PR002004

AUSTRALIAN INDUSTRIAL RELATIONS COMMISSION

Workplace Relations Act 1996

s.113 applications for variations

s.108 references to Full Bench

Australian Liquor, Hospitality and Miscellaneous Workers Union

THE HOSPITALITY INDUSTRY-ACCOMMODATION, HOTELS, RESORTS AND GAMING AWARD 1998

(ODN C No. 389 of 1975)

[AW783479  Print P9138]

(C2003/6455)

BUILDING SERVICES (VICTORIA) AWARD 2003

(ODN C No. 21726 of 1992)

[AW822844  PR929372]

(C2003/6456)

CHILD CARE INDUSTRY (AUSTRALIAN CAPITAL TERRITORY) AWARD, 1998

(ODN C No. 3697 of 1985)

[AW772250CRA  Print Q2724]

(C2003/6457)

LAUNDRY INDUSTRY (VICTORIA) AWARD 1998

(ODN C No. 21626 of 1992)

[AW787052  Print Q6887]

(C2003/6458)

The Australian Workers' Union

HORSE TRAINING INDUSTRY AWARD 1998

(ODN C No. 3039 of 1975)

[AW783476  Print R5058]

(C2003/2845)

Textile, Clothing and Footwear Union of Australia and another

CLOTHING TRADES AWARD 1999

(ODN C No. 696 of 1980)

[AW772144CRA  Print S1147]

(C2003/2508)

Automotive, Food, Metals, Engineering, Printing and Kindred Industries Union

VEHICLE INDUSTRY-REPAIR, SERVICES AND RETAIL-AWARD 2002

(ODN C No. 1339 of 1974)

[AW801827CRA  PR931545]

(C2003/6563)

VEHICLE INDUSTRY AWARD 2000

(ODN C No. 1522 of 1979)

[AW801818  Print T3920]

(C2003/6570)

METAL, ENGINEERING AND ASSOCIATED INDUSTRIES AWARD, 1998-PART I

(ODN C No. 2568 of 1984)

[AW789529  Print Q2527]

(C2003/6741)

GRAPHIC ARTS-GENERAL-AWARD 2000

(ODN C No. 22956 of 1995)

[AW782505CAN  Print S1716]

(C2003/6742)

National Union of Workers

GROCERY PRODUCTS MANUFACTURE-MANUFACTURING GROCERS AWARD 2003

(ODN C No. 1152 of 1985)

[AW820730  PR926383]

(C2003/6230)

COMMERCIAL SALES (VICTORIA) AWARD 1999

(ODN C No. 31107 of 1993)

[AW772623  Print R1382]

(C2003/6231)

RUBBER, PLASTIC AND CABLE MAKING INDUSTRY-GENERAL-AWARD 1998

(ODN C No. 1800 of 1982)

[AW794720  Print R4420]

(C2003/6232)

STORAGE SERVICES-GENERAL-AWARD 1999

(ODN C No. 32518 of 1992)

[AW796791  Print R1040]

(C2003/6233)

Australian Municipal, Administrative, Clerical and Services Union

VICTORIAN LOCAL AUTHORITIES AWARD 2001

(ODN C No. 36277 of 1989)

[AW811556  PR910776]

(C2003/6318)

CLERICAL AND ADMINISTRATIVE EMPLOYEES (VICTORIA) AWARD 1999

(ODN C No. 34749 of 1995)

[AW773032  Print S1367]

(C2003/6320)

Transport Workers' Union of Australia

TRANSPORT WORKERS AWARD 1998

(ODN C No. 1520 of 1982)

[AW799474CRN  Print Q8693]

(C2003/6348)

Shop, Distributive and Allied Employees Association

RETAIL AND WHOLESALE INDUSTRY-SHOP EMPLOYEES-AUSTRALIAN CAPITAL TERRITORY-AWARD 2000

(ODN C No. 30030 of 1993)

[AW794740CRA  Print T3309]

(C2003/6321)

Construction, Forestry, Mining and Energy Union

TIMBER AND ALLIED INDUSTRIES AWARD 1999

(ODN C No. 31 of 1950)

[AW800937  Print R5055]

(C2003/2846)

Health Services Union of Australia

HEALTH SERVICES UNION OF AUSTRALIA (ABORIGINAL AND TORRES STRAIT ISLANDER HEALTH SERVICES) AWARD 2002

(ODN C No. 32659 of 1990)

[AW819920  PR924564]

(C2003/6527)

Various industries
JUSTICE GIUDICE, PRESIDENT
VICE PRESIDENT ROSS
VICE PRESIDENT LAWLER
SENIOR DEPUTY PRESIDENT MARSH
SENIOR DEPUTY PRESIDENT LACY
COMMISSIONER LARKIN
COMMISSIONER GRAINGER	MELBOURNE, 5 MAY 2004

CONTENTS

LIST OF MAIN ABBREVIATIONS

THE ACTU'S CLAIM

OUTLINE OF RESPONSES TO THE CLAIM

Employers

Governments

Other Interveners

THE ECONOMY

Overview of Economic Submissions

Indicators of Past Economic Activity

Economic Outlook

COST OF THE CLAIM

WAGES MOVEMENTS

AWARD EMPLOYEES AND AWARD INDUSTRIES

Output

Profits

Employment

Productivity

General Conclusion

ECONOMIC EFFECTS

Academic Research

Surveys

ACCI Analysis of Safety Net Adjustments on Inflation

THE NEEDS OF THE LOW PAID AND LIVING STANDARDS GENERALLY

Senate Poverty Inquiry Report

Social Policy Research Centre Budget Standards

Other Issues

DEAC AND THE SUPPORTED WAGE

CONCLUSION ON THE ACTU'S CLAIM

STATEMENT OF PRINCIPLES AND RELATED MATTERS

ADDITIONAL MATTERS

ORDERS

ATTACHMENT A - STATEMENT OF PRINCIPLES

Charts

Chart 1: Past Negative Factors

Chart 2: Consumer Price Index-All Groups CPI, Tradable and Non-Tradable Components

Chart 3: Annual Wages and Earnings Growth-Main Indicators September 1998-September 2003

Chart 5: Movements in Profit and Wages Shares of Total Factor Income Including Gross Mixed Income

Chart 6: Output Growth-Award Dependent Industries 2002-03

Chart 7: Employment Growth 1996-2003 (Selected Industries)

Chart 8: Change in Numbers Employed Under Agreements and Awards by Industry May 2000-May 2002

Chart 9: Real After-Tax Wages

Tables

Table 1: Key Economic Indicators Comparison

Table 2: Summary of Outlook for the States and Territories for 2003-04

Table 3: Economic Growth-Real Gross Domestic Product Growth (Trend Estimates)

Table 4: Contributions to Growth in Gross Domestic Product

Table 5: Private Consumption

Table 6: Private Investment

Table 7: Exports and Imports (Trend)

Table 8: Manufacturing Production

Table 9: Net Balance of Firms Reporting Improvements in Selected Economic Indicators

Table 10: Consumer Price Index

Table 11: Measures of Wage Movements

Table 12: Comparisons of Various Wage Movements

Table 13: Employment and Unemployment

Table 14: Employed Persons by Industry-Trend Data

Table 15: GDP Per Hour Worked-Market Sector

Table 16: Company Profits

Table 17: MYEFO Domestic Economic Forecasts(a)

Table 18: Wages Measures

Table 19: Proportions of Award-Reliant Employees by Industry

Table 20: Differences between Minimum Wages Report and EEH Survey

Table 21: Relative Cost of Goods and Services in Rural Australia

Table 22: Effect of 2003 Safety Net Adjustment on Respondents' Business

Table 23: Budget Standards for Various Household Types

LIST OF MAIN ABBREVIATIONS

In this decision the following abbreviations are used:

AACS:	Award and Agreement Coverage Survey 1999
AAWI:	Average Annualised Wage Increases
ABARE:	Australian Bureau of Agricultural and Resource Economics
ABS:	Australian Bureau of Statistics
ACCER:	Australian Catholic Commission for Employment Relations
ACCI:	The Australian Chamber of Commerce and Industry
ACIRRT:	Australian Council for Industrial Relations Research and Training
ACOSS:	Australian Council of Social Service
Act:	Workplace Relations Act 1996
ACTU:	Australian Council of Trade Unions
AENA:	Average Earnings on a National Accounts Basis
AiG:	The Australian Industry Group and the Engineering Employers Association, South Australia
AWE:	Average Weekly Total Earnings
AWOTE:	Average Weekly Ordinary Time Earnings
CPI:	Consumer Price Index
DEAC:	Disability Employment Action Centre and the National Council on Intellectual Disability
DEWR:	Department of Employment and Workplace Relations
EEH Survey:	ABS Survey of Employee Earnings and Hours, Australia
GDP:	Gross Domestic Product
GOS:	Gross Operating Surplus
GSP:	Gross State Product
Lewis Report:	A Report on the Effect of Raising Minimum Wages on Rural Businesses
May 2000 decision:	Safety Net Review-Wages May 2000 decision [Print S5000; (2000) 95 IR 64]
May 2002 decision:	Safety Net Review-Wages May 2002 decision [PR002002; (2002) 112 IR 411]
May 2003 decision:	Safety Net Review-Wages May 2003 decision [PR002003; (2003) 121 IR 367]
Metal Industries Award:	Metal, Engineering and Associated Industries Award, 1998-Part I [AW789529]
Minimum Wages Report:	Minimum Wages in Australia: An Analysis of the Impact on Small and Medium Sized Businesses
MPMSAA:	The Master Plumbers' and Mechanical Services Association of Australia
MYEFO:	Mid-Year Economic and Fiscal Outlook
National Motor Industry:	Victorian Automobile Chamber of Commerce, the Motor Traders' Association of New South Wales, the Motor Trades Associations of South Australia, Queensland, the Australian Capital Territory, the Northern Territory and Western Australia, and the Service Station Association of New South Wales
NFF:	National Farmers' Federation
NMI:	National Motor Industry
OECD:	Organisation for Economic Co-operation and Development
RBA:	Reserve Bank of Australia
SARS:	Severe Acute Respiratory Syndrome
SPRC:	Social Policy Research Centre, University of New South Wales
States and Territories:	States of New South Wales, Queensland, South Australia, Tasmania, Victoria, Western Australia and the Australian Capital Territory and Northern Territory
WCI:	Wage Cost Index

DECISION

THE ACTU'S CLAIM

[1] The Australian Council of Trade Unions' claim for 2004, like those of the past five years, is for a flat dollar adjustment to the safety net award rates of pay. The amount of the claim is $26.60, with commensurate adjustments in wage-related allowances. Underpinning the claim are a number of union applications pursuant to s.113 of the Workplace Relations Act 1996 (the Act) to vary certain awards. The applications were filed in November and December 2003.

[2] In support of its claim the Australian Council of Trade Unions (the ACTU) argued that, in the context of strong job growth, moderate growth in prices and wages and a growing economy, the amount of the increase is fair, both to employees and employers alike. More particularly, it was said that granting the claim will:

• have a negligible impact on aggregate earnings of 0.1 per cent;
• provide an average increase for full-time award workers of 4.5 per cent and an average increase for all award workers of 4.7 per cent;
• result in an average increase for award workers in the five-year period 2000 to 2004 (inclusive) of 3.4 per cent, the same as the average annual increase in the Wage Cost Index (WCI) for the period June 1999 to June 2003; and
• provide a real, but modest increase, which will help restore the relative value of after-tax wage income for the lowest paid award workers whose real after-tax wages have been essentially stagnant since 1999.

OUTLINE OF RESPONSES TO THE CLAIM

Employers

[3] With the exception of the National Farmers' Federation (NFF), each employer association, in opposing the ACTU's claim, conceded that a moderate increase in award rates of pay is appropriate. The Australian Chamber of Commerce and Industry (ACCI) identified a moderate increase as an amount of up to $10 per week in award rates up to and including the tradesperson classification - C10 ($542.20 per week) in the Metal, Engineering and Associated Industries Award, 1998-Part I¹ (the Metal Industries Award). The Australian Industry Group and the Engineering Employers Association, South Australia (jointly AiG) supported an increase of $10 per week in all award rates.

[4] The essence of the ACCI position is that the ACTU's claim ignores its economic impact, the plight of those without jobs and the liability of the employers for on-costs associated with wage increases. In support of its position ACCI contended that wages have been rising as a proportion of Gross Domestic Product (GDP) and profitability has been falling. According to ACCI, increases in interest rates and the value of the dollar are causing a contraction in manufacturing, while the rural sector is underperforming as a legacy of the remaining pockets of drought affected areas and the rising value of the dollar.

[5] In the alternative ACCI contended that if, contrary to its support for a capped increase of $10 per week, the Commission ordered an increase in all award rates and determined that a tiered increase was appropriate, then the cut-off point for the higher amount of the tier should be somewhat lower than the C2(a) rate in the Metal Industries Award. ACCI proposed also an alteration to Principle 8(a), the effect of which would be to defer for 28 days the operative date of a variation order to give effect to this decision.

[6] AiG's picture of the economic landscape is similar to that presented by ACCI. The ACTU's claim, it contended, is not sustainable because of weakening productivity growth, falling profits and deteriorating business competitiveness. AiG proposed a $10 per week increase in all award rates of pay, which, it said, would reflect the competitive pressures that industry faces in 2004 and accommodate the needs of the low paid.

[7] AiG proposed also that, as a matter of principle, unions be required to commit to continuous improvement in productivity, efficiency and flexibility as a condition precedent to the granting of an order in an application to give effect to this decision.

[8] NFF contended that the ACTU's claim is not justified. Its contention was made on the grounds that there is still a degree of uncertainty in the farm sector and, with the rising value of the dollar and a downturn in commodity prices, a generally gloomy forecast for the sector. NFF supported the ACCI proposal for an amendment to Principle 8(a).

[9] The Master Plumbers' and Mechanical Services Association of Australia (MPMSAA) expressed support for ACCI's submission and its position on an adjustment to minimum rates of pay.

[10] The Victorian Automobile Chamber of Commerce, the Motor Traders' Association of New South Wales, the Motor Trades Associations of South Australia, Queensland, the Australian Capital Territory, the Northern Territory and Western Australia and the Service Station Association of New South Wales (jointly the National Motor Industry) contended that the ACTU's claim is excessive. Its position was that the National Motor Industry (NMI) is comprised predominantly of small businesses and any increase in the safety net should be moderate. NMI said it would support an increase of $10 per week.

