Chapter 8 – Input Costs

8) Input Costs


9.The input costs factors take account of non-policy differences in costs of inputs to service provision: wages and salaries[1], office accommodation and electricity. These inputs are common to all State functions, and input cost disabilities are thus assessed for most expenditure categories.

10.In concept, the input cost assessment measures the ‘cost’ disabilities (the differences in unit price of inputs to services), as opposed to the ‘demand’ disabilities (the differences in the per capita demand for State services).

11.Given the above average wage costs faced by the ACT in the provision of government services, and hence the relative importance of the input costs assessment, the ACT has given a priority to the examination of this input cost methods.

12.In the 1999 Review several States attempted to dilute the intent of the input costs assessment. This position appears not to have changed in the 2004 Review, with Queensland in their Workplace Discussion Briefing Notes disputing the current methods of assessment. The ACT endorses the current approach, albeit with some minor changes.



Introduction

13.The ACT supports the application of the current input costs assessment to the 2004 Review, and considers that no substantial changes are required given the veracity of the current approach.

14.Importantly, the ACT views the non-State sector (as opposed to the private sector alone) as representing the best possible proxy for assessing wages and salaries disabilities. The ACT considers that other approaches suggested by a number of States for the determination of disabilities are poor in comparison to the non-State sector. For example:

  • the use of private sector wages alone would result in disabilities that are not reflective of circumstances faced by all States; and
  • State public sector wages would compromise the policy neutral approach which the Commission has been able to achieve through the use of the non-State sector.

15.In regard to the input costs wages assessment component, the ACT requests that wage disabilities continue to be assessed:

  • for both the general and specific provision of government services given the extent of the disability faced by the ACT; and
  • based on the non-State sector as it represents the best policy neutral measure of wage differentials between the States. In doing so, the ACT also requests that:

current practice of standardising State employment by the national employment distribution as it may not necessarily reflect a truly policy neutral position on which to assess wage input cost disabilities; and

the discounting of the raw wage input cost factor be revised downwards from 50% to 25% given the:

improvement in the reliability of the Average Weekly Ordinary Time Earnings statistical series as advised by the Australian Bureau of Statistics (ABS); and

the changes in the composition and mix of the workforce since the 1999 Review which suggest that changes to the heterogeneity judgements in the assessment are required.

Support for the input costs assessment

The need for State a wage differential assessment

16.The ACT supports the continued assessment of wages disabilities given that the Territory’s above average wages are a cost disability affecting the provision of all government services.

17.The ACT faces relatively high public sector wages for a selection of key occupations.

Structural Nature of Wage and Salary data

18.The ACT considers that the major drivers of State wages are the labour market conditions and the socio-economic characteristics existing in each State. The variables that have greatest amount of influence on wages include both demand and supply elements.

Demand elements include:

  • education and skill levels;
  • union influence/industrial strength;
  • the relative size of the State labour force; and
  • levels of unemployment.

19.Supply elements include the:

  • number and size of firms;
  • industry composition of all firms;
  • cost of living; and
  • level of economic activity.

20.In the ACT’s case, non-State employment comprises a much greater proportion of total employment than any other sector. The ACT Government regularly competes against Federal Government Departments in the local labour market in many similar occupations where more homogenous employment opportunities exist between State and non-State areas of employment.

21.The ACT considers that as there are non-policy influences which affect the wage differentials between the States, an input costs factor for each category (for both the scale affected component and other relevant components) should continue to be assessed.

Assessment of the Non-State Sector

22.The ACT supports the conclusion contained in the 1999 Review of General Revenue Grant Relativities. That is, that a non-State sector proxy for wage differentials, which combines Commonwealth public sector wages and private sector wages, should be used to assess wage disabilities.

23.The ACT supports the view put forward by a number of the States in the 1999 Review that the non-State sector proxy is a robust way of measuring wage disabilities. The use of non-State wages represents a policy neutral position and therefore reflects the best measure available for State public sector wage differentials.

24.The use of private sector wages by itself:

  • increases the variability of the sample, with a consequent reduction in the veracity of the assessment; and
  • would result in disabilities that are not reflective of circumstances in all States.

25.In some cases, State wages are largely influenced by Commonwealth wages, and it could be argued that State public sector wages are more reflective of Commonwealth public sector wages, than of private sector wages. The importance of Commonwealth wages is primarily due to the:

  • similarity and homogeneous nature of public sector occupations between the States and the Commonwealth; and
  • fact that a reasonably large proportion of Australia’s workforce (around 38%) are employed pursuant to federal enterprise awards.

