic outcomes arise in countries where social solidarity is well developed, and the coordination of common interests is well advanced. These outcomes require modest levels of wage and income inequality and the future in Australia does not bode well.⁷⁶