Governments

[11] The Commonwealth opposed the ACTU's claim on the grounds of its magnitude and an anticipated adverse economic impact. However, the Commonwealth did not oppose an increase of up to $10 per week in award rates up to and including the tradesperson classification - C10 ($542.20) in the Metal Industries Award.

[12] The States of New South Wales, Queensland, South Australia, Tasmania, Victoria and Western Australia and the Australian Capital Territory and Northern Territory (jointly the States and Territories) supported an increase of $20 per week in all award rates of pay.

Other Interveners

[13] As in past safety net cases a number of community-based organisations made submissions in support of a safety net adjustment.

[14] The Australian Council of Social Service (ACOSS) supported substantial increases in minimum wages over time and contended that, in the short-term, the Commission should increase minimum wages to prevent them falling any further behind movements in average wage rates. ACOSS did not nominate an increase of any particular amount.

[15] The Australian Catholic Commission for Employment Relations (ACCER) renewed its 2003 claim that the Commission should set a benchmark or appropriate guidelines within which needs could be established. While supporting an increase of $26.60 per week in the federal minimum wage, an amount that it regarded as insufficient to provide a fair minimum rate of pay for award-only employees, ACCER said:

"[I]t does not support a uniform increase across all classifications because it believes that the increases should be directed to those who are most in need. The primary beneficiaries should be those at the lower paid classifications."

[16] The Disability Employment Action Centre, a Commonwealth funded community legal centre, and the National Council on Intellectual Disability, a national body representing people with disabilities, (jointly DEAC) supported the ACTU's claim. DEAC also urged the Commission to introduce measures to secure the inclusion of the supported wage clause in all awards, to eliminate discrimination in awards and agreements and to establish a safety net of fair minimum wages and conditions for workers with disabilities.

THE ECONOMY

Overview of Economic Submissions

The ACTU

[17] The ACTU submitted that the economic data point to a strong and robust economy with the near and medium-term outlook continuing to be positive. It further submitted that the economy is stronger than last year and predicted that the situation would continue to improve. The strength of the economy has occurred in an environment characterised by an abatement of factors such as SARS and the drought. In particular, drawing on national accounts and other statistical data, the ACTU submitted that:

• growth in domestic consumption and a strong rural sector contributed to strong growth (GDP) of 1.4 per cent seasonally adjusted (1.1 per cent trend) for the last quarter and a very solid 4.0 per cent seasonally adjusted (3.5 per cent trend) for the year. Non-farm GDP grew by 0.8 per cent for the December quarter to be 3.0 per cent higher over the year;
• company profits have continued to grow strongly reaching a 27.5 per cent increase over the year to the December quarter 2003. The Gross Operating Surplus (GOS) measure of company profits continued to grow sharply despite the slowdown in economic growth in the first half of 2003;
• inflationary pressures remained within the Reserve Bank of Australia's (RBA) target range for the medium-term. The all groups consumer price index (CPI) for the year ending December 2003 at 2.4 per cent was lower than that recorded for the same period in the previous year;
• there is evidence to support the forecast that the downward trend in inflation will continue before rising again in 2005 (RBA Statement on Monetary Policy);
• productivity growth has increased over the 12 months to the December quarter 2003, whether measured by GDP per hour worked throughout the economy (2.3 per cent) or GDP per hour worked in the market sector (3.3 per cent);
• the labour market remains strong with the total number of employees increasing by 1.8 per cent to January 2004; over 90 per cent of new employment has been in full-time jobs. Employment growth remained strong in the latest monthly data for January 2004. The unemployment rate has remained below 6.0 per cent for the last six months, reducing to 5.6 per cent for January 2004;
• wages growth has remained moderate with the WCI increasing 3.6 per cent for the year to December 2003. This is broadly consistent with previous yearly outcomes; and
• the external sector recorded an increase in the current account deficit largely as a result of the fall in exports which was a consequence of the drought and the appreciation of the Australian dollar. Australia also recorded a strong trade deficit in November 2003.

[18] The ACTU relied on the Mid-Year Economic and Fiscal Outlook (MYEFO) mid-year forecast to demonstrate a positive picture of the Australian economy over the foreseeable future. The drought has broken and there has been a marked improvement in the global economy which should boost Australia's economy and further alleviate some of the difficulties that have been incurred by exporters (due to the rising dollar) through an increase in demand for Australian exports. It relied on the Treasurer's comments following the release of the MYEFO forecasts set out in the review of the economic outlook later in this decision.²

[19] It also relied on the Prime Minister's statement:

"The economy remains powerful and robust. There has been a tiny rise in the unemployment level, but there has been some reclassification carried out by the ABS. But it remains under six per cent. It has been so since September. We have a very strong employment outlook, fuelled and supported by a very strongly growing economy."³

[20] The ACTU in reply said:

"Economic data released since the ACTU filed its original submissions confirms the ACTU's analysis of Australia's recent economic performance. The December quarter National Accounts show GDP growth in that quarter of 1.4 per cent (seasonally adjusted), giving Australia's economy an annualised growth rate in the last six months in excess of 5 per cent. Prices and wages growth have continued to be moderate with little change in either the CPI or the Wage Cost Index from the September quarter figures referred to in our original submissions. CPI is currently at 2.4 per cent and the most recent Wage Cost Index is 3.6 per cent in trend terms (3.7 per cent seasonally adjusted). Labour market conditions continue to be strong, unemployment remains at its lowest level for 22 years."

[21] Table 1 sets out the comparison of key economic indicators for the past two years, together with forecasts for 2003-04.

Table 1: Key Economic Indicators Comparison

Notes:

(a) In the second and third columns trend figures have been used (except for CPI and GDP (Seasonally Adjusted)). GDP, CPI and WCI are year to December 2002 and 2003 respectively.

Labour Force statistics are year to January 2003 and 2004 respectively.

[Source: Exhibit ACTU 4, Table R1.1 at para R1.6.]

The States and Territories

[22] The States and Territories submitted that the Australian economy continues to perform well and that the outlook is generally positive. They submitted that since the Safety Net Review-Wages May 2003 decision⁴ (the May 2003 decision) was handed down:

• a number of potential threats to the Australian economy (including the drought, the war in Iraq and the advent of SARS) have eased; and
• the Australian economy continues to improve - this is demonstrated by a range of economic forecasts.

[23] The States and Territories relied on and supported the ACTU's submissions. In summary, the States and Territories submitted that:

• expectations for economic growth range from 3.5 per cent to 4 per cent in 2003-04, an improvement in the performance from the previous year;
• the economy is estimated to grow by 3.5 per cent in 2004-05, with growth expected to come from exports, easing but solid consumption and business investment. Housing investment is expected to detract from growth;
• net exports are forecast to detract 1.5 percentage points from GDP growth in 2003-04 which is an improvement on the 3 per cent fall in net exports in the previous year. Rural exports are expected to recover and tourism to pick up. Offsetting these factors is the appreciation in the Australian dollar which is likely to stimulate import demand. Net export demand is forecast to rise in 2004-05 but not expected to add to economic growth;
• the recovery in farm production expected in 2003-04 of 27 per cent is expected to continue into 2004-05;
• employment and unemployment are expected to remain relatively stable;
• wage pressures are expected to remain moderate, balanced by increases in productivity;
• headline inflation for 2004-05 is expected to grow by around 2 per cent (year average). Although capacity utilisation is high, capacity constraints may not emerge in the coming year, although in New South Wales some employers are reporting labour shortages;
• the CPI may fall by mid 2004 but drift back by December 2004 and by mid 2005 be subject to upward pressure; and
• forecasts for world growth are strong, largely due to a United States-led recovery.

[24] Risks for 2004 were identified as:

• slowdown in dwelling construction;
• higher interest rates;
• delay in major recovery of farm output; and
• re-emergence of major foreign security risks.

[25] Table 2 sets out the predictions for major economic indicators on a state and territory basis:

Table 2: Summary of Outlook for the States and Territories for 2003-04

Notes:

(a) Gross State Product.

(b) Wage Cost Index.

(c) Not comparable with ABS statistics.

[Source: Derived from written submissions, see Exhibits G1 at paras 48, 59, 67, 73, 81, 88 and G2 at para 11.]

ACCI

[26] ACCI submitted that the Australian economy will continue to grow in 2003-04 and that economic conditions will be reasonable. However, economic growth will not be as strong as over the last 12 months because the momentum for sustained economic growth has dissipated, especially having regard to data on expectations of investment growth. The strong growth which the Australian economy has recently recorded has been fuelled by consumption expenditure and farm income. Given the negative household savings ratio, high increases in consumer demand cannot be sustained, nor can farm income which is volatile. ACCI also submitted that dwelling investment and new private business investment will grow less slowly than in the past year and that on the basis of actual and expected capital expenditure, there may be an actual contraction in investment relative to the level experienced in the current financial year.

[27] ACCI submitted that, whilst investment indicators were generally performing well, a number of circumstances impacting on investment indicators raise the question of whether the recent performances will be sustainable. These include:

• gross fixed capital formation;
• new non-dwelling construction;
• investment in machinery and equipment; and
• new private business investment.

[28] ACCI also submitted that the labour market was performing very well, particularly the growth in full-time employment, and the reduction in the unemployment rate to 5.6 per cent was a tremendous improvement to be below 6 per cent. However, ACCI warned that the rate of underemployment introduces a degree of concern when considering labour market outcomes.

[29] ACCI, whilst acknowledging that the growth in the inflation rate as measured by the CPI had declined, nevertheless cautioned that this rate of increase in the CPI could not be sustained at such moderate levels. This is because the contribution to that moderation by the appreciation of the Australian dollar may be reversed in the coming quarters. The effect of the rising dollar on dampening price increases is evident from a comparison of the rise in prices in the tradable and non-tradable goods sectors. The former, being exposed to international competition, has recorded zero price increases whilst the latter, with no such competition, recorded a 4.4 per cent increase in the year to the December quarter 2003.

[30] ACCI submitted that the RBA warned that the CPI may react in a volatile way with further falls followed by rises to an annual rate of around 2.5 per cent. The concern here is that with the deflationary impact of the dollar out of the picture, increased inflationary pressures may emerge as the price of imports increases. The RBA target rate of 2 to 3 per cent may be exceeded.

[31] ACCI submitted that exports have not grown in real terms since December 2003 largely due to the drought conditions in Australia. With imports rising there is a problem with international competitiveness since the gap between exports and imports cannot continue indefinitely.

[32] ACCI further submitted that the impact of the safety net adjustment decision is reflected in the WCI, which has spiked in the September quarter each year since its publication. This is evidence that in an aggregate sense the safety net adjustments have an economic impact. ACCI submitted that the real WCI demonstrates that since 1997 the level of wages has grown by 1 per cent a year or 5.9 per cent for the total period.

[33] ACCI acknowledged that labour productivity over the past year of 2.4 per cent has been quite healthy. ACCI's preferred measure of market sector productivity has grown by 3.3 per cent compared with a longer run average of 2.7 per cent. However, hours worked in the market sector have recorded very low growth since December 1999.

[34] ACCI recognised the change in rural conditions with the easing of the drought but submitted there are still weaknesses within the farm sector which are affecting the economy as a whole.

[35] On the basis of the latest Business Barometer, ACCI concluded that the economy is showing more strength than last year but there is no major acceleration taking place.

AiG

[36] AiG submitted that in the past 12 months international competition has intensified such that Australia has experienced pressures arising out of high levels of imports, weaker exports and deteriorating business competition. The December quarter national accounts record no growth in the manufacturing sector in 2003 and a complete absence of export growth in manufacturing exports which have experienced their worst downturn in 40 years.

[37] Moreover, there has been a significant decline in manufacturing employment of 70 000 jobs. Profit growth has been non-existent since June 2003.

[38] The pressures on the manufacturing sector are described as:

• manufacturing production has gone into reverse;
• the sale price for Australian produced manufacturing goods has gone backwards;
• for the first time in many years the manufacturing sector has experienced a complete absence in exports growth; and
• exports have fallen.

[39] AiG submitted that the key factor relevant to an assessment of Australia's economic performance and outlook is the significant decline in business competitiveness. This has resulted from the sharp appreciation of the Australian dollar, the impact of global capacity and China's emerging industrial strength, together with an environment of increasing interest rates. These factors "are encouraging the trifecta of higher import competition, weaker export competitiveness and deteriorating business competitiveness".⁵

[40] AiG emphasised the speed with which the Australian dollar has appreciated, particularly over the past 12 months (33 per cent) and the corresponding increase in the Australian trade weighted index, a key indicator of international competitiveness which has increased by 23 per cent. It identified major consequences of the higher Australian dollar - reduced earnings from exports and weakened competitiveness in key markets such as the United States; intensification of competition within the Australian market by making imports cheaper and more attractive to both business and the consumer; and import prices falling dramatically putting a hold on selling prices.

[41] In AiG's view international competition has resulted in an increasing number of exporters reporting an intention to move employment offshore - this trend is particularly apparent in the textile, clothing and footwear sector.

[42] AiG further submitted that the glut of excess capacity in global manufacturing was compounded by the rapid emergence of the Chinese economy which is characterised by manufacturing and construction. Over the past year the intensity of competition from China has accelerated due in large part, but not limited to, the appreciation of the Australian dollar.

[43] AiG submitted that a tough trading environment exists for manufacturing firms. It relied on the latest AiG quarterly survey of manufacturing to confirm that manufacturing activity has continued to ease, companies are struggling to lift profitability and there is slower production growth with signs of a softening in domestic demand. In short, the export and manufacturing sectors are under quite serious pressure:

"The data including the quarterly survey and the national accounts, demonstrates the point we have been making, namely that the export and manufacturing sectors are under quite serious pressure."⁶

[44] AiG shared ACCI's concern for labour market outcomes by highlighting the level of underemployment.

NMI

[45] NMI submitted that the economy was growing strongly:

"[I]t is true that there have been record new vehicle sales in 2003, and good economic conditions, and that generally many observers would assume that the retail motor industry businesses have enjoyed several years of high profitability."⁷

[46] It submitted however that the Commission should take into account the economic circumstances confronting small business, in particular the national motor industry. Despite record new vehicle sales in 2003 and good economic conditions the motor vehicle retail and services industry has recorded low profitability and low profit margins. In particular, profit margins have fallen for two consecutive quarters to stand at 1.9 per cent compared with 6.1 per cent for the all services industries within the Australian economy. In a joint AC Neilson Survey (November 2003) respondents recorded that almost half of retail motor industry businesses halved their business profitability with just over a half reporting poor or very poor profitability.