26.The ACT is also opposed to the use of State public sector wages, in whole or part, to derive disabilities as this approach would compromise the policy neutrality method which has been achieved through the use of the non-State sector.

Police Wages and Salaries

27.In the ACT’s case, the wages of police officers represents a very strong example of the relationship between State public sector wages and Commonwealth public sector wages.

28.The ACT, via an agreement with the Commonwealth, remunerates Federal police officers, to provide community policing functions in the Territory. The Australian Federal Police (AFP) is, for all intents and purposes then, the ACT’s State police force as it provides the same functions as the forces in the other States (this matter is discussed in more detail later in this chapter).

29.While the strategic policy decisions regarding the AFP’s employment are still undertaken by the Commonwealth Government, the agreement has the effect of limiting the role the ACT has in relation to setting employment conditions and controlling expenses. The ACT is therefore obliged to remunerate ACT police at levels imposed by the Commonwealth. This example highlights the extent of one of the influences which the Commonwealth has on the ACT’s State wages. This gives further support to the need for the non-State sector to include Commonwealth wages.

Nursing and Teachers Wages and Salaries

30.In regard to other occupations such as nurses and teachers, the situation is no different. It is important for the ACT to remain competitive with the Commonwealth and prevent a departure of vast numbers of talented and qualified staff to equivalent, higher paying Commonwealth Government occupations. For example, teachers might seek Commonwealth positions in a range of occupations such as with the:

  • CSIRO; and / or
  • Federal Department of Education, Training and Youth Affairs; and / or
  • ACT’s six post secondary educational institutions (driven largely by the fact that the Commonwealth Government is located in Canberra):

the Australian National University;

the University of Canberra;

the Australian Catholic University (Signadou Campus);

Canberra Institute of Technology;

Australian Defence Force Academy (University of NSW);

Australian Defence College; and / or

  • various other training and other education based positions.

Public Service Wages and Salaries

31.Increasingly, it has become necessary for the ACT to respond to moves by the Federal Government regarding raising wages for Commonwealth employees. The ACT must respond to these actions in order to attract and retain high quality staff.

32.The close proximity of the Commonwealth and the ACT Public Sectors within the ACT, and their interrelationship, lends itself to a high degree of workforce mobility. It is common for public servants to span their careers within both sectors, particularly because of the adoption of similar employment conditions and superannuation arrangements.

33.In addition to its proximity, the significance of the Commonwealth Public Sector in the ACT [2] impacts on the ACT Public Sector (ACTPS) wages and employment conditions.

34.The consequences of a significant gap in wages and conditions are a combination of both:

  • serious deterioration in the quality of the workforce; and
  • associated decline in productivity.

35.States regularly compete in the labour market for employment. Employment and wages in each of the states reflect the different socio-economic characteristics of the states. Essentially, States compete for employment in the same geographical labour market and it is the non-State sector that determines wages.

36.The ACT considers that a non-State sector proxy for wage differentials, which combines Commonwealth public sector wages and private sector wages, is the most appropriate policy-neutral measure for assessing wage costs disabilities.

37.The ACT considers that while the flow-on effects of increased wages from the Commonwealth to the States occurs in each jurisdiction, it is stronger and more direct in the ACT than in any other State.

38.The uniqueness of particular arrangements, such as the community policing arrangements with the AFP, and the proximity of the Commonwealth and ACT Public Sectors, only serves to strengthen the evidence that:

  • there are strong linkages between State and Federal wages; and
  • the non-State proxy should include both private and Commonwealth wages.

39.In summary, the ACT supports the:

  • retention of input costs as an assessment for the 2004 Review;
  • assessment of the non-State Sector as the most appropriate measure of State wage differentials; and
  • use of ABS Average Weekly Ordinary Time Earnings to derive wage disabilities.

Support for various aspects of the input costs assessment

Superannuation and Non-Wage Payments

40.The ACT supports the finding that the assessment of input costs for wages should not include superannuation contributions and other nonwage payments.

41.The ACT considers that the inclusion of these payments might compromise the current input costs assessment to a degree as some evidence to date has suggested that data availability on these payments is not satisfactory.

42.The ACT considers that unless reliable data can be found that can indicate the overall impact which these non-wage payments have on each States’ wage input costs, any attempt to include contributions from superannuation and other non–wage payments may result in the determination of inaccurate wage relativities.

Dominant Employer Effect

43.Several States have argued against the inclusion of Commonwealth wages in the non-State sector proxy as they consider that the Commonwealth has the ability to exert excessive influence over wages. In other words, it has a dominant employer effect.

