Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Have tariff reductions increased earnings inequality?

Download PDFDownload PDF



Embargoed until 00.01am, 16 February 1998


Reduced trade barriers do not appear to have been a major contributor to the increasing inequality of earnings or of unemployment, according to a staff research paper released today by the Industry Commission.

The paper — Trade Liberalisation and Earnings Distribution in Australia — uses a wide range of data and several analytical approaches to investigate how cuts to tariffs and other trade barriers have affected low skill workers relative to other employees.

Increased dispersion of earnings has occurred while trade barriers have been lowered. But reduced trade barriers were found to have a minor effect on income distribution and employment. This is partly because the industries most affected since the mid 1980s account for only a small share of total employment in Australia.

The effects of reduced barriers were overshadowed by other structural changes in the economy. These led to both a shift in the distribution of employment away from industries which are intensive users of the least skilled occupations and a substitution within industries toward the highest skill occupations and salespersons and personal service workers.

No more than around one-twentieth of these occupational changes were found to be related to reduced assistance. Among the primary causes were adverse shifts in international trading conditions for primary commodities and technical change which favoured greater use of high skill labour.

The paper concludes that retaining or increasing trade barriers would have been of little use as a policy instrument directed at aggregate employment or income distribution.


For further information contact: Professor Richard Snape (media comment) - 015 880 013 (mobile) - 03 9653 2342 (office) Mr Greg Murtough (background) - 03 9653 2163

Media copies o f this report are available from Clair Angel on 02 6240 3239. Other requests should be directed to Government Info Shops. The paper can also be accessed via the Internet at


Embargoed until 00.01am, 16 February 1998

Changes in the distribution of earnings from employment

The general consensus of recent research is that the distribution of earnings from full-time employment in Australia has become more unequal and that this trend is most evident for male workers (pages 3-5 of the paper).

The figure below (from page 5 of the paper) shows how there has been a ‘hollowing out’ of the earnings distribution.

Figure 1.4 on page 6 of the paper compares the rise in Australia’s earnings inequality to other economies. Inequality has grown faster in the United Kingdom and the United States but it appears to have fallen in Germany.

Growth in the inequality of earnings from employment did not cause a similar rise in the inequality of disposable household incomes between 1981-82 and 1993-94 because of changes in household sizes, taxes and welfare payments (figure 1.5, page 7 of the paper).

Cumulative distribution of real total weekly earnings for full-time employees, 1983 and 1996a


Έ 25


Real earnings per week ($)

a Earnings for 1983 are converted to 1996 dollars using the consumer price index for the June quarter. Data source: ABS (Distribution and Composition o f Employee Earnings and Hours, Cat. no. 6306.0); ABS (Consumer Price Index, Cat. no. 6401.0).


Embargoed until 00.01am, 16 February 1998

Trends in wages and employment

There have been significant changes in the composition of employment. These include a shift from blue collar to white collar occupations; a greater share of employment in the services sector; and a shift towards part-time work (pages 30­ 34 of the paper).

Most of the growth in employment has been in services. Between 1987 and 1996 there were 1.3 million new jobs in services and 8 900 in agriculture. There was a fall of 16 000 jobs in mining and 55 300 in manufacturing (figure 2.7, page 33 of the paper).

The figure below shows the growing relative importance of the two highest skill occupations and the (low skill) category of Salespersons and personal service workers and the relative decline of the middle skill and two lowest skill occupations

Earnings relativities between different occupations were generally stable between 1986 and 1996 (pages 26-29 of the paper).

More detailed information is available in appendix A of the paper (pages 69-89).

Share of employment3 by occupation, 1987 and 1996fc

Managers & administrators


Para professionals



Salespersons & personal service workers

Plant & machine operators & drivers

Labourers & related workers

8 10

Per cent

14 16 18

a Note the data relates to employees, rather than employed, so excludes self-employed persons. “ Labour force data for May

1987 and May 1996. c Managers and administrators excludes Farmers and farm managers (ASCO 14). Source: IC estimates adapted from ABS (Labour Force Australia, unpublished data, 1987 to 1996, Cat. no. 6203.0).


Embargoed until 00.01am, 16 February 1998

Monash model results

• The Monash model of the Australian economy was used to analyse how eight factors contributed to changes in employment between 1986-87 and 1993-94 (the eight factors are detailed in box 4.1 on page 46 of the paper).

• The results show how structural change would have contributed to shifts in the distribution of employment between industries if there had been no growth in total Australian employment. The results indicate that between 1986-87 and 1993-94 total hours worked would have fallen in manufacturing (by 12.7 per cent) and agriculture and mining (by 15.8 per cent) and that this would have been offset by an increase in services (by 3.9 per cent).

. The above changes contributed to an increase in total hours worked by managers and administrators (by 2.2 per cent) and salespersons and personal service workers (by 5.9 per cent) but reduced hours worked by the least skilled occupations of plant and machine operators and drivers (by 8.8 per cent) and

labourers and related workers (by 1.8 per cent).

. No more than one-twentieth of these occupational changes were attributed directly to changes in assistance (summary, pages xii-xiii and figure 4.3, page 60 of the paper)

• Reduced industry assistance was only one of a number of (often larger) forces which caused a redistribution of employment from manufacturing to other industries (figure below and figures 4.1 & 4.2, pages 50 & 51).

Estimated sources of change in manufacturing employment, 1986-87 to 1993-94*

Manufacturing employment

Trading conditions


Technical change

Import/domestic prefs

Consumer preferences

Macro composition of GDP

Rates of return

-5 -4 -3 -2 -1 0

Per cent

a Excludes the impact of growth in employment and the number of households. Source: Industry Commission estimates based on results of the Monash model.