[47] NMI expressed concern over the Commission granting the ACTU's monetary claim arguing that the industry would become less efficient and less capable of responding to competition with more businesses being at risk of closure. The situation would be exacerbated in Victoria due to proposed changes in workers' compensation premium rates.

[48] Other negative factors facing the industry include the risks in rural Australia such as the ongoing severity of the drought, especially in New South Wales and Queensland, and the rising Australian dollar resulting in lower incomes in the retail sector. Further pressure is being put on the industry by skills shortages as competition pushes up wage rates. Sections of the industry, including farm machinery, service stations, body repairers and automobile repairers are facing specific challenges.

NFF

[49] NFF submitted that the Commission should have particular regard to the farming sector and a less "metro centric" focus on the effects of the safety net adjustment. It submitted that uncertainty still exists for the farming sector in Australia.

[50] Despite a rebound in the farming sector, drought-declared areas in New South Wales and Queensland are extensive and have reappeared. The Australian Bureau of Agricultural and Resource Economics (ABARE) has predicted a tough outlook for agriculture in 2004-05, including falling farm incomes and farm output and prices received by farmers.

[51] NFF acknowledged that there has been an improvement from last year, particularly in the cropping sector, but the rebound has not been as strong as was hoped. Negative impacts are the appreciation of the Australian dollar and a downturn in commodity prices.

[52] NFF's submission stressed the impact of a "cash drought" which occurs during a drought recovery phase. The longer-term impact of the drought is still to be felt.

[53] NFF submitted that the appreciation of the Australian dollar has had a negative impact on the rural economy. It calculated that every 1 per cent appreciation in the Australian dollar against the United States dollar cuts annual farm increases by $115m.

[54] NFF highlighted the impact of the drought on employment levels which fell from 443 000 people employed in agriculture in the November quarter 2001, compared with 369 000 in the November quarter 2002 and 376 000 in the November quarter 2003.

The Commonwealth

[55] The Commonwealth submitted that the economy is forecast to grow by 3¾ per cent in 2003-04. It submitted that:

"2.1 . . . The positive outlook is underpinned by expected robust growth in domestic demand and a stronger performance from the external sector while non-farm GDP is expected to grow solidly."

and further submitted that:

"2.2 Employment growth should remain moderate with the unemployment rate remaining steady. The forecast for year-ended inflation in 2003-04 is 2¼ per cent, reflecting the appreciation of the Australian dollar and continued moderate wages growth.

2.3 Near-term risks around the outlook have diminished over the course of 2003 and are now more evenly balanced. However, some medium-term risks around both the domestic and international outlooks have become more pronounced." ⁸

[56] In relation to the domestic forecast, the Commonwealth submitted that recent economic releases are consistent with MYEFO forecasts. Economic and employment growth have recovered since the final half of 2003 with over 109 000 jobs created in the six months to February 2004.

[57] Employment growth is forecast to grow around 1½ per cent in 2003-04 and 1¾ per cent in 2004-05. The unemployment rate is expected to remain at around, or a little below, 5¾ per cent over the forecast horizon.

[58] Inflation is expected to decline to around, or possibly below, the bottom of the target band of 2 to 3 per cent through the forecast horizon in line with modest growth in wages and solid productivity growth.

[59] Australia's low inflation rate is due in part to a decline in the price of imported goods as a result of the appreciation in the Australian dollar. The price of domestic goods and services has grown slightly over the year (due to cost pressures in some service industries, as well as increases in construction costs).

[60] In relation to the external economy, the pace of the global economic recovery has picked up consistent with MYEFO forecasts. However, a downside risk remains for the medium-term with the global recovery still over-reliant on US growth.

[61] The risks around the near-term outlook are evenly balanced. However, a number of issues weigh on prospects for stronger domestic growth and more broad-based investment growth over the medium-term, including:

• how the current housing cycle plays out;
• the trading conditions of some exporters and firms in import-competing industries being affected by the appreciation in the Australian dollar;
• failure of the rural economy to generate forecast increased farm production;
• over-dependence on the US for global recovery; and
• the high global price of oil.

[62] The Commonwealth submitted that the downside risks should be balanced in the short and medium-term by upside risks:

• the housing market may hold up longer than expected;
• ongoing momentum of domestic demand could produce stronger than anticipated growth; and
• global imbalances could unwind in an orderly fashion and global recovery could be stronger and more widespread than forecast.

Indicators of Past Economic Activity

Economic Growth

[63] On a quarterly basis the annual growth in GDP has improved from 2.9 per cent in September 2003 to 3.5 per cent in December 2003. This contrasts with 3.1 per cent annual increase in the year ended December 2002. The less reliable quarterly data show an increase for the December quarter 2003 of 1.1 per cent, the same as recorded for the September quarter 2003 and over double the 0.5 per cent increase in December 2002.

[64] Non-farm GDP recorded a 3 per cent annual increase to the December quarter 2003, lower than the annual growth figures recorded for each of the quarters since September 2001 and slightly lower than 3.7 per cent for the period March 1996 to December 2003. The quarterly figures demonstrate that the December quarter 2003 growth of 0.8 per cent was consistent with recorded increases over the past four quarters.

[65] Farm GDP grew by 21.1 per cent in the year ended December 2003. This contrasts with -2.6 per cent for the year ended September 2003 and -27 per cent for the annual figure to December 2002. The December quarter 2003 data recorded an 8.8 per cent increase building on strong growth which has occurred since June 2003.

Table 3: Economic Growth-Real Gross Domestic Product Growth (Trend Estimates)

Notes: Reference year for chain value measures is 2001-02. All figures are trend estimates.

[Source: ABS Cat. No. 5206.0.]

Contribution to Growth

[66] The latest data show that total final consumption expenditure was a strong contributor to growth. The annual figures to the December quarter 2003 show a 3.8 percentage point contribution to GDP total growth of 3.5 percentage points. Whilst total gross capital formation contributed 2.2 percentage points, this was overwhelmingly from the private sector, particularly machinery and equipment. The strong consumption and investment growth was offset by a 3.3 percentage point subtraction in net exports, primarily reflecting stronger imports and lower exports.

Table 4: Contributions to Growth in Gross Domestic Product

Notes: Reference year for chain volume measures is 2001-02. All figures are trend estimates.

[Source: ABS Cat. No. 5206.0.]

Consumption

[67] Private consumption expenditure increased by 5.4 per cent for the year ending the December quarter 2003 and the corresponding increase for the December quarter 2003 was 1.6 per cent. The annual growth in consumption expenditure is the largest recorded in the period set out in the table. This robust growth continues a trend of strong growth recorded throughout 2002 and 2003. The quarterly increase for December 2003 is the fourth consecutive increase and stands at twice the March 2003 increase.

Table 5: Private Consumption

[Source: ABS Cat. No. 5206.0.]

Investment

[68] Growth in each component of investment remained solid during the December quarter 2003, notwithstanding very high rates of growth recorded in previous quarters particularly throughout 2002. The easing of quarterly growth is reflected in the annual figures to the December quarter 2003. The annual growth rates reflect strong growth across the board with the exception of dwelling investment with the anticipated cooling off of the housing market from the very strong growth previously recorded.

[69] Australian Bureau of Statistics (ABS) estimates of private new capital investment indicate that investment expectations are lower with a fall-off in investment being anticipated on the latest estimates for 2003-04. Early estimates for 2004-05 indicate that investment may actually fall.

Table 6: Private Investment

[Source: ABS Cat. No. 5206.0.]

Exports and Imports

[70] The value of exports in real terms remains stagnant but the latest quarterly data show an increase of 0.7 per cent, which reverses the negative growth recorded on a quarterly basis since the December quarter 2002, whilst the annual data on exports for the year ending December quarter 2003 was -2.2 per cent.

[71] Import growth is very strong, recording an increase in the December quarter 2003 of 2.9 per cent, which is the same order of magnitude as has occurred on a quarterly basis since late 2001. The import growth for the year to December 2003 is a high 12.0 per cent which reflects the continuation of the reversal in import growth since the start of calendar year 2002.

Table 7: Exports and Imports (Trend)

[Source: ABS Cat. No. 5206.0.]

Manufacturing and Export Sector

[72] ABS manufacturing data show that whilst manufacturing production increased by 2.3 per cent in the 2002-03 financial year, growth in the year to the September and December 2003 quarters recorded 0.2 per cent and -1.3 per cent respectively. There have been three consecutive quarters of negative growth.

Table 8: Manufacturing Production

Notes: Reference year for chain volume measures is 2001-02. All figures are trend estimates.

[Source: ABS Cat. No. 5206.0.]

[73] The March quarter 2004 Survey of Australian Manufacturing⁹ showed a reduction in the number of firms experiencing an increase in production and employment. The survey indicated that "the weakness in profit growth remains largely due to the strength of the exchange rate, which is contributing to weakness in exports, encouraging higher import penetration and intensifying the squeeze on margins".¹⁰ The low net balance of firms experiencing an increase in selling prices records the fourth consecutive quarter of easing.

[74] The outlook is somewhat rosier when regard is had to forecasts for the June quarter 2004 with the net balance of firms expecting a lift in production and new orders and sales the highest since mid 2002. Exports are forecast to strengthen and profits to weaken. Investment is forecast to increase both in the period immediately ahead and in the next twelve months. Employment is forecast to remain steady in the June quarter 2004.

Table 9: Net Balance of Firms Reporting Improvements in Selected Economic Indicators

[Source: AiG & PricewaterhouseCoopers, Survey of Australian Manufacturing at p. 12.]

[75] A new series known as the DHL Export Barometer is a survey of Australia's leading exporters aimed at providing insight into the factors impacting on Australia's export trade. The impact of the appreciation of the Australian dollar was one of the major negative factors identified by exporters. In the latest quarter and over the past 12 months, 52 per cent of exporters recorded the exchange rate as a negative factor. This factor broadly equates with the proportion of exporters viewing economic/political conditions abroad as a negative factor. At the other end of the scale very few reported SARS and the drought as negatives, whilst 12 per cent considered there were no negative factors.

Chart 1: Past Negative Factors

[Source: DHL Export Barometer November 2003-April 2004.]

[76] Looking to the future, two-thirds of exporters expect an increase in profitability over the next 12 months, whereas 31 per cent indicated the exchange rate would affect output and 28 per cent indicated that their future investment and business plans would be adversely affected.

Prices

[77] Inflation, as measured by the CPI, increased for the December quarter 2003 by 0.5 per cent which fed into the annual increase of 2.4 per cent. This is the third consecutive fall in the CPI annual rate of increase. This is well within the RBA target range for inflation of 2 to 3 per cent.

Table 10: Consumer Price Index

[Source: ABS Cat. No. 6401.0.]

[78] The tradable component of the CPI has been stable over the 12 months to December 2003 except for falls in the index in the June and September quarters. This outcome reflects the impact of the appreciation of the exchange rate. In contrast, the non-tradable component of the CPI has increased by 4.4 per cent on an annual basis.

Chart 2: Consumer Price Index-All Groups CPI, Tradable and Non-Tradable Components

[Source: ABS Cat. No. 6401.0.]

[79] It is anticipated that once the effects of the appreciation of the Australian dollar are worked through, domestic factors may put upward pressure on the inflation rate.

Earnings

[80] Average weekly ordinary time earnings (AWOTE) increased by 5.7 per cent for the year to December 2003 which is consistent with recorded annual data since 2000. The WCI which does not reflect the impact of compositional change evident in AWOTE data also demonstrates consistent increases on annual data which are slightly lower than the average increases of wages in federal agreements. The growth in wages in the agreement-covered workers sector has been at a consistent level, recording 3.9 per cent for the year to December 2003.

Table 11: Measures of Wage Movements

Notes:

(a) Relates to February, May, August, November.

(b) Source: May 2003 decision, Table 7 at p. 24.

(c) Source: Exhibit Commonwealth 4 at paras 3.15-3.17.

(d) Source: Exhibit Commonwealth 5 at para R3.5.

[Source: ABS Cat. Nos 6345.0 and 6302.0.]

[81] Table 12 shows a broad range of earnings data covering wide sections of the community for the latest year which shows a comparison with the 3.6 per cent increase recorded for the WCI.

Table 12: Comparisons of Various Wage Movements

[Source: ABS Cat. Nos 6302.0, 5206.0 and 6345.0.]

[82] Chart 3 shows a comparison of annual wages and earnings growth for a number of wages series including AWOTE, WCI and average annualised wage increases for employees covered by federal certified agreements (AAWI) which measures the wage increases in certified agreements.

Chart 3: Annual Wages and Earnings Growth-Main Indicators
September 1998-September 2003

[Source: ABS Cat. Nos 6345.0, 6302.0 and 5206.0.]

The Labour Market

[83] Total employment has continued to reflect the strong growth which has been sustained for the last two years. The annual growth of 2 per cent to the December quarter 2003 has been achieved primarily through full-time employment growth of 2.3 per cent which has grown at almost double the part-time employment growth of 1.2 per cent. The unemployment rate for December 2003 of 5.6 per cent is the second consecutive monthly rate below 6 per cent.

[84] ACCI relied on ABS data on hours worked in the market sector which demonstrate that there has been virtually no increase since the end of 1999. The ACTU disputed the relevance of this data because the market sector, as defined by the ABS, is not a reliable proxy for the private sector since it accounts for only about 60 per cent of all employees and excludes 40 per cent of award-only employees.

[85] The latest data on underemployment indicate that in September 2003 there were 610 700 part-time workers who wanted to work more hours. This figure represents 23 per cent of all part-time workers and 6 per cent of people employed in Australia.

Table 13: Employment and Unemployment

Notes: All figures are trend estimates. Annual and quarterly figures are averages of monthly data.

[Source: ABS Cat. Nos 6202.0 and 6202.0.55.001.]

Employed Persons by Industry

[86] Employment growth in the year to February 2004 has been 1.3 per cent in both the market and the non-market sectors. The market sector has had strong annual increases in mining, construction, and transport and storage. The non-market sector has been fuelled by property and business services, and health and community services. Quarterly data show a mixed outcome with mining the clear growth industry across all sectors. Other industries recorded very low or negative growth. In particular, the zero quarterly growth in property and business services may be a reflection of the cooling-off of the property market.