44.It has been determined by the Commission that for a dominant employer effect to in fact exist, a number of assumptions would have to be positive. Principally, there would need to be substantial labour market segmentation as well as a relatively immobile workforce. The ACT does not believe that this is the case.

45.The ACT notes that research conducted by the Commission indicated, in relation to the quantitative tests, that no amount of wage differentiation that enabled a dominant employer effect to be present, actually existed.

46.In relation to the qualitative tests undertaken on teacher and nurse salaries in NSW, some similarities between the movements in private and public sector wages was always found. “On the whole we think that there is some dominant employer effect for salaries paid to teachers and nurses in NSW and Tasmania.”[3]

47.Despite lengthy tests being conducted on this issue, many of the other tests conducted to trace dominant employer effects remain inconclusive. The ACT considers that it was correctly concluded in the 1999 Review that wage differentials resulting from dominant employer effects are due to non–policy influences, and are therefore, dealt with appropriately through the current assessment methods.

48.Wage differentials are predominantly a result of the influence of market forces and the increased expenditure associated with maintaining a high quality workforce, and the other demand and supply influences referred to earlier on in this chapter, rather than the actions of States.

49.The ACT agrees with research indicating that wages are largely determined through external labour market forces and that therefore, no dominant employer effect exists.

National Benchmarking

50.Smaller States such as WA, SA and Tasmania have previously indicated that movement towards a national award scheme would remove wage differentials among States. The argument put forward is that State workers will be more motivated to argue for wages which reflect the characteristics and national circumstances for their occupations.

51.The system of industrial relations in Australia has undergone substantial transition in the last twenty years. The introduction of enterprise bargaining throughout the country since 1991 has increased the impetus for employers and employees to negotiate wages and conditions that will deliver them the most desirable outcomes given their individual circumstances.

52.In vast numbers, each State’s labour force has started to embrace enterprise bargaining.

53.The ACT considers that with the acceptance of enterprise bargaining, and its reflection of individual workers’ circumstances, wage differentials will continue to exists between the States.

Suggested Changes to the Input Costs Assessment

Industry, Occupation Structure and Education

54.As part of the recommendations in the 1999 Review, it was determined that the variables of occupational structure and industry composition had a significant effect on wage differentials. The influence of differing occupational structures was removed by standardising each States’ employment structure by the national distribution of employees by industry.

55.It appears that reasons for introducing the national employment distribution were based largely on the work undertaken by Borland and Lye [4], where an econometric analysis was undertaken estimating earnings differentials for all States.

56.Larger states such as NSW and Victoria are not affected to a great extent by this exercise, mainly because their employment structures dominate national averages. Given the relative size and diversity of employment in some of the smaller States, however, this exercise is likely to have the greatest impact upon the ACT:

“Clearly the most populous states (NSW and Victoria) are hardly affected by the standardisation (since their employment data dominate the national averages), while the greatest effect is for the ACT.” [5]

57.The ACT requests that the results of the econometric analysis by Borland and Lye be discounted largely because of the effects on the ACT and other smaller States.

58.The ACT considers that the inclusion of the Commonwealth sector within the non–State sector proxy is enough to remove the effects of differing occupational composition from salary and wage data.

59.The ACT requests that the current practice of standardising State employment by the national employment distribution be examined as it may not necessarily reflect the real underlying wage input cost disabilities.

Discounting

Data Quality and sampling variability

60.The ACT supports the current treatment of discounting, with a few minor amendments. In the 1999 Review, it was considered appropriate to discount the raw wage cost factor by 50% in order to:

  • minimise the impact of variable AWOTE data; and
  • take account of market segmentation issues between the private and public sectors.

61.The ACT requests that the extent to which the raw wage cost factor is discounted, be examined. The ACT considers that changes in relation to data quality have improved to such an extent since the 1999 Review, that the raw wage discount factor should be reduced from 50% to 25%.

62.Principally, the relatively small population in the ACT compared with the larger States results in a much greater standard error being allocated to the ACT by the ABS in the Survey of Employee, Earnings and Hours. Currently, the assessment takes into account the effects of sampling variability by extending the coverage of data to nine years in order to generate more accurate information from the time series.

63.The Standard error is defined as the degree to which an estimate may vary from the value that may have been obtained from the full population. Another method of estimating the sampling variability is the relative standard error, this method involves expressing the standard error as a percentage of the estimate to which it refers.

“The relative standard error is a useful measure in that it provides an immediate indication of the percentage errors likely to have occurred due to sampling, and thus avoids the need to refer also to the size of the estimate.” [6]