BY JONATHAN SINGER Establishment newspaper headlines still proclaim Australia's “powerhouse economy”, but have workers really benefited from all the increased wealth of the past decade?
Measured by growth in gross domestic product (GDP), the economy has been expanding continuously since the 1990-91 recession, and at a annual rate of more than 4% for the last three years.
The growth is real enough, even after taking into account an increasing population: according to research by the government's Productivity Commission, income per person grew by 2.9% each year in the 1990s, underpinned by annual productivity rises averaging 2.2%.
Uneven distribution
But while aggregate figures of employees' incomes do suggest some gains, the distribution is uneven and the gains come at the cost of an extended, and intensified, working day.For example, the total compensation (wages, salaries and superannuation) paid to all employees increased by 35% in real terms between the 1990-91 financial year and 1998-99, mirroring GDP growth over the period.
As a share of GDP, employees' compensation was relatively stable at about 54%. This may be compared with the 1980s, when the wages share fell from about 60% at the start of the decade to a low point of 52.8% in 1988-89.
The 35% rise in total compensation is due to both a 22% increase in pay rates for wage and salary earners and an 11% increase in the absolute numbers of wage and salary earners. Given that wage and salary earners' proportion of the work force and the population hasn't changed, however, this doesn't signify any direct loss of income for workers.
The rate of pay for non-managerial workers, however, when calculated against GDP, rose by perhaps only about 14% in that period, and other Australian Bureau of Statistics (ABS) data suggest similar amounts. Again, the number of these workers increased, but the total hours of their work increased at an even faster rate: the stable wages share is being spread more thinly as compensation for workers' efforts.
Rates of pay have also tended to become more unequal during the 1990s. Much of this has been based on an increasing variety of legal structures for determining employees' pay.
Before the 1990s, arbitration court awards set most pay levels. Over-award payments could also be made. In 1991 the then federal ALP government introduced enterprise agreements negotiated between employers and unions, and later extended this to include non-union agreements.
The Coalition government's 1996 Workplace Relations Act introduced an individual contracts system, called Australian Workplace Agreements (AWAs), and restricted workers' rights to organise and be active in unions.
As a result, awards generally became less important in determining pay. The federal workplace relations department's 1999 survey of award and agreement coverage, involving organisations with five or more employees, showed 22% of employees were paid solely according to awards, the same amount got over-award payments or pay under some other kind of unregistered agreement, 42% worked under registered collective agreements and 14% had other pay agreements, presumably mostly individual ones.
Industry variation
How many people worked under these different pay arrangements varied across industries.Employees in the mining and construction industry were mainly subject to either individual or unregistered arrangements. Government administration and manufacturing industry employees predominated among those working under registered collective agreements. Finance, property and business services employees largely worked under one or other form of collective agreement.
The highest proportion of award rate workers were found in education, health and community services and in hospitality, retail and similar industries.
One consequence of the reduced influence that awards have on wages is that, in March 1997, the ABS abandoned its previous statistical measure that tried to reflect changes in labour costs to employers through surveying a sample of jobs, an award rates index.
No doubt this also helped to hide some figures that would suggest that some workers' living standards were going backwards, not forwards, during the 1990s “boom”: in the seven years from December 1989 to December 1996, the index fell by 3% in real terms.
Award rates
Awards, however, are still much more significant in directly governing the pay of lower-paid workers. Also, the lower the pay of a worker not under a registered agreement, the greater the award as a proportion of their pay.For example, according to a table published in the June 1999 Journal of Australian Political Economy by John Buchanan, Ron Callus and Chris Briggs from the Australian Centre for Industrial Relations, Research and Training (ACCIRT), in 1989 the award rate of pay as a proportion of actual pay rates for a non-agreement fitter (the award's C10 classification) in the metals industry was 77.4% and was still this in 1996, while the same proportion for a metals process worker (C13) rose from 87.7% to 91.8%.
This also suggests both classifications of workers were falling behind the general movement in wages, since the award fell by as much as a sixth in comparison with average earnings in the industry.
For these lower-paid workers the national increases in award rates granted in the 1990s were proportionally greater than for the work force as a whole because the increases were, after a 2.5% rise in 1991, flat dollar amounts. Three $8 per week amounts were granted from 1994 to 1996, $10 in the first of three annual living wage cases in 1997, $14 in 1998 (but only $12 for those on between $550-700 per week and $10 for those with awards above that) and $12 in 1999, with a further $15 granted this year.
These award increases took the minimum wage — the C14 classification of the metals industry award — from less than $320 per week in 1990 to $385.40 in 1999 and then to $400.40 in 2000. This has meant the minimum wage has more or less kept pace with price rises over the decade.
Wage increases in enterprise agreements in the 1990s were larger: ACCIRT data shows agreements averaged above 4% from 1993 into 1998, peaking at 5.7% in 1996, and falling further to about 3% in 1999. These created real increases in pay rates, as inflation fell to 2% or less at the end of the decade.
But particular agreements provided for annual increases as great as 24% or as little as 1%. In general, agreements influenced by and struck with unions involved higher pay increases than those driven by employers, including non-union agreements.
The increases in individual agreements varied even more wildly. Higher-level executives were racing ahead of average wage movements. Many employees' AWAs, however, had no provisions for pay increases during the contract's lifetime.
Performance-based pay
A feature of both collective and individual agreements is the extent to which pay is put “at-risk” through making pay conditional on performance. John Shields told an ACCIRT conference in March that a 1999 remuneration survey found an average 6.1% of pay for non-managerial employees was subject to provisions such as merit pay rises and bonuses, goal sharing plans for work teams and share purchase plans.ABS data on employee benefits shows that the number of employees receiving shares as a part of their remuneration package grew from 2% to 6% over the decade.
Performance pay measures tend to be found in larger, non-unionised private firms in industries such as finance and insurance, wholesale and retail trade, mining and property and business services, and also in the manufacturing and cultural and recreational services sector.
Among certified agreements, however, references to measures such as tying pay to productivity increases are more common where there is union involvement.
A result of enterprise and individual agreements and performance-related pay is a growing differentiation between basic hourly pay rates and total pay rates.
Average increases in hourly rates paid have been between 1% and 2% per year — that is, more or less stagnant against price rises — according to Melbourne Institute for Applied Economic and Social Research surveys which began in 1997. Total pay increases, in the same surveys, have ranged from 2% to 5%.
This suggests pay increases won in the 1990s are precarious, based largely on bonuses and productivity rises, and may be difficult to retain in a new recession without strong workers' organisations.
Nevertheless, the ABS's new wage cost index, which incorporates non-award payments and began in September 1997, does show a real increase in pay: in the two-and-a-half years until March 2000, the index has risen by 3%.
This, along with pay increases from enterprise agreements, has fed into real increases in workers' earnings in the 1990s (as shown in Table 1), despite its increasingly unequal distribution.
[This article is the first in a series on wages and working conditions in the 1990s. The next article will examine earnings inequality, hours and working conditions.]
Ìý