Table 14: Employed Persons by Industry-Trend Data

[Source: ABS Cat. No. 6291.0.55.001.]

Household Savings Ratio

[87] The level of household savings as a proportion of the level of disposable income recorded its sixth consecutive decline in the December quarter 2003. This reflects the long-term downward trend in savings. Since September 2002 the savings ratio has been negative. Although caution must be exercised in the accuracy of this data, it is clear that households are saving less income than previously.

Chart 4: Household Savings Ratio

[Source: Exhibit ACCI 3, Chart 12.]

Productivity

[88] Market sector productivity for the financial year 2002-03 was somewhat lower than that recorded for 2001-02. For the year to December quarter 2003 the increase was 3.2 per cent, a substantial improvement on the year to September 2003. The latest quarterly data for both September and December 2003 reflect growth of 1.5 per cent following several quarters of low or negative growth.

Table 15: GDP Per Hour Worked-Market Sector

[Source: ABS Cat. No. 5206.0.]

Profits

[89] GOS data demonstrate strong growth of 2.3 per cent in the December quarter 2003 with the increase in the year to the December quarter at 8.1 per cent. The latest annual data reinforce very solid annual growth since the year ending December 2001.

[90] The GOS measure of company profits recorded an 8.1 per cent increase in the year ending December quarter 2003. This also demonstrates consistent strong growth for each financial year since 1998-99.

[91] The ACTU's preferred measure of profit share as a proportion of total factor income shows a record share at around 25 per cent in the year to the December quarter 2003. Profits as a share of total factor income has shown a very gradual increase since 1998-99.

[92] In recent years there has been a trend to incorporation among Australian businesses. The ACTU's measure of profit share excludes unincorporated enterprises measuring only the profit share for corporations. If gross mixed income, which includes the profits of unincorporated entities, is included in the measure of profit share a different picture emerges because gross mixed income has experienced a long-term downward trend.

Table 16: Company Profits

[Source: ABS Cat. Nos 5651.0 and 5206.0.]

[93] Chart 5 includes the gross mixed income in the calculation of the profit share.

Chart 5: Movements in Profit and Wages Shares of Total Factor Income
Including Gross Mixed Income

[Source: ABS Cat. No. 5206.0.]

[94] When unincorporated businesses are included in the measure, it can be seen that the profit share has been stable at around 34 per cent of total factor income since 1990. This is a lower share than the 40 to 45 per cent experienced in the 1960s.

[95] The criticism of using this measure of profit share is that gross mixed income includes income received by the owner as a return to capital (profits) as well as the return to the owner's labour (income). As such it is not a pure measure of profits.

Conclusions from Aggregate Data

[96] The Australian economy has continued to grow strongly over the course of 2003. The greatest contributors to growth have been private consumption and farm output, the latter reflecting a turnaround from the low base which resulted from the drought conditions and sluggish international growth. Private sector business investment growth remains solid with strong contributions being made across the board, notwithstanding a substantial cooling-off of the housing sector in line with predictions.

[97] The labour market remains buoyant with recent growth in employment being achieved primarily by growth in full-time employment. Unemployment is at the lowest level for over 20 years.

[98] The export and import competing sectors have felt the negative effects of the appreciation of the Australian dollar which is also reflected in high import growth.

[99] In the year to December 2003 productivity growth remains strong.

[100] Profit growth is strong and the profit share is at record levels on conventional measures although we note there is controversy over whether the conventional measure is still appropriate.

Economic Outlook

[101] The most recently released and widely relied on forecasts of the economic outlook for Australia are the Treasury (MYEFO) forecasts for 2003-04.

Table 17: MYEFO Domestic Economic Forecasts^(a)

Notes:

(a) Percentage change on preceding year unless otherwise indicated.

(b) Calculated using data.

(c) Chain Volume measure.

(d) Excluding transfers of second-hand asset sales between the public and private sectors.

(e) Percentage point contribution to growth in GDP.

(f) For presentation purposes, forecast change in inventories held by privatised marketing authorities. Included with the inventories of the farm sector and public marketing authorities.

(g) Average earnings (national accounts basis).

[Source: Treasury, Mid-Year Economic and Fiscal Outlook, December 2003.]

[102] In summary, it can be seen that the Australian economy is expected to continue to perform strongly throughout the 2003-04 financial year with gross domestic product forecast to increase by 3¾ per cent compared with the 3¼ per cent earlier budget estimate. Household consumption is forecast to grow at a slightly higher rate than in 2002-03 whilst total business investment is anticipated to grow at a slower rate than in the previous year, albeit at a healthy 7 per cent. Net exports are forecast to make a negative contribution to growth. The labour market is forecast to grow and unemployment to again fall. The CPI is expected to record a financial year average of 2¼ per cent increase, a reduction on the 3.1 per cent increase in the previous year. Earnings growth will be in line with last year's outcome.

[103] The Treasurer made the following comments on the MYEFO forecasts:

"The 2003-04 Mid-Year Economic and Fiscal Outlook (MYEFO) released today demonstrates the continued resilience of the Australian economy and reflects the Government's ongoing commitment to a strong economy through sustainable fiscal management.

The Australian economy is forecast to grow by 3¾ per cent in 2003-04, an upward revision from the 3¼ per cent forecast at the 2003-04 Budget. The upward revision comes against a backdrop of improving sentiment about domestic and external growth prospects. Generally the near term outlook for the Australian economy has brightened since the Budget with increasing signs of global recovery, particularly in the United States, and an easing of drought conditions over much of Australia.

. . .

The initial forecast for 2004-05 is for economic growth of around 3½ per cent. Domestic demand is expected to be solid and external conditions should be supportive of growth. Employment growth is forecast to be moderate and inflation is expected to fall a little further.

There are still risks around the outlook, particularly regarding the sustainability of the international recovery, but overall the near-term risks have diminished since Budget and are now more evenly balanced."¹¹

[104] In his Statement on Monetary Policy in February 2004, the Governor of the Reserve Bank provided an overview of the direction the economy was expected to take based on the MYEFO forecasts:

"Overall, the growth of the economy has been driven by well-above-average growth in domestic spending, while the main factors that were holding back growth, particularly in the period up to around mid 2003, were the drought and the unfavourable international environment. These negative factors, however, are now being reversed. With improved rainfall in most areas, farm production is adding significantly to growth. Reflecting the strengthening international environment, Australia's export earnings have begun to increase gradually after their decline in the first half of 2003. In light of these trends and the more promising global outlook, the prospects are that the Australian economy will continue to grow at a strong pace during 2004. Exports are likely to continue their gradual recovery as a result of stronger trading-partner growth, even though progress in this area will be dampened to some extent by the higher exchange rate now prevailing, and also by capacity constraints in the resources sector. Domestic demand growth is likely to moderate a little but remain quite strong.

Australia's recent inflation performance has been marked by contrasting influences from domestic conditions and the exchange rate. In the December quarter the CPI increased by 0.5 per cent, and by 2.4 per cent over the year, down from an annual rate of around 3 per cent a year earlier. Inflation in the non-traded sector of the economy remains relatively high, at over 4 per cent, reflecting the overall strength of the domestic economy, strong demand conditions in the housing sector and continuing cost pressures in some service industries. At the same time, the overall inflation rate is being held down by the gradual pass-through of the exchange rate appreciation, with prices of tradable items in the CPI declining slightly in recent quarters.

The experience of recent years has been that these exchange rate effects have tended to be smaller than initially expected, but even so, it is likely that the dampening effect from the exchange rate still has some way to run. As a result, inflation is likely to decline further over the coming year, and could fall as low as 1½ per cent, before returning to around 2½ per cent in 2005 and then continuing on a gradually rising trajectory."¹²

[105] The Organisation for Economic Co-operation and Development (OECD) underpinned the positive expectations for the Australian economy:

"The projections are for an acceleration of economic growth, accompanied by a rebalancing of aggregate spending from domestic to foreign sources.

. . .

With the global economy recovering and the drought likely to break, exports are set to accelerate, narrowing the current external deficit from 6¾ per cent of GDP in mid-2003 to 4¾ per cent of GDP in 2005. The improved global environment should bode well for business investment, given existing high capacity utilisation, favourable company profitability and low corporate gearing. Inflation should remain within the Reserve Bank's 2 to 3 per cent target range, underpinned by the recent exchange rate appreciation, modest wage increases and a projected pick up in labour productivity."¹³

[106] Despite positive predictions on the economic front, the Commission has been cautioned by some parties that strong positive aggregated outcomes may mask emerging or evident risks which should be taken into account in determining the monetary claim before us.

[107] Australia's immediate economic outlook is positive with support coming from both a strengthening domestic market and an improvement in the international economy which should obviate some earlier concerns that our growth rate may not be sustainable. There is every reason to believe that there will be further strength in the labour market with employment numbers on the increase and the favourably low levels of unemployment at least maintained.

[108] The recent trend in full-time jobs growth may alleviate the relatively high levels of recorded underemployment. A continuation of moderate wages and prices growth should not impede employment growth.

[109] The RBA anticipates that the inflation rate will continue to fall, but rise again beyond the 2003-04 financial year as the dampening affects from the rise in the Australian dollar work their way through the index. The RBA in its inflation forecasts has factored in the end of the deflationary effect of price changes on the tradables sector, which in recent times has recorded zero or negative price increases.

[110] The farm sector will continue to grow making a strong contribution from an abnormally low base as the negative effects of the drought recede, although the pace may be slower in some geographical areas.

[111] The manufacturing sector has suffered in the past 12 months from the exchange rate, global overcapacity and rising interest rates which have manifested as a loss of 70 000 jobs. However, survey data indicate that there are strong expectations of an improvement in the performance of the sector.

[112] Despite the cooling-off of the domestic housing sector, capital expenditure by businesses is expected to grow but by a lower rate than the very high rises experienced in recent years. Notwithstanding preliminary expectations of a fall-off in investment growth to negative numbers, the lowering of the cost of imported capital equipment may be an offsetting factor to investment measured in real terms. In addition, a slowdown in the domestic housing market will have a favourable impact on further interest rate rises.

[113] Overall, there are grounds for optimism that the Australian economy will perform strongly in the immediate future. We do not foreshadow, on the material currently available, any major impediments to the realisation of the Treasury forecasts. Any downside risks should be counterbalanced by strong aggregate performance.

COST OF THE CLAIM

[114] The ACTU submitted that its claim for a $26.60 increase in all award rates "will have a negligible impact on aggregate earnings of 0.1 per cent". This statement was based on estimates of a net impact on aggregate wages of 0.1 per cent and a gross impact of 0.4 per cent. Seen in historical terms, the ACTU submitted, the gross impact of the current claim is the same as the combined effect of the increases in the superannuation guarantee charge and the safety net adjustments in 2000 and 2002. Furthermore, the cost of the claim is in the region of the cost of the safety net adjustments actually awarded by the Commission in recent years.

[115] The ACTU submitted that its claim is a moderate one in historical and absolute terms. It estimated that if granted the claim would result in an average award increase of 4.7 per cent. On that basis, the average annual increase in award rates from 2000 to 2004 would be 3.4 per cent for all award workers and 3.2 per cent for full-time award-only workers. Over the four years to June 2003 the average annual increase in the WCI was 3.4 per cent. The claim would produce a real average annual increase of 0.6 per cent for award-only workers and a real average annual increase of 0.4 per cent for award-only full-time workers over the last five years.

[116] The Commonwealth did not provide an estimate of the aggregate cost of the claim, preferring to concentrate on the cost impact in the award-reliant sectors of the economy. It nevertheless contended that the ACTU's costing understates the true cost. It relied on two matters in particular. First, it argued that the gross cost should be given more weight than the net cost. Secondly, it argued that the ACTU should have included an allowance for flow-on of the increase to employees who are in receipt of overaward payments or paid under certified agreements. In relation to the second matter, the Commonwealth relied on the Minimum Wages Report in submitting that safety net adjustments influence the wages of around 2.5 million employees who are paid overaward wages. (We refer to the Minimum Wages Report in detail below.)

[117] As in previous years ACCI provided an estimate of the cost of the ACTU's claim based on the cost to the private sector only, submitting that almost the whole cost would be borne by that sector. ACCI estimated that if granted the claim would add 0.67 per cent to private sector costs, rising to 0.81 per cent when flow-on effects are added. It submitted that:

"Such an increase is not negligible, it adds substantially to costs of production, it increases inflationary pressures and it keeps the economy growing more slowly than would otherwise be the case."¹⁴

[118] Like the Commonwealth, ACCI submitted that the Commission should be more concerned about the impact of any increase on the sectors of the economy in which award reliance is highest.

[119] NMI also made submissions about cost. Based on survey evidence, it submitted that a total of 46.7 per cent of employees in the retail motor industry would receive any safety net adjustment in full. The annual cost to the industry would be $184 million, without allowing for increases in non-wage costs. No estimate was given of the overall labour cost increase. An increase of the magnitude proposed "with no requirement for an offsetting productivity improvement" would reduce profitability and efficiency in the industry.

[120] The only economy-wide cost estimate is that provided by the ACTU. It was challenged on two substantive bases. The first issue is the assumption to be made about the extent to which any safety net adjustment would flow to employees who are not paid at the award rate. The second issue is whether the gross or the net cost should be given greater weight.

[121] In relation to the issue of flow-on, sometimes referred to as part of the indirect cost of a safety net adjustment, various parties provided estimates. We have noted the Commonwealth's reliance on the Minimum Wages Report for the conclusion that safety net adjustments influence the wages of 2.5 million employees who are paid overaward wages. For reasons discussed elsewhere, we are reluctant to place too much weight on the report.

[122] ACCI relied on the results of its latest survey in support of a submission that around 5 per cent of private sector employees who are not award dependent receive the safety net adjustment, in addition to 24.6 per cent who are award dependent. NMI, also relying on survey evidence, submitted that the safety net adjustment awarded last year was paid to some 18.5 per cent of employees in the industry, in addition to the 28.2 per cent who were paid exactly the award rate.

[123] Reliable economy-wide data on the indirect costs of safety net adjustments is hard to find. In previous safety net reviews the Commission has not relied too heavily on surveys like those which ACCI and NMI advanced in this case.¹⁵ Data from the Award and Agreement Coverage Survey 1999 (AACS), discussed in the Safety Net Review-Wages May 2000 decision (the May 2000 decision), suggest that flow of safety net adjustments to employees in receipt of overaward payments or paid pursuant to certified agreements is limited.¹⁶

[124] In relation to the second issue, whether the gross or the net cost should be given greater weight, we think that both estimates are relevant. We approach the assessment in this case on the basis that if granted in full the ACTU claim would lead to an increase in aggregate labour costs of no less than 0.1 per cent in net terms and no less than 0.4 per cent in gross terms. The actual cost is likely to be more but it is not possible to be more precise about the issue.

WAGES MOVEMENTS

[125] The ACTU relied on movements in various earnings measures including Average Weekly Total Earnings (AWE), Average Weekly Ordinary Time Earnings (AWOTE), Average Earnings on a National Accounts Basis (AENA) and the WCI. We were presented with historical data showing the growth in each of the measures over various periods. The Commonwealth contended that we should place more weight on the WCI which, it was submitted, accurately measures changes in wage rates for performing the same work at the same level. It submitted that the earnings measures, AWE and AWOTE, are growing faster because of the increasing proportion of workers in higher paying occupations and because earnings measures are affected by extraneous factors such as length of employment, skills and qualifications.

[126] The ACTU also submitted that award-only employees received an average increase of 3.1 per cent as a result of last year's decision and the rest of the community fared better. The Commonwealth took issue with this claim. It submitted that dissection of the WCI shows a range of increases above and below 3 per cent and that sizeable increases in the minimum wage and C10 since 1997 should be taken into account. In particular, movements in the minimum wage and C10 have kept pace with movements in the WCI over that period. The Commonwealth further contended that it should be borne in mind that award coverage is disproportionately concentrated among labourers and related workers and elementary, clerical, sales and service workers. Workers in those occupations tend to be relatively low paid whether paid pursuant to awards or agreements.

[127] The ACTU also relied on the level of wage increases in certified agreements. The average annual wage increase per employee for agreements certified in the four quarters to the September quarter 2003 averaged 4 per cent.¹⁷ The average annual wage increase per agreement for agreements certified during the four quarters to the September quarter 2003 averaged 4.1 per cent.¹⁸ It submitted that comparisons with earnings measures are appropriate when a consideration of living standards is apposite. The Commonwealth, on the other hand, cautioned against comparing average annual wage increases in certified agreements with movements in award wages. It pointed out, however, that the average annual wage increase for all agreements current as at 30 September 2003 was 3.8 per cent.

[128] The Commonwealth also contended that wage increases under agreements reflect the market and should not be used to adjust award rates. Average measures of wage increases, including increases achieved through bargaining, are an unreliable and inappropriate guide to fixation of safety net rates in awards. If the Commission were to grant the ACTU's claim the resulting increase would exceed the wage rises achieved by many workers though federal agreements, particularly employees at the lower levels in industries such as accommodation, cafes and restaurants, and retail. Increases in the award safety net of the magnitude the ACTU proposes would reduce the incentive for employees in the award-reliant industries to bargain and decrease the opportunities for productivity improvement.

[129] The ACTU also pointed to various measures of increases in management and executive remuneration. On those measures management and executive remuneration has been growing at an average of between 4 and 4.5 per cent per annum. It also relied upon Remuneration Tribunal decisions providing for increases of a similar magnitude and in some cases considerably more. The Commonwealth, on the other hand, submitted that movements in management and executive remuneration and Remuneration Tribunal determinations relate to small, highly specialised segments of the community and are not a useful guide in adjusting the safety net. In the same way, ACCI alleged that the material is selective and disputed its relevance.

[130] Although we have considered all of the historical material, it seems to us that the indicators for the latest year are the most relevant. Each year for the last seven years the Commission has adjusted the safety net of minimum wages in light of the submissions and the data available at the time of the hearing. While it is helpful to some extent to compare increases in award rates with increases in other measures over the longer-term, it would not be appropriate, under that guise, to reopen past decisions. For similar reasons it would not be appropriate to place too much weight on average award increases expressed in percentages. Each safety net adjustment since 1997 has been a flat dollar adjustment rather than a percentage. On some occasions the increase has been higher at the lower classification levels although on one occasion the increase was lower at the lower levels and higher at the higher levels. The overall effect of these adjustments has been that wages at the lower classification levels have increased more in percentage terms than wages at the middle and higher levels. The ACTU has consistently sought flat dollar adjustments in recent years and again in this case. In the circumstances we think it would be inappropriate to pay too much attention to the average increase in award rates. It is also important to point out that in several safety net review decisions the Commission has emphasised that changes in relativities brought about by flat adjustments cannot form the basis for future claims.

[131] Furthermore, while all of the measures relied on by the ACTU are of some relevance, the broader measures are more helpful than those measuring changes in particular segments of the labour market, such as management and executive salaries. In relation to those broader measures, the most recent figures are shown in the following table:

Table 18: Wages Measures

Notes:

(a) Source: ABS Cat. No. 6302.0.

(b) Source: ABS Cat. No. 5206.0.

(c) Source: ABS Cat. No. 6345.0.

[132] It is accepted that the AWE and AWOTE data are affected by compositional changes in the workforce. While the AENA and WCI data both record the increase in wages for a given job, and therefore are not affected by compositional change, AENA includes some non-wage costs such as superannuation, redundancy payments and workers' compensation payments. We agree with the Commonwealth that the WCI data are the most direct measure of changes in wage costs for a particular position and therefore the most useful indicator for our purposes.

AWARD EMPLOYEES AND AWARD INDUSTRIES

[133] The declining incidence of award-reliant employees in the workforce is evident from the ABS Employee Earnings and Hours, Australia survey (the EEH Survey).¹⁹ Data from the May 2002 EEH Survey show that 20.5 per cent of employees, or about 1.6 million employees, were award-reliant in May 2002, the corresponding figure from the May 2000 survey, was 23.2 per cent.²⁰ The proportion of award-reliant employees in each industry is set out in Table 19 below:

Table 19: Proportions of Award-Reliant Employees by Industry

Notes:

(a) Estimate has a relative standard error greater than 50% and is considered too unreliable for general use.

(b) Estimate has a relative standard error of between 25% and 50% and should be used with caution.

[Source: ABS Cat. No. 6306.0.]

[134] Employees classified as "award only" had the main part of their pay set by an award and were not paid more than the award rate of pay.²¹

[135] Award-reliant employees are:

• generally paid less than the rest of the working community;²²
• more likely to be women;²³
• concentrated in three industry sectors: accommodation, cafes and restaurants, retail trade, and health and community services;
• overwhelmingly employed in the private sector;²⁴ and
• overwhelmingly employed in non-managerial occupations and concentrated in lower-skilled occupations.

[136] Table 19 shows that award-reliant employees are most heavily concentrated in the industry sectors of accommodation, cafes and restaurants, retail trade, and health and community services. It was submitted that these three sectors account for nearly 60 per cent of all award-reliant employees. About a quarter of all such employees are in the retail industry and more than 15 per cent in each of accommodation, cafes and restaurants, and health and community services.

[137] The ACTU contended that the economic performance of the most award-reliant sectors has been strong and in some key respects exceeds the average for all industries. In particular it is argued that the data show that:

• the three most award-reliant industries have performed better than all-industry averages in terms of growth in output and employment for the period 1996-2003;
• in the same industries, and for the same period, productivity growth has outpaced the growth in real award wages and as a result real unit labour costs have fallen;
• in the two most award-dependent industries profits have increased by more than 80 per cent for 1996-2003, while real award wages have grown by less than 9 per cent; and
• the recent performance of the three most award-reliant sectors has continued to be strong, with output and productivity growth for 2002-03 exceeding the all-industry average.

[138] We now turn to briefly consider each of these contentions.

Output

[139] The ACTU submitted that growth in output in the three most award-reliant industries has exceeded the all-industry average for the period June 1996 to June 2003.²⁵

[140] Chart 6 shows that industry gross value added increased in accommodation, cafes and restaurants by 3.9 per cent, in retail trade by 4.6 per cent and in health and community services by 3.6 per cent, exceeding the overall growth in GDP of 2.8 per cent over the 2002-03 financial year.

Chart 6: Output Growth-Award Dependent Industries 2002-03

[Source: Exhibit ACTU 1, Figure 2.1 at para 2.6.]

[141] In its oral submissions, the ACTU argued that recent national accounts data for the year to December 2003 show growth in accommodation, cafes and restaurants of 8.3 per cent and in the retail trade of 7.4 per cent, compared to growth in GDP of 3.5 per cent, with growth in health and community services slightly lower at 3 per cent.²⁶

Profits

[142] The ACTU contended that in 2002-03 the increase in gross operating profits in two of the most award-reliant sectors - accommodation, cafes and restaurants (12.7 per cent) and retail trade (21.7 per cent) - exceeded the average increase for all industries (8.2 per cent).²⁷ The ABS does not measure gross operating profits for the health and community services sector.

[143] In the period 1996 to 2003 gross operating profits in the accommodation, cafes and restaurants sector increased by 82.1 per cent and by 99 per cent in retail trade.

[144] The Commonwealth criticised the profit measure relied upon by the ACTU on the basis that it only covers incorporated businesses with 20 or more employees and ignored unincorporated businesses. In the restaurant and catering component of the accommodation, cafes and restaurants sector - which accounts for about half of the sector's total employment - over 90 per cent of businesses have fewer than 20 employees. The Commonwealth contended that profit should be measured as GOS and gross mixed income, minus capital consumption, as a proportion of end-year net capital stock (the accounting rate of return as opposed to the economic rate of return).

[145] Based on the Commonwealth's data, average annual nominal profit growth from 1995-96 to 2002-03 was 1.0 per cent for retail trade, 6.6 per cent for accommodation, cafes and restaurants and 5.3 per cent for health and community services. The comparable figure for all industries was 5.9 per cent. In 2002-03 nominal profit growth was stronger than the national average for the three award-reliant sectors - particularly in retail trade where it increased by 14.3 per cent - reflecting a recovery in profit levels from previous years.

[146] A difficulty with the Commonwealth's suggested measure of profit is that gross mixed income not only captures returns to capital for unincorporated businesses, but returns to labour as well. The combination of GOS and gross mixed income provides a measure of profits in all businesses and returns to labour in unincorporated businesses.

[147] The second difficulty with the Commonwealth's approach concerns its measure of profit as a proportion of end-year net capital stock. The Commonwealth acknowledged the difficulty with using accounting rates of return and attempted to minimise the deficiencies by subtracting capital consumption from profits.²⁸

[148] Even if it were accepted that the accounting rate of return was an appropriate measure of profits, the ABS publication Business Operations and Industry Performance, which covers all public trading and private employing businesses and provides a measure of profits as a proportion of assets, shows that the return on assets in the most award-dependent industries exceeded the all-industry average for the financial year ending June 2001. However, the ABS highlights the limitations of this measure due, in part, to differences in accounting policy and practice across businesses and industries.

[149] Given the limitations in the data used by both the ACTU and the Commonwealth, the submissions made in relation to profit levels in the award-reliant sectors are of little assistance.

Employment

[150] Chart 7 compares employment growth between the three highest award-reliant industries and the three least award-reliant industries since May 1996, using a common index base:

Chart 7: Employment Growth 1996-2003 (Selected Industries)

[Source: ABS Cat. No. 6291.0.55.001.]

[151] Employment growth in the three most award-reliant industries over the period 1996-2003 has exceeded growth in the three least award-reliant industries and has exceeded the all-industry average. However, as the Commonwealth noted in its submission, a range of factors affect the rate of employment growth in an industry sector including productivity, changes in consumer preferences and industry output.²⁹ The Commonwealth also contended that it is too simplistic to conclude that because employment growth in the award-reliant industries has outstripped total employment growth, previous safety net adjustments have had no impact on employment.

[152] AiG relied on the dispersion of growth in employment for different occupations ranked according to whether they are high or low paying as evidence that safety net adjustments have impacted adversely on employment. The data utilised are somewhat outdated - the latest yearly data is 2000 which is of no relevance to the safety net increases in recent years. The earlier data demonstrate that relative growth in hours was lower in the bottom decile in the period 1996-2000 than 1986-95. However, because the impact on employment of safety net adjustments has not been isolated as a discrete factor, we are unable to draw conclusions on the cause of the relative employment growth. This point was acknowledged by the Commonwealth's view that "wage levels are just one of many variables that affect the rates of employment growth".³⁰

[153] The Commonwealth undertook an exercise of measuring the differential impact of increases in award wages upon employment by comparing changes in the employment of award-reliant and agreement-based workers within a sector. The approach adopted was said to help account for all the other factors which may affect the employment growth of industries which have confounded such analysis in the past.

[154] Chart 8 demonstrates that non-award employment increased in all but three industries with a net increase of 327 000 jobs. Award employment fell in all but five industries, with a net decline of 133 000 jobs. The Commonwealth concluded that because the increase in numbers in employment covered by agreements exceeded the decrease in numbers covered by awards "it therefore seems reasonable to conclude that most new employees are engaged under agreements, not awards".³¹

Chart 8: Change in Numbers Employed Under Agreements and Awards by Industry
May 2000-May 2002

[Source: ABS Cat. No. 6306.0 and ABS, Supertables, E06.]

[155] This could be so. It could also be the case that award-reliant employees are increasingly being covered by agreements. Such a conclusion would be consistent with the decline in the proportion of employees covered by awards. It would follow that increases in the safety net are not inhibiting the movement away from award coverage to agreement coverage.

Productivity

[156] The ACTU contended that labour productivity increased in major award-reliant industry sectors in 2002-03. Productivity growth in accommodation, cafes and restaurants was 5.9 per cent for that financial year, whilst in health and community services productivity increased by 2.0 per cent and in the retail trade by 1.1 per cent. In each case productivity growth in these heavily award-reliant sectors exceeds average productivity growth for all industries for the financial year 2002-03 of 0.8 per cent.

[157] In the period of safety net adjustments, real unit labour costs have fallen in each of the three most award-dependent sectors, that is the increase in productivity has outpaced the increase in real wages.

[158] This year the ACTU supplements its consideration of real increases in gross award wages with a consideration of real increases in after-tax award wages. Chart 9 shows the real after-tax wage for all wage rate classifications in the Metal Industries Award at or below the C10 level from June 1996 to September 2003. As the chart illustrates, the real after-tax wage for C14 has only increased slightly since 1999 and wage rates for C11 and C10 are currently worth less in real terms than they were in June 1999.

Chart 9: Real After-Tax Wages

Notes:

(a) All figures September 2003 dollars.

(b) After-tax wages includes low income tax off-set (where applicable) but excludes Medicare levy.

[Source: Exhibit ACTU 1, Figure 1.1 at para 1.14.]

[159] The real after-tax wage for an employee on the federal minimum wage (C14) in September 2003 was $391.12, just $5.47 more than the corresponding June 1999 figure, an increase of 1.4 per cent in over four years - an average annual real increase of 0.5 per cent. Award workers in the C11 and C10 classifications have suffered real declines in after-tax income in the period ($0.58 and $1.63 respectively) with workers at C12 and C13 essentially at the same position in September 2003 as they were in June 1999. (The real after-tax wage for C12 has increased $0.68 in the four-year period and for C13 has increased $2.04.)

[160] The Commonwealth conducted a series of regression analyses to test the relationship between productivity performance and award coverage. Its analysis of seven industries concluded that:

• nearly 60 per cent of the difference in the average annual rate of productivity growth between 1990-2003, between the seven industries modelled, is explained by the change in award coverage;
• different industries have different capacities for productivity growth. Both capital deepening and an industry's intrinsic capacity for productivity growth influence productivity outcomes;
• industries with high levels of award-reliance had lower productivity growth rates regardless of whether they can be considered intrinsically high or low productivity industries; and
• a reduction in award-reliance is associated with an increase in productivity. A 10 percentage point reduction in award-reliance in an industry between 1990 and 2002 was associated with an increase in the average annual productivity growth of 0.5 percentage points.

[161] The Commonwealth undertook a second regression analysis whereby it included service industries and disaggregated the retail and wholesale industry such that eleven industries were analysed.

[162] The analysis showed that:

• the level of award-reliance in 2002 was highly significant in explaining the differences in industries' productivity growth, even after taking into account the effects of capital deepening and the intrinsic capacity for productivity growth as measured by average levels of productivity; and
• a 10 percentage point reduction in award-reliance in 2002 in an industry was associated with a 0.6 percentage point rise in that industry's average annual productivity growth between 1990 and 2003.

[163] Professor W Mitchell, Professor of Economics and Director of the Centre of Full Employment and Equity at the University of Newcastle, in a critique included in the ACTU's reply submissions, raised a number of limitations in the methodology applied in regression analyses, including:

• failure to fully report diagnostic statistics;
• the number of observations were below the professionally accepted level of 30 in all but one of the models;
• measurement problems and error; and
• failure to control for factors which may affect both variables and produce bias and endogeneity problems.

[164] Given the technical limitations of the exercises undertaken, the correlations, which the Commonwealth regression analyses produce, do not allow us to reach any conclusions on the causes of the productivity growth in particular industries. In particular, we do not think it is safe to conclude that a reduction in award coverage resulted in higher productivity increases, or, as a corollary, that higher productivity increases arose out of a higher level of coverage in the enterprise bargaining sector within an industry.

[165] We accept the Commonwealth's submission that its analysis "makes good use of the limited data available". However, in the absence of more reliable and appropriate data it is not possible to confidently reach conclusions about causality from these regression analyses.

[166] Based on the material before us, we adhere to the conclusion reached in the May 2003 decision that it has not been demonstrated that there is a negative association between safety net adjustments and productivity growth. There is no necessary association between award coverage, safety net adjustments and productivity growth.³²

General Conclusion

[167] Having regard to the range of material in respect of the economic performance of the most award-reliant industry sectors, we are unable to discern any negative impact from the safety net adjustments which have been awarded in recent years.

ECONOMIC EFFECTS

Academic Research

[168] As has been the case in previous safety net review proceedings, the parties relied on academic papers and other research material in support of their contentions. We now turn to deal with some of this material.

The Minimum Wages Report

[169] In the May 2003 decision the Commission encouraged the parties, and in particular the Commonwealth, to give consideration to facilitating a comprehensive, representative and technically robust survey to provide direct and contemporary information relevant to the Commission's task in adjusting the wages safety net.³³

[170] The Commonwealth responded by commissioning research intended to provide answers to the following questions:

• How many businesses and employees have their wages adjusted through the annual safety net case and its flow-on effects?
• What is the effect of the annual safety net decision on wage setting and labour costs?
• What was the effect of the 2003 safety net adjustment on employment? and
• What is the effect of safety net adjustments on employment levels?

[171] Turning Point Research Pty Ltd was commissioned by the federal Department of Employment and Workplace Relations (DEWR) to undertake two tasks. The first was to design a questionnaire, to be included in the October-November 2003 Yellow Pages survey of small and medium sized businesses, that could provide information to help answer these research questions. The second task was to analyse the data collected from that survey.

[172] The resultant report titled Minimum Wages in Australia: an analysis of the impact on small and medium sized businesses (the Minimum Wages Report) was tendered during the course of the proceedings and one of the authors of the report, Dr Don Harding, gave evidence.

[173] The survey panel consisted of some 1800 businesses, being 1200 small businesses (i.e. those with 19 or fewer full-time employees) and 600 medium sized businesses (i.e. those with 20 to 200 full-time employees). The response rates to the survey, at 20 to 22 per cent, were lower than expected.³⁴ It took Sensis Pty Ltd (Yellow Pages), who administered the survey, 9887 business contacts to obtain the required 1800 interviews.

[174] The survey was piloted on seven firms in metropolitan New South Wales, but ultimately was only trialled with two of these firms because five of the firms contacted proved to either have no minimum award wage rate employees or so few such employees that they did not move beyond the screening questions in the survey.³⁵

[175] Some of the key findings of the Minimum Wages Report are:

• About 1.8 million employees of small and medium sized businesses are paid minimum award rates.³⁶ A further 1.1 million employees are paid overaward wages and receive an automatic flow-on of the May 2003 safety net adjustment to their wages³⁷ and an additional 1.4 million employees received or expected to receive, discretionary flow-on of the May 2003 safety net adjustment.³⁸ Hence the annual safety net adjustments to minimum award wage rates influence (directly or indirectly) about 4.3 million people employed by small and medium sized businesses.³⁹
• Over a three-month period the short-run impact of the May 2003 safety net adjustments on employment demand was 14 000 fewer job places than would otherwise have been the case. This represents an elasticity of demand of about -0.2 per cent, that is, for every 1 per cent increase in award wages employment demand for award workers will fall by 0.2 per cent.⁴⁰ The estimate of 14 000 jobs was subsequently revised to a figure of "almost 13,000" in Dr Harding's supplementary statement of 15 April 2004.
• A guarantee that there would be no safety net adjustments awarded for a period of five years would result in an increase in employment demand of 245 000 jobs.⁴¹

[176] In his response to criticism of Dr Ian Gordon, Director, Statistical Consulting Centre, University of Melbourne, which formed part of the ACTU's reply, Dr Harding calculated confidence intervals. He concluded that almost 13 000 jobs were lost, or not created, because of the 2003 safety net adjustment. Applying confidence intervals he found that at the 95 per cent level the number of jobs lost was between 7067 and 18 807. Dr Harding concluded that the "main point to emerge . . . is that the job loss is statistically significantly different from zero" that is, a positive number of jobs were lost, or not created, as a result of the 2003 safety net adjustment.⁴²

[177] The Commonwealth relied on the Minimum Wages Report in support of its contention that a cautious approach should be taken to adjusting minimum wages because of negative employment consequences.

[178] The ACTU contended that "no weight whatsoever" should be placed on the Minimum Wages Report because of a number of serious deficiencies in the methodology adopted. We do not propose to set out each of the points advanced by the ACTU in support of this contention. We need only deal with three of the points advanced.

[179] First, the accuracy of the survey responses is dependent on the extent to which respondents to the survey understood what a safety net adjustment is. Dr Harding agreed that without such an understanding a person could not meaningfully answer the key survey questions.

[180] In his evidence Dr Harding dealt with this issue in these terms:

"[I]f you look at the preamble at the beginning of the report, the objective of the preamble is to make sure that people understand the context of the next set of questions that are coming. So it is put into - they are put into that context. People who are unsure of that context have an opportunity to say that they are unsure of that context, and they have opportunities through the report, to say, don't know.

So that you can be assured that these are responses that the people who responded understood the context of the question that was asked."⁴³

[181] The preamble referred to is in the following terms:

"Federal Minimum award wages, often referred to as safety net wages, are set each year by the Australian Industrial Relations Commission and also passed on to State award wages. I would like to ask you some questions about how wages are determined in your business, about whether the increases in minimum award wages influence your business in any way."⁴⁴

[182] There is reason to doubt whether at least some of the survey respondents had a sufficient understanding of the concept of a safety net adjustment to enable them to accurately answer the questions put.

[183] In its May 2002 EEH Survey, the ABS asked questions regarding both overaward employees and the flow-on of safety net adjustments. The responses to these questions have not been published due to data quality concerns. The ABS has advised the ACTU as follows:

"A number of data items collected in EEH 2002 are not available for release as a result of data quality concerns. Included among the data items deemed not available for release are data on `Award (paid more than the award rate)' and the `Safety net wage adjustment'. For both of these data items, quality concerns were raised during the editing process and confirmed during the subsequent Post Enumeration Survey. Contact with providers gave a strong indication that there were frequent instances of incorrect reporting, usually resulting from a misunderstanding of the question or the associated notes and definitions. In particular, many respondents were identified as having incorrectly reported employees being paid by `award (paid more than the award rate)' when there was either clearly no link between the employee's rate of pay and the relevant award rate, and/or some other form of agreement was in place which took precedence over the award. With respect to the safety net adjustment, it became apparent that this term and the associated concept were not widely understood by respondents. In addition to under reporting of entitlement to the safety net adjustment among award-only employees, there was frequent incorrect reporting of safety net entitlements among employees who had no link between their remuneration and an underlying award."⁴⁵

[184] No party sought to challenge the accuracy of this advice.

[185] The EEH Survey has a significantly larger sample size than the Minimum Wages Report survey (8000 employer units representing 53 000 employees, as opposed to 1800 employer units) and has a response rate of 95 per cent.⁴⁶

[186] The EEH Survey is a mail-out survey whereas the Minimum Wages Report questionnaire was administered by telephone interview. The wording of the explanation given for the relevant questions in the EEH Survey is not materially different from the preamble in the Minimum Wages Report survey, though the placement is different. In the Minimum Wages Report survey the preamble is read at the beginning of the relevant section of the questionnaire whereas, in the EEH Survey, it appears immediately before the questions which relate to the impact of the safety net adjustment. In our view these are not material points of difference.⁴⁷

[187] In his oral evidence Dr Harding suggested that the EEH Survey, which was conducted in May 2002, took place at a different time in the public understanding of safety net adjustments. There was no other evidence to support the assertion that the public understanding of these issues has materially increased between May 2002 (when the EEH Survey was conducted) and October-November 2003 (when Sensis administered the relevant questionnaire).

[188] Dr Harding also said that the survey respondents were given every opportunity to reply "don't know" in respect of the questions put. But the difficulty with this proposition is that a survey respondent may believe that he or she knows what a safety net adjustment is, but their understanding may be incorrect. Nor is there any record of how many of the respondents said that they did not understand the preamble.

[189] There was no cross-check question in the Minimum Wages Report survey to check the respondent's understanding in this regard. Such questions were apparently not included due to budgetary constraints on the number of questions that could be asked. Nor was a post-enumeration survey conducted to check the veracity of the answers given. Further, as we have noted, the trial of the survey was very limited and the Commonwealth acknowledged that the trial "provided less help than usual in fine tuning the design of the survey".⁴⁸

[190] In considering the effect of these issues, it is important to bear in mind that the employment impact of the May 2003 decision suggested by the Minimum Wages Report is based on an extrapolation of the responses of just 37 firms who reported an adverse economic impact from the May 2003 safety net adjustment. Two of the 37 firms said that the effect was zero, thus 35 firms assigned a negative employment impact to the 2003 safety net adjustment.⁴⁹ Hence, if only some of the respondents answering on behalf of these 35 firms had a flawed understanding of what a safety net adjustment is, this might have had a substantial effect on the estimate of the employment impact of the May 2003 decision.

[191] The second point to mention is that there are significant differences between a number of industry sector estimates extrapolated from the Minimum Wages Report questionnaire and those from established ABS surveys.⁵⁰ We briefly summarise some of these differences below:

Table 20: Differences between Minimum Wages Report and EEH Survey

[192] Given the robust nature of the ABS survey these discrepancies raise serious doubts about the reliability of the Minimum Wages Report.

[193] The third point relates to the low response rate. The response rate of the Minimum Wages Report survey was 20 to 22 per cent. In the opinion of Dr Gordon, a response rate lower than 30 per cent means that the survey must be regarded as unreliable due to possible response bias. As pointed out by Dr Gordon, response bias will arise when respondents to a survey have different opinions and characteristics to those who do not respond. Dr Gordon referred to the ABS examination of data pertaining to business surveys to establish whether response bias is a problem. The ABS concluded that:

"Overall, we can conclude that non-response bias can have significant detrimental effects on the accuracy of survey estimates. These effects can be reduced through higher response rates. From this, it is recommended that steps should be taken to ensure that the response rates achieved in any particular survey are as high as possible."⁵³

[194] In summary, Dr Gordon stated that in the absence of a survey of non-respondents, there could be no certainty that non-response bias would not be present. There is no basis for the assertion that because the survey is stratified any biases between stratum may cancel out because they operate in different directions. It is possible that a non-response bias will operate across all strata and this will occur when the bias is not related to characteristics captured by the stratification.

[195] Dr Harding acknowledged that the response rate of the survey was low but relied on an article on low response rates in the economic literature to support his assertion that the survey results could be valid notwithstanding the low response rate. The article by Johnson and Owens, Survey Response Rate in the Professional Literature,⁵⁴ reported the results of a survey of editors of journals ascertaining policy positions held on publishing articles with low or non-reporting of response rates. The survey arose out of concern that response rates were declining yet they are viewed as the most widely compared statistic for judging the quality of a survey. They stated that "many would argue that the issue of declining response rates is a growing crisis in the profession". The survey results confirmed the authors' concerns because none of the journals had a minimum response rate standard. Most editors rely on the expertise of peer reviewers to determine whether information on response rates is necessary. In our view the article relied on by Dr Harding to justify the publication of the survey does little to overcome the conclusion that the response rate of his survey was at a level which means non-response bias may be present.⁵⁵

[196] Having regard to the methodological limitations identified we think it would be unwise to place any reliance on the Minimum Wages Report.

The New Zealand Minimum Wage Experience

[197] The ACTU relied on a paper by Hyslop and Stillman⁵⁶ to support its contention that increases in minimum wages do not harm employment.⁵⁷

[198] The Hyslop and Stillman paper analyses the impact of the increases in minimum wages affecting youth workers in New Zealand since 2001. Prior to those increases the youth minimum wage, applying to 16-19 year olds, was set at 60 per cent of the adult rate. Two changes were made. The first lowered the eligible age for the adult minimum wage from 20 to 18 years, and resulted in a 69 per cent increase in the minimum wage for 18 and 19 year olds.

[199] The second change raised the minimum wage in two annual steps from 60 to 80 per cent of the adult minimum, and resulted in a 41 per cent increase in the minimum wage for 16 and 17 year olds over a two-year period.⁵⁸

[200] Using data from the New Zealand Household Labour Force Survey for the period 1997-2003, the authors examined changes in the labour market experiences of the two groups of teenagers directly affected by these changes (i.e. 16 and 17 year olds, and 18 and 19 year olds) and compared these with the changes experienced by young adults aged between 20 and 25.

[201] The paper concludes "we find no consistent and robust evidence of any adverse effects of the changes on teenage employment".⁵⁹

[202] For a number of reasons we have concluded that the Hyslop and Stillman paper has little relevance to the determination of the matter before us. Five points may be noted in this regard.

[203] First, the paper analyses the impact of increases in minimum wages in circumstances which are quite different to those which are before us. We think there are inherent limitations in comparing the impact of a single statutory minimum rate with an award system that sets minimum wages for a number of skill levels.

[204] The second point to note is that it may be premature to conclude that increasing the minimum wage for young workers in New Zealand will have no adverse effect on employment in that country. In that regard Hyslop and Stillman sound a note of caution in their concluding remarks "given the recent increases, whether such benign effects continue going forward remains a moot point".⁶⁰

[205] Third, there is a real issue about whether the minimum wage changes actually had an impact on the rate of pay of teenagers and as Burkhauser, Couch and Wittenberg observe in their 2000 paper A Reassessment of the New Economics of the Minimum Wage Literature:

"If changes in the minimum wage are to affect teenage employment, they must first have an impact on teenage wage rates."⁶¹

[206] Although there were increases of 69 and 41 per cent in the statutory minimum wage for 18-19 and 16-17 year olds respectively, the actual increase in average real hourly rates of pay for these two groups was much less. Hyslop and Stillman also acknowledge that "the pre-reform youth minimum wages were comparatively low".⁶² At footnote 17 of their paper they state:

"[T]he post-reform average wage for 16-17 year-olds is 7 percent higher than the pre-reform average, while the increase for 18-19 year-olds is only 4 percent."⁶³

[207] The authors make the following comments about their results:

"This reform appears to have shifted the lower-end of the wage distribution to the right for both 16-17 and 18-19 year-olds relative to older adults. Also, the shift in the wage distribution is larger for 16-17 year-olds than for 18-19 year-olds, partly because the minimum wage reform was less binding on 18-19 year-olds' wages."⁶⁴

[208] The fourth point is that the study documents an increase in minimum wage non-compliance for teenage workers since the increases came into effect. The proportion of workers in each age group reporting sub-minimum wages increased substantially. In the 18-19 year old group the proportion increased from 1 per cent to 12 per cent in 2001 and 2002, and to 13 per cent in 2003. The proportion of 16-17 year olds receiving sub-minimum wages increased from 3 per cent in 2000 to 6 per cent in 2001, 9 per cent in 2002 and 8 per cent in 2003. The authors conclude that "these increases imply a significant increase in non-compliance with the statutory minimum wage". As the Commonwealth suggested in its submission, increased compliance may come at the expense of employment.

[209] Finally, the Hyslop and Stillman results could have been affected by the fact that the unemployment rate in New Zealand declined from 7.5 per cent in 1998 to 5.4 per cent in 2003. As ACCI pointed out in its submission, the employment outcomes relied on by Hyslop and Stillman could have been due to other factors such as the general strengthening of the New Zealand economy or a tighter labour market.

The Western Australian Minimum Wage

[210] A recent paper by Leigh⁶⁵ has sought to estimate the effect of raising the Western Australian statutory minimum wage on employment.

[211] Leigh looked at six statutory minimum wage increases between 1994 and 2001. By comparing employment rates before and after the increases, he sought to test the proposition that increases in minimum wages cost jobs. Leigh made use of a control group (the rest of Australia) to validate his "before" and "after" results. The before and after minimum wage increase difference in employment rates for Western Australia was calculated by subtracting the employment rate three months after a minimum wage increase from the employment rate which prevailed three months before. This was done for six occasions when minimum wage increases occurred in Western Australia between 1994 and 2001. The before and after differences were also calculated for the control group, i.e. the rest of Australia. Leigh then calculated the difference between the Western Australian figure and the figure for the rest of Australia.

[212] An example will serve to better illustrate the methodology used. In the case of the March 2001 increase (8.8 per cent) Leigh found that the employment rate in Western Australia fell by 0.037 per cent, while employment for the rest of Australia fell by a smaller amount (0.020 per cent). The relative change in Western Australia is the difference between these two numbers, i.e. 0.017 per cent.

[213] Leigh's overall finding was that a 1 per cent rise in the minimum wage leads to a 0.15 percentage point fall in employment.⁶⁶ Leigh's study has been trenchantly criticised by Professor Junankar⁶⁷ and Dr Watson⁶⁸ in separate papers. In his March 2004 paper, Dr Watson describes Leigh's study as fundamentally flawed, with major methodological and empirical weaknesses.

[214] We do not propose to canvass all of the criticisms made of the Leigh study. For present purposes we note four particular issues:

1. There is reason to doubt the appropriateness of using the rest of Australia as a control group, as the other Australian states have widely differing industry and employment characteristics to Western Australia. Leigh has not controlled for other factors that may be influencing the rate of employment growth in Western Australia versus the rest of Australia, such as different compositional structures in GDP and different policy settings.

2. The six minimum wage increases which Leigh examined averaged 6.3 per cent hence, on his own estimation, we are looking at employment losses of around 1300 persons. Given the month-to-month variability in employment it seems unlikely that such an impact would be discernable. Further, the ABS table of standard errors for Western Australian employment statistics shows that figures of about 1000 persons are subject to standard errors of 450 persons.⁶⁹

3. In at least four of the minimum wage increases studied, the gap between the employment rate in Western Australia and the rest of Australia was already trending against Western Australia before the relevant increase in the minimum wage.

4. Leigh's results appear sensitive to a change in the months before and after the minimum wage increases which are analysed. The ACTU submitted that if a different time frame is chosen to analyse the effect, for example by comparing employment rates one month before an increase and three months after, then the results are quite different. On four of the 10 occasions when the statutory minimum wage increased, Western Australia had a better employment performance relative to the rest of Australia.

[215] In the circumstances we think that the appropriate course is not to place any weight on the study until it has been the subject of further debate. In any event we note that even if the methodological problems which have been identified are satisfactorily addressed, the study will only be of limited relevance. It focuses on the effect of adjusting a minimum rate and, as we have noted, the circumstances we are dealing with are quite different.

NFF Research

[216] NFF submitted that the Commission should consider a range of factors impacting across Australia to provide an overall determination of a fair outcome arising out of those proceedings.

[217] NFF commissioned a report entitled A Report on the Effect of Raising Minimum Wages on Rural Businesses by Professor Lewis, from the Centre of Labour Market Research at the University of Canberra (the Lewis Report).

[218] Professor Lewis considered a range of factors operating upon the rural sector including population and employment growth. Professor Lewis concluded:

• rural business statistics highlight the differential between rural and metropolitan sectors. In 1999-2000, the average revenue for rural businesses was only 76 per cent of the average for city businesses;
• rural businesses are not able to reap the economies of scale of urban businesses. Revenue streams are most likely to be limited by population size and growth in rural towns;
• average profit for rural businesses are 59 per cent of non-rural business profits;
• metropolitan businesses are less in debt than rural businesses;
• rural businesses are subject to greater seasonal fluctuations;
• rural businesses have greater input costs due to higher transportation costs;
• the welfare of rural businesses is critical to the welfare of rural towns; and
• farm businesses face declining terms of trade.

[219] NFF emphasised two key aspects arising out of the Lewis Report:

• the relative cost of living; and
• the effect of wages on demand for labour.

Cost of Living

[220] NFF submitted that the cost of living for rural Australia is 4 per cent lower than metropolitan Australia. Professor Lewis compiled a basket of goods and services and weighted it to reflect the expenditure pattern for rural Australia. His results are shown in the following table:

Table 21: Relative Cost of Goods and Services in Rural Australia

[Source: Caltex Australia (2003), Commonwealth Bank-HIA (2003) and Queensland Office of Economic Statistical Research (2001).]

[221] We make some brief observations about the conclusions reached:

• as the ACTU pointed out, the price index will depend on weights afforded the components in the index. A slight variation in the weight given to others to equate with the weight given to food, alcohol and tobacco, and petrol, rather than assigning the weight given to transportation, would reduce the differential from 4 per cent to 3 per cent;
• the resultant difference in the index is not out of kilter with the differences recorded in the level of the CPI for the capital cities. The level of the CPI varies for the December quarter 2003 by between 143.6 percentage points for Sydney and 138.5 percentage points for Darwin - a range of around 5 percentage points; and
• the level of the price index for the rural sector tells us just that - the level of prices. It does not record the rate of increase.

[222] NFF submitted that the Commission should take the lower cost of living in rural Australia into consideration in determining the level of a fair minimum wage.

Effect of Wages on Demand for Labour

[223] In his report, Professor Lewis relies on a number of studies which have measured the relationship between wages and the demand for rural labour. In a recent paper Professor Lewis, together with Garnett, summarised the studies and the results are included in his report. He concluded that all the results unambiguously imply that a rise in wages results in a "significant fall in employment". Lewis and Garnett also found that structural and technological change are important. They conclude that for hired labour the best estimate is that the effect of a 1 per cent rise in wages is a fall in employment of between 0.6 and 1.5 per cent.⁷⁰

[224] We make some observations about the conclusions reached by Professor Lewis:

• Professor Lewis himself notes that many of the studies are "somewhat dated".⁷¹ The studies summarised by Professor Lewis were published over the time span 1974-87.
• As acknowledged by Professor Lewis, the results of the studies differ somewhat due to the differences in definitions of employment, the industry coverage and the period over which the elasticities are estimated. The results for long-run hired labour varied from -0.58 to -2.28 per cent.
• All the studies modelled aggregate wages, they do not measure award-only employees. The outcomes are of little relevance to the quantification of the employment effects of a safety net adjustment to the award structure. In addition, the periods over which the studies were conducted were characterised by very different wage fixing mechanisms from current wage settings.
• In his paper with Garnett, Professor Lewis identifies a number of methodological problems in some of the studies which they rely on to determine the range of elasticities.

Other Academic Studies

[225] A number of academic studies relied on in previous years were again referred to, together with new research, which is primarily based on the effect of a change in teenage unemployment rates as a consequence of an increase in the minimum wage.

[226] For instance, ACCI relied on Burkhauser, Couch and Wittenburg's reworking of the Card and Krueger research by controlling for the effects of general economic conditions on employment to show that there were "significant but modest negative effects on teenage employment".⁷²

[227] The Card and Krueger study was criticised by the Commonwealth which relied on a recent study by Bazen and Marimoutou.⁷³ It is contended that this study overcomes deficiencies in early studies to successfully model teenage unemployment levels in the United States over a longer time frame. It finds a negative relationship between teenage employment and the minimum wage over the period 1954 to 1999. The resultant model shows that a 10 per cent increase in the minimum wage leads to a 1 per cent fall in teenage employment in the short-run, rising to a 2 to 3 per cent fall in employment in the longer run.

[228] The ACTU also relied upon a book by Professor Alan Manning, Professor of Economics and Director of the Labour Markets Program in the Senate for Economic Performance at the London School of Economics, titled Monopsony in Motion - Imperfect Competition in Labour Markets, in which he provides a theoretical basis under which employment might increase in response to an increase in minimum wages. This will arise where the employer has the power to determine the level of wages. If a minimum wage is set at a market clearing level, then employment may increase concurrently with increases in the minimum wage. Since the ACTU provided no empirical research which relates this theory to the fixation of award rates in Australia, we are not assisted by it.

Conclusions on Academic Studies

[229] After yet another survey of the literature and other material on the economic effects of increases in minimum wages it has clearly emerged, once again, that most of the information is of very limited assistance. The research is either largely irrelevant, is limited in scope, or has serious methodological flaws.

[230] On the question of relevance, it is only necessary to point out, as we have done several times in the course of this decision, and has been pointed out by the Commission in previous safety net review decisions, that studies based on movements in a single minimum wage are unlikely to shed much light on the economic effects of safety net adjustments. This is because safety net adjustments apply to a range of minimum rates at various levels throughout the award system. In the same way, studies which model a single statutory minimum wage increase for a particular sector of the labour market, such as teenage employment, are of limited relevance in determining the effect of increases in award rates of pay across a range of skill levels. To take one issue, the employment effects of safety net adjustments, there may well be a range of elasticities of demand for labour throughout the award classification structures and the elasticities may differ at various points in time. We add that there seems to be a significant measure of agreement that much of the research is about the effects of increasing minimum wages in circumstances which are quite different to those which characterise the Australian industrial relations systems. Despite this apparent consensus we have again been presented this year with material of that kind.

[231] As to the scope of the material, studies which deal with employment effects should be treated with caution unless they also deal with countervailing effects which might be equally significant. That proposition is well expressed in this passage from Leigh's paper:

"Naturally, this study has focused only on the employment effect of raising the minimum wage. But this is of course only half the story. Policy makers must weigh the employment cost of minimum wage increases against the higher earnings that accrue to low-wage workers and their families. This involves learning more about those workers who hold minimum wage jobs, and how minimum wage increases affect household incomes - issues about which we currently know surprisingly little."⁷⁴

[232] To similar effect is this passage from Dr Watson's summary of the literature:

"Obviously it is difficult to reach a firm theoretical conclusion about the impact of the minimum wage on employment. A great deal depends on numerous assumptions, including the composition of the workforce, the degree of competition in the labour market, and the impact of higher wages on consumer demand. In particular, the size of the minimum wage increase can be critical: below a certain threshold its impact may be insignificant."⁷⁵

[233] It is unnecessary to refer in detail to the methodological limitations which undermine confidence in much of the material. Those limitations have been explored in this and previous safety net review decisions. It is sufficient to refer to two persistent problems with surveys - unacceptably low response rates and lack of testing for non-response bias.

[234] We note that the Commonwealth has responded to concerns expressed by the Commission in the May 2003 decision about the lack of probative material in relation to the effect of safety net adjustments on employment. It commissioned the research referred to as the Minimum Wages Report. Unfortunately the utility of the report is severely restricted because of methodological limitations. Nevertheless we are encouraged by the Commonwealth's initiative in this area.

[235] We repeat what the Commission said in relation to the effect of safety net adjustments on employment in the May 2003 decision:

"[177] Various parties have submitted that the employment effects of safety net adjustments should weigh more heavily in the Commission's deliberations. That submission would obviously be assisted by high quality research into those effects. Some degree of consultation, and hopefully agreement, between the major parties and interveners as to the scope, content and technical specification of any such research would enhance the value of the results."⁷⁶

[236] The material to which we have been referred does not undermine the conclusion expressed by the Commission in the May 2003 decision that there is a continuing controversy amongst academics and researchers about the employment effects of minimum wage improvements. There is nothing before us to indicate that the controversy has been resolved. We endorse the view expressed in that decision that:

"[161] . . . Taking all of the research into account, it has not been established that moderate increases in the wages of the low paid, of themselves, will diminish aggregate employment outcomes, although some studies suggest that some negative effects might occur for employees receiving the minimum wage."⁷⁷

Surveys

[237] ACCI and NMI relied on telephone surveys of their memberships which sought to provide data on the impact of the May 2003 decision on employment and other indicators of economic performance.

ACCI Survey

[238] ACCI as in previous years conducted a survey of its members in order to develop an understanding of the labour market consequences of the May 2003 decision. It sought to provide evidence on the proportion of firms which passed on the adjustment, whether employment losses occurred and consequences for prices and profits.

[239] The survey was modified to address a number of methodological concerns which the Commission raised in previous decisions.

[240] The survey was conducted by ACCI member organisations using a sample frame of its members nationwide. The five largest businesses in each of the ACCI member organisations were deliberately included in the sample but all other businesses were randomly chosen.

[241] ACCI received 289 survey results representing a 57.8 per cent response rate, a substantial improvement on last year's 30 per cent response rate.

[242] The survey results showed:

• 58.9 per cent of employers passed the May 2003 safety net adjustment on to at least one employee;
• 34.3 per cent of employers passed the adjustment on to other employees within the firm;
• 32.2 per cent of employers passed the adjustment on both directly and indirectly;
• 42.1 per cent of employees did not receive an increase in wages resulting from the safety net decision;
• 69.3 per cent of non-award or overaward employees did not receive an adjustment in wages;
• 92.4 per cent of employers stated that the increase did not result in a change in full-time employment; 2.0 per cent of employers indicated employment had increased; 4.8 per cent of employers indicated employment had decreased;
• 85.5 per cent of employers indicated that the number of part-time and casual employees had remain unchanged; 10 per cent of employers indicated that employment had decreased; 4.5 per cent of employers indicated that employment had increased;
• 19.3 per cent of employers indicated they had increased prices as a result of the adjustment; and
• 34.8 per cent of employers indicated that profitability had decreased.

[243] ACCI submitted that the results showed a "small but definite reduction in employment". They also showed a "larger and more insidious effect on price level" and a "much larger effect" on the number of firms where profitability decreased.

[244] ACCI submitted that focusing on aggregate measures fails to take proper account of the damage caused to firms where the increase actually takes place.

[245] The conclusions to be drawn from the survey with respect to measuring the real impact of the safety net increase are limited by the parameters of the survey itself. Those limitations include:

• The survey results do not reveal the magnitude of any economic effects, whether adverse or otherwise. By recording the proportion of firms reporting particular effects, it is not possible to draw any conclusions with respect to numbers affected. To do so would require information on the size of a firm's workforce and numbers affected by the safety net increase.
• Similarly, we cannot draw any conclusions on the size of a price increase or a decrease in profitability which some employers reported.
• As pointed out by the ACTU in its reply submissions, the difference between estimates for responses to question 5 regarding increases and decreases in full-time employment is not statistically significant - the observed difference between the increase and decrease responses may be due to sampling variability rather than a "real" difference. With respect to question 6 regarding increases and decreases in part-time and casual employment, basing the statistics on the correct number of firms responding would produce a statistically insignificant result.
• The number of firms from which the proportions are calculated excludes some or all firms which did not pay safety net increases, thus producing higher proportions than if all firms were included in the survey results.
• In the absence of pre-testing the questionnaire, cross-check references and post-enumeration surveys, it may be that the ACCI survey suffers from the same deficiency as the Minimum Wages Report survey if the respondents failed to comprehend the concept of a safety net adjustment.

[246] For these reasons it would be inappropriate to attempt to extrapolate the survey results to provide an estimate of the effect of the safety net increase on numbers employed in Australia's workforce.

NMI Survey

[247] NMI submitted that the direct and indirect flow-on of award safety net decisions has a comparatively large impact on the industry which is characterised by about 90 per cent being small businesses and close to half of which (43 per cent) are located in rural and regional Australia.

[248] NMI relied on the ABS November 2003 survey of retail motor industry businesses which were asked about the economic effects of the May 2003 award safety net adjustment.

[249] A summary of responses is set out in Table 22 as follows:

Table 22: Effect of 2003 Safety Net Adjustment on Respondents' Business

[Source: AC Nielsen National Wage Case Survey, November 2003.]

[250] NMI acknowledged "that it is difficult to directly link the impact of the safety net adjustment to changes in a respondent's business".⁷⁸ However, it concluded that the survey results identify profitability and cash flow as two key aspects of NMI business operations which were adversely affected by the May 2003 decision. It challenged the ACTU's view that safety net adjustments will not have adverse economic effects.

[251] As with the ACCI survey, care must also be taken in interpreting the results which are confined to businesses where the safety net adjustment was passed on. The survey results are not percentages of all businesses in the sector. The survey also shares another limitation with the ACCI survey in failing to quantify the effects.

[252] Notwithstanding these limitations, the survey shows that the total number of employment hours worked and the number of permanent employees both declined in 9 per cent of businesses, with 91 per cent recording no change or an increase. Much higher proportions of firms recorded a decrease in cash flow (27 per cent) and profitability (40 per cent). It is not possible to draw conclusions on the magnitude of these changes.

Conclusion on Surveys

[253] The results of the ACCI and NMI surveys confirm the broad conclusions reached in the May 2003 decision:

• "there are employment effects of safety net increases with respect to some employers;
• such effects operate differentially; and
• adverse employment effects are more evident in relation to those employers directly affected by safety net increases."⁷⁹

[254] This year's surveys also demonstrate that as a result of the May 2003 decision:

• some firms increased prices; and
• a larger number of firms experienced a decrease in profitability.

ACCI Analysis of Safety Net Adjustments on Inflation

[255] ACCI produced data on award coverage by state and inflation by capital cities which it submitted demonstrate a positive correlation between the proportion of total employees whose pay is determined by award only and the rate of inflation. It also undertook a similar analysis on an industry basis which, it submitted, demonstrated that the percentage of employees in industries whose earnings are set by award only is positively correlated with the rate of inflation of prices for the goods produced in those industries.

[256] ACCI fairly acknowledged that this analysis was at a preliminary stage and more work was required to validate the conclusions arising from it. We agree but make a number of observations:

• Whilst acknowledging that in the material presented the two sets of data show a correlation between the extent of award coverage and the rate of inflation, it is not possible to draw any conclusions as to causality. There are a range of factors at work within a particular state or specific industry which may affect the level of inflation. Without controlling for these factors, it is not possible to draw inferences about the impact of award coverage on inflation.
• There are a number of methodological difficulties:
• The award coverage data relate to May 2002 whilst the CPI data relate to the year ended December quarter 2003.
• A small number of observations have been relied on. Earlier in this decision we referred to Professor Mitchell's comments on the validity of regression analyses which rely on less than thirty observations.
• The exercise has an element of artificiality in it partly due to the need to develop industry equivalents from the CPI components of goods and services. For example, the ACTU in its reply submissions removed cultural and recreation services and added a proxy for accommodation, cafes and restaurants which was not included in ACCI's original regressions, and used the CPI measure for the September 2003 quarter which is the quarter when the safety net adjustment had an impact. The regression resulted in a negative relationship between award coverage and inflation.

[257] The material in the form presented is of little assistance to us in explaining the role which the safety net adjustment has on the inflation rate award in reliant industries.

THE NEEDS OF THE LOW PAID AND LIVING STANDARDS GENERALLY

[258] Sections 88A, 88B and 90 of the Act require the Commission to balance a number of competing interests. The Commission is not concerned purely with the need to provide fair minimum standards for employees in the context of living standards generally prevailing in the Australian community and the needs of the low paid, the Commission must also have regard to economic factors, including levels of productivity and inflation, and the desirability of attaining a high level of employment and the state of the national economy and the likely effects on the national economy of any award or order particularly any effects on inflation and the level of employment.

[259] It ought now be regarded as well established that the expression "low paid" in s.88B(2)(c) refers to the low paid in employment and does not extend to include the low paid who are not employed.⁸⁰

[260] We agree with ACCI that there is no "statutory pre-eminence" accorded to the needs of the low paid in connection with the discharge of the Commission's functions pursuant to s.88B(2) of the Act. The Commission's primary obligation - indicated by the word "must" - is to "ensure that a safety net of fair minimum wages and conditions of employment is . . . maintained". In discharging that obligation the Commission has always proceeded on the basis that each of the matters specified in ss.88B(2)(a), (b) and (c) are matters to which the Commission must have regard. The weight to be accorded to each of those matters will depend upon the evidence and the particular circumstances prevailing in relation to the application(s). ACCI's apparent concern that the Commission will engage in an "over-consideration" of needs is unwarranted.

[261] At this point it is appropriate to note that, in relation to the questions posed in the emphasised portions of the extract from the May 2003 decision set out at paragraph 263, ACCI stated that it "does not necessarily accept the legitimacy of the questions raised by ACCER, nor that they constitute a complete range of considerations".⁸¹ The Commission regards the questions as legitimate and relevant to its task under s.88B(2), albeit that they are not exhaustive of the considerations arising in connection with the needs of the low paid. To the extent that ACCI regards the questions as illegitimate, it has not articulated a basis for that contention. The Commission will always receive submissions as to the range of considerations that it ought to take into account in the discharge of its functions under s.88B(2).

[262] Referring to the Commission's statutory powers under s.88B of the Act, ACCER submitted that the Commission has a duty to provide for a fair minimum wage in the context of fair living standards generally prevailing in the Australian community. It submitted that the Commission should not fix a poverty wage and that it would be failing to carry out its statutory duty if it merely had regard to bare minimum standards in setting the federal minimum wage. ACCER submitted that the Commission ought to be setting benchmarks by which needs can be established.

[263] In the 2003 Safety Net Review, ACCER and ACOSS submitted that the Commission should conduct an inquiry into the needs of the low paid for the purpose of determining benchmarks. In the May 2003 decision the Commission dealt with those submissions in the following way:

"[222] Our rejection of the proposals for an inquiry should not be taken as a rejection of the utility of empirically determined `benchmarks' such as the poverty line. Indeed, it seems to us that the use of such measures is relevant to an assessment of the needs of the low paid. In this context we also note that in their oral submissions ACCER argued that the Commission must ensure the minimum rates it sets (and in particular the federal minimum wage) do not fall below the poverty line. It was put that this task involved determining questions such as `what are needs, who are the low paid, what is the poverty line, what is living in poverty and how does the federal minimum wage compare to the poverty line?' We acknowledge the relevance of the questions posed by ACCER and would be assisted by submissions and material directed to them. As we have already noted empirical studies dealing with these matters would be of more assistance to the Commission in addressing the specific matters mentioned in the Act than the type of illustrative evidence adduced by the ACTU in these proceedings. There is no impediment to ACOSS and ACCER, or any other party, bringing forward such material in any future safety net review. It is not, however, desirable for the Commission to establish a separate inquiry for that purpose particularly in view of the absence of any support for the proposal from any other party or intervener." [emphasis added]

[264] In response to the emphasised comments above, the ACTU tendered evidence which it, and other parties, contended met the purpose identified by the Commission. In particular, ACCER now submits that the ACTU case in this safety net review has addressed these questions noting:

"[T]hey are important questions, but ultimately, their utility depends upon an ability to identify the kind of evidence needed, the compilation of that material and the preparedness of the Commission to discharge its statutory function on the basis of the evidence."⁸²

Senate Poverty Inquiry Report

[265] At the time of the 2003 Safety Net Review, the Senate Community Affairs Committee was conducting an inquiry into poverty and financial hardship. The Committee's report was tabled on 11 March 2004. The ACTU sought, in its supplementary submission, to rely upon that report. The Committee divided on party lines. The Majority Report of the opposition and minor party senators noted:

"4.72 Severely limited opportunities are often part of the life experiences of low wage working poor individuals and their families. A lack of financial resources often has adverse flow-on effects for workers and their families. A lack of money can lead to reduced access to preventative health and other services; reduced educational opportunities for their children and a disincentive for them to participate in post-secondary education; and a reduced ability to participate in social activities and the wider society generally. Lack of financial resources also reduces a worker's asset base with more likelihood that their financial difficulties will persist in to old age . . .

4.87 Minimum full-time wages have fallen well behind average wages over the last 20 years, especially in the early years of the shift to enterprise bargaining, before the present round of `Living wage' cases was instituted in 1996. The minimum wage has now fallen to just 50 per cent of average earnings, a reduction of 15 per cent since 1983."⁸³

[266] The Majority Report included the following recommendation:

"Recommendation 6

That the Australian Industrial Relations Commission establish a new minimum wage benchmark based on a wage level that enables a single full-time worker to achieve an adequate standard of living relative to contemporary community standards."⁸⁴